Full text of Essays in Honor of Marvin Goodfriend : Economist and Central Banker

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

Essays in Honor of
Marvin Goodfriend:
Economist and Central Banker
Edited by Robert G. King and Alexander L. Wolman

|1

2 |

|3

Essays in Honor of
Marvin Goodfriend:
Economist and Central Banker
Edited by Robert G. King and Alexander L. Wolman

4 |

|5

The views expressed here are those of the authors and do not
necessarily reflect the views of the Federal Reserve Bank of Richmond
or the Federal Reserve System.
Copyright © 2022
All rights reserved.

Managing Editor: Lisa Davis
Copy Editor: Katrina Mullen
Designer: Cecilia Bingenheimer
Photograph of Marvin Goodfriend courtesy of Carnegie Mellon University
6 |

|7

Biography

8 |

Marvin Goodfriend (1950-2019) joined the Research Department at
the Federal Reserve Bank of Richmond in 1978 after graduate study
at Brown University. He was director of research from 1993 to 1999
and senior vice president and policy advisor from 1999 to 2005.
After joining Carnegie Mellon University as a professor of political
economy, he became a leading academic figure in
international central banking circles.

|9

Table of Contents
Introduction.......................................................................................................15
Contributing Authors......................................................................................17
Personal Reflections
Marvin Goodfriend at the Richmond Fed: Recollections .......................25
J. Alfred Broaddus Jr.
The Federal Reserve’s New Monetary Policy Framework.......................43
Donald Kohn
What Does the Fed Know, and When Does it Know It?...........................59
William Poole
Perspectives on Policy and Research Contributions
Central Bank Lending and Incentives..........................................................79
Kartik Athreya and Stephen D. Williamson
Marvin Goodfriend and the Zero Lower Bound........................................95
Ben S. Bernanke
Federal Reserve Structure and Economic Ideas.......................................105
Michael D. Bordo and Edward S. Prescott
Banking Policy and Monetary Policy..........................................................119
Douglas W. Diamond
Interest Rate Smoothing................................................................................129
Michael Dotsey, Andreas Hornstein, and Alexander L. Wolman
Paying Interest on Bank Reserves................................................................143
Huberto M. Ennis and John A. Weinberg
Reconsidering the Case for Price Stability.................................................163
Vitor Gaspar and Frank Smets
The 2007 Monetary Policy Consensus in Retrospect .............................185
Mark Gertler
From the Great Inflation to the Great Moderation................................. 203
Robert L. Hetzel
10 |

| 11

Taming Inflation Scares................................................................................. 217
Athanasios Orphanides and John C. Williams
Federal Reserve Independence:
Is it Time for a New Treasury-Fed Accord?................................................ 229
Charles I. Plosser
Industrial Development and Convergence.............................................. 271
Sergio T. Rebelo and Pierre-Daniel G. Sarte
Rational Expectations and Volcker’s Disinflation.................................... 279
Thomas J. Sargent
Monetary Mystique and the Fed’s Path
Toward Increased Transparency.................................................................. 289
Lars E.O. Svensson
Interest Rate Policy......................................................................................... 303
John B. Taylor
The Implications of Optimal Prediction Formulae................................. 315
Mark W. Watson
The New Neoclassical Synthesis
and the Role of Monetary Policy................................................................ 323
Michael Woodford
Credibility and Explicit Inflation Targeting............................................... 347
Robert G. King and Yang K. Lu

12 |

| 13

Introduction
Robert G. King and Alexander L. Wolman, editors
Marvin Goodfriend inspired many with his ideas about central banking and economics. This memorial project, sponsored by the Federal
Reserve Bank of Richmond, begins with three personal reflections by
central bankers who knew him well: Al Broaddus, Don Kohn, and Bill
Poole. Marvin joined the Richmond Fed in 1978 after studying with
Poole at Brown University, rising to be Broaddus’s senior policy adviser.
Combining research excellence with a knowledge of both the Fed’s
history and its contemporary challenges, Marvin was passionate about
Richmond’s contributions to the Federal Open Market Committee
meetings in Washington. Attending these meetings at Broaddus’s side
in the 1990s and early 2000s, Marvin developed a friendship — based
on mutual respect rather than complete coincidence of views — with
Kohn. After Marvin joined the Carnegie Mellon faculty in the fall of
2005, his presence in international central banking circles, sometimes
as an adviser, only grew.
An online collection gathers two dozen of Marvin’s best-known
papers with 18 topical essays based on these papers by his colleagues
and contemporaries in central banks and academia. The personal reflections and topical essays are collected in this volume.
Since Marvin’s analyses were frequently unconventional and at
times controversial, the authors of these essays did not always agree
with him at the time and do not always agree with him now. But many
describe how they have been stimulated by his thinking and through
personal interactions over many years.
Building this memorial project led us to invite contributions from a
group of influential and busy economists. As we did so, a remarkable
pattern emerged: acceptance was immediate, and we were thanked
for creating the project as well as for the invitation to participate. In itself, that is a tribute to Marvin Goodfriend as a person and a researcher.

14 |

We had the good fortune to work with Marvin in various ways over
many years. It is our hope that this volume and the online collection
will help readers gain new insights both from Marvin's work and the
related essays.
| 15

Contributing Authors
Kartik Athreya is executive vice president and director of research at
the Federal Reserve Bank of Richmond.
Ben S. Bernanke is a distinguished senior fellow with the Economic
Studies program at the Brookings Institution. From February 2006
through January 2014, he was chairman of the Board of Governors of
the Federal Reserve System.
Michael D. Bordo is a Board of Governors Professor of Economics and
director of the Center for Monetary and Financial History at Rutgers
University. He is the Ilene and Morton Harris Distinguished Visiting
Fellow at the Hoover Institution, Stanford University, a research associate at the National Bureau of Economic Research, and a member of the
Shadow Open Market Committee.
J. Alfred Broaddus Jr. was president of the Federal Reserve Bank of
Richmond from 1993 to 2004. Prior to this, he was the Bank’s director
of research.
Douglas W. Diamond is the Merton H. Miller Distinguished Service
Professor of Finance at the University of Chicago’s Booth School of
Business. He is a research associate of the National Bureau of Economic
Research and a visiting scholar at the Federal Reserve Bank of Richmond.
Michael Dotsey is executive vice president, director of research, chief
economic advisor, and director of the Consumer Finance Institute at
the Federal Reserve Bank of Philadelphia. He is also an associate editor
for the International Journal of Central Banking.
Huberto M. Ennis is group vice president for macro, micro, and financial economics in the Research Department at the Federal Reserve
Bank of Richmond. He is an expert on financial economics and serves
as co-editor of the International Journal of Central Banking and associate editor for the Journal of Monetary Economics and Macroeconomic
Dynamics.

16 |

| 17

Vitor Gaspar is director of the Fiscal Affairs Department at the International Monetary Fund. Previously, he was Portuguese Minister of State
and Finance, head of BEPA at the European Commission, directorgeneral of research at the European Central Bank, director of Economic
Studies and Statistics at the Central Bank of Portugal, and director of
Economic Studies at the Portuguese Ministry of Finance.
Mark Gertler is the Henry and Lucy Moses Professor of Economics at
New York University and a research associate of the National Bureau of
Economic Research. He currently serves as a co-director of the NBER’s
program on Economic Fluctuations and Growth and as a consultant to
the European Central Bank.
Robert L. Hetzel was an economist in the Research Department at the
Federal Reserve Bank of Richmond from 1975 to 2018. He is author of
The Federal Reserve: A New History (forthcoming, University of Chicago
Press); The Monetary Policy of the Federal Reserve: A History (2008); and
The Great Recession: Market Failure or Policy Failure? (2012).
Andreas Hornstein is a senior advisor in the Research Department at
the Federal Reserve Bank of Richmond. His work has been published in
the American Economic Review, the Review of Economic Studies, and the
NBER Macroeconomic Annual.
Robert G. King is professor of economics at Boston University, after
time at the University of Rochester and the University of Virginia. He
is a research associate at the National Bureau of Economic Research, a
Fellow of the Econometric Society, former editor of the Journal of Monetary Economics, and long-time visitor to the Richmond Fed.
Donald Kohn holds the Robert V. Roosa Chair in International Economics and is a senior fellow in the Economic Studies program at the
Brookings Institution. He is a 40-year veteran of the Federal Reserve
system, serving as member and then vice chair of the Board of Governors from 2002-2010.
Yang K. Lu is associate professor of economics at Hong Kong University of Science and Technology. She obtained her PhD from Boston
University and has been a Max Weber fellow at the European University Institute.
18 |

Athanasios Orphanides is a professor of the practice of global economics and management at the MIT Sloan School of Management and
co-chair of the Board of Governors of Asia School of Business. He is also
an honorary advisor to the Bank of Japan’s Institute for Monetary and
Economic Studies, a member of the Shadow Open Market Committee,
a research fellow at the Centre for Economic Policy Research, a senior
fellow at the Center for Financial Studies, and a research fellow at the
Institute for Monetary and Financial Stability.
Charles I. Plosser is a visiting fellow at the Hoover Institution at Stanford University. He was the president and CEO of the Federal Reserve
Bank of Philadelphia from 2006 until his retirement in 2015. He also
served as a public governor for FINRA from 2016-2021.
William Poole was president of the Federal Reserve Bank of St. Louis
from 1998-2008. He is currently Distinguished Senior Scholar at the
Mises Institute and Senior Advisor to Merk Investments. Previously, he
taught at Brown University and Johns Hopkins University, was a senior
economist at the Board of Governors, and was a member of the Council of Economic Advisers.
Edward S. Prescott is a senior economic and policy advisor in the
Research Department of the Federal Reserve Bank of Cleveland. From
1995-2015, he worked in the Research Department of the Federal
Reserve Bank of Richmond. He received a PhD in economics from the
University of Chicago.
Sergio T. Rebelo is the MUFG Distinguished Professor of International
Finance at the Kellogg School of Management at Northwestern University. He is the co-director of Northwestern’s Center for International
Macroeconomics. He is a fellow of the Econometric Society, the National Bureau of Economic Research, and the Center for Economic Policy
Research.
Thomas J. Sargent is the W.R. Berkley Professor of Economics at New
York University and a senior fellow at the Hoover Institution at Stanford
University. He was awarded the 2011 Nobel Prize in Economics with
Christopher Sims.

| 19

Pierre-Daniel G. Sarte is a senior advisor in the Research Department
at the Federal Reserve Bank of Richmond. He has taught at the University of Richmond and the University of Virginia and has served as a
visiting scholar at the Reserve Bank of New Zealand and the University
of Queensland. He is an associate editor of the Journal of Monetary
Economics and has served as a co-editor of Economics Letters.
Frank Smets is director-general of DG Economics at the European
Central Bank and professor of economics in the Department of Economics at Ghent University in Belgium. At the ECB, he is also secretary
for monetary policy of the ECB's Governing Council and chair of the
Monetary Policy Committee.
Lars E.O. Svensson is affiliated professor at the Stockholm School of
Economics. He is a fellow of the Econometric Society, a fellow of the
European Economic Association, a research associate of the National
Bureau of Economic Research, and a research fellow of the Centre for
Economic Policy Research, London.
John B. Taylor is the Mary and Robert Raymond Professor of Economics at Stanford University and the George P. Shultz Senior Fellow
in Economics at the Hoover Institution. He is director of the Stanford
Introductory Economics Center. He formerly served as director of the
Stanford Institute for Economic Policy Research, where he is now a
senior fellow.

was previously president and CEO of the Federal Reserve Bank of San
Francisco.
Stephen D. Williamson is the Stephen A. Jarislowsky Chair in Central
Banking in the Department of Economics at the University of Western
Ontario. He is also a Bank of Canada Fellow, a member of the Monetary
Policy Committee at the C.D. Howe Institute, co-editor of the Canadian Journal of Economics, and senior associate editor of the Journal of
Monetary Economics.
Alexander L. Wolman is vice president for monetary and macroeconomic analysis at the Federal Reserve Bank of Richmond. He joined the
Richmond Fed as an economist in 1996, after earning his doctorate
from the University of Virginia. He has served as an associate editor of
the Journal of Monetary Economics and the Journal of Money, Credit and
Banking.
Michael Woodford is the John Bates Clark Professor of Political Economy at Columbia University. He co-directs the NBER’s working group on
Behavioral Macroeconomics.

Mark W. Watson is the Howard Harrison and Gabrielle Snyder Beck
Professor of Economics and Public Affairs at Princeton University, a
research associate at the National Bureau of Economic Research, and a
long-term consultant with the Federal Reserve Bank of Richmond.
John A. Weinberg is a policy advisor emeritus in the Research Department at the Federal Reserve Bank of Richmond. Previously, he served
as director of research, senior vice president, special advisor to the
president, and policy advisor at the Richmond Fed.
John C. Williams is the president and CEO of the Federal Reserve
Bank of New York. In that capacity, he serves as the vice chairman and
a permanent member of the Federal Open Market Committee. He
20 |

| 21

Personal Reflections

22 |

| 23

Recollections

Marvin Goodfriend at the Richmond
Fed: Recollections
J. Alfred Broaddus Jr.
Marvin Goodfriend joined the Richmond Fed Research Department
as an economist in 1978, in the middle of the Great Inflation, and
worked there until his retirement in 2005, when he became professor
of economics at Carnegie Mellon University. He viewed the 20th century monetary policy experience as an “odyssey” from the gold standard to today’s inconvertible paper standard supported by a credible
Fed commitment to price level stability.1 While in Richmond, Marvin
participated meaningfully in the latter stages of that odyssey and the
substantial progress it represented. I was privileged to work closely
with him for most of the time he was in Richmond. I’m happy to have
this opportunity to share a few memories of what our Reserve Bank
achieved in those years with Marvin’s extraordinary intellectual and
personal leadership along with a few details of how we achieved it.
I think I can recall the exact time and place I first met Marvin. It was
Wednesday, December 28, 1977. Bob Hetzel and I were in a room in the
NY Hilton interviewing candidates to join us in our Richmond Fed Research Department. Bob stepped out for a minute. When he returned,
he looked at me and said, “G. William Miller.” A big question mark must
have appeared in a bubble over my head, so Bob added that President
Carter had just nominated Miller to succeed Arthur Burns as Fed chair.
We returned to interviewing candidates. Little did I know that one of
the most important and fortunate events of my life and in the history
of the Richmond Fed as a policy unit was about to occur.
When the final candidate of the day knocked on the door, it was getting dark in Manhattan, and I remember being weary. Marvin was the
candidate. I don’t recall for sure, but I’d bet he was wearing a beige and
brown argyle patterned sweater vest. I do recall clearly that my first
impression of him was altogether positive. There was none of the
1

24 |

Goodfriend (1997).

| 25

Broaddus
awkwardness or resume padding that often characterizes these
interviews. I remember thinking “with this guy, what you see is what
you’ll get.” I liked what I saw, and Bob did as well. At the time, we were
hiring with the objective of strengthening our department’s ability to
contribute meaningfully to Fed monetary policy; we wanted the significant resources we were devoting to policy research to be justified
by increased influence in the broader policymaking process. We were
therefore talking to a number of very well-qualified people that we
knew would be hard to attract to Richmond. In the interview, Marvin
expressed some interest in what we were doing, but as he left, I recall
thinking our chances of attracting him were slim.
What a pleasure to learn not long after that Marvin would accept our
offer. I would have been even happier had I known then that he would
stay almost 30 years and that my Richmond colleagues and I would enjoy the extraordinary stimulation he would bring to the department’s
intellectual environment throughout those years.
The central theme that motivates virtually all of Marvin’s work is the
overarching importance of credibility in conducting monetary policy successfully. To achieve credibility, Marvin thought it essential to
keep the Fed independent within the government so that its policy
decisions and actions were well separated from fiscal policy, Treasury
actions outside the Fed’s purview, and partisan politics. As I see it, this
theme — which was firmly aligned with Marvin’s core personal values
— was not only a guiding principle but an enabler in practice of most
of his policy positions and proposals. Credibility was not a soft concept
to Marvin but a critical precondition of effective monetary policy that
should be built into the expectations components of policy models. These views are conventional now, but Marvin embraced them
well before they became standard and argued for them relentlessly
throughout his career. More compactly, in his own words,
	Fed (and other central bank) policies only have lasting effectiveness if
the policies are credible to the public, i.e., the public is confident that
the Fed’s actions are free of political influence or manipulation and
seek consistently to advance attainment of the Fed’s central mandates of maintaining price stability and promoting maximum sustainable economic growth.2
26 |

Recollections
This theme was a natural and powerful elaboration of the policy
positions advocated by Robert Black, Richmond Fed president when
Marvin arrived in 1978. It came to encapsulate the Bank’s permanent
overall position on how the Fed should pursue its mandates

Gearing up
Marvin’s arrival required some adjustments by all of his colleagues in
the department — essentially a higher bar for our policy research and
advice to Black. Actually — and fortunately — an initial round of such
adjustments was occasioned by the arrival of Bob Hetzel about four
years earlier, fresh from writing a PhD thesis under Milton Friedman. I
will be forever grateful to Bob for helping get me ready for Marvin.
In the years following their respective arrivals, both Marvin and Bob
argued persuasively that influencing broader Fed monetary policy
positions meaningfully would require articulating our Bank’s views
more forcefully and visibly to economists — inside and outside the Fed
— focused on monetary policy. The department initiated a concerted, continuing effort to attract recent PhD graduates from university
economics departments recognized for their effective research and influence on monetary policy issues. We set an expectation that all economists would produce high-quality and relevant policy research, seek
its publication in top professional journals, and present it at influential
meetings and conferences inside and outside the Fed. We also began
to hire leading monetary and banking economists as consultants and
gathered them in Richmond for several weeks during the summer.
Through seminars, lunch roundtables, and office visits with our economists, the consultants strengthened our research and broadened our
contacts in the profession.3 (They also helped Marvin grow and thrive
in Richmond, I believe, despite the absence of the richer economic
2
3

 oodfriend (2010).
G
Many prominent economists participated in this program including Bennett McCallum, Robert King, and Douglas Diamond.

| 27

Broaddus
research environments enjoyed by Reserve Banks in larger cities such
as Boston, New York, and Chicago.)
It took some time to build, but over the course of the late 1970s and
1980s, a critical mass of research bearing the Richmond Fed imprimatur was produced and the Bank’s influence and recognition in the
policy arena increased. This reflected in significant part Marvin’s own
research since his arrival, and it provided a suitable platform for what
was to come.

Research and Preparation for FOMC Meetings
While the department’s policy team recognized the need to broaden the Bank’s influence, it also understood that the central channel
through which any Reserve Bank influences monetary policy is the
Bank president’s participation in meetings of the Federal Open Market
Committee (FOMC), the Fed’s principal monetary policymaking body.
Even before Marvin’s arrival, the department had developed an effective procedure for preparing the president for these meetings including a “pre-FOMC” briefing attended by the president and the full policy
team late in the week before a meeting and a subsequent final briefing
on Sunday afternoon. The Sunday briefing was attended by the president and the research director — traditionally the president’s principal
advisor, who typically attended FOMC meetings with him — and two
or three other senior members of the policy team.
Over the years, as he gained greater experience with the FOMC,
Marvin raised the level of these preparatory meetings substantially,
especially after he became my principal advisor when I was appointed Bank president in 1993. Team members presented high-quality
memos on various topics relevant to the upcoming meeting. Some of
these memos addressed matters expected to be the principal focus of
the meeting. Some of the most valuable, however, provided broader
relevant background on issues like the inflation process, operating procedures used in conducting monetary policy, and labor and financial
market conditions.4 I attended these pre-FOMC briefings throughout
4

28 |

The department’s distinguished historian of economic thought, Tom Humphrey,

Recollections
my term as president and benefited greatly from them not only for
their content, but also because the interaction reinforced my relationships with individual members of the Research Department and the
policy team. These initial stages of our FOMC preparations were central
to our ability to participate effectively in FOMC meetings and constructively influence their outcomes.
The heart of our Bank’s policymaking process occurred after the
pre-FOMC briefing, culminating in the Sunday afternoon session.
During my tenure, Marvin led these meetings. Before the meeting, he
would prepare a draft proposal for the statement I would make at the
upcoming FOMC meeting. This statement summarized our Bank’s view
of the economic outlook and our recommendations for the policy that
would be implemented following the meeting.5 The team members
and I would then discuss Marvin’s draft in detail and hone it to a “final”
product.6 The discussions were lively and expert. By the latter stage of
my tenure, Marvin had established himself as a widely respected macroeconomist and constructive critic of Fed policy. He spent substantial
time drafting the statement and did not readily agree to modifications.
I, in turn, pushed for modifications I thought were necessary for me
to present our positions comfortably and effectively to the FOMC. A
typical exchange went like this:
 roaddus: Marvin, I think we'd be more persuasive if we said such
B
and such rather than what you have.

often
provided relevant insights from the broader, long established economics
literature. Especially important contributions were made in the pre-FOMC briefings
by Tim Cook. Tim retired shortly after I became Bank president in 1993. Before that, he
worked closely with me, Marvin, and others keeping Black abreast of developments
regarding the Fed’s operating procedures for implementing monetary policy —
particularly the Volcker “monetarist” policy framework inaugurated in October 1979.
Tim wrote or coauthored several important papers on the relationships between Fed
policy actions and financial markets, and he had a significant influence on Marvin’s
and my thinking about operational policy issues.
5
In this period, participant statements in FOMC meetings tended to be scripted to a
greater degree than in the post-Greenspan era.
6
In addition to Marvin and me, team members who participated regularly in the
Sunday meetings included Bob Hetzel, Roy Webb, and Jeff Lacker. Alex Wolman
and Andreas Hornstein also attended during parts of the period as did Mike Dotsey
before his departure to the Philadelphia Fed.

| 29

Broaddus
	Goodfriend: You can make that change if you want to, Al, but it will
gut the whole point I'm trying to make.

After some good-natured but serious back and forth, we’d reach an
agreement Marvin could live with. These brief but sometimes intense
fine-tunings played to our respective strengths and, in my view, contributed greatly to our effectiveness in the FOMC meeting “go-around”
discussions. They were reinforced by innumerable one-on-one discussions with Marvin, during walks in Richmond or Washington or late at
night in one of our offices, where we hammered out our joint positions
on core monetary policy issues.
Following the FOMC meeting in Washington, Marvin would grade
my performance on the drive back to Richmond — always fairly, but as
anyone who knows Marvin would expect, no punches Marvin felt were
needed were pulled. There’s no question in my mind, though, that
these critiques served me well and elevated my ability to represent the
Bank and present its positions effectively.

Making the case to the FOMC — some examples
Against this background, the following sections describe several
experiences over about a quarter-century that I recall especially well,
where Marvin and our Richmond policy team sought to convey some
of Marvin’s core policy themes persuasively to the FOMC and the
broader Fed’s policymaking staff.7
a. Policy Transparency and Marvin's Secrecy Paper
The first of these experiences was the writing and eventual publication
of Marvin’s seminal paper on transparency, “Monetary Mystique: Secrecy and Central Banking.”8 If memory serves, Marvin began thinking
about this paper not long after arriving in Richmond. As many readers
are probably aware, early drafts of this paper made some people in the
 ther team members such as those mentioned in the preceding footnote also made
O
important contributions, but Marvin was the dominant and unifying force.
8
Goodfriend (1986). Lars E.O. Svensson covers the paper in detail in his essay in this
volume.
7

30 |

Recollections
Fed uncomfortable, not least in Richmond. At the time, the cultural
consensus inside the Fed was that it was inherently and continuously
exposed to often politically motivated external scrutiny and attack.
Many felt the wagons needed to be circled pretty much all the time.
Moreover, it was generally recognized that responsibility for protecting the Fed and the part of the public interest the Fed served resided
primarily with the Fed Board of Governors in Washington.
Marvin’s paper challenged important aspects of this consensus
when an early version was sent to the Board for review. In brief, the
1979 Merrill lawsuit forced the Fed — more precisely the FOMC — to
provide an explicit defense of its routine delay (i.e., secrecy) in releasing policy directives after FOMC meetings.9 Marvin scrutinized this
defense rigorously using relevant tools of economic analysis, including
rational expectations. His paper caused discomfort, I think, because
the FOMC’s defense of its secrecy took the form of affidavits submitted
during the litigation by one of the Board members and senior staff at
the Board. Therefore, Marvin was directly (and potentially publicly)
challenging these statements and their authors, albeit in a balanced
and professional manner. Board positions had certainly been challenged before, by the St. Louis Fed in particular, but not as directly and
forcefully by us.
Over time — a fair amount of it, actually — things worked out.
Marvin never wavered, and the department’s and the Bank’s leaders
consistently supported making the paper available to the public. It was
published in 1986 in the Journal of Monetary Economics and played a
significant role in the advancement of transparency in Fed monetary
policy. It also permanently set a higher standard for our Bank’s effort to
contribute meaningfully to monetary policy.
b. Preemptive Policy in 1994
A second experience Marvin and our policy team shared was our
especially active participation in FOMC meetings in 1994. This was an
eventful year for the FOMC and for us. It was my first year as a voting
9

Federal Open Market Committee v. Merrill, 443 U.S. 340 (1979).

| 31

Broaddus
FOMC member and Marvin’s first accompanying me to Washington as
my official advisor. That last point might suggest to some that Marvin
reported to me. But most readers of this article are probably aware that
on matters of monetary policy, in most respects effectively, I reported
to Marvin. In any case, I was now positioned to present and advocate
Marvin’s views, often edited by me to soften their hardest edges to
make them more palatable to my colleagues at the FOMC table. My
statements typically included Marvin’s thoughts, along with my own,
on immediately current policy issues, but also Marvin’s longer-term
core principles summarized above. He sat directly behind me, and I felt
a strong need to convey these principles accurately, frequently, and
convincingly.
Several issues arose during the year. The first was whether the Fed
would act promptly and with sufficient force to preempt any material increase in inflation or, equally importantly, emerging inflation
expectations as the economy completed its recovery from the 199091 recession. Marvin had been greatly impressed by the preemption
of inflation that Chair Volcker had overseen in 1983 and 1984, when
for the first time the Fed had increased its policy federal funds rate
materially without a sustained prior increase in inflation. With longterm Treasury bond rates rising since October 1993, signaling a rise in
inflation expectations, Marvin wanted a preemptive encore in 1994. To
that end, we argued for relatively aggressive tightening at each meeting throughout the year, which occurred, although not as aggressively
as we wanted at the September meeting, when I dissented for the first
time.10
Marvin believed that the FOMC’s preemptive policy rate increases
in 1994 anchored inflation expectations in the US and prevented the
increase in actual inflation that appeared possible, even likely, at the
beginning of the year. This was a substantial accomplishment for the
10

32 |

T he federal funds rate rose from approximately 3 percent to approximately 6 percent
over the course of 1994.

Recollections
FOMC, and I believe that Marvin's advocacy of this strategy, through
my FOMC statements and other channels, played an important role in
making it happen.
The year 1994 was also the year in which, at Marvin’s instigation, we
argued strongly against Fed involvement in Treasury initiatives to lend
money to Mexico to assist the country in dealing with its peso crisis
and prevent Mexico’s problems from destabilizing broader international financial markets.11 While persuasive arguments were made supporting such actions, Congress had explicitly declined to authorize them,
and Marvin was appropriately concerned that Fed involvement would
threaten the Fed’s independence. Later in the year Marvin pushed me
to dissent against renewal of the Fed’s foreign exchange swap lines
because they facilitated foreign exchange market intervention, which
he felt undermined monetary policy credibility. He also believed they
didn’t work well, as illustrated especially clearly by an unsuccessful
joint intervention with several countries in June of that year to support
the dollar. Marvin insisted that I repeat the language of that dissent
verbatim, annually, for the remainder of my time at the Fed. I would
object, annually, that the FOMC had already heard it more than once.
With a hint of annoyance, he would tell me, in effect, to “play it again,
Sam.”
c. What Assets Should the Fed Buy if Treasury Bonds are in Short Supply?
In 2001 and 2002, not long before both Marvin and I left the Fed, tax
revenues arising from the late 1990s technology boom produced
federal budget surpluses. Remarkable as it seems now, Treasury debt
outstanding was declining and the FOMC began to worry about
what it would do in the event there were insufficient Treasury securities available for the Fed’s routine open market purchases. An FOMC
subcommittee led by Don Kohn and Peter Fisher suggested several
potential ways to address the problem.
11

 ur broader views on the Fed’s foreign exchange operations are presented in GoodO
friend and Broaddus (1996).

| 33

Broaddus
As he made clear in his proposal later for an “Accord” on Federal
Reserve credit policy, Marvin believed passionately that the Fed should
avoid putting private assets on its books.12 Consequently, faced with
the prospect of a shortage of bonds, at the January 2001 FOMC meeting I summarized in detail Marvin’s proposal for having the Treasury
issue additional bonds to keep the Fed’s operations “Treasuries Only.”
Chair Greenspan, true to his jazz band accounting experience, wanted
to know what the Treasury would do with the proceeds of such bond
issuance.13 Wouldn’t the Treasury then have to buy private assets?
I mumbled something; Greenspan repeated the question. Marvin,
clearly alarmed that I was about to blow the opportunity, raised his
hand and asked if he could respond. I said “sure,” but I wasn’t sure what
the chair would say. In almost 30 years of attending FOMC meetings,
this was the only time I ever saw a Reserve Bank advisor intervene in
an FOMC meeting without an invitation to do so. I suppose I should
have been embarrassed since it revealed who was driving policy in
Richmond, at this meeting certainly. But I was proud of Marvin, and the
point that needed to be made was made — that the Treasury would
either need to buy private assets or the proceeds would have to be
eliminated by increasing government spending or reducing taxes. I
was also impressed that Greenspan showed no displeasure at the deviation from protocol and had a brief but natural exchange with Marvin.

Recollections
At this point, naturally enough, questions arose. First, what should
the ultimate numerical objective for the desired steady-state inflation
rate be? Second, and related, with inflation now persistently at historically low levels with market interest rates trending downward, was
there a level below which a further decline in inflation might harm the
economy?
Attention thus turned to the idea that the Fed should consider
setting an explicit numerical inflation target.15 Not surprisingly, Marvin
viewed inflation targeting favorably as a way of reinforcing a central
bank’s credibility for low inflation, and he pushed me to indicate our
Bank’s support for the concept when I became an FOMC member/participant in 1993. When the idea gained traction among some members
of Congress, Chair Greenspan asked then Fed Governor Janet Yellen
and me to lead a discussion of the pros and cons of inflation targeting
at the FOMC meeting in January 1995.
In the week preceding this meeting, Marvin enthusiastically drafted
the case for targeting that I delivered at the meeting. There, I summarized the benefits that a credible Fed precommitment to low inflation
via a target would foster, most notably higher economic growth and
employment. In particular, I argued that the greater credibility resulting from a target would reduce the sacrifice ratio:

d. The Initial Inflation Targeting Debate in the FOMC, 1995-9814
As indicated above, the progress toward bringing inflation down
began with Chair Volcker’s decisive actions in the early 1980s. While it
took well over a decade, by the late 1990s there was at least an implicit
consensus within the Fed that we were in the neighborhood of “price
stability.”

	What it would do – and this is probably the most important thing I’ll
say today – is discipline us to justify our short-term actions designed
to stabilize output and employment against our commitment to
protect the purchasing power of our currency.16

I nflation targets had been established in New Zealand in 1990, Canada in 1991, the
UK in 1992, and subsequently in many other countries.
16
FOMC Transcripts, January 31-February 1, 1995, pp. 39-41.

15

 oodfriend (2008).
G
Greenspan was a musician in Henry Jerome’s dance band for several years early on.
He also kept the band’s financial books and helped other band members with their
taxes.
14
This section closely follows Marvin’s description of the early debate regarding inflation targeting in the FOMC in Goodfriend (2010), pp. 17-23.
12
13

34 |

| 35

Broaddus
Governor Yellen made the case against a target, which she believed
would downgrade the Fed’s high employment goal.17 She also doubted that a target would increase the Fed’s credibility and thereby reduce
the sacrifice ratio, i.e., the loss in jobs and output required to resist
short-run increases in inflation.
In the discussion that followed, the committee was about evenly
divided on the desirability of a target. A subsequent discussion at the
July 1996 meeting, however, briefly encouraged Marvin and me at
one point when it appeared that the committee was approaching a
consensus on “holding the line” at the then current 2 percent inflation
rate (as measured by the Fed’s preferred personal consumption expenditures (PCE) price index), which would have locked in the significant
reduction in inflation already achieved. Our hopes were dashed when
the discussion ended without an explicit recognition of the progress
just achieved toward consensus on the issue. Marvin was especially
disappointed.
With Marvin’s encouragement, I again proposed an inflation target at
the February 1998 FOMC meeting. At this point, the core PCE inflation
rate had declined to below 2 percent and concerns about “unwelcome
disinflation” and the risk of deflation had begun to arise. With this in
mind, we argued for an explicit lower bound for any inflation target
that might be put in place. This was a precursor of Marvin’s subsequent, influential work on conducting monetary policy at the zero
lower bound (ZLB).18
I n my opening statement favoring a target, I recommended defining it in the terms
of the earlier Neal Amendment, which omitted a numerical target and defined price
stability as a condition where expectations regarding future inflation do not play a
significant role in economic decisions. My hope was that softening the proposal in
this way would increase its acceptability to the committee. My recollection is that
Marvin agreed with this strategy but with reservations. Yellen, in contrast, was arguing specifically against any proposal that made price stability the sole objective of
policy, which appeared to include any proposal that included a numerical target.
18
The FOMC adopted an explicit 2 percent inflation target in 2012, initially as a standard point target. In August 2020, the committee modified the target to an “Average
Inflation Target.” See Williams (2021).
17

Recollections
e. Monetary Policy at the ZLB
As just discussed, the almost 20-year effort to achieve credibility for
low inflation appeared to bear fruit in the late 1990s. Attention began
to turn — slowly at first — to the challenges posed for monetary policy
by persistently low inflation and short-term interest rates approaching
the ZLB. Marvin was at the forefront of path-breaking research about
these challenges and ways to deal with them.19
In October 1999, at a Fed conference in Woodstock, Vermont, Marvin
presented the results of his initial research in this arena, “Overcoming
the Zero Bound on Interest Rate Policy.”20 In the paper, he described
three approaches to retaining the ability of monetary policy to cushion a deflationary downturn in the price level and economic activity:
a "carry tax" on cash balances and two forms of "quantitative" policy
actions, either large-scale purchases of longer-term government securities (or similary illiquid private assets), or direct transfers of money to
the public. At the January 2002 FOMC meeting, Marvin was invited to
summarize his two quantitative policy alternatives. Both alternatives
derive from the idea that long-term government securities offer what
Marvin calls “broad liquidity” services in that they can be converted to
liquid assets or used as collateral to borrow liquidity. Consequently,
Marvin argued, the Fed (or any central bank) can affect the economy
by purchasing broad liquidity assets and thereby affecting their yield.
While I did not participate directly in Marvin’s preparation for this
meeting, I was present and listened attentively to his presentation and
the discussion that followed. Participants generally indicated interest
in his results but seemed to view them as preliminary and academic
rather than of immediate relevance to the policy issues of the day.
Their relevance soon became apparent, however, during the financial
crisis of 2008-09 and its aftermath, as Marvin’s ZLB research provided a
starting point for the Fed and other central banks as they confronted
the radically different policy environment that emerged following the
crisis.
19
20

36 |

F or an engaging summary of Marvin’s contributions in this area, see Williams (2020).
Goodfriend (2000). Ben Bernanke covers the paper in detail in his essay in this volume.

| 37

Broaddus
Against this background, it is interesting and satisfying to consider
Marvin’s views regarding the role of Fed credibility in conducting policy at the ZLB, thus returning full circle to his core credibility principle
discussed at the beginning of this article.21 As Marvin indicates, the
quantitative policy approaches described above are likely to increase,
significantly, the public debt and the monetary base. Therefore, in resisting excessive disinflation or a deflation, the Fed may create a risk of
a rapid reemergence of inflation. Its willingness to mount a successful
defense against deflation, then, requires that it sustain, permanently,
its credibility against inflation. As Marvin put it to me (and to many others), full credibility against deflation requires continuing full credibility
against inflation. It was at this moment, late in my career at the Fed,
that I grasped fully the comprehensive power of Marvin’s credibility
principle for monetary policy.

Recollections
it available to a broad international audience. Most importantly, I hope
that what I’ve summarized here will confirm again what I and many
others, inside and outside the Fed, have long believed: that among
those who have labored to improve the conduct of American and
global monetary policy, Marvin Goodfriend is a giant.

Concluding comments
Hopefully the preceding discussion has conveyed adequately the
range and depth of Marvin’s contributions to Federal Reserve monetary policy and central banking more broadly. All of us who worked
with Marvin at the Richmond Fed are proud that our Reserve Bank was
the setting for much of his most important research. We are also grateful for what he did to raise the standing of our Bank and enrich our
individual careers. By the same token, I know also from conversations
with Marvin that he greatly appreciated the supportive research environment he enjoyed in Richmond just as he appreciated Chair Greenspan’s willingness to foster a collaborative culture across the Federal
Reserve, understanding that each strengthened his research and made

21

38 |

See especially Goodfriend (2016).

| 39

Broaddus

References
Goodfriend, Marvin. 1986. “Monetary Mystique: Secrecy and Central
Banking.” Journal of Monetary Economics 17, no. 1 (January): 63-92.

Recollections
Williams, John C. 2021. “The Theory of Average Inflation Targeting.”
Remarks at Bank of Israel/CEPR Conference on “Inflation: Dynamics,
Expectations, and Targeting,” July 12.

Goodfriend, Marvin. 1997. “Monetary Policy Comes of Age: A 20th Century Odyssey.” Federal Reserve Bank of Richmond Economic Quarterly
83, no. 1 (Winter).
Goodfriend, Marvin. 2000. “Overcoming the Zero Bound on Interest
Rate Policy.” Journal of Money, Credit and Banking 32, no. 4, Part 2
(November): 1007-35.
Goodfriend, Marvin. 2008. “We Need an Accord for Federal Reserve
Credit Policy.” Paper prepared for a Shadow Open Market Committee
Symposium, Cato Institute, April 24.
Goodfriend, Marvin. 2010. “Policy Debates at the FOMC: 1993-2002.”
Paper prepared for the Federal Reserve Bank of Atlanta-Rutgers University Conference “A Return to Jekyll Island: The Origins, History and
Future of the Federal Reserve,” November 5-6: 1-2.
Goodfriend, Marvin. 2016. “The Case for Unencumbering Interest Rate
Policy at the Zero Bound.” Paper presented at the Federal Reserve Bank
of Kansas City Symposium on “Designing Resilient Monetary Policy
Frameworks for the Future,” Jackson Hole, Wyoming, August 25: 127160.
Goodfriend, Marvin, and J. Alfred Broaddus Jr. 1996. “Foreign Exchange
Operations and the Federal Reserve.” Federal Reserve Bank of Richmond Economic Quarterly (Winter): 1-19.
Federal Open Market Committee v. Merrill, 443 U.S. 340 (1979).
Federal Open Market Committee. 1995. “Meeting of the Federal Open
Market Committee, January 31-February 1.” pp. 39-41.
Williams, John C. 2020. “Research, Policy and the Zero Lower Bound.”
Remarks at the Shadow Open Market Committee Spring Meeting, New
York City, March 5.
40 |

| 41

New Fed Framework

The Federal Reserve’s New Monetary
Policy Framework
Donald Kohn
I very much appreciate the opportunity to participate in this volume
honoring Marvin Goodfriend’s contributions. Marvin and I interacted
often as we both served in the Federal Reserve System, comparing
notes on developments in macroeconomics and monetary policy at
numerous conferences and in informal contacts along the margins
of Federal Open Market Committee (FOMC) meetings and elsewhere.
Marvin had a huge influence on the study and practice of monetary
policy within the Federal Reserve and more widely. Even where his
ideas took time to filter into actual policy, he often framed the debate,
forcing the skeptics to examine their arguments more closely.
For many years, I was one of those skeptics when it came to explicit inflation targeting. In fact, conference organizers liked to position
Marvin and me as the pro and con on this topic. While I shared Mavin’s
objective of anchoring inflation and expectations around 2 percent,
I saw advantages in keeping the goal implicit rather than explicit;
expectations were becoming anchored at close to 2 percent in any
event, and an implicit target might afford greater flexibility to respond
to some types of shocks.
These debates were great learning opportunities for me. Marvin
marshalled empirical evidence and embedded that evidence in the
theory and practice of central banking over history. Marvin was open
and honest about his views and the supporting evidence. And however much you might have differed, you couldn’t doubt his focus on and
devotion to bringing his considerable intelligence and deep learning
to serving the interests of the Federal Reserve and the United States.
In the end, Marvin’s analysis prevailed. In the aftermath of the financial crisis, I was won over to the view that the benefits of an explicit
inflation target would exceed its costs. In 2012, after I left, the Fed
adopted a 2 percent target.
42 |

| 43

Kohn
My conversion and the explicit target emerged from the threat of
the Federal Reserve missing both its employment and inflation targets
on the low side, rather than from building the bulwark against high
inflation that mostly animated Marvin’s advocacy over the years. The
major risk to inflation expectations as the country slowly recovered
from the global financial crisis of 2007-09 was that they would fall below 2 percent, reducing nominal rates and limiting the scope for policy
easing in the future. I saw an explicit target as helping gain support
for additional monetary policy action from the members of the FOMC
who feared that unconventional policies might cause much higher
inflation down the road.
But the essence of Marvin’s vision has been realized. The Federal Reserve has made an explicit public commitment to achieving 2 percent
inflation over time, which should help discipline policy and firm up
expectations against deviations from the target in either direction.
Importantly, the inflation target that Marvin advocated for rested on
two closely related pillars of his analysis. First, that economic welfare
was fostered by effective price stability and by public expectations that
prices would remain stable (avoiding “inflation scares”). Second, that
those expectations would be more durably anchored, and democratic
accountability better served, by central bank transparency about its
targets and its plans for meeting them.
Clearly the Federal Reserve has embraced both of those propositions. In addition to the explicit inflation target, policymakers have
taken a number of steps in recent decades to be more open about
their analysis and rationale for policy. In his essay introducing Marvin’s
paper on “Monetary Mystique” in this volume, Lars E.O. Svensson outlines what the Federal Reserve has done to realize Marvin’s objective
of transparent policymaking. The major actions include: (1) announcing policy decisions immediately (1994); (2) publishing quarterly the
projections of FOMC participants for output growth, inflation, the
unemployment rate, and the appropriate path of the federal funds rate
to achieve the FOMC’s legislated objectives for “maximum employment and stable prices” (2008-2012); and (3) adopting and publishing a
“Statement on Longer-Run Goals and Monetary Policy Strategy” (2012)
44 |

New Fed Framework
that outlines the FOMC’s view of its objectives and how it intends to
pursue them.
Goals may be largely fixed by legislation, but strategies and the communication around them need to adapt to changing circumstances.
In recent years — before the Covid pandemic — the Federal Reserve
had been wrestling with how to adapt its targets, strategy, and transparency to a world in which central banks, including the Fed, struggled
with getting inflation up to the 2 percent target in an environment of
persistent disinflationary pressures. Those disinflationary forces were
marked by weak demand, the effects of globalization and technology
on costs, and surprisingly muted responses of inflation to low unemployment rates. During this time, very low equilibrium nominal interest
rates raised pressing questions about whether existing policy strategies could be consistently successful in achieving the Fed’s legislative
objectives when the zero lower bound (ZLB) on rates could frequently
limit the scope for easing policy in response to negative demand
shocks.
The constraint on policy easing presented by the ZLB creates a potential asymmetry toward missing both the inflation and employment
goals on the low side on average over time if, as in the strategy adopted in 2012, policy is always aiming just at its 2 percent inflation target.
That bias may not be reliably overcome using unconventional policy
measures, like asset purchases and forward guidance. In fact, inflation had persistently fallen short of the Fed’s goal in the decade from
2009 through 2019, despite interest rates at zero and substantial asset
purchases over much of that period. Moreover, inflation misses on the
low side occurred even with unemployment rates that had declined to
much lower levels than previously thought consistent with maintaining price stability. To be sure, the inflation misses were generally small,
but they also were accompanied toward the end of this period by a
downward drift in some measures of inflation expectations, raising
questions about whether these expectations would continue to be
anchored around the 2 percent target.
In response to this experience, the Federal Reserve ran a very public and transparent process to assess how it should alter its monetary
| 45

Kohn
policy framework — its monetary policy strategy, tools and communications — to raise the odds on achieving its legislated price stability
and maximum employment objectives more consistently in this low
natural rate environment. It announced the results in August 2020 in
a revised version of its “Statement on Longer-Run Goals and Monetary
Policy Strategy” and in a speech by Chair Powell that explained the
changes and their rationale.2
Lars E.O. Svensson touches on the new framework at the end of his
essay. In this piece I will dig a little deeper, evaluating it through the
prism of the two pillars of Marvin’s work previously cited — sustaining price stability and being very transparent about how that will be
achieved.

The new framework
With respect to the objectives of monetary policy, the new framework retains the critical elements of the old framework. It kept the 2
percent inflation target as its definition of its price stability mandate.
With respect to maximum employment, it continued to acknowledge
that specifying an explicit numerical goal is unwise because the level
of maximum employment consistent with stable prices is not directly measurable, is not under the control of the Federal Reserve, and
changes over time for reasons unrelated to monetary policy.
2

46 |

See https://www.federalreserve.gov/monetarypolicy/review-of-monetarypolicy-strategy-tools-and-communications.htm for an explanation and background
material. The following description is based on the “Statement of Longer-Run Goals
and Monetary Policy,” https://www.federalreserve.gov/monetarypolicy/
review-of-monetary-policy-strategy-tools-and-communications-statement-onlonger-run-goals-monetary-policy-strategy.htm, and on Chair Powell’s speech introducing the new framework, https://www.federalreserve.gov/newsevents/speech/
powell20200827a.htm. At its meeting on January 25, 2022, the FOMC unanimously
reaffirmed the statement adopted in August 2020, https://www.federalreserve.gov/
newsevents/pressreleases/monetary20220126b.htm.

New Fed Framework
But responding to the 2009-19 experience, the FOMC made some
key changes in the specification of the maximum employment goal
and the strategy for achieving its objectives. The previous statement
hadn’t defined maximum employment, but many observers, including
many members of the FOMC, gave heavy weight to the unemployment rate and looked at the historical relationship of this variable to
changes in inflation to gauge how close the economy was to this goal.
Reflecting this approach, the old statement gave the median of FOMC
members’ recent estimates of the normal long-run rate of unemployment as an example of a measure of maximum employment. Because
the history of this relationship had not been a good guide to future
inflation in recent years, the new statement dropped this reference to
the unemployment rate and added that “the maximum level of employment is a broad-based and inclusive goal.” Although the old statement said that the FOMC looked at a “wide range of indicators” of labor
market tightness, the new statement and its exposition by a number of
FOMC participants has seemed to suggest not only a de-emphasis of
the unemployment rate, but also increased attention to a wider array
of other indicators, including labor market outcomes across population subgroups.
In addition, the monetary policy strategy for dealing with labor
markets was altered in an important way. Policy would take account
of shortfalls from maximum employment (e.g., unemployment rates
above the estimated normal level), but not necessarily of estimated
overshoots (e.g., unemployment rates below the estimated normal
level).3 That’s because in the 2009-19 period policy had been tightened
in low unemployment periods to head off inflation, but experience had
been that inflation would be quiescent at much lower levels of unem3 ”… the Committee’s policy decisions must be informed by assessments of the shortfalls of employment from its maximum level…” The old statement had said that policy
would be informed by “assessments of the maximum level of employment,” implying attention to both sides of the level. And later in the new statement: “In setting
monetary policy, the Committee seeks to mitigate shortfalls of employment from
the Committee’s assessment of its maximum level and deviations of inflation from its
longer-run goal.”

| 47

Kohn
ployment than had been expected. The FOMC continued to acknowledge that policy affected employment and inflation with a lag, but it
would not run a tight policy to preempt projected inflation overshoots
based only on actual and projected labor market conditions. Tight
policy — interest rates being moved above the estimated neutral rate
— would depend on already seeing unsatisfactory inflation outcomes.
On the inflation goal itself, the FOMC’s new statement emphasized
the importance of keeping inflation expectations well-anchored at its
goal of 2 percent, but it worried that periods of below-target inflation
would be more prevalent than above-target inflation given the ZLB
problem, which would tend to pull expectations under 2 percent. To
avoid this outcome, it would now seek to achieve inflation that averaged 2 percent over time. That means that when inflation has been
running below 2 percent, monetary policy will aim to achieve inflation
“moderately above 2 percent for some time.” This has been labeled
flexible average inflation targeting, or FAIT.
Sadly, we cannot know what Marvin would have thought about
the new framework. I suspect he would have been very pleased with
how the review was conducted: the process for arriving at the new
framework and statement was announced ahead of time; the Fed held
public “FedListens” sessions to get input from the public and from academics; it reported on the progress of its deliberations in the minutes
of the FOMC; some members of the FOMC used speeches to keep us
informed about the evolution of their own thinking; and, simultaneously with the rollout of the new framework, it published the staff
analysis that the FOMC had as it considered its options. I also suspect
he would have liked the emphasis on keeping inflation expectations
anchored at the 2 percent target and the explicit rejection of raising
the target, as some academics had been suggesting.4
4

48 |

F or example, see the paper by Eberly, Stock, and Wright at the FedListens conference
in June 2019, https://www.federalreserve.gov/conferences/conference-monetarypolicy-strategy-tools-communications-20190606.htm.

New Fed Framework
Beyond these key elements I will not try to guess at how Marvin
would have reacted. Below, in examining the new strategy, I will
channel the principles he imbued in the Federal Reserve for securing
price stability and enhancing transparency. But I know that my analysis
would have been much stronger had I been able to benefit from the
give and take with Marvin that was so important to the evolution of
my thinking and that of countless others at the Fed.

Anchoring inflation expectations at 2 percent
As noted, the new framework grew out of a period in which inflation
fell short of the 2 percent target and inflation expectations drifted
down, despite very low interest rates and much lower unemployment
than had previously been thought consistent with low, stable inflation. Keeping expectations from moving below 2 percent is especially
important when the real equilibrium interest rate appears also to be
quite low, making the ZLB an increasingly salient policy constraint.5
The FAIT framework is well designed to counter the disinflationary
bias imparted by policy being constrained by the ZLB from time to
time. FAIT promises to make up for inflation below 2 percent by aiming
to run it “moderately above 2 percent for some time” — a flexible form
of price-level targeting. That implies easier policies for longer than if
the Fed were simply aiming to return inflation to 2 percent without
the makeup. The point of the averaging is to make sure expectations
are indeed anchored at 2 percent. In effect, deliberately aiming for
inflation to exceed 2 percent for some time and likely allowing the
overshoot of maximum employment necessary to achieve that results
in an upward inflation bias that offsets the downward bias arising from
the ZLB.

5

S everal reasons have been given for a very low natural real rate of interest (r-star).
They include weak investment demand in a more service-dominated economy with
slower growing populations; increased saving as populations age, as governments
of developing countries accumulate reserves as a precaution against sudden stops
of capital flows, and as people downshifted consumption in the wake of the global
financial crisis; and slowed productivity growth after 2005. The r-star estimates of the
Laubach-Williams model are given at https://www.newyorkfed.org/research/policy/
rstar.

| 49

Kohn
But there are other asymmetries in the new framework — beyond
the makeup for actual undershoots — that also lean toward taking upside risks on inflation and raise questions about how well adherence to
the framework would anchor expectations in circumstances in which
inflation wasn’t so quiescent.
One key problem is that the strategy does not address what is to
happen if there are persistent overshoots of the 2 percent target. It
reaffirms that “the Committee seeks to achieve inflation that averages
2 percent over time” but follows that statement with the aim of making
up for undershoots and doesn’t address the opposite situation. Making up for overshoots would be required if persistent inflation over 2
percent resulted in longer-term inflation expectations rising above 2
percent. It would involve difficult economic decisions as it implies a
need to deliberately run the economy below its sustainable potential
for a time to lower inflation and inflation expectations.
One suspects the FOMC might opt for “opportunistic disinflation”
in such circumstances — waiting for an external shock, rather than
monetary policy, to create slack and lower inflation. That was the view
of many members of the FOMC in the 1990s when inflation exceeded
2 percent, but at that time the FOMC was working in the context of an
implicit target, not an explicit one, which afforded a greater degree
of flexibility. In any event, a framework that addresses making up for
undershoots but not overshoots would seem to risk a bias toward inflation over 2 percent if circumstances differ materially for a time from
the disinflation pressures of 2009-19.6
 t his press conference of January 26, 2022, Chair Powell appeared to confirm the
A
one-sided character of the make up in the new framework — though still with some
ambiguity since he also emphasizes having inflation average 2 over time:
	MICHAEL MCKEE: ……ask you, as you start to reverse policy, what your goal is.
Are you going to be raising interest rates until you get inflation to 2 percent? Do
you want to go below 2 percent so that, on average, you get a 2 percent inflation
rate? ….
	CHAIR POWELL: So, no. There’s no — there’s nothing in our framework about having
inflation run below 2 percent so that we would do that, try to achieve that outcome.
So the answer to that is, is “no.” What we’re trying to do is get inflation, keep inflation
expectations well anchored at 2 percent. That’s, that’s always the, the ultimate goal.
And we do that in the service of having inflation — we get to that goal by having
inflation average 2 percent over time.

6

50 |

New Fed Framework
A second upside inflation bias arises from the asymmetry of the
response to deviations of labor markets from estimates of maximum
employment. Shortfalls of employment from estimated maximums
weigh on the side of accommodative policy, but actual or projected
overshoots, by themselves, do not call for tight policy. In the past, the
FOMC had increased the federal funds rate by enough to head off
possible future inflation when projections suggested that declines in
the unemployment rate in the absence of tightening were likely to
result in future above-target rates of inflation — even before inflation
or inflation expectations had risen into unacceptable territory. Lags in
the effects of monetary policy made such preemptive moves desirable
to avoid having to impose future output losses to bring inflation back
down. In the new framework, the FOMC will continue to get ahead of
unwelcome declines in inflation by running accommodative policy
when it judges there to be slack in labor markets, but its scope to head
off future unwanted increases in inflation would appear to be more
constrained, risking overshoots of the target under some circumstances and the resulting greater variability in output.7
The choice of this asymmetric reaction function grows out of the
experience from 2009-19 when the committee had overestimated the
unemployment rate consistent with low stable inflation and in retrospect felt that had it waited to tighten more jobs would have been
created more quickly with inflation still contained. If the relationship
between slack and inflation is as attenuated as it seemed to be from
2009-19, then waiting to see actual inflation rise might not be very
costly in terms of unanchoring expectations on the upside, because
any rise in inflation would be small. But a steeper Phillips curve, say
because of greater uncertainty and eroding credibility around the 2
percent target or a larger decline in unemployment relative to the natural rate, would impart a more definite inflation bias to policy.

7

T o be sure, the new framework does not rule out tightening policy — raising the
funds rate toward its equilibrium rate — when employment is rising toward its
maximum sustainable level, but it does seem to rule out tight policy — r above r* —
unless actual inflation is already unacceptably high.

| 51

Kohn
The forward guidance on interest rates provided during the pandemic implemented this asymmetric response to labor markets in a
particularly aggressive way. In September 2020, the FOMC adopted
language that promised to keep interest rates close to zero “until labor
market conditions have reached levels consistent with the Committee’s assessments of maximum employment and inflation has risen
to 2 percent and is on track to moderately exceed 2 percent for some
time.” This meant that short-term rates would remain deeply negative
even in the run up to full employment with inflation at the target and
predicted to rise further, leading to a likely overshoot of maximum
employment and continued upward pressure on inflation. This was
more aggressively easy guidance than required by the new framework,
which only called for policy to be accommodative before the economy had reached maximum employment, but it reflected the new
emphasis on reaching full employment and not projecting increases in
inflation.
As I write in early 2022, the economy has rebounded from the initial
shock of shutdowns in the wake of the onset of the Covid-19 pandemic, inflation has surged to levels well in excess of the 2 percent target,
and the labor market has tightened considerably faster than expected.
That surge has reflected both strength in demand and Covid-related
constraints on supply.8 The situation is complex and unprecedented,
containing elements of adverse supply shocks that are always difficult
for monetary policy to navigate.
To the credit of the Federal Reserve, as the persistence of inflationary
pressures became increasingly evident, it pivoted quite rapidly toward
an accelerated removal of the extreme accommodation it had put in
place in the initial stages of the pandemic. It began to reduce its
8

52 |

 emand has been boosted by highly supportive fiscal policy, very accommodative
D
monetary policy and its associated effects raising equity prices and wealth, and
by spendable household savings accumulated from fiscal payouts and limits on
opportunities to spend on services. On the supply side, waves of Covid-19 infections have adversely affected the supply chains for the goods so much in demand.
Early retirements along with health concerns, child care and school disruptions, and
Covid-19 infections of workers and their families have limited the rebound in labor
force participation, leading to very tight labor markets.

New Fed Framework
purchases of securities sooner and more rapidly than it had previously
expected, and it is clearing the way for beginning to raise its target
interest rate in March 2022, also much sooner than it or many observers had expected as recently as summer 2021. Although measures
of short- and medium-term inflation expectations have risen to well
above 2 percent, longer-term expectations remain anchored at levels
consistent with 2 percent, suggesting the Federal Reserve retains credibility for achieving its long-run objective over time.
Still, this experience suggests that the asymmetries of the new
framework could risk an upward drift from the 2 percent inflation
anchor when the surrounding macroeconomic circumstances deviate
from the damped demand, low-inflation environment of 2009-19. The
Federal Reserve needs to consider and then spell out its approach to
achieving its price stability goal if the prior disinflationary forces do
not reemerge. To be sure, the FOMC can never anticipate and discuss
all the situations it might face in the future. But it should be able to
describe in a general way how it would expect to react if, for example,
inflation persisted above its target and expectations began to drift
higher despite estimated slack in labor markets or if wages and prices
suggest that labor markets are approaching a level of “maximum employment” that is lower than previously expected.

Transparency
As indicated by the previous discussion of asymmetries and changing circumstances, the new framework is more complex than the
old one, in which monetary policy was always targeted at 2 percent
inflation and goal conflicts posed by supply shocks were subject to
a “balanced approach” to both goals based on deviations on either
side of inflation from target and of employment from the estimate of
maximum.
Complexity challenges transparency. As Marvin emphasized, transparency is critical for policy effectiveness — the more accurately the
public can predict what the central bank is going to do, the more
rapidly the economy is likely to move toward central bank objectives.
Transparency is also critical in a democratic society for accountability
| 53

Kohn
and preserving central bank independence. The previous regime could
be approximated by relatively simple policy rules that could serve as
benchmarks and guides for the FOMC and its observers. In the new
framework, policy responses depend on how long the policy rate has
been pinned at zero and inflation has fallen short of target and whether “maximum employment” is being approached from above or below,
greatly complicating the explanation of strategy.
Although the process of coming up with and rolling out the new
framework was very open, in some respects the new framework is
a step back in transparency. To an extent, this is an understandable
consequence of the inherent complexity and lack of experience with
the new framework. And some of the lack of transparency results from
a deliberate attempt to preserve flexibility.
Yet, I am convinced that the Federal Reserve can improve the transparency of policy under the new framework.
First, as I’ve already discussed, it needs to outline how it will deal
with circumstances that differ materially from those that prompted the
rethink — for example, adverse supply shocks, persistent and material inflation overshoots, and the waning effects of the disinflationary
forces of the 2009-19 period. A crucial question is how monetary policy
strategy would evolve in the framework if the problem becomes one
of reducing inflation rather than getting it up to target.
Second, the new framework has redefined its ”maximum employment” goal in ways that are less transparent. Both the old and new
statements note that maximum employment isn’t under Fed control
but rather is “defined by nonmonetary factors that affect the structure
and dynamics of the labor market.” The previous statement, however, gave the most recent estimate of participants for the long-term
unemployment rate consistent with its mandates as a reference point.
To be sure, that estimate had to be interpreted along with other data
on labor markets, wages, and prices to get a sense of whether the
unemployment rate consistent with stable prices was shifting, but the
long-term unemployment rate provided a guidepost.

54 |

New Fed Framework
The current statement characterizes maximum employment as a
“broad-based and inclusive goal that is not directly measurable and
changes over time” and omits the reference to participants’ projections
of the sustainable unemployment rate. The FOMC has yet to define
what measures are encompassed by “broad-based and inclusive,” with
some participants emphasizing that they will be paying close attention
to the labor market experience of low-wage and minority groups without discussing how that view intersects with the FOMC’s price stability
mandate. If economic agents are to understand and accurately anticipate monetary policy, the FOMC needs to spell out more clearly what
it means by broad-based and inclusive.
Third, another aspect of the new strategy that would benefit from
additional explanation is how flexible average inflation targeting will
work in the real world. I have considerable sympathy with the “flexible”
piece of FAIT — it’s in line with my position on inflation targeting in my
discussions with Marvin that an implicit target gave the FOMC greater
scope to deal with unexpected and unusual shocks arising, for example, from financial market developments. But the FOMC has an explicit
target — 2 percent inflation over the longer run — and emphasizes
the gains for both parts of the dual mandate from having expectations anchored at that level. In the circumstances that the framework
was designed for — avoiding a downward drift in expectations from
persistent undershoots of the target — the committee has said that if
inflation did fall persistently short of target, it would “aim to achieve
inflation moderately above 2 percent for some time.”
As inflation came to exceed 2 percent over the course of 2021,
market participants were struggling to gauge the FOMC’s definitions
of “moderately” and “for some time.” I wouldn’t expect precise definitions of those words — that would take the F out of FAIT — but some
guidance would be a useful enhancement of transparency. Are there
ranges around the level of inflation or the length of its persistence
over 2 percent that would stretch the definition of these words too
far? Also, after “some time” of “moderate overshoots” would the FOMC
resume targeting 2 percent inflation, or would it be content with inflation moderately over 2 percent and count on future shortfalls at the
| 55

Kohn
ZLB to bring the average back to 2 percent? How would FAIT work over
time — what’s the end game once the average has been secured at 2
percent?

		

Conclusion
The new framework is well adapted to the circumstances that
faced monetary policy from 2009 through 2019. But it is complex and
incomplete, with a deliberate lean toward taking upside inflation risks
to offset the downside risks inherent in very low interest and inflation
rates. It has been implemented initially in a global pandemic that is
unprecedented in modern times, which has had complex and difficult-to-predict effects on aggregate demand and supply and prompted extraordinarily expansionary fiscal and monetary policies. These
forces — the virus and the policy response — have produced very high
and persistent inflationary pressures, much higher and more persistent
than the Federal Reserve or most mainstream economists had predicted or had been contemplated in the new statement. To the Fed’s credit, the FOMC has reacted relatively quickly to the emergence of those
pressures by beginning to dial back its highly accommodative policy.
The Federal Reserve has said that it will review the new framework
after five years. That review and the opportunities to clarify the framework in the next few years should be used to address the work that still
needs to be done to meet Marvin’s objectives. They were to assure that
policy would act to keep inflation expectations anchored at 2 percent
under a wide variety of circumstances and that the Fed would gain the
full benefits of transparency for policy effectiveness and democratic
accountability. We honor Marvin’s memory by trying to live up to his
high standards for analysis and policy. We can best contribute to the
public welfare by applying his ideas and identifying ways to narrow the
inevitable gap between Marvin’s standards and actual policy practice.

56 |

| 57

What Does the Fed Know? When Does it Know It?

What Does the Fed Know, and When
Does it Know It?
William Poole
Marvin Goodfriend was my student at Brown University; he was
awarded his PhD in 1980. I am pleased to contribute an essay to honor
his memory.
When I was appointed to the Council of Economic Advisers (CEA)
in 1982, I immediately reached out to Marvin and persuaded him to
spend a year on the CEA staff in 1984. During my time as president of
the Federal Reserve Bank of St. Louis (1998-2008), Marvin and I crossed
paths regularly. We often chatted briefly at Federal Open Market Committee (FOMC) meetings before he left the Richmond Fed in 2005.
Shortly after returning to the Richmond Fed from the CEA, Goodfriend began working on a paper evaluating the Fed’s case for secrecy.1
He picked apart the arguments the Fed presented as it attempted to
fend off the FOIA suit brought by David R. Merrill in 1975. That litigation found its way to the Supreme Court and back to Federal District
Court before being finally concluded in 1981. That paper of Marvin’s is
my motivation for this essay.
As St. Louis Fed president I attended every meeting of the FOMC
between March 1998 and my final one in January 2008. The committee often grappled with the wording of policy statements, as anyone
can see from reading transcripts of the meetings. For example, at the
end of my first meeting in March 1998, Ned Gramlich, a Fed governor
appointed in 1997, interrupted the roll call vote on the policy directive.
As reported in the transcript of the meeting:
 R. GRAMLICH. Do we have to have the “slightly lower” phrase? Am I
M
out of order? [Laughter]
	CHAIRMAN GREENSPAN. That is the conventional rhetoric.
1

58 |

Goodfriend (1986). See the essay by Lars E.O. Svensson elsewhere in this volume.

| 59

What Does the Fed Know? When Does it Know It?

Poole
	MR. GRAMLICH. Yes, but—
	CHAIRMAN GREENSPAN. We have been butchering the English language in this directive for years, but let’s not change it just yet. Why
don’t you bring that up at a later meeting? [Laughter]
	MR. GRAMLICH. On that advice, I vote “yes.” [Laughter]

During my entire tenure we struggled with how best to communicate with the markets and the general public. I gave several speeches
on the subject (see Poole [2003] for an example). By 1998, when I
arrived in St. Louis, the case for secrecy in the abstract was dead but
saying that does not per se make a case for disclosure. Disclosure
of what, and when? We agreed that putting the FOMC meeting on
C-SPAN made no sense because doing so would inhibit a full discussion of the policy issues the committee faced and would simply move
the real discussion into the hallways.
Most of the literature on Fed communication is normative — what
should the Fed disclose, when, and why? As illustrated by Ned Gramlich’s question, the issue was how we could best communicate the
policy issues we faced and why we made the decisions we did. Goodfriend examined the Fed’s case for secrecy; my purpose in this essay
is to approach the policy communication issues by first concentrating
on what the Fed knows and when. When does the Fed have an information advantage, or disadvantage, relative to the market? What is the
nature of the advantage, or disadvantage, and why does it matter? This
essay is more along the lines of a positive than a normative analysis of
the role of information in the Fed’s monetary policy.
Proposition 1: Absence of a systematic Fed information advantage.
The Fed has a minimal advantage, if any, in terms of knowledge of
facts. The Fed chair gets the unemployment report through the chair
of the CEA late Thursday afternoon, just a few hours before the Bureau
of Labor Statistics releases the data publicly on Friday at 8:30 a.m.
This same “advantage” of a few hours is true of all the data released by
federal statistical agencies. Today, the Fed, governments at all levels,
and private firms are wrestling with COVID-19. It is hard to believe that
the Fed has a COVID-19 information advantage over other entities; the
60 |

Fed has made this point clear on many occasions as it has discussed
the pandemic.
Federal statistical data are the raw material of macroeconomic and
financial analysis. While it is true that Fed economists can tap directly
into employees at the statistical agencies for additional insight, it is
also generally true that private economists and analysts can reach the
same experts. As a close approximation, the Fed has no advantage
whatsoever over the private market in access to data. It is laughable
to believe that the Fed has an inside track on future fiscal policy when
the federal government itself seems so obviously dysfunctional in
planning just about everything. All too much of what the federal government does is a consequence of last-minute deals brokered among
many competing interests. Relative to political experts, the Fed is probably at a disadvantage in predicting what will happen.
It is also true that certain private parties sometimes have access to
data that the Fed does not. An example is the story spun by Michael
Lewis in The Big Short.2 Lewis explains how a few hedge fund managers developed a firm conviction that many housing-related securities
would crater in a financial crisis. It is not infrequently the case that
industry experts understand developments that will make an important difference to the macroeconomic environment in the months and
quarters ahead. Another example might be the expansion of oil and
gas production through fracking. That expansion affected the outlook
for energy prices and, through those prices, the direction of the international economy. Yet another example might be the development of
COVID-19 vaccines. The companies involved understood the process
and prospects much better than the Fed. Because of insider trading
restrictions, pharmaceutical firms had to be very careful about the
release of information.
I have long held the view that the best way to think about the arrival
of new information is to assume that the market and Fed get the information at the same time. Since the rational expectations revolution in
macroeconomics in the early 1970s, it has been standard to model
2

Lewis (2011).

| 61

Poole
new information as “innovations” that push the state of knowledge one
way or the other from previous expectations for every variable in the
economy.
This view of the information process motivated my 1999 speech,
“Synching, not Sinking, the Markets.” In that paper, I argued that the
Fed ought to think of policy in the context of how to change the
federal funds rate in response to arriving information. With Fed clarity
and transparency, one of its goals should be leadership and, through
disclosure, alignment with private sector reactions to incoming information.
Proposition 2. Absence of a systematic Fed processing advantage.
Analysts must process the flow of raw data to develop useful ideas as
to what new data might mean for the direction of the financial markets
and the economy. The Fed has a very large and expert staff. Does the
Fed have a processing advantage over the private market? Can the
Fed’s experts distill knowledge out of raw data more quickly and more
accurately than the private market?
My instinct is that the Fed does not, overall, have a processing advantage. Fed economists and private economists read the same professional papers and attended the same graduate schools. They go to the
same conferences. Fed and private sector economists move back and
forth in their employment. The literature on the accuracy of economic
forecasts does not suggest that Fed forecasts are clearly more accurate
than private forecasts.3
Obviously, there are occasions when the Fed is ahead of the private
market and some when the Fed is behind. In thinking about the normative aspects of information sharing and secrecy, both the average
situation and the special cases are relevant. My general proposition is
that the Fed and the markets have the same information base and the
same ability to process that information.

3

62 |

 hang and Hanson (2015); Stephen K. McNees of the Boston Fed had earlier written
C
several papers on the topic. By now, the literature is voluminous.

What Does the Fed Know? When Does it Know It?
Obviously, not everyone in the market has the same information and
the same skills in processing information. And that is also true of Federal Reserve officials. What makes sense to me is that it is best to assume,
absent evidence to the contrary, that market prices accurately reflect
implications of available information for all the reasons discussed at
length in the rational expectations and efficient markets literature.
The Fed does have one power the market does not — the power to
print money. In the short run, the Fed can set almost any interest rate
on government securities at almost any level it wants, and the same
is true for foreign exchange rates. “At almost any level” is obviously an
exaggeration, but it is not far off if we think in terms of basis points for
a few weeks rather than percentage points for a few years. Basis points
for a few weeks are of critical importance to market speculators.
However, the Fed does not have the power to set real variables, except real interest rates temporarily, at any level it wants. This proposition goes far back in the monetary theory literature. Its corollary is that
an effort by the Fed to set real variables, such as the unemployment
rate, will fail. Moreover, a determined Fed effort to set real variables at
levels materially different from market equilibrium levels will create
large problems.

Illustrative examples
The terrorist attacks on Tuesday, September 11 illustrate these
themes. Neither the Fed nor the private market anticipated the attacks.
Here was an information innovation writ large. Because the government shut down air travel, the Fed was forced to shut down the clearing of checks. At that time, the Fed clearing process involved shipping
physical checks by air from Fed processing facilities in Reserve Banks
and their branches across the country to Reserve Banks and then by
van to banks on which checks were written. Commercial banks make
payments from accounts that have received funds. With no funds coming in, banks could not honor checks written on accounts on which
firms were making payments. The Fed made clear to banks that they
could borrow from the Fed whatever amounts they needed to be able
to honor checks on September 12 and later.
| 63

Poole
The Fed’s power to print money prevented the terrorist attacks from
creating a financial disaster. No private entity could have done what
the Fed did. A similar and more familiar story arose after the failure of
Lehman Brothers in September 2008. The Fed established several special facilities, and expanded others, to provide funds in vast amounts
to the markets. The Fed does have powers that permit it to respond to
information in ways the market cannot. That said, the Fed’s powers are
not indefinitely large.
There is a flip side to these examples. When the Korean War broke
out in June 1950, the Fed found itself stuck with the World War II policy
of pegging yields on long-term Treasury bonds. That policy was unsustainable as the market began to dump bonds on the Fed in massive
quantities. Looking at the Federal Reserve Bulletins for June 1950 and
January 1951, we see that the Fed’s holdings of government securities
at the end of 1950 were 28 percent above the level at the end of May
1950. In March 1951, the Fed was able to negotiate the Treasury-Fed
Accord that discontinued the Fed’s obligation to peg Treasury bond
rates. Clearly, the Fed must be very careful about the commitments it
makes or implies. As I write, the Fed needs to find a way to extract itself
from its announced plans to buy bonds and hold the fed funds rate
near zero.
My personal view at this time is that waiting until spring 2022 to
begin the process of raising the federal funds rate increases the risk,
dramatically, that 2021’s surge in inflation will continue. More generally, though, isn’t this situation of exactly the sort that led many
economists to favor an announced inflation target? I will return to this
question below.
A somewhat similar process to the one in 1950 occurred in the late
1970s, as the Fed attempted to hold down interest rates as inflationary
pressures blossomed. The inflation accumulated to the point of creating a variety of ills in the economy. The Fed’s power to print money was
then the problem and not the solution. The Fed had no information
advantage or processing advantage over the private market. It took
Paul Volcker’s leadership advantage to deal with the growing inflation
turmoil. No private entity could fix the inflation problem.
64 |

What Does the Fed Know? When Does it Know It?
A more recent Fed processing failure is worthy of comment, especially since it illustrates my own failure to understand what was happening with the house price bubble that ran from 2000 to 2006. I was a
member of the FOMC during those years and still have a vivid memory
of the special FOMC study of the housing situation. It is a sobering
exercise to review the FOMC transcript of the meeting of June 29-30,
2005. By coincidence, this was also the last FOMC meeting that Marvin attended, as Chairman Greenspan noted at the beginning of the
meeting.
The staff analysis of the housing situation, and my own contributions to the FOMC debate, demonstrate how wrong the FOMC can be.
I had given several speeches, as had Greenspan, worrying about the
potential for financial chaos should Fannie Mae and/or Freddie Mac
find themselves approaching insolvency. They did become insolvent
in 2008, but the problem was easy to fix. All the government had to do
was to take them into conservatorship, making the implicit guarantee
explicit.
A much more serious problem was the accumulation of weak mortgage paper on the books of investment and commercial banks. There
was not a word of that development in the FOMC record until 2008,
and even then the severity of the problem was not understood, at least
from my reading of the FOMC transcripts of this period. The private
market understood the developing problem before the Fed did. The
Fed and the Treasury did eventually deal with that problem by recapitalizing major banks and taking other steps familiar to students of the
2008 financial crisis. However, there was no satisfactory way of dealing
with the insolvency of several million households that defaulted on
mortgage and other debt. The FOMC and its staff seemed not to have a
clue about these risks until they arose.

Market speculation on what the Fed knows
I am often amused by press accounts guessing as to what the Fed
knows that the private market does not know. During most of my
tenure in St. Louis, I was fortunate to have Robert Rasche as research
director. Bob and I had known each other for many years, and he was
| 65

What Does the Fed Know? When Does it Know It?

Poole
a full coauthor on most of my speeches, many of which were mini
research projects.
Particularly after a Fed surprise policy adjustment, Bob and I would
joke, before release of the news, about the likely effects on the markets. I wish I had kept a record of what we guessed would be the
impact on the stock market. We generally got the direction right but
neither of us had confidence as to the magnitude of the effect. I doubt
that any member of the FOMC or its staff then or since has a better
track record than we did. The Fed’s lack of ability to forecast the effects
of its policy changes was one of the reasons I was always opposed to
the Fed creating policy surprises.
Given that private participants in financial markets are deeply
knowledgeable and highly motived to understand the significance of
Fed policy adjustments, the observation just made should be a warning. If market experts cannot figure out the likely effects of Fed policy
adjustments, why should economists and other outside observers
believe that the Fed itself has a good idea about the effects of its policy
adjustments?
One of the complications Fed policymakers face is that some market participants interpret policy adjustments as evidence that the Fed
knows something the market in general does not know. As already emphasized, that view seems to me to be mistaken most of the time.
Market participants sometimes interpret Fed policy adjustments as
“sending a message” of some sort, or “priming the market.” That interpretation is probably correct on occasion, but it is an example of the
Fed’s failure to explain its policy clearly. Fed words and policy actions
should be consistent with one another to avoid confusing the market.

Application to current policy situation
Marvin’s 1986 paper concerns interactions with the markets of Fed
statements, policy actions, and stance on disclosure. The Fed’s adoption in 2012 of a formal inflation target was a new phase in its communication with the markets.4
In 2012, the FOMC settled on a target of 2 percent annual increase
measured by the annual change in the price index for personal consumption expenditures. In August 2020, the FOMC changed the
target to be an undefined average over time: “The Committee seeks
to achieve inflation that averages 2 percent over time, and therefore
judges that, following periods when inflation has been running persistently below 2 percent, appropriate monetary policy will likely aim
to achieve inflation moderately above 2 percent for some time.” How
moderate and how long? The problem is not that the Fed won’t provide an answer, but that it does not know the answer.
Why the change in the inflation target? I’ll offer a speculation. The
Fed wanted to pull out all the stops to make clear that it was doing
everything possible to fight the COVID-19 pandemic. Easy money was
the only tool it had. However, the Fed badly miscalculated the inflation risk. As the months passed, how could the Fed backtrack without
appearing to give up its battle against the pandemic or admitting that
it had miscalculated the inflation risk?
As many have noted, by changing to an ill-defined average over
time the FOMC has debased the clarity of the original target. Also, the
committee’s turn to vague language has walked away from one of
the important original arguments for an inflation target — providing
discipline to the FOMC itself. As I write, with an Excel spreadsheet in
front of me, I see that (using continuous compounding) the average
annual percentage change over the past 36 months of the Fed’s favorite inflation gauge is 2.5 percent for data through October 2021. The
three-year average breached 2 percent in June 2021, and yet the FOMC
continues to buy assets and continues to hold the federal funds
4

66 |

S hapiro and Wilson (2019) provide a convenient short history of FOMC deliberations
on the subject.

| 67

Poole
rate near zero. The Fed has achieved its objective — real interest rates
are now substantially negative.
The Fed, by taking variance and uncertainty out of the financial markets with its policy of asset purchases and near-zero fed funds rate, has
increased variance and uncertainty in both quantities and prices in the
goods and labor markets. Fifty years ago, I published a paper5 on this
exact topic. FOMC members who are especially concerned about the
real economy have the story upside down — Fed policy has increased,
not reduced, variance and uncertainty in the goods and labor markets.
Why should they be so terribly solicitous of the financial markets while
ignoring what is going on in the goods and labor markets?
The June 2021 FOMC meeting was June 15-16. By that time the committee knew that its previous projections had gone seriously wrong.
On June 10, the BLS had released data for the consumer price index
for May, which showed a year-over-year increase of 5.0 percent and
ex food and energy of 3.8 percent. The index for used cars was up a
remarkable 29.7 percent. On May 28, the BEA had released data on the
personal consumption expenditures price index for April, showing a
year-over-year increase of 3.6 percent (3.1 percent ex food and energy).
The median of the FOMC’s inflation projections (PCE price index) for
2021, released December 16, 2020, were already obsolete. The FOMC
had projected for all of 2021 an increase of 1.8 percent. In June 2021,
the projection for the full year was 3.4 percent. By December 2021,
the projection was 5.3 percent. Yet, at that time, the FOMC continued
to purchase billions of dollars of bonds and continued to maintain a
near-zero fed funds rate.
Without question, COVID-19 has created a severe disturbance. That
said, we should not forget that the start of the great inflation in the
mid-1960s also saw severe disturbances. The Vietnam War and President Johnson’s efforts to prevent disputes over the war from disrupting his plans for expansion of Great Society programs were active
Federal Reserve concerns.
5

68 |

Poole (1970).

What Does the Fed Know? When Does it Know It?
Anyone who doubts that statement should spend time studying the
detailed economic history of this era. I was an economist on the Fed’s
Washington staff from May 1969 through June 1973. I used to commute to the Fed on Rock Creek Parkway and remember antiwar demonstrators throwing park benches off overpasses onto the parkway.
In 1965 and thereafter, the Fed was also concerned with the solvency of the savings and loan industry. Pushing up interest rates would
lead to disintermediation. Then, in 1971, Nixon imposed wage-price
controls. Then, in October 1973, OPEC’s oil embargo sent petroleum
prices surging. I was among those who had problems finding an open
gas station. And then there was the Watergate affair. Pumping money
into the economy did not help to resolve any of these nonfinancial
disturbances.
I do not mean to downplay the importance of COVID-19, which has
killed an estimated 800,000 Americans as of this writing. That number
may be compared with about 50,000 Americans dead from the Vietnam War. The point is that the Fed cannot affect a real variable — the
number of vaccinated Americans — by monetary expansion. Nor will
monetary expansion alleviate supply chain problems. The economy is
very liquid. Bank credit is readily available for any trucking company
that wants to buy more trucks to get goods off Pacific Coast docks.
Eighteen wheelers going down the highway are traveling billboards
for trucking companies trying to hire more drivers.
A policy of drift, month after month, for one reason after another,
created the Great Inflation of 1965-80. The same policy today is likely
to yield the same outcome.
Continuing to pump cash into the economy will not encourage
more vaccinations and will not alleviate supply chain disruptions. As
I write this just before Christmas 2021, the Federal Reserve is at least
six months behind in responding to the flow of formal data and the
many anecdotal observations familiar to anyone who is looking. Help
wanted signs are everywhere, and employee turnover is the highest in
the history of the JOLTS data. Residential property prices month after
month have been rising at a rate higher than the peak rate during the
house price bubble before the 2008 financial crisis.
| 69

What Does the Fed Know? When Does it Know It?

Poole
So far, inflation expectations data suggest that the market has not
lost confidence in the Fed. The risk is there, however; Goodfriend explained the process in his 1993 and 2007 papers, and elsewhere.
Knowledge of this history, in the United States and abroad, led many
economists (including me) to advocate a Federal Reserve commitment to an announced inflation target. The announced target would
communicate more clearly with markets and — very importantly —
constrain the Fed to act as it had promised. The literature on inflation
targeting is voluminous; I will refer to just one paper, by David Archer
(2000).
Archer — assistant governor at the Reserve Bank of New Zealand
at the time — begins his paper with four observations about the New
Zealand inflation targeting regime.
1. A nominal variable (such as the price level or the inflation rate)
is recognized as the sole achievable medium-term objective for
monetary policy.
2. An attempt to drive policy directly at the medium-term objective
via a tightly specified inflation target, rather than indirectly via an
intermediate target.
3. An institutional structure that clearly articulates the respective
roles and responsibilities of the key actors (the central bank and
the government).
4. Heavy reliance on transparency to support the arrangement and
cover the “weak points” in the institutional structure.
In the context of Goodfriend’s work and this essay, points (3) and (4)
are especially relevant. Just following the above points, Archer refers
to, “the interactions of ‘public choice’ incentives and expectations.” The
US experience of 1965-80 displays these interactions in spades. The
Federal Reserve today faces the same sorts of issues. The public debate
about how best to deal with COVID-19 is confused. The Fed, of course,
has expressed its commitment to do “whatever it takes” — to use all
the “tools” at its disposal. The public and the Fed itself seem baffled as

70 |

to what these phrases mean. Has the Fed published a list of the tools it
can use to tackle COVID-19?
If the Fed had not confused matters by fuzzing up its inflation target
in August 2020 — walking away from transparency — it would have
been in a position in mid-2021 to end asset purchases immediately
and to begin to raise its fed funds rate target. The Fed could have said
that the situation was extremely difficult with the new delta variant
and the best monetary policy was not clear. The Fed could have said
that given the inflation data and its commitment to an inflation target,
it was time to begin raising rates. If the increase in rates turned out to
be premature, the policy adjustment could be reversed in the future.
Some will be reminded of the quote attributed to Keynes: “When
the facts change, I change my mind. What do you do, sir?” As a bit of
research will reveal, there is controversy over whether the quote is
accurate, but it fits here anyway. By June 2021, the facts had changed
dramatically.
I am also reminded of the legend of Odysseus. “The technique is
called Odyssean self-control, and it is more effective than the strenuous exertion of willpower, which is easily overmatched in the moment
by temptation.”6 In August 2020, the Fed cut the ropes by which it had
lashed itself to the mast, and now all of us will pay for that mistake.
An economist who discusses his outlook is always risking ridicule
in retrospect. That understood, I have put my cards on the table. I am
well aware of the first law of forecasting: if you name a number, do not
name a date; if you name a date, do not name a number. I say “here
goes” because the only true test of any empirical proposition intended to be taken seriously is an out-of-sample forecast. My outlook is
dependent on the data I have observed and on Federal Reserve statements about its policy. I believe that the inflation rate in 2022 will be at
least as high as the 2021 rate.

6

Pinker (2021), p. 55.

| 71

What Does the Fed Know? When Does it Know It?

Poole
My sense of the policy environment as I write in the fourth quarter
of 2021 is that the Fed has created a generalized asset price inflation.
At the time of the FOMC’s December 2021 meeting, stock prices were
high, as measured by P/E ratios. Bond prices were high — interest rates
low — as compared with data over past decades. Residential property prices were rising at a more rapid rate than at any time during
the house price bubble of 2000-06. The second derivative of house
prices — the speed with which house price inflation year over year
has increased — is higher than at any time in the history of the broad
repeat-sale house price indexes. Farmland prices are increasing. However, gold prices — a traditional measure of inflation concerns — have
not moved by much.
Here I am writing in late December 2021, and at best the Fed has
indicated that it might begin to increase the federal funds rate in the
middle of 2022. I am well aware that I am offering pointed criticism
of the Fed in a volume to be published by the Federal Reserve Bank
of Richmond. Marvin Goodfriend spoke his mind, as evidenced especially in his paper on the Merrill case, which was highly critical of Fed
leadership. It was gutsy for him to write this paper while he was in the
Research Department at the Richmond Fed.
I would like to think that a small fraction of his approach to policy
came from his old professor with the initials WP. I am honored that I
was invited to contribute a paper to this volume.

References
Archer, David J. 2000. “Inflation Targeting in New Zealand.” IMF seminar
on inflation targeting, March 20-21.
Board of Governors of the Federal Reserve System. 1950. Federal Reserve Bulletin. June.
Board of Governors of the Federal Reserve System. 1951. Federal Reserve Bulletin. January.
Chang, Andrew C., and Tyler J. Hanson. 2015. “The Accuracy of Forecasts Prepared for the Federal Open Market Committee.” Board of
Governors of the Federal Reserve System Finance and Economics
Discussion Series 2015-062.
Federal Open Market Committee. 2005. “Meeting of the Federal Open
Market Committee, June 29-30.”
Federal Open Market Committee. 2012. “Statement on Longer-Run
Goals and Monetary Policy Strategy, Adopted effective January 24,
2012.”
Federal Open Market Committee. 2020. “Statement on Longer-Run
Goals and Monetary Policy Strategy, Adopted effective January 24,
2012; as amended effective August 27, 2020.”
Goodfriend, Marvin. 1986. “Monetary Mystique: Secrecy and Central
Banking.” Journal of Monetary Economics 17, no. 1 (January): 63-92.
Goodfriend, Marvin. 1993. “Interest Rate Policy and the Inflation Scare
Problem: 1979–1992.” Federal Reserve Bank of Richmond Economic
Quarterly 79, no. 1 (Winter): 1-23.
Goodfriend, Marvin. 2007. “How the World Achieved Consensus on
Monetary Policy.” Journal of Economic Perspectives 21, no. 4 (Fall): 47-68.

72 |

| 73

Poole
Lewis, Michael. 2011. The Big Short: Inside the Doomsday Machine. New
York: W.W. Norton & Company. Kindle Edition.
Pinker, Steven. 2021. Rationality: What It Is, Why It Seems Scarce, Why It
Matters. New York: Penguin Random House. Kindle Edition.
Poole, William. 1970. “Optimal Choice of Monetary Policy Instruments
in a Simple Stochastic Macro Model.” Quarterly Journal of Economics 84,
no. 2 (May): 197-216.
Poole, William. 1999. “Synching, Not Sinking, the Markets.” Speech at
the meeting of the Philadelphia Council for Business Economics, Federal Reserve Bank of Philadelphia, August 6.
Shapiro, Adam, and Daniel J. Wilson. 2019. “The Evolution of the FOMC’s Explicit Inflation Target,” Federal Reserve Bank of San Francisco
FRBSF Economic Letter, April 15.

74 |

| 75

Perspectives on Policy and
Research Contributions

76 |

| 77

Lending and Incentives

Central Bank Lending and Incentives
Kartik Athreya and Stephen D. Williamson
Marvin Goodfriend was a remarkably original thinker and did much
to advance the science of monetary policy. We both benefited not
only in key ways from Marvin’s published work in economics, but also
just as much from his contributions at seminars, conferences, and over
lunch or dinner. Our essay primarily concerns Marvin’s work with Jeff
Lacker, in Goodfriend and Lacker (1999). It analyzes the roles of central bank lending and central bank credit policy, describes what can
go wrong with central bank lending, and suggests how to fix these
problems.

Central bank lending
Lending to the financial system has been a critical feature of central
banking for a very long time. Typically, central banks are constrained
to hold assets to back their liabilities, and those assets include loans to
the private sector. Indeed, in some central banking systems, the principal mechanism by which the central bank controls the quantity of
its liabilities in circulation is through lending to private banks. Notably,
at its inception in 1914, the Federal Reserve System was constructed
as an organization of semiautonomous regional banks that financed
lending to member banks by issuing currency. These regional Federal
Reserve Banks became the sole issuers of circulating currency. Each
regional Reserve Bank in turn held the liabilities of its member banks
to back the currency. Government debt was not initially a key asset in
the Fed’s portfolio, and open market operations were not an important
component of Fed activities until the 1920s. Today, the European Central Bank intervenes primarily through central bank lending to banks
in the euro area, by way of the ECB’s main refinancing operations. So,
one possible design for central banks includes the use of central bank
lending in day-to-day or week-to-week financial market intervention.
But — and this is the most important aspect of Marvin Goodfriend’s
research that we want to address — lending to financial institutions is
78 |

| 79

Athreya and Williamson
a key component of central bank crisis intervention. The lender of last
resort role for central banks was essentially invented by the Bank of
England (BoE), as documented for example in Bagehot (1873). In the
later 19th century, the BoE was a private institution, which had been
granted a special place in the British financial system. It had a different
financial structure from a typical private bank in the United Kingdom
at the time and had been granted a monopoly on currency issue by
the Crown through Peel’s Bank Act of 1844.
The BoE had also built a reputation for safety by the late 19th century. So, during the recurring financial crises in the UK in the later
19th century, consumers and firms typically fled from bank liabilities,
perceived to have increased in their riskiness, to BoE liabilities, and
this inflow of funding at the BoE was then used to finance lending to
Banks. But the BoE was presumably more well-informed than the general public about which banks were insolvent and which were merely
illiquid and so could profit from judicious lending under Bagehot’s
(1873) principles: lend freely, at a high rate, against good collateral.
In the United States, a principal objective of the authors of the
Federal Reserve Act of 1913 was preventing, or at least mitigating, the
effects of the banking panics that occurred in the US in the late 19th
century and early 20th century. Central bank lending to private banks
was seen as the principal crisis mechanism at the Fed’s disposal. This
was a key element in research done under the National Monetary
Commission leading up to the Federal Reserve Act legislation in 1913.
But Fed policy during the Great Depression is typically viewed as a
failure,1 in part because the Fed did not lend adequately to the private
banking system. Seemingly, Fed leadership absorbed the lessons of
the Great Depression in subsequent years, and the global financial
crisis was a quite different story. Indeed, Ben Bernanke has argued that
much of the Fed’s lending policy during the global financial crisis was
motivated by Bagehot’s principles.2 The argument for central bank
crisis lending generally rests on advantages the central bank might
1
2

80 |

S ee for example Friedman and Schwartz (1963).
See Bernanke (2017).

Lending and Incentives
have relative to the private sector. For one example, the central bank
could have superior information on banks’ creditworthiness, due to its
supervisory role in the banking system. For another, the deep pockets
of the central bank are important, given the backstop of the federal
government’s power to tax.
There were at least two unusual elements in the Fed’s lending strategy during the global financial crisis. The first was lending through the
Term Auction Facility (TAF). An undesirable feature of lending through
the Fed’s discount window is that a bank’s discount window borrowings can send a signal to the market that the bank on the receiving
end of the loan is distressed. This “stigma” can deter borrowing, which
works against the intent of central bank lending in a crisis.3 Stigma can
occur because, even though the details of discount window lending
are not public, it may be possible to infer which banks are borrowing at
the discount window, particularly if the banks are large. So, the goal of
the TAF program was to auction off discount window funds to member banks willing to post the appropriate collateral. The assumption
was that both distressed and nondistressed banks would be on the
receiving end of TAF funds, thus eliminating, or at least mitigating, the
stigma effect.
A second example of unusual lending during the global financial
crisis was the extension of Fed loans to financial institutions that were
not commercial banks. In historical banking panics prior to the founding of the Fed, for example, those that occurred in the US in the late
19th and early 20th centuries, the key problem was massive outflows
of deposits from commercial banks that caused the disruption of payments, bank failures, and the forced sale of bank assets. Such negative
effects, where they occurred in solvent but illiquid banks, could in
principle have been mitigated by central bank lending to commercial
banks if there had been a US central bank during this period. The global financial crisis was different, however. Early on, panics appear
to have occurred at the wholesale level, for example, when asset
3

Ennis (2019).

| 81

Athreya and Williamson
portfolios of asset-backed securities were funded by rolling over overnight repos. The perception that some asset-backed securities were
much riskier than previously thought generated the withdrawal of
funding for such “shadow-banking” entities, in part inducing fire sales
of assets. Later, problems involved systemic risk, which had been latent
until 2008. For example, the Fed’s direct lending to American International Group (AIG) — primarily an insurance company — was intended
to address a novel crisis problem not directly related to retail banking.
That is, through the sale of financial derivatives, AIG had made itself
highly sensitive to aggregate risk and had thus created a threat to the
entire financial system. Whether the Fed has a legitimate role in lending to nonbank financial institutions is not clear. For example, it seems
harder to make the case that the central bank has an information
advantage in lending to financial firms that it does not regulate, or that
the Fed somehow has an advantage relative to large private financial
institutions in such lending.

Goodfriend-Lacker and central bank lending
Crisis lending by the central bank might on its surface appear
straightforward. The basic nature of banking is captured nicely by the
classic model of Diamond and Dybvig (1983). Banking, by its nature,
involves the transformation of illiquid assets into liquid ones. Such
transformation is socially useful, as retail payments work efficiently if
consumers and firms can trade the widely acceptable liabilities of third
parties — here, banks — for goods and services. These third-party
liabilities are viewed as highly liquid, despite the underlying assets
held by banks being difficult to exchange for goods and services. But
in conducting liquidity transformation, banks leave themselves open
to runs, as shown by Diamond and Dybvig (1983). If each depositor
anticipates that all other depositors will run to the bank to withdraw
their deposits, then the bank — which would otherwise be solvent
— cannot satisfy all requests for withdrawal and it fails. Though a key
motivation for establishing the Federal Reserve System was to prevent
or mitigate banking panics through central bank crisis lending, the
banking theory literature appears to have given central bank lending
short shrift. For example, though the Diamond-Dybvig (1983) model
82 |

Lending and Incentives
has received much attention, it does not directly address the role of
central bank crisis lending. Thus, Goodfriend and Lacker (1999), while
not making a contribution in a formal modeling sense, use available
theory to analyze the incentive problems inherent in central bank
lending to private financial institutions and provide useful recommendations for improving this aspect of central bank policy.
Goodfriend and Lacker (1999) recognize that, while we know less
than we would like about the pitfalls of central bank lending, we
know a lot about how private loan commitments work. And in some
ways private loan commitments are not so different from central bank
lending. For example, discount window arrangements between the
Fed and private commercial banks are essentially loan commitments.
The Fed specifies the terms under which banks can borrow, including
interest rates, admissible collateral, and collateral haircuts, and then
commits to lending to banks on these terms, with some restrictions.
The key difference, however, is that private banks making loan commitments are concerned with their own profits, while the central bank has
public policy goals in mind, for example, the systemic implications of
the failure of large financial institutions. But since the Fed’s loan commitments to banks are in part insurance, Fed loan commitments are
subject to the same moral hazard issues as private loan commitments,
with no profit motive for the Fed to motivate structuring its lending to
deal with this moral hazard.
But what could central banks learn from the structure of private
loan commitments? First, as with borrowing from the central bank,
borrowing from a private bank under a loan commitment may become
desirable when a firm is under financial stress. For example, a large
firm may normally have access to the commercial paper market, but
if it becomes widely known that the firm’s financial state is precarious,
borrowing by way of commercial paper may become more costly or
impossible. Under these circumstances, taking down a loan commitment may be desirable for the firm. Loan commitments may have implications for the commercial paper market, ex ante, as well. That is, a
firm may obtain more favorable terms in the commercial paper market
because it has a standing loan commitment with a bank. Subsequently, in the event of financial distress on the firm’s part, loans under the
| 83

Athreya and Williamson
commitment from the bank could serve to insure the holders of the
firm’s commercial paper.
Of course, the bank making a loan commitment does not want to be
in such a position. So typically, as Goodfriend and Lacker (1999) point
out, loan commitment arrangements have terms that give the firm
incentives not to take out loans — or prevent the firm from taking out
loans — if the firm has a high probability of defaulting. First, collateral
can play an important role, both in insuring the bank against losses in
the event of the firm’s default and giving the firm the incentive not to
default. Second, covenants in the loan commitment can allow the bank
not to lend to the firm given the firm’s financial state or limit lending in
various ways. Third, the bank can have the right to monitor the firm’s
activities, which, in conjunction with covenants, could serve to limit
lending in particular circumstances.
For central banks, the incentive problems are similar to those with
private loan commitments, with some caveats. First, the central bank’s
lending behavior has implications for other bank creditors, though in
the case of a regulated bank, this is muted by the existence of deposit
insurance. However, lending by the central bank to a precarious commercial bank can provide a window of time when uninsured depositors can conveniently exit. Effectively, the central bank acts to provide
implicit insurance to the uninsured depositors. But the central bank’s
lending behavior can ultimately change the payoffs to the stakeholders in important ways in the event of a bank default. The Fed protects
itself by requiring that banks pledge good collateral against central
bank loans, but that implies that this collateral cannot be used to pay
off the bank’s other creditors in such an event. Further, if the Fed lends
to a bank that ultimately fails, the Fed becomes a senior creditor, which
matters for the other regulator, the Federal Deposit Insurance Corporation (FDIC). Once the FDIC steps in to resolve the failed bank, it pays off
the bank’s debt to the Fed. Though the FDIC then retains the collateral
pledged against the central bank loan, this may not cover the loss.
Second, the central bank, like private banks making loan commitments, may want to commit to limiting central bank lending under
particular circumstances. It is well-accepted, for example, that lending
84 |

Lending and Incentives
to insolvent banks is a bad idea, and that central bank crisis lending
should be limited to cases where there is just a liquidity problem. Of
course, it may be difficult to tell the difference between an insolvent
bank and an illiquid one, but the Fed is in a good position to discriminate, given the information it acquires through its supervisory role
in the commercial banking system. But commitment to limit central
bank lending may be difficult nevertheless, and Goodfriend and Lacker
(1999) are particularly pessimistic about this. When push comes to
shove, they argue, the Fed will typically opt for the path of generous
lending except in the most egregious cases. The key issue is that moral
hazard problems associated with borrowing by banks are accentuated
as actual insolvency becomes more likely for an individual bank. But,
in general, the probability of bank insolvency is related to aggregate
activity, so many banks will be in a precarious state at the same time.
This is exactly the type of circumstance where the Fed is likely to lend
freely, rather than cutting off banks that are likely on the brink of insolvency.
We highlight a particularly prescient passage in Goodfriend and
Lacker (1999, p. 15):
	The financial stability mandate can create pressure to expand the
scope of central bank lending to nonbank financial institutions.
Nonbank financial intermediaries are capable of amassing sizable
financial market positions. The liquidation of these positions could
be seen as a threat to the stability of asset prices and the solvency of
many other financial institutions, including insured banks. A central
bank with no formal authority to lend outside a narrowly defined set
of institutions is, of course, well positioned to resist influence. Otherwise, we might see a tendency to expand the range of institutions
receiving central bank line-of-credit assistance.

This is an accurate prediction of part of what happened during the
global financial crisis, when the Fed expanded its lending beyond the
commercial banking sector. For example, after the Fed extended a
large loan to AIG, Goldman Sachs and Morgan Stanley became bank
holding companies and thus eligible for Fed lending. Of course, a key
difficulty is that it can be easy to see the short-term gains but hard to
| 85

Athreya and Williamson
see the long-term costs of central bank crisis intervention that goes
beyond the commercial banking sector. In theory, we understand
the implications of moral hazard in central bank lending for excessive
risk-taking and the expansion of already-large, too-big-to-fail financial
institutions. But these effects can be difficult to measure in practice.

Dealing with moral hazard in central bank lending
As we have argued, the view of Goodfriend and Lacker (1999) is that
there are issues in Fed lending to the private sector that need to be addressed. So, what to do about it? One possibility they consider is that
the Fed could forgo lending to the private sector entirely and focus
instead on facilitating the easing of liquidity problems in large financial
institutions and on orderly resolution in the event of large financial failures. A few examples in the past two decades, however, suggest that
this would be a radical alternative.
The role of the Fed in coordinating privately financed emergency
lending and support in the financial sector is an aspect of the use
of the Fed’s “good offices.” For example, in 1999, the New York Fed
participated in the effort to prevent the failure of a hedge fund, LongTerm Capital Management (LTCM). At the time, LTCM was viewed as
a systemically important financial institution that was encountering
liquidity problems and thus faced a potential forced sale of assets.
That is, LTCM was in a position that several large financial institutions
would find themselves in during the global financial crisis in 2008. The
Fed did not participate financially in propping up LTCM, but it helped
facilitate an arrangement by which a group of private financial institutions recapitalized the troubled hedge fund. Was the Fed’s intervention
during this episode necessary? Was there risk to the Fed in injecting
itself into negotiations over the potential failure of a private hedge
fund that was well outside the Fed’s normal supervisory purview?
An instance where the Fed’s intervention with respect to a large
troubled financial firm outside the commercial banking sector moved
from mere facilitation to key financial participation was the Bear
Stearns failure in spring 2008. As discussed in more detail in Goodfriend (2011), the Fed facilitated an orderly resolution in the Bear
Stearns failure by setting up a limited liability company, Maiden Lane I,

86 |

Lending and Incentives
which then proceeded to purchase troubled assets then held by Bear
Stearns, with funding coming from a loan from the Fed. This made it
more attractive for JPMorgan Chase to take over what remained of
Bear Stearns. The motivation for the Fed’s Bear Stearns intervention
was similar to that for the LTCM intervention, in that Bear Stearns was
a systemically important financial institution and a disorderly failure
would potentially have much wider effects in the financial sector. But
this illustrates what happens when the Fed becomes involved in negotiations involving large troubled financial institutions. In such negotiations, the Fed is the elephant in the room and can end up more financially involved than it initially intends, left holding the bag by private
financial institutions or passing on losses to taxpayers.
As Goodfriend and Lacker (1999) point out, the Fed may protect
itself by requiring good collateral with appropriate haircuts to secure
central bank lending, but safety for the Fed in this respect can be to
the detriment of the FDIC and a bank’s uninsured creditors. Potentially, there could be stricter capital thresholds for closing distressed
banks, though this is problematic due to the subjective nature of asset
valuation. Goodfriend and Lacker (1999) also suggest that the Fed
might choose to be “constructively ambiguous.” This means the Fed
could be deliberately vague about the conditions and terms on which
it will lend. Presumably, this could limit the quantity of lending, reducing moral hazard, and causing banks to bear less risk. But this would
have to be balanced against the increase in perceived risk faced by the
banking system due to the Fed’s unpredictable behavior.
During the global financial crisis, the Fed not only intervened in
important ways with respect to large nonbank financial institutions —
Bear Stearns and AIG in particular — but it made large loans to large
banks, Citigroup and Bank of America. The latter two banks certainly
fall in the too-big-to-fail category and are therefore important examples of the moral hazard problems associated with central bank
lending, as discussed by Goodfriend and Lacker (1999). Much of the
lending to large financial institutions — nonbanks and banks alike
— during the financial crisis fell under Section 13(3) of the Federal
Reserve Act, which gave the Fed very broad lending powers. In line
with Goodfriend and Lacker’s (1999) emphasis on limits to Fed lending,
| 87

Lending and Incentives

Athreya and Williamson
the Dodd-Frank Act of 2010 included provisions to constrain the Fed’s
powers under Section 13(3). However, Goodfriend and Lacker (1999)
might suggest we worry about the array of Fed lending programs
introduced during the COVID-19 pandemic, which included lending
to nonfinancial businesses (the Main Street Lending Program), to state
and local governments, and to money market mutual funds. In most
cases, there were explicit arrangements for the Treasury to absorb
potential losses, but this raises some of the key issues discussed in
Goodfriend and Lacker (1999), relating to moral hazard and the redistribution of creditor losses.
Finally, using an analogy to inflation control, Goodfriend and Lacker
(1999) argue that the Fed could establish a reputation for limiting its
lending to financial institutions, and that this could ultimately curb
the moral hazard problem associated with Fed lending. Unfortunately,
inflation control as a central bank goal has the advantage of simplicity
— for example a 2 percent inflation target — and it is relatively easy
for people to evaluate the Fed’s success in achieving the goal. With
respect to Fed lending, it is much more difficult both to establish what
the goal is and to evaluate the Fed’s performance relative to the goal.
Perhaps the only limits on Fed lending that can have force are those
specified explicitly in the Federal Reserve Act.

Other approaches to dealing with moral hazard in central
bank lending
It is possible that the key problems discussed by Goodfriend and
Lacker, associated with moral hazard and central bank lending, could
be solved by simple (though perhaps radical) changes in institutional
structure. We will consider two: one that changes central bank interventions that we term “repos only,” and another that reforms private
financial institutions that is conventionally called “narrow banking.”

auctions. As the Fed has learned since establishing a floor system for
monetary policy following the global financial crisis, intervention in
the overnight repo market — on either side of the market — proved
important for achieving the Fed’s overnight interest rate target. So, under the proposed system, the Fed would choose the size of its balance
sheet and the target for the overnight interest rate, and then the two
standing facilities would look after the rest. Thus, the Fed could conduct balance sheet policy independent of interest rate policy.
Such a system appears to solve some of the problems discussed by
Goodfriend and Lacker. In particular, discretion would be removed
from central bank lending, which would be done at arm’s length
through third parties. Thus, lending would be limited, and any lending
to troubled banks would be made on the same terms as by private
repo market lenders. Perhaps a defect of such a system is that collateral would be restricted relative to what is normally acceptable for
discount window lending, although this might be what Goodfriend
and Lacker had in mind.
The Standing Repo Facility, established by the Fed in July 2021, is
related to this proposal in that it increases the likelihood of regular
central bank lending. However, lending through this facility is currently
at the same rate as the discount rate, which is set above the interest
rate on reserves, whereas our proposal would have lending at the
policy rate. Note also that the Standing Repo Facility accepts only a
restricted set of collateral — Treasuries, agency securities, and agency
mortgage-backed securities.
Another simplified approach would be narrow banking. Proposals
for narrow banking have existed since at least the 1930s, as put forward by the Chicago Banking School.4 A key proponent of narrow
4

See, e.g., Hart (1935) and Irving Fisher (1935), among others.

One simple approach would involve a central bank lending policy
that restricts intervention to the repo market. Under this setup, the
central bank sets a target for the overnight interest rate and then
achieves that through two standing facilities, a repo facility and a
reverse repo facility, both involving fixed rate and full allotment
88 |

| 89

Athreya and Williamson
banking was Milton Friedman, who argued in 1960 that monetary
control would be improved if all private transactions accounts were
backed 100 percent by reserves. In general, a narrow bank is a financial
intermediary that backs all liabilities used as means of payment with
safe assets, typically central bank reserves or safe government debt.
Such safe backing could be a legal restriction or it could be an unconstrained choice of the bank. A useful survey of narrow banking proposals is in Pennacchi (2012).
Generally, narrow banking separates money from credit. For example, under a narrow banking proposal requiring that all means-ofpayment liabilities be backed 100 percent by central bank reserves
and government debt, much of the structure of banking regulation
could be eliminated. There would be no need for capital requirements,
leverage requirements, or deposit insurance, for example, as all meansof-payment liabilities of banks would be essentially risk free. There
would of course be other financial intermediaries holding risky asset
portfolios but, according to narrow-banking proponents, the liabilities of such institutions would be efficiently priced and not subject to
flights to safety.
There are two standard issues with narrow banking. One is that a
narrow banking structure potentially increases the demand for safe
assets, in that banking would no longer be about transforming risky
assets into safe ones, but guaranteeing the safety of bank deposits by
backing those deposits with safe assets only. This would be particularly
problematic in the current environment, in which the world is suffering a scarcity of safe assets.5 The second issue is that narrow banking
potentially causes disintermediation effects in crowding out the risky
assets that are currently held by regulated banks.

Lending and Incentives
from lending excessively to financial institutions deemed to be systemically important. Anticipating that, those systemically important
financial institutions will behave in suboptimal ways.
Both proposals in this section have the flavor of the ideas in Goodfriend and King (1988), who argue that most of the goals of the central
bank can be accomplished through conventional monetary policy.
Goodfriend and King (1988), for example, cast doubt on the value of
crisis intervention, arguing that such lending is prone to moral hazard
problems, and thus dominated by indirect injections of liquidity in a
crisis, through open market operations.

Conclusion
Much of Marvin Goodfriend’s work was both innovative and prescient. That certainly applies to his work with Jeff Lacker in 1999. Nine
years before the financial crisis, Marvin and Jeff grappled with issues
of moral hazard and central bank intervention that would be key to
how the crisis unfolded in 2008-09. They may not have seen the global
financial crisis coming, but their analysis helped provide important
background for policymakers during the crisis and afterward.

In addition, narrow banking (as well as the simplified, repo market
approach to monetary policy outlined previously in this section) would
not solve all the problems discussed by Goodfriend and Lacker. In particular, if the central bank is still permitted to lend outside the banking
sector, there is nothing in the proposal to prevent the Fed
5

90 |

See Andolfatto and Williamson (2015).

| 91

Athreya and Williamson

References
Andolfatto, David, and Stephen Williamson. 2015. “Scarcity of Safe
Assets, Inflation, and the Policy Trap.” Journal of Monetary Economics 73
(July): 70-92.
Bagehot, Walter. 1873. Lombard Street: A Description of the Money Market. New York: Scribner, Armstrong & Co.

Lending and Incentives
Hart, A.G. 1935. “The ‘Chicago Plan’ of Banking Reform: A Proposal for
Making Monetary Management Effective in the United States.” Review
of Economic Studies 2, no. 2 (February): 104-116.
Pennacchi, George. 2012. “Narrow Banking.” Annual Review of Financial
Economics 4 (October): 1-34.

Bernanke, Ben. 2017. The Courage to Act: A Memoir of a Crisis and its
Aftermath. New York: W.W. Norton & Co.
Diamond, Douglas W., and Philip H. Dybvig. 1983. “Bank Runs, Deposit
Insurance, and Liquidity.” Journal of Political Economy 91, no. 3 (June):
401-419.
Ennis, Huberto M. 2019. “Interventions in Markets with Adverse Selection: Implications for Discount Window Stigma.” Journal of Money,
Credit and Banking 51, no. 7 (October): 1737-1764.
Fisher, Irving. 1935. 100% Money. New York: Adelphi.
Friedman, Milton. 1960. A Program for Monetary Stability. New York:
Fordham University Press.
Friedman, Milton, and Anna Schwartz. 1963. A Monetary History of the
United States, 1867-1960. Princeton: Princeton University Press.
Goodfriend, Marvin. 2011. “Central Banking in the Credit Turmoil: An
Assessment of Federal Reserve Practice.” Journal of Monetary Economics
58, no. 1 (January): 1-12.
Goodfriend, Marvin, and Robert G. King. 1988. “Financial Deregulation,
Monetary Policy, and Central Banking.” Federal Reserve Bank of Richmond Economic Review May/June: 3-22.
Goodfriend, Marvin, and Jeffrey Lacker. 1999. “Limited Commitment
and Central Bank Lending.” Federal Reserve Bank of Richmond Economic Quarterly 85, no. 4 (Fall): 1-27.

92 |

| 93

The Zero Lower Bound

Marvin Goodfriend and the Zero
Lower Bound
Ben S. Bernanke
The zero lower bound (ZLB) on short-term interest rates has become
a central issue in contemporary monetary economics. Fundamentally,
the ZLB exists because households and businesses can choose to hold
cash, which pays zero nominal interest, rather than accept a negative
return on their short-term investments.1 (Other considerations, including concerns about the possibly adverse effects of negative rates on
the financial system, have also made some central banks unwilling to
cut short-term rates below zero or very far below zero.) Together with
long-term global declines in the so-called neutral rate of interest —
the interest rate consistent with full employment and price stability —
the ZLB has significantly reduced the ability of monetary policymakers
to ease policy through traditional short-term interest rate cuts.
For example, the Fed’s reaction to a typical recession before 2000
was to cut its target for the federal funds rate by 5 to 6 percentage
points. Today, in contrast, the Fed’s scope for rate cuts is probably at
most 2 or 3 percentage points on average — even less in Europe or
Japan. The reduction in policy “space” available through traditional
methods has led central banks to experiment with alternative tools,
including quantitative easing and detailed forward guidance. Central
banks have also developed new policy frameworks aimed at mitigating the effects of the ZLB, such as the Fed’s flexible average inflation
targeting approach.
1

94 |

 ith the advent of modestly negative rates in some jurisdictions, the lower bound is
W
now commonly referred to as the effective lower bound. Here I’ll stick with the older
ZLB terminology.

| 95

Bernanke
During the decades following World War II, the ZLB was treated by
most economists as a Depression-era curiosity. However, it became
relevant again in the 1990s, when the Bank of Japan — in the aftermath of the collapse of the country’s stock and real estate markets —
struggled with deflation and the lower bound with little success. Many
at the time saw the Japanese economy as quite different from that of
the United States. Its monetary institutions and practices differed from
ours, and key structural features — including high saving rates, slow
population growth, and a sharp slowdown in productivity — were
conducive to a lower neutral interest rate. Nevertheless, the Federal
Reserve was sufficiently interested in the implications of low rates
of inflation and interest to organize an October 1999 conference in
Woodstock, Vermont, on the subject of “Monetary Policy in a Low-Inflation Environment.” Then an academic, I was fortunate to attend. It
was a stimulating few days, full of untrammeled discussions and new
and sometimes radical ideas. Our debates, in the formal sessions and
informal get-togethers, plumbed deep issues in monetary theory. I suspect that most of the attendees enjoyed the blue-sky thinking at the
conference but underestimated the practical significance that those
discussions would have a few years later. I know that I did.
Marvin Goodfriend’s paper at the Woodstock conference, “Overcoming the Zero Bound on Interest Rate Policy,” was a highlight. Marvin’s
article — which would appear in 2000 in the Journal of Money, Credit
and Banking — anticipated a number of the approaches that central
banks would use (or contemplate) during and after the global financial
crisis. Like Irving Fisher, John Maynard Keynes, Silvio Gesell, and other
great economists who had thought about these issues in earlier eras,
Marvin did not hesitate to recommend radical institutional reforms
— notably changes that would eliminate the constraint on monetary
policy posed by the availability of currency. He would return to these
issues in a number of subsequent writings, including in a paper presented at the August 2016 Jackson Hole conference, entitled “The Case
for Unencumbering Interest Rate Policy at the Zero Bound.” Rereading
these papers today, I am impressed by their prescience and insight. I
am also struck by Marvin’s conviction that his job was to get the economic analysis right, without concern about the politics.
96 |

The Zero Lower Bound
A general approach to overcoming the ZLB, favored by some economists, is to try to manage inflation or inflation expectations directly,
through forward guidance or by announcing changes to the central
bank’s policy framework. For example, raising the central bank’s
medium-term inflation target — if fully credible — should also increase the neutral nominal interest rate and thus increase policy space.
As Marvin pointed out in his 2000 paper, though, even putting aside
issues of credibility, this tactic is inefficient. Raising the target (and
assuming the new target can be reached) imposes higher inflation at
all times but benefits the economy only at times when the ZLB would
otherwise bind. Moreover, it reduces but does not eliminate the adverse effects of the ZLB in severe recessions, in which deeply negative
real interest rates are needed. Alternative policy frameworks that are at
least theoretically more efficient, though also more complex, depend
on the central bank’s ability to vary inflation and inflation expectations
according to the state of the economy, for example, through price-level targeting (Wolman, 1998; Krugman, 1998; Eggertsson and Woodford, 2003).
Marvin was skeptical of monetary policymakers’ ability to manage
expectations in this way, especially since the theoretically optimal
policies are typically time-inconsistent and thus may not be credible.
His preferred approaches instead involved concrete policy actions or
institutional changes whose effectiveness does not depend on convincing the public and markets that the central bank’s future policy
strategy has changed.
Marvin’s 2000 paper proposed several specific policies that could
loosen the constraint of the ZLB. An idea that particularly appealed to
him — he would reconsider related issues in detail in his Jackson Hole
paper 16 years later — was for the central bank to impose a variable
charge on bank reserves to create what would effectively be a negative
short-term interest rate on that asset. Following the imposition of a
charge, banks’ efforts to substitute into alternative liquid assets should
lead the negative rate to spill over into money markets and perhaps to
longer-term assets as well. Indeed, this approach would ultimately be
successfully adopted by several central banks, including the European
| 97

The Zero Lower Bound

Bernanke
Central Bank and the Bank of Japan, and actively studied by others,
including the Bank of England.

ibility of deposits into cash, bank deposits and currency often traded at
varying, market-determined rates.

However, monetary theorist that he was, Marvin worried that the effectiveness of taxing bank reserves would be undermined by depositor
arbitrage. If banks tried to pass on their negative returns to depositors,
say by imposing new fees on checking accounts, people would have an
incentive to hold cash instead of deposits, thus disintermediating the
banking system. That arbitrage in turn would limit the central bank’s
ability to impose negative rates. In his 2000 and 2016 papers, Marvin
proposed three possible solutions to this problem.

One can only be impressed by the intellectual rigor and creativity
of these ideas. With the benefit of hindsight, though, it does not seem
that any of these additional measures are needed to achieve meaningfully negative rates (though I take Rogoff’s point about the negative
social externalities caused by the circulation of large bills). If anything,
Marvin underestimated the effectiveness of his proposal to impose
a variable charge on bank reserves, without supplementary steps to
limit currency hoarding. In recent years, central banks have been able
to push beyond the ZLB without provoking the hoarding that Marvin
worried about. In part, that reflects the fact that banks rely substantially on wholesale funding markets. The costs to lenders in those markets
of holding assets in physical cash is substantial, taking into account
security, insurance, and the inconvenience of making large payments
to global counterparties in cash. Banks have thus been able to pass
on negative returns to many of their largest funders, even if not to all
retail depositors.

First, simply abolish cash (or large bills, which are easiest to store).
That would solve the problem, presumably, but Marvin worried about
the social costs — increased transactions costs for the unbanked, for
example — of such a step. Incidentally, this solution would later be
explored in detail by Rogoff (2016), who emphasized that the social
costs would be balanced by important gains, including the reduction
of money laundering and tax evasion.
Second, impose a carry tax on cash, allowing a negative return on
currency at times when a negative rate was needed to support economic activity. In his 2000 paper, Marvin suggested that the carry
tax be implemented through magnetic strips attached to bills, which
at the time seemed technologically implausible. Today, the Federal
Reserve is discussing the possibility of creating a central bank digital
currency (CBDC). If a CBDC were to bear a variable interest rate, and if
it supplanted physical currency, it would bring Marvin’s idea into the
realm of technical feasibility.
Finally, in his 2016 paper, Marvin suggested that the Fed eliminate
the fixed one-for-one exchange rate between currency and bank
deposits and instead allow the exchange rate to be determined by
market conditions. By varying the supply of currency, the central bank
would then be able to keep the return on currency close to the return
on deposits during periods of negative rates. This approach sounds
strange to modern ears, but there is historical precedent. Before the
creation of the Fed, when bank runs led to the suspension of convert98 |

Also with the benefit of hindsight, two other issues would have to
be addressed before adopting any of Marvin’s proposals regarding
currency. First, much of the recent discussion about negative rates has
centered on possible risks to financial stability. If significantly negative
rates make banks, money market funds, or other institutions unprofitable or unstable, it is possible that there is a “reversal” interest rate
below which rate cuts are no longer beneficial.2 Neither of Marvin’s papers mention this issue. Personally, I am less alarmed than some about
this risk: Negative rates, at the levels we have seen them, appear to
have eased broad financial conditions without creating serious problems for financial institutions, over and above the effects of low rates
in general. But this issue is one that would require more study before
rates were cut much more deeply.
2

Brunnermeier and Koby (2018).

| 99

Bernanke
The other barrier to the implementation of Marvin’s proposals is the
difficulty of gaining political support. Changes to the currency, even
regarding (economically) trivial matters like whose face is on the bill,
can be highly controversial. The Fed is independent, but Congress has
ultimate control over the currency and will oppose unpopular measures. Here is my opportunity to praise Marvin’s intellectual integrity.
The proposal in his Woodstock, Vermont, paper to impose a carry tax
on currency created a media firestorm, which was soon followed by a
statement from the Libertarian Party and a proposed bill from Congressman Ron Paul (R-TX) to prohibit fees on currency. The Federal
Reserve Board staff had to assure legislators that research by Fed economists is conducted independently and does not necessarily represent
the views of the Federal Reserve System. I have been told (though
cannot document) that Marvin received threats to his personal safety.
Marvin believed in free intellectual inquiry and was not dissuaded.
He made similar recommendations in the high-profile Jackson Hole
meeting in 2016.
Even though Marvin’s work on the ZLB focused on the constraint
imposed by the availability of zero-interest cash, he looked at other
possible approaches to overcoming the constraint, again always with
thoughtful prescience. In his 2000 paper, beyond negative rates, Marvin anticipated another key tool used by all major central banks at the
ZLB — namely, purchases of longer-term securities, or what today we
call quantitative easing (QE).
Marvin’s early exposition of how QE might work had a more monetarist flavor than most modern accounts. But he did not make the
error of assuming that adding liquidity — in the narrow sense of assets
useful primarily as transactions media — when the economy is in a
liquidity trap would have much stimulative power. He instead took a
broader view of the liquidity services provided by longer-term securities such as Treasuries, including not only the ease with which they can
be turned into cash, but their usefulness as collateral and in providing
other services. He saw central bank purchases of longer-term securities
as affecting the convenience yields of those securities (as reflected in
term premiums) and their close substitutes, in a manner very similar
100 |

The Zero Lower Bound
to the portfolio balance channel emphasized by Bernanke (2020) and
most contemporary discussions of QE. In particular, by raising the values of assets broadly, Marvin saw QE as raising borrower net worth and
thus reducing the external finance premium — the wedge between
the safe rate of return and the cost of borrowing — in the sense of
Bernanke and Gertler (1995).
Does QE have costs as well as benefits? In his 2000 paper, Marvin
focused primarily on the fiscal risks of the central bank holding long
bonds. The possibility of losses on the central bank’s portfolio is indeed a concern for monetary policymakers but mostly for optical and
political reasons. Losses on securities bought in QE programs are not
first-order social costs, any more than are the losses that arise when the
maturity mix of new securities issued by the Treasury turns out not to
have been cost-minimizing. Moreover, any paper losses on central bank
portfolios should be set against the benefits of QE for economic growth
and thus for tax revenues. Finally, central banks can continue to operate
with losses or low levels of capital, so agreements with the Treasury
to replenish central bank capital — as we have seen in the United
Kingdom, for example — relate more to questions of governance and
independence rather than to the feasibility of QE policies.
The fiscal and political concerns about QE are shared by many central bankers. More puzzling is Marvin’s view, stated in 2000 and reiterated in his 2016 paper, that QE risks becoming “inflationary finance.” In
the depressed conditions that followed the global financial crisis, and
with the ZLB binding, the use of QE proved insufficient to get inflation
even up to target, much less to uncomfortable levels. In particular,
with rates very low, the velocity of money fell significantly during the
post-crisis period. Marvin was perhaps concerned that QE would not
be slowed or reversed after the economy no longer required support.
It is true, at least, that unwinding a QE program without disrupting
markets too much can take some time, during which the central bank’s
securities holdings continue to provide stimulus.

| 101

The Zero Lower Bound

Bernanke
Finally, Marvin’s 2000 paper considered yet another strategy for
defeating the ZLB — so-called helicopter money, Milton Friedman’s
term for money-financed tax cuts. Helicopter money — or monetary
transfers, in Marvin’s language — is a combination of tax cuts and QE,
a combination we have seen in many countries during the recent pandemic. We have good reasons to expect that transfers and tax cuts, on
the one hand, and central bank securities purchases, on the other, will
stimulate the economy. The question is whether the combination of
the two policies is any more powerful than the policies taken separately, that is, whether monetary-fiscal coordination buys anything in this
instance. Marvin appreciated in 2000, very early in this debate, that the
answer may be no. Unless the public sees monetary-fiscal coordination
as changing the goals of the central bank in the medium term — for
example, by increasing the amount of inflation it is willing to accept
— the stimulus provided by the combined policies will be roughly
the same as that of the two policies taken separately. For helicopter
money to have extra stimulative power, there must be either a real or
perceived increase in fiscal dominance (loss of central bank independence), which the public believes will lead the central bank to accept
higher inflation than it otherwise would, at least for a time.
The effects of the ZLB, and strategies for overcoming it, was just
one of many critical issues in monetary theory and policy that Marvin
Goodfriend tackled in his career. He was an extraordinary economist.
He was also an extraordinarily kind person, with the gift of telling you
that you are completely wrong while making you feel good about it. I,
among his many friends, will miss him.

References
Bernanke, Ben. 2020. “The New Tools of Monetary Policy.” American
Economic Review 110, no. 4 (April): 943-83.
Bernanke, Ben, and Mark Gertler. 1995. “Inside the Black Box: The Credit
Channel of Monetary Policy Transmission.” Journal of Economic Perspectives 9, no. 4 (Fall): 27-48.
Brunnermeier, Markus, and Yann Koby. 2018. “The Reversal Interest
Rate.” NBER Working Paper 25406, December.
Eggertsson, Gauti, and Michael Woodford. 2003. “The Zero Bound on
Interest Rates and Optimal Monetary Policy.” Brookings Papers on Economic Activity no. 1 (Spring): 139-211.
Goodfriend, Marvin. 2000. “Overcoming the Zero Bound on Interest
Rate Policy.” Journal of Money, Credit and Banking 32, no. 4, part 2 (November): 1007-35.
Goodfriend, Marvin. 2016. “The Case for Unencumbering Interest Rate
Policy at the Zero Bound.” Jackson Hole Policy Symposium, Federal
Reserve Bank of Kansas City, August 26-27.
Krugman, Paul. 1998. “It’s Baaack: Japan’s Slump and the Return of the
Liquidity Trap.” Brookings Papers on Economic Activity no. 2 (Fall) 137-87.
Rogoff, Kenneth. 2016. The Curse of Cash: How Large-Denomination Bills
Aid Crime and Tax Evasion and Constrain Monetary Policy. Princeton, N.J.:
Princeton University Press.
Wolman, Alexander. 1998. “Staggered Price Setting and the Zero
Bound on Nominal Interest Rates.” Federal Reserve Bank of Richmond
Economic Quarterly 84 (Fall), 1-24.

102 |

| 103

Fed Structure and Economic Ideas

Federal Reserve Structure and
Economic Ideas
Michael D. Bordo and Edward S. Prescott1
Marvin Goodfriend’s essay, “The Role of a Regional Bank in a System
of Central Banks,” was written for the November 1998 Carnegie-Rochester conference series on public policy.2 The title of the conference
was “Issues Regarding European Monetary Integration,” which focused
on the European monetary union experiment that was just underway at the time.3 Marvin’s essay was about the institutional design of
the European monetary institutions. His strategy was to describe the
partially decentralized structure of the Federal Reserve System, laying
out the respective roles of the headquarters in Washington and the
regional Reserve Banks, and then to use the American experience to
provide lessons for the newly formed European system. The analogy
was apt due to the strikingly parallel structure of the nascent European
monetary institutions with a headquarters institution — the European
Central Bank in Frankfurt — and numerous regional institutions in the
form of the preexisting national central banks.
While ostensibly about the design of the European monetary system, the essay is also a statement of Marvin’s views about the proper
role of a central bank and a defense of the federal structure of the
Federal Reserve System. The decentralized structure of the Federal Reserve periodically comes under attack from various interests who want
a centralized, less federal system. Marvin’s essay provides an important
antidote to these attacks by laying out the many advantages of the
federal system.
 e would like to thank the editors, Bob King and Alex Wolman, for helpful comments.
W
See Goodfriend (1999a). Marvin’s essay was also reprinted in the Federal Reserve Bank
of Richmond’s annual report in 1999 (Goodfriend, 1999b).
3
The conference agenda is available at https://www.sciencedirect.com/journal/
carnegie-rochester-conference-series-on-public-policy/vol/51/suppl/C.
1
2

104 |

| 105

Bordo and Prescott
In the essay, Marvin stated his philosophy of how a central bank
should operate,
	The overarching principle is that a central bank should provide the
necessary monetary and financial stability in a way that leaves the
maximum freedom of action to private markets. In keeping with this
principle, monetary policy is implemented by direct means, with an
interest rate policy instrument rather than with direct credit controls.
In the banking sphere every effort is made to minimize as far as possible the regulatory burden associated with financial oversight.4

He also stated what a central bank needed to operate in this way.
Marvin believed that a central bank needed independence, credibility,
and an ability to learn about economic ideas and markets. Furthermore, for the United States, he argued that the Reserve Banks played
an important role in meeting these needs. He discussed how the Fed’s
decentralized structure enhanced credibility and supported independence because “… the diffusion of power makes it more difficult for
outside pressures to be brought to bear on a central bank.”5 He also
believed that the regional structure helped gather information and
disseminate information to the various regions of the United States
and helped with bank supervision. Finally, he argued that “… a system
of regional banks led by the center institution harnesses competitive
forces to encourage innovative thinking within the central bank.”6
In this essay, we discuss Marvin’s last point. The other benefits of the
Reserve Banks, while important, are already well known. However, the
idea that the decentralized structure encourages innovative thinking is less appreciated, but it is, we think, one of the System’s biggest
strengths. In his essay, Marvin planted the seed for this idea.7
 oodfriend (1999a), p. 51.
G
Goodfriend (1999a), p. 52.
6
Goodfriend (1999a), pg. 53.
7
See also Wheelock (2000).
4
5

Fed Structure and Economic Ideas
Furthermore, when it comes to innovative thinking, there is no better
person to be honoring than Marvin. As we both know from personal
experience, and many others know as well, Marvin was full of ideas. He
thought for himself, followed economic logic to its conclusions, and
was willing to advocate for his ideas even if they challenged central
bank orthodoxy.
Marvin’s intellectual contributions to central banking and monetary
economics are well known and many of them are described in companion essays in this volume. What we want to emphasize is how the
semi-independence of the regional Reserve Banks allowed a creative
thinker like Marvin to flourish and led to a transformation in thinking about policy both in the System and among central banks more
generally. The key elements provided by a Reserve Bank were direct
exposure to policy problems, via a Reserve Bank’s role on the Federal
Open Market Committee (FOMC) and in the banking and payment
systems, and enough independence from headquarters so that ideas
that challenged an existing orthodoxy could be developed, explored,
and supported over time.
The most striking example of this institutional dynamic with Marvin’s work is his 1986 Journal of Monetary Economics paper in which
he derived from economic principles the costs and benefits of FOMC
transparency.8 While transparency is now taken for granted, it was not
at the time. The prevailing central bank view was that secrecy was valuable for central banking, and, consistent with that view, the reaction
from the Board in Washington to this publication was strong disapproval. However, by being at Richmond, which supported him, his
career in the Federal Reserve was not slowed down and he was able to
flourish. The subsequent change in views by the Federal Reserve and
the central banking community on transparency is a testament to the
value of Marvin’s insights and a prominent example of how an idea can
develop from a Reserve Bank, gestate, and later lead to good reforms
for the institution.
8

106 |

See the essay by Lars E.O. Svensson in this volume.

| 107

Bordo and Prescott
It was no accident that during his time at the Richmond Fed, that
bank developed a reputation for being independent within the System
and its presidents dissented on numerous FOMC votes. Indeed, a creative economist like Marvin would probably have contributed far less
to monetary policy if he had worked at a highly centralized institution.
At a monolithic central bank, Marvin might not have even thought of
proposing these ideas because they would likely have been stopped
before seeing the light of the day.
While it is not unusual for an institution to have internal debate, it
is unusual for an institution to allow some of that debate to appear in
public. For this reason, we think there is some value to describing how
the System evolved to the point where the public competition of ideas
could exist and flourish.9 In this evolution, the Richmond Fed was one
of the early innovators.
The Federal Reserve was designed as a decentralized institution in
1913 with 12 privately chartered Reserve Banks and a Federal Reserve
Board in Washington, DC, that had limited oversight. The structure
was explicitly designed to distribute power throughout the country.
However, as with many other federal institutions, power was centralized by the Roosevelt administration during the Great Depression. The
Banking Act of 1935 moved monetary policy primarily to the Board
of Governors by creating the FOMC and increased the oversight of
the Reserve Banks. However, Congress retained a role for the Reserve
Banks by giving them, on a rotating basis, five of the 12 votes on the
FOMC. Furthermore, they left the Reserve Bank’s corporate structure
with its unique quasi-public governance relatively untouched.10
For the next 15-20 years, the Reserve Banks, other than New York,
played a relatively minor role in monetary policy. This was partly
because the Federal Reserve had become subservient to the Treasury
in the 1930s under Secretary Morgenthau and Chairman Eccles, who
believed in fiscal dominance. And partly because during World War II,

Fed Structure and Economic Ideas
the Fed accommodated Treasury’s war expenditures by setting a low
interest rate peg.11 The subservience was ended by the Treasury-Fed
Accord of 1951 that gave the Federal Reserve monetary independence
and made William McChesney Martin the chairman of the FOMC.12
While the Accord reasserted the Federal Reserve’s independence,
the role of Reserve Banks (other than New York) for monetary policy
remained relatively minor. The FOMC met infrequently, and most decisions were made by an executive committee consisting of the chairman, the New York Fed president, and a few other members. For a variety of reasons, including a battle for control over monetary policy with
the New York Fed, Chairman Martin instituted reforms to the FOMC in
the mid-1950s in which the executive committee was eliminated and
decisions were made by the entire FOMC. This change in operating
procedures gave the presidents of the non-New York Reserve Banks a
more prominent role in monetary policy. Previously, their responsibilities focused on providing banking services and supervising banks in
their regions.13
The other development at this time was external. Starting in the late
1950s, and even more so in the 1960s, the economics profession was
becoming increasingly professionalized. Keynesian ideas for macroeconomic policy were developing in academia, young PhDs were
bringing these ideas into the Federal Reserve, and more formal analysis was being used by the FOMC. Furthermore, the Council of Economic Advisers under the economist Walter Heller was pushing to appoint
Keynesian economists to the Board of Governors.14 The result was an
increased role for economists in leadership positions throughout the
Federal Reserve.15
This change in the internal and external environment created the
conditions that encouraged a Reserve Bank to innovate on monetary
Meltzer (2002).
For more information about the Accord, see Hetzel and Leach (2001a and 2001b)
and Meltzer (2009).
13
Business Week (1956).
14
Bremner (2004).
15
Whittlesey (1963).
11
12

9

The subsequent analysis is based on Bordo and Prescott (2019).
See Bordo and Prescott (2019) for details on Reserve Bank governance.

10

108 |

| 109

Bordo and Prescott
policy. The first innovator was President D.C. Johns of the St. Louis Fed.
Johns felt that he and the other presidents were being ignored by
the Board, so in 1958 he hired Homer Jones, who had taught Milton
Friedman, and soon that bank became closely tied to monetarist ideas
on monetary policy.16 The bank served as a conduit for the monetarist
ideas of prominent economists like Karl Brunner, Milton Friedman, Allan Meltzer, and Anna Schwartz, but it also made important monetarist
contributions such as Andersen and Jordan (1968) on monetary versus
fiscal policy. A sequence of presidents and the research department
leadership provided enough organizational continuity that the St.
Louis Bank was able to support monetarist ideas through at least the
1990s. A progression like this could exist only with enough separation
from Washington to develop and maintain independent ideas.
The next innovator was the Minneapolis Fed, which, starting in the
1970s, became closely associated with and contributed to the rational
expectations and dynamic stochastic general equilibrium revolution
in macroeconomics. Even more so than St. Louis, Minneapolis was
actively involved in the development of academic ideas, particularly
rational expectations. Much of this work was done jointly and in partnership with the University of Minnesota, which was only about two
miles away from the bank. There are three especially notable examples.
Tom Sargent and Neil Wallace worked on rational expectations while
professors at the University of Minnesota and consulting with the Minneapolis Fed.17 Ed Prescott’s real business cycle methodology (developed with Finn Kydland) led to important changes in macroeconomic
methods, and their identification of time consistency problems with
optimal macroeconomic policy led to a renewed emphasis on monetary policy credibility and the role that institutional structure plays in
providing that credibility.18 Chris Sims developed pathbreaking time
series methods.19
S ee Melzer (1989). Following D.C. Johns, President Daryl Francis, who served 19661976, championed monetarist ideas at FOMC meetings.
17
See, for example, Sargent and Wallace (1975, 1981).
18
Kydland and Prescott (1977, 1982). Alesina and Summers (1993) showed that independent central banks did a better job of controlling inflation.
19
Sims (1980).
16

110 |

Fed Structure and Economic Ideas
The monetarist ideas associated with St. Louis and the rational
expectation ideas associated with Minneapolis gained increased
attention in the 1970s due to the high inflation and other failures of
the Keynesian ideas of the time. It was during this period of intellectual
and economic ferment that Marvin was hired by the Richmond Fed in
1978. The late 1970s was a particularly auspicious time for an energetic
and creative economist like Marvin to start at the Fed. Inflation was
over 8 percent in 1978 and would reach 14 percent in 1980. Paul Volcker would become chairman in 1979 and the FOMC would then start to
dramatically raise the fed funds rate. The Monetary Control Act of 1980
would change the role of the Federal Reserve in the payment system,
and the increasing number of thrift and bank failures would highlight
the importance of bank regulation, deposit insurance, and Federal Reserve lending facilities. It was an exciting time intellectually to work on
money and banking research and it was an exciting time to do policy,
both at which Marvin excelled.
The environment in Richmond when Marvin arrived was not that of
a modern research department with an emphasis on academic publications, but it was moving in that direction. The intellectual interest
was there. The president at the time was Bob Black, who was trained
as an economist and had become sympathetic to monetarist thinking. His colleagues in the department included Al Broaddus, who led
the macroeconomists and later became research director and then
president; Bob Hetzel, who was a student of Milton Friedman’s and a
monetarist; and Tom Humphrey, who, with his background in history
of economic thought, was resurrecting intellectual interest in the Fed’s
role as lender of last resort by studying lessons from Henry Thornton
and Walter Bagehot. When Marvin joined, the spark was lit for the
department to take off.
With the support of an ambitious institution and amid the exciting
intellectual debates at the time, Marvin thrived. Anyone who worked
with him will remember his excitement when discussing economic
ideas, or monetary policy operating procedures, or just about any

| 111

Bordo and Prescott
other topic associated with the Fed. In these discussions, he consistently linked research and monetary policy.20 Relative to St. Louis and
Minneapolis, Richmond’s innovation was to closely integrate research
and the policy process. During Marvin’s tenure there, this was best
represented by the team he made with Al Broaddus, who was president from 1993 to 2004. As Mark Gertler mentions in his essay in this
volume, at the time he may have been unique within the System in
both actively doing research while being the senior monetary policy
advisor.21
The value of the St. Louis, Minneapolis, and Richmond models have
been recognized within the System. Today, virtually every Reserve
Bank has a thriving research department and to varying degrees policy
and research complement each other in the way that Marvin exemplified.22 Some Reserve Banks partner with local universities, and all
interact with academics. There is a regular flow of ideas within the Fed
and with the outside, and research results are actively part of FOMC
discussions. For example, at a 2005 FOMC meeting, San Francisco Fed
President Janet Yellen stated,
 considerable body of research — most conducted within the FedA
eral Reserve System — has examined the possibility that the last recession and recovery were characterized by unusually large structural
shifts, resulting in an exceptional degree of mismatch in the labor
market. If an unusually small fraction of the currently unemployed
are qualified for existing or emerging job vacancies, the true degree
of slack in the labor market is overstated by measured unemployment. In effect, the NAIRU has risen. This possibility motivates one of
the alternative simulations in the Greenbook. At the AEA [American
Economic Association] meetings in Philadelphia last month, I chaired

Fed Structure and Economic Ideas
	a session in which four teams of Fed economists subjected this structural-shifts hypothesis to close scrutiny. I emerged from this session
a skeptic. I see this recent research as casting considerable doubt on
the hypothesis that the jobless recovery was a period of pronounced
economic restructuring.23

As we said earlier, Marvin took some controversial stands. In his case,
we can see the important role of Fed structure in supporting debate
and differing views. His transparency work challenged the orthodox
view at the Board at the time, and his work would likely not have been
published if he had worked in a more centralized institution. However,
due to the Reserve Bank’s structure, each with its own president and
Board of Directors, the bank was able to support him and keep him at
Richmond, which was to the long-term benefit of the Richmond Fed
and the System as a whole.
What is striking in rereading Marvin’s essay is that his ideas are not
just abstract arguments weighing the pros and cons of the System’s
structure. Instead, they are based on what he observed and experienced during his career. For those of us who were fortunate to have
been able to talk over these and so many other topics with him, we
deeply miss him.

23

 hile Marvin is most associated with monetary policy, he also wrote on banking and
W
payments policy (Goodfriend and Hargraves, 1983, and Goodfriend, 1990).
21
We don’t want to give the impression that other Reserve Banks didn’t develop ideas
during this period as well. For example, in the late 1980s, Chicago became associated with deposit insurance reform and Cleveland became associated with inflation
targeting. For more details and other examples, see Bordo and Prescott (2019).
22
Marvin’s innovations were also recognized internationally. For example, the European Central Bank asked him to undertake a review of their research activities (Goodfriend, König, and Repullo, 2004).
20

112 |

F OMC Meeting Transcript, February 1-2, 2005, p. 87. https://www.federalreserve.gov/
monetarypolicy/files/FOMC20050202meeting.pdf.

| 113

Bordo and Prescott

References
Alesina, Alberto, and Lawrence H. Summers. 1993. “Central Bank
Independence and Macroeconomic Performance: Some Comparative
Evidence.” Journal of Money, Credit and Banking 25, no. 2 (May): 151-162.
Anderson, Leonall C., and Jerry L. Jordan. 1968. “Monetary and Fiscal
Actions: A Test of Their Relative Important in Economic Stabilization.”
Federal Reserve Bank of St. Louis Economic Review 50 (November): 1124.
Bordo, Michael D., and Edward S. Prescott. 2019. “Federal Reserve
Structure, Economic Ideas, and Monetary and Financial Policy.” National Bureau of Economic Research Working Paper 26098, July.
Bremner, Robert P. 2004. Chairman of the Fed: William McChesney Martin
Jr. and the Creation of the American Financial System. New Haven: Yale
University Press.
Business Week. 1956. “The Banker’s Bank in Cleveland: A Leader’s Role.”
March 17, pp. 187-196.
Federal Open Market Committee. 2005. “Meeting of the Federal Open
Market Committee, February 1-2.”
Goodfriend, Marvin. 1986. “Monetary Mystique: Secrecy and Central
Banking.” Journal of Monetary Economics 17, no. 1 (January): 63-92.
Goodfriend, Marvin. 1990. “Money, Credit, Banking, and Payments System Policy.” In The U.S. Payments System: Efficiency, Risk and the Role of
the Federal Reserve, edited by David B. Humphrey. Philadelphia: Kluwer
Academic Publishers.

Fed Structure and Economic Ideas
Goodfriend, Marvin, and Monica Hargraves. 1983. “A Historical Assessment of the Rationales and Functions of Reserve Requirements.” Federal Reserve Bank of Richmond Economic Review 69 (March/April): 3-21.
Goodfriend, Marvin, Reiner König, and Rafael Repullo. 2004. “External
Evaluation of the Economic Research Activities of the European Central
Bank.” February 20.
Hetzel, Robert L., and Ralph F. Leach. 2001a. “The Treasury-Fed Accord:
A New Narrative Account.” Federal Reserve Bank of Richmond Economic
Quarterly 87, no. 1 (Winter): 33-55.
Hetzel, Robert L., and Ralph F. Leach. 2001b. “After the Accord: Reminiscences on the Birth of the Modern Fed.” Federal Reserve Bank of
Richmond Economic Quarterly 87, no. 1 (Winter): 57-64.
Kydland, Finn E., and Edward C. Prescott. 1977. “Rules Rather than Discretion: The Inconsistency of Optimal Plans.” Journal of Political Economy 85, no. 3 (June): 473-492.
Kydland, Finn E., and Edward C. Prescott. 1982. “Time to Build and Aggregate Fluctuations.” Econometrica 50, no. 6 (November): 1345-1370.
Meltzer, Allan H. 2002. A History of the Federal Reserve, Vol. 1, 1913-1950.
Chicago: University of Chicago Press.
Meltzer, Allan H. 2009. A History of the Federal Reserve: Vol. II, Book One,
1951-1969. Chicago: University of Chicago Press.
Melzer, Thomas C. 1989. “The Making of a ‘Maverick’ Monetary Policymaker.” Remarks to the Newcomen Society of the United States, St.
Louis, Missouri, November 8.

Goodfriend, Marvin. 1999a. “The Role of a Regional Bank in a System of
Central Banks.” Carnegie-Rochester Conference Series on Public Policy 51
(December): 51-71.

Sargent, Thomas J., and Neil Wallace. 1975. “‘Rational’ Expectations, the
Optimal Monetary Instrument, and the Optimal Money Supply Rule.”
Journal of Political Economy 83, no. 2 (April): 241-254.

Goodfriend, Marvin. 1999b. “The Role of a Regional Bank in a System of
Central Banks.” Federal Reserve Bank of Richmond Annual Report, 2-15.

Sargent, Thomas J., and Neil Wallace. 1981. “Some Unpleasant Monetarist Arithmetic.” Federal Reserve Bank of Minneapolis Quarterly
Review 5, no. 3 (Fall): 1-17.

114 |

| 115

Bordo and Prescott
Sims, Christopher A. 1980. “Macroeconomics and Reality.” Econometrica
48, no. 1 (January): 1-48.
Svensson, Lars E.O. “Monetary Mystique and the Fed’s Path Toward
Transparency.” In Essays in Honor of Marvin Goodfriend: Economist and
Central Banker, edited by Robert G. King and Alexander L. Wolman.
Richmond: Federal Reserve Bank of Richmond. Available at www.richmondfed.org/goodfriend.
Wheelock, David C. 2000. “National Monetary Policy by Regional Design: The Evolving Role of the Federal Reserve Banks in Federal Reserve
System Policy.” In Regional Aspects of Monetary Policy in Europe, edited
by Jurgen von Hagen and Christopher J. Waller, 241-274. Philadelphia:
Kluwer Academic Publishers.
Whittlesey, C.R. 1963. “Power and Influence in the Federal Reserve
System.” Economica 30, no. 117 (February): 33-44.

116 |

| 117

Banking Policy and Monetary Policy

Banking Policy and Monetary Policy1
Douglas W. Diamond
Marvin Goodfriend was a great economist.2 He had a deep understanding of formal theory, but he was also very intuitive and policy
oriented. He was rigorous, but he did not use large amounts of math in
his theoretical work. One of Marvin’s especially insightful and influential papers was Goodfriend and King (1998), hereafter referred to
as GK. It discusses a framework for determining if monetary policy,
implemented with open market purchases of Treasury securities, is a
necessary and sufficient response to a banking crisis. The other possible response is banking policy. Banking policy in GK, whenever it is
not redundant, is lending to banks at a rate below the one they could
obtain in the market (especially lending to banks that cannot borrow
in the market). Marvin later labeled such interventions as a component
of “credit policy,” but I stick to the earlier terminology and consider the
specific interventions discussed in GK. This essay discusses the ideas
and contributions of GK and relates it to some of Marvin’s other research on banking and monetary policy.

Core elements of GK and my own perspectives
GK provides what is almost a Modigliani and Miller (1958) theorem
for banking policy. GK gives conditions under which banking policy
has no beneficial effects. The result has two parts: one for idiosyncratic
banking policy and the other for system-wide banking policy. The first
part deals with lending to individual banks experiencing idiosyncratic
funding problems (lending to banks that cannot raise enough deposits
or interbank loans to survive). GK outlines a proof that there is no social benefit to this (given an appropriate monetary policy) if individual
bank failures can be treated like individual business failures. This is true
in partial equilibrium and it seems to me that this is true more generally if individual bank failures generate no externalities.
I am grateful to Bob King, Alex Wolman, and Luke Zinnen for helpful comments on an
earlier draft.
2
I met Marvin when I was an undergraduate and will discuss my interactions with him
more fully at the end of this essay.
1

118 |

| 119

Diamond
Banking policy is a special form of credit policy that targets the
rescue of individual banks, primarily by lending to them when the
market will not. This is a case of lending to (bailing out) the exogenously insolvent that are worth more dead than alive and is consequently
undesirable.3 Even worse, when anticipated, it requires ex-ante regulation to prevent excessive ex-ante risk-taking that takes advantage of its
subsidized bailout.
The second part of the paper is about banking policy in system-wide
financial crises. It is very short but makes several important points. It
argues that monetary policy implemented by open market operations
is necessary and sufficient to respond to some types of banking crises.
The basic model is close to the one implicit in the celebrated Great
Contraction chapter in Friedman and Schwartz (1963), which describes
the period around 1929 in the United States. In essence, Friedman and
Schwartz (1963) views a system-wide banking crisis as a temporary
increase in the demand for currency. A monetary policy that provides
an elastic currency response to the demand increase (via open market
operations expanding the supply of high-powered money) will prevent such crises from bringing down the banking system. In Friedman
and Schwartz (1963), the macroeconomic losses from a crisis are due
to the implications of the induced fall in the stock of money due to
bank failures.
GK considers more general reasons for social losses. I read the GK
analysis to include the fire-sale losses from selling assets that arise
when the supply of currency is too small to meet the crisis-induced
demand. The key conditions for the sufficiency of open market operations are fully integrated markets for government bonds, bank deposits, and interbank loans. Injecting bank reserves into these integrated
markets will offset increases in the demand for currency that cause a
banking crisis.
As a part of this analysis, GK shows that the United States Federal
3

120 |

I f the authority conducting a bailout could have better information than depositors,
GK points out that the argument becomes more nuanced.

Banking Policy and Monetary Policy
Reserve System’s interest rate smoothing policy (providing an elastic
currency to offset all temporary shocks to the demand for currency)
will work just as well as Bagehot’s lender of last resort policy. Bagehot
allows moderate spikes in interest rates (spikes consistent with bank
solvency) when borrowing from the lender of last resort. Much of this
part of GK is under the assumption that banking crises are caused by
an increase in the demand for currency relative to deposits and that
there are no other real shocks. GK shows that monetary policy is necessary and sufficient to offset these monetary shocks.
GK also considers a real, nonmonetary shock: a temporary increase
in the real rate of interest. This shock could cause bank failures that
monetary policy (without banking policy) could not prevent. In essence, an increase in the real rate required by bank deposits can cause
a run by making some banks insolvent. No amount of liquidity can
overcome this insolvency because the problem is a real shock and not
a shock to the demand for currency versus deposits. They argue that
the bank failures could cause economic dislocations, but that these are
similar to the dislocations caused by the effect of solvency of industrial
firms. This makes bailouts via banking policy somewhat unlikely to be
desirable. The desirability of banking policy is a cost-benefit analysis
based on magnitudes. An alternative view of the implications of an
increased real rate of interest described in GK is that many banks might
be on the border between full insolvency and insolvency only due to
the illiquidity of their assets — these banks are solvent but illiquid. A
real rate shock pushes some banks over the insolvency border. If there
is incomplete information on the exact value of the individual banks,
it could lead all banks near the border to fail in a crisis. If the problem
is this incomplete information about which banks are fully solvent, GK
suggests that it could be beneficial to have detailed ex-ante supervision and monitoring of banks to determine in advance which ones are
insolvent.
I have a somewhat different view of the desirability of banking
policy and favor a lender of last resort policy that goes beyond open
market operations. Nonetheless, the analysis in GK greatly clarifies
the important issues in understanding banking policy. In my view, the
| 121

Diamond
case for saving a just-insolvent bank is almost identical to a bank that is
slightly solvent if there are large social and external costs to bank failure.
I view illiquidity and solvency as part of a general equilibrium effect
with externalities, as in Diamond and Rajan (2005, 2012). In addition, a
forecast of a run can be self-fulfilling if enough depositors believe that
no one will lend directly to the bank or banking system, as in Diamond
and Dybvig (1983). One way to approximate these two points in the GK
framework is that bank failures themselves could have an impact on the
real rate of return that depositors require on bank deposits and other assets. An open market operation to buy Treasury bills with high-powered
money in response to a banking crisis will not work if the holders of the
high-powered money or currency are not willing to lend to the banks at
an interest rate that leaves the bank solvent.
If the bad effects of bank failures are very large, then banking policy
can be desirable ex-post. GK notes that banking policy is a form of desirable insurance beyond the insurance against money demand shocks
provided by open market operations.4 Because a lender of last resort
that goes beyond open market operations provides a discretionary
injection of subsidized funds with ex-post benefits to society, it (like
monetary policy) faces a problem of limited commitment. Marvin and
Jeff Lacker provided a very nice analysis of this in Goodfriend and Lacker (1999)5; see also, the related analysis in Diamond and Rajan (2012).
GK notes that insurance that provides a subsidy generally requires
ex-ante pricing and regulation. GK does not discuss deposit insurance, but it would be interesting to understand how their perspective
applies to this pervasive feature of modern banking systems. Unlike a
lender of last resort, deposit insurance is explicit and contractual. This
commits the deposit insurer and avoids the discretionary element of
lender of last resort policy.

 recent analysis of lender of last resort policy as explicit insurance is presented in
A
Mervyn King (2016).
5
See the essay by Athreya and Williamson elsewhere in this volume.
4

122 |

Banking Policy and Monetary Policy

Later work on banking policy, monetary policy,
and macroeconomics
Marvin was one of the pioneers who brought ideas from banking
theory into macroeconomics. I thought it was a sad situation when
“Money and Banking” (the former name for this field of research)
became a completely inaccurate description of the current research in
macroeconomics.6 Many know Marvin best from his work on monetary
economics, but I found him to be aware of developments in banking,
both practical and theoretical. Marvin did continue to work on banking topics intermittently, and I briefly comment on two later, related
contributions. Goodfriend (2005) describes how frictions in the banking sector can impact the effect of shocks to the economy and also
influence the information content of various interest rates as indicators
of macro shocks. Goodfriend and McCallum (2007), GM hereafter, took
these ideas and made them quantitative. It has very useful implications for monetary policy, providing a different approach to using yield
spreads as measures of economic conditions than those provided in
Bernanke, Gertler, and Gilchrist (1999).
GM introduces a wedge between banking markets and Treasury
markets and studies the interest rate spread between bank rates and
Treasury rates. Marvin had an open mind. He believed that open market operations were the right way to implement monetary and banking policy, with the least moral hazard, but that bank liquidity services
and the collateral role of banks complicated the argument. He viewed
the banking sector as important, along with the final goods production sector, because loan production was subject to a moral hazard
problem as in Diamond (1984).
The GM model allows differentiated financial products where loans
differ from bonds and where collateralized bank deposits are different
from uncollateralized liabilities. GM’s integrative model features New
Keynesian sticky price frictions and a deposit in advance constraint
(similar to a cash in advance constraint).
6

 s I describe below, Marvin had previously encouraged me to attempt to bring more
A
of banking theory into macroeconomics.

| 123

Diamond
GM does not assume that currency, government bonds, bank
deposits, and bank loans are in an integrated market as in GK. There
are separate markets for each, with separate shocks to supply and
demand. There can be yield spreads between them, which could be
time varying. Open market operations would be sufficient to offset any
shock in the markets, but the size of the intervention cannot be determined by examining a single rate of interest. For example, interest rate
smoothing of government bonds would not be sufficient to offset all
shocks. The central bank also needs to look at the various yield spreads
to determine the economic shocks that require a monetary policy
response.
GM does not consider banking policy, but it does generalize the
framework of GK. The focus in GM is on the measurement needed to
calibrate an appropriate monetary policy. Implicitly, an appropriate
open market operation can deal with any bank funding problems that
need to be addressed.
Reviewing the GM framework, I have some similar reservations as
those I expressed earlier about the GK setup. I’d prefer a framework
that could allow bank failures or runs to temporally impact the real
rate of interest required by those individuals and institutions funding banks. My view is that the supply of short-term funding to banks
and other financial institutions is not sufficiently integrated with the
market for Treasury securities (substitution across the markets is too
limited) for increased purchases of Treasuries to substitute for a lender
of last resort lending directly to all or most banks using bank loans as
collateral. It might even be possible that lending to them in times of
crisis could be profitable for a central bank and could result in smaller
disruptions than an alternative policy that is restricted to open market
operations in Treasury security markets.

Banking Policy and Monetary Policy
Grossman and Bill Poole. I did not know either of them well, and these
banking issues were not ones that any of us were then thinking about.
Much later, I talked to Marvin many times when he was a visiting
professor at the University of Chicago, and to both Marvin and Bob
after 1990 when I became a long-term visiting scholar at the Federal
Reserve Bank of Richmond. Marvin was an incredibly kind person who
was very generous with his time. In addition to his outstanding skill as
an economist, he was a great electric guitar player.
Marvin said to me many times that he saw important insights for
macroeconomics in the microeconomic theory of financial intermediation, but that the theories needed to be embedded into the standard
quantitative macroeconomic models in order to actually have that
impact. He encouraged me to work in this area, but I did not have
the skills or the inclination to undertake this integration. But working
alone and with Ben McCallum, Marvin later provided his own approach
to this integration, as described above.

Summing up
Marvin Goodfriend had deep insights into money and banking. I
regret that Marvin did not continue to work to integrate the analysis
in GM with the study of a lender of last resort as in GK, but, of course,
there are opportunity costs in research as elsewhere. His combination
of a clear theoretical perspective, a detailed understanding of institutions, and an open mind led him to make major contributions to
economic policy analysis.

Marvin as a wonderful colleague
I have had many long discussions with Marvin about research. I met
Marvin at Brown University in 1974 or 1975 when he and Bob King
were graduate students and I was an undergraduate student taking
graduate macroeconomic and monetary theory courses with Herschel
124 |

| 125

Diamond

References
Bernanke, Ben S., Mark Gertler, and Simon Gilchrist. 1999. “The Financial Accelerator in a Quantitative Business Cycle Framework.” In Handbook of Macroeconomics, vol. 1C, edited by J.B. Taylor and M. Woodford,
1341-93. Amsterdam: North-Holland Publishing Co.

Banking Policy and Monetary Policy
King, Mervyn. 2016. The End of Alchemy: Money, Banking, and the Future
of the Global Economy. London: W.W. Norton.
Modigliani, Franco, and Merton H. Miller. 1958. “The Cost of Capital,
Corporation Finance and the Theory of Investment.” American Economic
Review 48, no. 3 (June): 261-97.

Diamond, Douglas W. 1984. “Financial Intermediation and Delegated
Monitoring.” Review of Economic Studies 51, no. 3 (July): 393-414.
Diamond, Douglas W., and Raghuram G. Rajan. 2005. “Liquidity Shortages and Banking Crises.” Journal of Finance 60, no. 2 (April): 615-47.
Diamond, Douglas W., and Raghuram G. Rajan. 2012.“ Illiquid Banks,
Financial Stability and Interest Rate Policy.” Journal of Political Economy
120, no. 3 (June): 552-91.
Friedman, Milton, and Anna Schwartz. 1963. A Monetary History of the
United States 1867-1960. Princeton: Princeton University Press.
Goodfriend, Marvin. 2005. “Narrow Money, Broad Money, and the
Transmission of Monetary Policy.” In Models and Monetary Policy: Research in the Tradition of Dale Henderson, Richard Porter, and Peter Tinsley, edited by J. Faust, A. Orphanides, and D. Reifschneider. Washington,
DC: Board of Governors of the Federal Reserve System.
Goodfriend, Marvin, and Robert King. 1988. “Financial Deregulation,
Monetary Policy, and Central Banking.” In Restructuring Banking and
Financial Services in America, edited by William S. Haraf and Rose Kushmeider, 216-53. Washington, DC: American Enterprise Institute.
Goodfriend, Marvin, and Jeffrey M. Lacker. 1999. “Limited Commitment
and Central Bank Lending.” Federal Reserve Bank of Richmond Economic Quarterly 85, no. 4 (Fall): 1-27.
Goodfriend, Marvin, and Bennett T. McCallum. 2007. “Banking and
Interest Rates in Monetary Policy Analysis: A Quantitative Exploration.”
Journal of Monetary Economics 54, no. 5 (July): 1480-1507.

126 |

| 127

Interest Rate Smoothing

Interest Rate Smoothing
Michael Dotsey, Andreas Hornstein, and Alexander L. Wolman
The concept of interest rate smoothing behavior by central banks
is now standard, but it was not when Marvin Goodfriend wrote in
the early 1980s.1 His “Interest Rate Smoothing and Price Level Trend
Stationarity” made the concept into a distinct phenomenon to be
explained, first in working papers that appeared in 1983 or earlier and
then in a 1987 publication in Journal of Monetary Economics. At the
time, the conduct of monetary policy as well as the interest rate and
inflation outcomes produced by it were changing rather dramatically.
We begin with a summary of Marvin’s “Interest Rate Smoothing”
paper. We then use the paper as a window into the changes that were
taking place, and continued to take place, both in the practice and
the analysis of monetary policy. Next, we discuss how the literature
on interest rate smoothing evolved after Marvin’s work and reflect on
how the practice of interest rate smoothing seems to have evolved in
recent years in the US. In the concluding section, we offer some personal recollections from our time working with (and for) Marvin at the
Richmond Fed.

Summary of the paper
In the early 1980s, the US economy was emerging from a period of
high and volatile inflation, accompanied by high unemployment and
low output growth — the so-called stagflation of the 1970s. This was
fertile ground for studying why the Fed made the policy choices it
1

128 |

 Google search on “interest rate smoothing” turns up only a handful of references
A
that predate Marvin’s paper. Most of those are also by Marvin and coauthors, the exception being an NBER working paper version of Ben McCallum’s classic paper (1981)
on price level determinacy with an interest rate instrument. But smoothing is not a
central component of McCallum’s analysis.

| 129

Interest Rate Smoothing

Dotsey, Hornstein, and Wolman
made, and how those choices determined the behavior of inflation and
influenced the behavior of real variables. A large part of Marvin’s career
was devoted to studying these issues.
In “Interest Rate Smoothing,” he focused specifically on the nontrend
stationarity of the price level: amid calls for constant money growth
rules that would presumably guarantee price-level trend stationarity as
well as limit inflation, Marvin asked why was such a policy not chosen?
As a student of central banking, Marvin believed it unlikely that
failure to follow a constant money growth rule was due to ignorance
on the part of the central bank. Thus, to answer the question, Marvin
developed a simple model where the monetary authority can choose a
rule for money growth that may or may not result in a trend-stationary
price level. He shows that if the monetary authority is only interested
in stabilizing the macroeconomy, then the optimal policy will indeed
result in a trend-stationary money stock and price level. But if the
authority also cares about financial market stability, then they would
accept random drift in the money stock and the price level would no
longer be trend stationary.
His theoretical foundations were those of the times: a linear rational
expectations model with a Lucas supply curve, a Fisher equation for
nominal interest rates, and a money demand equation. Each one of
the model’s three equations is subject to an iid shock. The monetary
authority observes the current interest rate but not the current price
level or output, or the current shocks for that matter. The monetary
authority chooses the parameters of a money supply growth rule that
responds to contemporaneous interest rate surprises. In addition, the
monetary authority can choose to offset last period’s unanticipated
money stock change that follows from the response to surprise movements in the interest rate. The objectives of the monetary authority
relate to the stabilization of the macroeconomy and the financial
sector. Instability of the macroeconomy is represented by the variance
of price-level surprises and the variance of expected inflation. Instability of the financial sector is represented by the variance of nominal
interest rates.

130 |

First, Marvin shows that if the monetary authority is only concerned
with macroeconomic stability, the optimal policy will exactly offset any
past surprise money stock change such that the money stock becomes
trend stationary. With a trend-stationary money stock, the price level
is also trend stationary. In particular, the price level is an iid random
variable, so unconditional expected inflation is zero. Conditional on
observing the current price level though, expected inflation is the opposite of the current price-level surprise, and therefore the variance of
expected inflation and the price-level surprise are the same. The monetary authority’s optimal contemporaneous response to the interest rate
surprise is then set such that there is no expected price-level surprise.
That is, based on the monetary authority’s current information, the expected price level is the same as the unconditional expectation of the
price level, which then implies that expected inflation is also zero.2
After showing that optimal macroeconomic stabilization indeed
implies trend stationarity of the money stock and the price level, Marvin then demonstrates that adding the additional objective of financial market stabilization — which translates into smoothing nominal
interest rates — introduces random drift into the optimal money stock
and the price level. To understand this result, first note that the Fisher
equation determines the nominal interest rate as the sum of expected
inflation and the stochastic real rate, which is iid. With the previously
optimal trend-stationary policy, expected inflation conditional on the
observed interest rate is always zero. This means that the observed
nominal interest rate reflects the conditional expectation of the real
rate shock. Now suppose that the nominal interest rate is above average and the inferred real rate shock is positive. Then the monetary
authority could reduce the nominal interest rate by creating expected
deflation, that is, an expected reduction in next period’s price level. To
reduce the price level, the monetary authority would reduce the money stock by more than expected; that is, it would more than offset
2

 onditional on their information set, the monetary authority always expects next
C
period’s inflation rate to be zero. However, the monetary authority knows that in the
current period it is offsetting last period’s price-level surprise. This means that the
monetary authority generally expects current-period inflation to be nonzero conditional on their information set, which does not include the current price level.

| 131

Dotsey, Hornstein, and Wolman
last period’s unanticipated money stock change. But this introduces
drift into the money stock. Optimal policy then trades off reductions in
interest rate volatility against increases in expected inflation volatility.
The monetary authority’s objectives are crucial for Marvin’s account
of the stochastic drift in the price level under optimal policy. Given
the reduced-form nature of the model, these objectives don’t arise
from the economic environment as in modern macroeconomics, but
as a student of central banking Marvin motivates them by referring
to expressed or inferred preferences of central banks. Minimizing the
variance of price-level surprises is related to stabilizing employment,
which seems reasonable in the context of the Lucas supply curve.3
Minimizing the variance of expected inflation is motivated by a reference to central banks’ concern arising from imperfect indexation of
nominal contracts. Finally, minimizing the variance of nominal interest
rates is motivated by central banks’ preference for “smooth nominal
interest rates to maintain ‘orderly money markets.’” Note that such behavior has the central bank working against “natural rate” movements
to some degree, rather than simply tracking them, as in Goodfriend
and King (1997) and much subsequent analysis with New Keynesian
models.4

Marvin’s evolving views about monetary policy
Marvin’s paper provides a window, circa 1985, into how he thought
about monetary policy and into the tools that were widely used for
studying monetary policy. Both the conduct of monetary policy and
the analytical tools changed substantially over the course of Marvin’s
career. As discussed in other essays in this volume, Marvin’s work at the
 n the other hand, Woodford (2003a, p. 92) notes that price-level trend stationarity
O
is more attractive if the monetary authority’s objective is the variance of long-horizon
price-level forecast errors. In this case, the desire to smooth interest rates may well
be consistent with trend stationarity, because otherwise the variance of long-horizon
forecast errors is increasing with the horizon. Marvin does acknowledge that low
variance of long-horizon forecast errors might be desirable since it reduces the real
rate of return variance of long-term nominal assets.
4
Goodfriend and King (1997) is the subject of an essay by Michael Woodford elsewhere in this volume.
3

132 |

Interest Rate Smoothing
intersection of theory and policy put him at the center of these changes. We think it’s interesting to use the paper as a departure point for a
brief discussion of how the practice and analysis of monetary policy
changed from the time Marvin began work on “Interest Rate Smoothing” to the early 21st century.

The monetary policy instrument
Although the title of Marvin’s paper refers to “interest rate smoothing,” it is notable that the money supply is the policy instrument in
his model. This may suggest that Marvin’s analysis reflects large differences in the way monetary policy was implemented in the 1980s
compared to today. It is more likely though, that it reflects Marvin
straddling the boundary of academic monetary economics and central
bank practice.
At the time, there was much less transparency about policy than
there is today (see Goodfriend, 1986, and Lars E.O. Svensson’s essay in
this volume). To the extent that the Fed did make public statements
about policy, those concerned broad target ranges for money supply
growth and the federal funds rate.5 Subsequently, policy has become
much more transparent, and — at least away from the zero bound
— policy decisions are framed in terms of short-term interest rates.
Attention to the money supply gradually faded until it essentially disappeared from policy discussions by the early 2000s, if not earlier.
But it is also true that most academic research on monetary policy
at the time treated the money supply as the central bank instrument.
In particular, Marvin’s paper is all about why optimal monetary policy
would deviate from one leading optimal policy prescription, which
called for a constant money growth rule. There was also an ongoing
debate over whether the choice of the interest rate as a policy instrument would lead to instability, especially in the case of interest rate
pegs, for example, in Sargent and Wallace (1975) and McCallum (1981).
At the same time, as we noted above, Marvin was an avid student
5

S ee, e.g., the record of policy actions from February 1983 https://www.federalreserve.
gov/monetarypolicy/files/fomcropa19830209.pdf.

| 133

Dotsey, Hornstein, and Wolman

Interest Rate Smoothing

of central banking, and he saw that actual US monetary policy was
implemented by means of an interest rate instrument. In fact, footnote
six in “Interest Rate Smoothing” previews the main themes of his 1991
paper, “Interest Rates and the Conduct of Monetary Policy.”6 There,
Marvin writes that the Fed has “always employed either a direct or indirect Federal Funds Rate policy instrument.”

rates besides financial stability considerations. One of the reasons is
Mankiw’s (1987) observation that it is optimal to smooth the inflation
tax and thus make inflation a random walk. Barro (1989) analyzed how
a central bank would implement such a policy. Although the idea of
an optimal inflation tax is intriguing, we know of no evidence that US
monetary policymakers ever considered smoothing the Treasury’s
revenue stream as an important consideration for policy.

The meaning of interest rate smoothing

More persuasive is Marvin’s observation that it is longer-term interest rates that primarily influence macroeconomic behavior, and
having a substantial impact on longer-term rates generates a desire for
persistence in short-term rates. Because financial markets are forward
looking, inertial behavior by the central bank can translate a small
change in the current funds rate into meaningful changes in longer-term rates of longer maturities. This feature is also highlighted by
Sack and Wieland (2000). Not wanting to “whipsaw” financial markets
based on every new piece of information also can explain why rates
tend to be adjusted gradually and unidirectionally. Affecting market
behavior at longer horizons is also part of the reason that starting in
the early 2000s the FOMC has increasingly resorted to forward guidance.

Clearly, even in 1987 Marvin viewed interest rate smoothing as being
related to the adjustment of the interest rate instrument over time. But
the absence of any true dynamics in his theoretical framework constrained him to identify interest rate smoothing with minimizing the
variance of interest rates. The development of New Keynesian dynamic
stochastic general equilibrium models, with nominal rigidities and interest rate rules for monetary policy, allowed for an expanded analysis
of interest rate smoothing.

Subsequent literature on smoothing
Over time, central banks moved to explicit interest rate policy and
became more transparent. As analytical tools were developed, a large
literature arose using the broader notion of interest rate smoothing
as representing high persistence of policy interest rates. That the fed
funds rate target is highly persistent at a quarterly frequency, in the
sense of first-order autocorrelation, is an established fact. The interpretation, however, is not obvious. The more recent literature has addressed theoretical explanations for this persistence and then attempted to evaluate those candidate explanations with empirical studies.

Why make interest rates persistent?
Subsequently, in Goodfriend (1991), Marvin considered additional
reasons for why central banks might want to smooth interest

Other potential reasons for interest rate smoothing involve the facts
that data are imperfectly measured and often revised. Additionally,
some of the important underlying variables such as the output gap
and the natural rate of interest are not directly observable. A number
of researchers have shown that these data features generate a rationale for interest rate smoothing, as can the presence of model uncertainty.7
In addition to these heuristic motivations for interest rate smoothing, Woodford (2003b) has shown that even if the policymaker merely
wants to minimize interest rate volatility, interest rate persistence
7

6

134 |

Goodfriend (1991) is the subject of an essay by John Taylor elsewhere in this volume.

S ee, for example, Sack and Wieland (2000). The effects of model uncertainty are sensitive to the details; if the policymaker has a preference for robust policies (minimizing
the probability of the worst outcomes), model uncertainty tends to work in favor of
more aggressive policy.

| 135

Dotsey, Hornstein, and Wolman
may arise through history dependence of optimal policies with commitment. And it is history dependence that allows policy to influence
views about the far future and therefore allows less aggressive policy
to achieve desirable outcomes.
Woodford also shows that in the absence of full commitment, introducing an additional preference for low volatility of changes in the
interest rate leads to time-consistent policies that achieve outcomes
close to those that can be achieved by full commitment. Thus, far from
being an unduly timid policy, interest rate smoothing can enhance the
conduct of monetary policy.
Woodford’s work builds on Marvin’s insights, formalizing some of the
ideas in “Interest Rate Smoothing” and Goodfriend (1991). Woodford
argues that another reason for preferring low interest rate volatility,
besides a preference for orderly markets, is to reduce the likelihood
of very high interest rates (and their associated distortions) and the
likelihood of hitting the zero lower bound. Of course, the latter consideration was not on the horizon of policymakers when Marvin originally
wrote his paper.

Empirics: purposeful smoothing or inherited persistence?
Broadly speaking, the theories rationalizing interest rate smoothing
can be classified as purposeful vs. incidental.8 In the latter case, the
central bank has no inherent preference for smoothing, and the observed persistence reflects the nature of the data to which the central
bank responds (with some of those data being perhaps unobserved
by the researcher). Two notable attempts to sort out these issues are
Rudebusch (2006) and Coibion and Gorodnichenko (2012).9
Rudebusch (2006) characterizes the problem in the context of a
Taylor-style rule for monetary policy. He asks: Is the lagged interest
8
9

 udebusch (2006) refers to the former as endogenous and the latter as exogenous.
R
See also Bernanke (2004) for a useful survey.

Interest Rate Smoothing
rate an argument of that rule (purposeful smoothing) or is there a
persistent error term representing some misspecification (an omitted
variable that is persistent)? As he explains, this is a classic econometric problem, and it is notoriously difficult to distinguish the two cases
without auxiliary information.10 Rudebusch brings in auxiliary information from the yield curve and argues that purposeful smoothing would
show up as predictability in short-term rates to a greater degree than
is evident in the term structure. He thus concludes that persistence of
short-term rates is inherited from other variables to which the policy
rate responds.
Coibion and Gorodnichenko (2012) conduct an exhaustive study
broadly in the same spirit as Rudebusch but reach a different conclusion. They first take a direct approach, using a more general specification of both the policy rule and shock persistence, and find that the
evidence favors purposeful smoothing. With respect to the yield curve,
they find that predictability may be reduced if the public has different
information about the economy or policy than the Fed. They also show
that Federal Reserve Board staff forecasts more accurately predict
future interest rates than private sector forecasts. Their summary of the
evidence favors purposeful smoothing, and they support this conclusion with narrative evidence from the public and private comments of
FOMC members.

Narrative evidence and recent policy history
The statistical analysis in Coibion and Gorodnichenko (2012) is
impressively thorough. And yet, given the fundamental challenge
described by Rudebusch (2006), it is conceivable that another study
could push the dial back toward inherited persistence. For this reason,
narrative evidence — of which there is a plethora — has great potential to inform the debate. By narrative evidence we mean material such
as speeches by FOMC members and transcripts of FOMC meetings,
which are released with a five-year lag.
10

136 |

 arvin himself studied this econometric issue in the context of money demand
M
equations (Goodfriend, 1985). See Mark Watson’s essay in this volume.

| 137

Interest Rate Smoothing

Dotsey, Hornstein, and Wolman
Coibion and Gorodnichenko summarize the narrative evidence up
to approximately 2011. Since then, 10 more years of FOMC transcripts
have become available. In addition, over that period, the observed
persistence of policy rates has been unusually high: the fed funds rate
target range remained at its lower bound until December 2015, and
then the Fed raised that range in nine 25-basis point increments from
December 2015 to December 2018. On the surface, this certainly looks
like purposeful smoothing, with Federal Reserve policy becoming
more persistent as opposed to economic fundamentals becoming
more persistent. Increased persistence on the part of the Fed is probably related to its experience with the effective lower bound. With rates
constrained by the lower bound for some time, the FOMC then kept
rates at that lower bound until it was nearly certain that the appropriate funds rate was positive. And once it lifted off, it behaved cautiously
to avoid having to reverse course and once again be constrained by
the lower bound. One does not need to comb the transcripts to find
support for this view: from December 2015 to May 2018, the FOMC
statements contained slight variations on the following sentence: The
Committee expects that economic conditions will evolve in a manner that
will warrant only gradual increases in the federal funds rate; the federal
funds rate is likely to remain, for some time, below levels that are expected
to prevail in the longer run.

Concluding remarks
The three of us worked with Marvin at the Richmond Fed for a collective total of almost 40 years. He played a role in hiring each of us. We
can’t stress enough his role in guiding policy work and setting the tone
for the research environment at the Richmond Fed, first as an economist and then as research director and policy advisor. If we had to
sum up that tone concisely, it would be that research and policy work
are complementary. Along with much of Marvin’s work, “Interest Rate
Smoothing” exemplifies that complementarity.
The lunch table was an important forum for Marvin. During his many
years in Richmond, conversations with as many as 12 people crowded
around a table meant for six were a critical ingredient in creating the
unique mix of a policy and research environment. And Marvin as much
as anyone else facilitated those conversations. Marvin’s contributions
were wide ranging, sometimes heated, sometimes bewildering, and
always with his tie tucked into his shirt (when we still wore ties).

As longtime Federal Reserve economists, we view narrative evidence
as important and as supporting the view that the FOMC does purposefully impart persistence in the short-term rate. A simple way to make
this case is to consider the following question: Does the FOMC view
the short-term interest rate as a meaningful “state variable”? If there
is not purposeful interest rate smoothing, then the answer should be
“no.”
To us, the answer is unambiguously “yes.” Virtually all policy discussions (when the interest rate is in play) are of the form “how much – if
at all – should the policy rate rise or fall?” If there were no concern for
smoothing, then the discussion would be about the proper level of the
interest rate, not how much of a change is appropriate.

138 |

| 139

Dotsey, Hornstein, and Wolman

References
Barro, Robert J. 1989. “Interest-rate targeting.” Journal of Monetary Economics 23, no. 1 (January): 3-30.
Bernanke, Ben S. 2004. “Gradualism.” Remarks at an economics luncheon cosponsored by the Federal Reserve Bank of San Francisco (Seattle Branch) and the University of Washington, Seattle, Washington,
May 20.
Coibion, Olivier, and Yuriy Gorodnichenko. 2012. “Why Are Target Interest Rate Changes So Persistent?” American Economic Journal: Macroeconomics 4, no. 4 (October): 126-62.
Goodfriend, Marvin. 1985. “Reinterpreting Money Demand Regressions.” Carnegie-Rochester Conference Series on Public Policy 22: 207-241.

Interest Rate Smoothing
Rudebusch, Glenn D. 2006. “Monetary Policy Inertia: Fact or Fiction?”
International Journal of Central Banking 2, no. 4 (December): 85-135.
Sack, Brian, and Volker Wieland. 2000. “Interest-Rate Smoothing and
Optimal Monetary Policy: a Review of Recent Empirical Evidence.” Journal of Economics and Business 52, no. 1-2 (January-April): 205-228.
Sargent, Thomas J., and Neil Wallace. 1975. “‘Rational’ Expectations, the
Optimal Monetary Instrument, and the Optimal Money Supply Rule.”
Journal of Political Economy 83, no. 2 (April): 241–254.
Woodford, Michael. 2003a. Interest and Prices. Foundations of a Theory
of Monetary Policy. Princeton and Oxford: Princeton University Press.
Woodford, Michael. 2003b. “Optimal Interest-Rate Smoothing.” Review
of Economic Studies 70, no. 4 (October): 861-886.

Goodfriend, Marvin. 1986. “Monetary Mystique: Secrecy and Central
Banking.” Journal of Monetary Economics 17, no. 1 (January): 63-92.
Goodfriend, Marvin. 1987. “Interest Rate Smoothing and Price Level
Trend-Stationarity.” Journal of Monetary Economics 19, no. 3 (May): 335348.
Goodfriend, Marvin. 1991. “Interest Rates and the Conduct of Monetary Policy.” Carnegie–Rochester Conference Series on Public Policy 34
(Spring): 7–30.
Goodfriend, Marvin, and Robert G. King. 1997. “The New Neoclassical
Synthesis and the Role of Monetary Policy.” In NBER Macroeconomics
Annual 1997, edited by Ben S. Bernanke and Julio J. Rotemberg, 231283. Cambridge and London: National Bureau of Economic Research.
Mankiw, N. Gregory. 1987. “The Optimal Collection of Seigniorage:
Theory and Evidence.” Journal of Monetary Economics 20, no. 2 (September): 327-341.
McCallum, Bennett T. 1981. “Price Level Determinacy with an Interest
Rate Policy Rule and Rational Expectations.” Journal of Monetary Economics 8, no. 3: 319-329.
140 |

| 141

Interest on Reserves

Paying Interest on Bank Reserves1
Huberto M. Ennis and John A. Weinberg
Since the 2008 financial crisis, the Fed’s main tool for exercising
control over short-term interest rates has been the rate it pays on the
now large reserve balances held by banks. This new approach marks
a substantial change from the Fed’s previous operating regime, in
which the supply of reserves was tightly controlled to ensure that the
equilibrium rate in the federal funds market was at, or near, the Fed’s
target. In 2002, well before this change and before the idea was under
active consideration by the Fed, Marvin Goodfriend set out a proposal
for such an approach to policy operations in the Federal Reserve Bank
of New York’s Economic Policy Review. This proposal served as a cornerstone for planning and discussions around the System leading up
to implementation of an interest on reserves mechanism in October
2008. While interest on reserves, as implemented in the US, has yielded
some surprises, Marvin’s discussion remains an important benchmark.
In this essay, we review Marvin’s 2002 proposals and provide some
perspectives on the evolution and implementation of those ideas. In
particular, we recount the 15-year history of interest on reserves at the
Fed, showcasing the influence of Marvin’s ideas. We document how
those ideas were adapted as policymakers learned from the actual
implementation of policy.
The issues identified by Marvin in the early 2000s remain central
today to the assessment of an operating regime based on the
management of interest on reserves. One particular set of issues —
having to do with preserving independence of monetary policy from
political motivations — has only grown in relevance.
1

142 |

 e would like to thank Todd Keister, Bob King, Beth Klee, Jeff Lacker, and Alexander
W
Wolman for comments. All errors are our own.

| 143

Ennis and Weinberg

Interest on reserves
Bank reserves are balances held by depository institutions in their
accounts at the corresponding Reserve Bank. Reserves are freely
convertible, dollar-for-dollar, into currency and play a critical role in
the settlement of financial transactions intermediated by banks. Banks
trade reserves in the fed funds market.
The idea of paying interest on bank reserves first arose as a tool for
reducing the distortions from the tax on money (as modeled by Lacker
[1997]). Central bank money that is dominated in rate of return will still
be held by economic agents because of its privileged role in the payment system. Eliminating the tax on currency is the motivation behind
the Friedman rule.2 In his Program for Monetary Stability, Friedman
(1959) also proposed paying interest on reserves held by banks.3

Evolution of an idea: implications for conduct of monetary policy
Goodfriend’s (2002) paper goes beyond this public finance perspective to a concern for the operational conduct of monetary policy. At
the time of his writing, major central banks around the world had adopted the approach of conducting monetary policy by manipulating
short-term interest rates and reserves earned zero interest. The supply
of reserves was one of the main levers used for affecting short-term
interest rates. By contrast, Marvin’s proposal was to peg the fed funds
rate by paying a positive interest on reserves and assuring a plentiful
reserve supply. An important argument in Goodfriend (2002) is that
paying interest on reserves could allow the central bank to separately
manage short-term interest rates and the supply of monetary liabilities.
Goodfriend (2002) traces his thinking on this topic to his earlier 2000
paper on monetary policy at the lower bound on nominal interest
rates,4 which proposes supplementing a policy rate set at its effective

Interest on Reserves
floor with the expansion of reserves. With the short-term rate at its
lower bound and reserves abundant, he argued, the convenience yield
on money goes to zero — making reserves and fed funds lending
perfect substitutes. With the demand for reserves thus satiated, the
banking system will hold whatever quantity the Fed supplies.5
At the lower bound, Goodfriend characterized expanding the supply
of reserves as an alternative means of policy accommodation when
interest rate reductions are off the table. More generally, in his interest-on-reserves (IOR) paper,6 he argued that interest rate and reserves
policy could operate independently, serving different objectives. By
setting its interest rate on reserves, the Fed could conduct interest rate
policy in much the same way central banks had grown accustomed to
doing, systematically responding to economic conditions to maintain
price stability.

Separation of interest rate and balance sheet policy
Importantly, once reserves were large enough to ensure that demand is satiated at the IOR rate, further adjustments to reserve supply could respond to liquidity conditions in broader money markets.
Marvin dubbed this “managing the supply of broad liquidity.”7 Here,
he recognized the possibility of frictions affecting the efficiency with
which market participants might exchange alternative short-term
instruments. Importantly though, his view of the central bank’s role
in responding to such frictions was limited to managing the overall
supply of liquidity by creating reserve balances through open market operations. Inherent in this view is a belief that markets will do a
reasonably good job of distributing the liquidity that the central banks
supply — money market problems are adequately addressed by managing that overall supply.8
 eister and McAndrews (2009).
K
Goodfriend (2002).
7
Keister et al. (2008) provides a detailed discussion of these ideas.
8
Goodfriend and King (1988). See the essay by Douglas Diamond elsewhere in this
volume.
5
6

F riedman (1969).
See also, Ireland (2019).
4
Elsewhere in this volume, Ben Bernanke discusses Marvin's initial work on the zero
lower bound, Goodfriend (2000) and a follow-up Jackson Hole paper, Goodfriend
(2016).
2
3

144 |

| 145

Ennis and Weinberg
The separation of interest rate and balance sheet policy was an
important part of Marvin’s vision for an IOR operating regime. Marvin,
of course, had made significant contributions to our understanding
of how modern central banks achieved and maintained low inflation
by systematically adjusting short-term rates in response to macroeconomic conditions.9 So being able to continue reacting to the economy
in the same manner was vitally important to him. Note that an implication of the separation of interest rate and balance sheet policy is that if
the Fed expanded reserves in response to a perceived need for liquidity in money markets, there would be no expectation that interest rate
moves — in particular, rate increases, if economic conditions warranted them — would need to wait for an unwinding of the reserves
expansion. This observation is consistent with the Fed’s approach to
allowing a very gradual balance sheet run-off while at the same time
raising interest rates.

Desirable features of an IOR regime
One particular virtue of an IOR operating regime, according to Goodfriend (2002), is that central bank interest rate control would be more
robust to technological changes in the payment system that reduce
the demand for central bank balances as a means of settlement. Marvin argued that the traditional Fed approach could become increasingly difficult in the face of such changes. This is because the traditional
approach involved manipulating the supply of reserves so that the
market-determined fed funds rate settled near the target. By contrast,
in a regime in which the opportunity cost of holding reserves was
eliminated, banks would be willing to hold the reserves supplied by
the Fed, independently of the role of reserves in payment settlement.
Goodfriend (2002) also noted that the abundant reserves in such an
operating regime could reduce the need for central bank credit that
can arise when banks face unexpected payment flows. Historically, in
the US interbank system, such credit extensions took the form of both

9

146 |

Interest on Reserves
intraday overdraft allowances and overnight discount window loans.10
Goodfriend had elsewhere expressed the view that modern money
markets could function with less reliance on central bank credit.11 So,
he saw a reduced reliance on Fed lending as a means to solidify the
separation of monetary and credit policies. The IOR regime, then, could
also serve as a means of facilitating that goal.

IOR and Fed income
One potential difficulty that Goodfriend (2002) anticipated involved
the effects of reserve remuneration and larger Fed balance sheets on
the Fed’s income.12 However, he expected that under normal conditions, the Fed would continue to earn a positive spread on its balance
sheet. The Fed’s assets would have an average maturity well above the
overnight maturity of reserves and so would usually bear a yield greater than IOR. Also, he took as given that (non-interest-bearing) currency
would remain a nontrivial part of the Fed’s liabilities for the foreseeable
future. While the Fed’s net interest margin would be reduced by paying
a positive rate on reserve balances, that spread would be earned on an
expanded balance sheet.
So Goodfriend did not see paying interest on reserves as a fundamental problem for the Fed’s income, on average. That said, he did see
the possibility of periods with negative Fed income when appropriate
monetary policy called for relatively quick and significant increases in
short-term interest rates.13 He was nervous that such episodes would
mean that the Fed would have to go to Congress or the Treasury to
obtain funding to cover operating costs. That is, such a situation could

I n Ennis and Weinberg (2007), we investigate formally the link between excess
reserves and intraday credit. For the link between reserves and discount window
lending, see Ennis and Klee (2021).
11
Goodfriend and King (1988) and Goodfriend and Lacker (1999).
12
For a recent thorough study of this issue see Carpenter et al. (2015).
13
Goodfriend (2014).
10

E lsewhere in this volume, John Taylor discusses Goodfriend (1991) and Michael Dotsey, Andreas Hornstein, and Alexander Wolman discuss Goodfriend (1987).

| 147

Ennis and Weinberg
threaten the political independence of monetary policy. While acknowledging the risk, he saw it as manageable so long as the Fed
carried an appropriately sized capital buffer or surplus account.14

Interest on reserves at the Fed
In 2006, Congress enacted a collection of regulatory relief measures
for banks and other financial institutions. As part of this package — the
Financial Services Regulatory Relief Act of 2006 — the Fed was granted
authority to pay interest on the reserves held by banks.15 Part of the
motivation for this provision was to reduce banks’ incentives to engage
in costly account management practices that served only to reduce
reserve requirements. This practice involved automated procedures
for shifting customer overnight balances out of reservable deposit
accounts and into other instruments that did not carry reserve requirements — so-called sweep accounts. Changes in sweep account practices were a focus of the Congressional Budget Office’s (CBO) analysis
of the likely implications of this provision for the federal budget.16

Key aspects of the 2006 act
The language in the act (Section 201) is very brief, but there are a
few specific conditions worth noting. First, the possibility of interest on
reserves was not made available to all holders of reserve balances with
the Fed — only to depository institutions. Importantly, this left out
government sponsored enterprises (GSEs) such as Fannie Mae, Freddie
Mac, and the Federal Home Loan Banks. These entities hold significant
balances with the Fed and are typically lenders of overnight funds in
the fed funds market and other segments of the short-term
 eis (2015) provides a good overview of central bank solvency and its relationship
R
with independence. In the years after Marvin wrote his piece, a number of political
actions by Congress have called into question the ability of the Fed to independently
manage its surplus account (Fessenden, 2015; and George, 2020). Indeed, in later
writings, Marvin expressed greater concern about the independence consequences of cash flow volatility and about the true availability of future unencumbered
seigniorage (Goodfriend, 2014).
15
This essay only discusses the experience with IOR in the US. For an international
perspective, see Bowman et al. (2013) and the references therein.
16
Available here: https://www.congress.gov/congressional-report/109th-congress/
senate-report/256.
14

148 |

Interest on Reserves
money market. The distinction between banks and GSEs would prove
significant for the eventual workings of the IOR regime in ways not
fully anticipated.
Second, the act specified that the rate paid by the Fed was not to
exceed the general level of short-term market rates. At the time, this
seemed like an innocuous requirement. In the model Goodfriend used
to discuss his proposal,17 the rate paid by the central bank essentially
becomes the market rate. This is because the demand for reserves
is satiated, and interest-bearing central bank balances replace interbank loans for the most part. A market rate above the rate on reserves
would tend to arise in other systems in which reserves remain scarce
and there is still an active interbank market. So, standard analysis at the
time did not foresee the possibility of markets rate below IOR. Indeed,
the CBO’s analysis assumed a rate on reserves that was on average 10
to 15 basis points below overnight money market rates. The experience since implementation of IOR has been quite different, in large
part due to the behavior of the GSEs in the money markets.
Finally, the 2006 act gives the authority to set the rate paid on
reserves to the Federal Reserve Board of Governors, as opposed to the
monetary policymaking body — the Federal Open Market Committee
(FOMC). Congress likely saw the provision as more of an operational
modernization — to reduce adverse incentives created by the tax on
reserves — than a fundamental change in the way monetary policy
would be implemented.18 Yet, in the proposal put forward by Goodfriend (2002), the rate on reserves becomes the instrument for interest
rate policy. The act’s designation of decision-making authority creates
the need for potentially delicate coordination among different bodies
within the Federal Reserve System.

Consideration by a Fed workgroup
The 2006 act set 2011 as the start date for the Fed’s new authority,
giving the Fed time to work out the details of using this tool. Shortly

17
18

 oole (1968).
P
Ireland (2019).

| 149

Ennis and Weinberg
after its enactment, the Fed created a System workgroup, consisting of
staff from the Board of Governors and the Reserve Banks, to study the
issue and prepare a number of alternatives to be considered by policymakers. A central question in designing an implementation regime
was whether the Fed would continue to target an interbank rate by
manipulating the supply of (relatively scarce) reserves. Doing so would
result in a so-called corridor system, in which the target rate was above
the rate paid by the Fed (and typically below the rate charged by the
Fed for discount window loans).
The main alternative to this approach, as proposed by Goodfriend, is
typically referred to as a floor system. In such a system, the Fed would
provide enough reserves to essentially eliminate the need for interbank lending. Movements in short-term interest rates would be driven
by the rate on reserves, and moderate fluctuations in the quantity of
reserves would have little to no effect on those short-term rates.
The workgroup examined the relative merits of floor and corridor
systems (as well as some hybrids). A technical appendix to the workgroup report provided an analytical framework for this comparison
(and was later published as Ennis and Keister [2008]). The workgroup
also devoted attention to an array of technical details, including the
mechanics of monitoring reserve requirements and the treatment of
different categories of reserves (required, clearing balances, and excess
reserves).19

Accelerating the implementation of IOR
The Fed intended a long deliberative process before it implemented
IOR. But by the time the workgroup presented its analysis to policymakers in early 2008, attention had shifted to dealing with the unfolding financial crisis. The Fed’s crisis response led directly to an accelerated adoption of IOR. Beginning in late 2007, the Fed dramatically
expanded the provision of central bank credit in various forms. In the

Interest on Reserves
interest rate targeting regime it maintained at the time, the Fed could
not let credit expansions increase the size of its balance sheet and the
supply of reserves. Accordingly, it offset its growing book of loans by
selling Treasury securities from its portfolio (and in that way sterilizing
that credit growth). When the crisis deepened in October 2008, around
the time of the failure of Lehman Bros. and the bailout of AIG, the rapid
rise in credit extensions overwhelmed the Fed’s ability to sterilize — it
risked running out of Treasuries to sell.
Partly to address this issue, the Fed asked for and received permission to accelerate its authority to pay interest on reserves — and in
that way control interest rates without sterilizing the credit programs.20
In October 2008, then, the Fed started paying interest on reserves.
The conditions under which this happened were not ideal for a test of
using the rate paid on reserves to target and control market rates. With
the extreme market volatility at the time, and with the Fed both rapidly
expanding its balance sheet and taking its interest rate target to its
effective lower bound, the choice between a corridor and a floor soon
became trivial. Indeed, with the beginning of quantitative easing programs in 2009, it became clear that it would be some time before there
was a relevant option to make reserves anything other than abundant.

A surprise: the overnight rate below the floor
As the Fed entered the regime of interest paid on abundant reserve
balances, something else became clear. The theoretical floor on market
overnight rates provided by the rate on reserves turned out not to be
a firm floor. When the Fed moved its policy rate to the effective lower
bound, it started targeting a range instead of a given number. From
December 2008 until December 2015, the FOMC target range for the
effective fed funds rate was 0 to 25 basis points and IOR was fixed at
the top of the policy range. During this period, the effective fed funds

20
19

150 |

T he document produced by the workgroup for the Board and the FOMC is available
at https://www.federalreserve.gov/monetarypolicy/files/FOMC20080411memo01.
pdf.

S ee Ireland (2019) for a critical assessment of the policy of paying interest on reserves. Somewhat ironically, the motivation for paying interest on reserves as a way
to finance credit programs is in sharp contrast with Marvin’s thinking — he suggested that the ability to adjust the amount of excess reserves could actually reduce the
urge of central banks to pursue other more objectionable credit policies.

| 151

Ennis and Weinberg

Interest on Reserves

rate consistently traded below IOR by, on average, 10 to 15 basis points.
And, while theory would predict that with abundant reserves fed funds
trading would vanish, activity did decline significantly but did not
disappear.

reserves would reliably pull up the broader array of short-term rates
that were seen as important for affecting economic activity. This lack
of confidence ultimately led to the creation of the Overnight Reverse
Repurchase Agreement facility, or ON RRP.22

These surprising facts are due primarily to the participation of the
GSEs in the fed funds market. Since they could not earn interest on
their reserve balances, there was an opportunity for them to lend
those balances to banks that could earn interest from the Fed. While
one would expect competition among borrowing banks to bid the rate
on those loans up close to the rate on reserves, at least two market features are thought to have limited this arbitrage. First, a bank borrowing
reserves from a GSE to earn overnight interest incurs costs from expanding its balance sheet. One direct source of costs is that banks pay
FDIC premiums on the overall size of their balance sheet (not just on
their insured deposits). This places a cap, below the yield on reserves,
on what a bank would be willing to pay to a GSE. Second, competition
may not even have driven fed funds loan rates as far as the ceiling
implied by the cost of balance sheet expansion. One explanation for
this is that the GSEs, holding market power in fed funds lending, were
restrictive in which banks they would lend to and how much. The
resulting small number of bidders, then, resulted in imperfect competition that may have further held down the effective fed funds rate.21

The ON RRP facility allows an expanded set of counterparties to
exchange excess cash for securities held by the Fed in an overnight
transaction that pays a rate to the cash lender that is a bit below the
rate paid on bank reserves. The expanded set of counterparties includes, importantly, the GSEs that are also active lenders in the fed
funds market. Their access to the ON RRP was intended to place a more
reliable floor under the fed funds rate — as a GSE would presumably
not want to lend reserves to a bank at less than it could earn by doing
a repo transaction with the Fed.

The prospects of liftoff and a new facility
Still, for the first several years of IOR, there really was no opportunity
to assess the Fed’s ability to conduct interest rate policy exclusively (or
primarily) through manipulating the rate paid on reserves. Eventually,
though, the likelihood of a rate increase being necessary became more
apparent. In 2012, when the Fed began including the policy interest
rate as one of the variables in the Summary of Economic Projections, a
majority of participants anticipated that some rate increases would be
appropriate by the end of 2015. As the FOMC discussed prospects for
lifting its target rate off of its effective floor, some members and senior
staff expressed uncertainty as to whether simply lifting the rate on

The liftoff and transition to a new policy
When the FOMC did eventually begin raising rates near the end of
2015, market rates generally followed its increases in the rate paid on
reserves.23 As to whether the extra precaution of the ON RRP proved to
be a necessary tool for interest rate control, the evidence is somewhat
mixed. Most of the time, overnight market rates remained above the
ON RRP floor. But at ends of months, it was common for market rates
to fall to that floor. Correspondingly, those dates would see significant
pick up in volume for the Fed’s ON RRP facility. This pattern appears
to have been due to incentives created by calendar-based reporting
requirements for many of the financial institutions that participate in
these markets.
Once rate increases were underway, the FOMC returned to the
question of how it would implement its interest rate policy over the
longer run. Would it maintain the abundant reserves regime that had
emerged during the crisis response, in which the rate paid on reserves
is the main tool for influencing market rates? Or would it move to a
regime in which the supply of reserves was more restricted, so that
22
23

21

152 |

Bech and Klee (2011).

F rost et al. (2015).
See, for example, Haltom and Wolman (2016).

| 153

Interest on Reserves

Ennis and Weinberg
an active interbank funds market would emerge, and market rates
would depend both on the rate paid by the Fed and the supply of reserves? And what would be the implications of this choice for the size
of the Fed’s balance sheet?
These discussions led to the January 2019 “Statement Regarding
Monetary Policy Implementation and Balance Sheet Normalization.”
There, the committee stated its intent to continue to operate with an
“ample” supply of reserves, so that rate control would be achieved
“primarily through setting the Fed’s administered rates.” It also said that
it would continue to use the target range for the federal funds rate to
express the stance of its interest rate policy.
Relative to the approach proposed by Goodfriend, this use of the
funds rate range might seem like an unnecessary complication. Why
not simply use the administered rate — the rate paid on reserves — as
the policy rate? Here, the rate governance created by the legislation allowing interest on reserves might have come into play. Recall that the
enabling act gave the Fed’s Board of Governors the authority to set the
rate on reserves. The January 2019 statement, by contrast, emphasized
the rate decision made by the FOMC. This statement can be seen as a
reassurance that adopting a regime in which administered rates are
the main tool does not move authority for interest rate policy from the
FOMC to the Board of Governors.
After describing its plan for how it would proceed to implement
monetary policy, the committee made a series of announcements
about the intended size of its balance sheet and the supply of reserves.
The Fed intended to allow the balance sheet to run down through the
redemption of securities without reinvestment of the proceeds. This
process was made gradual and predictable by capping the amount of
unreinvested redemptions. The Fed’s announcements generally avoided projecting a specific number for the ultimate size of its balance
sheet. They did indicate that the average level of reserves would likely
be “somewhat above the level of reserves necessary to efficiently and

154 |

effectively implement monetary policy.”24 In the context of the January
2019 statement on implementation plans, an interpretation is that
reserves supply would remain large enough so that the fed funds rate
would remain at or below the rate paid on reserves.

Turbulence and renewed expansion of the balance sheet
Turbulence in money markets in September 2019, including spikes
in repo rates and in the fed funds rate, led some to question whether
the Fed had already taken the supply of reserves too low relative to its
stated intentions. As a result, the Fed ended its runoff of the balance
sheet and began a modest pace of growth in reserves. In its October
2019 announcement, the Fed included a plan to build back reserves at
least to their level before the market stress of September. This plan was
intended to extend through “at least the second quarter” of 2020. The
Fed did not change the character of its long-run plans (as expressed
earlier in 2019), but perceptions about how large the supply of reserves would need to be on an ongoing basis did change some.

Monetary policy implementation after COVID-19
The Fed’s intent remained to operate with reserves no larger than
necessary to effectively and efficiently conduct monetary policy.25 Yet,
money markets events in September 2019 produced a reassessment
of the desired long-run level of reserves. Further, with the onset of the
COVID-19 pandemic in early 2020 and the economic disruption that
followed, the normalization process for the Fed’s balance sheet was
again interrupted. The uncertainty caused by the public health crisis
led many money market yields to rise as investors sought to hold only
the safest, most liquid assets (in a so-called “dash for cash” episode).
The Fed’s response included an array of credit facilities as well as a substantial increase in the pace of asset purchases. At the same time,
S ee the March 2019 Fed press release, “Balance Sheet Normalization Principles and
Plans” (https://www.federalreserve.gov/newsevents/pressreleases/
monetary20190320c.htm).
25
See, for example, the Fed’s online press release from January 30, 2019. For a detailed
discussion of monetary policy implementation in a system with “ample” reserves, see
Ihrig et al. (2020).

24

| 155

Ennis and Weinberg
the Fed brought the interest rate paid on reserves to its lowest level to
date.
The Fed’s pandemic actions brought its balance sheet size to new
highs. Recently, take-up in the ON RRP facility also rose to more persistently high levels.26 As of this writing (January 2022), the Fed has begun to taper its asset purchases, but there has been little discussion as
yet about when the Fed will begin to let its net holdings of assets run
down and whether the envisioned long-run level of reserves remains
the same as it did pre-pandemic.

A new repo facility
In the wake of the pandemic, the Fed has added another standing
facility to its toolbox for interest rate control. It created the Standing
Repurchase Agreement (repo) Facility (SRF) that lends cash overnight
to a broad set of counterparties against Treasury and agency securities.
By lending at a rate somewhat above the rate on reserves, this facility
is expected to limit the kinds of money market rate spikes experienced
in September 2019. It is still an open question whether the SRF will
become heavily used when the Fed normalizes the size of its balance
sheet. Interestingly, earlier discussions of a standing repo facility
included the argument that such a facility would permit the Fed to operate effectively even with a lower level of reserves.27 It seems unclear
that this is the current intent of the SRF.
So, the current regime is somewhat more complicated than the one
laid out by Goodfriend in 2002, which saw rate control as being adequately accomplished by managing the rate paid on reserves. Now, out
of an abundance of caution, the Fed has added supports in the form
I nitially, the ON RRP facility was envisioned to function mainly as a backstop in
short-term secured credit markets (Frost et al., 2015). However, with the current
configuration of interest rates and the very large Fed balance sheet, the ON RRP has
become heavily used — and functions as a way for the Fed to pay interest directly
on cash reserves held by money market funds and other financial institutions (such
as the GSEs), which are eligible to use the ON RRP but are not eligible to receive IOR.
This situation seems to be hardly something that Marvin (or we) would have recommended or even anticipated.
27
Andolfatto and Ihrig (2019).
26

156 |

Interest on Reserves
of standing facilities in the repo market.28 While we of course cannot
be sure, it seems likely to us that the supply of reserves will remain for
some time well above levels that Marvin would have viewed as necessary for implementing an effective IOR regime.29

Conclusion
As on many topics related to modern central banking, Goodfriend’s
discussion of interest on reserves as a tool for monetary policy has
proved to be prescient. The issues he identified as potential areas
of concern have all remained relevant, and his thinking has influenced the learning process of policymakers every step of the way.
For instance, his consideration of the implications for the demand for
reserves of payment innovations — written at a time when an array of
payment tools now common were barely in use, if at all — identified
legitimate concerns that are being voiced today by those studying the
merits of a central bank digital currency.30 His conclusion — that an IOR
regime diminishes the concern by making reserves an interest-bearing
asset and hence rendering their payment settlement role less dominant — suggests that monetary control might survive the advent of
digital currencies.
It is important to stress that Goodfriend’s proposal was offered in a
context where the credibility of the central bank was unquestioned.
In such an environment, inflation expectations are firmly anchored.
But when there is more uncertainty about the central bank’s goals and
its likely conduct of interest rate policy in pursuit of those goals — as
some have argued coming out of the pandemic crisis — the possibility of drifting inflation expectations can be more of a concern. In this
context, as Ennis and Wolman (2010) noted, the consequences of an
interest rate policy that falls “behind the curve” can be magnified by
S ee Ennis and Huther (2021) for a discussion.
It has become a real concern that the Fed will deem it necessary to go back to
quantitative easing policies before the size of its balance sheet has had the time to
fully normalize from its very high current level. This possibility of a “ratchet effect”
suggests to us that normalization should become a priority as soon as conditions
allow it.
30
See, for example He (2018) and Benigno (2019).

28
29

| 157

Ennis and Weinberg
the existence of large excess reserve balances. This provides another
reason to be mindful of the overall size of reserve balances.
Among the most salient issues discussed by Goodfriend (2002) are
those related to the possibility of negative Fed net income, especially
as that might affect its independence to conduct appropriate monetary policy.31 The Fed’s relationship to the legislative and executive
branches of government was a central concern in much of Goodfriend’s work.32 While in his 2002 article he saw the income and expense challenges created by interest on reserves as being manageable,
his vision was certainly of a much smaller balance sheet. He also advocated a regime in which the Fed’s intervention in money markets was
simpler than one with multiple standing facilities. Again, we cannot be
sure, but we think that on those issues our insightful and experienced
colleague would be more concerned now than he was in 2002.
As a larger balance sheet and greater involvement in the money
markets raise the risks to Fed independence, another common theme
from Marvin’s work is worth remembering. In many instances, Marvin
advocated creating formal understandings between the Treasury and
the Fed delineating the central bank’s autonomy and accountability.
He would often link these proposals to the 1951 Fed-Treasury Accord
that ultimately allowed the Fed to conduct the independent interest rate policy to which we have become accustomed.33 As the Fed’s
operating regime expands in its reach, an accord that establishes
reasonable and commonly understood limits to Fed activities could be
a useful tool.34
 s a purely technical matter, negative cash flow does not create a problem for the
A
Fed, as it can use reserves to pay its interest expenses. The Fed has discussed accounting for such a practice by means of a “negative liability” account to the Treasury
representing the withholding of future remittances until the effects of the negative
cash flow are offset. Still, as Marvin has noted (Goodfriend, 2014), the optics of creating reserves to pay interest on reserves could stress both the Fed’s credibility for low
inflation and its political independence.
32
George (2020).
33
See the 2001 issue of the Richmond Fed Economic Quarterly on this topic, particularly
Lacker’s (2001) introduction.
34
Elsewhere in the volume, Charles Plosser independently makes a specific proposal
for such an accord.
31

158 |

Interest on Reserves

References
Andolfatto, David, and Jane Irhig. 2019. “Why the Fed Should Create a
Standing Repo Facility.” On the Economy Blog, Federal Reserve Bank of
St. Louis, March 6.
Bech, Morten L., and Elizabeth Klee. 2011. “The Mechanics of a Graceful
Exit: Interest on Reserves and Segmentation in the Federal Funds Market.” Journal of Monetary Economics 58, no. 5 (July): 415-431.
Benigno, Pierpaolo. 2019. “Monetary Policy in a World of Cryptocurrencies.” Centre for Economic Policy Research Discussion Paper 13517,
February.
Bowman, David, Michiel De Pooter, Etienne Gagnon, and Mike Leahy.
2013. “Update on Foreign Central Bank Operating Procedures and the
Foreign Experience with Using Interest on Reserves as a Monetary
Policy Instrument.” April 19.
Carpenter, Seth, Jane Ihrig, Elizabeth Klee, Daniel Quinn, and Alexander Boote. 2015. “The Federal Reserve’s Balance Sheet and Earnings: A
Primer and Projections.” International Journal of Central Banking 11, no.
2 (March): 237-283.
Ennis, Huberto M., and John A. Weinberg. 2007. “Interest on Reserves
and Daylight Credit.” Federal Reserve Bank of Richmond Economic
Quarterly 93, no. 2 (Spring): 111-142.
Ennis, Huberto M., and Todd Keister. 2008. “Understanding Monetary
Policy Implementation.” Federal Reserve Bank of Richmond Economic
Quarterly 94, no. 3 (Summer): 235-263.
Ennis, Huberto M., and Alexander L. Wolman. 2010. “Excess Reserves
and the New Challenges for Monetary Policy.” Federal Reserve Bank of
Richmond Economic Brief 10-03, March.
Ennis, Huberto M., and Jeff W. Huther. 2021. “The Fed’s Evolving Involvement in the Repo Markets.” Federal Reserve Bank of Richmond
Economic Brief 21-31, September.
Fessenden, Helen. 2015. “A Bridge Too Far?” Federal Reserve Bank of

| 159

Ennis and Weinberg
Richmond Econ Focus, Third Quarter.
Friedman, Milton. 1959. A Program for Monetary Stability. New York:
Fordham University Press.
Friedman, Milton. 1969. The Optimum Quantity of Money. New York:
Macmillan.
Frost, Joshua, Lorie Logan, Antoine Martin, Patrick E. McCabe, Fabio
M. Natalucci, and Julie Remache. 2015. “Overnight RRP Operations as
a Monetary Policy Tool: Some Design Considerations.” Federal Reserve
Bank of New York Staff Report 712, February.
George, Esther L. 2020. “Perspectives on Balance Sheet and Credit
Policies: A Tribute to Marvin Goodfriend.” Cato Journal 40, no. 3 (Fall):
589-594.
Goodfriend, Marvin. 1987. “Interest Rate Smoothing and Price Level
Trend-Stationarity.” Journal of Monetary Economics 19, no. 3 (May): 335348.
Goodfriend, Marvin. 1991. “Interest Rates and the Conduct of Monetary
Policy.” Carnegie-Rochester Conference Series on Public Policy 34 (Spring):
7-30.
Goodfriend, Marvin. 2000. “Overcoming the Zero Bound on Interest
Rate Policy.” Journal of Money, Credit and Banking 32, no. 4, Part 2 (November): 1007-35.
Goodfriend, Marvin. 2002. “Interest on Reserves and Monetary Policy.”
Economic Policy Review 8, no. 1 (May): 77-84.
Goodfriend, Marvin. 2014. “Monetary Policy as a Carry Trade.” Monetary
and Economic Studies 32 (November): 29-44.
Goodfriend, Marvin, and Robert G. King. 1988. “Financial Deregulation,
Monetary Policy, and Central Banking.” Federal Reserve Bank of Richmond Economic Review 74 (May/June): 3-22.
Goodfriend, Marvin, and Jeffrey M. Lacker. 1999. “Limited Commitment
160 |

Interest on Reserves
and Central Bank Lending.” Federal Reserve Bank of Richmond Economic Quarterly 85, no. 4 (Fall): 1-27.
Haltom, Renee Courtois, and Alexander L. Wolman. 2016. “How Did
Short-Term Market Rates React to Liftoff?” Federal Reserve Bank of
Richmond Economic Brief 16-09, September.
He, Dong. 2018. “Monetary Policy in the Digital Age.” IMF Finance &
Development 55, no 2 (June).
Ihrig, Jane, Zeynep Senyuz, and Gretchen C. Weinbach. 2020. “Implementing Monetary Policy in an ‘Ample-Reserves’ Regime.” FEDS Notes,
October 2.
Ireland, Peter N. 2019. “Interest on Reserves: History and Rationale,
Complications and Risks.” Cato Journal 39, no. 2 (Spring/Summer): 327337.
Keister, Todd, Antoine Martin, and James McAndrews. 2008. “Divorcing
Money from Monetary Policy.” Federal Reserve Bank of New York Economic Policy Review 14, no. 2 (September): 41-56.
Keister, Todd, and James McAndrews. 2009. “Why are Banks Holding So
Many Excess Reserves?” Current Issues in Economics and Finance 15, no.
8 (December): 1-10.
Lacker, Jeffrey M. 1997. “Clearing, Settlement and Monetary Policy.”
Journal of Monetary Economics 40, no. 2 (October): 347-381.
Lacker, Jeffrey M. 2001. “Introduction to Special Issue [on the Treasury-Fed Accord].” Federal Reserve Bank of Richmond Economic Quarterly 87, no. 1 (Winter): 1-6.
Poole, William. 1968. “Commercial Bank Reserve Management in a Stochastic Model: Implications for Monetary Policy.” Journal of Finance 23,
no. 5 (December): 769-791.
Reis, Ricardo. 2015. “Different Types of Central Bank Insolvency and the
Central Role of Seignorage,” Journal of Monetary Economics 73, no. 1
(June): 20-25.
| 161

Reconsidering Price Stability

Reconsidering the
Case for Price Stability
Vitor Gaspar and Frank Smets
From the beginning of the Economic and Monetary Union (EMU) in
1999, Marvin Goodfriend was a frequent visitor at the European
Central Bank (ECB). As the ECB started its operations, his advice —
grounded in long policy experience as well as innovative and highly
relevant research — was sought after by ECB policymakers and researchers alike.
The Treaty on European Union gives the ECB the primary mandate
of maintaining price stability. The quantitative formulation of the price
stability objective was, however, left to the ECB’s Governing Council.
In its original monetary policy strategy at the start of EMU, the ECB
defined price stability as “a year-on-year increase in the Harmonised
Index of Consumer Prices (HICP) for the euro area of below 2 percent.”1
It was thus natural when the ECB organized its first central banking conference that the topic should be “Why price stability?” The
ECB sought to discuss with a diverse group of economists what price
stability should be taken to mean. In their contribution to the conference, Marvin Goodfriend and Robert King (hereafter referred to as GK)
made the case for literal price stability.2 This conclusion built on the
general equilibrium optimal taxation literature in the spirit of Ramsey
(1927) and Lucas and Stokey (1983). Following earlier work,3 they used
a general equilibrium model augmenting the approach of the real
business cycle literature with imperfect competition and costly price
setting.4 To connect to the optimal taxation literature, GK interpreted
the imperfect competition wedge between price and marginal cost —
the markup — as analogous to a tax rate.
E CB (1999), p. 46.
Goodfriend and King (1999).
3
Goodfriend and King (1997).
4
GK called this approach the “New Neoclassical Synthesis” and their 1997 paper is
discussed by Michael Woodford in another contribution to this volume.
1
2

162 |

| 163

Gaspar and Smets
They then explored under which circumstances this markup should
be uniform across time and across states of nature, concluding that
optimal policy stabilizes real marginal costs within their model in most
circumstances. Moreover, given their presumed link between marginal
cost and inflation, such policy also stabilizes the price level. Therefore,
GK suggested that a central bank would stabilize markups by credibly
maintaining price stability and thus deliver optimal policy. The basic intuition is that markup constancy corresponds to “tax smoothing” over
the business cycle or, more generally, to the case for uniform taxation
in public finance.
The ECB’s original definition of price stability did not exclude such
literal price stability or a zero inflation target. However, in the 2003
strategy review, the ECB introduced an inflation aim of “below, but
close to 2 percent” within the price stability definition.5 Three economic factors were viewed as relevant for justifying a small inflation buffer:
i) the existence of downward nominal price and wage rigidities; ii) the
persistence of sustained inflation differentials across euro area countries; and iii) the constraint imposed by the zero lower bound (ZLB) on
nominal interest rates.
While this formulation of the price stability objective was effective in
maintaining long-term inflation expectations close to 2 percent in the
inflationary environment of the first decade of EMU, it was problematic
in terms of anchoring expectations when disinflationary forces prevailed following the Global Financial Crisis (GFC) in 2008 and the sovereign debt crisis in 2010-2013. In the new ECB monetary policy strategy,
announced on July 8, 2021, this formulation was therefore replaced
by a symmetric 2 percent inflation target.6 Such a symmetric target is
simple, clear, and easy to communicate and should therefore help to
better anchor long-term inflation expectations. The ECB Governing
Council sees the level of 2 percent as representing a good balance
between providing a safety margin against the risks of deflation and

Reconsidering Price Stability
the welfare costs posed by excessive inflation. With this new symmetric
2 percent inflation target, the ECB has joined many central banks in
advanced economies that flexibly target a 2 percent inflation rate.
Why did central banks not move to literal price stability? For example, the Bank of Canada investigated the option of price-level targeting
regularly in its periodic review of Canada’s inflation-control target, but
so far it has always continued with a 2 percent inflation target.7
To understand why, our essay highlights four factors not considered
in the GK case for literal price stability. Section 2 analyzes the impact
of the effective lower bound (ELB) on nominal interest rates, which was
central to the revised formulation of the price stability objectives both
in the United States and the euro area. Section 3 reconsiders aspects
of the GK model, which focused on price rigidities and assumed a
perfectly competitive labor market. GK argued this is a reasonable
approximation when wages, at any point in time, are nonallocative. We
review some recent evidence on wage rigidity and analyses of its implications for the optimal inflation target. Section 4 considers the role
of fiscal policy. The main message is that when the ELB limits the ability
of monetary policy — acting on its own — to deliver price stability in
a timely way, fiscal policy can help to overcome the constraint. Hence,
interactions of fiscal and monetary policy become central for the
conduct of monetary policy at the ELB. Finally, GK’s analysis suggested
that literal price stability would also be optimal for a monetary union,
arguing that an integrated financial market would provide the necessary insurance against country-specific shocks. Section 5 makes a few
observations on the evolution of financial risk sharing in EMU against
that background.
Overall, we conclude that these factors significantly undermine the
case for literal price stability and support the current practice in most
advanced economies of targeting a small positive inflation rate.

7

E.g., Bank of Canada (2011).

E CB (2003), p. 81.
6
ECB (2021a).
5

164 |

| 165

Gaspar and Smets

The relevance of the effective lower bound on
nominal interest rates8
Marvin Goodfriend was an early advocate of the importance of interest
rates for central bank policy practice and analysis, but the GK analysis
was mainly concerned with the behavior of inflation and real activity.9
In his comments on GK at the conference, Gali (2001) observed that a
zero inflation policy would lead to a very low steady-state level of the
nominal interest rate and that GK had not considered the fact that a
central bank could not push down nominal interest rates below zero
(the ZLB). The assumption in GK was in line with the policy wisdom
at the time. For example, at the conference, Jose Vinals (2001) wrote:
“In sum, zero bound problems are very rare events and most of their
negative consequences can and should be avoided by preventive
measures.” In the same spirit, when reviewing and revisiting the ECB’s
monetary policy strategy in 2003, the consensus from a number of
studies was that the ZLB was unlikely to bind if the inflation norm was
set at 1 percent or higher.10 Such conclusions were predicated on a 2 to
3 percent range for the equilibrium or natural real interest rate.
Today the perspective is very different as a consequence of policy
experience and empirical evidence. Although Japanese short-term interest rates had come close to zero during the mid-1990s, it was only in
1999 that a zero interest rate policy was adopted by the Bank of Japan.
At the time, Japan was seen as an interesting but isolated case. Later,
the perspective changed as more and more central banks confronted limits on the use of interest rates in monetary policymaking. For
example, the ECB encountered the ELB during the sovereign debt crisis
in 2010-2011 and has yet to exit it, echoing the Japanese experience.
Many other central banks have faced similar circumstances, including
the Federal Reserve beginning in late 2008. COVID-19 once again
S ee ECB (2021b, p. 35) for a more elaborate discussion and analysis of the impact of
the ELB in the euro area.
9
Goodfriend’s 1991 “Interest Rates and Conduct of Monetary Policy” is a notable contribution that is reviewed by John Taylor elsewhere in this volume.
10
Issing (2003), p. 17.
8

Reconsidering Price Stability
made the ELB relevant for most advanced economies. Today, the ELB
has become a fact that policymakers have to take into account in the
normal conduct of policy.
The most important reason for why nominal interest rates have been
close to the ELB is the gradual fall in the natural real interest rate, frequently called r*. This has been a global phenomenon driven by a combination of factors such as lower population and productivity growth,
rising inequality, and higher demand for safe assets following the GFC.
Brand et al (2018) survey a range of estimates of r* for the euro area
from 1999 to 2019. While the estimated level of r* differs across methodologies, all estimates point to a significant decline over this period
from a range of 2 to 3 percent to one of 0 to -2%.
A lower natural rate implies that the ELB is more likely to keep a
central bank from lowering real rates to offset disinflationary forces. On
the basis of stochastic simulations using a variety of macroeconomic
models for the euro area, ECB (2021b, p. 36) shows that — for an inflation target of 2 percent — the time spent at the effective lower bound
increases from 10 percent to more than 30 percent as the equilibrium
real interest rate falls from 2 to 0 percent. The likelihood of a binding
ELB has also increased due to changes in estimated macro volatility.
In 2003, the variance of the demand and supply shocks affecting the
economy was assumed to be relatively low, consistent with the experience in the Great Moderation period. Following the GFC, the volatility
of the economy has increased. This higher volatility may be related
to the fall in the equilibrium real interest rate. Adam (2020) finds that
with a low r*, the sensitivity of the business cycle to asset price bubbles increases, which in turn increases the volatility of the economy,
the time spent at the ELB, and the optimal inflation target. A higher
inflation target reduces the relevance of this constraint as, for example,
also shown in Andrade et al (2019), who find that the optimal inflation
target increases by 0.9 percentage point for each 1 percentage point
fall in r*. An additional factor that may have contributed to an increase
in the optimal inflation target is increased inequality. It not only may
have contributed to the fall in the equilibrium real rate,11 but can also
11

166 |

Mian et al (2021).

| 167

Gaspar and Smets
make the economy more sensitive to the real interest rate increases
that occur at the ELB as low-income households typically are more
affected and have a larger propensity to consume than richer ones.12
Without the use of nonstandard monetary policy measures such
as forward guidance or asset purchases, a 2 percent inflation target is
likely not enough to avoid a disinflationary bias when the equilibrium
rate is zero. Depending on the model used, this disinflation bias can
be sizeable, even with a 2 percent inflation target.13 Forward guidance
and other nonstandard policy measures such as asset purchases and
targeted long-term lending operations can however help to overcome
this bias as, for example, shown in Coenen et al (2020, 2021), Gerke,
Kienzler, and Scheer (2021), and Mazelis, Motto, and Ristiniemi (2021).
The ECB’s new monetary policy strategy consequently takes into
account the implications of the ELB for its reaction function.14 In particular, when the economy is close to the lower bound, the commitment
to the symmetry of its 2 percent inflation target requires especially
forceful and persistent monetary policy measures. These preclude
negative deviations of inflation from the target becoming entrenched.
Such forceful action may include the use of large-scale asset purchases
or targeted long-term refinancing operations, which are now part of
the central bank’s instrument set and were implemented in response
to the pandemic crisis. In addition, closer to the ELB, a more persistent
use of such instruments may be necessary, which could lead to a
transitory period in which inflation is moderately above target. The Fed
has implemented the more persistent use of its instruments through
an asymmetric average inflation targeting framework. By contrast, the
ECB has implemented the need for persistence and patience through a
strengthened threshold-based interest rate forward guidance.

Reconsidering Price Stability

Downward nominal wage rigidities and the inflation buffer 15
GK assumed a perfectly competitive labor market. While they acknowledge that there may be labor market frictions that lead to equilibrium unemployment, they argue that wages — at any given point in
time — do not play an allocative role. In such an environment, nominal
rigidity in wage formation does not matter. At the conference, though,
Wyplosz (2001) argued that the importance of downward nominal
wage rigidity (DNWR) may justify a positive inflation target, following
the work of Akerlof, Dickens, and Perry (1997). DNWR leads to a nonvertical long-run Phillips curve and introduces an exploitable monetary policy trade-off at low levels of inflation. Based on estimates of the
slope of a long-run Phillips curve for the euro area, Wyplosz suggested
an optimal inflation target of 4 percent.
Since 2001, a lot of empirical evidence has been collected on the relevance of price and wage rigidities in the euro area and beyond. Consolo et al (2021) review the accumulated evidence on price and wage
rigidities in the euro area since the early 2000s: they find evidence of
both price and wage rigidities in the euro area. However, price flexibility may have increased during the EMU period, in particular in the
more traded nonenergy industrial goods category, and there is little
evidence of pervasive downward price rigidity. By itself, this would
suggest that a zero inflation target is optimal to avoid misallocations
due to inefficient relative price changes when inflation is positive.16
In contrast, evidence from the ECB’s Wage Dynamics Network (WDN)
surveys suggest that the length of wage contracts may have increased
and, more importantly, that nominal base wages are very sticky downward. According to the WDN surveys, nominal base wage cuts are very

15
16

F ernández-Villaverde et al (2020).
ECB (2021b), p. 37.
14
ECB (2021a).
12
13

168 |

T his section is based on Consolo et al (2021).
Note, however, that studies combining the frequency of price changes with the fact
that goods prices tend to decrease over their life cycle suggest that a substantial
positive inflation target would still be needed to minimize misallocations over time.
Adam et al. (2021) estimate that the positive inflation buffer needed to account for
these effects might well be above 1 percent in the euro area.

| 169

Gaspar and Smets
rare among euro area firms. Remarkably, this was the case even during
the period 2010-2013 despite the length and severity of the sovereign
debt crisis.17 Downward nominal rigidity during the crisis is further
suggested by the fact that the percentage of firms freezing base
wages increased dramatically, reaching its peak during 2008-09, before
declining over the period 2010-13. The WDN surveys also find that the
wages of new hires are closely related to those of incumbents suggesting that wages do play an allocative role.18
For the US economy, using high-quality administrative data, Grisgby
et al. (2021) finds that downward wage rigidity is more pervasive than
previously measured with the nominal base wage declining only for 2
percent of job stayers. These researchers also find that the flexibility of
base wages of new hires is similar to that of incumbent workers.
This euro area and US evidence suggests that higher real wages due
to a binding downward nominal wage constraint may lead to persistent effects of aggregate demand on unemployment, with relatively
high real wages depressing hiring and increasing unemployment duration. For instance, according to Daly and Hobijn (2014), DNWR provides
a rationale for persistent US output losses during the GFC.
Turning to the implications for monetary policy, we stress that
DNWR provides a rationale for a positive inflation buffer as it “greases
the wheels” of the labor market.19 New Keynesian DSGE models with
exogenous growth that embed DNWR find that the optimal inflation
rate is positive, although it is usually below 2 percent. Specifically, the
calibrated DSGE model developed in Consolo et al (2021) provides a
point estimate for the optimal inflation rate of about 1.2 percent with a
confidence band ranging from 0.2 to 1.6 percent.

S ee Consolo et al. (2021), section 2.2.
See, e.g., Galuscak et al. (2012).
19
E.g., Akerlof, Dickens, and Perry (1997).
17
18

170 |

Reconsidering Price Stability
A surprising conclusion of some recent research is that the introduction of DNWR in a model with the ZLB reduces the optimal inflation
rate.20 The mechanism is that wage rigidities limit the frequency and
the persistence of the ZLB by keeping marginal costs relatively higher.
In the quantitative analysis of Consolo et al (2021), discussed earlier,
the introduction of the ZLB leads to a lower optimal inflation rate from
1.2 to 0.3 percent. These results are, however, overturned in DSGE
models that feature equilibrium unemployment and endogenous
growth as in Abritti and Consolo (2021). Such models support a symmetric inflation buffer around 2 percent.21 From a welfare perspective,
the optimal rate of inflation balances welfare costs of price inflation
distortions and hysteresis effects on output and unemployment. Over
all, recent empirical studies and investigations with quantitative models lead to a robust conclusion that DNWR leads to a positive average
optimal inflation rate, even in the presence of the ELB — reinforcing
the logic that has led the ECB to choose a 2 percent rather than 0 percent target over the years.

Fiscal-monetary policy interaction and the ELB
The GK framework identifies a clear division of labor between fiscal
and monetary policy. Optimal fiscal policy compensates for the permanent distortions in the economy. For example, a wage subsidy can
compensate for the average markup. Optimal monetary policy, in contrast, focuses on stabilizing the markup over the course of the business
cycle. Such a perspective leads to fiscal policy as structural policy. Fiscal policy is crucial to provide incentives to work and save, accumulate
knowledge, and innovate. Fiscal policy is, therefore, aimed at growth,
employment, resource allocation, and the distribution of income and
wealth. In contrast, monetary policy focuses on keeping the economy
close to potential at business cycle frequencies.22
 illi and Galí (2020) and Amano and Gnocchi (2021).
B
See also ECB (2021b), p. 34.
22
See Tabellini (2001).
20
21

| 171

Gaspar and Smets
Recent reviews of monetary policy strategy by the Federal Reserve
and by the ECB explicitly recognize the relevance of the ELB for monetary-fiscal interactions.23 Specifically, when the optimal policy interest rate is substantially lower than allowed by the ELB, the monetary
authority typically uses forward guidance and communicates that
interest rates will be kept low for an extended period. Fiscal multipliers
are larger under these conditions than when monetary policy adjusts
the policy rate according to a typical reaction function. Fiscal policy
thus can assist monetary policy in avoiding recessions and restoring
price stability. The power of fiscal policy is greatest when it is needed
the most.
Ramey and Zubairy (2018) provide empirical evidence of a government spending multiplier of 1.5, for the United States, at the ZLB.
Similarly, Klein and Winkler (2021), using an historical panel, also find a
multiplier of 1.5 at the ZLB. There is a vast model-based literature arguing that multipliers can be very large at the ZLB. It includes Christiano,
Eichenbaum, and Rebelo (2011), Eggertsson (2011), Woodford (2011),
and Woodford and Xie (2020). Although Boneva, Braun, and Waki
(2016) show that the effect of fiscal policy may be smaller than multipliers reported in the literature, they still come out with a multiplier
above 1, at 1.05.
ECB (2021c, p. 72 and ff ) presents simulations using the ECB-BASE
model that shed light on monetary-fiscal policy interactions at the ELB.
Looking at negative scenarios, the simulations show that fiscal stabilization policy stays effective at the ELB. In contrast, monetary policy
rates are constrained so that the contribution from monetary policy
is limited to nonstandard measures. In the concluding section, ECB
(2021c) argues that there are substantial benefits from monetary-fiscal
policy coordination that would result from well-calibrated changes to
the euro area macroeconomic policy architecture.
23

See, especially, ECB (2021c).

Reconsidering Price Stability
The bottom line is that the ELB limits the ability of monetary policy
— acting on its own — to deliver price stability in a timely way. But fiscal policy can help to overcome the constraint. Hence, fiscal-monetary
policy interactions are becoming central for the conduct of monetary
policy at the ELB.

Financial frictions, monetary operations,
and credit market imperfections
GK use principles of public finance to determine whether price
stability can be expected to be a good approximation of optimal
monetary policy under more general circumstances than afforded
by the basic closed economy model. One very interesting extension
discussed by GK is the case of a small open economy, taken to mean
that private agents and the government can trade in complete world
financial markets at given prices. In such a setting, financial variables
and wealth are stabilized by access to world finance. The small open
economy gets full insurance, at fair prices, against idiosyncratic shocks.
The setup is not without problems. Tirole (2002) argues that a country’s
moral hazard limits its access to financing. In the presence of asymmetric/incomplete information and government incentive problems, the
complete market assumption may not be a particularly useful benchmark. In general, the financial structure will have to reflect government
information and incentives.
But, in 2001, monetary unification was seen as a major force leading
to financial market integration. Financial union — a single European
market — might not yet be realized but, with time, deeper financial
integration would amplify the benefits from the euro. Fast progress
since the late 1990s toward the integration of bond markets and interbank money markets seemed in line with such an optimistic view. At
the global level, too, there was the Great Moderation. It was a period of
relative stability from the late 1980s to the GFC. After the first 10 years
of the euro, the prospects looked bright.
Unfortunately, the process of gradual financial integration in Europe
was put into reverse as the GFC morphed into sovereign debt crises. As
feared by the founding fathers of the euro, financial markets abruptly

172 |

| 173

Reconsidering Price Stability

Gaspar and Smets
fragmented under stress.24 A strong correlation emerged between
sovereign risk and bank risk. The phenomenon became so salient that
it got a special name: the doom loop.25 One version of the mechanism
is as follows. First, the sovereign comes under stress, so that bank balance sheets deteriorate given their exposure to the sovereign. Second,
financing conditions tighten as credit spreads widen. Third, investment
and economic activity declines, leading loan defaults to increase. In
turn, the balance sheets of banks deteriorate further. Finally, public
debt and deficits widen, closing the loop. This loop is one explanation
of the well-known high correlation between private and public credit
default swaps.
Now, it is impossible for us to see the abstract complete markets
model as a relevant benchmark. During the decade beginning in 2008,
the fragilities of the European pattern of financial integration came
into sharp focus. For central banking, these fragilities affected the
monetary transmission mechanism and, through it, the single monetary policy.
One particular challenge may come from self-fulfilling debt crises.
When the sovereign is vulnerable and there is the possibility of a
rollover crisis, there may be multiple equilibria. In such a crisis, there
is a new role for a central bank: it can tilt the balance toward the good
equilibrium.26 In some cases, action through intervention in markets
may not even be required. The euro area provided the perfect example
in 2012. In the early summer, bond yields on Spanish and Italian bonds
were heading to levels that in previous experience had triggered bond
Delors Report, p. 20.
T here is a voluminous literature on the sovereign-bank nexus. See Schnabel (2021)
for a recent survey from an influential policymaker. References include Acharya,
Drechsler, and Schnabl (2014), L. Bocola, (2016), Brunnermeier et al (2016), and Farhi
and Tirole (2014).
26
See Corsetti and Dedola (2016), Bocola and Dovis (2019), and Lorenzoni and Werning
(2019).
24
25

market crises. It was in this context that Mario Draghi famously stated:
“Within our mandate, the ECB is ready to do whatever it takes to preserve the euro.” The statement was enough to calm markets. “Whatever
it takes” became the three most famous words in the history of European monetary and financial integration.27
An important implication for central bankers is that, given multiple
financial market frictions and the fragmentation of the single financial
market, the operational framework for monetary policy implementation becomes very relevant. The simplicity of monetary policy implementation through a money market interest rate is lost. The management of the central bank’s balance sheet becomes a matter of concern.
The details of monetary policy implementation matter. This is particularly the case in crises. The last decade provides a rich illustration.

Summing up
As presented at the first ECB central banking conference in 2001, the
GK analytical policy framework based on imperfect competition and
price rigidity provided an intuitive and elegant rationale for central
banks pursuing literal price stability. In this essay, we argue that
additional frictions — including the ELB on nominal interest rates and
DNWR — can explain why central banks in the advanced economies
have instead targeted a small positive inflation rate of 2 percent.
We also argue that monetary policy may not be enough to achieve
the 2 percent target. With a zero or negative natural real interest rate,
fiscal policy may also have to take on a macroeconomic stabilization
role to avoid inflation becoming trapped too close to zero. Moreover,
incomplete banking union and capital markets union makes departures from the complete contingent markets’ paradigm employed by
GK painfully visible. The magnitude of financial instability and its consequences for economic activity and employment were salient during
the sovereign debt crises in the euro area from 2010-13.
27

174 |

A useful reference is Saka, Fuertes, and Kalotychou (2015).

| 175

Gaspar and Smets
Yet, our bottom line is that the case for price stability, understood as
low and stable inflation, at say 2 percent, remains strong. The gradual
fall of the real natural interest rate, in the euro area, to the range of
0 to -2 percent, makes it more likely that policy interest rates will be
constrained by the ELB. In such circumstances, it is important that the
conduct of policies be patient and persistent, with particular attention
paid to the interaction of monetary and fiscal policy.

Reconsidering Price Stability

References
Abbritti, M., A. Consolo, and S. Weber. 2021. “Endogenous Growth,
Downward Wage Rigidities and Optimal Inflation.” International Monetary Fund Working Paper 208, August 6.
Acharya, Viral, Itamar Dreschler, and Philipp Schnabl. 2014. “A Pyrrhic
Victory? Bank Bailouts and Sovereign Credit Risk.” Journal of Finance 69,
no. 6 (December): 2689-2739.
Adam, K. 2020. “Monetary Policy Challenges from Falling Natural
Interest Rates.” Paper prepared for the ECB Forum on Central Banking,
European Central Bank, November 11-12.
Adam, K., E. Gautier, E. Santoro, and H. Weber. 2021. “The Case for a
Positive Euro Area Inflation Target: Evidence from France, Germany and
Italy.” European Central Bank Working Paper 2575, July.
Akerlof, G.A., W.T. Dickens, and G.L. Perry. 1996. “The Macroeconomics
of Low Inflation.” Brookings Papers on Economic Activity 1, no. 27: 1–59.
Amano, R., and S. Gnocchi. 2021. “Downward Nominal Wage Rigidity
Meets the Zero Lower Bound.” Journal of Money, Credit and Banking,
forthcoming.
Andrade, P., J. Galí, H. Le Bihan, and J. Matheron. 2019. “The Optimal
Inflation Target and the Natural Rate of Interest.” Brookings Papers on
Economic Activity, Fall, 173-255.
Bank of Canada. 2011. “Renewal of the Inflation-Control Target: Background Information.” November.
Billi, R.M., and J. Galí. 2020. “Gains from Wage Flexibility and the Zero
Lower Bound.” Oxford Bulletin of Economics and Statistics 82, no. 6: 123961.
Bocola, Luigi. 2016. “The Pass-Through of Sovereign Risk.” Journal of
Political Economy 124, no. 4 (August): 879-926.
Bocola, Luigi, and Alessandro Dovis. 2019. “Self-Fulfilling Debt Crises: A

176 |

| 177

Gaspar and Smets
Quantitative Analysis.” American Economic Review 109, no. 12 (December): 4343-77.

Reconsidering Price Stability
ing Paper 2013-08, January.

Boneva, Lena, R. Anton Braun, and Yuichiro Waki. 2016. “Some Unpleasant Properties of Log-linearized Solutions When the Nominal Rate is
Zero.” Journal of Monetary Economics 84 (December): 216-232.

Delors, Jacques. 1989. “Report on Economic and Monetary Union in the
European Community (commonly called the Delors Plan or Report).”
Committee for the Study of Economic and Monetary Union, presented
April 17.

Brand, C., M. Bielecki, and A. Penalver (eds). 2018. “The Natural Rate of
Interest: Estimates, Drivers, and Challenges to Monetary Policy.” European Central Bank Occasional Paper Series 217, December.

European Central Bank. 1999. “The Stability-Oriented Monetary Policy
Strategy of the Eurosystem.” European Central Bank Monthly Bulletin,
January: 39–50.

Brunnermeier, Markus, Luis Garicano, Philip Lane, Marco Pagano,
Ricardo Reis, Tano Santos, David Thesmar, Stijn Van Nieuwerburgh, and
Dimitri Vayanos. 2016. “The Sovereign-Bank Diabolic Loop and ESBies.”
American Economic Review 106, no. 5 (May): 508-512.

European Central Bank. 2003. “The Outcome of the ECB’s Evaluation of
its Monetary Policy Strategy.” European Central Bank Monthly Bulletin,
June: 79-92.

Christiano, L., M. Eichenbaum, and S. Rebelo. 2011. “When Is the Government Spending Multiplier Large?” Journal of Political Economy 119,
no. 1 (February): 78-121.
Coenen, F., C. Montes-Galdón, and F. Smets. 2020. “Effects of State-Dependent Forward Guidance, Large-Scale Asset Purchases and Fiscal
Stimulus in a Low-Interest-Rate Environment.” European Central Bank
Working Paper 2352, January.
Coenen, G., C. Montes-Galdón, and S. Schmidt. 2021. “Macroeconomic
Stabilisation and Monetary Policy Effectiveness in a Low-Interest-Rate
Environment.” European Central Bank Working Paper 2572, July.
Consolo, A., G. Koester, C. Nickel, M. Porqueddu, and F. Smets. 2011.
“The Need for an Inflation Buffer in the ECB’s Price Stability Objective
– the Role of Nominal Rigidities and Inflation Differentials.” European
Central Bank Occasional Paper Series 279, September.

European Central Bank. 2021a. “An Overview of the ECB Monetary Policy Strategy.” European Central Bank Economic Bulletin Issue 5/2021.
European Central Bank. 2021b. “The ECB’s Price Stability Framework:
Past Experience, and Current and Future Challenges.” European Central
Bank Occasional Paper Series 269, September.
European Central Bank. 2021c. “Monetary – Fiscal Policy Interactions
in the Euro Area.” European Central Bank Occasional Paper Series 273,
September.
Eggertsson, G. 2011. “What Fiscal Policy is Effective at Zero Interest
Rates?” In NBER Macroeconomics Annual 2010, Volume 25, edited by
Daron Acemoglu and Michael Woodford, 59-112. Chicago: University
of Chicago Press.
Farhi, Emmanuel, and Jean Tirole. 2014. “Deadly Embrace: Sovereign
and Financial Balance Sheet Doom Loops.” Working Paper.

Corsetti, G., and L. Dedola. 2016. “The Mystery Of The Printing Press:
Monetary Policy And Self-Fulfilling Debt Crises.” Journal of the European
Economic Association 14, no. 6 (December): 1329-1371.

Fernández-Villaverde, J., J. Marbet, G. Nuño, and O. Rachedi. 2021. “Inequality and the Zero Lower Bound.” Banco de Espana Working Paper,
March.

Daly, M.C., and B. Hobjin. 2014. “Downward Nominal Wage Rigidities
Bend the Phillips Curve.” Federal Reserve Bank of San Francisco Work-

Galí, J. 2001. “Comment on Goodfriend and King.” In Why Price Stability? Proceedings of the First ECB Central Banking Conference, edited by

178 |

| 179

Gaspar and Smets
A. Garcia-Herrero, V. Gaspar, L. Hoogduin, J. Morgan, and B. Winkler.
Frankfurt, Germany: European Central Bank.
Galuscak, K., M. Keeney, D. Nicolitsas, F. Smets, P. Strzelecki, and M.
Vodopivec. 2012. “The Determination of Wages of Newly Hired Employees: Survey Evidence on Internal Versus External Factors.” Labour
Economics 19, no. 5: 802-812.
Gerke, R., D. Kienzler, and A. Scheer. 2021. “Unconventional Monetary
Policies When the Policy Rate is Constrained by a Lower Bound.” Deutsche Bundesbank Technical Paper, forthcoming.
Goodfriend, M., and R.G. King. 1997. “The New Neoclassical Synthesis
and the Role of Monetary Policy.” In NBER Macroeconomics Annual 1997,
Volume 12, edited by Ben S. Bernanke and Julio J. Rotemberg, 231-283.
Chicago: University of Chicago Press.
Grisgby, J., E. Hurst, and A. Yildirmaz. 2021. “Aggregate Nominal Wage
Adjustments: New Evidence from Administrative Payroll Data.” American Economic Review 111, no. 2 (February): 428-471.
Issing, Otmar, ed. 2003. Background Studies for the ECB’s Evaluation of its
Monetary Policy Strategy. Frankfurt, Germany: European Central Bank.
Klein, Mathias, and Roland Winkler. 2021. “The Government Spending Multiplier at the Zero Lower Bound: International Evidence from
Historical Data.” Journal of Applied Econometrics 36, no. 6 (September/
October): 744-759.
Lorenzoni, Guido, and Iván Werning. 2019. “Slow Moving Debt Crises.”
American Economic Review 109, no. 9 (September): 3229-63.
Lucas, R.E., and N. Stokey. 1983. “Optimal Fiscal and Monetary Policy in
an Economy without Capital.” Journal of Monetary Economics 12, no. 1:
55–93.
Mazelis, F., R. Motto, and A. Ristiniemi. 2021. “The Price Stability Framework: Quantitative Evaluation of Key Policy Parameters and Approaches.” mimeo.
180 |

Reconsidering Price Stability
Mian. A., L. Straub, and A. Sufi. 2021. “What Explains the Decline in r*?
Rising Income Inequality versus Demographic Shifts.” Paper prepared
for the Federal Reserve Bank of Kansas City Symposium, Jackson Hole,
Wyoming, August 27.
Ramey, Valerie, and Sarah Zubairy. 2018. “Government Spending Multipliers in Good Times and in Bad: Evidence from US Historical Data.”
Journal of Political Economy 126, no. 2 (April): 850-901.
Ramsey, F.P. 1927. “A Contribution to the Theory of Taxation.” Economic
Journal 37, no. 145 (March): 47–61.
Saka, Orkun, Ana-Maria Fuentes, and Elena Kalotychou. 2015. “ECB Policy and Eurozone Fragility: Was De Grauwe Right?” Journal of International Money and Finance 54, no. C: 168-185.
Schnabel, Isabel. 2021. “The Sovereign-Bank-Corporate Nexus – Virtuous or Vicious?” Speech at the LSE Conference on “Financial Cycles,
Risk, Macroeconomic Causes and Consequences,” Frankfurt am Main,
Germany, January 28.
Tabellini, G. 2001. “Comments.” In Why Price Stability? Proceedings of
the First ECB Central Banking Conference, edited by A. Garcia-Herrero,
V. Gaspar, L. Hoogduin, J. Morgan, and B. Winkler. Frankfurt, Germany:
European Central Bank.
Tirole, Jean. 2002. Financial Crises, Liquidity and the International Monetary System. Princeton: Princeton University Press.
Viñals, J. 2001. “Monetary Policy Issues in a Low Inflation Environment.”
In Why Price Stability? Proceedings of the First ECB Central Banking Conference, edited by A. Garcia-Herrero, V. Gaspar, L. Hoogduin, J. Morgan,
and B. Winkler, Chapter 3. Frankfurt, Germany: European Central Bank.
Woodford, M. 2011. “Simple Analytics of the Government Expenditure
Multiplier.” American Economic Journal: Macroeconomics 3, no. 1 (January): 1-35.
Woodford, M., and Y. Xie. 2020. “Fiscal and Monetary Stabilization Policy
at the Zero Lower Bound: Consequences of Limited Foresight.” National
| 181

Gaspar and Smets
Bureau of Economic Research Working Paper 27521, July.
Wyplosz, C. 2001. “Do We Know How Low Inflation Should Be?” In Why
Price Stability? Proceedings of the First ECB Central Banking Conference,
edited by A. Garcia-Herrero, V. Gaspar, L. Hoogduin, J. Morgan, and B.
Winkler, Chapter 1. Frankfurt, Germany: European Central Bank.

182 |

| 183

Monetary Policy Consensus

The 2007 Monetary Policy
Consensus in Retrospect1
Mark Gertler
Keynes famously ends the General Theory with a description of the
“academic scribbler” whose ideas from “a few years back” eventually
find their way into policymaking. When Marvin Goodfriend wrote “How
the World Achieved Consensus on Monetary Policy” in 2007, that time
lag had largely disappeared, at least in central banking. For example, in
the Federal Reserve System it had become the norm for a substantial
fraction of those in important decision-making positions to hold PhDs
in economics. Not only did these officials have advanced degrees, they
often earned the stature that led them to be chosen for their position
by doing cutting edge research. The most prominent example was the
Federal Reserve chairman at the time, Ben Bernanke. It is also now the
case that the research done by staff at the Fed and other central banks
is as sophisticated as any that occurs in academia. As a result, ideas
flow freely and instantly between the halls of academia and central
banks. The time lag is gone.
An important theme of Marvin’s 2007 paper is how the development
of this symbiotic relationship between academic research and central
bank policymaking led to a consensus on a new framework for monetary policymaking, a framework that has proved highly useful and
remains with us today. What Marvin leaves out is the significant role
that he played in this development. To my knowledge, Marvin was the
first economist to simultaneously contribute to the modern literature
on monetary policy and hold a nontrivial policymaking job in the
Federal Reserve System. A happy coincidence is that at Brown he was
classmates with his future coauthor Bob King. The two would play an
important role in developing the New Keynesian framework and
1

184 |

Thanks to Bob King and Alex Wolman for helpful comments.

| 185

Gertler
making it operational for analyzing monetary policy. Being able to
work with Bob kept Marvin in close proximity to academia. At the
same time, Marvin’s experience with the policymaking process provided him with important insights into how to make their work, as well as
his other research, most useful in practice.

The early years: disarray and revolution in macroeconomics
When Marvin joined the Richmond Fed in 1978, communication
between central banking and academia may have been at an all-time
low. The failure of the large econometric models developed more than
a decade earlier to anticipate the stagflation of the late 1960s and ‘70s
left central banks without a framework to provide guidance for monetary policy, at least one in which they could have confidence. In academia, the rational expectations revolution was heating up. Popular at
the time was the Phelps/Friedman natural rate theory, which related
output to unanticipated movement in inflation. It was standard to
assume that expectations were formed adaptively, so that a monetary
expansion in the short run would increase real output temporarily but
then produce a subsequent increase in inflation. Rational expectations
turned things upside down: Within the context of the Phelps/Friedman
framework, predictable movements in the money supply (which produced predictable movements in inflation) had no effect on real output, as Robert Lucas (1972) famously showed. To put it mildly, central
bankers were not particularly hospitable to the idea that only unpredictable movements in the money supply could affect real activity.
A more extreme development from the vantage of central bankers
was the advent of real business cycle (RBC) theory, which involved
the use of the stochastic competitive equilibrium growth model to
explain business cycles. The virtue of the approach is the explicit use
of microfoundations to build a macroeconomic framework. A striking
implication, however, is the total irrelevance of monetary and financial factors. Another dramatic implication was that business cycles,
while unfortunate, represented efficient responses of the economy to
exogenous disturbances. Needless to say, this development did not
exactly enhance academic interaction with monetary policymakers.
186 |

Monetary Policy Consensus

The Volcker disinflation: consequences for research
and policymaking
As Marvin emphasizes, a critical turning point was the shift to tight
monetary policy in late 1979, engineered by Paul Volcker. The aim was
to bring the era of high inflation to an end. As Marvin describes, the
sudden and unexpected tightening can be thought of as a kind of natural experiment to study the impact of monetary policy on output and
inflation. The tightening succeeded in reducing inflation, though with
a lag. But in the process it induced the largest recession of the postwar
period up to that point. As Marvin notes, the episode sent a clear message to central bankers: they did have the ability to control inflation. At
the same time, disinflations were not costless, even if the factors that
determined these costs were not clearly understood.
I would add that the Volcker disinflation also had a profound effect on the course of academic research. It was clear that neither the
Phelps/Friedman model with rational expectations nor RBC theory
could easily account for the effect of the Volcker tightening on output
and inflation dynamics. The need for a new framework was obvious.
But it was also clear that the field could not retreat from the methodological advances ushered in by the rational expectations/RBC
revolution. These considerations led to an effort to rebuild Keynesian
economics using microfoundations. Out of this effort would emerge a
framework that could be used — and eventually would be used — in
the policymaking process. No, the framework has not come anywhere
close to the point where it can be used to put monetary policy on automatic pilot. But it has reached the point where it does play a significant role in helping organize thinking about policy implementation.
As a result, the relationship between academic research and central
bank policymaking has become highly symbiotic. Economic events
influence the development of the model. The model in turn informs
policymaking.

| 187

Gertler
Marvin’s work with Bob King (Goodfriend and King, 1997) played a
significant role in the development of what is now widely known as
the New Keynesian (NK) model.2 Marvin and Bob perhaps more aptly
refer to this paradigm as the New Neoclassical Synthesis, as it begins
with an RBC model and then adds three crucial ingredients. First, money is introduced so that the model can account for nominal variables.
Second, monopolistic competition is incorporated so that it is possible to characterize price setting by firms. Third, nominal rigidities are
added, which gives rise to the nonneutraliy of money and inefficient
fluctuations in output.3 Absent nominal price rigidity, the framework
behaves essentially as an RBC model. With nominal price rigidity, the
Keynesian features emerge.

The interest rate as the policy instrument
Another important component of the consensus that Marvin emphasizes is the use of the short-term interest rate as the instrument of
monetary policy, in keeping with actual practice at central banks. As
late as the 1980s, it was still commonplace in academic work to model
the money supply as the policy instrument. However, central banks
have learned through practical experience that trying to directly regulate monetary aggregates was problematic. Broad monetary aggregates were difficult to control due to the endogeneity of inside money.
Controlling narrow aggregates like reserves generated wild gyrations
in interest rates due to fluctuations in reserve demand. These wild fluctuations in interest rates, in turn, wreaked havoc on the economy.
Given his proximity to policymaking, Marvin quickly saw that to get
the attention of central bankers, academic work needed to treat the
interest rate as the policy instrument. Indeed, Marvin was among the
 elated work includes Rotemberg and Woodford (1997), Clarida, Gali, and Gertler
R
(1999), Woodford (2003), Christiano, Eichenbaum, and Evans (2005), and Gali (2015).
For a collection of the early contributions to the New Keynesian framework, see
Mankiw and Romer (1991). For a critique, see Chari, Kehoe, and McGrattan (2009).
3
The most common way to incorporate nominal rigidities is via the staggered contracting approach of Calvo (1983), which is a more tractable version of Taylor’s (1980)
overlapping contracts framework.
2

188 |

Monetary Policy Consensus
earliest researchers to interpret monetary policy actions through the
lens of interest rate decisions, not only about current rates settings, but
also about communication of the paths of future rates.4

A monetary policy framework
Overall, Marvin Goodfriend played a key role in developing a framework for monetary policy that facilitated interaction between academic researchers and policymakers at central banks. I now employ a
version of this approach that, while simple, is sufficiently rich to help
organize thinking about some of the key issues facing central bankers.
Let denote log real output, the natural (flexible price) level of output, and the output gap, where each variable is a log deviation from
the state. Next, let be the nominal interest rate, inflation, the
central bank’s target inflation rate, the natural real rate of interest,
and a cost push shock.5 Then we can express the model as follows:
								(1)

Equation (1) decomposes output into the sum of the output gap and
the natural level of output. Simply put, the New Keynesian features
determine while the RBC framework characterizes the variation
in . A key theme of Goodfriend-King was that there is no reason to
think should evolve as a smooth linear trend, as was the traditional
approach in policy circles. Rather, should fluctuate in a manner that
RBC theory suggests. A classic application of this thinking occurred in
the mid-1990s when a productivity boom ushered in a period of
S ee Marvin’s paper “Interest Rates and the Conduct of Monetary Policy” (1991), which
is also discussed in this volume, in an essay by John Taylor.
5
The cost push shock can be interpreted as a transitory fluctuation in the desired
markup. See Gali (2015).
4

| 189

Gertler
strong growth. Pressure mounted on the Greenspan Federal Reserve
to raise rates to slow down growth for fear of subsequent inflationary
pressures. However, Greenspan correctly perceived that supply side
factors were generating the boom and wisely chose to accommodate
it.6
Equation (2) is the familiar New Keynesian IS curve that relates the
output gap inversely to the gap between the real interest rate and
the natural rate plus the expected future output gap.7 From a policy
perspective, there are two key features of this formulation. First, as is
recognized in both theory and practice, an important benchmark for
rate setting is the natural interest rate . The notion of a natural or
“equilibrium” real rate is not new: it dates back to Wicksell. What is new
is the use of to judge the stance of policy. Of course, like ,
is not directly observable. As the model implies, however, one can
use the behavior of inflation to infer the direction of the error in the
estimate. For example, if is lower than forecast, then the central
bank may be setting interest rates higher than desired, resulting in a
lower than desired output gap. The net effect, as can be seen from the
aggregate supply curve (3) is that inflation will be lower than expected.
Hence, the surprise in inflation can be used to update the estimate of
. The use of as benchmark in the policy process is now standard,
as the theory would predict.
A second key insight from the New Keynesian IS curve, one that
Marvin strongly emphasizes, is that credible communication about
the future path of policy is critical. To illustrate, let’s consider the case
where the central bank’s target inflation rate is zero. As any first-year
graduate student knows, one can iterate equation (3) forward to obtain an expression that links the output gap inversely to the expected
path of the interest rate gap. The expression makes clear that
T o be fair, the issue becomes murkier later on in the boom as inflationary pressures
mounted. According to Alex Wolman, Marvin argued at the time that the high productivity growth had pushed up , suggesting it was time to raise rates, as the Fed
did shortly after.
7
As is well known, the relation comes from the consumption Euler equation, given an
economy with consumption goods only. Gali and Gertler (2007) show how to generalize to the case where investment is present as well.
6

190 |

Monetary Policy Consensus
monetary policy management is not simply about current rate setting,
but also about managing market expectations of the path of future
rates. To close the output gap, for example, it is not only necessary
to set the nominal rate equal to the natural rate, the central bank
must also credibly promise to set the future path of equal to
.
Communication about future policy, the importance of which comes
naturally out of this simple model, indeed plays a central role in the
monetary policymaking process. The framework also makes clear
why communication — or “forward guidance” as it is known today —
should not take the form of promising a path of rates: the natural rate
is likely to vary in unexpected ways, which will affect the appropriate
rate setting. Hence, as it has evolved in practice, communication must
always stress the “data dependence” of rate setting.
The movement of the economy to the zero lower bound in 2008
pushed forward guidance to center stage. As the simple framework
makes clear, when the natural rate becomes negative, the zero lower
bound constraint, equation (5), becomes binding. Aside from unconventional policies — which we briefly mention later — the central
bank’s only option for stimulating the economy at the ZLB is to use
forward guidance. In particular, the central bank must promise to keep
rates low after the economy has emerged from the ZLB. The tension
is that since keeping rates low after the storm has passed could be
inflationary, the central bank may be tempted to renege on its promise. As made clear by Eggertsson and Woodford (2003) and Werning
(2012), a central bank confronting a liquidity trap must commit to keep
rates “lower for longer,” which will involve some overshooting of the
inflation target, once the economy leaves the liquidity trap. Again, we
have another example of how theory meets policymaking in practice.
Throughout the recent history of operating at the ZLB, central banks
in the industrializesd world have opted for forward guidance with an
emphasis on a lower for longer strategy for interest rates, along with a
temporary overshooting of the inflation target.

Inflation targeting and trend inflation
As Marvin emphasizes, a critical reason for reaching consensus on
monetary policy was the success in moving from high and volatile
| 191

Gertler
inflation to a prolonged period of low and stable inflation. Here also
the academic work provided useful insight to guide policy. Though I
add the caveat that what it had to offer was not completely satisfactory, particularly with regard to how the central bank can manage
private sector beliefs about trend inflation, as I discuss shortly.
At a most basic level, the challenge for a central bank in maintaining
inflation is finding an appropriate nominal anchor. For this purpose,
from 1944 until 1973, a number of the major central banks agreed to
maintain a fixed exchange rate against the dollar while the Federal Reserve tied the dollar to the price of gold. The loss of monetary independence eventually made the system unworkable, especially as inflation
pressures had been building in the US in the late 1960s/early 1970s.
There was a brief flirtation with using money growth as the nominal
anchor. But as noted earlier, broad monetary aggregates proved difficult to control while targeting narrow aggregates like reserves typically
introduces disruptive gyrations in interest rates. The failure of these
traditional nominal anchors led both central bankers and academics to
view an inflation target as the most effective nominal anchor.8 Indeed,
in the early 1990s a number of central banks adopted an explicit
inflation target. The Federal Reserve began communicating as if it had
a 2 percent inflation target in the early 1990s before eventually formalizing this policy under Chair Bernanke in 2012. Now virtually all the
major central banks in the industrialized world have adopted a formal
inflation target.
The challenge that inflation targeting poses for both central banks
and academics is twofold: First, if trend inflation differs from the desired target, what is the best way to engineer a convergence to target?
Second, if indeed trend inflation is in close range of the target, how
should the central bank manage policy to achieve the dual mandate of
price and output stability. Both these issues have received enormous
attention in the academic literature.
8

192 |

See, for example, Bernanke, Laubach, Mishkin, and Posen (1998).

Monetary Policy Consensus
To sharpen the focus, it is useful to express the Phillips curve (3) in a
way that allows for variable trend inflation. Let =
market
be market expectations of trend inflation,
the cyclical
expectations of the trend output gap, and
component of the output gap. Assume that both and obey stationary first order processes with serial correlation parameters, and
respectively. Finally, suppose the inflation target is zero. Then following Hazell, Herreno, Nakamura, and Steinsson (2021), we can express
inflation as
		
where

(6)

is given by

		
and with
. As equation (6) makes clear,
inflation depends not only on excess demand captured by and cost
push shocks captured by , but also on market expectations of trend
inflation. Indeed, as Hazell, Herreno, Nakamura, and Steinsson and
others have shown, most of the variation in inflation over the postwar
has been due to the trend term.
As Marvin argues, central bank credibility is key to understanding
may differ persistently from target. It is also key to understandwhy
ing the costs of engineering it to target (in terms of undesired output
fluctuations). Here the academic literature took the lead. The classic
paper by Kydland and Prescott (1977) motivates how positive trend
inflation could emerge when the central bank is operating under discretion and is tempted to push output above its flexible price equilibrium.9 With a credible commitment toward not generating a surprise
inflation, the problem disappears. Central banks quickly adopted the
idea that credibility was critical for controlling inflation. Indeed, one
could argue further that this literature provided the foundation for the
move toward inflation targeting.10
I n the New Keynesian framework, due to imperfect competition, the flexible price
equilibrium level of output is below the efficient level, creating an incentive for the
central bank to want to push output above .
10
See, for example, Bernanke, Laubach, Mishkin, and Posen (1999).
9

| 193

Gertler

Establishing credibility
Where the literature has been somewhat silent, however, is on exactly how a central bank establishes credibility. History suggests that
central banks cannot simply announce that they are going to make a
credible commitment.
Rather, they must earn the private sector’s trust through experience.11 Here Marvin’s description of the Volcker disinflation is instructive. What the theory suggests is that if Volcker had been perfectly
credible at the outset, the announcement of the monetary tightening
would have induced a drop in inflation to target with minimal cost in
terms of output loss. But there was little reason for the private sector to
take Volcker’s promises at face value. Using Marvin’s terminology, the
late 1960s-1970s was an era of “stop-go” policy: the central bank would
periodically tighten but then let up as the economy weakened even
though inflation remained high. Compounding matters for Volcker was
his initial policy reversal: after the aggressive move toward tightening
in October 1979, there was an equally dramatic drop in rates in 1980.
This move was likely costly in terms of central bank credibility, having
the practical effect of slowing the convergence of beliefs about trend
inflation to target. The implication, as equation (6) makes clear, is that
the disinflation would entail a costly recession before inflation reached
target. Marvin’s broader point, I think, is that central banks cannot
simply be bestowed with credibility; they need to earn it by showing
through experience that they can deliver on their promises.
Another way that the central bank can enhance its credibility is by
setting the policy instrument in a way that is clearly consistent with
its objectives. As Taylor (1993) notes, a policy rule that achieves this
objective is the simple interest rate feedback given by equation (4) but
is set equal to the
with two key restrictions. First, the trend term
inflation target (in our example zero). Second, the feedback coefficient
on inflation,
exceeds unity. As a result, whenever inflation exceeds
target, the central bank increases the nominal rate sufficiently to raise
11

194 |

E rceg and Levin (2001) make some progress in analyzing how a central bank might
establish credibility: they assume the private sector updates its beliefs about the
central bank’s time varying trend inflation rate by using the variation in the policy

Monetary Policy Consensus
the real rate. This action reduces demand, pushing inflation back to
target. Taylor showed that the Greenspan Federal Reserve — a central
bank determined to establish and maintain credibility — set rates in a
manner consistent with this rule. From Marvin’s perspective, the Taylor
rule offered a practical guideline for implementing inflation targeting.
To be clear, it is nowhere near the point of being a mechanical rule that
central banks can use to put monetary policy on automatic pilot, especially given that two key ingredients, and , are not directly observable. Nonetheless, the rule does offer a guideline for framing the policy
discussion in a way that connects to the general inflation targeting
framework. It is not an exaggeration to suggest that at least from the
early 1990s to the eve of the Great Recession in 2007, the great majority of central banks in the industrialized countries adopted the inflation
targeting/Taylor rule (guideline) approach. The prolonged period of
low inflation and stable output growth only served to reinforce the
consensus.

After the consensus: developments from 2007-2021
Marvin wrote “How the World Achieved Consensus...” in 2007, just
before the global financial crisis of 2008-2009. Of course, the New
Keynesian model could not directly capture the crisis, given the absence of financial market frictions. Nor was it useful for understanding
the myriad of unconventional credit market interventions aimed at
containing the crisis.
Nonetheless, in certain dimensions it provided important insights.
Central bank interest rate strategy came directly out of the New
Keynesian analysis of the ZLB, which featured forward guidance and
“lower for longer.” The establishment of a credible inflation target, as
the theory prescribes should be done, helped keep inflation stable in
the face of a sharp contraction in real activity. As a result, a destructive
deflation was avoided.
Indeed, the inflation targeting/Taylor rule framework appears to remain intact today at many central banks, with some adjustments that
 rate. Exactly why the central bank’s preferred trend inflation is exogenous and time
varying remains an open question, though.

| 195

Gertler

Monetary Policy Consensus

take into account the experience of financial crisis: the policy toolkit
now includes some of the unconventional tools employed in the financial crisis, and macroprudential policy occupies a significant role.

into beliefs about trend inflation. Exactly how the Federal Reserve
should manage this situation is no doubt an issue Marvin would have
been all over.

Monetary policy analysis: never a dull moment

Summing up

But as Marvin cautions, the monetary policy framework remains a
work in progress. Perhaps the most important issue outstanding is that
we still have at best only a rough idea of how central banks can effectively anchor inflation expectations.12 In this regard, Japan’s inability
to escape low inflation/deflation after decades poses a challenge,
especially since the Bank of Japan introduced a Western-style inflation
targeting framework in 2013. Part of the answer surely is that since
the late 1990s, the Bank of Japan had done nothing to convince the
public that it could indeed engineer an escape from a deflation trap,
given the persistently low inflation since the 1990s.13 This could lead to
hardening of long-term inflation expectations at or below zero. Nonetheless, it remains a puzzle as to why Japan appears stuck. At the core
of the problem is an incomplete understanding of what determines
expectations of trend inflation.

Marvin’s role as a policymaker sharpened his thinking as a researcher. His active engagement in research sharpened his thinking as a policymaker. As can be seen from his example, academic scribblers are no
longer so remote from the policy process. Marvin is among the central
figures responsible for this development.

A related, though less dramatic, example involves the inability of
the central banks of the industrialized economies in the West to reach
the 2 percent inflation target during the recovery period following the
Great Recession. In the decade-long recovery, both core PCE inflation
and the five-year breakeven inflation rate hovered between a 1 and 2
percent annual rate, without ever consistently reaching the 2 percent
target. The target miss was larger for Europe. The inability to consistently reach the target over such a long period is something we still
don’t understand. Complicating matters is that inflationary pressures
have picked up considerably over the current year. Associated with this
pickup has been an increase in both the five- and 10-year breakeven
inflation rate from 1.5 percent annually on the eve of the pandemic to
currently 2.5 percent. It appears that the increase in inflation is feeding
S ee Candia, Coibin, and Gorodnichenko (2021) for evidence that the inflation expectations of US firms remain far from anchored.
13
See Gertler (2017) for an analysis of this issue.
12

196 |

| 197

Gertler

References
Bernanke, Ben, Thomas Laubach, Frederic Mishkin, and Adam Posen.
1998. Inflation Targeting: Lessons from the International Experience.
Princeton: Princeton University Press.
Calvo, Guillermo. 1983. “Staggered Prices in a Utility-Maximizing
Framework.” Journal of Monetary Economics 12, no. 3 (September): 383398.
Candia, Bernardo, Olivier Coibin, and Yuriy Gorodnichenko. 2021. “The
Inflation Expectations of U.S. Firms: Evidence from a New Survey.”
mimeo.
Chari, VV., Patrick Kehoe, and Ellen McGrattan. 2009. “New Keynesian
Models: Not Yet Useful for Policy Analysis.” American Economic Journal:
Macroeconomics 1, no. 1 (January): 242-266.
Christiano, Lawrence, Martin Eichenbaum, and Charles L. Evans. 2005.
“Nominal Rigidities and the Dynamic Effects of a Shock to Monetary
Policy.” Journal of Political Economy 113, no. 1 (February).
Clarida, Richard, Jordi Gali, and Mark Gertler. 1999. “The Science of
Monetary Policy: A New Keynesian Perspective.” Journal of Economic
Literature 34, no. 4 (December): 1661-1707.
Eggertsson, Gauti, and Michael Woodford. 2003. “The Zero Bound on
Interest Rates and Optimal Monetary Policy.” Brookings Papers on Economic Activity 1.
Erceg, Christopher, and Andrew Levin. 2003. “Imperfect Credibility and
Inflation Persistence.” Journal of Monetary Economics 50, no. 4 (May):
915-944.
Galí, Jordi. 2015. Monetary Policy, Inflation and the Business Cycle. Princeton: Princeton University Press.

198 |

Monetary Policy Consensus
Galí, Jordi, and Mark Gertler. 2007. “Macroeconomic Models for Monetary Policy Evaluation.” Journal of Economic Perspectives 21, no. 4 (Fall):
25-46.
Gertler, Mark. 2017. “Rethinking the Power of Forward Guidance:
Lessons from Japan.” Monetary and Economic Studies 35 (November):
39-58.
Goodfriend, Marvin. 1991. “Interest Rates and the Conduct of Monetary
Policy.” Carnegie Rochester Conference Series on Public Policy 34 (Spring):
7-30.
Goodfriend, Marvin. 2007. “How the World Achieved Consensus on
Monetary Policy.” Journal of Economic Perspectives 21, no. 4 (Fall): 47-68.
Goodfriend, Marvin, and Robert G. King. 1997. “The New Neoclassical
Synthesis and the Role of Monetary Policy.” In NBER Macro Annual 1997,
Volume 12, edited by Ben S. Bernanke and Julio Rotemberg, 231-296.
Cambridge: MIT Press.
Hazell, Jonathon, Juan Herreño, Emi Nakamura, and Jón Steinsson.
2022. “The Slope of the Phillips Curve: Evidence from U.S. States.” Quarterly Journal of Economics, conditionally accepted.
Kydland, Finn, and Edward C. Prescott. 1977. “Rules Rather than Discretion: The Inconsistency of Optimal Plans.” Journal of Political Economy
85, no. 3 (June): 473-492.
L ucas, Robert. 1972. “Expectations and the Neutrality of Money.” Journal of Economic Theory 4, no. 2 (April): 103-124.
Mankiw, N. Gregory, and David Romer. 1991. New Keynesian Economics.
Cambridge: MIT Press.
Rotemberg, Julio, and Michael Woodford. 1997. “An OptimizationBased Econometric Framework for the Evaluation of Monetary Policy.”
In NBER Macro Annual 1997, Volume 12, edited by Ben S. Bernanke and
Julio Rotemberg, 297-361. Cambridge: MIT Press.

| 199

Gertler
Taylor, John. 1980. “Aggregate Dynamics and Staggered Contracts.
Journal of Political Economy 88, no. 1 (February): 1-23.
Taylor, John. 1993. “Discretion versus Policy Rules in Practice.” Carnegie
Rochester Conference Series on Public Policy 39: 195-214.
Werning, Iván. 2012. “Managing a Liquidity Trap.” MIT, mimeo.
Woodford, Michael. 2003. Interest and Prices: Foundations of a Theory of
Monetary Policy. Princeton: Princeton University Press.

200 |

| 201

Great Inflation to Great Moderation

From the Great Inflation to
the Great Moderation
Robert L. Hetzel
During Marvin Goodfriend’s more than 25 years at the Richmond
Fed, US monetary policy changed from the activist go-stop policy of
the 1970s to the Volcker-Greenspan policy of creating an expectation
of price stability. Marvin both documented this radical transformation
and played a key role — through policy analysis and analytical work —
in its becoming the consensus view of academics and policymakers. In
this essay, I review the Volcker and Greenspan policy accomplishments
and Marvin’s contributions, referring to 10 of his papers in the process.
Marvin’s genius and gift for making substantive contributions lay
in his ability to draw from diverse methodological traditions. Milton
Friedman identified two broad classes of approaches to identifying
causation in economics. The Walrasian approach requires from the
beginning explicit specification of a general equilibrium model, in contemporary terms, an optimizing, dynamic, stochastic, general equilibrium model. The Marshallian approach, favored by Friedman, requires
from the beginning a search for empirical regularities and then uses
an historical narrative organized around events that have the element
of semi-controlled experiments.1 Marvin’s ability to meld these two
approaches is illustrated by the way in which he integrated theory and
historical narrative, respectively, in Goodfriend and King (1997) and
Goodfriend (2005).
While Marvin’s research used both theory and history, his approach
to both was grounded in monetarist principles, Bill Poole having been
an important influence in the Brown PhD program.
	The consensus among monetary economists that central banks are
responsible for inflation is built on both theory and evidence. Above
all, there is the substantial body of evidence from the inflationary
experiences of a great many nations, including the widespread
1

202 |

Friedman and Schwartz (1963).

| 203

Hetzel

Great Inflation to Great Moderation

	inflation in the industrialized world during the 1960s and 1970s,
showing that sustained inflation is always associated with excessive
money growth. The evidence also clearly indicates that inflation is
stopped by slowing the growth of the money supply.2

Money remains in the background for much of this essay, but I will
argue that these monetarist principles are consistent even with Marvin’s work on the New Keynesian models in which money was ostensibly absent.

	The absence of an anchor for inflation caused inflation expectations
and long bond rates to fluctuate widely.… [It] became increasingly
difficult to track the public’s inflation expectations to tell how nominal federal funds rate policy actions translated into real rate actions.5

Paul Volcker became FOMC chairman in August 1979. On October 6,
1979, the FOMC announced nonborrowed reserves procedures designed to ensure a deceleration in money growth. As Marvin wrote,
	In October 1979 it was not at all clear how quickly the Volcker Fed
could acquire credibility for low inflation, how costly a disinflation
might be, or even whether it could succeed at all, given the pressure
that would be brought to bear on the Fed as a result of the accompanying recession.6

The transformation of monetary policy
from pre- to post-Volcker
Marvin used the term “go-stop” to describe FOMC policy in the
pre-Volcker period.3 He describes it as follows:
	Inflation would rise slowly as monetary policy stimulated employment in the go phase of the policy cycle. By the time the public and
Fed became sufficiently concerned about rising inflation for monetary policy to act against it, pricing decisions had already begun to
embody higher inflation expectations. At that point, a given degree
of restraint on inflation required a more aggressive increase in shortterm interest rates, with greater risk of recession.4

The go-stop character of monetary policy was based on achieving
a socially desirable low rate of unemployment and on the consensus
that cost-push pressures drove inflation. The latter assumption meant
that achievement of price stability would incur a high cost in terms of
unemployment. With go-stop policy, inflation drifted up in the 1970s,
causing the loss of a stable nominal anchor.

Goodfriend (1997), 8.
T hrough innumerable conversations, Marvin helped the author work out his own
ideas. See Hetzel (2008, 2012, and forthcoming 2022).
4
Goodfriend (2005), 244.

Proponents of rational expectations argued that credibility for disinflation would greatly reduce the cost of disinflation. Goodfriend and
King (2005) pointed out that the disinflation was not a good test of
the proposition. Volcker could not commit to price stability because of
uncertainty over the support of the political system.
The Volcker disinflation was ultimately successful. The Volcker-Greenspan monetary policy that followed ended the instability of the 1970s
and created the Great Moderation. Marvin documented that transition,
for example, in Goodfriend (2005). In Marvin’s telling, monetarism was
central to the Volcker disinflation:
	Monetarist theory and evidence on money supply and demand,
and on the relationship between money and inflation, encouraged
the Volcker Fed to act against inflation. The successful stabilization
and eventual elimination of inflation at reasonable cost in light of
subsequent benefits, without wage and price controls, and without
supportive fiscal policy actions, vindicated the main monetarist
message…. By assembling a convincing body of theory and evidence
that controlling money was necessary and sufficient for controlling

2
3

5
6

204 |

 oodfriend (2005), 245 and 247.
G
Goodfriend (2005), 247.

| 205

Hetzel
	inflation, and that a central bank could control money, monetarists
laid the groundwork for the Volcker Fed to take responsibility for
inflation after October 1979 and bring it down.7

Pre-Volcker, the activist policy of discretionarily balancing off independent targets for low unemployment and low inflation foundered as
the inflationary expectations of the public rose and offset the stimulative effects of monetary expansion, leaving only higher inflation with
no benefit in terms of lower unemployment. Marvin wrote,
	Over time, deliberately expansionary monetary policy in the ‘go’
phase of the policy cycle came to be anticipated by workers and
firms. Workers learned to take advantage of tight labor markets to
make higher wage demands, and firms took advantage of tight product markets to pass along higher costs in higher prices. Increasingly
aggressive wage- and price-setting behavior tended to neutralize the
favorable employment effects of expansionary policy.8

This process led to expected and actual inflation becoming unanchored.
Reanchoring inflationary expectations required abandoning the
prior activist policy with its characteristic of expansionary monetary
policy in recoveries and contractionary monetary policy in response to
the resulting inflation. As made evident in the preemptive increases in
the funds rate in economic recoveries intended to prevent the emergence of inflation and to forestall an increase in expected inflation,
policy maintained a neutral character. Marvin summarized:
	For the Volcker Fed … its room to maneuver between fighting inflation and fighting recession disappeared. In effect, the Fed lost the
leeway to choose between stimulating employment in the go phase
of the policy cycle and fighting inflation in the stop phase.9
 oodfriend (2005), 243 and 246.
G
Goodfriend (2005), 6-7.
9
Goodfriend (2005), 247.

Great Inflation to Great Moderation
The discipline imposed on monetary policy in the Volcker-Greenspan era came from the desire to short circuit the process whereby expected inflation would rise with monetary stimulus and rising inflation.
Stabilizing expected inflation in a way consistent with a return to price
stability required preemptive increases in the funds rate. Once a policy
of price stability became fully credible starting in 1995, this meant
raising interest rates vigorously in economic recovery when signs of
tightness emerged in labor and product markets. “The Fed has learned
to adjust interest rates more preemptively since October 1979 … and
inflationary go-stop policy cycles are no more.”10

The Great Moderation through the lens of the NK model
The shift in the monetary regime to focusing on price stability from
discretionarily trading off between unemployment and inflation
changed the intellectual consensus. The profession became receptive
to the development of the New Keynesian (NK) dynamic stochastic
general equilibrium (DSGE) model. Goodfriend and King (1997) used
the term “New Neoclassical Synthesis” to characterize the model and to
emphasize the sharp break with the prior class of Keynesian models.11

Optimal policy in the NK model
A central property of the basic NK model, as exposited by Goodfriend and King (1997), is that optimal monetary policy is neutral
(see also King and Wolman, 1999). That is, it is nonactivist in that the
optimal policy does not move between expansionary and contractionary monetary policy. Instead, policy remains focused on price stability
and turns over the behavior of the real economy to the real business
cycle core of the economy. In the basic NK model, a credible policy that
stabilizes the price level keeps output at potential, even for shocks to
aggregate supply.

7
8

10
11

206 |

 oodfriend (2005), 256.
G
Marvin’s 1997 paper with Bob King was a key landmark in the development of this
class of models; that paper is the subject of Michael Woodford’s essay in this volume.

| 207

Hetzel
Blanchard and Gali (2007) refer to the simultaneous occurrence of
price stability and a zero output gap in the basic NK model as “divine
coincidence.” They treat markup shocks as classic supply shocks requiring a trade-off between inflation and employment. Price stability is
nonoptimal in that it requires increases in unemployment in the event
of supply shocks. In the spirit of the Goodfriend-King (1997) version
of the NK model, however, the central bank can let the shocks pass
through to the price level and over time they average out as noise.
“If one argues that some costs flow directly to prices in a perfectly
competitive sector, then theory suggests that the central bank should
consider stabilizing only a ‘core’ index of monopolistically competitive
sticky prices.”12 In Goodfriend (2005, p. 254) Marvin lists modifications
to the basic model that can produce a short-run trade-off between
inflation and unemployment and explains why they are unlikely to be
important in practice.
Because the NK Phillips curve makes current inflation depend upon
expected future inflation and the markup (the excess of price over
marginal cost), a policy of price stability entails stabilization of firms’
markup.13 The markup is a latent (nonobservable) variable. However, a
rule that maintains the output gap equal to zero through price stability
is equivalent to a rule that maintains actual and expected real rates of
interest equal to their natural counterparts. Practical implementation
of a rule that provides for price stability then requires a reaction function that provides for a stable nominal anchor and that causes the real
funds rate to track the natural rate of interest.

Great Inflation to Great Moderation
potential as evidenced by a sustained increase in the rate of resource
utilization (a declining unemployment rate), the FOMC raised the
funds rate in measured increments. It then watched bond markets to
ascertain that markets believed the FOMC would raise rates over time
sufficiently to maintain price stability.15 A converse statement holds for
weakness.
The NK model provides the economic intuition for LAW procedures
in that above-trend output growth is associated with optimism about
the future and below-trend growth with pessimism about the future.
Optimism is associated with a relatively high natural rate of interest
while pessimism is associated with a relatively low natural rate of
interest, as Marvin discusses, for example, in Goodfriend (2004).16 LAW
procedures are then the foundation for tracking the natural rate of
interest. The central modification to LAW that differentiated go-stop
monetary policy from the successor policy directed at the reestablishment of price stability was preemptive increases in the funds rate
intended to prevent the revival of inflation. In terms of the NK model,
such preemptive tightening allows the FOMC to track the natural rate
of interest.17
Preemption began with the FOMC’s response to inflation scares.18
Marvin wrote:
	The successful containment of the 1983-84 inflation scare was the
most remarkable feature of the Volcker disinflation. The Fed had
succeeded in reducing inflation temporarily in many preceding gostop policy cycles. Preemptive interest rate policy actions in 1983-84
finally put an end to inflationary go-stop policy. This success was particularly important for the future because it showed that well-timed,
aggressive interest rate policy actions could defuse an inflation scare
and preempt rising inflation without creating a recession.19

How policy worked in practice
In the Great Moderation of the Volcker-Greenspan era, the FOMC
did not literally implement a rule that directly targeted the price level.
Since the chairmanship of William McChesney Martin, the FOMC’s reaction function has always worked on the lean-against-the-wind (LAW)
principle.14 Specifically, when the economy was growing above

 oodfriend (1991).
G
Goodfriend (2002), 33-34, 37-38
17
A simple feedback rule with which the Fed changes its policy instrument in response
to misses of the price level from target would run afoul of the Friedman (1960, 87-88)
critique of long and variable lags, which he specifically applied to a price level target.
18
Goodfriend (1993).
19
Goodfriend (2005), 249.
15
16

 oodfriend (2005), 255. See also Aoki (2003).
G
See Goodfriend (2002), 36.
14
Hetzel (2022).
12
13

208 |

| 209

Hetzel

Great Inflation to Great Moderation

Preemptive increases in the funds rate do not limit sustainable
growth in employment. Marvin also highlighted the preemptive tightening in 1994.
T he economic expansion gathered strength in late 1993. The zero
real federal funds rate was no longer needed and would become inflationary if left in place. The Fed began to raise the federal funds rate
in February 1994, taking it in seven steps from 3 percent to 6 percent
by February 1995. Inflation showed little tendency to accelerate and
remained between 2.5 percent and 3 percent. Thus, the Fed’s policy
actions took the real federal funds rate from zero to a little more than
3 percent. The move raised real short-term interest rates to a range
that could be considered neutral to mildly restrictive. In spite of the
policy tightening, real GDP grew by 4 percent in 1994, up from 2.6
percent in 1993, and the unemployment rate fell from 6.6 percent to
5.6 percent from January to December 1994. ...
	The 1994 tightening demonstrated that a well-timed preemptive increase in real short-term interest rates is nothing to be feared. In this
case, it was needed to slow the growth of aggregate demand relative
to aggregate supply to avert a buildup of inflationary pressures. By
holding the line on inflation in 1994, preemptive policy actions laid
the foundation for the boom that followed.20

This discipline imposed on underlying LAW procedures contrasted
with the discretion that characterized the go-stop policy of the 1970s.
“[D]iscretion leads inexorably to go-stop policy that brings rising and
unstable inflation and inflation expectations, with adverse consequences for interest rates and employment.” 21
However, there is an issue of political economy. Preemptive increases
in the funds rate inevitably arouse populist criticism charging that the
FOMC is increasing unemployment to fight a nonexistent inflation.
20
21

 oodfriend (2002), 5.
G
Goodfriend (1997), 17.

	By successfully keeping inflation in check, preemptive policy actions
necessarily appear to be busting ghosts. So the appearance of ghost
busting is a consequence of good monetary policy.22

What about money?
One notable aspect of NK models is that they do not contain money.
Is there then a contradiction between Marvin’s willingness to use NK
models for policy analysis and his view that monetarism was central to
the Volcker disinflation? No: under optimal policy in the basic NK model of Goodfriend and King (1997), money is a veil. It need not appear
in the model.23 The reason is that the optimal monetary rule ensures
monetary control. “Under a neutral policy, the monetary authority
accommodates variations in money demand to insure that excesses or
shortages of money do not create aggregate demand disturbance.”24
With the output gap equal to zero, there is no excess demand or
supply in the goods market to spill over and create a corresponding
excess supply or demand in the bond market. With no excess supply or
demand in the bond market, there are no excesses or deficiencies for
the Fed to monetize or demonetize as a consequence of defending its
interest rate target, thereby creating destabilizing changes in money.
Contrary to the expectations of monetarists, the required monetary
control in the Volcker-Greenspan era did not occur through adoption
by the Fed of a reserves aggregate as a target for achieving substantive money targets. The control occurred indirectly through a rule that
provided for a stable nominal anchor through commitment to maintenance of an expectation of nominal stability, ultimately price stability.
With that nominal expectational stability, the Fed could implement
operating procedures that caused the real funds rate to track the
natural rate of interest, thereby turning over the determination of real
variables such as output and employment to the unfettered operation
of the price system. In the 1970s, the monetary regime entailed
 oodfriend (1997), 17, italics in original.
G
Marvin does say, “[T]he Fed should have a contingency plan for returning to monetary targeting in the event that high and volatile inflation and inflation expectations
cause trouble again.” (Goodfriend, 2005, 257).
24
Goodfriend and King (1997), 267.
22
23

210 |

| 211

Hetzel
discretionarily trading off between objectives of low unemployment
and low inflation presumed driven by cost-push forces. This activist
policy created destabilizing fluctuations in money. The neutral policy
pursued in the Volcker-Greenspan era disciplined money creation so
that it did not become a source of instability.

Conclusion
I close with a passage from Goodfriend (1997), which seems especially relevant today, in light of the FOMC’s movement away from
preemptive funds rate increases. Marvin wrote,
	The Fed has acquired credibility since the early 1980s by consistently
taking policy actions to hold inflation in check. Experience shows
that the guiding principle for monetary policy is to preempt rising
inflation. The go-stop policy experience teaches that waiting until
the public acknowledges rising inflation to be a problem is to wait
too long. At that point, the higher inflation becomes entrenched
and must be counteracted by corrective policy actions more likely to
depress economic activity.25

It is indeed a tragedy that Marvin is no longer here to help us learn
from the “Odyssey” of monetary policy not only in the 20th century,
but also the 21st century. But we do have the gift of his body of work,
whose continued relevance for thinking about monetary policy I have
tried to convey in this essay.

Great Inflation to Great Moderation

References
Aoki, Kosuke. 2001. “Optimal Monetary Policy Responses to Relative-Price Changes.” Journal of Monetary Economics 48, no. 1 (August):
55-80.
Blanchard, Olivier, and Jordi Gali. 2007. “Real Wage Rigidities and the
New Keynesian Model.” Journal of Money, Credit, and Banking 39 (February): 35-65.
Friedman, Milton. 1960. A Program for Monetary Stability. New York:
Fordham University Press.
Friedman, Milton, and Anna J. Schwartz. 1963. A Monetary History of the
United States, 1867-1960. Princeton: Princeton University Press.
Goodfriend, Marvin. 1991. “Interest Rates and the Conduct of Monetary
Policy.” Carnegie-Rochester Conference Series on Public Policy 34 (Spring):
7-30.
Goodfriend, Marvin. 1993. “Interest Rate Policy and the Inflation Scare
Problem.” Federal Reserve Bank of Richmond Economic Quarterly 79,
no. 1 (Winter): 1-24.
Goodfriend, Marvin. 1997. “Monetary Policy Comes of Age: A 20th Century Odyssey.” Federal Reserve Bank of Richmond Economic Quarterly
83, no. 1 (Winter): 1-22.
Goodfriend, Marvin. 2002. “The Phases of U.S. Monetary Policy: 1987 to
2001.” Federal Reserve Bank of Richmond Economic Quarterly 88, no. 4
(Fall): 1-16.
Goodfriend, Marvin. 2004. “Monetary Policy in the New Neoclassical
Synthesis: A Primer.” Federal Reserve Bank of Richmond Economic Quarterly 90, no. 3 (Summer): 21-45, reprinted from International Finance,
2002, 5, 165-92.

25

212 |

Goodfriend (1997), 14.

Goodfriend, Marvin. 2005. “The Monetary Policy Debate since October
1979: Lessons for Theory and Practice.” Paper prepared for “Reflections
on Monetary Policy: 25 Years after October 1979,” Federal Reserve Bank
of St. Louis Review 87, no. 2 (March/April): 243-62.

| 213

Hetzel
Goodfriend, Marvin. 2013. “The Great Inflation Drift.” In The Great
Inflation: The Rebirth of Modern Central Banking, edited by Michael D.
Bordo and Athanasios Orphanides, 181-215. Chicago: The University of
Chicago Press.
Goodfriend, Marvin, and Robert G. King. 1997. “The New Neoclassical
Synthesis.” In NBER Macroeconomics Annual 1997, Volume 12, edited by
Ben S. Bernanke and Julio Rotemberg, 231-296. Cambridge: MIT Press.
Goodfriend, Marvin, and Robert G. King. 2001. “The Case for Price Stability.” In Why Price Stability? Proceedings of the First ECB Central Banking
Conference, edited by A. Garcia-Herrero, V. Gaspar, L. Hoogduin, J. Morgan, and B. Winkler, 53-94. Frankfurt, Germany: European Central Bank;
NBER Working Paper 8423, National Bureau of Economic Research,
August.
Goodfriend, Marvin, and Robert G. King. 2005. “The Incredible Volcker
Disinflation.” Journal of Monetary Economics 52, no. 5 (July): 981-1015.
Hetzel, Robert L. 2008. The Monetary Policy of the Federal Reserve: A
History. Cambridge: Cambridge University Press, 2008.
Hetzel, Robert L. 2012. The Great Recession: Market Failure or Policy Failure? Cambridge: Cambridge University Press, 2012.
Hetzel, Robert L. 2022. The Federal Reserve System. A New History. University of Chicago Press, forthcoming.
King, Robert G., and Alexander L. Wolman. 1999. “What Should the
Monetary Authority Do When Prices Are Sticky?” In Monetary Policy
Rules, edited by John B. Taylor, 349-404. Chicago: University of Chicago
Press.

214 |

| 215

Taming Inflation Scares

Taming Inflation Scares
Athanasios Orphanides and John C. Williams
The beginning of the 1990s marked a singular moment when academic researchers and policymakers critically examined the most basic
questions of monetary policy for the post-Bretton Woods world. Researchers analyzed and debated the choices of policy instrument and
goals, central bank independence and transparency, and the relative
merits of rules vs. discretion. John Taylor’s seminal 1993 paper “Discretion versus Policy Rules in Practice” exemplified this era.
At the Federal Reserve, Chair Greenspan acknowledged the unfinished work of anchoring sustained low inflation, saying “It is an open
question whether we have learned enough to skirt the dangers of
budgetary and monetary excess that have triggered past episodes of
debilitating inflation.”1 Likewise, central banks around the world were
redesigning institutional frameworks with the aim of reestablishing a
nominal anchor. These included the European Monetary System, which
was under severe stress at the time, and the adoption of inflation targeting, starting with the Reserve Bank of New Zealand.
It was in this context that Marvin Goodfriend’s essay “Interest Rate
Policy and the Inflation Scare Problem: 1979-1992” entered the discussion, synthesizing strands of theory and practice with the goal “to
distill observations to guide future analysis of monetary policy with
the ultimate objective of improving macroeconomic performance.”2
Our essay connects Goodfriend’s important and timely paper to the academic and policy debates of the period. It then traces its influence on
subsequent monetary policy research and the evolution of the Federal
Reserve’s monetary policy strategy and communication.
Goodfriend focused on the interplay of interest rate policy and “inflation scares,” which he defined to be adverse movements in long-term
1
2

216 |

 reenspan (1993), p. 5.
G
Goodfriend (1993), p. 2.

| 217

Orphanides and Williams
inflation expectations inconsistent with the Federal Reserve’s inflation
goal. He highlighted a critical dilemma that arises when the public and,
in particular, financial markets doubt the central bank’s commitment
to achieve price stability. He argued that resisting inflation scares is
costly because it requires a tightening of monetary policy that would
not otherwise be warranted by economic fundamentals. On the other
hand, failing to effectively address an inflation scare engendered the
risk of persistent, undesirably high inflation. He illustrated this dilemma with specific examples from the US disinflationary period of 19791992.
Goodfriend drew a number of profound conceptual and policy
implications from his analysis of inflation scares. At the time, these
were novel and not entirely uncontroversial. But they have stood the
test of time, becoming part of the canon of central banks’ approaches
to monetary policy. First and foremost, he stressed the importance of
formulating a strategy that clearly communicates the commitment to
price stability and anchors inflation expectations at the desired level.3
He argued that such a strategy affords the central bank greater flexibility to respond forcefully to recessionary shocks, thereby improving
economic stability, without risking an adverse shift in longer-term
inflation expectations. Second, he identified the central role inflation
expectations play in the conduct of monetary policy and highlighted,
by example, the value of measuring, monitoring, and analyzing inflation expectations. Third, building on his earlier work in Goodfriend
(1991), he emphasized the primacy of the short-term interest rate as
the instrument of monetary policy, at a time when many economic
models assumed a measure of money supply was the policy instrument. Finally, he used the example of inflation scares to illustrate the
need to recognize the role of imperfections in information and credibility for the formulation of a successful monetary policy strategy. His
arguments highlighted that establishing and maintaining an enduring
nominal anchor is at the heart of such an effort.
3

218 |

T hese themes were developed further by Goodfriend in collaboration with Robert
King (Goodfriend and King 1997), and in his work advocating for central bank transparency regarding its inflation goal and strategy (Goodfriend 2004).

Taming Inflation Scares

The policy context
Monetary policy strategy in the United States has undergone a sea
change over the past four decades.4 By the late 1970s, Federal Reserve
policy had inadvertently contributed to macroeconomic instability
and the central bank’s credibility was in question. Starting with Fed
Chairman Paul Volcker’s monetary reform in 1979, the subsequent long
journey was marked by a series of dramatic changes and subtle refinements. The immediate challenge that Volcker faced was to reestablish
the Federal Reserve’s credibility and restore low and stable inflation. By
the mid-1980s, with the credibility of Federal Reserve’s commitment
to price stability improved, the policy debate could once again return
to the overarching monetary policy challenge: What institutional
framework and monetary policy strategy can best enhance economic
stability, promote high growth and employment, and deliver a stable
nominal anchor and price stability?
A related question was whether a clearer formal mandate for price
stability would be necessary to assure improvement and defend
against the risk of policy backsliding to the unfortunate experience of
the 1970s. By the end of the 1980s, the evidence and practical experience worldwide supported the benefits of an independent central
bank with a clear mandate to preserve price stability. However, there
was then no clear consensus on the relative merits and costs of changing central bank mandates in that direction, or of the necessity of such
changes in legislation for improving policy practice.
In the United States, the zero-inflation resolution, proposed in 1989,
represented one specific legislative effort toward that end. Although
Chair Greenspan supported the legislation, it was not enacted.5 Opponents of the legislation argued that greater emphasis on price stability
would come at the cost of employment and growth. By contrast,
4
5

 eltzer (2009), Williams (2015), and Orphanides (2020).
M
In his testimony, Chairman Greenspan (1989) welcomed the clarity of the legislation in supporting the Federal Reserve’s disinflation effort and noted that the mere
adoption of the legislation would be helpful in reducing inflation expectations and
achieving price stability, but he also cautioned that eliminating inflation too quickly
would entail a cost in the form of suppressed growth.

| 219

Orphanides and Williams
Goodfriend argued that a more explicit mandate for price stability
would enhance economic stability:
	The preceding observation suggests an attractive argument in favor
of a congressional mandate for price stability. By reducing the risk
of inflation scares, such a mandate would free the funds rate to
react more aggressively to unemployment in the short run. Thus, a
mandate for price stability would not only help eliminate inefficiencies associated with long-run inflation, it would add flexibility to the
funds rate that might improve countercyclical stabilization policy as
well.6

At the center of the policy debate was an increasing appreciation
of the key role of the interplay between monetary policy and inflation
expectations in shaping the effectiveness of monetary policy to counteract economic shocks and secure desirable economic outcomes.
While the zero-inflation legislative initiative failed in the United
States, legislation stipulating monetary policy mandates was enacted
in other countries that had experienced high inflation in the 1970s and
1980s and where disinflation efforts were less successful than in the
United States. The most prominent example proved to be the case of
New Zealand, which originated the inflation targeting framework. The
Reserve Bank of New Zealand Act 1989, which became law in 1990,
instructed the central bank to focus on a single objective, an inflation
goal, and enhanced the central bank’s operational independence to
achieve this goal.
The adoption and clear communication of an explicit numerical
inflation target, the distinguishing characteristics of the inflation
targeting approach, were subsequently widely adopted, including by
the Federal Reserve. At the time Goodfriend wrote the inflation scares
article, however, the inflation targeting framework was in still in its
6

220 |

Goodfriend (1993), p. 17.

Taming Inflation Scares
infancy.7 Opacity in goals and, in the case of the Federal Reserve, even
in policy instruments was common.8 Inflation targeting gradually came
to be seen as the preferred way to communicate the quantitative definition of price stability that Goodfriend’s analysis called for.
Over time, the Federal Reserve came to recognize that a policy strategy that anchored inflation expectations in line with a clearly communicated definition of price stability enhanced the overall effectiveness
of monetary policy, and furthermore, that this could be achieved with
appropriate interpretation of the Federal Reserve’s mandate within the
existing institutional framework.
On January 25, 2012, the Federal Open Market Committee (FOMC)
formally adopted a 2 percent inflation goal as its definition of price
stability:
	Communicating this inflation goal clearly to the public helps keep
longer-term inflation expectations firmly anchored, thereby fostering
price stability and moderate long-term interest rates and enhancing
the Committee's ability to promote maximum employment in the
face of significant economic disturbances.9
T he article was published in early 1993, first in the Annual Report of the Federal
Reserve Bank of Richmond for 1992 and then in the Bank’s Economic Quarterly.
Goodfriend had completed a draft by the summer of 1992. In between, he had the
opportunity to present the work at the Bank of England and the Riksbank, two central
banks that adopted inflation targeting in response to the ERM crisis that erupted in
the fall of 1992. Goodfriend’s arguments in support of a “mandate for price stability”
and of the short-term interest rate as the instrument of monetary policy were pertinent to the inflation targeting debate.
8
As discussed in the essay by Lars E.O. Svensson elsewhere in this volume, in his
influential article on “Monetary Mystique,” Goodfriend (1986) argued that the Fed’s
preference for secrecy, as practiced in the mid-1980s, was difficult to justify. The Fed
subsequently became more transparent, but this was a gradual process. While the
Federal Reserve calibrated policy with a target federal funds rate, the timing of policy
changes was not immediately disclosed until February 1994, and the actual setting
of the policy instrument — the target federal funds rate — was only communicated
starting in July 1995. Lindsey (2003) presents a history of the evolution of FOMC communication that spans this period.
9
Federal Reserve (2012).
7

| 221

Orphanides and Williams
In 2020, the FOMC stated, “The Committee judges that longer-term
inflation expectations that are well anchored at 2 percent foster price
stability and moderate long-term interest rates and enhance the Committee’s ability to promote maximum employment in the face of significant economic disturbances” (Federal Reserve, 2020). The language in
these statements echoes the arguments of Goodfriend’s article.

Advances in models for monetary policy analysis
Today, the merits of a monetary policy strategy based on the communication of a clear goal for inflation as the central bank’s definition
of price stability and aiming to solidly anchor inflation expectations in
line with this goal are uncontroversial.10 At the time of the writing of
Goodfriend’s article, however, economic theory had not yet provided
clear support for moving policy in this direction. During the 1980s, the
field of macroeconomics was disjointed regarding monetary policy.
On the one hand, it was recognized that available models employed
for policy analysis by central banks did not adequately capture the
endogeneity of inflation expectations to monetary policy strategy and
its communication. On the other hand, while the rational expectations
revolution had highlighted the critical importance of policy regimes
for shaping expectations, the prevalent assumption of full information and perfect knowledge made available models too simplistic to
overcome reasonable policymaker doubts about the resulting policy
advice.
Assuming rational expectations with full information and perfect
knowledge, as was common in monetary policy models at the time of
Goodfriend’s article, created an apparent disconnect between theory
and practice. Such models typically assumed that inflation expectations were well-anchored, effectively ruling out the inflation scares
documented by Goodfriend. On the other hand, policy practitioners
10

222 |

 ith the recent adoption of a 2 percent goal by the European Central Bank, all major
W
advanced economy central banks now share this element in their policy strategy.
In its policy strategy statement, the ECB highlighted the role of a clear goal for
anchoring inflation expectations: “The two per cent inflation target provides a clear
anchor for inflation expectations, which is essential for maintaining price stability”
(ECB 2021).

Taming Inflation Scares
learned to pay close attention to private inflation expectations, as
these could be inferred, albeit imperfectly, from surveys and financial
market data. Indeed, the increasing appreciation of the central role of
credibility for the effectiveness of monetary policy in recent decades
has resulted in more resources allocated by central banks to improving
the measurement of inflation expectations11 as well as support for the
development of related financial markets.12 It also prompted the development of models that would close the apparent gap between theory
and practice.
One strand of this literature introduced imperfect credibility by
positing that the central bank’s implicit inflation goal shifted over time
and tracing the evolution of beliefs about this goal. Along these lines,
Kozicki and Tinsley (2001) demonstrated that “shifting endpoints”
modeled in this fashion improved our understanding of expectations
embedded in the term structure of interest rates. Explicit introduction
of learning about policy and other aspects of the economy improved
the ability of canonical models to explain inflation persistence and
business cycle dynamics.13 Another strand focused on the policy implications of departures from rational expectations with the introduction
of a process of perpetual learning by private agents as a mechanism
governing the formation of expectations.14 These models show that
modest deviations from the assumption of rational expectations with
perfect knowledge introduce a layer of complexity in inflation dynamics that can give rise to the type of inflation scares envisaged by Goodfriend.15 As a result, policies that would appear to be efficient under
S uch efforts include the development of new surveys as well as models that combine information from surveys and financial markets, such as indexed debt, to arrive
at estimates of inflation expectations. (See, e.g., Armantier et al 2017, D’Amico, Kim
and Wei 2018, and Ahn and Fulton 2020.)
12
The Federal Reserve supported the Treasury’s development of “inflation-indexed”
debt in the mid-1990s. In his article, Goodfriend cited a proposal by Hetzel (1992)
about how the improved measures of long-term inflation expectations that could
be derived once indexed debt became available would assist the Fed in setting
monetary policy.
13
Erceg and Levin (2003), Milani (2007), Orphanides and Wei (2012), and Slobodyan
and Wouters (2012).
14
Orphanides and Williams (2004), and Gaspar, Smets, and Vestin (2010).
15
Orphanides and Williams (2005).

11

| 223

Orphanides and Williams
rational expectations can instead yield poor results when knowledge is
imperfect. In these models, consistent with Goodfriend’s policy recommendations, better outcomes can be achieved with policies that reflect
a greater commitment to price stability.

The continuing importance of taming inflation scares
We conclude with one key message from Goodfriend’s analysis
of inflation scares relevant for monetary policy strategy: systematic
anchoring of inflation expectations can improve the achievement of
both price stability and economic activity goals. In contrast, the Fed’s
imperfect credibility during the disinflationary period of 1979-1992
led to bouts of inflation scares that increased the economic costs of
restoring price stability. Goodfriend argued that when its commitment
to price stability is credible, the Fed enjoys “remarkable latitude” (p. 17)
for easing monetary policy in response to a recessionary shock without triggering an inflation scare. A policy strategy that keeps inflation
expectations solidly anchored in line with a clearly communicated
definition of price stability supports both growth and employment.
Providing an enduring nominal anchor succeeds in taming inflation
scares. The Federal Reserve’s monetary policy strategy and communication has evolved since Goodfriend highlighted the inflation scare
problem, heeding these important lessons.
Marvin Goodriend’s article on inflation scares was characteristic of
his approach to policy research “with the ultimate objective of improving macroeconomic performance.” His eagerness to explore new ideas,
debate, listen, and debate some more, and his openness to questioning central bank orthodoxy, were inspiring for younger economists
like us who had the good fortune to interact with him. His passion for
principled, research-based policy in the quest for improving the contribution of an independent central bank to society is part of his lasting
legacy.

Taming Inflation Scares

References
Ahn, Hie Joo, and Chad Fulton. 2020. “Index of Common Inflation Expectations.” FEDS Notes, September 2.
Armantier, Olivier, Giorgio Topa, Wilbert van der Klaauw, and Basit
Zafar. 2017. “An Overview of the Survey of Consumer Expectations.”
Federal Reserve Bank of New York Economic Policy Review 23, no. 2
(December): 51-72.
D’Amico, Stefania, Don Kim, and Min Wei. 2018. “Tips from TIPS: The
Informational Content of Treasury Inflation-Protected Security Prices.”
Journal of Financial and Quantitative Analysis 53, no. 1 (February).
European Central Bank. 2021. “The ECB’s Monetary Policy Strategy
Statement.” July 8.
Erceg, Christopher, and Andrew Levin. 2003. “Imperfect Credibility and
Inflation Persistence.” Journal of Monetary Economics 50, no. 4 (May):
915-944.
Federal Open Market Committee. 2012. “Statement on Longer-Run
Goals and Monetary Policy Strategy, Adopted effective January 24,
2012.”
Federal Open Market Committee. 2020. “Statement on Longer-Run
Goals and Monetary Policy Strategy, Adopted effective January 24,
2012; as amended effective August 27, 2020.”
Gaspar, Vitor, Frank Smets, and David Vestin. 2010. “Inflation Expectations, Adaptive Learning and Optimal Monetary Policy.” In Handbook of
Monetary Economics, edited by B. Friedman and M. Woodford, Chapter
19. North-Holland: Elsevier.
Goodfriend, Marvin. 1986. “Monetary Mystique: Secrecy and Central
Banking.” Journal of Monetary Economics 17, no. 1 (January): 63-92.
Goodfriend, Marvin. 1991. “Interest Rates and the Conduct of Monetary
Policy.” Carnegie-Rochester Conference Series on Public Policy 34 (Spring):
7–30.

224 |

| 225

Orphanides and Williams
Goodfriend, Marvin. 1993. “Interest Rate Policy and the Inflation Scare
Problem: 1979-1992.” Federal Reserve Bank of Richmond Economic
Quarterly 79, no. 1 (Winter): 1-23.
Goodfriend, Marvin. 2004. “Inflation Targeting in the United States?” In
The Inflation-Targeting Debate, edited by B. Bernanke and M. Woodford,
311-352. Chicago: The University of Chicago Press.
Goodfriend, Marvin, and Robert G. King. 1997. “The New Neoclassical
Synthesis and the Role of Monetary Policy.” In NBER Macroeconomics
Annual 1997, Volume 12, edited by Ben S. Bernanke and Julio J. Rotemberg, 231-283. Chicago: University of Chicago Press.
Greenspan, Alan. 1993. “Remarks.” Remarks at the Economic Club of
New York, New York, NY, April 19.
Hetzel, Robert L. 1992. “Indexed Bonds as an Aid to Monetary Policy.”
Federal Reserve Bank of Richmond Economic Review 78 (January/February): 13–23.
Kozicki, Sharon, and Peter Tinsley. 2001. “Shifting Endpoints in the Term
Structure of Interest Rates.” Journal of Monetary Economics 47, no. 3
(June): 613-652.

Taming Inflation Scares
University of Chicago Press.
Orphanides, Athanasios, and John C. Williams. 2005. “Inflation Scares
and Forecast-Based Monetary Policy.” Review of Economic Dynamics 8,
no. 2 (April): 498-527.
Orphanides, Athanasios, and Min Wei. 2012. “Evolving Macroeconomic
Perceptions and the Term Structure of Interest Rates.” Journal of Economic Dynamics and Control 36, no. 2 (February): 239-254.
Slobodyan, Sergey, and Raf Wouters. 2012. “Learning in a Medium-Scale DSGE Model with Expectations Based on Small Forecasting
Models.” American Economic Journal: Macroeconomics 4, no. 2 (April):
65-101.
Taylor, John. 1993. “Discretion versus Policy Rules in Practice.” Carnegie-Rochester Conference Series on Public Policy 39 (December): 195-214.
Williams, John C. 2015. “Monetary Policy and the Independence Dilemma.” Federal Reserve Bank of San Francisco Economic Letter 2015-15,
May 11.

Lindsey, David. 2003. “A Modern History of FOMC Communication:
1975-2002.” FOMC memorandum, June 24.
Meltzer, Allan H. 2009. A History of the Federal Reserve: Volumes 1 and 2.
Chicago: The University of Chicago Press.
Milani, Fabio. 2007. “Expectations, Learning and Macroeconomic Persistence.” Journal of Monetary Economics 54, no. 4 (October), 2065-2082.
Orphanides, Athanasios. 2020. “Monetary Policy Strategy and its Communication.” Paper prepared for the Federal Reserve Bank of Kansas
City Symposium on “Challenges for Monetary Policy,” Jackson Hole,
Wyoming, August 22-24, 2019.
Orphanides, Athanasios, and John C. Williams. 2004. “Imperfect Knowledge, Inflation Expectations, and Monetary Policy.” In The Inflation-Targeting Debate, edited by B. Bernanke and M. Woodford. Chicago:
226 |

| 227

Federal Reserve Independence

Federal Reserve Independence:
Is it Time for a New Treasury-Fed
Accord?
Charles Plosser*
In March 1951, after a long, and at times acrimonious, debate, the US
Treasury and the Federal Reserve reached an agreement that allowed
the central bank to end nearly a decade of pegging the interest rate
on government debt.1 The country was facing uncomfortably high
inflation following World War II and the Fed was frustrated by the fiscal
demands of the Treasury that, in its view, rendered it unable to ensure
price stability. The Treasury-Fed Accord of 1951 was an institutional
arrangement, not a legal agreement, that established an understanding of how both parties would conduct policy, and it was an important
milestone in the transformation of the Fed into an independent central
bank. As described by Allan Meltzer (2003, p. 738), it “prevented an
administration from deciding unilaterally to use monetary expansion
to gain temporary political advantage or to finance too much of the
budget at the central bank.” Its goal was to permit the Fed to control its
own balance sheet rather than have it be controlled by the Treasury for
the purposes of debt management.2 Following the agreement, the Fed
reduced the growth of bank reserves, allowing interest rates to rise to
slow inflation.
* The comments and suggestions of Michael Bordo, Robert Hetzel, Jeffrey Lacker,
Mickey Levy, Bill Nelson, and especially Robert King and Alex Wolman are greatly
appreciated.
1
See Hetzel and Leach (2001) and Meltzer (2003, pp. 699-724) for further insights
surrounding the creation and details of the 1951 Accord.
2
Allan Sproul, then president of the Federal Reserve Bank of New York, was asked
if the Fed’s efforts to break the link to Treasury funding demands might usurp the
debt-management responsibilities of the Treasury. He responded: “Certainly not. The
essence of debt-management is to tailor your offerings to the market in terms of
current economic conditions, not to have the market tailored to your offerings by the
central bank.” See Board of Governors (1951).

228 |

| 229

Plosser
The Accord reached its 50th anniversary in March 2001. At the time,
it seemed that central bank independence with respect to monetary
policy was increasingly secure and price stability was widely accepted as the primary objective of the Fed and many other central banks.
Concerns about entanglements of monetary and fiscal policy were of
more historical interest than pressing issues, at least in the US. But by
the 70th anniversary in March 2021, following massive central bank interventions in the recession of 2008-09 and again beginning in March
2020, the traditional boundaries between monetary and fiscal policy
had blurred.
In this essay, I discuss some of the important changes in the use of
the Fed’s balance sheet and ask if the 1951 Accord remains a sufficient
framework to ensure the Fed’s independence. I conclude that it does
not. Through its expansion of credit policies, the Fed has effectively
engaged in fiscal policy actions that more appropriately belong to
Congress. Congress, as well as the Fed, have taken actions that violate
at least the spirit of the 1951 Accord. Taken together, these actions
undermine the independence of monetary policy decision-making by
the Fed and open the door to political and fiscal abuse of the central
bank’s balance sheet. Thus, it is important to strengthen Fed independence through the appropriate assignment of decision-rights and
accountability required of the institution in a democratic society. Later
in this essay I will lay out a concrete proposal aimed at fostering these
goals.

Some background
I began publicly speaking and writing on these matters during my
time at the Federal Reserve (2006-15), as the Fed began its first round
of quantitative easing (QE).3 But like any other research, the ideas I
express here have antecedents in the work of others and, in particular,
that of Marvin Goodfriend. Marvin spent most of his career as a mone3

230 |

For early examples, see Plosser (2009a, 2009b, 2009c).

Federal Reserve Independence
tary economist working at the Federal Reserve Bank of Richmond.4
He served as the Bank’s research director and chief policy advisor
during the tenure of President Al Broaddus. Marvin had deep knowledge and experience in policymaking as well as economic theory. His
forward-looking research often identified important issues before they
rose to the forefront of policy debates.
Goodfriend and King (1988) and Goodfriend (1994), for example,
stress the importance of distinguishing two elements of central bank
policy. The first is monetary policy, which is reflected in changes in
the overall size of the Fed’s balance sheet. Such actions are frequently
framed in terms of interest rate policies intended to bring about desired changes in the balance sheet. The second is credit policy, which
is captured by changes in the composition of assets held.5 This decomposition is consistent with a long tradition in monetary economics that
views central banks as unique because they alone can directly alter
the amount of government-created money. Credit policy by the Fed
is more correctly viewed as debt-financed fiscal policy as it inevitably
(and presumably intentionally) favors one party over another and
places taxpayer funds at risk. It amounts to off-budget spending since
it does not go through the usual congressional appropriation process.
The notion of placing taxpayer funds at risk is important for understanding the consequences and dangers of central bank credit policy.
In the case of the Fed, credit policies almost always involve substituting more risky assets for less risky assets on the balance sheet, thus
shifting credit risks from individual entities (usually in the private sector) to the taxpayer. Credit policy decisions place taxpayer funds at risk
and often involve complex and highly political choices. As such, they
I n 2005, he joined Carnegie-Mellon University, adding to the proud tradition in monetary economics established by Allan Meltzer and Ben McCallum. He subsequently
assumed a leadership role in the Carnegie-Rochester Conference on Public Policy
Series and joined the Shadow Open Market Committee (SOMC), two organizations in
which I was deeply involved until my departure for the Federal Reserve in 2006.
5
Douglas Diamond (2022), in this volume, discusses Goodfriend and King (1988) using
their terminology of “banking policy” rather than the more modern “credit policy”
label that I adopt. Banking policy also includes regulations, but that will not be discussed here.
4

| 231

Federal Reserve Independence

Plosser
are more appropriately thought of as the responsibility of Congress
and the Treasury, not the central bank. So it is important to clearly
address the decision-rights and accountability of credit policies. Some
examples help illustrate these points.
The government, and Congress more specifically, frequently engages in an array of credit policy actions that put taxpayer funds at risk.
Some take place through the tax code, but many occur through loans
and loan guarantees to private entities. This is the most common form
of credit allocation by the government. One high-profile example was
a Department of Energy loan guarantee of $535 million made in 2009
to the Solyndra Corp, a developer of solar panels using a new technology. It was touted as a great investment in “green” technology. The
company filed for bankruptcy in 2011, and the government (taxpayer)
took a loss of $528 million.6
Goodfriend and King (1988), however, discuss central bank credit
policies. One common example is lending to individual banks through
the discount window. Federal Reserve Banks make such loans based
on collateral posted by individual banks. The interest rate is called the
primary credit rate and must be approved by the Board of Governors.
The rate is typically set somewhat above the fed funds target rate
determined by the Federal Open Market Committee (FOMC). Such
loans are traditionally very short term, fully collateralized, and can only
be made to solvent depository institutions. A discount window loan
provides the individual bank with reserves so that both the assets and
liabilities of the Fed increase. But such lending need not increase the
balance sheet since the Fed could simultaneously sell Treasuries from
its portfolio to offset or “sterilize” the transaction, leaving monetary
policy unchanged. Thus, such sterilized lending would be
6

232 |

See Geman (2015) for an interesting account of this episode.

characterized as pure credit policy.7 8 The resulting composition of
Fed assets would likely be riskier than the Treasuries it held before. So
selling Treasuries to purchase riskier assets, credit policy by the Fed is
essentially debt-financed fiscal policy.
Another example discussed by Goodfriend (1994) and Broaddus
and Goodfriend (1996) is Fed participation in foreign exchange intervention. Such interventions can be sterilized (through an offsetting
purchase or sale of Treasury securities) or unsterilized, which allows the
balance sheet size to respond. The empirical evidence suggests that
sterilized foreign exchange interventions have small and temporary
effects at most. Requiring or expecting the Fed to engage in credit policy in the form of sterilized foreign exchange interventions that expose
the Fed to credit risk and place taxpayer funds at risk would seem to be
of questionable value without explicit congressional approval.
In a remarkable stroke of foresight, Goodfriend (1994) suggested
that the Fed and Treasury consider a “new Accord” to address Fed
credit policies. He worried that “large federal budget deficits, a deposit
insurance crisis, or a significant foreign exchange market intervention”
might give rise to increased fiscal pressures on the Fed and specifically
on its credit policies.9
Goodfriend (1994) recommended that such a “new Accord” be based
on the following principles: “(1) liquidity assistance should not fund insolvent institutions; (2) credit policy should not fund expenditures that
ought to get explicit Congressional authorization; (3) Congress should
 n August 16, 2007, the FOMC had just such a conversation. In discussing a proO
posed reduction in the discount rate from 100bp above the fed funds target to 50bp,
Jeffrey Lacker, then president of the Federal Reserve Bank of Richmond and longtime
colleague of Goodfriend, asked if any additional loans would be sterilized or allowed
to increase the balance sheet. Bill Dudley, then manager of the System Open Market
Desk, responded that there would be “offsetting adjustments.” See Board of Governors (2007, p. 4).
8
As will be discussed further below, many of the lending programs pursued between
2007 and September 2008 can be thought of as pure credit policies as they were
largely sterilized, thus having little direct consequence for the stance of monetary
policy.
9
See also Broaddus and Goodfriend (1996).
7

| 233

Plosser
not direct the Fed to transfer assets to the Treasury in order to reduce
the Federal debt.” Such an agreement, of course, has not come to pass.
Events since 2008 have strengthened the case for greater clarity of
the boundaries between the Fed and both the Treasury and Congress
regarding the decision-rights and accountability of credit policies.
In the remainder of this essay, I offer some general, mostly well-known,
observations about the importance of an independent central bank
and the critical role of institutional constraints in preserving
independence. Then I highlight how changes in the use of the Fed’s
balance sheet by both the Fed and the fiscal authorities (the Treasury
and Congress) have potentially undermined the fragile balance established by the 1951 Accord. This discussion is followed by some suggestions — largely compatible with those in Goodfriend (1994, 2009) and
Plosser (2009a, 2009b, 2017b) — for reclaiming both independence
and accountability of the Fed by strengthening the boundaries around
credit policy.

Fed independence and the role of institutions
The case for central bank independence largely stems from monetary policy’s unique role in providing price stability. Since a central
bank can also play a role in financing government expenditures, the
potential for conflicting interests between the monetary and fiscal
authorities is clear, as was evident in the events leading up to the 1951
Accord. Governments can finance spending in three ways: taxation,
debt (future taxes), or printing money. In this sense, monetary and
fiscal policy are intertwined through the government’s budget constraint. The historical record offers a strong case for the independence
of monetary policy. It teaches us that without institutional or constitutional constraints of some form, governments often resort to the
printing press to avoid difficult fiscal decisions, potentially undermining monetary policy’s responsibility for ensuring price stability.10

10

234 |

S ee Plosser (2016a) for a brief discussion of recent events in Argentina as one example.

Federal Reserve Independence
Thus, there are good reasons to maintain a healthy separation
between monetary and fiscal policy. The 1951 Accord is a form of
institutional constraint that acknowledges that the Fed has the decision-rights to control the size of its own balance sheet to protect
monetary policy from dominance by the fiscal or political authorities.
But it also suggests that the Fed should refrain from active engagement in issues that fundamentally are under the purview of Congress
and the fiscal authorities.
Plosser (2010, 2012, 2014) discusses some dimensions of these institutional arrangements but stresses that independence does not mean
that the Fed should have unrestricted powers, nor does it mean the
institution is unaccountable. In a democracy, independence requires
that there be constraints on the breadth of the central bank’s responsibilities and its powers.
Limits to central bank authorities can take different forms. One
important element involves the breadth and scope of the bank’s
mandate. Narrow mandates focus the central bank on a limited set of
objectives that make it easier to hold the institution accountable for
success or failure. Narrow mandates also limit the range of the central
bank’s responsibilities that it can use to justify its actions. This argues
for a mandate that focuses more narrowly on price stability.11
It is also common to see restrictions on the types of assets that the
central bank can buy and sell to limit its interference with market
allocations of scarce capital and generally to avoid actions more appropriately left to the fiscal authorities or the market. For example, the
Federal Reserve Act (FRA) limits asset purchases by the Fed to specific
classes of securities.12 The Fed does not have the general authority to
buy private sector securities, such as equities or corporate bonds, nor
can it buy securities issued by state or local governments
 nfortunately, the trend is an expanding mandate for central banks and the Fed, in
U
particular. They are being pressured to consider all sorts of distributional issues in
the real economy from wealth inequality, inclusive employment, as well as broader
issues such as climate change. Levy and Plosser (2020), for example, discuss how this
arises in the Fed’s new monetary strategy adopted in August 2020.
12
See Section 14(2(b)) of the FRA.
11

| 235

Plosser
unless they are revenue bonds or tax anticipation bonds with maturities of six months or less.13
Between 1950 and 2007, the Fed followed a policy of buying and
selling almost exclusively Treasury securities. In fact, between 1951
and 2000, outright holdings of Treasury securities accounted for over
85 percent of the growth in the balance sheet and in 2000, 83 percent
of total assets.14 By August 2007, Treasury securities held outright
accounted for 90 percent of the Fed’s balance sheet and 92 percent of
the asset growth since 1950. Thus, the Fed operated a de facto “Treasuries-only” approach to its balance sheet. The Fed’s rationale for this
approach has been an explicit desire to make its buying and selling
as neutral as possible on the allocation of capital by the private sector
and to ensure its portfolio is liquid, avoiding large amounts of interest
rate risk inherent in longer-term bonds.15
Legal constraints on the central bank can sometimes provide protection from fiscal interference because these give the Fed grounds for
denying requests from the fiscal authorities. Of course, Congress can
change the law, but such constraints raise the bar for getting the Fed
to do something that might undermine its independence. An example,
which will be discussed further below, occurred in December 2008
when Senator Chris Dodd wrote a letter to Fed Chair Ben Bernanke
requesting that the Fed lend money to the failing automobile companies. Bernanke said no, on the grounds that certain provisions of the
FRA made it inappropriate for the Fed to intervene in the case or to
engage in industrial policy. His case was weakened by the Fed’s own
actions to rescue Bear Stearns and AIG earlier in the year and its
T here are exceptions such as loans at the discount window and, under extraordinary
circumstances, lending under Section13(3) of the FRA, which is considered further
below.
14
Outright holdings of Treasuries exclude repurchase agreements collateralized by
Treasuries, which were largely temporary in nature. Including them would raise these
percentages slightly.
15
The Fed has frequently debated whether its holdings of Treasuries should be predominately in the form of short-term bills or a more balanced range of maturities
that mimic the actual distribution of the outstanding public debt. In the 1950s the
portfolio was more heavily weighted toward longer maturity Treasuries as a consequence of the accumulation of wartime bonds and the interest rate peg prior to
13

236 |

Federal Reserve Independence
financial support for the GSE’s through the purchases of agency debt.16
Of course, constraints and restrictions can also be placed on the fiscal authorities that limit interference in monetary policy. The 1951 Accord was just such an example, as it declared that the Treasury would
not use the Fed’s balance sheet to directly fund fiscal deficits. The
prominent role credit policy has played since the financial crisis makes
it important to reconsider the limits to the Fed’s scope for such policies
and do so in a way that respects and protects its independence.

The fraying boundaries
The boundaries established in the FRA and strengthened by the
1951 Accord enhanced the credibility of the Fed as an independent
central bank and its ability to achieve its mandates. Of course, performance has not been perfect, as the high inflation of the late 1960s and
1970s clearly illustrates. The fiscal authorities, including the executive
branch and members of Congress, frequently express their views on
the appropriateness of Fed policy, but often with little or no effect on
policy. Yet in the years leading up to the Great Inflation, President Lyndon Johnson exerted increasing pressure on Federal Reserve Chair William McChesney Martin to keep rates low to help support his spending
agenda that included the Great Society programs and the Vietnam
War. Martin became concerned about inflationary pressures, and many
on the FOMC were even more concerned. The pressure from the White
House was intense: Martin succumbed by delaying FOMC action, in
t he 1951 Accord. Gradually, the portfolio tilted toward shorter maturities, in part to
improve its liquidity. There was also a desire to make the portfolio “neutral,” that is
similar to the maturity structure of the outstanding public debt. There were occasions when the Treasury would try to get the Fed to extend its maturity structure
to accommodate Treasury’s funding desires. As of 2002, the Fed’s Treasury portfolio
was pretty close to neutral and thus was heavily weighted toward the short end of
maturity spectrum. About 65 percent of the Treasury securities had maturities of two
years or less (compared to about 55 percent of the public debt) and a little over 10
percent had maturities of 10 years or more (compared to about 17 percent of the
public debt). See Huther, Ihrig, and Klee (2017) for an interesting discussion of the
evolution of the Fed’s Treasuries portfolio.
16
As will be discussed later, Congress ended up using fiscal policy tools to deal with
the challenges facing the auto industry.

| 237

Plosser
part, by telling his colleagues that the Fed should keep the Treasury
market on an “even keel.”17 The Fed finally raised the discount rate on
December 3, 1965. The president was not pleased. The political pressure from the executive branch was clearly at odds with the spirit of
the 1951 Accord and was largely responsible for beginnings of the
Great Inflation. In October 1969, with inflation running at nearly 6 percent, President Richard Nixon replaced Martin, who by then was committed to restraining inflation, with Arthur Burns, whom he thought
would be more compliant with his political wishes. And indeed, he
was. The Accord was strained, and political pressure dominated Fed
decision-making. These periods under presidents Johnson and Nixon
were consequential breakdowns in the traditional boundaries that
had developed to support independence and avoid monetary and
fiscal policy entanglements. They bear significant responsibility for the
subsequent inflation.18
Following the Great Inflation, the wisdom and spirit of the Accord
was mostly restored with the strong support of Fed Chair Paul Volcker
and President Ronald Reagan. As mentioned at the outset, by 2001,
the 50th anniversary of the 1951 Accord, central bank independence
seemed well established as a principle of sound central banking and
price stability was increasingly accepted as the primary objective.
Yet, beginning with the financial crisis and the accompanying
recession, and continuing into the pandemic and economic turmoil it
engendered, we again witnessed substantial deterioration in the traditional institutional barriers between fiscal and monetary policy, especially the willingness of the Fed to use its balance sheet to conduct
credit policy. In the remainder of this section I trace how the Fed’s credit policies shaped its balance sheet during the 2007 to 2021 period.
“ Even keel” policy during the 1950s and 1960s was Fed speak for avoiding policy
actions that might disrupt financing operations of the Treasury.
18
See Levin and Taylor (2013) and Meltzer (2009) for a more in-depth discussion of this
historic period.
17

238 |

Federal Reserve Independence

Balance sheet responses to the financial crisis
The Early Crisis and Sterilized Lending. Early in the financial crisis, August 2007 to August 2008, the Fed aggressively used its balance sheet
to conduct sterilized credit policy. As a consequence, these early credit
policies had little impact on the size of the balance sheet but did significantly change its composition. As pointed out previously, in August
2007, outright holdings of Treasury securities accounted for about 90
percent of the Fed’s balance sheet. One year later, in August 2008, Fed
holdings of Treasuries had declined by $305 billion, largely to fund its
credit extensions. Treasuries fell to just 53 percent of assets held by the
Fed. On net, the balance sheet increased by less than 4 percent, about
$37 billion over the year.
Much of the lending went to depository institutions through discount window loans and through a new program created in December
2007 called the Term Auction Facility (TAF) that supplemented traditional discount window lending but permitted longer terms. By August
2008, such lending increased by almost $170 billion.19 20
The most significant event during this period was the Fed’s steps to
rescue the creditors of the investment bank Bear Stearns. Like its other
actions during this period, this lending arrangement (about $30 billion) was credit policy as it was primarily the purchase of non-Treasury
securities (mostly high-yield/subprime mortgages) financed by the
sale of Treasury securities. In other words, it was debt-financed fiscal
policy, yet the spending did not show up as funds appropriated by
 ther assets that increased during this period included dollar swap lines with foreign
O
central banks and repurchase agreements.
20
Some ask why the Fed was concerned about sterilization and increasing the size
of the balance sheet at that time. It is useful to remember that in December 2007
the CPI year-over-year rate of inflation was 4.1 percent and by August 2008 it was
5.4 percent. The Fed did reduce the funds rate from 4.25 percent, where it started
the year, down to 2 percent by the end of April 2008. So, it is not entirely a surprise
that the Fed was weighing the prospects for inflation as well as the risks financial
instability.
19

| 239

Plosser
Congress. The rescue was undertaken by the Board of Governors (not
the FOMC) under the authority granted in Section 13(3) of the FRA.21
By the end of August 2008, sterilized lending by the Fed led its holdings of short-dated Treasuries to fall to $22 billion. This decline made
it much more difficult for the Fed to engage in further sterilization
without selling long-term Treasuries.
The powers under Section 13(3) of the FRA allowed for the expanding role of the Fed into credit allocation. The provisions permitted the
Board of Governors, “under unusual and exigent circumstances” to lend
to private firms, individuals, and partnerships. It was originally put in
place in the 1930s as a complement to the Fed’s role as lender of last
resort. Yet it was almost never used.22 Even in periods of severe financial strain, such as the savings and loan crisis, the failures of Enron and
WorldCom, the stock market crisis in 1987, the collapse of the NASDAQ
accompanying the bursting of the so-called tech bubble, and the
financial stress of the terrorist attacks on 9/11, the Fed did not resort
to the discretionary powers of Section 13(3) to purchase private sector
risky securities.23 The Fed earned a reputation, developed over nearly
three-quarters of a century, that it would not use this provision to
conduct credit policy. The Fed’s conduct starting in March 2008 undermined this reputation and the long-standing boundaries recognized
by the Fed and the Treasury.

T he lending to support the Bear Stearns rescue was consolidated into the Maiden
Lane, LLC facility and presented as a separate item on the Fed’s balance sheet. At the
time of the Bear Stearns rescue the Fed created two other Section 13(3) facilities in
support of primary dealers, the Term Securities Lending Facility (TSLF) and the Primary Dealer Credit Facility (PCDF).
22
Prior to 1980 the Fed did use Section 13(3) on a few occasions to permit nonmember
banks to borrow at the discount window. The Monetary Control Act of 1980 eliminated the prohibition of nonmember borrowing from the Fed so any depository bank
could have access to the discount window.
23
During this period the Fed also declined to use its Section 13(3) authority to lend
to Penn Central, New York City, Lockheed, or Chrysler, despite political pressure and
claims of potential financial contagion. See Fettig (2002) and Schwartz (1992) for
some additional history and discussion.

Federal Reserve Independence
In many ways, the rescue of Bear Stearns was the watershed moment in the crisis. The Board used its powers under Section 13(3)
for the first time since the Great Depression to engage in lending to
rescue a failing private entity outside the traditional banking system.
This created uncertainty about future policy actions and moral hazard
that did not exist before the intervention. That is, firms could be rescued with public funds by the Federal Reserve. It set the stage for the
turmoil that followed in September of 2008 surrounding the rescue of
AIG and the failure of Lehman Bros.24
Thus, early in the financial crisis the stage was set for a much more
active role for the Fed in credit allocation and thus, a more active involvement in fiscal policy.
The Financial Crisis Stage Two: More Credit Policy and Balance Sheet
Expansion. In September to December 2008, following the failure of
Lehman Bros. on September 15, the Fed invested $85 billion in the
rescue of AIG using the Board’s powers under Section 13(3). Over
the course of the fall of 2008, the Board also created, again using
Section13(3), an alphabet soup of lending programs in an effort to
support the broader real economy by investing in private sector
securities.25 It included programs to make loans to private investors
to purchase asset-backed securities, commercial paper, and to support money market funds, for example. These programs would not be
sterilized and thus would involve balance sheet expansion, an action
impacting monetary policy.

21

240 |

T he question is would Lehman Bros. (and other financial actors) have behaved
differently had Bear Stearns not been rescued by the Fed? We do not know. We do
know that following the failure of Lehman, other major investment banks became
banks with access to the discount window, yet they were subject to more regulation
and oversight (Goldman Sachs and Morgan Stanley), and another (Merrill Lynch) sold
itself to a bank (Bank of America). Might Lehman, along with the others, have done
the same thing earlier in the year had Bear Stearns been allowed to fail, perhaps
reducing some of the ensuing turmoil?
25
In addition to the previously announced TSLF and PDCF, these new programs included the Term Asset Backed Loan Facility (TALF)), Commercial Paper Funding Facility
(CPFF), Asset-Backed Commercial Paper Money Market Mutual Fund Lending Facility
(AMLF), and the Money Market Investor Funding Facility (MMIF).
24

| 241

Federal Reserve Independence

Plosser
It is important to recognize that these programs took time to establish and could have been implemented directly by the Treasury.
Fundamentally, the new programs shifted credit risks from private entities to the taxpayer without explicit congressional approval. As such,
they represented off-budget spending by the Fed. Even in fast-moving
crises, where it may be desirable to have the Fed act with alacrity, the
programs (including the lending to rescue Bear Stearns and AIG) could
have been shifted over to the Treasury after a few months in exchange
for government securities. This would have required legislation, but
it would have meant Congress was responsible for the oversight and
accountability of these taxpayer financed investments. I suggested just
such an action in the FOMC meeting in December 2008 and publicly
in Plosser (2009a).26 I will discuss this strategy more below as it is a key
part of my “New Accord” recommendations to clarify and constrain the
use of Fed credit policies in an emergency.
Another significant step was taken in November 2008 when the
Board announced it would purchase $500 billion in agency MBS and
$100 billion of agency debt beginning in January 2009.27 This was an
entirely new and significant step into credit allocation. As discussed
previously, the Fed had essentially operated with a “Treasuries-only”
portfolio since the 1951 Accord. Because the agency MBS purchases
also expanded the balance sheet, it was also an important monetary

policy action, although this was not emphasized by the Board of
Governors at the time.28 In the announcement, for example, the Fed
said, “This action is being taken to reduce the cost and increase the
availability of credit for the purchase of houses, which in turn should
support housing markets and foster improved conditions in financial
markets more generally.”29 Chair Bernanke (2009) spoke specifically on
this issue, saying that the Fed was using its powers to engage in “credit
easing.” As he explained, the tools “…. make use of the asset side of the
Federal Reserve’s balance sheet. That is, each involves the Fed’s authorities to extend credit or purchase securities.” The actions were clearly
stated in terms of the Fed’s intention to allocate credit to the housing
sector relative to other sectors of the economy. The alternative, of
course, would be to buy Treasuries.
In March 2009, the FOMC announced it would increase the intended
agency MBS purchases by $750 billion and agency debt by $100 billion. The announcement also indicated the intention to purchase $300
billion in longer-term Treasury securities. Purchases of agency MBS
continued until the spring of 2010. At that point, the financial markets’
functioning had mostly normalized, and the economy was beginning
to slowly recover. The Fed, however, continued to purchase long-term
Treasuries and lengthen the maturity of its portfolio.
T here is an important underlying issue that has not received much public attention.
The announcement of MBS purchases in November was made by the Board of Governors (BOG) not the FOMC. In fact, there seems to be no FOMC document or record
of prior approval by the FOMC. Consequently, the BOG appears to have made an
announcement of a major expansion of the Fed’s balance sheet without the explicit
approval of the FOMC, which is the body responsible for monetary policy. While the
expansion was approved by the FOMC at its December 2008 meeting, it was accompanied by significant discussion scattered throughout the meeting (see Board of
Governors (2008b, pp. 16-103, 166-167)). The commentary on pages 30-35 by Jeffrey
Lacker (president of the Richmond Fed) concerning governance and the roles and
responsibilities of the BOG and the FOMC regarding monetary policy and the size of
the balance sheet is particularly relevant. Outside of commitments by the chair to
work cooperatively with the FOMC in the future on such matters, I was unable to find
subsequent documentation that these governance issues have been discussed or
further clarified.
29
From 2015 through 2017, the Fed did purchase some agency MBS as a means of
stabilizing the size of its MBS portfolio.
28

I n the FOMC meeting on December 15-16, 2008, I said: “As I have articulated before, I
believe we need to remain cognizant of the line between monetary policy and fiscal
policy. I would prefer to see us purchasing Treasuries rather than riskier assets, as I
would favor the purchases of long-term Treasuries over new 13(3) facilities. ... To the
extent that some of our lending programs are targeted at aiding specific markets, my
preference would be to shift those assets from the Fed’s balance sheet to the Treasury and substitute Treasury securities. This would help distinguish monetary policy
from credit policy and preserve our ability to conduct independent monetary policy.”
Board of Governors (2008b, p. 41).
27
See Board of Governors (2008a). The term agency debt refers to the direct obligations of housing-related government-sponsored enterprises (GSEs), including Fannie
Mae, Freddie Mac, and the Federal Home Loan Banks. Agency MBS are the mortgage-backed securities backed by Fannie Mae, Freddie Mac, and Ginnie Mae. The
credit policy dimensions of the program were made all the more apparent by the
purchase of GSE debt. But this aspect of the program had multiple implications. For
example, at this point the GSEs had already been placed in conservatorship, so by
purchasing agency debt the Fed was supporting the fiscal policy dimensions of the
government takeover of the institutions.
26

242 |

| 243

Federal Reserve Independence

Plosser
By the summer of 2012, the Fed was dissatisfied with the pace of the
recovery and wanted to apply more monetary stimulus, so the rationale for MBS purchases was modified “to support a stronger economic
recovery and help ensure inflation, over time, is most consistent with
its dual mandate.”30 Purchases continued through the end of 2014.31 By
December 2014, the Fed’s portfolio of agency MBS had grown to $1.74
trillion, accounting for almost 40 percent of Fed assets.32
The Treasury and Fed Joint Statement of March 2009. After concerns
expressed by FOMC members and others, the Fed and Treasury issued
a joint press release on March 23, 2009, seeking to clarify the role of
the Fed.33 The statement made four points. First, the Fed and Treasury
should cooperate in "improving the functioning of credit markets and
fostering financial stability.” Second, the Fed “should seek to avoid
credit risk and credit allocation,” noting that “(g)overnment decisions to
influence the allocation of credit are the province of the fiscal authorities.” Third, it acknowledged the need for the Fed to preserve monetary
stability and that lending under emergency circumstances (that is, Section 13(3)) that increased the balance sheet, “must not constrain the
exercise of monetary policy.” It added that the Fed and Treasury were
seeking to provide “additional tools the Federal Reserve can use to
sterilize the effects of its lending or securities purchases on the supply
of bank reserves.” The fourth and final point was to recognize the need
for a “resolution regime for systemically critical financial institutions.”
The concluding paragraph offered a tantalizing hint at Treasury’s financial responsibilities: “In the longer run and as its authorities permit, the
Treasury will seek to remove from the Federal Reserve’s balance sheet,
or to liquidate, the so-called Maiden Lane facilities.”
 oard of Governors (2012).
B
From 2015 through 2017, the Fed did purchase some agency MBS as a means of
stabilizing the size of its MBS portfolio.
32
The total size of the balance sheet in December 2014 had grown to $4.5 trillion, more
than a fivefold increase from its pre-crisis size of $850 billion.
33
See Board of Governors (2009) for the joint press release. See Lacker (2009) and
Plosser (2009a) for public expressions of concerns in early 2009 and footnotes 26 and
39 regarding comments by Plosser in the transcript of the December 15-16, 2008,
FOMC meeting (Board of Governors (2008b)).
30
31

244 |

This joint statement made some good points but did not change or
constrain Fed or Treasury actions in any meaningful way. The Fed continued its Section 13(3) credit policies, including its agency MBS purchases, and non-Treasury assets were never transferred to the Treasury.
Moreover, as will be discussed further below, the Fed’s response to the
pandemic in 2020 followed the same playbook, constructing credit
programs and purchasing agency MBS in a manner similar to what it
had done in the post-2007 period, except on a larger scale.
Section 13(3) and the Dodd-Frank Act of 2010. The Dodd-Frank Act
of 2010 made some changes to provisions in Section 13(3) to address
concerns about credit allocation and the use of emergency lending.
Three provisions are relevant. First, the Fed must get prior approval
from the secretary of the Treasury for any program proposed under
Section 13(3).34 Second, any program or facility established must be
designed to provide liquidity to the financial system, have broadbased eligibility, and not be constructed in such a way as to target a
single firm or entity. Third, all transactions conducted under Section
13(3) of the FRA must be with “solvent” institutions or entities only.
These changes place some restrictions on emergency lending but, as
we will see, they are not very limiting.
Other Forms of Risk Taking on the Balance Sheet. As previously noted, by the late 1990s and up until 2007, the Fed managed its Treasury
portfolio in a “neutral” fashion. That is, it had a maturity structure that
looked similar to the maturity structure of the overall public debt.35
Beginning in 2007, the duration of the Fed’s portfolio began to lengthen as its credit programs expanded and it sold short-term Treasuries
to sterilize the impact on the balance sheet. So not only was the Fed
taking on more credit risk, but it was also acquiring more interest rate
risk. A deliberate effort to lengthen the duration of its Treasury portfolio began in March 2009 when the Fed announced that it intended to
purchase $300 billion in long-term Treasury securities. The rationale
 rior to this change, the Board of Governors could determine if “circumstances” could
P
be categorized as an “unusual and exigent.”
35
See Huther, Ihrig, and Klee (2017) for some supporting evidence.
34

| 245

Plosser
was an attempt to reduce long-term interest rates and “improve conditions in the private credit markets.” 36
The Maturity Extension Program and Reinvestment Policy was initiated in September 2011 and sought to lower the longer-term interest
rate by selling Treasuries from the Fed’s portfolio with maturities of less
than three years and purchasing Treasuries with maturities of greater
than six years. This changed the composition or maturity structure of
Treasury securities on the Fed’s balance sheet with the goal of flattening the yield curve of government debt. The Fed tried this once before
in the 1960s in what was known as Operation Twist. At the time it was
widely viewed as unsuccessful, although some suggest it may have
been more effective than previously thought.37 But regardless of its
efficacy, the operation is an exercise in maturity management of the
public debt, traditionally executed by the Treasury and not the Fed.
Moreover, the Treasury could effectively offset the Fed’s effort by issuing more long-term debt relative to short-term debt, resulting in the
same distribution of maturities in the hands of the public as existed
before the Fed’s program. Of course, the effect on the Fed, in either
case, is to increase the interest rate risk of its own portfolio by creating
a greater maturity mismatch between the Fed’s liabilities (bank reserves and currency) and its assets.38
Congressional Responses to Credit Policies and Balance Sheet Expansion by the Fed. In addition to the revision to Section 13(3) already
discussed, the Fed’s plunge into credit allocation generated policy
responses by Congress. For example, as discussed previously, Congress
asked the Fed to consider financial aid to the automobile companies.
The companies had been under financial stress prior to the crisis, but it
turned more acute in 2008. Chair Bernanke turned down the request.
 f course, the purchase of agency MBS was also lengthening the duration of the
O
Fed’s portfolio.
37
See Swanson (2011).
38
The System Open Market Account (SOMA), comprised of securities held by the
Fed, had an average duration of 2.8 years in August 2007. By December 2014 it had
grown to almost six years, and by December 2020 it stood at about five years.
36

246 |

Federal Reserve Independence
In the end, Congress addressed the matter directly using its fiscal
powers.39 However, it is not difficult to understand the request from
Congress. The Fed had already rescued Bear Stearns and AIG, although
Lehman was allowed to fail; it had committed to providing support to
the housing sector, including purchases of agency debt of the GSEs;
and it was purchasing (through the CPFF) commercial paper directly
from issuers. So why not provide support to the automobile industry?
This illustrates the confusion created in the eyes of the public, and
Congress, over the limits or boundaries surrounding Fed credit policies.40

T he big three auto companies received about $80 billion in assistance through the
Troubled Asset Relief Program (TARP). However, restructuring attempts failed, and in
June 2009 GM and Chrysler went through a forced bankruptcy and restructuring. See
Klier and Rubenstein (2012) for a review of the crisis in the auto industry.
40
This issue of acceptable lending and its impact on intermediation was discussed
more broadly in the December FOMC meeting and is illustrated by the following
exchange. Board of Governors (2008b, pp. 235):
	MR. PLOSSER. But in some sense, just to follow up on this point, the limits are
what is really important here because, as long as we don’t define some limits
and we just say limited by TARP capital, well, that doesn’t really answer the
question. As long as the markets act as if we or someone else is going to step
in and rescue them from any more lending arrangements they happen to be
facing, the incentives for the intermediary system to repair itself or to gradually adjust are going to be limited. I’m worried about the lack of definition
about what constitutes a legitimate market or instrument or firm that we
wouldn’t save.
	CHAIRMAN BERNANKE. That’s a good point, and I think one thing that is a
problem now is the transition between Administrations. We’ll soon have a
new Treasury Secretary and a new Administration. I think it’s very important—I’ve discussed this with Tim Geithner and others—that as soon as possible we lay out a broad strategy. What are the components of our strategy?
What are we going to do going forward?
MR. PLOSSER. And what are the limits to it?
39

| 247

Plosser
Unfortunately, efforts to exploit the Fed’s balance sheet for fiscal
purposes continued. In the Dodd-Frank Act of 2010, Congress relied on
the Fed for direct fiscal support. As part of the legislation, the Consumer Financial Protection Bureau (CFPB) was created. It was a vigorously
debated provision. In the end, Congress created the agency and decided the Fed should fund it but gave the Fed no control or oversight
authority. The result is that the CFPB is exempt from the annual appropriation process.41 Of course, this mandated expenditure reduces Fed
payments to the Treasury each year. Thus, the taxpayers still pay, but
the agency becomes an off-budget expenditure. Without a change in
the enabling legislation, the CFPB budget is determined by Fed operating expenditures and rises with them.
In the Fixing America’s Surface Transportation Act (FAST) of 2015,
Congress also used the Fed’s balance sheet as a means of funding.
Specifically, the act required that the Federal Reserve Bank surplus account not exceed $10 billion. This resulted in almost $20 billion being
transferred directly to the Treasury to help fund the act.42 In the Bipartisan Budget Act of 2018, Congress further decreased the Fed’s surplus
account to $7.5 billion, resulting in another $2.5 billion transfer to the
Treasury.43
The Fed is required to pay all expenses of the CFPB up to the equivalent of 12 percent of the Fed’s own operating budget.
42
The 2015 act also required the Fed to reduce the dividend it was paying to member
banks. These changes (the permanent cap on the surplus account and reduction in
the dividends) constituted a permanent increase in flow of funds from the Fed to the
Treasury.
43
Congress has exploited the Fed’s balance sheet in this manner before. One example
was in 2000 when Congress passed a budget that transferred $3.75 billion from the
Fed’s surplus account to the Treasury. Goodfriend (1994) considers an even earlier example in 1993. This motivated him to include as his third principle for a new Accord
quoted earlier, “Congress should not direct the Fed to transfer assets to the Treasury
in order to reduce the Federal deficit.” These actions to exploit the Fed’s balance
sheet clearly violate the spirit of the 1951 Accord.
41

248 |

Federal Reserve Independence
Managing a Large Balance Sheet and the Potential for Abuse. Unsterilized credit policy initiatives, including the large volume of agency
MBS purchases, contributed to unprecedented growth in bank reserves and hence the Fed’s balance sheet. The Fed has never seriously
considered reducing the balance sheet sufficiently to enable a return
to its precrisis operating regime, commonly referred to as a “corridor
system.” In this prior regime, the Fed adjusted the volume of bank
reserves up and down to ensure that the fed funds rate (the rate that
banks trade reserves among themselves) remained close to its target
set by the FOMC. Most major central banks used this regime prior to
the financial crisis.
Once bank reserves became large, this framework could not be used,
as modest changes in bank reserves would have no impact on the
effective fed funds rate.44 As long as the fed funds target is effectively
zero, this is not a major issue. It becomes a significant issue when the
Fed wishes to raise the fed funds target above zero. How can it do that
when the banking system is flooded with reserves? Goodfriend (2002)
suggested that paying interest on reserves (IOR) would be a way to
raise rates even if a large volume of reserves existed in the banking system.45 46 The IOR acts as a floor on short-term rates under which banks
have no incentive to lend. Thus, raising the IOR would encourage other
rates to increase without shrinking reserves. In central bank parlance
this is called a “floor system.” The essential policy instrument in this
regime is the interest rate paid on reserves.47
In theory, this allows the Fed to manage the interest rate and bank
reserves separately. The danger of this approach is that it increases the
temptation to use the Fed’s balance sheet for other purposes. The
I n December 2006, bank reserves were about $55 billion, and by December 2015,
they had reached over $2.5 trillion. They declined to $1.5 trillion in September 2018
when they began to rise again, reaching over $4.4 trillion in October 2021.
45
The Fed was granted the ability pay interest on reserves in October 2008.
46
Ennis and Weinberg, in this volume, discuss in detail the recent Fed experience with
paying interest on reserves, in the context of Goodfriend's seminal 2002 paper.
47
This creates another governance challenge because IOR is set by the Board of Governors. So technically this leaves the BOG, not the FOMC, with the ultimate authority
to determine monetary policy. It seems that the FOMC can suggest the stance of
monetary policy, but the BOG does not have to concur. See Plosser (2020) for further
discussion.
44

| 249

Federal Reserve Independence

Plosser
Fed was forced into this arrangement by its QE policies, but the Fed
has now explicitly adopted the floor system as an operating regime,
although they had to give it another name. So it became the “ample
reserve” regime.48 The Fed has been vague as to how it will decide on
the size of the balance sheet and what factors might motivate changes
in the volume of purchases up or down.
A series of official statements by the FOMC on policy normalization
began in 2014.49 None anticipated outright sales of securities. Most
who opposed sales argued that simply allowing runoffs of maturing
securities would be sufficient to gradually shrink the balance sheet.50
Others were concerned that the balance sheet may need to fall faster
and sales may be necessary. The fact that runoffs are likely to be slower
for the MBS portfolio (especially as interest rates rise) means that the
role of credit policy as measured by the share of MBS would likely rise
over time.
All versions of the normalization statement have similar versions of
“the Committee intends to maintain securities holdings in amounts
needed to implement monetary policy efficiently and effectively.” The
statement also declares that “in the longer run, the Committee intends
to hold primarily Treasury securities in the SOMA.” These statements
do not contain much information or constraints on credit policy. In
this new operating regime, the Fed seems to have even more latitude
to fluctuate the size and composition of its balance sheet, but it is
not entirely clear to what end. Based on the arguments in this essay,
the adoption of the new “ample reserve” operating regime reinforces
the importance of a new credit Accord that more tightly constrains
the composition of the Fed’s balance sheet. Put differently, the ample
reserve regime offers no limiting principle on the size or composition
balance sheet. The corridor regime constrains the balance sheet to be
a size that delivers an effective funds rate close to the target set by the
S ee Board of Governors (2019a and 2019b).
See Board of Governors (2022a) and Board of Governors (2022b) for the most recent
incarnation.
50
The Fed has frequently cited concerns that outright sales of MBS risk disruptions in
the financial market and the housing sector.
48
49

250 |

FOMC to implement a monetary policy that achieves the Fed’s macroeconomic mandates.
A large unconstrained balance sheet is ripe for abuse. The fiscal
authorities will be tempted to look to that balance sheet for their own
purposes, including credit policy and off-budget fiscal policy. This
would undoubtedly lead to the Fed being drawn into political debates
on how to best “allocate” or “diversify” the Fed’s balance sheet. This
further politicization of the Fed would lead to a loss of independence
that could interfere with accomplishing its congressionally mandated
goals.51

Balance Sheet Responses to the Pandemic and Shutdowns
Despite the concerns over the Fed’s engagement in credit policy
and the entanglements with fiscal policy, during the pandemic the Fed
mostly relied on the same strategy it had adopted during the financial
crisis. It rapidly expanded the balance sheet and aggressively employed unsterilized credit programs authorized by the Board of Governors under Section 13(3) and approved by the Treasury secretary.
Uncertainty over the risks surrounding the pandemic emerged in the
financial markets in March 2020. On March 3, 2020, the Fed announced
a 50 basis point reduction in its target range for the fed funds rate.
This was followed on March 15, 2020, by an additional 100 basis point
reduction to a range of 0 to 0.25 percent. At the same meeting, the Fed
announced its intention to increase its holdings of Treasury securities
by $500 billion and its holdings of agency MBS by $200 billion. The
stated purpose was to support the “smooth functioning of markets” for
these securities.52
In June 2020, the Fed extended, indefinitely, the asset purchases of
Treasuries and agency MBS at a pace of $80 billion and $40 billion
S ee Plosser (2020) for more discussion of the floor system and the risks it poses to
independence.
52
From the end of March to the end of June the Fed purchased almost $560 billion in
of agency MBS and about $240 billion of Treasuries.

51

| 251

Plosser
per month, respectively. By June 2020, financial markets were largely
functioning normally so, like the shift in 2012, the Fed indicated the
rationale for continued purchases was “fostering effective transmission
of monetary policy to broader financial conditions.”53 The Fed’s balance
sheet grew by $4.4 trillion in just 21 months (January 2020 to September 2021). Treasuries accounted for about $3.1 trillion of the increase
and agency MBS purchases were about $1.1 trillion. These purchases
were huge: following the financial crisis, it took the Fed about six years
to purchase $1.0 trillion in agency MBS while expanding the balance
sheet by a total of about $3.8 trillion.
The virus rapidly spread across the US and the world, and, by midMarch, shutdowns began across the country that resulted in economic
disruption. The Fed rapidly increased and expanded the reach of its
credit policies. The range of coverage was quite remarkable and extended far beyond the banking system. Various lending programs developed during the financial crisis to aid the availability of credit were
revived, including those supporting the commercial paper market, the
asset-backed paper market, money market mutual funds, and securities lending for the primary dealers.54
But there were a number of additional programs created to target
specific sectors of the economy. Several targeted the business sector
by lending directly to corporations through the purchase of newly
issued debt securities, including short-term commercial paper, from
investment grade issuers.55 Another new funding facility enabled
the purchase of existing corporate bonds as well as exchange-traded-funds (ETF) in the secondary market, including those that primarily
invested in risky US high-yield securities.56 Support for midsized companies, including profit and not-for-profit entities, was expanded
 oard of Governors (2020). The housing market was at the center of the financial
B
crisis in 2007-10. In the recession of 2020, housing was not central to the crisis in any
fundamental way.
54
These revived programs included the TALF, CPFF, AMLF, which was renamed Money
Market Liquidity Facility (MMLF), and PDCF.
55
These programs included the CPFF and the Primary Market Corporate Credit Facility
(PMCCF).
56
This program was the Secondary Market Corporate Credit Facility (SMCCF).
53

252 |

Federal Reserve Independence
through additional Section 13(3) programs.57 The Fed also created a
program to lend directly to state and municipal governments.58 According to Section 13(3), as updated by the Dodd-Frank legislation,
the programs had to be approved by the Treasury. Many of them were
structured so that the Treasury committed first-loss backing of about
10 percent.59 Nonetheless, these programs all constituted fiscal policy
actions by the Fed to allocate credit across the economy.
As with many of the funding facilities created during the financial
crisis, all of these facilities could have been set up and administered
by the Treasury without Fed involvement. They were basically debt-financed fiscal policy that transferred risk from private-sector debtholders to the government. Some argue that only the Fed can play the role
in emergencies, but the Paycheck Protection Program (PPP) demonstrates that it can be done. PPP loans made to small businesses were
administered and guaranteed by the Small Business Administration
(SBA).60 This could be an emergency model that does not use the Fed
balance sheet to conduct fiscal policy.
There is an interesting twist to these most recent programs. The
programs were ended in December 2020. Treasury Secretary Steven
Mnuchin recommended that the programs not be renewed. He viewed
the programs as having served their purpose and as no longer being
the best use of funds.61 Fed Chair Jerome Powell preferred to continue
the programs. Congress was split on the issue. One interpretation of
this episode is that, at the end of the day, the fiscal authorities ended
the credit policies of the Fed over the objections of Chair Powell. The
T his was the focus of the Main Street Lending Program (MSLP).
This refers to the Municipal Liquidity Facility (MLF).
59
This backing by the Treasury does not really change anything, as the government
would likely have to absorb any losses (or gains) the Fed incurs in any event. Its
major effect is to acknowledge publicly that the Treasury and Congress bear some
responsibility.
60
The Fed’s involvement was a little different in that the Fed would lend to financial
institutions and take PPP loans guaranteed by the SBA as collateral. As of October
2021, approximately 76 percent of all PPP loans have been forgiven by the SBA ($602
billion) and reimbursements paid to the lenders.
61
The funds in this case were those Congress had dedicated as a limited first-loss backstop for certain 13(3) facilities.
57
58

| 253

Federal Reserve Independence

Plosser
macroeconomic consequences of this action seemed undetectable.
But the debate was clouded by who had the decision rights to do
what. While this episode sheds some light on how Congress can exercise ultimate control over certain credit policy actions by the Fed, it
does not address the potential for abuse of the Fed’s balance sheet for
fiscal purposes. The operation of the PPP would be the better model
for emergency credit policy.

A new Treasury-Fed Accord
The use of the Fed’s balance sheet as a tool of credit policy has
taken on new dimensions since the financial crisis, entangling the Fed
deeper into the realm of fiscal policy. The traditional boundaries between monetary policy and fiscal policy have been breached in ways
not envisioned in the 1951 Accord. Once a central bank ventures into
credit allocation and off-budget financing of fiscal initiatives, it is likely
to find itself under increasing pressure from the private sector, financial markets, or the government to use its balance sheet to substitute
for other fiscal decisions. In essence, groups will seek to capture the
Fed’s balance sheet to further their own economic and political interests. Allocating credit through its lending practices or asset purchases
means the Fed can create its own form of moral hazard, as markets
and governments come to see the central bank as a source of financial
support or a tool of fiscal policy, thus undermining private and public
fiscal discipline. This pressure will undermine central bank trust, invite
politicization, and severely threaten monetary policy’s effectiveness
and independence.
To restore the boundaries between monetary and fiscal policy and
to safeguard Fed independence, there needs to be a new Accord that
clarifies and circumscribes the role of credit policy actions by the Fed.
As noted at the outset of this essay, Goodfriend (1994) recognized that
credit policy undertaken by the Fed poses risks to the institution. The
importance of this issue has grown since the financial crisis, and it is
apparent that there needs to be a clear statement of the roles and responsibilities of the Fed and Treasury to reinforce the integrity of both
fiscal and monetary policy, and reduce uncertainty and moral hazard.
254 |

The original principles laid out in Goodfriend (1994) are useful,
but the challenges that have materialized since the financial crisis go
beyond anything Goodfriend likely envisioned in 1994. In my view, a
new Accord must raise the bar on the fiscal authorities, as well as the
monetary authority, to reverse the growing abuse of the Fed’s balance
sheet and to support and maintain Fed independence.
Goodfriend (2009), updated Goodfriend (1994) in light of the
events in 2007-08. His three principles became six. Briefly, they were
as follows: (1) the Fed should “adhere to a Treasuries-only acquisition
policy except for occasional and limited discount window lending;” (2)
the Fed and Treasury should “co-operate” to “shrink the central bank’s
lending reach” through program runoff or by moving the associated
assets to “be managed elsewhere in the government;” (3) the Fed and
Treasury should “co-operate” so that the credit policy actions do not
“undermine price stability;” (4) the Fed and the Treasury should “agree
on a low long run inflation objective to anchor inflation expectations;”
(5) “the Treasury should help the Fed to secure the power of ‘interest on
reserves;’” (6) the Fed and Treasury should co-operate quickly on these
matters to “secure the commitment to price stability.”62
These ideas are principled but short on specificity, and the institutional mechanisms that would help ensure the boundaries on credit
policies are transparent and effective. The call to “co-operate” does
not change incentives or clarify the boundaries of credit policies but
is most likely to result in maintaining the status quo. The Treasury-Fed
joint press release in 2009, which was a response to internal and external criticism, acknowledged many important concepts but did not
change the incentives or behavior of the Fed or the Treasury. Without
institutional mechanisms or constraints that limit or define the scope
for actions, the incentives to abuse the Fed’s balance sheet through
credit policy will remain a threat to the Fed and monetary policy independence.

62

I t is worth noting that in January 2012 the Fed established, for the first time, an
explicit inflation target (see Lacker [2020] for a discussion of how that came about).
Also, the Fed received the authority to pay interest on reserves in October 2008.

| 255

Federal Reserve Independence

Plosser
Early in the rush to transform the Fed’s balance sheet into a tool of
credit policy, I stressed two themes that would help clarify the boundaries of Fed credit policy and the resolution of conflicts.63 The two
themes are similar in spirit to Goodfriend’s first and second principles
given above. The first is to require the Fed to follow a Treasuries-only
policy when conducting monetary policy. This is not a new idea but
a return to the practice followed by the Fed for most of the postwar
period. The agency MBS purchases were a dramatic departure from
that practice but were legal under the FRA. My intention is to require a
Treasuries-only portfolio with the exception of collateralized lending to
solvent financial institutions through the discount window. The second
theme is to address the role of the Fed as a “lender of last resort,” a traditional function of a central bank, and particularly the powers embedded in Section 13(3) in emergencies.
The framework has three key features:64
1.	The Federal Reserve should be required to maintain a
Treasuries-only policy as it pertains to the conduct of monetary policy.
2.	The Federal Reserve should be prohibited from purchasing
non-Treasury securities or lending against private collateral
except through traditional discount window operations with
solvent depository institutions.
3.	Emergency lending under Section 13(3) should be eliminated
and replaced with a new arrangement where the Treasury is
the responsible agency. The Treasury, however, may request
assistance from the Fed in an emergency. The new provisions
would require the Treasury to exchange (at book value) Treasury securities for any private or non-Treasury securities temporarily acquired by the Fed in the process.

S ee footnote 26 (Board of Governors (2008b, p. 41) and Plosser (2009a, 2009b, and
2009c).
64
This material follows Plosser (2017b).

The Fed Should Maintain a Treasuries-Only Policy in the conduct of
Monetary Policy. From 1951 until 2007, as discussed earlier, the Fed
conducted monetary policy through what can be described as a Treasuries-only policy, that is through the purchase and sale of Treasury
securities. In August 2007, the balance sheet was $873 billion, and
Treasuries accounted for about 90 percent ($785 billion) of Fed assets.
As of October 2021, Treasuries only accounted for 64 percent of the
balance sheet and agency MBS represented about 30 percent of Fed
assets. These holdings comprised about 30 percent ($2.5 trillion) of all
outstanding agency MBS, indicating how large the credit allocation
program of the Fed had become. The purpose of a Treasuries-only
policy is to prevent the Fed from subsidizing the housing sector (or
any other sector) through its conduct of monetary policy. Yet, as noted
earlier, the Fed explicitly stated that the purchases of agency MBS and
agency debt were intended as a policy to give special credit preference
to the housing sector over other sectors through monetary policy.
Such credit allocation should be a fiscal policy decision and not left to
the discretion of the monetary authorities.
One counterargument is that the Fed should not be limited to Treasuries, in fact, should not purchase Treasuries at all, as it ties monetary
policy directly to the funding of fiscal deficits.65 Allowing the Fed to
purchase private sector securities (private equities, corporate bonds,
etc.), so the argument goes, would help protect Fed independence
by breaking the close link with government finance. Yet, this strategy seems worse than the alternative. Central banks with significant
holdings of private sector securities would likely come under even
more pressure from those who seek to use the balance sheet and the
powers of the central bank for credit allocation in support of political
or economic advantage.
The risk is real and already in play in other countries. For example,
the Swiss National Bank (SNB) has come under pressure in the last decade from various groups, including the government, to manage their
portfolio of investments (mostly comprised of foreign exchange

63

256 |

65

See Selgin (2018) for example.

| 257

Plosser
reserves) to satisfy other objectives. There have been calls for the SNB
to invest more in Swiss firms to support the Swiss economy, to use
its portfolio to support “green” investments or sell assets in fossil fuel
industries to promote a climate change agenda, and the list of requests seems to grow over time. If the Fed were free to invest in private
sector assets, there would be no end to the requests from Congress
and elsewhere for the Fed to tailor its asset holdings to suit a variety
of interest groups, both private and public. The Fed is already under
pressure by the financial sector to respond to stock prices and other
asset prices during volatile times. The Fed would likely come under
enormous pressure to stabilize or boost the market if it were actively buying and selling large amounts of equities, even ETFs. Imagine
future confirmation hearings for the Board of Governors that focused
on how a nominee would manage the asset allocation of the Fed’s
multitrillion-dollar balance sheet rather than on the conduct of monetary policy.66 Recently, the Fed has already accepted the premise that
climate change should be a priority. There seems to be little doubt that
it will come under intense political pressure to use its balance sheet to
reflect this priority.67
The Fed’s new “ample reserves” regime and its intention to maintain
a large balance sheet increases the temptation for abuse.68 For all these
reasons, restricting the Fed to a Treasuries-only portfolio would be an
T his is another reason the Fed’s ample reserves approach to an operating regime is
potentially dangerous as it gives the Fed more latitude to expand the balance sheet
for purposes other than monetary policy. See Plosser (2017a and 2020). On this
point, Plosser (2020) and Selgin (2018) seem to agree in preferring the Fed return to
a corridor or channel regime for achieving its interest rate target.
67
This pressure is evident in the early 2022 confirmation hearings of nominees to the
Board.
68
An alternative would be for the Fed to shrink its balance sheet so it can return to its
precrisis corridor regime, but this seems increasingly unlikely.
66

258 |

Federal Reserve Independence
important step in restricting its role in credit allocation, reducing moral
hazard, and helping protect its independence from political encroachment.69
It would be best to implement this requirement by amending
Section 14(2) of the FRA that pertains to open-market operations. The
change would limit the scope of securities eligible for sale and purchase by the Federal Reserve Banks. In particular, it would remove the
provisions that allow the Fed to purchase MBS and certain short-term
obligations of state or local municipalities.
This change does not replace the need for the continuation of the
1951 Accord, which deals with the size of the balance sheet. The interest rate peg implemented through the 1940s, for example, was implemented with a Treasuries-only strategy. With the freedom to expand
the balance sheet enabled by the ample reserve regime, there will be
even more pressure to use the Fed for off-budget public spending. To
protect the independence of monetary policy and allow it to best address its mandated goals, the fiscal authorities (Treasury and Congress)
must refrain from actions that move funds directly from the Fed’s
balance sheet to the Treasury. There also should be a reaffirmation of
the 1951 Accord that pressuring the Fed to add government debt to its
balance sheet through the required purchases of Treasuries to address
fiscal demands is inappropriate. Such actions violate the spirit of the
1951 Accord.
Limited Lending to Financial Institutions. One role of central banks
is to serve as a lender of last resort (LOLR). This notion dates back, at
least, to the early 19th century and the work of Henry Thornton (1802)
and later Walter Bagehot (1873). The basic idea is that central banks are
69

 roaddus and Goodfriend (2001) argued for a Treasuries-only policy, not because the
B
Fed was violating the idea in practice, but in reaction to the unlikely event that the
federal government would pay off the public debt. The impetus for such a discussion
was projections by the Congressional Budget Office (CBO) in 2000 suggesting that
government surpluses might be sufficient to retire all of the public debt within a
decade. They argued from the work of Goodfriend and King (1988) and Goodfriend
(1994), much as is argued here, that the Treasuries-only policy is a good one and that
the Fed and Treasury should cooperate to ensure that there remained a sufficient
stock of Treasuries to allow the Fed to conduct monetary policy using Treasuries
rather than purchasing private assets.

| 259

Federal Reserve Independence

Plosser
well suited to ensure the elasticity of the currency to address banking
crises, which occur when banks, because of the maturity mismatch
between their assets and liabilities, find themselves short of liquidity.
In so doing, central banks can help ensure the integrity of the payment
system through their provision of cash. Bagehot (1873) is credited with
the recommendation that in the face of a banking crisis, the central
bank should lend to solvent institutions at a penalty rate against good
collateral to make sure solvent banks can easily accommodate depositors’ demands for cash.
I will refer to the LOLR as the central bank lending to banking institutions. This essentially describes the functions of the long-standing
discount window where the Fed lends to banks against a wide range
of eligible collateral. The guidelines for borrowing from the discount
window are that the firm must be solvent and the interest rate must
be set above the traditional interbank lending rate, or federal funds
rate. There is some logic to this as the central bank has a responsibility
to support the continued functioning of the payments system in an
emergency. Banks are regulated, the Fed has more ability to assess the
solvency of the institution, and banks remain integral to the payment
system. There are debatable questions as to the breadth of firms that
should have access to the discount window. I believe the criteria
should hinge on the role the institution plays in the payment system,
not simply that it is part of the financial sector in general. But this is not
a question that will be addressed in this essay.

We saw this was possible with the PPP. Even with the modified language of Section 13(3) enacted under Dodd-Frank, the Fed retains
wide discretion as illustrated by the scope of programs developed in
2020. A new Accord must confront the possibility of emergencies but
carefully recognize the appropriate placement of decision-rights and
accountability. In Plosser (2009a), as the Fed was rapidly expanding its
credit policies, I suggested that to safeguard Fed independence and
ensure the integrity of fiscal policy, the Fed and the Treasury should
agree to “an arrangement whereby the Treasury takes the non-Treasury
assets and non-discount window loans from the Fed’s balance sheet
in exchange for Treasury securities.” This would transfer funding for the
credit programs to the Treasury, ensuring that credit policies that put
taxpayer funds at risk are under the control of the fiscal authorities. It
would also return the control of the Fed’s balance sheet to the Fed so
that it can continue to conduct independent monetary policy.
This strategy suggests replacing Section 13(3) with a new arrangement to address emergency lending that would clarify the boundaries
for the Fed and the Treasury.70
•	In an emergency, a request could be made by the Treasury for
Fed assistance in facilitating government policies to support
lending to individuals, partnerships, and corporations.
•	If Congress has not previously granted contingency funds for
such use, the Treasury would be required to immediately seek
congressional approval and funding for the projected expenditures within 30 days of the action.

Limitations on Emergency Lending to Nonfinancial Borrowers. Many
people have come to accept the notion that the central bank’s LOLR
function implies that it should be a backstop lender to all manner of
private institutions that are in stress. Section 13(3) is considered an
emergency provision that allows the Fed to lend to the private sector
under “unusual and exigent circumstances.” Apart from agency MBS
and agency debt purchases, almost all the facilities discussed in this
essay were established under Section 13(3).
These programs were forms of credit policy that should be the
responsibility of and accountable to the Treasury or Congress. There
is little reason they could not have been undertaken by the Treasury.
260 |

•	Upon congressional approval, the Treasury would, within 14 days,
arrange to exchange (at book value) Treasury securities for any
non-Treasury securities or assets that may have been temporarily
acquired by the Fed. Any gains or losses would thus accrue to the
Treasury.
•	Should Congress not approve the necessary funds within the 3070

Such a rewrite should have occurred in the Dodd-Frank legislation.

| 261

Federal Reserve Independence

Plosser
day window, the Fed would be required to liquidate such securities
within 60 days of their acquisition and the Treasury would be required
to terminate the program.

Over the years, but especially after 2008, the Fed and the Treasury have
significantly changed their approach to the central bank’s balance
sheet.

By exchanging the private sector assets on the Fed’s balance sheet
for Treasuries, the Treasury-only requirement is restored, and the Fed
can control the size of its balance sheet through normal operations.
The exchange does not alter the program in any way; the credit policy
or distributional aspects remain in place. The Treasury was responsible for the program and it remains responsible. If the Fed funded the
acquisition of the non-Treasury assets by the sale of Treasuries (that is
sterilized), the exchange would leave the balance sheet back where
it stood prior to the initiation of the program. If the assets purchased
were funded by an expansion of the balance sheet (that is unsterilized), the exchange would permit the Fed to shrink its balance sheet,
through the sale of Treasuries, back to its previous level or not depending on what it thought was the appropriate size. These provisions incorporated into a new Section 13(3) ensure that credit policies remain
the responsibility of the Treasury and Congress, reducing the threats
that the Fed will be pressured to undertake further action on its own
discretionary authority or that the composition of the balance sheet
could impinge on monetary policy decisions.

Some might argue that the new Fed-Treasury approach was all driven by the confrontation with effective lower bound and the financial
crisis that required unusual steps and unconventional policies. This
certainly has an element of truth. This essay, however, is not intended
as an evaluation of the efficacy of the programs themselves. Rather,
my concern is that the deep involvement and reliance on the Federal
Reserve to initiate and administer fiscal policy — in the form of the
allocation of credit — puts at risk the political independence of the
central bank and the integrity of fiscal policy.

This arrangement would clarify the boundaries of Fed credit policies in terms of decision-rights and accountability. The Fed could be
a facilitator on behalf of the Treasury but would not make discretionary decisions on credit actions. Congress would have to recognize its
responsibilities by agreeing in advance to the process and recognizing
that the Fed would be required to sell assets if the decisions on the
funds are not forthcoming.

Closing thoughts
The Treasury-Fed Accord of 1951 was a significant turning point in
establishing the Federal Reserve as an independent central bank. By
abandoning the Treasury’s requirement that the Fed maintain the peg
on government bond yields, it allowed the Fed to control its own balance sheet and freed it to tighten monetary policy to control inflation.
262 |

In early 2020, more than a decade after the financial crisis, the Fed
was a long way from what most would call a normal policy framework,
most notably in its vast holdings of agency MBS. Its new ample reserve
regime does not offer much guidance on how the balance sheet will
be managed in terms of its size or its composition. Will balance sheet
expansion occur at times other than when constrained by the zero
bound? Will MBS purchases become a staple of future expansions?
Without sales, they will remain on the balance sheet for many years.
Will credit policies become a more standard feature of the operating
regime? The Fed’s behavior since 2007 has surely affected the public’s
expectations along these lines and the Fed may find it difficult to say
no.71 When the pandemic came and the government shut the economy down, the Fed pulled out the same playbook (zero rates, massive
asset purchases combined with credit allocation policies, including
purchases of MBS, and emergency lending facilities) even though the
shock was quite different from the financial crisis. While to some this
may appear to be a natural and even desirable step, it is both troubling
and risky from the standpoint of Fed independence and the conduct
of monetary policy.
71

T his confusion surrounding the ample reserve regime and how it will be implemented adds uncertainty to the conduct of monetary policy, and it suggests that the Fed
should consider returning to a corridor system. This would help strengthen the Fed's
position against those that might seek to take advantage of the Fed's balance sheet
for other purposes.

| 263

Plosser

Federal Reserve Independence

To ensure and strengthen the independence of the Fed, clearer
boundaries between monetary and fiscal policies should be established. Without such boundaries, the Fed will come under increasing
pressure to use its balance sheet, and especially credit policy, to further
the special interests of politicians or those in the private sector. Given
this “new” world and the expanded role the Fed and the Treasury have
taken for the balance sheet, we must consider amending the 1951
Accord to address the new challenges to independence.

References

Credit policy poses a threat to Fed independence just as did the
practice of pressuring the Fed to pursue an interest rate peg in the
post-WWII era. This essay seeks to address the shortcomings of the
1951 Accord by restricting the Fed to holding only Treasury securities.
It also recommends that Section 13(3) be replace with a rule that credit
programs should be initiated and managed by the Treasury, not the
Fed. The Treasury could seek the Fed’s assistance in facilitating such
programs. But consequences must be clear in the statute. In particular,
if such facilitation results in the Fed acquiring non-Treasury assets on
behalf of the Treasury, then the Treasury must replace those assets
with government securities within a relatively short, predetermined
time period, or they would be sold in the market, returning the Fed’s
balance sheet to its Treasuries-only status. These steps would strengthen the institutional framework supporting Fed independence and the
integrity of both monetary and fiscal policy.

Board of Governors of the Federal Reserve System. 1951. “Minutes
of the Executive Committee of the Federal Open Market Committee,
February 26, 1951.”

Bagehot, Walter. 1873. Lombard Street: A Description of the Money Market. New York: Scribner, Armstrong & Co.
Bernanke, Ben. 2009. “The Crisis and the Policy Response.” Speech at
the Stamp Lecture, London School of Economics, London, England,
January 13.

Board of Governors of the Federal Reserve System. 2007. “Transcript:
Meeting of the Federal Open Market Committee on August 16, 2007.”
Board of Governors of the Federal Reserve System. 2008a. “Federal
Reserve announces it will initiate a program to purchase the direct
obligations of housing related government-sponsored enterprises and
mortgage-backed securities backed by Fannie Mae, Freddie Mac, and
Ginnie Mae.” Press Release, Board of Governors, November 25.
Board of Governors of the Federal Reserve System. 2008b. “Transcript:
Meeting of the Federal Open Market Committee on December 15-16,
2008.”
Board of Governors of the Federal Reserve System. 2009a. “FOMC
Statement.” Press Release, Board of Governors, March 18.
Board of Governors of the Federal Reserve System. 2009b. “The Role
of the Federal Reserve in Preserving Financial and Monetary Stability
Joint Statement by the Department of the Treasury and the Federal
Reserve.” Press Release, Board of Governors, March 23.
Board of Governors of the Federal Reserve System. 2012. “FOMC Statement.” Press Release, Board of Governors, September 13.
Board of Governors of the Federal Reserve System. 2019a. “Statement
Regarding Monetary Policy Implementation.” Press Release, Board of
Governors, October 11.

264 |

| 265

Plosser
Board of Governors of the Federal Reserve System. 2019b. “Minutes of
the Federal Open Market Committee, October 29-30, 2019.” November
20.

Federal Reserve Independence
Operations and the Federal Reserve.” Federal Reserve Bank of Richmond Economic Quarterly (Winter): 1-19.

Board of Governors of the Federal Reserve System. 2020. “FOMC Statement.” Press Release, Board of Governors, April 29.

Goodfriend, Marvin, and Robert G. King. 1988. “Financial Deregulation,
Monetary Policy, and Central Banking.” Federal Reserve Bank of Richmond Economic Review (May/June): 3-22.

Board of Governors of the Federal Reserve System. 2022a. “History of
the FOMC's Policy Normalization Discussions and Communications.”
January 26.

Hetzel, Robert L., and Ralph F. Leach. 2001. “The Treasury-Fed Accord: A
New Narrative Account.” Federal Reserve Bank of Richmond Economic
Quarterly 87, no. 1 (Winter): 33-55.

Board of Governors of the Federal Reserve System. 2022b. “FOMC Communications Related to Policy Normalization.” January 26

Huther, Jeffrey, Jane Ihrig, and Elizabeth Klee. 2017. “The Federal Reserve’s Portfolio and its Effect on Interest Rates.” Finance and Economics Discussion Series Working Paper 2017-075, June.

Broaddus Jr., J. Alfred, and Marvin Goodfriend. 2001. “What Assets
Should the Federal Reserve Buy?” Federal Reserve Bank of Richmond
Economic Quarterly 87, no. 1 (Winter): 7-22.
Fettig, David. 2002. “Lender of More Than Last Resort.” Federal Reserve
Bank of Minneapolis, December 1.
Geman, Ben. 2015. “Solyndra: What a Mess.” The Atlantic, August 26.

Klier, Thomas H., and James Rubenstein. 2012. “Detroit Back from the
Brink? Auto Industry Crisis and Restructuring, 2008-11.” Federal Reserve
Bank of Chicago Economic Perspectives 36, no. 2: 35-54.
Lacker, Jeffrey M. 2009. “Government Lending and Monetary Policy,”
Speech at the National Association for Business Economics, 2009 Washington Economic Policy Conference, Alexandria, Virginia, March 2.

Goodfriend, Marvin. 1994. “Why We Need an ‘Accord’ for Federal Reserve Credit Policy: A Note.” Journal of Money, Credit and Banking 26, no.
3 (August): 572-580.

Lacker, Jeffery M. 2020. “A Look Back at the Consensus Statement.” Cato
Journal 40, no. 2 (Spring/Summer): 285-319.

Goodfriend, Marvin. 2002. “Interest in Reserves and Monetary Policy.” Federal Reserve Bank of New York Economic Policy Review 8, no. 1
(May): 1-8.

Levin, Andrew, and John Taylor. 2013. “Falling Behind the Curve: A Positive Analysis of Stop-Start Monetary Policies and the Great Inflation.”
In The Great Inflation: The Rebirth of Modern Central Banking, edited by
M. Bordo and A. Orphanides, 217-244. Chicago: University of Chicago
Press.

Goodfriend, Marvin. 2009. “We Need an ‘Accord’ for Federal Reserve
Credit Policy.” Paper prepared for the Shadow Open Market Committee,
April 24.
Goodfriend, Marvin. 2011. “Fiscal Dimensions of Inflationist Monetary
Policy.” Paper prepared for the Shadow Open Market Committee, October 21.
Goodfriend, Marvin, and J. Alfred Broaddus Jr. 1996. “Foreign Exchange
266 |

Levy, Mickey D., and Charles Plosser. 2020. “The Murky Future of Monetary Policy.” Paper prepared for the Hoover Institution’s Monetary Policy
Conference, October 1.
Meltzer, Allan H. 2003. A History of the Federal Reserve, Volume 1 19131951. Chicago: University of Chicago Press.
Meltzer, Allan H. 2009. A History of the Federal Reserve, Volume 2, Book 1,
1951-1969. Chicago: University of Chicago Press.
| 267

Plosser
Plosser, Charles I. 2009a. “Ensuring Sound Monetary Policy in the Aftermath of Crisis.” Speech at the US Monetary Policy Forum, New York,
New York, February 27.
Plosser, Charles I. 2009b. “Improving Financial Stability.” Speech at the
Distinguished Speaker Series, University of Chicago Booth School of
Business, Chicago, Illinois, March 31.
Plosser, Charles I. 2009c. “Sound Monetary Policy for Good Times and
Bad.” Speech at Stanford Institute for Economic Policy Research, Stanford University, Palo Alto, California, October 20.
Plosser, Charles I. 2010. “The Federal Reserve System: Balancing Independence and Accountability.” Speech at the World Affairs Council of
Philadelphia, Philadelphia, Pennsylvania, February 17.
Plosser, Charles I. 2012. “Fiscal and Monetary Policy: Restoring the
Boundaries.” Speech at the U.S. Monetary Policy Forum, New York, New
York, February 24.

Federal Reserve Independence
Plosser, Charles I. 2019. “A Cautionary Note on Price-Level-Targeting.”
Defining Ideas, A Hoover Institution Journal, May 1.
Plosser, Charles I. 2020. “Operating Regimes and Fed Independence.”
Cato Journal 40, no. 2 (Spring/Summer): 361-371. Originally presented
at the 2019 Cato Institute’s 37th Annual Monetary Policy Conference.
Schwartz, Anna J. 1992. “The Misuse of the Fed’s Discount Window.”
Federal Reserve Bank of St. Louis Review 74, no. 5 (September/October): 58-69.
Selgin, George. 2018. “On Fiscally Neutral Monetary Policy.” Alt-M: Ideas
for an Alternative Monetary Future, November 27.
Swanson, Eric. 2011. “Let’s Twist Again: A High Frequency Event-study
Analysis of Operation Twist and Its Implications for QE2.” Brookings
Papers on Economic Activity 42, no. 1 (Spring): 151-207.
Thornton, Henry. (1802) 1965. An Inquiry into the Nature and Effects of
the Paper Credit of Great Britain. Reprint, New York: Augustus M. Kelley.

Plosser, Charles I. 2014. “A Limited Central Bank.” Cato Journal 34, no.
2 (Spring/Summer): 201-211. Originally presented at the 2013 Cato
Institute’s 31st Annual Monetary Policy Conference.
Plosser, Charles I. 2016a. “Argentina Redux?” Defining Ideas, A Hoover
Institution Journal, July 26.
Plosser, Charles I. 2016b. “Balancing Central Bank Independence and
Accountability.” In Central Bank Governance and Oversight Reform, edited by J. H Cochrane and J. B. Taylor, Chapter 6, part 1. Stanford, California: Hoover Institution Press.
Plosser, Charles I. 2017a. “The Risks of a Fed Balance Sheet Unconstrained by Monetary Policy,” In The Structural Foundations of Monetary
Policy, edited by M. D. Bordo, J. H. Cochrane, A. Seru, Chapter 1, section
1. Stanford, California: Hoover Institution Press.
Plosser, Charles I. 2017b. “Why the Fed Should Only Own Treasuries.”
Defining Ideas, A Hoover Institution Journal, June 10.
268 |

| 269

Industrial Development

Industrial Development and
Convergence
Sergio T. Rebelo and Pierre-Daniel G. Sarte
Marvin Goodfriend was an economist with broad interests. He is
well known for his contributions to macroeconomics and monetary
economics. But he was also an avid reader of economic history and a
student of the forces that drive economic growth.
Goodfriend wrote two papers on growth with his Brown University
classmate John McDermott: “Early Development” and “Industrial Development and the Convergence Question,” published in the American
Economic Review in 1995 and 1998, respectively.
Early Development (Goodfriend and McDermott, 1995) examines the
transition from a traditional, stagnant economy to a modern, growing
economy. This work is part of an extensive literature on poverty traps
and industrialization that includes Murphy and Shleifer (1989), Becker,
Murphy, and Tamura (1990), Galor and Weil (2000), and Lucas (2002).
Industrial Development (Goodfriend and McDermott, 1998) highlights the importance of familiarity with the technologies of other
countries in the returns to human capital accumulation and the
growth process. This work is part of a large literature on the connections between trade and growth.1
In this essay, we describe the Industrial Development model and
ponder its lessons for contemporary policy discussions.

Model structure
Goodfriend and McDermott (1998) develop a model in which human capital accumulation is the engine of growth, as in Lucas (1988).

1

270 |

Grossman and Helpman (1991, 2015) and Eaton and Kortum (2001).

| 271

Rebelo and Sarte
The novel dimension is an interaction between the size of the population and the technology for human capital accumulation.
Final output is produced by combining efficiency units of labor with
a continuum of differentiated goods produced by monopolists. There
is free entry into the production of differentiated goods, so profits are
zero in equilibrium. But, unlike Romer (1990), Goodfriend and McDermott assume in both of their papers that firms do not have to incur
research and development costs to produce a new differentiated good.
The key mechanism is that a larger population results in higher
demand for differentiated goods. The prospects of higher sales and
profits induce entry so that a rise in population increases the number
of differentiated-good producers. In turn, the number of differentiated goods produced has a positive impact on the ability of workers to
accumulate human capital. Goodfriend and McDermott (1998) assume
that this driver of human capital accumulation depends on how many
goods are produced at home and how many of the goods produced
abroad are familiar to domestic agents.
Ultimately, the setup is one in which countries that are open to the
rest of the world, and familiar with technologies used or developed
elsewhere, find it easier to accumulate know-how. As a result, these
countries develop at a rate that allows them to catch up and even
overtake the per capita output levels of more advanced trading partners. By contrast, countries that are less inclined to promote openness,
and are less familiar with technologies used elsewhere, fall persistently
behind the leading countries and potentially never catch up.

Context and contributions
In thinking about the broader arch of economic history, from the eve
of the Industrial Revolution to recent times, Goodfriend and McDermott (1998) sought to identify key mechanisms underlying economic
development that would not necessarily be evident over a shorter
time period or in an individual country. They also recognized that these
mechanisms needed to be consistent with periods of both divergence
and convergence across countries during the process of world growth.
272 |

Industrial Development
Goodfriend and McDermott (1998) isolates one of the potential
ingredients responsible for the Japanese, South Korean, and Chinese
growth miracles. In the context of their model, the familiarity of firms
and workers in these economies with goods produced in the US increased the returns to human capital accumulation, fostering a period
of fast growth. The concept of familiarity with foreign economies, in
particular, which plays a central role in Goodfriend and McDermott
(1998), now seems prescient given the explosion of information-sharing that the information technology revolution was about to make
possible.
The Goodfriend and McDermott (1998) model can be used to think
about some grand experiments implemented in the 20th century:
Japan, South Korea, and other Asian countries chose export promotion
as their development strategy in the 1970s and, starting in 1979, China
also went this route. By contrast, India, Latin America, and Africa chose
import substitution as their development policy.
Import substitution had the appeal of combining economic growth
with political independence from the US and other large economies.
In particular, Latin American countries hoped that import substitution
would be a solution to the problem of being, in the words of the Mexican president Porfirio Dias, “too far from God, too close to the United
States.” After becoming independent in 1947, India hoped that import
substitution would allow it to achieve “swadeshi,” i.e., a genuine self-reliance that would make India independent of both communist and
capitalist countries.
Import substitution produced some early wins. In 1954, Hindustan
Motors, a state-owned Indian company, started producing automobiles. Imposing high tariffs on foreign goods and fostering the development of a domestic industry helped create a fast-growth decade in
Brazil during the 1970s.
But these successes proved ephemeral. The Ambassador, the car
first produced in India in 1954, remained almost unchanged for 50
years. Import substitution turned Brazil into one of the world’s most
closed economies. By protecting its computer industry, Brazil made
| 273

Rebelo and Sarte
the domestic price of computers much higher than the international
price,2 reducing the productivity benefits from adopting information
technologies in a wide range of industries. The success of Brazil in the
1970s was followed by decades of stagnation that lent credence to the
famous quip by the writer Stefan Zweig: “Brazil is the land of the future
and always will be.”
There are many aspects of import substitution regimes that might
have contributed to their poor performance. The extensive government involvement in the economy gave entrepreneurs an incentive to
invest in political connections rather than learning new technologies
or adopting from abroad. At the same time, the protection from competition reduced incentives to be efficient and invest in new technologies.
Goodfriend and McDermott (1998) add two important factors to
this standard list of the perils of import substitution. The first factor is
scale. In many countries, the internal market is too small. According
to Scitovsky (1969), the 1960s saw 90 Latin American companies in
eight countries producing cars and trucks with a combined volume
of about 600,000 vehicles. Yet, he documents that during this period,
the efficient scale for an automobile plant was estimated to require a
minimal annual production of 250,000 vehicles. In the Goodfriend and
McDermott (1998) model, scale matters indirectly: it affects the returns
to human capital accumulation and the associated growth prospects.
In earlier work, Goodfriend and McDermott (1995) argued that a
large enough scale, achieved through population growth, was key to
increased specialization, which eventually raised learning productivity
enough to kick start a period of self-sustaining technological progress.
The second factor highlighted by Goodfriend and McDermott is
familiarity with foreign goods and technologies. Countries that close
their doors to trade also shut their doors to foreign technologies. In the
Goodfriend and McDermott (1998) model, low familiarity with foreign
goods depresses the returns to human capital accumulation and
2

274 |

Industrial Development
dims growth prospects. Here, familiarity is envisioned as stemming in
part from active commercial relations, but also from a more pervasive
knowledge of foreign cultures and commodities.
Reading this paper brings to mind themes that are very relevant today. Just as the information technology revolution was about to hit its
stride, Goodfriend and McDermott (1998) wrote that “... it is easy to see
that familiarity would grow over time with the economy. Nations become more familiar with each other as technological progress lowers
transports costs and communication costs.” After achieving remarkable
success with export promotion, China is now turning to import substitution with its campaign “Made in China 2025.” Goodfriend and McDermott’s (1998) model predicts that this choice will penalize China’s
growth prospects. But will progress in information technology allow
China to stay inside its great walls and maintain familiarity with the
technological world outside?
As it turns out, forces that contributed to familiarity with the outside
world also brought to the fore new vulnerabilities such as cybercrime
and intellectual property theft. Intellectual property products, in
particular, are now a key component of investment goods in the US
National Income Accounts following, in part, a reclassification of industries in 1999 and again in 2013. Therefore, will the ease with which
information can be obtained increase the incentive to accumulate human capital in countries that are still relatively closed to foreign trade?
In Goodfriend and McDermott (1998), world growth is maximized
when all countries are perfectly familiar with each other. Would they
still see the world in this way 25 years later?
These are fascinating topics that can be discussed in the context of
the Goodfriend and McDermott (1998) model. We only wish that Marvin could be here to debate them with us, as he would have undoubtedly done enthusiastically.

Luzio and Greenstein (1995).

| 275

Rebelo and Sarte

References
Becker, G.S., K.M. Murphy, and R. Tamura. 1990. “Human Capital, Fertility, and Economic Growth.” Journal of Political Economy 98, no. 5, part 2
(October): S12-S37.
Eaton, Jonathan, and Samuel Kortum. 2001. “Technology, Trade, and
Growth: A Unified Framework.” European Economic Review 45, no. 4-6
(May): 742-755.
Galor, Oded, and David N. Weil. 2000. “Population, Technology, and
Growth: From Malthusian Stagnation to the Demographic Transition
and Beyond.” American Economic Review 90, no. 4 (September): 806828.

Industrial Development
Murphy, Kevin M., Andrei Shleifer, and Robert W. Vishny. 1989. “Industrialization and the Big Push.” Journal of Political Economy 97, no. 5
(October): 1003-1026.
Romer, Paul M. 1990. “Endogenous Technological Change.” Journal of
Political Economy 98, no. 5, Part 2 (October): S71-S102.
Scitovsky, Tibor. 1969. The Process of Industrialization in Latin America:
Round Table, Inter-American Development Bank. Guatemala:
Inter-American Development Bank.

Goodfriend, Marvin, and John McDermott. 1995. “Early Development.”
American Economic Review 85, no. 1 (March): 116-133.
Goodfriend, Marvin, and John McDermott. 1998. “Industrial Development and the Convergence Question.” American Economic Review 88,
no. 5 (December): 1277-1289.
Grossman, Gene M., and Elhanan Helpman. 1991. “Trade, Knowledge
Spillovers, and Growth.” European Economic Review 35, no. 2-3 (April):
517-526.
Grossman, Gene M., and Elhanan Helpman. 2015. “Globalization and
Growth.” American Economic Review 105, no. 5 (May): 100-104.
Lucas Jr., Robert E. 1988. “On the Mechanics of Economic Development.” Journal of Monetary Economics 22, no. 1 (July): 3-42.
Lucas Jr., Robert E. 2002. “The Industrial Revolution: Past and Future.” In
Lectures on Economic Growth, 109-188. Cambridge: Harvard University
Press.
Luzio, Eduardo, and Shane Greenstein. 1995. “Measuring the Performance of a Protected Infant Industry: The Case of Brazilian Microcomputers.” Review of Economics and Statistics 77, no. 4. (November):
622-633.
276 |

| 277

Volcker's Disinflation

Rational Expectations and Volcker’s
Disinflation
Thomas J. Sargent
I am grateful to the authors of Goodfriend and King (2005) for letting
me watch two masters of modern macroeconomics at work as they
marshal both quantitative and narrative evidence and then carefully
choose simple, but not too simple, theoretical tools to interpret the
evidence. I lived through the turbulent times they describe, and their
account rings true, as I tell at the end of this essay. I write this with
mixed feelings because the theory that they use with so much success
abandons the rational expectations equilibrium concept that I love so.1
People inside a rational expectations model share a unique statistical model. For given parameters that describe technologies, preferences, and information flows, there is typically a manifold of rational
expectations equilibria that are indexed by distinct budget-feasible
government policies. Credible government plans are special rational
expectations equilibria in which government policies are chosen by
a sequence of policymakers who always choose to continue a possibly history-dependent plan.2 A government strategy is a sequence of
history-dependent functions that map a history at time t into government actions at time t. A government strategy is effectively a sequence
of conditional distributions for future actions that government decision-makers choose to confirm. In models of credible government policies, government decision rules thus play two roles, one as decision
rules for policymakers and another as the public’s forecasting functions.
The “communism of statistical models” that prevails within a rational
expectations equilibrium and the subgame perfection that prevails
 axter (1985) and Ball (1994, 1995) also either refined or abandoned rational expectaB
tions equilibria in order to understand episodes in which the credibility of fiscal-monetary policies was the focus of public discussions.
2
Chang (1998) provides sharp definitions of a manifold of rational expectations equilibria and of credible government policies within an elementary monetary model.
1

278 |

| 279

Sargent
within a model of credible government policy have the consequence
that it is subtle, if not impossible, to tell who chooses a government
policy — government decision-makers or private forecasters. In a
rational expectations model of credible public policy, the conditional
probability distributions that the public uses to forecast are the very
same ones that government policymakers want to confirm.3 In such an
equilibrium, a Federal Open Markets Committee (FOMC) would never
want statistically to disappoint or surprise the market. Such an FOMC
feels no urge to “acquire credibility.”4

Volcker's Disinflation
lower long-term nominal interest rates by permanently lowering inflation rates.5 Disconnecting the market’s forecasts from those of policymakers is impossible in a rational expectations equilibrium. These
considerations led Goodfriend and King to abandon a rational expectations equilibrium concept in creating their model of a central banker
striving to “acquire credibility.” Thus, Goodfriend and King (2005, p. 34)
summarized their paper in this way:
	In contrast [to what goes on in a rational expectations model], during
the Volcker disinflation the Fed needed to acquire credibility for low
and stable inflation. We studied this episode without having a firm
understanding of Fed behavior, so instead we adopted an analytical
strategy that focused on the interplay between inflation, expected
inflation, credibility and real activity without specifying the monetary policy rule. We sought to document how the Volcker FOMC
tried to acquire credibility: with an initial appeal to monetary targets
as a nominal anchor, with new operating procedures designed to
allow greater scope for short-term interest rates to be determined
by market forces, and ultimately by employing an interest rate and
reserve aggregate policy mix to work the actual inflation rate down.
Our methodology for studying the disinflation without a firm understanding of the Fed’s behavioral rule places us in a position similar to
the public and the FOMC itself. To improve our understanding of the
Volcker disinflation, it will be necessary to specify Fed behavior explicitly and to model the interaction of Fed policy with the dynamics
of private sector beliefs about inflation. Requiring these beliefs to be
consistent with the financial market data will allow a clearer understanding of the role of imperfect credibility in the Volcker disinflation.

The authors of Goodfriend and King (2005) were accomplished
architects of rational expectations models. But for good reasons, they
chose not to interpret the Volcker disinflation by constructing a rational expectations model. Their thorough readings of FOMC transcripts
and other sources left Goodfriend and King (2005) without a coherent
description of an FOMC decision rule or evidence that the Fed thought
systematically about designing one. It described disagreements and
confusions about macroeconomic structures among FOMC members.
It documented FOMC concerns that the market’s expectations about
inflation and other outcomes differed systematically and persistently
from FOMC targets. Goodfriend and King spotted “inflation scares” in
high long-term interest rates that had disappointed the FOMC’s intention that by pushing short-term nominal interest rates up it could
3
4

280 |

See Chari and Kehoe (1990) and Stokey (1991).
T his statement is an accurate description of a large class of plain vanilla rational
expectations models in the Ramsey plan tradition and its extension to models of
credible policies by Chari and Kehoe (1990) and Stokey (1991) as refined by Bassetto
(2005). For example, see Chang (1998). These models have a single type of policymaker who, together with private agents inside the model, all trust the same single
statistical model. It is a less accurate description of rational expectations models that
like Backus and Driffill (1985a,b) and Lu et al. (2008, 2016) posit multiple types of
policymakers who have different objective functions. In these models, a policymaker
has an incentive to pretend to be another type. The claim in the text is also dubious in
rational expectations models in which policymakers have multiple statistical models
about which they are uncertain, so that what constitutes “the unique model” for a rational expectations equilibrium concept is possibly what a Bayesian statistician would
dub a “hierarchical model” that involves Bayesian model averaging over the distinct
statistical models.

Goodfriend and King’s model combines an artfully parameterized
inflation “forecast credibility gap” with an expectational Phillips curve
and a Fisher equation. By intentionally taking its “forecast credibility
5

S ee Goodfriend (1993) for a definition of an inflation scare and a technique for diagnosing one.

| 281

Sargent
gap” equation as a primitive and intentionally not providing “microfoundations” for it, it got a convenient tool for precisely defining and
quantifying a credibility gap. The paper calibrated its model to do a
good job of approximating inflation, unemployment, and long- and
short-term interest rate paths under the Volcker-led FOMC and inferred
private sector beliefs about prospects for inflation. Its concluding
section, part of which I just quoted, called for more work to learn about
the imperfect credibility that challenged Volcker’s FOMC and our assessment of how well the FOMC had coped with it.
I admire Goodfriend and King’s thoughtful marshalling of the
evidence that led to their sparsely parameterized non-rational expectations model as well as the road map it provides for further work. In
the spirit of its concluding section, I mention just two routes opened
up by its abandoning of rational expectations, each of which involves
belief heterogeneities and model uncertainties. The first, exemplified
by the Bayesian model-averaging setup of Cogley and Sargent (2005),
acknowledges that in the 1970s and 1980s neither the academic
macroeconomic community nor the Federal Reserve staff nor FOMC
members had settled on a macroeconomic model. Competing views
about dynamic trade-offs between inflation and unemployment were
embedded in alternative conceptions of a Phillips curve. Cogley and
Sargent describe a setting in which part of the FOMC’s information is
a Bayesian posterior over three distinct Phillips curves, each of whose
coefficients are themselves updated as data accrue. The FOMC’s initial
1960 prior puts almost all probability on a Samuelson-Solow Phillips
curve and very little on a Lucas rational expectations version of an
expectational Phillips curve.6 An intertemporal objective function tells
the FOMC to pay attention not only to posterior probabilities attached
to each model, but also to the continuation values implied by each of
them. Then a peculiar thing happens. Even though observations gathered during the 1970s tell the policymaker to put most weight on the
Lucas model, the updated Samuelson-Solow model tells the policymaker that very bad continuation values would be associated with the
rapid inflation stabilization policy that the Lucas model recommends.
6

282 |

Goodfriend and King’s model includes such a Phillips curve.

Volcker's Disinflation
It is only after enough data caused updated coefficients of the Samuelson-Solow model to imply that less adverse outcomes would occur
under the recommended actions from the Lucas model that the FOMC
ultimately decided to sharply reduce inflation toward zero. Thus, posterior probabilities attached to the three models are not all that matters.
When they predict sufficiently adverse consequences of following
advice delivered by higher posterior probability models, models with
low posterior probabilities can derail recommendations that come
from higher posterior models. Consequently, the FOMC will decide
to stabilize inflation only after estimated coefficients of the Samuelson-Solow model have adapted to imply low enough inflation-unemployment “sacrifice ratios.” Note how this special type of “model averaging” recommendation system gives special weight to models that
set off Cassandra warnings like those proclaimed by Arthur Okun (see
Goodfriend and King [2005, pp. 982-83]) and many others.
“Expectations management” is a second research agenda opened
by Goodfriend and King’s decision not to use a rational expectations
model. To make progress on this topic requires a setting in which, first,
private agents and the government have different beliefs; second,
the government has a model of how its actions affect private agents’
beliefs; and third, discrepancies of beliefs between government and
private agents can be rationalized. Filling all three of these requirements simultaneously is a tall order. Karantounias (2013, 2018) offers
a promising approach based on a multi-agent application of robust
control theory. In his setting, a representative agent and a Ramsey
planner share a common approximating statistical model, just as they
do in a rational expectations model. But now one or both of them
distrusts the approximating model. In Karantounias (2013), the Ramsey planner completely trusts the approximating model, but private
agents don’t; the Ramsey planner knows this. Because private agents
choose policies that are best responses not to the shared approximating model, but to their worst-case model, an object that is affected by
the Ramsey planner’s policy, the Ramsey planner is thrust into manipulating private agents’ beliefs. In a similar vein, Presno and Orlik (2016)
study expectations management as components of credible government policies in a Chang (1998) monetary environment that they alter
| 283

Sargent

Volcker's Disinflation

by having private agents manage their distrust of an approximating
model by using techniques from robust control theory.7

asked the other academics present to comment on the Fed staff report.

Goodfriend and King’s story rings true to me as an eyewitness to
events they describe so well. In June 1976, I attended a meeting of a
group of “academic consultants” at the Federal Reserve Board in Washington, DC. I had been invited by George Leland Bach, the organizer of
the meeting, at the urging of fellow attendee Milton Friedman. I think I
was invited because Milton Friedman had told Lee Bach that I was one
of “those crazy economists up in Minnesota” who talked about macro
models in which the FOMC is just a decision rule that maps its information into its actions, a rule that everyone inside and outside a macroeconomic model knows.8 Arthur Burns chaired the meeting.

Before my turn, Governor Henry Wallich said, “Mr. Sargent, your tribe
always talks about monetary policy rules and rational expectations.
OK. Please tell me what you think our rule is.” I answered, “I can’t tell.”
Goodfriend and King’s careful sifting of more documentary evidence
than I certainly had at that time indicates that, like me at the time,
they could not fathom a coherently thought-out FOMC decision rule.
The gap between actions at the FOMC and how we “crazy Minnesota
economists”9 thought choices should be framed did not reflect well on
the FOMC.10

Academics were supposed to discuss a report the Fed staff had
prepared. Lead speakers for the academics were Milton Friedman and
Arthur Okun, representing contending “monetarist” and “Keynesian”
perspectives. Friedman spoke first and offered a scholarly discussion of
the Fed staff report, drilling down especially on two or three footnotes.
Then Arthur Okun brought out fireworks based on his passionate
belief in sacrifice ratios like those summarized by Goodfriend and King
(2005, pp. 982-83). Okun did not mention the Fed staff report. Instead,
he lambasted the FOMC for the crusade that he said it was now pursuing to stamp out inflation quickly, while foolishly ignoring what he
said were big social costs in terms of GDP and unemployment. He said
that the Fed had decided to do that on its own authority against what
Okun understood to be the preferences of the public and Congress.
Okun concluded by warning Burns and the other governors that if they
chose to persist in so abusing their independence, they would have
nobody to blame but themselves if Congress were soon to take away
their independence. After moments of silence as he puffed his pipe
while looking straight at Okun, Chairman Burns said, “Would the nextspeaker please confine himself to economics?” Burns then sequentially
 obert King and others have rolled up their sleeves and worked on this problem usR
ing a quite different approach than I describe here, and one that I think is very fruitful.
See King et al. (2008) and Lu et al. (2016).
8
According to Silber (2012, p. 176), the quotes indicate what Paul Volcker thought of
those of us then working in Minnesota.
7

284 |

9

 nd of course also at Chicago and Carnegie-Mellon and Rochester.
A
For an extended account of Volcker’s struggle permanently to lower US inflation that
mostly agrees with Goodfriend and King’s, see Silber (2012, Part III). Silber recently
offered an ominous comparison of discrepancies between current bond traders’ market-revealed inflation forecasts and those of monetary policy analysts like himself
and the opposite sign of such discrepancies that Volcker confronted. Silber fears that
a “credibility gap” of opposite sign now threatens us. See Silber, William L., “Why Last
Week’s Higher Inflation Left Bond Yields Unchanged.” LinkedIn post, July 18, 2021,
available at https://www.linkedin.com/in/william-silber-0a854b158/detail/recentactivity/shares/.

10

| 285

Sargent

References
Backus, David, and John Driffill. 1985a. “Inflation and Reputation.”
American Economic Review 75, no. 3 (June): 530–38.
Backus, David, and John Driffill. 1985b. “Rational Expectations and
Policy Credibility Following a Change in Regime.” Review of Economic
Studies 52, no. 2 (April): 211–21.
Ball, Laurence. 1994. “Credible Disinflation with Staggered PriceSetting.” American Economic Review 84, no. 1 (March): 282–89.

Volcker's Disinflation
Karantounias, Anastasios G. 2018. “Optimal Fiscal Policy with Recursive
Preferences.” Review of Economic Studies 85, no. 4 (October): 2283–2317.
King, Robert G., Yang K. Lu, and Ernesto Pasten. 2008. “Managing Expectations.” Journal of Money, Credit, and Banking 40, no. 8 (December):
1625–666.
Lu, Yang K., Robert G. King, and Ernesto Pasten. 2016. “Optimal Reputation Building in the New Keynesian Model.” Journal of Monetary Economics 84 (December): 233–49.

Ball, Laurence. 1995. “Disinflation with Imperfect Credibility.” Journal of
Monetary Economics 35, no. 1 (February): 5–23.

Presno, Ignacio, and Anna Orlik. 2016. “On Credible Monetary Policies
under Model Uncertainty.” Society for Economic Dynamics 2016 Meeting Papers 1280.

Bassetto, Marco. 2005. “Equilibrium and Government Commitment.”
Journal of Economic Theory 124, no. 1 (September): 79–105.

Silber, William L. 2012. Volcker: The Triumph of Persistence. New York,
London, New Delhi, Sydney: Bloomsbury Press.

Baxter, Marianne. 1985. “The role of Expectations in Stabilization Policy.” Journal of Monetary Economics 15, no. 3 (May): 343–62.

Stokey, Nancy L. 1991. “Credible Public Policy.” Journal of Economic
Dynamics and Control 15, no. 4 (October): 627–56.

Chang, Roberto. 1998. “Credible Monetary Policy in an Infinite Horizon
Model: Recursive Approaches.” Journal of Economic Theory 81, no. 2
(August): 431–61.
Chari, V. V., and Patrick J. Kehoe. 1990. “Sustainable Plans.” Journal of
Political Economy 98, no. 4 (August): 783–802.
Cogley, Timothy, and Thomas J. Sargent. 2005. “The Conquest of US
Inflation: Learning and Robustness to Model Uncertainty.” Review of
Economic Dynamics 8, no. 2 (April): 528–63.
Goodfriend, Marvin. 1993. “Interest Rate Policy and the Inflation Scare
Problem: 1979-1992.” Economic Quarterly 79, no. 1 (Winter): 1–24.
Goodfriend, Marvin, and Robert G. King. 2005. “The Incredible Volcker
Disinflation.” Journal of Monetary Economics 52, no. 5 (July): 981–1015.
Karantounias, Anastasios G. 2013. “Managing pessimistic expectations
and fiscal policy.” Theoretical Economics 8, no. 1 (January): 193–231.
286 |

| 287

Monetary Mystique

Monetary Mystique and the
Fed’s Path Toward Increased
Transparency1
Lars E.O. Svensson
Marvin Goodfriend’s paper “Monetary Mystique: Secrecy and Central Banking”2 is a masterpiece: It is an extremely well-written, meticulous, and fair analysis and critique of the Federal Reserve’s defense
of its practice of secrecy in monetary policy and central banking. His
critique was devastating, and he completely demolished the Federal
Open Market Committee’s (FOMC) arguments in the most precise and
convincing way. Nevertheless, it took the Fed many years to reach the
current standards of transparency and accountability in monetary
policy.

The paper and its background
When Marvin wrote and published his paper, central banking was
generally cloaked in mystery. In a much-quoted paragraph, Karl Brunner had written in 1981:
	Central Banking [has been] traditionally surrounded by a peculiar and protective political mystique. . . The mystique thrives
on a pervasive impression that Central Banking is an esoteric
art. Access to this art and its proper execution is confined to
the initiated elite. The esoteric nature of the art is moreover
revealed by an inherent impossibility to articulate its insights
in explicit and intelligible words and sentences.3

I thank Alan Blinder for help with the chronology of the Fed’s path toward increasing
transparency and Robert King and Alexander Wolman for helpful comments. Support
from the Jan Wallander and Tom Hedelius research foundation and the Tore Browaldh
research foundation is gratefully acknowledged. Views expressed and any errors are
my own.
2
Goodfriend (1986).
3
Brunner (1981), p. 5; Goodfriend (1986) provides a longer quote.
1

288 |

| 289

Svensson
A specific background to Marvin’s paper was that in March 1975, the
FOMC of the Federal Reserve System was sued under the Freedom of
Information Act to make public, immediately following each FOMC
meeting, the policy directive and minutes for that meeting. At the
time, the policy directives were available to the public 90 days after
their decision. In response to the suit, the FOMC did shorten the publication lag to 45 days, which was a few days after the next regularly
scheduled meeting. But the Committee stated that it was not prepared
to disclose policy actions and minutes immediately after an FOMC
meeting.4 The suit went to court. After six years of court proceedings,
including a hearing before the US Supreme Court, the case was decided in 1981 in favor of the FOMC.
Nevertheless, even though the FOMC won the legal case, it lost the
intellectual case. The suit forced the Federal Reserve, for the first time,
to provide a detailed written defense of its secrecy. The court records
were made public. This allowed Marvin to summarize and evaluate the
FOMC’s arguments for continued secrecy. He collected the arguments
found in the FOMC’s affidavits into five categories:
1.	
Unfair speculation: Only the large speculator is in a position to
benefit from disclosure of the current policy directive.
2.

I nappropriate market reaction: Current disclosure would cause
the market to overreact or to react contrary to the intention of
the FOMC; in general, market reaction would be more difficult
to predict with current disclosure.

3.

 arm to the government’s commercial interest: Current discloH
sure would cause market reactions that would raise the cost of
marketing US Treasury debt and make open market operations
more costly.

4.	
Undesirable precommitment: The FOMC does not wish to precommit its future policy actions, and current disclosure of the
policy directive would tend to precommit the FOMC.
4

290 |

T he FOMC also discontinued its Memoranda of Discussion, the detailed written minutes, removing them as an issue in the case (Goodfriend 1986, footnote 13).

Monetary Mystique
5.	
More difficult interest rate smoothing: Current disclosure would
make it more difficult for the FOMC to smooth interest rates.
The FOMC’s arguments are laughable by today’s standards. They
were probably laughable to Marvin at the time, but he took them very
seriously. He meticulously and fairly evaluated each category of arguments in the light of existing economic theory and “recent theoretical
work related to the secrecy issue.” And he ended up completely demolishing the arguments in the most precise and effective way.
In the abstract of the paper, Marvin wrote (my emphasis): “The discussion highlights a number of potential benefits and costs of central
bank secrecy, and identifies some conditions under which secrecy could
be socially beneficial.” In the summary at the end of the article, Marvin
furthermore wrote (my emphasis):
	[My critique of the FOMC’s defense], based heavily on rational expectations reasoning, supported some FOMC contentions and pointed out
some theoretical weaknesses in others. In order to investigate the
secrecy issue further, theoretical papers related to the secrecy issue
were reviewed. The discussion highlighted a number of potential
benefits and costs of central bank secrecy, and identified some conditions under which secrecy could be socially desirable. At best, however,
given the inconclusiveness of the theoretical arguments and the presumption that government secrecy is inconsistent with the healthy
functioning of a democracy, further work is required to demonstrate
that central bank secrecy is socially beneficial. (p. 90.)

However, as far as I can see in Marvin’s main text, he did not find any
reasonable conditions under which secrecy would be socially beneficial
and some FOMC contentions would be supported. The abstract and
summary seems to have been written in order to somewhat hide the
true force of the critique in the main text. If so, this is understandable.
That Marvin wrote and published this paper must be seen as an act of
considerable courage. Marvin was a vice president at the Richmond
Fed at the time. That an insider of the Fed, a vice president of a Federal
Reserve Bank, publicly criticized — even implicitly ridiculed — a major
FOMC position was not a trivial thing.
| 291

Svensson
As far as I know, Marvin had the full support of his boss, J. Alfred
Broaddus, who had been appointed director of Research and senior
vice president at the Richmond Fed in 1985. Broaddus presumably also
had the support of his boss, President Robert P. Black. But the paper
must have been very unpopular with the FOMC and the chairman of
the Federal Reserve Board in 1986. It seems that one could easily have
anticipated then that the paper could have negative consequences for
Marvin’s career in the Federal Reserve System.

Monetary Mystique

The Federal Reserve’s path to increased transparency
It took quite a few years for the Fed to become more transparent and
less secret, in what Yellen (2012) called a “recent revolution and continuous evolution” and Blinder (2020) called a “slow-motion revolution.”
In February 1994, immediately after the FOMC meeting, the committee surprised Fed watchers and media by issuing a statement under
Greenspan’s name:8
	Chairman Alan Greenspan announced today that the Federal Open
Market Committee decided to increase slightly the degree of pressure on reserve positions. The action is expected to be associated
with a small increase in short-term money market interest rates.

Paul Volcker, who was chair until August 1987, was hardly a friend of
transparency. He “liked to blow smoke — both literally and figuratively
— in his congressional testimonies.”5 Neither was the new chair, Alan
Greenspan. He boasted that he had learned to “mumble with great incoherence,” and famously told an audience that “if you think what I said
was clear and unmistakable, I can assure you you’ve probably misunderstood me.”6 More specifically, in 1989 — three years after the publication of Marvin’s paper — an apparently unconvinced Greenspan
vigorously argued in Congress against immediate announcements of
changes in the federal funds target:
	The immediate disclosure of any changes in our operating targets
would make this information available more quickly to all who were
interested, but it would have costs. Simply put, this provision would
take a valuable policy instrument away from us. It would reduce
our flexibility to implement decisions quietly at times to achieve a
desired effect while minimizing possible financial market disruptions.
Currently, we can choose to make changes either quite publicly or
more subtly, as conditions warrant. With an obligation to announce
all changes as they occurred, this distinction would evaporate; all
moves would be accompanied by announcement effects akin to
those currently associated with discount rate changes.7
 linder (2020).
B
Blinder et al. (2001).
7
Greenspan (1989), pp. 14–15.

	The decision was taken to move toward a less accommodative stance
in monetary policy in order to sustain and enhance the economic
expansion.
	Chairman Greenspan decided to announce this action immediately
so as to avoid any misunderstanding of the Committee’s purposes,
given the fact that this is the first firming of reserve market conditions by the Committee since early 1989.

After this, there were announcements immediately after the FOMC
meetings at which the federal funds rate was changed.
Thus, eight years after the publication of Marvin’s paper, the Fed
gave up its strong resistance toward disclosing its policy decisions
immediately after each meeting. The Fed thereby implicitly acknowledged that Marvin had been right. Marvin had completely won the
case.9
However, this was not the end of the Fed’s steps toward increased
transparency. As noted by Blinder (2020, p. 43), in August 1994, the Fed
offered its first bit of forward guidance — although that term

5
6

292 |

8
9

F OMC (1994).
Broaddus would be appointed president of the Richmond Fed in 1993, and Marvin
was the same year promoted to senior vice president and director of Research. Ex
post, neither Marvin’s nor Broaddus’s career seems to have suffered.

| 293

Svensson
would not appear until years later: In announcing that it was raising
both the discount rate and the federal funds rate by 50 basis points,
rather than the expected 25, the committee added that “these actions
are expected to be sufficient, at least for a time, to meet the objective
of sustained, noninflationary growth,” where “at least for some time”
was a new phrase.
More importantly, in 1999, the FOMC started to issue a short explanatory statement also when the interest rate was not changed. The
FOMC also started to reveal whether there was a “bias” toward future
tightening, easing, or neither in its thinking about the near-term future
of interest rates, that is, effectively a near-term forward guidance.10
During Greenspan’s time, a more specific much-noted forward guidance appeared in the August 2003 statement and subsequently:
	The Committee judges that, on balance, the risk of inflation becoming undesirably low is likely to be the predominant concern for the
foreseeable future. In these circumstances, the Committee believes
that policy accommodation can be maintained for a considerable
period.11 12

There were several further steps toward more information and transparency with Ben Bernanke as chairman (see Blinder, 2020, and Yellen,
2012, for details). In November 2007, the first “Summary of Economic Projections” (SEP) — a summary of FOMC participants’ individual
projections of GDP, unemployment, and inflation produced four times
a year — was included in the minutes published three weeks after the
October 2007 FOMC meeting. Any individual projections of the federal
funds rate were not included at this time. After the federal funds rate
hit its perceived effective lower bound (0–25 basis points) in December
2008, the statement was amended in 2009 to include decisions on asset purchases and more explicit forward guidance for the interest rate,
 linder (2020).
B
FOMC (2003a), my emphasis.
12
The expression “for a considerable period” had first appeared in Greenspan’s July
2003 testimony before Congress (Greenspan 2003). There is some interesting, lively,
and perhaps even tense discussion in the August 2003 FOMC meeting about whether or not to include that sentence in the FOMC statement (FOMC 2003b, pp. 86–95).
(I am grateful to Alexander Wolman for alerting me to this.)

Monetary Mystique
first in the form of initial qualitative language similar to the 2003 language: “for some time,” then “for an extended period,” which remained
until August 2011, when a specific date was first mentioned. This “calendar-based” forward guidance was later changed to be “data-based.”
In April 2011, the first press conference after a meeting was held. Then
an advance version of the SEP table on the ranges and central tendencies of the participants’ projections was released.
In January 2012, there were two major steps. First, the “Statement on
Longer-Run Goals and Monetary Policy Strategy” was issued. It was an
extremely well-written and concise summary of the Federal Reserve’s
goals and strategy. It has since been reaffirmed by the FOMC every
year in January.13 The statement was the result of a subcommittee
chaired by then Vice Chair Janet Yellen (see Yellen, 2012, for details
and the case for transparency). Second, the SEP was amended to also
include the individual FOMC participants’ projections of the federal
funds rate, the “dot plot” of individual participants’ assessment of the
appropriate policy-rate settings over the next three years and the long
run.14
With these two steps, the Federal Reserve had arguably become a
full-fledged flexible inflation targeter, including doing “forecast-targeting” (see Svensson, 2020b, for details). The Fed’s loss function was
well understood, and the “balanced approach” indicated equal weight
on stabilizing inflation around the inflation target and employment
around its estimate of full employment.15 It was publishing projections
of its target variables, inflation and employment (unemployment), and
of its main instrument, the federal funds rate.
To be precise, the SEP includes the FOMC participants’ individual projections before the meeting; it is not the result of a joint decision about
a joint projection at the meeting. In particular, the projections are not
linked together: the SEP shows the distributions for each variable but

10
11

294 |

F OMC (2012b).
FOMC (2012a).
15
Yellen (2020).
13
14

| 295

Svensson
not the joint distribution (no table with numbered rows for participants and columns for variables, even unnamed). For that, one has to
wait five years until the minutes and SEPs amended with such unnamed tables are released. Nevertheless, the SEPs are quite informative: they provide forward guidance about the participants’ appropriate policy paths, reveal the participants’ long-run forecasts (including
the neutral unemployment and policy rates), and help to hold the
FOMC accountable for achieving its goals.16 Perhaps the median projections are not so different from what a vote on the projections would
result in?17

The new monetary policy strategy
In August 2020, the Federal Reserve announced a revision of its
monetary policy strategy and released a new “Statement on Longer-Run Goals and Monetary Policy Strategy,”18 a result of the strategy
review that it had initiated in 2019.19 With regard to the maximum
employment mandate, the FOMC now seeks over time to mitigate
“shortfalls” of employment from its maximum level, not “deviations.”
This means that a low unemployment rate by itself, unless accompanied by signs of unwanted increases in inflation, will not justify a policy
tightening. Focusing on shortfalls of employment instead of deviations
introduces an asymmetry in the maximum employment mandate, and
the statement drops previous language about “a balanced approach.”
S vensson (2020a,b).
A decision-making process whereby the FOMC arrives at an explicitly joint policy-rate path and corresponding inflation and unemployment forecasts would
be more consistent with forecast targeting. As discussed in Svensson (2020b, pp.
81–82), the FOMC has undertaken some experiments in constructing a consensus
policy-rate path and forecasts of inflation and unemployment. They are discussed in
some detail under the heading “Experimental Consensus Forecast” in the October
2012 transcripts, FOMC (2012c, pp. 201–79). There were several difficulties noted
about constructing consensus forecasts, including that the policymaking environment was unusually complex with both unconventional portfolio actions and forward guidance being important policy tools. In view of these difficulties, the FOMC
abandoned the consensus forecast exercise at the time — perhaps not permanently
— and instead focused on improvements to the SEP. This resulted in the first dot
plot, in FOMC (2012a).
18
FOMC (2020).
19
Powell (2020).
16
17

296 |

Monetary Mystique
With regard to the price stability mandate, the FOMC now “seeks to
achieve inflation that averages 2 percent over time.” This implies that
if inflation has been running persistently below 2 percent, the FOMC
would likely aim to achieve inflation “moderately above 2 percent for
some time.” The Federal Reserve has thus adopted an explicit “makeup”
strategy. As explained by Powell (2020), Clarida (2020), and Brainard
(2020), this introduces a strategy of “flexible average inflation targeting.”20 It is also made clear that it would not be appropriate to implement this strategy by using a mechanical Taylor-type instrument rule.21
Although the Fed has announced that it seeks to achieve inflation
that “averages 2 percent over time,” it has left itself with some flexibility
by not announcing an explicit period over which average inflation is
calculated. Dropping the language about a balanced approach leaves
some ambiguity about the relative weights on the maximum-employment and price-stability mandates. It is understandable if the Fed
prefers some flexibility in adapting the new framework, but eventually
a high level of transparency and accountability will most likely require
the Fed to become more explicit on these points.

S vensson (2020a) discussed the strategies of flexible price-level targeting, temporary price-level targeting when the effective lower bound for the policy rate binds,
flexible average-inflation targeting, and nominal-GDP targeting for the Fed and, in
conclusion, recommended flexible average-inflation targeting.
21
Clarida (2020); Brainard (2020). Such reservations were also expressed by FOMC
participants in the discussion of average-inflation targeting at the September 2019
FOMC meeting (FOMC 2019).
20

| 297

Svensson

References
Blinder, Alan S. 2020. “What Does Jerome Powell Know that William McChesney Martin Did Not—And What Role Did Academic Research Play
in That?” The Manchester School 88, no. S1 (September): 32–49.
Blinder, Alan S., Charles Goodhart, Philipp Hildebrand, David Lipton,
and Charles Wyplosz. 2001. How Do Central Banks Talk? Geneva Reports
on the World Economy 3. Oxford: Information Press.
Brainard, Lael. 2020. “Bringing the Statement on Longer-Run Goals and
Monetary Policy Strategy into Alignment with Longer-Run Changes
in the Economy.” Speech at the Hutchins Center on Fiscal and Monetary Policy at the Brookings Institution, Washington, DC (via webcast),
September 1.
Brunner, Karl. 1981. “The Art of Central Banking.” Working Paper GPB
81-6, Center for Research in Government Policy and Business, University of Rochester.
Clarida, Richard H. 2020. “The Federal Reserve’s New Monetary Policy
Framework: A Robust Evaluation.” Speech at the Peterson Institution
for International Economics, Washington, DC (via webcast), August 31.
Federal Open Market Committee. 1994. “FOMC Statement, February 4,
1994.” Statement, https://www.federalreserve.gov/fomc/19940204
default.htm.
Federal Open Market Committee. 2003a. “FOMC Statement, August 12,
2003.” Statement, https://www.federalreserve.gov/boarddocs/press/
monetary/2003/20030812/default.htm.
Federal Open Market Committee. 2003b. “Minutes of the Federal Open
Market Committee, August 12, 2003.” Minutes, https://www.federal
reserve.gov/monetarypolicy/files/FOMC20030812meeting.pdf.
Federal Open Market Committee. 2008. “SEP: Compilation and Summary of Individual Economic Projections, January 2008.” Report, https://
www.federalreserve.gov/monetarypolicy/files/ FOMC20080130
SEPcompilation.pdf.
298 |

Monetary Mystique
Federal Open Market Committee 2012a. “SEP: Compilation and Summary of Individual Economic Projections, January 2012.” Report,
https://www.federalreserve.gov/monetarypolicy/files/
FOMC20120125SEPcompilation.pdf.
Federal Open Market Committee. 2012b. “Statement on Longer-Run
Goals and Monetary Policy Strategy.” Statement, https://www.
federalreserve.gov/monetarypolicy/files/ FOMC_
LongerRunGoals_201201.pdf.
Federal Open Market Committee. 2012c. “Transcript of the Meeting of
the Federal Open Market Committee on October 23-24, 2012.” Transcript, https://www.federalreserve.gov/monetarypolicy/files/
FOMC20120801meeting.pdf.
Federal Open Market Committee. 2019. “Minutes of the Federal Open
Market Committee, September 17-18,” Minutes, https://www.federal
reserve.gov/monetarypolicy/files/fomcminutes20190918.pdf.
Federal Open Market Committee. 2020. “Statement on Longer-Run
Goals and Monetary Policy Strategy.” Statement, https://www.
federalreserve.gov/monetarypolicy/files/FOMC_LongerRunGoals.pdf.
Goodfriend, Marvin. 1986. “Monetary Mystique: Secrecy and Central
Banking.” Journal of Monetary Economics 17, no. 1 (January): 63–92.
Greenspan, Alan. 1989. “Statement before the Subcommittee on
Domestic Monetary Policy of the Committee on Banking, Finance and
Urban Affairs, U.S. House of Representatives.” Statement, October 25.
Greenspan, Alan. 2003. “Testimony before the Committee on Financial
Services, U.S. House of Representatives.” Testimony, July 15.
Powell, Jerome H. 2020. “New Economic Challenges and the Fed’s Monetary Policy Review.” Speech at the Federal Reserve Bank of Kansas City
Economic Policy Symposium “Navigating the Decade Ahead: Implications for Monetary Policy,” Jackson Hole, Wyoming, August 27.
Svensson, Lars E.O. 2020a. “Monetary Policy Strategies for the Federal
Reserve.” International Journal of Central Banking 16, no. 1 (February):
133-93.

| 299

Svensson
Svensson, Lars E.O. 2020b. “What Rule for the Federal Reserve? Forecast
Targeting.” International Journal of Central Banking 16, no. 6 (December): 35-95.
Yellen, Janet L. 2012. “Revolution and Evolution in Central Bank Communications.” Speech at the Hass School of Business, University of
California, Berkeley, November 13.

300 |

| 301

Interest Rate Policy

Interest Rate Policy
John B. Taylor
Marvin Goodfriend’s classic 1991 paper, “Interest Rates and the Conduct of Monetary Policy” was first published in the Carnegie-Rochester
Conference Series on Public Policy more than three decades ago. It is a
wide-ranging paper with an original analysis of interest rate policy that
was relevant in 1991 but is even more relevant today. His analysis was
informed by his experience in the Federal Reserve System as a policy
adviser at the Federal Reserve Bank of Richmond. He took this unique,
first-hand experience and translated it into practical monetary policy
proposals in a highly thoughtful and original way.
The Goodfriend paper begins with a history of the Fed’s interest rate
targeting procedures that is useful for monetary economists even today. He then reviews the instrument choice problem — money versus
interest rate — that had been studied in a classic article by William
Poole in 1970, describing how its results carried over to a modern
dynamic-rational-expectations model. He discusses the mechanics of
interest rate smoothing, showing how the persistence of the federal
funds rate results from the Fed’s macroeconomic stabilization policy.1
Finally, he provides evidence that the Fed implicitly had rules-based
monetary policy for the interest rate during most of the 1970s and
1980s.2
In this paper, I build on the analysis of Marvin Goodfriend and examine how the Fed can better engage in a rules-based monetary policy
going forward.
 otsey, Hornstein and Wolman discuss Goodfriend’s (1987) modeling of interest rate
D
smoothing in another essay in this volume.
2
Athanasios Orphanides and Volker Wieland later provided a detailed confirmation of
this view, stimulated by their work at the Fed to provide “Taylor Rule” memos to the
FOMC starting in the mid 1990s.
1

302 |

| 303

Taylor
Prior to the global financial crisis, policymakers within the Federal
Reserve System had adopted elements of the rules-based approach
to interest rate policy that I advocated in my 1993 Carnegie-Rochester
paper. For example, during his time as president of the Federal Reserve
Bank of St. Louis, William Poole used “the Taylor rule” as a guide to his
thinking about policy actions to be taken in upcoming meetings and
as a vehicle for explaining the Fed’s decisions to the public.3 But then
there was a move away from such an explicit use as the Fed and the
government more generally used a wide range of policies to deal with
the Great Recession, not all of which I view as desirable.4
More recently, starting around 2017, the Federal Reserve returned
to a more rules-based monetary policy that had worked well in the
United States in the 1980s and 1990s, as Goodfriend observed. Many
papers were written at the Fed and elsewhere reflecting this revival
and showing the benefits of rules-based policies. In 2017, the Fed
began to report on rules-based policy in its Monetary Policy Report, and
favorable comments about rules-based policy were made by many
policymakers.
One explanation for the revival was simply a revealed preference for
such an approach on the part of monetary policy officials and others
interested in monetary policymaking. Another explanation for the
revival was the desire to figure out how to deal with the effective or
zero lower bound on the interest rate that Goodfriend (2000) had highlighted earlier: there was genuine concern at the Fed about the lower
bound in the case of a need for substantial easing. Another possible
explanation was the disappointment with monetary policy leading
to the Great Recession and especially the deviation from rules in the
2003-05 “too low for too long” period. Yet another explanation was
the recognition that rules are needed to evaluate quantitative easing
proposals.
3
4

304 |

See Poole (2007).
Taylor (2009).

Interest Rate Policy
The Fed began a helpful reporting approach in the July 2017 Monetary Policy Report when Janet Yellen was Fed chair. Each report contained the policy rate implications of five well-known rules embedding
reactions to inflation and unemployment.

An interruption
However, that move toward rules-based policies was interrupted
when COVID-19 hit the American economy. The Fed took a number of
actions to deal with the economic effects of the severe health crisis.5
By most accounts, these actions were special and were not consistent
with rules-based policies.
The Fed also stopped reporting on rules-based policy in its Monetary
Policy Report. The pandemic that started in the first quarter of 2020
was a jolt to the American economy and to many other economies.
It interrupted the revival of rules-based policies at the Fed and most
other central banks. The actions at the Fed included a rapid reduction
in the target for the federal funds rate from 1.75 percent to .25 percent
during the weeks of March 2020. Both M1 and M2 measures of the
money supply grew rapidly. It also included large-scale purchases of
Treasury and mortgage-backed securities, causing a large expansion of
the Fed’s balance sheet with assets rising rapidly from about $4 trillion
to about $7 trillion during the second quarter of 2020 and then continuing to grow to about $9 trillion at the end of 2021.
The Federal Reserve’s Monetary Policy Report after the first year of
the pandemic, released on February 19, 2021, however, contained a
whole section on monetary policy rules. That policy rules reentered the
Report was a welcome development, restoring the helpful reporting
approach from the July 2017 Monetary Policy Report. The approach
continued in 2018, 2019, and early 2020, but it was dropped in July
2020.
5

See Taylor (2021).

| 305

Taylor
Five rules were discussed in the February 2021 Monetary Policy
Report on pages 45 through 48. To quote the Report, these include “the
well-known Taylor (1993) rule, the ‘balanced approach’ rule, the ‘adjusted Taylor (1993)’ rule, and the ‘first difference’ rule.” In addition to these
rules, there was a new “‘balanced approach (shortfalls) rule,’ which represents one simple way to illustrate the Committee’s focus on shortfalls
from maximum employment.” Table 1 shows the five rules from the
February 2021 Report. There were also five rules in the earlier Reports,
but the February Report left one out and added the new balanced
approach (shortfalls) rule in its place. As stated in the Fed document,
this simple new rule
	would not call for increasing the policy rate as employment moves
higher and unemployment drops below its estimated longer-run
level. This modified rule aims to illustrate, in a simple way, the Committee’s focus on shortfalls of employment from assessments of its
maximum level.
Table 1. Five Policy Rules in the February 2021 Monetary Policy Report, p. 47

Interest Rate Policy
include the usual section on monetary policy rules. The Fed had included the section on policy rules in its Reports since July 2017, except for
July 2020 during its initial response to Covid — a total of eight times
going back to Janet Yellen’s term as Fed chair.
This omission was significant. It occurred at the same time that the
Fed fell well “behind the curve,” and inflation has risen as a result.6 In
fact, the removal happened as the discrepancy between standard policy rules, including the Taylor rule listed in the Monetary Policy Report,
and actual Fed policy is as large as it has ever been. The removal thus
diverted attention from this big discrepancy. Several members of Congress brought attention to this omission when Chair Powell testified on
March 2 and March 3, 2022, and Powell’s answers were very important.
While he did not provide reasons for the omission, in the House he
answered Rep. Bill Huizenga by pledging “We’ll have it in the next one.”
He then followed up accordingly with Rep. French Hill. In the Senate,
Powell answered Sen. Bill Hagerty by pledging “We’ll bring them back
for the July thing.”
The recent Monetary Policy Report’s omission masks very large differences between the rules and the Fed’s current and forecasted policies.
Figure 1 shows the discrepancy. It gives the FOMC’s projection of the
federal funds rate and the rules-based paths for the federal funds rate
through 2024. This FOMC projection is the “value of the midpoint of
the projected appropriate target range for the federal funds rate or
the projected appropriate target level for the federal funds rate at the
end of the specified calendar year,” as stated in the Fed’s Summary of
Economic Projections.

Reporting rules is only a step toward systematic policy
It is good that rules were put back in the Fed’s Monetary Policy
Report, but it would be more helpful if the Fed incorporated some of
these rules or strategy ideas into its actual decisions. Apparently, this
has not yet happened.

6

I n 2013, Andrew Levin and I argued that “getting behind the curve” was central to the
Great Inflation of the 1960s and 1970s.

Even more troubling, as I write in March 2022, the Federal Reserve
has again eliminated the table and the discussion of rules: the Fed’s
Monetary Policy Report sent to Congress on February 25, 2022, did not
306 |

| 307

Taylor

Interest Rate Policy
difference in the first quarter of 2021, and the difference does not
diminish over time.

Figure 1. FOMC Projections of Federal Funds Rate and a Policy Rule

FOMC - Federal Funds Rate - April 2021
FOMC - Federal Funds Rate - December 2021
FOMC - Federal Fund Rate - September 2021
Taylor Rule - Federal Funds Rate - April 2021 (R*=1, CBO: Defl & Gap)

Percent
3.2
2.8

An optimal reentry

2.4

The policy rule parameters, even with the full percentage point
lower real equilibrium real interest rate, may not adequately reflect the
results of the Fed’s position and the new flexible average inflation rate
concept. To consider these alternatives and thereby come closer to
the new “flexible form of average inflation targeting” policy of the Fed,
we also consider the formulation of policy rules by David Papell and
Ruxandra Prodan (2021) in a recent paper.

2.0
1.6
1.2
0.8
0.4
0.0
-0.4

IV
2020

I

II

III

2021

IV

I

II

III

2022

IV

I

II

III

2023

IV

I

II

III

IV

2024

The dashed line in Figure 1 shows the federal funds rate using the
same parameters as those in the Taylor rule which is discussed in the
February 2021 Monetary Policy Report. Note that in the Monetary Policy
Report the Fed uses the difference between the unemployment rate
(ut) and the long-term natural unemployment rate (utLR) rather than the
output gap, and it thus modifies the coefficient on the difference to
reflect the regular and related movements of the rate difference and
the gap. The so-called equilibrium interest rate has been reduced from
2 percent to 1 percent. Such a reduction has been suggested at the
Fed but may be larger or smaller than assumed here. The line uses the
same percentage deviation of real GDP from potential GDP as in the
Congressional Budget Office (CBO) report, as well as the CBO inflation
forecast for the PCE. Many other economic forecasters have inflation
and real GDP forecasts close to those of CBO.
Even with this smaller equilibrium real interest rate (1 percent rather
than 2 percent in the original Taylor rule), the FOMC’s path for the
federal funds rate is well below any of these policy rules. There is a
308 |

There has been little mention of why the discrepancy exists between
the Fed’s projections and the rules. Does this mean the Fed will actually keep the rate so low under these circumstances regarding real GDP
and inflation? Will it then raise the rate sharply in 2023 or 2024?

Papell and Prodan (2021) consider a Taylor rule with shortfalls and a
balanced approach rule with shortfalls as introduced in the Monetary
Policy Report. In both cases, they consider the unemployment rate
relative to the long-run level rather than the GDP gap. For the Taylor (shortfalls) rule and the balanced approach (shortfalls) rule, they
replace the difference between the unemployment rate in the long run
and the actual unemployment rate with the minimum of that difference and 0. In other words, the focus is on the shortfall of unemployment from the long-run value rather than the deviation. Thus, if the
unemployment rate is 3.5 percent and the long-run level is 4.0 percent,
the interest rate is not raised as it would be in the rules without shortfalls. That is, zero is the minimum of .5 percent (=4.0-3.5) and zero. This
is as in the balanced approach (shortfalls) rule in the Monetary Policy
Report.
Papell and Prodan (2021) observe, however, that this adjustment
does not fully reflect the changes in policy strategy made by the
FOMC. They therefore also consider another adjustment that results
in the Taylor (consistent) rule and the balanced approach (consistent)
rule. This second adjustment defines the unemployment rate
consistent with maximum employment to be 3.5 percent rather than
| 309

Interest Rate Policy

Taylor
4.0 percent and also assumes an inflation rate that is moderately above
the target inflation rate. For example, if the target inflation rate is 2
percent, then they use a moderate inflation rate of 2.2 percent, using
a numerical example of Clarida (2021). This means the Fed would not
adjust the interest rate simply because the inflation rate was 2.0 or 2.1
percent; rather, it would watch for inflation going above 2.2 percent.
Papell and Prodan (2021) consider the behavior of the shortfalls
and the consistent rules over recent history using the actual historical
values of the unemployment rate, the inflation rate, and the federal
funds rate. It is helpful to look at the behavior of the rules going into
the future using forecasts of unemployment and inflation and comparing that with the FOMC's stated path for the interest rate. They look at
the period from the fourth quarter of 2020 through the fourth quarter
of 2023. It is also assumed that the equilibrium real interest rate is .5
percent rather than 1 percent, which reduces the interest rate.
Papell and Prodan also consider the Taylor rule, including the regular, shortfalls, and consistent rules, along with the FOMC path for the
federal funds rate, using a lower equilibrium real interest rate of .5 percent in these rules. The interest rate from the rules rises as the inflation
rate is forecast to rise and the unemployment rate is forecast to fall.
The balanced approach and the balanced approach (shortfalls) rule are
the same through the third quarter of 2022.
Looking out into the period in 2022 and 2023, a sizable gap emerges. That gap rises to 2.4 percent in the fourth quarter of 2022 and 2.8
percent in the fourth quarter of 2023. Also consider the balanced
approach (regular, consistent, and shortfalls) rule. There is little difference in the later years with the average difference between the rule
and federal funds rate being 3 percent in 2023Q4, compared with 2.8
percent and 2.7 percent with the Taylor rules. But the balanced approach rules rise faster. Thus, it indicates that the policy rate could be
held low through the fourth quarter of 2021. But even in this case, an
adjustment is warranted; perhaps for this reason, in the first quarter
of 2022 the Fed began to signal higher rates by the end of 2022. In its
March 16, 2022, Summary of Economic Projections, the Federal Open
Market Committee reported that the “the value of the midpoint of the
310 |

projected appropriate target range for the federal funds rate or the
projected appropriate target level” would be 1.9 percent by the end of
2022. But this is still low, and even lower if one adjusts the rule-based
path upward for the federal funds rate to take account of higher inflation rates observed in 2022.
To summarize, the analysis of optimal reentry takes into account the
shortfalls of unemployment rather than deviations and focuses on the
average inflation rate by looking at moderate inflation rates slightly
higher than the long-run target inflation rate. Nevertheless, the results
are similar to what was found by looking at the regular Taylor rule. The
results can be usefully summarized by looking at the average gap in
percentage points between the FOMC interest rate and the settings of
the three rules with modifications.

Conclusion
This paper has examined a return to a rules-based monetary policy
in the United States and has outlined methods to do so. By reviewing
the years leading up to the present monetary situation, it provides the
background for analyzing current and future monetary policy decisions.
The results indicate that the Fed should now engage in a strategy or
rule in which people and markets understand how it will adjust its policy interest rate if economic growth increases and inflation stays high
as it is now forecast to do. It would of course be a contingency plan
as all rules and strategies should be. By having clearly stated a shortfalls policy rule in its February 2021 Monetary Policy Report, the Fed
has prepared for such a strategy in practice. Explaining how its policy
rule or strategy would be consistent with its flexible average inflation
targeting would further clarify the Fed’s monetary policy and facilitate
the market adjustment when it takes place. It would remove uncertainty and remaining inconsistencies. The changes in the Reports have not
yet affected actual monetary policy decisions, and there is evidence of
a big difference between the rules-based policy and the actions of the
Fed.
By any measure it is time for reentry.
| 311

Taylor

References
Clarida, Richard. 2021. “The Federal Reserve’s New Framework: Context
and Consequences.” Remarks at “The Road Ahead for Central Banks,”
a seminar sponsored by the Hoover Economic Policy Working Group,
Hoover Institution, Stanford, California, January 13.

Interest Rate Policy
Time.” NBER Reporter no. 3 (July).
Taylor, John B. 2021. “The Optimal Reentry to a Monetary Policy Strategy.” Paper prepared for the Seminar in Applied Economics, The Graduate Center, City University of New York, April 13.

Goodfriend, Marvin. 1987. “Interest Rate Smoothing and Price Level
Trend-Stationarity.” Journal of Monetary Economics 19, no. 3 (May): 335348.
Goodfriend, Marvin. 1991. “Interest Rates and the Conduct of Monetary
Policy.” Carnegie-Rochester Conference Series on Public Policy 34 (Spring):
7–30.
Goodfriend, Marvin. 2000. “Overcoming the Zero Bound on Interest
Rate Policy.” Journal of Money, Credit and Banking 32, no. 4, Part 2 (November): 1007-35.
Levin, Andrew, and John B. Taylor. 2013. "Falling Behind the Curve: A
Positive Analysis of Stop-Start Monetary Policies and the Great Inflation." In The Great Inflation: The Rebirth of Modern Central Banking, edited by Michael D. Bordo and Athanasios Orphanides, 217-244. Chicago:
University of Chicago Press.
Papell, David H., and Ruxandra Prodan. 2021. “Policy Rules Consistent
with the FOMC’s Longer-Run Goals and Monetary Policy Strategy.” June
1.
Poole, William. 1970. “Optimal Choice of Monetary Policy Instruments
in a Simple Stochastic Macro Model.” Quarterly Journal of Economics 84,
no. 2 (May): 197–216.
Poole, William. 2007. “Understanding the Fed”. Federal Reserve Bank of
St. Louis Review 89, no. 1 (January/February): 3-14.
Taylor, John B. 1993. “Discretion Versus Policy Rules in Practice.” Carnegie-Rochester Series on Public Policy 39 (December): 195-214
Taylor, John B. 2009. “Empirically Evaluating Economic Policy in Real
312 |

| 313

Optimal Prediction

The Implications of Optimal
Prediction Formulae
Mark W. Watson
Marvin Goodfriend was not an econometrician, but he was a quantitative economist. He based his policy advice on the logic and quantitative implications of economic models. Like many economists who
came of age in the late 1970s and early 1980s, Marvin learned how
models were affected by the assumption of rational expectations. He
also realized that the optimal prediction formulae used to compute rational expectations had implications for econometric practice, and he
used these implications in his empirical research. Three of Marvin's papers include particularly novel applications of these insights. I'll discuss
these and then conclude with some brief comments about Marvin and
the research environment at the Federal Reserve Bank of Richmond.

Money demand and expected inflation
A key parameter determining the effect of money creation on prices
and seigniorage is the semielasticity of demand for real balances,
, with respect to expected inflation,
. This is the parameter
in the celebrated Cagan money demand function1 that is used to study
periods of hyperinflation:
			(1)
where
. An important challenge for estimating the
semielasticity is that
is unobserved. Marvin's 1982 Journal of
Monetary Economics paper2 presents a method for estimating under
the rational expectations assumption:
			(2)
where
denotes a time t information set. In its general form, (1) represents a canonical linear model involving unobserved future
1
2

314 |

 agan (1956).
C
Goodfriend (1982).

| 315

Watson

Optimal Prediction

expectations. Estimating such models under rational expectations was
an exciting and active area of research as Marvin developed his estimator for , and Marvin, along with several other researchers, proposed
estimators based on essentially the same insight. (Goodfriend references Hall [1978] and Hayashi [1979] and also discusses the important
contribution by Sargent [1977].) The basic estimation insight in all of
these papers is now widely appreciated: any variable can be decomposed as
, where is the prediction error with
. Applied to inflation, (2) implies the decomposition
, where
is uncorrelated with any variable in
. Solving for
using this decomposition and
the information set
substituting into (1) yields
			(3)
so the unobserved expectation of inflation in (1) is replaced with actual
inflation and a prediction error is added to the equation. The predicis positively correlated with
, so the OLS estimator
tion error
of from (3) is not consistent. What is required is an instrument. Ratiois uncorrelated with
nal expectations imply that any variable
, so the challenge is to choose so it is correlated with
. From
satisfies these two requirements, leading to the IV estimator
(1),
				(4)
which is a version of the estimator proposed by Goodfriend. Thus, from
a single equation, and without an explicit modeling of expectations,
one can estimate the semielasticity that was the objective of Cagan,
Sargent, and others. Notice that the structure of the model yields
, so under homoskedastic i.i.d. errors,
is the efficient IV estimator.
is the optimal instrument and
Marvin’s formulation was different from (1)-(4) in three respects. First,
and of no consequence, Marvin solved (3) for
, then regressed
on
using OLS to estimate
, and then inverted to find
. This is a “long-way-around” version of the IV estimator in (4). Secinto its
ond, and more interesting, Marvin decomposed
and components, leading to a test of an overidentifying restriction in
the model. Third, and most important, Marvin considered a more gen316 |

eral version of (1) that included an additional error , a “velocity” shock
to the money demand equation. In this case, and as noted in Marvin’s
is no longer a valid instrument and
is inconsistent.
paper,
that is correlated with
What is required is an instrument
but uncorrelated with the velocity shock. A more complete model (as
in Sargent [1977]) would yield such an instrument using, for example,
an exogenous shifter in the money supply function.
The estimator proposed by Goodfriend in this paper and the related estimators proposed by several others during the late 1970s and
early 1980s were important drivers of the study of GMM estimators.3
time
This analysis has largely been carried out for stationary (or
series. Far less work has been done on the properties of these estimators (and related inference procedures) in models with the explosive
and/or nonstationary data generated by hyperinflations — these were
the data of interest in Goodfriend (1982). Marvin’s use of the rational
expectations assumption yielded valid moment conditions and an
associated IV estimator, but statistical inference with locally explosive
data remains an understudied challenge, even 40 years after Marvin's
contribution. There is still work to do.
Invoking the properties of rational forecast errors to develop estimators is a direct implication of optimal prediction formulae. Marvin’s
other two papers use optimal prediction formulae in more subtle ways.

Money demand and partial adjustment
Marvin continued his study of money demand in Goodfriend (1985)
but in a stationary (non-hyperinflation) environment. A standard
formulation expresses the demand for real balances as a function of a
vector of variables, , that includes real income and the nominal rate
of interest:
			(5)
An empirical puzzle emerged when (5) was estimated using data
from countries like the United States during the 1950s through the
3

E.g., Hansen (1982) and Hayashi and Sims (1983).

| 317

Watson

Optimal Prediction

1970s and early 1980s: the model fit the data rather poorly, but the
fit improved substantially after augmenting the model with a lagged
value of m-p, say
(6)
A popular rationale for (6) is that the demand for real balances adjusts
,4 with a partial adjustslowly toward its target value given by
. A problem with this rationale is that
ment parameter given by
the estimated value of turned out to be large, implying an unreasonably long adjustment process. For example, Goldfeld (1973) reports
=0.72 from a benchmark specification estimated using quarterly data
from the US over 1952:Q2-1972:Q4. This implies an adjustment of only
within the quarter and only 70 percent within a
28 percent
year. Does money demand really adjust that slowly?
Marvin suggested that money demand might, in fact, adjust quite
rapidly, and he suggested that the OLS estimator of in (6) suffers
from errors-in-variables bias. Specifically, he asked: What if the measured value of is a noisy version of the relevant measures of income
and nominal interest rates, say ? Could the resulting errors-in-variables bias lead to large estimated values of , even though
when using the true value of ? Marvin uses optimal prediction formulae to buttress the case for this clever solution to the puzzle about
the apparent sluggish adjustment of money demand.
Classical errors-in-variables lead to well-known attenuation bias, so
the OLS estimators of the coefficients in (5) are biased toward zero. But
Marvin asked the more interesting question: What are the implications
of errors-in-variables for estimating the coefficients in (6)? Answering
,
this question requires specifying a joint stochastic process for
the true value of income and interest rates relevant for money de.
mand, , and the measurement error,
In practice, empirical researchers use proxies for the income (or expenditure) and interest rates relevant for money demand. For example,
Goldfeld used real GDP for income together with interest rates on
4

318 |

For example, see Goldfeld (1973).

commercial paper and time deposits; these were Goldfeld’s -measurements. These are arguably sensible proxies, but they are not perfect measurements of the expenditure and opportunity cost variables
determining money demand. Marvin used a variety of sensible calibrations for the
stochastic process, imposing
so that there is complete adjustment of money demand within the period. He then replaced the true
with the noisy measurement and computed
value of
, yielding the population values of
in (6). Interestingly, these calibrations yield values of that are large
and in line with those estimated in the empirical money demand literature.
Marvin’s explanation for this dynamic errors-in-variables finding
is enlightening: from (5),
is positively correlated with
(and highly so, if the error in (5) is small), is likely to be highly serially
correlated, so
has important predictive power for
,
even after controlling for the proxy measurements in . In Marvin’s
explanation, money demand adjusts rapidly to the fundamentals ,
and the large value of in the estimated regression (6) is not structurfor the
al but instead captures the predictive power of lags of
correctly measured fundamental .

Consumption and income
The final contribution that I highlight is Goodfriend (1992). The
substantive question Marvin addressed in this paper is an apparent
failure of the rational expectations version of the life-cycle model for
consumption when applied to economy-wide aggregate measures of
consumption and income. Specifically, Marvin considered a version of
the Hall (1978) random walk model of consumption that implies (under a set of assumptions) that consumption, , is a martingale, so that
consumption changes are unpredictable. Marvin’s paper studies the
robustness of the martingale property under aggregation: he postulates a model in which each individual’s consumption is a martingale,
and he asks whether the martingale property carries over to aggregate
consumption.
Using generic notation, write the model as

| 319

Watson
		

Optimal Prediction
				(7)

where
so that
. Marvin’s paper considers a
version of (7) with
and
. Equation (7) implies that the
will have a unit coefregression of onto and any variable
ficient on and a zero coefficient on . This is the insight underlying
the well-known Mincer-Zarnowitz test for optimal forecasts and the
related tests of efficient markets in finance.

References
Cagan, P. 1956. “The Monetary Dynamics of Hyperinflation.” In Studies
in the Quantity Theory of Money, edited by Milton Friedman, 25-117.
Chicago: University of Chicago Press.
Goldfeld, S.M. 1973. “The Demand for Money Revisited.” Brookings
Papers on Economic Activity 3: 577-638.

Marvin considers a case in which (7) holds for each of n members of
. He then studies the
a population, so
for
implications for the aggregates, say
.
onto
share the
Will a Mincer-Zarnowitz regression of
? Marvin shows
same properties as the regression of onto
that the aggregates will obey the optimal forecasting relationship if
individuals in the economy share the same information set, that is
for all , but as general matter, not otherwise. As he notes,
may
does not imply that
because
. Goodfriend
contain useful information about not contained in
(1992) uses this insight to discuss Mincer-Zarnowitz regressions using
aggregates and panel data models involving many individuals (large
), but over short time periods (small ). The results are interesting and
insightful.

Goodfriend, M. 1982. “An Alternate Method of Estimating the Cagan
Money Demand Function In Hyperinflation Under Rational Expectations.” Journal of Monetary Economics 9, no. 1: 43-57.

A consultant’s view

Hayashi, F. 1979. “A New Estimation Procedure Under Rational Expectations.” Economics Letters 4, no. 1: 41-43.

I became a regular consultant in FRB Richmond’s research department in 1995. Marvin was research director at the time, and I came
at his invitation. The research department was, and remains, a small,
friendly, and very serious place to work. Seminars are great, lunchtime
conversation is always focused, and a lot gets done. I learn something,
or better yet, get puzzled by something, during every visit. I can’t know
for sure how much of the department’s culture is because of Marvin,
or how much of Marvin was because of the department’s culture. I
suspect there was feedback.
Marvin’s research will have a lasting effect on economics, and his
collegiality and friendship will have a lasting effect on those of us who
were lucky enough to work with him.
320 |

Goodfriend, M. 1985. “Reinterpreting Money Demand Regressions.”
Carnegie-Rochester Conference Series on Public Policy 22: 207-242.
Goodfriend, M. 1992. “Information-Aggregation Bias.” American Economic Review 82, no. 3 (June): 508-519.
Hall, R.E. 1978. “Stochastic Implications of the Life Cycle - Permanent
Income Hypothesis: Theory and Evidence.” Journal of Political Economy
86, no. 6 (December) 971-987.
Hansen, L.P. 1982. “Large Sample Properties of Generalized Method of
Moments Estimators.” Econometrica 50, no. 4 (July): 1029-1054.

Hayashi, F., and C. Sims. 1983. “Nearly Efficient Estimation of Time Series
Models with Predetermined, But Not Exogenous, Instruments.” Econometrica 51, no. 3 (May): 783-798.
Mincer, J., and V. Zarnowitz. 1969. “The Evaluation of Economic Forecasts.” In Economic Forecasts and Expectations: Analysis of Forecasting
Behavior and Performance, edited by Jacob A. Mincer, 3-46. Cambridge:
National Bureau of Economic Research.
Sargent, T.J. 1977. “The Demand for Money During Hyperinflations
Under Rational Expectations.” International Economic Review 18, no. 1
(February): 59-82.
| 321

New Neoclassical Synthesis

The New Neoclassical Synthesis
and the Role of Monetary Policy
Michael Woodford
Marvin Goodfriend is probably best known for his contributions to
practical policy analysis during his long and distinguished career as an
economist and economic advisor within the Federal Reserve system.
But his influence was also great outside the Fed, and indeed outside
the community of central bankers. Marvin made fundamental contributions to the modern theory of monetary policy, which have greatly
influenced the scholarly literature as well. He was unusual in his ability
to bridge the worlds of practical policy debate and scholarly analysis,
providing academics like myself insight into the issues that needed to
be addressed in order for the academic literature to be of greater relevance for policy discussions, while also playing a crucial role in translating the conclusions from economic models for policymakers. My
own work was deeply influenced both by my study of Marvin’s writing
and by the many conversations that I was privileged to have with him
about our shared concerns.
His paper with Bob King, “The New Neoclassical Synthesis and the
Role of Monetary Policy,” is a landmark in the development of the
modern, welfare-based theory of monetary policy. It was one of two
papers1 published in the NBER Macroeconomics Annual for 1997 that
advocated a new approach to monetary policy analysis, using DSGE
models with a basic architecture taken from real business cycle (RBC)
theory,2 but introducing sticky prices in order to allow for real effects of
monetary policy.
1
2

322 |

 long with Rotemberg and Woodford (1997).
A
Kydland and Prescott (1982); Long and Plosser (1983).

| 323

Woodford
The two papers, while written independently, were largely complementary in the approaches that they proposed. However, the emphases of the two papers were different. The primary aim of Rotemberg
and Woodford (1997) was to demonstrate the possibility of the kind
of econometric policy evaluation of specific quantitative policy rules
promoted (most notably) by John Taylor (1979, 1993a), while deriving
both model equations and policy objectives from explicit microeconomic foundations as in RBC models. Goodfriend and King instead
focused primarily on the conceptual foundations of the new type of
model; on the general principles that should inform a welfare-based
approach to monetary policy analysis; and on the desirability of a
particular kind of target for monetary policy, without reference to the
kind of interest rate reaction function that might be involved in implementing it.3
These new papers can usefully be considered in the context of Julio
Rotemberg’s review of the emerging “New Keynesian Microfoundations” a decade earlier.4 That paper had highlighted a shift from an
emphasis on nominal wage rigidity (in the models of authors such as
Ned Phelps, Stan Fischer, and John Taylor in the late 1970s) to models
of sticky prices, an emphasis continued in the first wave of monetary
DSGE models. Rotemberg also emphasized the emergence of models
in which price adjustment results from the explicitly modeled optimizing decisions of firms, rather than being specified by a posited
dynamic response of “the market” to imbalances between supply and
demand or assuming that prices are predetermined by some shadowy
“auctioneer” at a level that is “expected to clear the market” at the time
that the prices are set. In this connection he argued for the value of
modeling individual suppliers as monopolistic competitors. Consideration of the price-setting decisions of suppliers naturally led to an
I n focusing on the quantitative evaluation of alternative interest-rate feedback rules,
Rotemberg and Woodford work within a program advocated by Taylor (1993b). The
discussion of principles for the conduct of monetary policy by Goodfriend and King
is instead more in line with the growing adoption by central banks throughout the
1990s of well-defined inflation targets, without a commitment to specific operating
procedures through which the targets should be hit. On the advantages of formulating rules for monetary policy as “targeting rules,” see Svensson (2003).
4
Rotemberg (1987).
3

324 |

New Neoclassical Synthesis
emphasis on the relationship between individual prices and firms’
(actual or anticipated) marginal costs rather than on a gap between
supply and demand; it also made it natural to consider the role of firms’
expectations regarding future market conditions as a central determinant of pricing dynamics. The paper briefly reviewed popular dynamic
models of staggered wage or price adjustment based on nominal
commitments for a fixed period of time, as in the influential models of
Taylor (1980) and Blanchard (1983) and early models of state-dependent pricing. For purposes of econometric modeling of aggregate time
series, however, Rotemberg advocated two approaches that both allowed flexible variation in the degree of stickiness of prices while preserving tractability of the analysis of dynamics: a model with quadratic
costs of price adjustment5 and one in which individual prices remain
fixed for random intervals, with a constant hazard for reconsideration
at any point in time.6 The 1997 NBER Macroeconomics Annual papers
represent a further stage of development of the program sketched by
Rotemberg.

A New Neoclassical Synthesis
Goodfriend and King call the approach they advocate a “New Neoclassical Synthesis.”7 The terminology recalls Paul Samuelson’s proposal
of a “neoclassical synthesis” in the mid-20th century, intended as a way
to reconcile the use of Keynesian models for practical policy analysis
with the Walrasian model of competitive equilibrium, the canonical
model of a market economy among economic theorists. Samuelson
proposed that the Walrasian model correctly described the long-run
equilibrium of a market economy, once prices and wages have all
adjusted in response to market forces, while the Keynesian model (or
more specifically, its IS-LM formulation by John Hicks) described
 otemberg (1982).
R
Calvo (1983).
7
Others working on related models around the same time proposed a variety of
names for the new style of modeling. Kimball (1995) called it “neomonetarist,” and
King and Wolman (1996) also stressed the monetarist influence on their model. I had
preferred the term “neo-Wicksellian” (Woodford, 2003), but the term that eventually
stuck was “New Keynesian,” probably because of its popularity as an epithet among
critics of the new approach.
5
6

| 325

Woodford
equilibrium in a short run over which wages and/or prices were predetermined. This formulation allowed economists to regard each model
as valid in its own (carefully delimited) sphere of application, but it
didn’t really integrate the two approaches; there was no accepted account of the dynamics of wage and price adjustment that should lead
from one situation to the other, and this left considerable ambiguity
about exactly when (if ever) either of the two limiting cases should be
empirically relevant. The lack of a model of wage and price adjustment
meant that the framework had little to say about the causes or consequences of inflation, a weakness that became glaring by the 1970s;
and the lack of any explicit modeling of dynamics made it hard to say
much about the determinants and effects of expectations, an increasing focus of attention by the 1970s as well. As Goodfriend and King
discuss, these weaknesses made the original neoclassical synthesis
particularly unsuitable as a guide to the conduct of monetary policy.
RBC theory offered a different answer to the question of how to
integrate a model of short-run fluctuations in business activity with
a model of long-run growth by developing a Walrasian model of a
complete intertemporal equilibrium (rather than using Walrasian competitive equilibrium only as a model of an essentially static “long run”),
with fluctuations in response to exogenous random disturbances to
productivity. The Kydland-Prescott (1982) model offered a complete
description of the dynamics of the economy’s response to a shock,
with no artificial separation of “short-run” from “long-run” analysis, and
at the same time provided complete choice-theoretic foundations for
all of the model’s equations, so that there was a clear answer (at least
in theory) to the question of which equations should be considered
“structural” in the face of a change in government policies. However,
RBC models of this kind provided no guidance for monetary policy. Indeed, Kydland and Prescott argued against any role for monetary policy as a determinant of economic activity, even in the short run, so that
a quarterly model of business fluctuations could safely ignore nominal
variables altogether. But the econometric estimation of the real effects
of monetary policy became an increasing focus of study in the late
1980s and throughout the 1990s, and most of this literature (reviewed
in Christiano et al., 1999) found real effects of identified monetary pol326 |

New Neoclassical Synthesis
icy shocks that were nontrivial both in size and persistence. The new
generation of models developed in the mid-1990s sought to make
DSGE models consistent with these facts.
Goodfriend and King argue that the new kind of models represent
an updated (and more articulated) version of Samuelson’s neoclassical
synthesis. A Walrasian model of market equilibrium (essentially, an
RBC model) is still at the heart of the synthesis model and represents
a limiting case of it (one in which the parameter that determines the
delay in price adjustment is set to zero). Moreover, even more than in
the original neoclassical synthesis, all model structural relations are
derived from explicit analysis of the optimization problems of households and firms (including an analysis of optimal price setting, on
those occasions when monopolistically competitive suppliers reconsider their prices), just as in Walrasian general equilibrium models (and
RBC models). Yet the fact that prices are not continually reoptimized
means that the short-run effects of shocks reflect the consequences of
optimizing behavior when some prices or wages are predetermined.
This means, as in the original neoclassical synthesis, that aggregate
demand — and, crucially for Goodfriend and King, monetary policy —
becomes an important determinant of economic activity in the short
run, even though the economy’s long-run growth path is determined
by factors such as productivity growth, growth of the labor force, and
incentives for capital accumulation, which are essentially independent
of monetary policy. The model also retains important features of an
RBC model in that “supply-side” factors (such as random variations in
productivity growth) continue to play an important role in the economy’s short-run dynamics.
Goodfriend and King introduce nominal rigidities into a DSGE model
using a variant of the model of staggered price adjustment originally
proposed by Calvo (1983) and adapted to a discrete-time DSGE framework by Yun (1994, 1996). In the Calvo-Yun model, firms are monopolistically competitive suppliers of differentiated goods and set the
prices of their own product so as to maximize the value to the owners
of the firm of the flow of profits generated by its pricing policy. Thus,
the model is one in which prices are determined on the basis of an
optimizing decision, as advocated by Rotemberg (1987), rather than
| 327

Woodford
being arbitrarily specified or adjusting “in response to market
pressures” through some arbitrarily specified process that is unclear
about who actually arranges for prices to change. This makes clear the
role of factors such as firms’ degree of market power (as well as their
information when decisions about prices are made) in price determination. The real effects of monetary disturbances result from an
assumption that prices are not continuously reconsidered, and, as in
the more ad hoc models of Taylor (1980) and Blanchard (1983), the persistence of these real effects is amplified by staggering of the times at
which different firms reconsider their prices. Yun (1994, chap. 1) further
showed that the empirical realism of the adjustment dynamics implied
by such a specification (when combined with an RBC core model) was
improved by assuming random intervals between price adjustments
rather than fixed-length price commitments as in the models of Taylor
or Blanchard. This made the Calvo-Yun specification convenient for use
in parsimoniously parameterized monetary DSGE models that were
intended to be compared with aggregate time series, such as King and
Watson (1996), Yun (1996), and Rotemberg and Woodford (1997).8 The
Calvo-Yun model also had the advantage of allowing a flexible specification of the time required for prices to adjust, while requiring only a
small number of state variables, so that analytical solutions remained
possible in the case of sufficiently simple policy rules (as illustrated, for
example, in Woodford, 1996 and 1999).
Goodfriend and King discuss how the Calvo-Yun framework can be
further generalized, to endogenize the timing of firms’ price adjustments rather than treating these as exogenously specified. The approach they suggest (citing an early version of Dotsey et al. [1999], in
8

328 |

 otemberg and Woodford modify the basic Calvo-Yun model of price setting to
R
assume that when prices are reconsidered, the new price that takes effect in quarter t
must be set on the basis of the economy’s state in quarter t-1. This assumption makes
their theoretical model consistent with an identifying assumption in their structural
VAR estimation of the effects of monetary policy shocks, which interprets contemporaneous correlation between inflation and interest rate innovations as necessarily
reflecting an effect of current inflation on the Fed’s interest rate target rather than any
possible effect of a policy surprise on price setting in that quarter. A similar time lag is
assumed in Christiano et al. (2005), for the same reason.

New Neoclassical Synthesis
which the analysis is more fully developed) incorporates elements of
“state-dependent” pricing models while retaining much of the tractability of the Calvo-Yun framework. In addition to providing more
complete microfoundations for the specification of price adjustment
dynamics, the richer framework of Goodfriend also nests models such
as those of Taylor and Blanchard as special cases, thus allowing a more
unified treatment of the literature on this topic.
This kind of microfounded model of price adjustment had consequences beyond those relating to the tractability of calculations, the
interpretability of macroeconomic structural relations in terms of measurable microeconomic variables, and the possibility of parameterizing
the model to allow for substantial persistence. One that was to prove
important for subsequent policy discussions followed from the fact
that firms are assumed to set prices in a forward-looking way, recognizing that they are unlikely to reconsider their prices again immediately, though it may already be predictable that market conditions are
changing. This makes expectations, and more specifically expectations
about other firms’ likely price increases over the near term, a crucial
factor in price setting, as Goodfriend and King emphasize. To the
extent that one accepts the realism of the assumptions of this kind of
model, it provides a powerful case for the potential value for stabilization policy of credible, public, and easily interpretable advance commitments about future policy, such as official inflation targets;9 it also
suggests that more ad hoc announcements about future policy, as in
the case of “forward guidance” in response to a crisis, can be effective.10
Another general implication of NNS models, highlighted by Goodfriend and King, is that they imply that an increase in relative price
dispersion has adverse effects similar to a negative productivity shock
and that instability of the general price level should increase such
dispersion. (To establish this result, they leverage the explicit demand
aggregation provided by Dixit-Stiglitz aggregators and the specific
 n the role of models of this kind in the theoretical case for inflation targeting, see in
O
particular Svensson (2011).
10
See for example Eggertsson and Woodford (2003).
9

| 329

Woodford
production aggregation result developed by Yun [1994, 1996].11) This
provides a rigorously microfounded basis for concern about the stability of the general price level. While this is probably not the only reason
that a variable price level complicates economic decision-making and
hence creates distortions, the Goodfriend-King model provides strong
support for the importance of price stability, even without taking into
account these other potential reasons.

Inflation and welfare
The greatest strength of a model of business fluctuations, and of
the short-run effects of monetary policy, with explicit microeconomic
foundations is that it becomes possible to evaluate alternative approaches to the conduct of monetary policy not simply in terms of
positive predictions (i.e., the extent to which various variables should
be stabilized to a greater or lesser extent), but in terms of economic
welfare (i.e., the extent to which people more successfully achieve
their private objectives, the ones revealed by their behavioral choices). Thus the theory of monetary policy can be treated as a branch of
welfare economics, using methods similar to, and fully consistent with,
the ones that had already been used for decades in theoretical public
finance (including the dynamic extensions of the theory that figured
extensively in the more recent literature).
It is in their discussion of the implications of the New Neoclassical
Synthesis (NNS) framework for a normative theory of monetary policy
that Goodfriend and King break the greatest amount of new ground.
Sections 7 and 8 of the paper take up a broad range of central issues in
the theory of monetary policy and provide novel insights about most
of them. Here I will mention only a few of the most striking of these
insights.
Many economic theorists have noted that, in principle, the money
prices charged for real goods and services should be of no significance
for decisions about quantities (only relative prices should matter), and
11

330 |

T hese are also the basis for the penalty for inflation variability in the microfounded
loss function derived by Rotemberg and Woodford (1997), as discussed further in
Woodford (2003, chap. 6).

New Neoclassical Synthesis
they have asked why, if that is so, the inflation rate (the rate of change
of prices in general) should be a matter of concern at all. Goodfriend
and King point out that the NNS model provides an answer by showing how the inflation rate is inevitably connected with changes in
relative prices that distort the allocation of resources (even on the assumption that households and firms are all perfectly rational and thus
not subject to “money illusion”). First of all, as already mentioned, their
model of staggered price setting implies that under any path of the
general level of prices other than perfect price stability, the fact that
different firms revise their prices at different times will result in relative
price differences (that do not reflect any differences in production
costs or utility from consumption of the different goods) and hence in
deadweight losses of the same kind as those resulting from distorting
taxes.
Second, and more subtly, they point out that their model of optimal
price setting implies a structural relationship between price changes and the gap between a good’s current supply price and the firm’s
current marginal cost of supplying the good. Hence there is a tight
connection between variations in the overall inflation rate and variations in the average markup of prices over marginal cost at each point
in time. The markup also has effects on the equilibrium allocation of
resources that are closely analogous to the effects of a tax distortion,
as standard public finance analyses of the deadweight losses associated with monopoly power have long emphasized.
These insights provide the basis for an analysis of what monetary
policy should seek to achieve, that is based on consideration of the
consequences of monetary policy for the deadweight losses associated with relative price distortions rather than taking as primitive
policymakers’ concerns for macroeconomic objectives such as control
of inflation or reduction of unemployment. Goodfriend and King draw
two important conclusions. The first is that monetary policy should
be used to ensure an average inflation rate near zero. This is based on
a consideration of the effects of steady inflation or deflation on the
average markup on the one hand and on the degree of dispersion of
relative prices on the other.
| 331

Woodford
With regard to relative price dispersion, they show that it has an
effect like a downward shift of aggregate productivity owing to the
fact that the “composite good” that matters for consumers’ utility from
consumption is produced using a less efficient mix of individual goods
(owing to their differing prices). This plainly reduces welfare (for any
assumed path of aggregate output, measured in terms of the “composite good”), and it is easy to show that in their model of staggered price
setting, relative price dispersion (and hence the productivity reduction) is minimized when the inflation rate is always zero. (In this case,
all firms can maintain identical prices even though they reconsider the
optimality of their prices at different points in time.) Hence from the
standpoint of this consideration, taken in isolation, an inflation rate of
exactly zero is clearly optimal.12
But it is also necessary to consider the consequences of different
constant (average) inflation rates for the average markup of prices over
marginal costs of supply. Here Goodfriend and King show that inflation
has two offsetting effects. On the one hand, for given expectations
regarding future inflation, a higher inflation rate (a greater rate of
increase of prices on average between period t-1 and period t) implies
a lower average markup in period t, because the firms that do not
reconsider their prices will fall further below the prices that they would
wish to set at that time (which is to say, their prices fall relative to
their marginal costs of supply to a greater extent), while those that do
increase their prices are simply keeping up with the faster growth of
nominal marginal costs (that must grow faster in order to bring about
a higher inflation rate). But on the other hand, for a given current rate
12

332 |

T he result depends on assuming that prices remain unchanged in nominal terms
between the occasions on which they are reconsidered. Yun (1996) proposes a more
complex model in which prices are automatically increased to reflect some “normal”
rate of inflation between the occasions on which they are reconsidered; in that
version of the model, price dispersion is minimized by choosing a steady inflation
rate equal to the “normal” rate that firms expect, which need not be zero. The argument that zero inflation results in minimal price dispersion also tacitly assumes that
one starts from a situation with zero price dispersion. If one starts from a different
distribution of relative prices — for example, because one has had positive inflation
up until now — then the policy that minimizes price dispersion will not be one that
jumps immediately to a zero inflation rate, though it should converge to zero inflation eventually.

New Neoclassical Synthesis
of inflation, a higher expected future rate of inflation (between periods
t and t + 1) will be associated with a higher average markup in period t,
because firms reconsidering their prices (and realizing that most likely
they will not reconsider them again as soon as period t + 1) will raise
them by more than the amount by which nominal marginal costs have
already risen to take account of the higher costs (and higher competitors’ prices) that they expect in period t + 1. These two forces roughly
balance one another, so that changes in the average rate of inflation
around zero don’t much change the average markup (assuming that
the rate at which firms discount future profits is relatively low).
This much they are able to establish analytically using log-linearized
structural equations relating the average markup to the path of inflation, which hold for an inflation rate not too far from zero.13 Goodfriend
and King go further and numerically solve for the deterministic steady
state of their model for different assumed constant inflation rates,
using the exact nonlinear model equations, and show in their calibrated model that while the steady-state markup is relatively constant for
a small range of inflation rates, it becomes significantly higher in the
case of inflation rates that are either much below zero or much above
zero, owing to nonlinearities.14 Hence consideration of relative price
distortions and of average markups lead to roughly similar conclusions: distortions should be larger if the average rate of inflation is
very far from zero in either direction. Goodfriend and King accordingly
argue that policy should strive to keep the average rate of inflation
near zero.
Their discussion of the issue is based on a comparison of alternative
possible stationary equilibria with constant inflation, but as subsequent literature was to show, the conclusion is also true if one asks
what inflation rate one should commit to maintain in the long run,
T he two counterbalancing effects are essentially the same as those that can be observed in the relationship between inflation and the output gap in the familiar “New
Keynesian Phillips curve.”
14
For the purposes of their numerical analysis, Goodfriend and King assume Taylor-style staggered price commitments that last for four quarters and calculate the
effect of steady-state inflation on relative price distortions, the reset price chosen by
adjusting firms, and two measures of the markup.
13

| 333

Woodford
even when transition dynamics are taken into account. Indeed, under
such an optimal policy commitment, one can show that in a model like
the one proposed by Goodfriend and King, the inflation rate should
converge in the long run to exactly zero rather than to a slightly higher
value as suggested by the comparative steady-states analysis given
in their paper. This was first shown in a related NNS model (with price
commitments that last for exactly two periods) by King and Wolman
(1999) and in a model with Calvo-Yun staggered price adjustment by
Woodford (2003, chap. 7).15
But perhaps more notably, the broad conclusion of Goodfriend and
King — that the optimal inflation target cannot be too far from zero
— has proven to be remarkably robust to the addition of a variety of
further complications to their basic monetary DSGE model, as reviewed by Schmitt-Grohé and Uribe (2011).16 Nowadays, the general
consensus is that an inflation target of a couple of percentage points
above zero is preferable to a target of zero. But the modern literature,
even when providing arguments for the preferability of a moderately
positive inflation rate, continues to use the basic method pioneered
by Goodfriend and King: analyzing the implications of different average inflation rates for the microeconomic distortions associated with
different degrees of misalignment of relative prices and prices relative
to costs by considering how trend inflation interacts with optimal price
setting by individual firms.17

Stabilization policy and welfare
The arguments just reviewed concern the average rate of inflation
but do not yet consider the extent to which it may be desirable to
allow inflation to vary around its average (or trend) rate in response to
the shocks that give rise to short-run fluctuations in business activity.
The second important conclusion of Goodfriend and King addresses

New Neoclassical Synthesis
this issue. They argue for a conception of “neutral monetary policy”
under which monetary policy is used to keep the average markup
constant at all times. Under at least some circumstances (which they
describe as their “benchmark” case), this corresponds to maintaining
a constant price level despite the occurrence of real shocks of various
types. Thus their prescription calls not only for an average inflation rate
near zero, but also for complete stabilization of the inflation rate.
This conclusion again follows from a consideration of how monetary
policy affects the economy through its implications for the path of the
average markup. Goodfriend and King argue that monetary policy cannot have much of an effect on the long-run average markup18 (that is,
its average over time, as opposed to the average across firms at a point
in time), but that it can determine how the average markup (across
firms) varies around this long-run value in response to different kinds
of shocks. With regard to the latter issue, they argue that in their model, absolutely any time path for the average markup consistent with
the long-run average level can be achieved by a suitably state-contingent monetary policy.
They then ask how one should want the average markup to vary
with shocks and argue that since the average markup has effects on
the allocation of resources similar to a distorting tax (such as a tax on
labor income), the familiar result in theoretical public finance that it is
desirable to smooth tax rates over time (and across states associated
with different shocks) suggests that it should similarly be optimal to
smooth the average markup over time and across states. The structural
relationship between the path of inflation and the average markup can
then be used to show that the average markup is constant, at the level
that occurs in a flexible price equilibrium, if and only if the inflation
rate is zero at all times. But an inflation rate of zero at all times means
18

S ee Benigno and Woodford (2005) for a more complete treatment of this issue.
Goodfriend and King themselves discussed some of these extensions in a follow-up
paper for the ECB’s First Central Banking Conference on the theme “Why Price Stability?” (Goodfriend and King, 2001). See the discussion of this contribution by Vitor
Gaspar and Frank Smets, elsewhere in the current volume.
17
For a recent example, see Adam and Weber (2019).
15
16

334 |

T hat is, as discussed above, they show that it is not possible to make the markup
be on average much lower than the markup associated with price stability, which
would also be the markup in a flexible-price economy (reflecting the market power
of monopolistically competitive suppliers). It is possible to use monetary policy to
make the average markup significantly higher than this, but that would not be desirable, since even the average markup level associated with price stability distorts the
equilibrium allocation of resources away from the social optimum.

| 335

Woodford
that firms’ prices do not get out of alignment simply because some
firms reconsider their prices and others do not; hence optimal policy
creates a situation with no relative price distortions and no differences
in the markups of different firms. It is thus not only the average markup
that must equal the flexible price markup, but each and every firm’s
markup at each point in time. Hence firms’ prices are at all times exactly the ones that they would choose if they were able to continuously
update their prices, and the equilibrium allocation of resources under
optimal policy will be the same as in an equilibrium with perfectly
flexible prices.
Thus the predictions of RBC theory remain relevant in the view of
Goodfriend and King. These are not simply the way that output, hours
worked, consumption, and so on would vary in response to real shocks
if prices were (counterfactually) fully flexible;19 they are also the way
that these variables should evolve, given the way that the economy
actually does work, in the case of a “neutral monetary policy” — which
Goodfriend and King suggest should be the welfare-maximizing monetary policy.
The proposed argument from an analogy with the theory of tax
smoothing is an important one but somewhat incomplete as presented. The fact that inflation variations must correspond to variations in
the average markup, and that the average markup has consequences
similar to a tax on production or on variable inputs, makes it relevant
to ask about the welfare consequences of variability of such a “tax rate.”
But this distortion is not the only one created by variations in inflation;
inflation variations also create relative price distortions, and so an analysis of the way in which it is optimal for inflation to vary in response to
shocks has to consider the welfare consequences of these effects
19

336 |

 ctually, the flexible-price limiting case of the NNS model is not exactly a canonical
A
RBC model, because it would still involve monopolistic competition, and hence a
positive markup, while the RBC model of Kydland and Prescott is a perfectly competitive economy. Nonetheless, the logic of equilibrium determination is extremely
similar in the two types of flexible-price DSGE models, and even their quantitative
predictions are similar if market power is not too extreme.

New Neoclassical Synthesis
as well.20 Nonetheless, in Goodfriend and King’s baseline case, consideration of the distortions created by variation in the average markup
leads to a conclusion that policy should fully stabilize the price level,
regardless of the shocks hitting the economy; this is also the policy
that minimizes the distortions created by relative price dispersion.
Hence even when one takes account of the relative price distortions
as well, one can conclude (under certain circumstances) that complete
price stability is optimal.21
The result that complete price stability is optimal, despite the occurrence of a variety of types of exogenous disturbances to “demand” and
“supply” factors, is perhaps less counterintuitive once one realizes that
this policy results in an equilibrium allocation of resources that is identical to the one in a flexible price economy that is subject to the same
exogenous shocks.22 At least in the case of a perfectly competitive RBC
model, the equilibrium allocation maximizes the expected utility of
the representative household, even in the presence of many types of
exogenous disturbances. The first welfare theorem does not hold, however, in the case of a flexible price model with monopolistic competition. And the result that perfect price stability is the optimal monetary
policy is also no longer quite correct once one adds staggered pricing
I n the approach introduced in Rotemberg and Woodford (1997), the expected
utility of the representative household is approximated by a quadratic loss function
(derived using a perturbation expansion around the zero inflation steady state). The
loss function has terms of two sorts each period: one proportional to the squared
deviation of aggregate output from its “natural rate,” and the other proportional to
the squared deviation of the inflation rate from zero (the optimal long-run rate).
The deviation of the “average markup” from its steady-state value, emphasized by
Goodfriend and King, is (to a log-linear approximation) proportional to the deviation
of output from its natural rate, as indeed Goodfriend and King note. Thus their consideration of the welfare losses associated with fluctuations in the average markup
corresponds to the terms in the Rotemberg-Woodford loss function that penalize
fluctuations in the output gap. But a full consideration of the welfare consequences
of optimal policy must take account of the terms proportional to the squared inflation rate as well. These terms represent a quadratic approximation to the impact on
productivity of the relative price distortions created by inflation variation, as noted
above.
21
See the discussion of this point in Woodford (2003, chap. 7).
22
King and Wolman (1996) had earlier described this result when using an NNS model
to analyze strict inflation targeting (i.e., price level targeting).
20

| 337

Woodford
to a model with monopolistic competition.23 Nonetheless, if the degree
of market power is not too great, the welfare-optimal responses of
quantities to real disturbances are still fairly close to the flexible price
equilibrium responses, and a policy of maintaining price stability is not
too bad an approximation of the optimal policy.
Goodfriend and King go on to discuss an important case in which
complete stabilization of inflation is not optimal — though the exception further demonstrates the fruitfulness of their general approach.
This is the case of an oil price shock, which they model as an increase
in production costs (negative productivity shock) in the oil-producing
sector. They further assume that the oil-producing sector has flexible
prices, while prices are sticky (with Calvo-Yun staggered price setting)
in the non-oil sector. Their analysis of this case proceeds by first positing that also in the case of this kind of sectoral productivity shock, the
flexible price equilibrium (i.e., the RBC equilibrium) should represent
a welfare optimum.24 They then ask if monetary policy can achieve this
outcome.
If the oil sector has perfectly flexible prices, the answer is that it can
by using monetary policy to ensure a completely stable index of prices
in the non-oil sector (i.e., the sticky price sector). In this case, all firms
in the non-oil sector set the same prices as they would in the flexible
price economy, as do all of the oil-producing firms; hence the equilibrium is the same as in the flexible price economy. Of course, stabilizing
the price index of the sticky price sector is not equivalent to using policy to stabilize a broader price index, which includes the oil price; the
broad price index must be allowed to go up. Thus, one can think of the
policy as one in which “headline inflation” is allowed to rise in response
to a “cost-push shock” in order to avoid having to contract activity
 ptimal policy is analyzed taking explicit account of the distortions due to monopO
olistic competition that remain even in steady state in the subsequent work of King
and Wolman (1999), using a model with two-period price commitments, and the
analysis of optimal policy in a model with Calvo-Yun price-setting by Benigno and
Woodford (2005).
24
Once again, this would be true if one were talking about a flexible-price model in
which both sectors are perfectly competitive. If there is instead monopolistic competition in the non-oil sector, it is not quite true, though the idea remains useful as
an approximation.
23

338 |

New Neoclassical Synthesis
more in the sticky price sector. Alternatively, one can describe it as a
strict inflation-targeting regime, in which however the inflation target is defined in terms of a measure of “core inflation” rather than the
headline rate of inflation.25 Interestingly, not only is the counterfactual
flexible-price allocation still useful as a normative benchmark in this
case, but the optimal policy can still be described as “neutral monetary policy” in the sense that Goodfriend and King propose — that is,
the monetary policy that maintains a constant level for the average
markup (corresponding to the markup in a flexible price equilibrium).
If we assume a similar degree of market power in both sectors (so that
the flexible price markup is the same for both kinds of firms), then
the fact that prices are flexible in the oil-producing sector means that
markups there are always equal to the flexible price markup, regardless
of monetary policy. Achieving an average markup for the economy as
a whole equal to the flexible price markup then requires that monetary
policy ensure a constant average markup in the non-oil sector that is
also equal to the flexible price markup; this is achieved by stabilizing
the price index for the sticky price sector.
The analysis provided by Goodfriend and King depends on assuming
that prices are perfectly flexible in the oil-producing sector. This is not
a bad assumption in the case of the oil sector, but one might also be
concerned about the “cost-push” effects of other kinds of asymmetric
real disturbances that similarly impact the relative costs of supplying
different goods, but none of which are goods with perfectly flexible
prices. In this more general case, it will in general not be possible for
any monetary policy to bring about the allocation of resources corresponding to a flexible price equilibrium; instead, one will have to
consider the trade-off between mitigating or exacerbating distortions
of several types, which cannot all be reduced to zero.26
25
26

T he welfare analysis leading to this conclusion is developed more fully in Aoki (2001).
Even in a one-sector model with only aggregate disturbances, such trade-offs exist,
of course, if the degree of market power is nonnegligible, as shown by Benigno and
Woodford (2005).

| 339

Woodford
Yet even then, “neutral monetary policy” as defined by Goodfriend
and King can provide a reasonable approximation to welfare optimal
policy. In calibrated numerical examples, Woodford (2003, chap. 6)
finds that in a model with two sticky price sectors subject to asymmetric disturbances, a monetary policy that completely stabilizes a particular price index provides a close approximation to the second-best optimal policy; however, as in the discussion of oil shocks by Goodfriend
and King, the price index that one should stabilize is not in general the
one that weights prices in the two sectors in proportion to their share
in the consumption basket of the representative household.27 Instead,
the nearly optimal policy stabilizes a price index that puts greater
weight on prices in the sector with stickier prices, but it does not put
sole weight on prices in only one sector except in the extreme case of
perfect price flexibility in one sector.28 Moreover, the principle of putting more weight on prices in the sector with stickier prices is exactly
what would follow from using monetary policy to stabilize the economy-wide average markup, since in the sector with more flexible prices,
a given range of variation in the sectoral inflation rate corresponds to
smaller variations in markups in that sector.

New Neoclassical Synthesis
spirit of their analysis — insisting not only on explicit microeconomic
foundations for the structural relations that define what policy can
possibly achieve and explicit microeconomic interpretations of the
“shocks” that shift those relationships among aggregate variables,
but also on using microeconomic analysis of the distortions created
by misaligned prices as the basis for welfare judgments with regard
to macroeconomic outcomes — has continued to guide much subsequent work. The paper remains a classic contribution to the theory
of monetary policy, and one from which much can be learned even
today.

Obtaining a more precise characterization of optimal policy, and
dealing with a larger number of complications (additional types of heterogeneity and additional market frictions), requires one to go beyond
the relatively informal discussion of welfare objectives provided in
this paper and develop a quantitative analysis in which the trade-offs
between distortions of different types can be explicitly represented.
However, the distortions identified by Goodfriend and King remain
central to analyses of monetary stabilization policy, even when these
make use of much more complex models. Even more importantly, the
S ee in particular Figures 6.2 and 6.3, and the discussion of these figures. The results
are based on the analysis of the optimal inflation target for a monetary union subject
to region-specific shocks in Benigno (2004).
28
Woodford (2011) provides further insight into the reason for such a policy to approximate an optimal policy commitment, showing analytically that the optimal policy
commitment implies long-run stability of a particular price index. In general, the second-best optimal policy involves transitory fluctuations in this index in response to
shocks but no permanent changes in it, even when the shocks result in permanent
shifts in the relative price of goods supplied by the two sectors.
27

340 |

| 341

Woodford

References
Adam, Klaus, and Henning Weber. 2019. “Optimal Trend Inflation.”
American Economic Review 109, no. 2 (February): 702-737.
Aoki, Kosuke. 2001. “Optimal Monetary Policy Responses to Relative
Price Changes.” Journal of Monetary Economics 48, no. 1 (August): 55-80.
Benigno, Pierpaolo, 2004. “Optimal Monetary Policy in a Currency Area.”
Journal of International Economics 63, no. 2 (July): 293-320.
Benigno, Pierpaolo, and Michael Woodford. 2005. “Inflation Stabilization and Welfare: The Case of a Distorted Steady State.” Journal of the
European Economic Association 3, no. 6 (December): 1185-1236.
Blanchard, Olivier J. 1983. “Price Asynchronization and Price-Level Inertia.” In Inflation, Debt, and Indexation, edited by R. Dornbusch and M.H.
Simonsen, 3-25. Cambridge: MIT Press.
Calvo, Guillermo A. 1983. “Staggered Prices in a Utility-Maximizing
Framework.” Journal of Monetary Economics 12, no. 3 (September): 383398.
Christiano, Lawrence J., Martin S. Eichenbaum, and Charles L. Evans.
1999. “Monetary Policy Shocks: What Have We Learned and to What
End?” In Handbook of Macroeconomics, Volume 1A, edited by J.B. Taylor
and M. Woodford, 65-148. Amsterdam: North-Holland.
Christiano, Lawrence J., Martin S. Eichenbaum, and Charles L. Evans.
2005. “Nominal Rigidities and the Dynamic Effects of a Shock to Monetary Policy.” Journal of Political Economy 113, no. 1 (February): 1-45.
Dotsey, Michael, Robert G. King, and Alexander L. Wolman. 1999.
“State-Dependent Pricing and the General-Equilibrium Dynamics of
Money and Output.” Quarterly Journal of Economics 114, no. 2 (May):
655-690.
Eggertsson, Gauti B., and Michael Woodford. 2003. “The Zero Bound
on Interest Rates and Optimal Monetary Policy.” Brookings Papers on
Economic Activity 1: 139-211.
342 |

New Neoclassical Synthesis
Goodfriend, Marvin, and Robert G. King. 2001. “The Case for Price
Stability.” National Bureau of Economic Research Working Paper 8423,
August; also published In Why Price Stability? Proceedings of the First
ECB Central Banking Conference, edited by A. Garcia-Herrero, V. Gaspar,
L. Hoogduin, J. Morgan, and B. Winkler. Frankfurt, Germany: European
Central Bank.
Kimball, Miles S. 1995. “The Quantitative Analytics of the Basic Neomonetarist Model.” Journal of Money, Credit and Banking 27, no. 4, part 2
(November): 1241-1277.
King, Robert G., and Mark W. Watson. 1996. “Money, Prices, Interest
Rates and the Business Cycle.” Review of Economics and Statistics 78, no.
1 (February): 35-53.
King, Robert G., and Alexander L. Wolman. 1996. “Inflation Targeting in
a St. Louis Model of the 21st Century.” Federal Reserve Bank of St. Louis
Review 78, no. 3 (May/June): 83-107.
King, Robert G., and Alexander L. Wolman. 1999. “What Should the
Monetary Authority Do When Prices Are Sticky?” In Monetary Policy
Rules, edited by John B. Taylor, 349-404. Chicago: University of Chicago
Press.
Kydland, Finn E., and Edward C. Prescott. 1982. “Time to Build and Aggregate Fluctuations.” Econometrica 50, no. 6 (November): 1345-1370.
Long, John B., and Charles I. Plosser. 1983. “Real Business Cycles.” Journal of Political Economy 91, no. 1 (February): 39-69.
Rotemberg, Julio J. “Monopolistic Price Adjustment and Aggregate
Output.” Review of Economic Studies 49, no. 4 (October): 517-531.
Rotemberg, Julio J. 1987. “The New Keynesian Microfoundations.” In
NBER Macroeconomics Annual 1987, Volume 2, edited by Stanley Fischer,
69-104. Cambridge: MIT Press.
Rotemberg, Julio J., and Michael Woodford. 1997. “An Optimization-Based Econometric Framework for the Evaluation of Monetary
Policy.” In NBER Macroeconomics Annual 1997, Volume 12, edited by Ben
S. Bernanke and Julio J. Rotemberg, 297-361. Cambridge: MIT Press.

| 343

Woodford
Schmitt-Grohe, Stephanie, and Martin Uribe. 2010. “The Optimal Rate
of Inflation.” In Handbook of Monetary Economics, edited by Benjamin
M. Friedman and Michael Woodford, Chapter 13. North-Holland: Elsevier.
Svensson, Lars E.O. 2003. “What Is Wrong with Taylor Rules? Using
Judgment in Monetary Policy through Targeting Rules.” Journal of Economic Literature 41, no. 2 (June): 426-477.
Svensson, Lars E.O. 2010. “Inflation Targeting.” In Handbook of Monetary
Economics, edited by Benjamin M. Friedman and Michael Woodford,
Chapter 22. North-Holland: Elsevier.
Taylor, John B. 1979. “Estimation and Control of a Macroeconomic
Model with Rational Expectations.” Econometrica 47, no. 5 (September):
1267-1296.
Taylor, John B. 1980. “Aggregate Dynamics and Staggered Contracts.”
Journal of Political Economy 88, no. 1 (February): 1-23.
Taylor, John B. 1993a. Macroeconomic Policy in a World Economy: From
Econometric Design to Practical Operation. New York: W.W. Norton.
Taylor, John B. 1993b. “Discretion versus Policy Rules in Practice.” Carnegie-Rochester Conference Series on Public Policy 39 (December): 195-214.
Woodford, Michael. 1996. “Control of the Public Debt: A Requirement
for Price Stability?” NBER Working Paper 5684, July.
Woodford, Michael. 1999. “Optimal Monetary Policy Inertia.” NBER
Working Paper 7261, August.
Woodford, Michael. 2003. Interest and Prices: Foundations of a Theory of
Monetary Policy. Princeton: Princeton University Press.

344 |

| 345

Credibility and Inflation Targeting

Credibility and Explicit Inflation
Targeting
Robert G. King and Yang K. Lu
Marvin Goodfriend believed that low and stable inflation should
be the primary objective of a modern central bank and it would lead
to good real outcomes. In the early 2000s, he built a public case that
the US should adopt an explicit inflation targeting system, which had
been advocated earlier — both in the FOMC and in speeches — by
Richmond Fed presidents Robert Black and Al Broaddus.1 In building
his case, Goodfriend drew on his knowledge of monetary history and
cutting-edge macroeconomic theory. He traced key episodes in US
history to the lack of central bank credibility: the inflation of the 1970s,
the costly disinflation of the 1980s, and the “inflation scares” of the
early 1990s.2
Communication of inflation targets to the public was important for
building and maintaining credibility more generally, he argued. With
explicit and credible inflation policy, the Fed would be able to conduct
stabilization of real activity and financial markets as necessary, without its actions being misinterpreted as inflationary or deflationary by
households, price setters, and bond markets.
These views were very different from those of Fed leadership and
many economists at the time. In 1994, Broaddus — with Goodfriend
at his side — had repeatedly advocated that the Federal Open Market
Committee should adopt a public long-run inflation objective and a
public system of shorter-term explicit inflation targets. During 1996,
 e laid out his views as part of a National Bureau of Economic Research initiative on
H
Inflation Targeting, organized by Ben Bernanke and Mike Woodford. The NBER conference was January 23-26, 2003; the published paper is Goodfriend (2004).
2
“Inflation scares” refer to sharp changes in expected inflation reflected in longer-term
yields, a phenomenon that is famously labeled by Goodfriend (1993).
1

346 |

| 347

King and Lu
in response, the FOMC coalesced on an internal long-run goal of 2
percent inflation, but they chose not to make it public. It also rejected
public and explicit inflation targets for shorter horizons. In fact, in the
mid-1990s, the FOMC settled on an “opportunistic approach” to inflation policy — in which planned future inflation was reduced during
recessions but left unaltered in expansions — rather than adopting the
explicit, public, and deliberate approach to disinflation advocated by
Black, Broaddus, and Goodfriend.
By the early 2000s, it was possible for the FOMC to bask in the success of the Fed under Alan Greenspan: following the Volcker disinflation, it had reduced core PCE inflation from the 3.5-5 percent range
in 1989-1990 to between 1 and 2 percent over 1991-2001. In various
speeches around 2000, Fed representatives3 explained that explicit
targeting was unnecessary because existing approaches had worked
in practice. They also voiced concerns that an explicit system would
unduly constrain stabilization policy.4
But, while the FOMC had turned away from explicit inflation targeting in 1996, Goodfriend was not shy in his 2004 inflation targeting
manifesto, advocating a relatively strict form of inflation targeting. This
is not because Goodfriend denied benefits from managing real activity. Instead, he described many situations in which there is no major
trade-off between stabilization of inflation and real objectives. Further,
he argued that the Fed can better manage real activity when its inflation policy is more credible.
In this essay, we review Marvin’s path to making the case that the
United States should adopt an explicit inflation targeting system, as
well as highlighting some key elements of his advocacy. We also
3
4

348 |

Greenspan, governors, regional bank presidents, and leading staffers.
 hile we attribute these views to Fed officials more generally, we frequently draw
W
on the writings of Donald Kohn in his 1996 Jackson Hole panel presentation that
discusses opportunistic disinflation and his discussion of Marvin’s NBER paper (2004).
In a chapter of personal reflections in this volume, Kohn notes that he was frequently
paired as an adversary to Marvin, but it is striking how friendly and good natured
their relationship was over many years. Kohn also notes Marvin’s influence in shaping
his own later support for a form of inflation targeting.

Credibility and Inflation Targeting
consider the links between credibility and explicit inflation targeting as
Goodfriend saw them in the early 2000s and as we see them today.
The organization of our discussion is as follows. We begin in the section “Richmond and inflation targeting” by placing Marvin within the
Richmond Fed tradition, specifically the expressed policy views of presidents Robert P. Black and J. Alfred Broaddus. These leaders fostered
Marvin’s intellectual development and relished the aspiration and
discipline that Marvin brought to their research departments. In “Evolving research at FRBR,” we describe how modern monetary economics
made its way into Marvin’s thinking and the Richmond Fed’s FOMC
process more generally. In “Goodfriend’s inflation targeting manifesto,”
we summarize core elements of his 2004 inflation targeting manifesto,
which notably stressed the importance of credibility and portrayed the
Greenspan Fed as practicing implicit inflation targeting.
Reviewing his manifesto, we extract six key ideas: (i) a definition
of implicit inflation targeting as a decision by a central bank to convey less accurate information to the private economy than under an
explicit system; (ii) the importance of inflation scares to Marvin’s view
of US history and to the case for an explicit inflation target; (iii) a view
that 1970s inflation was to be understood as a “breakdown in the mutual understanding” between the public and the Fed; (iv) a view that,
during intervals of low credibility, the Fed had been restricted in its
stabilization efforts and subject to inflation scares arising from market
uncertainty about its policies; (v) a view that an effective implicit inflation targeting regime would require the central bank to act preemptively but that it would be challenged to do so; and (vi) that a credible
inflation policy would allow the Fed flexibility to stabilize real activity
and the financial sector against undesirable shocks, as well as eliminating the real and nominal volatility that he saw the Fed as producing
itself during the 1960s and 1970s.
From this early 2000s starting point, in the “Credibility, inflation, and
real activity” section, we reconsider inflation targeting — implicit and
explicit — and its link to monetary policy credibility. To begin this process, it is necessary to have a concrete definition of credibility, which
we take to be the private sector’s likelihood that a specific policy plan
| 349

King and Lu
will be carried out. This concept can be most directly applied to explicit
inflation targeting regimes (such as those adopted by New Zealand,
Canada, the United Kingdom, and the Riksbank during the 1990s).
While these monetary policy frameworks were not identical, all included transparency about central bank inflation plans and objectives
as well as explanatory communication about the nature of inflation
outcomes. Some also specified current and future inflation targets
that varied over time, as envisioned in the early proposals of Black and
Broaddus. To an important extent, all involved a focus on managing
expectations about inflation and real activity.
Stimulated by Marvin’s ideas, we discuss some models of expectations management with imperfect credibility and of an implicit inflation targeting regime. Both imperfect credibility and imperfectly communicated policies reduce a central bank’s leverage over expectations
and reduce the effectiveness of its stabilization policies. We explain
how an implicit inflation targeting system leaves the central bank open
to inflation scares and even to a more complete breakdown of “mutual
understanding” between it and the public.

Richmond and inflation targeting
While our focal point is Marvin’s “Inflation Targeting in the United
States?”, we begin by documenting the Richmond Fed’s lengthy history
of support for low and stable inflation as the primary objective of monetary policy.

Robert P. Black: 1973-1992
Black is frequently portrayed as a “monetarist” and “inflation hawk”:
each element is important to understand policy analysis during his
presidency.5 But, like other simple characterizations, these labels mask
more important underlying beliefs, notably a faith in the strength of
market economies and an understanding of the limits of monetary
policy.
5

350 |

F or example, at Federal Reserve History, https://www.federalreservehistory.org/
people/robert-p-black.

Credibility and Inflation Targeting
Monetarism and the Mandate
In 1984, in the sixth quarter of recovery from the recession that
had ended in October 1982, Black spoke to the annual convention of
the Virginia Banker’s Association: he began by raising the question of
whether the Fed’s dual mandate was a help or hindrance for monetary policy. He then argued that (i) the existing broad mandate was a
limited practical guide and was sometimes an impediment; (ii) if the
public should choose to give the Fed a narrow objective that the best
choice would be price stability; and (iii) this outcome could be attained
by “slowly but surely” reducing the growth rate of M1. In addition, he
highlighted that such a narrow mandate would make it reasonable
to hold the Fed accountable for the behavior of the price level over a
period of two or three years. Finally, he noted that he was “enough of a
pragmatist to have absolutely no objection to switching to some other
monetary handle if it is ever demonstrated that something else has
become superior to M1.”6
Inflation targeting comes to FOMC meetings
The first time that the FOMC transcripts include a mention of “inflation targets” as an explicit policy proposal is in December 1986 comments by Black within a committee discussion of difficulties with monetary targeting: he outlined the “more radical idea” of “setting inflation
targets for the next three years or so,” which he described as “important
when we have to take the unpopular step of tightening.” In February
1987, he returned to the theme, linking it to preserving Fed credibility
in a time of rising actual and expected inflation.7
In 1989, Rep. Stephen Neal offered an amendment to the Federal
Reserve Act to require the Fed to transition to zero inflation within five
F ull text of the speech is available at https://www.richmondfed.org/publications/
research/economic_review/1984/er70040.
7
We searched the FOMC transcripts on this topic and others using strings such as
“inflation tar,” “inflation obj,” and “inflation goal” as participants sometimes used these
interchangeably. We then closely read pages of text preceding and following the
located string to understand the context. We thank Adam Shapiro of the FRB San
Francisco for his help with this activity.
6

| 351

King and Lu
years. Black joined four other regional bank presidents and Fed Chairman Alan Greenspan in testifying in support of the proposal.8 At the
time, inflation was running in the 5 percent range, having risen sharply at the end of Volcker’s term and the start of Greenspan’s. Over the
course of 1989, members of the FOMC had expressed interest in a
detailed staff analysis, which was presented at the December 19, 1989,
meeting.9 As the Fed’s staff economists assessed the real consequences of such a further disinflation, they highlighted the importance of
credibility.

Credibility and Inflation Targeting
percent in January 1994 to 6 percent in January 1995.11
Broaddus opened his comments by noting that the Fed’s Greenbook
inflation projection was 3-3/4 percent for the first quarter of 1995 and
that some private forecasts were for 4 percent or higher toward the
end of the year. He also pointed to evidence of rising inflation expectations in the sharp increase of the long bond rate since summer 1993,
noting that at 7.75 percent it stood at the highest level since 1991.12
Broaddus suggested that the bond rate indicated that
	the longer-term inflation expectations of market participants is
something closer to 4 percent than the 3 percent rate for the CPI that
the staff is projecting for the second half of 1995 and on into 1996.
That says to me that we still have a credibility gap. Market participants do not yet seem to be convinced that we are going to take
the actions we need to take to achieve our own internal inflation
forecast. So, I think it’s essential that we find a way to reaffirm our
commitment to price stability at an early date.

J. Alfred Broaddus: 1993-2004
During the mid-1990s, legislators Connie Mack and Jim Saxton introduced a series of increasingly specific bills. As discussed by Fed Governor Laurence Meyer in 2001, Saxton’s proposal involved “mandating
price stability as the ‘primary goal’ of the Federal Reserve and requiring
the Fed to establish an explicit numerical definition of inflation.”10
During 1993-1997, Broaddus tirelessly advocated for two ideas. First,
he argued that the FOMC should set a low long-run goal for inflation,
as with the Mack-Sexton proposals. Second, he argued that inflation
should be gradually reduced toward that goal using a publicly announced system of inflation targets, in line with Black’s earlier suggestions and the approach adopted in New Zealand, Canada, and other
countries.

Broaddus highlighted that
	one way to deal with the credibility problem might be to consider
announcing explicit multi-year inflation rate targets leading to price
stability, as has been done in some other countries–say 3 percent for
1995, 2-1/2 percent for 1996, and so forth.

He noted that
Advocating inflation targets at the FOMC in 1994
When it met in September 1994, the FOMC was halfway through a
tightening cycle that ultimately would take the funds rate from 3

F ull text of the speech is available at https://www.richmondfed.org/publications/
research/economic_review/1990/er760101.
9
See p. 1 of the 12-19-1989 meeting transcript at FRBoard, https://www.federalreserve.
gov/monetarypolicy/files/FOMC19891219meeting.pdf.
10
Meyer’s speech is at https://www.federalreserve.gov/boarddocs/speeches/2001/
20010717/default.htm. See also Gramlich: https://www.federalreserve.gov/
boarddocs/speeches/2000/20000113.html.

	If we announced explicit inflation targets and committed ourselves
clearly to achieving those targets, that might buy us a little more flexibility at least with respect to the timing of our short-term policy actions. In the absence of something like this, though, I think we need
seriously to consider some sort of policy action later in the meeting.

8

352 |

T his was during an interval of “preemptive policy” for the Greenspan Fed, as discussed in Goodfriend (2004, p. 320) that successfully “brought the economy to virtual
price stability.” We return to this topic in the section “Goodfriend’s inflation targeting
manifesto.”
12
Later, Goodfriend (2012) would identify this interval as part of an inflation scare in a
memo for the Shadow Open Market Committee. See SOMC, https://www.shadowfed.
org/wp-content/uploads/2012/04/Goodfriend-SOMC-Apr2012.pdf.
11

| 353

King and Lu
The meeting ended with no change in the funds rate target, although there was an agreement on an asymmetric upward tilt. However, this was before the FOMC began to immediately release the funds
rate target and to communicate its views about the near-term evolution of the policy rate.13
In December 1994, Broaddus emphasized the credibility effects of
the FOMC’s preemptive policy over the year, indicating that “the recent
behavior of the bond rate suggests, to me at least, that we have acquired some of late. In my view, that is the most encouraging development we have seen in some time. The trick is going to be to maintain it
going forward as we move into a situation where the risks are at least
a little more balanced than they have been. I might just note once
again if I may that precisely in this kind of situation, something like an
inflation target might be helpful.”14

Credibility and Inflation Targeting
Fed’s flexibility in dealing with short-term economic disturbances since
appropriate short-term actions could be taken without (or with much
less) concern about the potential loss of long-term credibility.”
Goals and targets at the FOMC in 1996
During 1995 and 1996, special components of FOMC meetings were
devoted to two major substantive questions. First, what long-run goal
should it have for inflation? Second, if that required a reduction in
inflation from the prevailing level — a disinflation — then what was
the best path?
Having previously endorsed price stability as the long-run objective,
Broaddus continued to advocate for an explicit inflation target during
1995 and 1996.16 By January 1996, seeing little prospect for the Mack
and Saxton Bill in Congress, he pushed for Greenspan to include
	in the Humphrey-Hawkins written report and hopefully in your
testimony, Mr. Chairman, a positive statement that the Committee
wants and expects the CPI inflation rate to remain below 3 percent
on average over the two-year 1996-1997 period and that beyond
that we intend to take steps to bring the inflation rate down further
over time. We could think of this, and describe this publicly, as a sort
of benchmark... Such a benchmark would give the Congress and
the public, and for that matter ourselves, something more concrete
than we have had in the past to hold ourselves accountable for. It
may seem like a small step, but I think this would be a significant
departure from what we have done in the past. I believe it would get
some attention and hopefully improve our credibility along with our
accountability...(I)f it would make the Committee more comfortable,

Speaking to the public in 1995
Broaddus also spoke to the public regularly: an excellent example
is his “Reflections on Monetary Policy” delivered to the Virginia Association of Economists in 1995.15 In such presentations, he stressed that
low inflation (“stable prices”) should be the long-run goal of monetary
policy, describing the importance of maintaining and increasing the
credibility for that objective. He advocated that this objective should
be publicly and unilaterally adopted by the Fed, despite the fact that
the Neal amendment had not been passed by Congress. He explained
that this objective is “fully consistent with the present Humphrey-Hawkins mandate since price stability would permit the economy to
achieve maximum growth in output and employment over time.” In
fact, he argued that the explicit long-run “objective would increase the
T he target began to be released July 6 1995. For a detailed history of evolving Fed
communication policy 1975-2003, see Lindsey (2003).
14
Transcript of 12-20-1994 FOMC meeting, p. 16, https://www.federalreserve.gov/
monetarypolicy/files/fomc19941220meeting.pdf.
15
Full text of speech is at https://www.richmondfed.org/press_room/speeches/
j_alfred_broaddus/1995/broaddus_speech_19950316.
13

354 |

16

 s he built his case in 1995-1996, though in the minority, he drew support from Jerry
A
Jordan (see the Jan 1996 meeting transcript, http://www.federalreserve.gov/monetarypolicy/files/FOMC19960131meeting.pdf ) and Tom Meltzer (see the July 1996
meeting transcript, https://www.federalreserve.gov/monetarypolicy/files/
fomc19960703meeting.pdf ).

| 355

King and Lu
	we could add a statement in the report that a benchmark like this
would not necessarily constrain us ... or prevent us from continuing
to take actions that are aimed at stabilizing employment and output
in the short run. I think it would make us evaluate such short-term
actions against our longer-term price stability objective rather than
evaluating efforts to contain inflation against an implicit unemployment objective, which I think has been the case in some past years.17

Opportunistic disinflation and a new goal
But during 1995 and 1996, new and very different ideas became
prominent at the FOMC that would shape its policy going forward.
First, to the extent that further reductions in inflation were to be necessary, it coalesced around an opportunistic disinflation strategy, as
opposed to the deliberate one advocated by Broaddus. Second, after
presentations by Broaddus and Yellen, it opted for a 2 percent longrun goal for inflation, although the committee did not fully settle on
whether this goal was for CPI or PCE and it chose not to make the goal
public.
Opportunistic disinflation
In 1989, FRB Philadelphia President Edward G. Boehne had suggested to his FOMC compatriots a strategy of disinflation that was later labeled “opportunistic disinflation.” In a speech in September 1996, new
Governor Laurence Meyer18 described “opportunistic disinflation” as
follows: “Under this strategy, once inflation becomes modest, as today,
S ee pp. 38-39 in the transcript of the 01-31-1996 meeting, http://www.federal
reserve.gov/monetarypolicy/files/FOMC19960131meeting.pdf.
18
President Clinton announced the renomination of Greenspan, as well as nominations of Alive Rivlin and Meyer as governors in mid-1996. Shortly before these
nominations, Steven Pearlstein of the Washington Post wrote that the renomination
of Greenspan promised a continuation of recent Fed policy. Pearlstein also described Meyer’s view that the Greenspan Fed had previously reduced inflation by an
asymmetric policy of acting cautiously to stimulate the economy during recessions
but acting early and decisively to “limit the economy’s upside potential, sacrificing
a measure of extra job and income growth in a way that most people never realize
they are being sacrificed. The aim is to reduce the long-term trend in inflation with
the minimum of political backlash.” https://www.washingtonpost.com/archive/
politics/1996/02/12/staying-the-course-with-greenspan-at-the-fed/d56fc1ec-c5b64e88-976c-0667e1e0a904/.

Credibility and Inflation Targeting
Federal Reserve policy in the near term focuses on sustaining trend
growth at full employment at the prevailing inflation rate. At this point
the short-run priorities are twofold: sustaining the expansion and preventing an acceleration of inflation. This is, nevertheless, a strategy for
disinflation because it takes advantage of the opportunity of inevitable
recessions and potential positive supply shocks to ratchet down inflation over time. Proponents of this strategy sometimes describe this
approach as reducing inflation cycle-to-cycle or describe the economy
as being one recession from price stability.”19
Some FOMC members and leading Board staff had been moving to
this perspective for some time.20 For the former, it must have appeared
as a way to end the lengthy debates over the importance of credibility
to deliberate disinflation. For the latter, it was congruent with their
accelerationist model of inflation, which made changes in inflation
negatively related to economic slack, measured either by the gap
between unemployment and its nonaccelerating level or by output
relative to potential.
In the decisive July 1996 meeting, Broaddus said he was “uncomfortable with the opportunistic approach” and offered three reasons why.
First, he challenged the accelerationist inflation approach used by its
proponents:
	keeping in mind that the ultimate goal is not temporary price stability but permanent price stability, an opportunistic strategy seems to
be premised on the idea that recessions are permanently rather than
just temporarily disinflationary. [...] In short, I am not sure that there
are autonomous recession opportunities out there, if I can use that
awkward phrase, that can be counted on to reduce inflation permanently in the absence of some deliberate effort to do so on our part.

17

356 |

F ull text of speech is at https://www.federalreserve.gov/boarddocs/
speeches/1996/19960908.htm.
20
In his confirmation hearing testimony in 1994, Alan Blinder had put forward related
ideas and the idea was much discussed during summer 1996 after the Wall Street
Journal highlighted a working paper on the topic by Orphanides and Wilcox (1996),
which included quotes from Boehne and Blinder. The topic figured prominently in
that summer’s Jackson Hole symposium on “Achieving Price Stability,” particularly
in the remarks of Donald Kohn concerning appropriate operating procedures to
maintain price stability.
19

| 357

King and Lu
Second, he challenged the political arguments made by its proponents
	one of the more persuasive arguments for following an opportunistic
policy would be that it might deflect some of the criticism we could
be expected to receive if we follow a more deliberate approach and
are perceived by the public as perhaps keeping policy tight and
keeping the economy slack as a way of reducing the inflation rate.
But if this kind of strategy is going to work, it would seem to imply
that in recessions we would not ease policy as aggressively as we
would if we were not trying to reduce the inflation rate permanently.
At first glance, it might look as if this approach would be less visible, less open to criticism, less of a lightning rod, and thus one that
would be more likely to succeed. But I think there is a risk here that
eventually the public would catch on, and then we would be open to
the criticism that we are not easing policy aggressively enough in a
recession. Think of the phrases that might come out – “we are kicking
the economy while it is down” and so forth. If we got that kind of
feedback, that could undermine the effectiveness of this strategy
over time. So, it is not really clear to me what we would be gaining
from this approach.

Third, he drew attention to its label: 21
	I have always thought that the word opportunistic had a mildly pejorative connotation. [...] So, if we decide to adopt this strategy, I would
hope that at least we would find another name for it. Better yet, I
think it would be better to follow a more deliberate, conventional
policy.22

 roaddus was harking to the conventional definition: “taking advantage of opporB
tunities as they arise: such as exploiting opportunities with little regard to principle”
but, for some, it had come to be used simply as “taking advantage of one’s opportunities.”
22
See pp. 48-49 of the July 1996 transcript at https://www.federalreserve.gov/
monetarypolicy/files/fomc19960703meeting.pdf.
21

358 |

Credibility and Inflation Targeting
The 2 percent long-run goal
At the July 1996 meeting, as the FOMC considered the appropriate
long-run rate of inflation, various members took into account their perceived transition costs, their sense of the benefits from permanently
low inflation, and their sense of the costs of permanently low inflation.
There was diversity in the views reflected in the statements of various
members on each of these topics.
In detailed prepared remarks, Governor Janet Yellen discussed a
cost-benefit approach to determining the optimal long-run rate of inflation and the transition path. Citing research by Akerlof, Dickens, and
Perry (1996), which argued that worker resistance to nominal pay cuts
produced a long-run Phillips curve with a negative slope at low rates
of inflation, Yellen argued for a positive rate of long-run inflation.23 The
idea that positive inflation was necessary “to grease the wheels of the
labor market” was compelling to some FOMC members.
Broaddus pointed out, even if there were disagreements about
near-zero inflation, there was a consensus that the long-run inflation
rate should not be higher than 3 percent. Broaddus and then Cleveland Fed President Jerry Jordan stressed the importance of explicit
public discussion of inflation objectives as a means of enhancing Fed
credibility and thus lowering the cost of further reductions in inflation.
The FOMC discussed how to define “price stability” as an objective of
monetary policy. Greenspan suggested that “price stability is that state
in which expected changes in the general price level do not effectively
alter business or household decisions,” but Yellen challenged him to
translate that general statement into a specific numerical value. He
responded that “the number is zero, if inflation is properly measured.”
23

She also noted that the Board’s new model indicated a cost of 2.5 point years of
unemployment for every 1 percent decline in the long-run inflation rate, under
imperfect credibility. To warrant a reduction in inflation, she argued that such a cost
of permanently lower inflation had to be less than the discounted value of a stream
of future benefits. See p. 42 of the July 1996 transcript at https://www.federalreserve.
gov/monetarypolicy/files/fomc19960703meeting.pdf.

| 359

King and Lu
Yellen said that she preferred 2 percent “imperfectly measured.” FOMC
members generally accepted the idea that there was an upward bias
of about 1/2 percent in annual CPI inflation relative to PCE inflation,
the measure that had begun to be more increasingly used by the Fed.
However, at the time, they never really settled on whether the 2 percent goal was for the CPI or the PCE.
Yet, the FOMC had coalesced around 2 percent as an interim goal.
Presumably, some members viewed it as the natural first step toward a
lower ultimate inflation objective, while others thought of it as an end
point. However, in the meeting, Greenspan and others noted that PCE
inflation was running in the 2 percent range.24 So, they pointed out,
the consensus outcome perhaps meant that the FOMC had already
achieved its objective for “price stability.”
On the second day of the two-day meeting, Greenspan urged that
the 2 percent objective be kept highly confidential. He noted that “the
discussion we had yesterday was exceptionally interesting and important” but warned that “if the 2 percent inflation figure gets out of this
room, it is going to create more problems for us than I think any of you
might anticipate.” He did not elaborate on whether he was concerned
about market or political reactions to the inflation goal.

Implicit inflation targeting at the Greenspan Fed
The FOMC had considered explicit inflation targeting, stimulated by
congressional initiatives and the constant prodding of Al Broaddus,
who was armed with arguments Marvin Goodfriend had helped develop. With its opting for an internal rather than public long-run goal and
opportunistic disinflation rather than pursuit of deliberate disinflation
with announced targets, the Greenspan Fed’s policy differed sharply
from the Richmond proposals. Goodfriend was later to describe the
practice as “implicit inflation targeting” because of the limited communication by the Fed and the lack of public accountability for inflation
24

360 |

S ee p. 59 of the July 1996 transcript at https://www.federalreserve.gov/monetary
policy/files/fomc19960703meeting.pdf.

Credibility and Inflation Targeting
performance. While PCE inflation was in the 2 percent range in 1996, it
was to fall closer to 1 percent in 1998 and 1999, motivating Goodfriend
and others to become concerned with deflationary scenarios.

Evolving research at FRBR
Al Broaddus has highlighted two elements of Marvin’s time in
Richmond.25 First, he described the importance that Marvin attached
to credibility if monetary policy was to be conducted successfully.
Second, he stressed the extraordinary intellectual energy that Marvin
brought to the Research Department’s intellectual environment during
his time in Richmond (1978-2005). In these years, Marvin grew as an
economist in the intellectual environment created by Black and Broaddus. We now trace some of the evolution of his thinking, as well as that
of the department more generally, as it relates to understanding the
case that he made for explicit inflation targeting in 2004.

The monetary instrument
Not too long after Goodfriend arrived in Richmond in Fall 1978, the
Volcker-led Fed announced its famous October 1979 “regime shift” that
emphasized bank reserve management and de-emphasized federal
funds rate control, with the aim of combating inflation and reducing
expectations of inflation.26 Some of Marvin’s early Fed working papers
were stimulated by those changes.27 However, he increasingly focused
on the implications of the Fed’s policies for the funds rate and the
macro economy.
In a PhD class on monetary economics that Marvin took at Brown,
Bill Poole had described his classic analysis of the choice between a
reserve instrument and an interest rate instrument under uncertainty.28 But Poole also explained that it was not possible to consider that
choice in a rational expectations model, because Sargent and Wallace
(1975) had shown that the price level was indeterminate when the
interest rate was taken as exogenous. The Sargent-Wallace finding was
S ee Broaddus’s personal reflections in this volume.
Lindsey, Orphanides, and Rasche (2004).
27
Goodfriend (1982) and Goodfriend et al. (1986).
28
Poole (1970).
25
26

| 361

King and Lu
surprising for researchers and controversial for central bankers. It was
not too long, though, before the monetarist economists Parkin (1978)
and McCallum (1981) recognized that there was no indeterminacy
if the interest rate instrument was part of a policy package with a
well-specified nominal anchor.29
Somewhat later, Goodfriend leveraged the Parkin-McCallum insights
to construct simple rational expectations models in which the central
bank purposefully chose the joint behavior of a monetary quantity,
the price level, and the nominal interest rate, updating the analysis of
Poole (1970) and Sargent and Wallace (1975). Goodfriend found that
when the central bank sought to smooth the price level and nominal
interest rate, its optimal policy gave rise to determinate but nonstationary price level and nonstationary money stock. That is, there was
a form of “base drift” similar to that which he and Al Broaddus had
described for the Fed’s monetary targets and also forcefully critiqued.30
But even though he — like the Fed — was moving away from money,
Marvin’s analysis featured an alternative nominal anchor: a coherent
central bank objective.31
As the decade unfolded, Goodfriend increasingly cast monetary policy decisions — current and historical — in terms of interest rate policy,
a perspective that he masterfully advocated in his 1991 Carnegie-Rochester article “Interest Rates and the Conduct of Monetary Policy.”32 In
making the shift from bank reserves and the money stock to the funds
rate and the price level, Marvin showed the intellectual flexibility
evidenced in Robert Black’s 1984 speech to the Virginia bankers. While
not forgetting his monetarist roots, he evolved along with many other

Credibility and Inflation Targeting
central bankers and economists and held on strongly to other Fisherian principles.33

The funds rate and the term structure
His Richmond Fed colleague Tim Cook substantially influenced Marvin’s evolution, with work on the link between the funds rate and the
term structure.34 To understand this research, one must recall that the
Fed did not release information on its funds rate target decisions after
FOMC meetings in the 1960s and 1970s. Based on his knowledge of
Fed procedures, Cook identified the 1974-1979 period as one in which
the “Desk” at the Federal Reserve Bank of New York had attained substantial day-to-day control over the funds rate. Hence, market participants were able to rapidly discern the effect of unannounced decisions
about the funds rate target, but news stories in the Wall Street Journal
cataloged the events. Based on careful collection and study of these
news reports, Cook and Hahn identified 76 target change events and
estimated the response of the term structure.
At the time, some were skeptical that the Fed had any ability to
affect the term structure. Others were entranced by estimating the
effects of money supply announcements. But after Cook and Hahn
(1989), many changed their views of relevant mechanisms and
events.35
In thinking through Cook’s results and related literature, Marvin developed an appreciation of the importance of learning from the term
structure about evolving expectations. He viewed the short end as
T hese Fisherian principles were strongly in the water in Richmond due to the tireless
efforts of Robert Hetzel and Thomas Humphrey.
34
Cook and Hahn (1989).
35
There has been a recent explosion of work on estimating term structure responses
to monetary policy events using high-frequency data. Cook’s work continues to
be highly cited, though the proponents of high-frequency identification are prone
to criticize it on the grounds that his events are somewhat forecastable. While this
view is econometrically correct, it misses the contribution of Cook’s work, which
showed non-zero term structure responses when these were thought to be absent.
Of course, the “errors in variables” problem associated with partly anticipated events
can be important. But it only attenuates the relevant coefficients: it does not produce
significant findings when no underlying relationship is present.
33

I n the latter part of the Volcker chairmanship, some say by Fall 1982, the Volcker-led
Fed began to de-emphasize monetary targets in internal decision-making due to
money demand instability, shifting to a borrowed reserve approach that also led to
closer control of the funds rate.
30
Broaddus and Goodfriend (1985).
31
In another essay in this volume, Michael Dotsey, Andreas Hornstein, and Alex Wolman discuss further research on “interest rate smoothing” linking to the ultimately
published version of this research (Goodfriend, 1987).
32
Elsewhere in this volume, John Taylor discusses this contribution, Goodfriend (1991).
29

362 |

| 363

King and Lu
dominated by central bank actions and the long end as containing
valuable information about long-horizon inflation expectations.36

Tracking the neutral rate
During the late 1980s and early 1990s, real business cycle analysis
grew from a few initial explorations into a vibrant — if controversial
— research program. Some prominent central bankers viewed it as a
negative shock, but that was not Marvin Goodfriend’s perspective nor
the practice at the FRB Richmond.
Instead, as Marvin became increasingly involved in preparing the FRBR’s policy positions,37 he recognized that the economy’s unobserved
natural rate of interest was critical and began to use elements of RBC
theory to guide his thinking.
For example, in the early 1990s, there were important changes in
US tax policy. In one pre-FOMC meeting during this period, a briefing memo by a young researcher described why the real interest rate
should rise as a result of various fiscal influences in a flexible price
model.38 To the Richmond economist this suggested that a rise in the
nominal policy interest rate target was warranted, drawing on the idea
that the nominal rate should track the “underlying real rate” along with
a targeted amount of expected inflation.

Openness to New Keynesian ideas
At Brown, Marvin had been schooled in a very unusual form of
Keynesian economics, with Herschel Grossman teaching a “general disequilibrium” approach to macroeconomics. Presented using chapters
from a monograph in progress, Grossman’s lectures featured household and firm dynamic optimization in consumption, investment, and
 oodfriend (1998).
G
Goodfriend was associate director of research from 1990-1992, director of research
from 1993-1999, and policy advisor from 1999-2005: in another essay, Al Broaddus
discusses his FOMC contributions. He played a key role in attracting two academic
consultants to FRBR: Bennett T. McCallum, then of the University of Virgina, and
Robert G. King, then of the University of Rochester.
38
Ching-Sheng Mao, now of National Taiwan University, was at FRBR from 1988-1991.
36
37

364 |

Credibility and Inflation Targeting
money demand in settings with gradual price and wage adjustment.39
In the late 1980s and early 1990s, as New Keynesian approaches
developed, Goodfriend absorbed these ideas, pouring over the twin
volumes edited by Mankiw and Romer (1991b, 1991a). Inspired, he
developed a theory in which the price level was sticky in a range of
mark-up indeterminacy, but without other impediments to wage or
price adjustment.40 In this approach, inflation arose only when employment was sufficiently stimulated and a moderate steady-state inflation
was desirable. Ultimately, though, like many others, he settled on using
exogenous pricing frictions as the basis for thinking about inflation
dynamics, particularly the interaction of inflation targets and central
bank credibility.

The productivity boom
In the history of the Greenspan years, perhaps the most widely
celebrated episode is his uncovering of a boom in productivity in the
latter half of the 1990s.41 Many FOMC members and much of the Board
staff saw rapid real output growth and a declining unemployment rate
as a sign of a demand shock: they sought to raise the funds rate to cool
off the economy. Explaining that rising productivity growth would lead
to lower inflation, Greenspan convinced the FOMC in July 1996 to hold
off on an interest rate increase.
By that time, so as to study ongoing developments and alternative
policy regimes, Goodfriend and others at FRB Richmond had adopted
a conceptual and quantitative model that combined monopolistic

 harles Plosser, author of another essay in this volume, received a similar exposure at
C
Chicago in a class from Grossman’s coauthor Robert Barro. Both Barro and Grossman
freely acknowledged the core weakness of their work, which was the absence of an
optimizing theory of price and wage adjustment, and taught the emerging literature
on market-clearing models with imperfect information and rational expectations.
40
Goodfriend (1997).
41
Chapter 11 of Bob Woodward’s 2000 book Maestro is an insider’s account of Greenspan’s unorthodox thinking and his struggle to convince the Board’s economists that
there was increased growth of productivity.

39

| 365

King and Lu
competition and sticky prices with a real business cycle core. Complementary lessons were sometimes also taken from an “optimizing IS-LM”
framework developed by Bennett T. McCallum and Edward Nelson,
which featured a forward-looking IS schedule based on a representative household’s consumption Euler equation.42 In the MN framework,
with an explicit production function and consumption as the dominant component of output, real supply factors such as productivity
and real demand factors such as government purchases affected the
natural rate of interest.
The fully articulated model was used to explore consequences of a
very strict inflation targeting system (a fixed price level path changing
at a constant rate) by King and Wolman (1996). They began with a neoclassical core that was an RBC model with variable labor supply and
capital formation with investment adjustment costs. They then added
Monetarist features43 and Keynesian features.44 The striking conclusion
was that strict price level targeting led to real activity close to that of
their rich core RBC model. An exact coincidence of output with its RBC
behavior could be obtained with an interest rate policy of tracking the
natural rate of interest and penalizing deviations of the price level from
target, with a unique rational expectations solution implied by the
analysis of Kerr and King (1996).45 Shifts in money demand would play
no role under the policy of tracking the natural rate. By contrast, with a
fixed money supply path, such shifts would affect real activity because
of the Keynesian sticky price mechanisms.
I n lunchtime sessions and other informal conversations with Goodfriend and others,
McCallum had long been advocating such an approach, which he had sketched in
his Monetary Economics (1989) and he had begun to develop in lectures in Vienna
(McCallum, 1994) using a log-linear approximation approach. His collaborative work
with Nelson was ultimately published in 1999.
43
A demand for money from a “shopping time” approach.
44
Monopolistically competitive firms with a pricing friction of the Calvo (1983) form.
45
The Kerr-King analysis is sometimes cited as the first to consider Taylor-style interest
rate rules in the now-familiar three equation NK model (they employed McCallum’s
forward-looking IS curve and explored both the 1980s and Calvo approach to price
stickiness). Leeper (1991) had developed a form of the “Taylor principle” in a flexible
price model. But Kerr and King reached the same conclusion for the basic linear NK
model (see, for example, the discussion on p. 391 of Clarida, Gali, and Gertler, 1999).
42

366 |

Credibility and Inflation Targeting
Advocating that central bankers and macroeconomists should adopt
the “new neoclassical synthesis” approach employed in FRB Richmond,
Goodfriend and King (1997) wrote:
	Central banks invariably use a short-term interest rate as their monetary policy instrument. The new synthesis says that central bankers
should manage a low-inflation targeting regime by making the shortterm nominal rate mimic the real short rate that would be ground
out by a well-specified RBC model with a low, constant markup.
	As [an] example of the value of RBC reasoning, consider this. Recently, a possible pickup in productivity growth has been cited as a reason why the Federal Reserve need not raise short-term real interest
rates to maintain low inflation. ...[T]he standard RBC component of
the NNS model suggests, at a minimum, that real rates would have
to rise one for one with an increase in trend productivity growth, e.g.,
a 50 basis point increase in the growth rate would be matched by a
50 basis point increase in real interest rates. Importantly, rates would
have to rise even if the economy were otherwise operating at a noninflationary potential level of GDP.

In evaluating the policy implications of productivity boom, the Richmond approach involved a sharply different form of analysis from that
elsewhere in the Fed.46 The Richmond analysis called for an increase in
 ore important for our discussion, however, Kerr and King also derived implications
M
for interest rate rules with feedback to the deviations of the price level from target.
In this case, the uniqueness conclusion holds irrespective of how strongly the central
bank interest rate rule responds to deviations of the price level from a target path, so
long as it does so positively. Interestingly in retrospect, the King-Wolman analysis of
price level targeting did not assume that the neutral real rate was tracked but instead
held constant at its steady state value. In that case, outcomes corresponded to the
RBC solution only with an aggressive response to deviations from the target path.
However, it was understood at the time that tracking the natural rate would have led
to an exact replication of RBC outcomes under strict inflation targeting.
46
Donald Kohn’s oral history interview in 2011 highlights two differences. First, Kohn
stresses that Broaddus was arguing from a different perspective during the productivity surge (p. 39). Second, David Small notes that it was fortuitous that the Board
was able to use FRB-US, the new forward-looking model introduced in about 1996
and then sharpened over the next few years, to think about the implications of
productivity shocks. Pp. 60-63 of the interview discuss Greenspan, the productivity
boom, and thinking through things using FRB-US. https://www.federalreserve.gov/
aboutthefed/files/donald-l-kohn-interview-20100527.pdf.

| 367

King and Lu
the policy rate to keep the economy on track with its potential. Goodfriend and others stressed that keeping the interest rate low would
lead economic activity to expand by more than was warranted by the
productivity boom.
Several years later, in the November 1999 FOMC meeting, Broaddus
noted that he
	was pleased to see the explicit recognition in the Greenbook that
faster trend productivity growth implies higher real interest rates. I
think one of the principal policy questions we need to ask ourselves
later in the meeting is whether the tightening actions we have taken
to date are sufficient to allow the upward adjustment in real rates
that is necessary to keep the economy in balance, given recent productivity developments. The alternative is that we may be holding
rates below where they need to be to accomplish that objective, with
all of the inflation risk that would imply.47

The Richmond rule and the Taylor rule
During this period, monetary economists increasingly began to
think about interest rate instrument settings through the lens of John
Taylor’s famous rule (Taylor 1993), which involved a high policy rate
when inflation exceeded a 2 percent target and a low policy rate when
output fell below potential.48 The intercept in the rule was a long-run
nominal rate, based on the inflation target and a long-run real rate.
This “rule of thumb” provides direct guidance to a central bank about
its policy settings based on current data, although choices must be
made about the inflation rate and the measurement of the output
gap.49 Many practical monetary economists have stressed that the
S ee p. 28 in the transcript of the 11-16-1999 meeting, https://www.federalreserve.
gov/monetarypolicy/files/FOMC19991116meeting.pdf.
48
The rule specified symmetric response to inflation below target and output above
potential.
49
Poole (2007) uses the Taylor rule to explain Fed behavior during the Greenspan
chairmanship, highlighting both systematic elements and special circumstances. Orphanides and Wieland (2008) draw attention to Poole’s practice as involving a “rule of
thumb” and also highlight the implications for implied Fed behavior of using its own
forecasts rather than historical values. It is of particular interest that their empirical
rule — based on Fed forecasts — better captures the 1994 period of “preemptive
47

368 |

Credibility and Inflation Targeting
central bank needs to adapt it to special circumstances, with financial
crises being the most frequently cited events.50
Comparison with Taylor’s rule provides additional perspective on
how very different the Richmond approach was. It started with desiderata for real activity and expected inflation, then provided an instrument setting within the context of these broader policy objectives.
In the passage cited above, GK argued that “central bankers should
manage a low-inflation targeting regime” by setting the policy rate
according to the Richmond rule.51
It is important to stress that advocates of the Richmond rule were
not naive. They understood that a crucial component of the rule of
“just track the natural rate” was that there was a credible and explicit
inflation targeting system in place. But, as highlighted by the excerpts
from Broaddus’s comments in the September 1994 meeting discussed
above, if the FOMC did not want to move to explicit inflation targeting,
then it had to undertake restrictive policy actions to combat rising
actual and expected inflation, as called for by the Taylor rule. Broaddus,
with Goodfriend at his side, called for raising the funds rate, at that
policy” than does a more standard Taylor rule.
S ome adherents of the Taylor rule would point out that the Richmond rule requires
detailed information on the natural rate of interest, an unobservable time-varying construct whose behavior differs across real theories. By contrast, they would
suggest, the Taylor rule has limited informational requirements and hence is more
robust, noting also that the real activity term in the Taylor rule may capture some
time variation in the natural rate of interest. But the case of productivity variation
provides a useful setting to think about such issues. First, high productivity growth
corresponds to high growth of potential output. So, with either the original “output
gap” form of the Taylor rule or the “output growth” form advocated by Orphanides
and Williams (2002), productivity adjustments to Taylor rule settings must be made
to avoid unsustainable real activity and undesirable variations in inflation. Second,
to gauge the implication of time-varying productivity growth for the policy rate as
a special circumstance, it is likely be preferable to think directly in terms of implications for the natural rate of interest, while recognizing the substantial uncertainty
involved in its shorter-run movements.
51
The behavior of inflation and real activity under optimal monetary policy was being
examined at Richmond as well, in the benchmark sticky price models of King and
Wolman (1999) and Khan, King, and Wolman (2003). Extending their earlier analysis
of price level targeting (1996), King and Wolman discussed implementation of optimal outcomes — differing from those in the underlying real business cycle framework — using a generalization of the Richmond rule, in line with the literature that
we discuss below in the section “Goodfriend’s inflation targeting manifesto.”
50

| 369

King and Lu
time and over the coming meetings, to combat inflation within the
policy regime that was in place.

High-quality policy-relevant research
A hallmark of the development of the Richmond research department during the late 1980s and the 1990s was the relatively seamless
integration of policy analysis and basic research.
Black's challenge to the FRBR research department
During the late 1980s and early 1990s, the FRBR invited its consultants to participate in the “pre-FOMC” briefings when they were in
residence.52 Marvin Goodfriend was the impetus for this initiative, and
he thought carefully about institutional design: the day-long briefing
was broken into a morning session on conceptual topics that would
include the consultants. Then, a staff-only afternoon session began the
process of developing the positions and statement for the President to
take to the FOMC.53 The consultants did not prepare briefing memos
or participate actively in the exchange between the president and his
research team during the meeting. But between the two sessions, at a
large cafeteria round table, lessons from the morning sessions would
be debated by the staff economists, with the president frequently
present.
At one of these FOMC meetings, Black articulated his philosophy
and posed a challenge to the research department after listening to
a policy-relevant but lengthy and conceptually demanding presentation. Essentially, he said: “That was very interesting and informative.
But I cannot take that message to the FOMC table, because no one will
be able to absorb it during the short time available. I need you to write
papers that the Board’s economists will find that they need to read.
S ome sessions at the time were attended by McCallum and King. The structure was
designed to protect both the Bank and the consultants against assertions that they
had access to confidential information, while including their input on substantive
issues of current importance.
53
Al Broaddus’s essay spells out the steps from pre-FOMC to the FOMC, highlighting
how he worked with Marvin and others during his presidency.
52

370 |

Credibility and Inflation Targeting
Once they have absorbed the lessons, they will be able to brief the
governors. Then, our statements at the FOMC can be understood and
influential.”54 It was a challenge that Marvin Goodfriend and his colleagues — growing in quantity and quality — took to heart.
Internal institutions and realizing Black's vision
On becoming research director in 1993, Marvin Goodfriend again
thought about institutional design, recognizing that the Fed required
high-quality research in both money and banking to face the challenges ahead. He made Michael Dotsey the head of the team working on
macroeconomics and monetary economics. At roughly the same time,
he made Jeffrey Lacker the head of Richmond’s team of banking and
financial economists, endorsing the concept that Richmond would
combine dynamic theory and practical studies in those areas as had
earlier been done in monetary economics. There were soon “banking
policy briefings” to complement the “monetary policy briefings.”
Dotsey sharpened the focus of pre-FOMC research, so that it soon
became routine for a briefing memo to evolve into a benchmark publication in the revamped Economic Quarterly and then sometimes into
an article in a top journal. Applied to both money and banking, the
research and policy model drew a diverse group of young, high-quality
economists to Richmond. During the tenure of Al Broaddus as president and Marvin Goodfriend as research director, President Black’s
vision was realized in a low-key, high-intensity research department.

Goodfriend’s inflation targeting manifesto
When Goodfriend began to work on “Inflation Targeting in the United States?” more than five years had passed since the FOMC meeting
where Al Broaddus made the case that the Fed should publicly adopt
a low long-run inflation goal and shorter-run inflation targets. Marvin and Al had come back from Washington disappointed: like Black
before them, they had been unable to convince the Fed chair and a
majority of the FOMC on a topic they saw as central to monetary policy
supporting good outcomes for inflation and real activity.
54

Personal recollections of Robert King from an early 1990s pre-FOMC meeting.

| 371

King and Lu

Explicit inflation targeting in theory
Explicit inflation targeting is characterized, according to Goodfriend,
“by the announcement of an official target for the inflation rate and by
an acknowledgment that low inflation is a priority for monetary policy.
Inflation targeting also involves ‘enhanced transparency of the procedures and objectives of monetary policy, and increased accountability of
the central bank for attaining those objectives.’”55
We begin our discussion of Goodfriend’s inflation targeting manifesto by considering how an explicit inflation targeting system operates
when it is perfectly understood by private agents and also perfectly
credible. This is one benchmark that Goodfriend employed in his 2004
contribution and other writings. For him, it was an important reference
point, as it became for the literature more generally.
EIT and the New Neoclassical Synthesis in 1997
Goodfriend first publicly built a case for an inflation targeting system
at an earlier NBER conference in 1997, when he and Robert King advocated the adoption of “new neoclassical synthesis” models for macroconomic analysis and monetary policy design. The NBER working
paper abstract for the GK paper reads in part:
	We find that the New Neoclassical Synthesis rationalizes an activist
monetary policy which is a simple system of inflation targets. Under
this neutral monetary policy, real quantities evolve as suggested in
the literature on real business cycles. Going beyond broad principles,
we use the new synthesis to address several operational aspects of
inflation targeting. These include its practicality, the response to oil
shocks, the choice of price index, the design of a mandate, and the
tactics of interest rate policy.

The NNS approach rationalized stabilizing the inflation rate of an index of sticky prices at close to zero. It also rationalized central bank use
a form of the Richmond rule, taking into account how various shocks
would affect the natural rate of interest.
55

372 |

Goodfriend (2004), p. 311.

Credibility and Inflation Targeting
New EIT concepts and tools in 2004
By the time Goodfriend presented his manifesto at the 2003 NBER
conference, there had been many conceptual developments in the
analysis of inflation targeting. One important one was the idea of
flexible inflation targeting associated with the practice of the Reserve
Bank of New Zealand and theoretical framing by Lars E.O. Svensson: an
inflation target was not an ironclad simple rule but was a component
of a policy package resulting from optimal choices by a central bank
and government that also placed weights on real objectives.56 Another
innovation was the concept of inflation forecast targeting developed
by Svensson (1997) to describe and analyze the practices of the Bank
of England, initially in a setting where inflation is predetermined and
then extended to settings with forward-looking inflation. The 2003
NBER conference contained important papers on these topics by Michael Woodford: one on forecast targeting mechanics (with Svensson)
and another on optimal inflation targeting rules (with Marc Giannoni).
These targeting frameworks utilized tools developed for optimal
policy under commitment with perfect credibility and were designed
for credible explicit inflation targeting regimes such as that in place at
the Riksbank, which had regular and highly public decision processes
and provided markets with a great deal of information about current
and future actions and planned outcomes. Goodfriend showed some
openness to lessons from the emerging synthetic theory of inflation
targeting and optimal policy, although, as we will see, he expressed
some reservations.
Goodfriend's revised long-run inflation goal
Starting from his earlier advocacy of a zero long-run inflation rate,57
Goodfriend had revised up the optimal long-run target in his inflation
F rom the earliest years of New Zealand experiment with inflation targeting, Svensson argued, central bankers had always stressed that they sought to bring about
desirable real outcomes as well as meeting inflation objectives. In 1996, when he
dubbed such behavior “flexible inflation targeting,” he noted wryly that this expression is more compact than “inflation-and-output-gap targeting.” The published
reference is Svensson (1999).
57
Goodfriend and King (1997, 2001).
56

| 373

King and Lu
targeting manifesto due to a recognition of the potential importance
of the zero lower bound for monetary policy, a topic that had absorbed
his attention in the late 1990s and early 2000s.58
His 2004 proposal was for “a range of 1 to 2 percent for core PCE
inflation monthly over twelve or twenty-four months earlier would be
a reasonable quantitative long-run target.”59 Writing in 2004, Governor
Ben Bernanke noted that “publicly expressed preferences by FOMC
members for long-run inflation have ranged considerably, from less
than 1 percent to 2.5 percent or more” so that Goodfriend’s range
included the views of many, but was perhaps slightly lower.60
Bernanke and Goodfriend both advocated that the “long-run inflation rate” objective of the Fed be made public. Each also pointed to the
continuing volatility of the long end of the term structure of interest
rates, carefully documented in the work of Gurkaynak, Sack, and
Swanson (2005), as an indication of the uncertainty that the public had
about the Fed’s long-run objective.
The strict inflation targeting benchmark
Goodfriend advocated keeping inflation within its long-run explicit
inflation target range even in the short run, a version of strict inflation targeting. In this regard, he differed substantially from Bernanke
(2003), who advocated that the Fed accompany an announcement of
an “optimal long-run inflation rate” of 2 percent with an explicit
 e alluded to the FOMC discussion of Yellen’s 1996 case for a positive but low goal
H
for inflation. But he added a new twist: with a positive inflation rate of about 2 percent, he calculated that downward nominal wage rigidity was irrelevant when there
was also trend productivity growth. Thus, if that form of wage rigidity was most prevalent, he suggested it would not provide an impediment to strict inflation targeting.
(Goodfriend, 2004, p. 331)
59
Goodfriend (2004), p. 327. In 2000, the Fed had highlighted the PCE as its preferred
measure of the inflation rate, as well as highlighting the stability of “core” PCE inflation relative to “headline” PCE inflation over the prior year (Monetary Policy Report,
https://www.federalreserve.gov/pubs/bulletin/2000/0300lead.pdf ). The earlier 1996
FOMC discussions had included the idea that a 2 percent goal for CPI inflation was a
1.5 percent goal for PCE inflation.
60
Remarks at the October 2003 FRB St. Louis Inflation Targeting conference published
in the 2004 Review.
58

374 |

Credibility and Inflation Targeting
statement that it would place “no unwanted constraint” on the shorter-run conduct of policy.61
From the perspective of flexible inflation targeting analyses such as
those just discussed and practical central bankers,62 the idea of a strict
inflation target seems obviously inefficient. Many studies of optimal
monetary policy are based on the central bank having a quadratic
objective for inflation stabilization around a long-run goal but seeking
to produce high levels of real activity. Svensson describes a “strict inflation targeter” as placing no weight on real activity, which Mervyn King
memorably termed an “inflation nutter.” Using a similar quadratic objective, Clarida, Gali, and Gertler (1999) alternatively describe “extreme
inflation targeting” as an outcome that is optimal when there are no
“cost push” shifts in their forward-looking model of inflation dynamics,
referencing only the GK NNS analysis as making such a claim. In the
literature, there are no references to specific inflation nutters except
to the hypothetical conservative central banker of Rogoff (1985). It is
sometimes suggested that Goodfriend was such a central banker, but
this is a misunderstanding.
As we will see, within an explicit inflation targeting system with the
full information and full commitment assumptions of these studies,
Goodfriend saw an important role for interest rate policy in real stabilization and portrayed strict inflation targeting as the anchor for necessary “constrained countercyclical stabilization policy.”63 He argued that
 ernanke’s specific suggestion was that the announcement be accompanied by two
B
provisos: ”(i) The FOMC believes that the stated inflation rate is the one that best promotes its output, employment, and price stability goals in the long run. Hence, in the
long run, the FOMC will try to guide the inflation rate toward the stated value and
maintain it near that value on average over the business cycle and (ii) However, the
FOMC regards this inflation rate as a long-run objective only and sets no fixed time
frame for reaching it. In particular, in deciding how quickly to move toward the longrun inflation objective, the FOMC will always take into account the implications for
near-term economic and financial stability.” (Remarks made at the 28th Annual Policy
Conference, https://www.federalreserve.gov/boarddocs/speeches/2003/20031017/
default.htm.)
62
See Kohn’s (2004) discussion of Goodfriend’s proposal.
63
Goodfriend (2004), p. 323.
61

| 375

King and Lu
core inflation was the appropriate argument in a central bank objective and worked through four prominent shocks that central banks
would experience and found no reason to turn away from strict inflation targeting.
First, he considered an increase in demand which could prove inflationary.64 He pointed out that the central bank can offset this pressure
with an increase in the policy rate, maintaining aggregate demand
equal to potential output so that no inflationary pressure arises. When
this response is systematic, even a serially correlated demand shift will
not bring about changes in expected inflation, so that the increase in
the policy rate is real as well as nominal.65
Second, he considered variations in productivity growth, as we have
described in detail previously, noting that these have little effect on
inflation with an appropriate interest rate adjustment. A corollary was
that time-varying productivity was no impediment to strict inflation
targeting.
Third, he noted that the Greenspan Fed had been able to reduce
interest rates substantially in response to financial crises, without “creating inflation or an inflation scare in bond markets.”66 So, he reasoned,
such responses were also not an impediment to credible strict inflation
targeting. Rather, by stabilizing expectations, inflation targeting would
enhance the central bank’s crisis-fighting capabilities.
Fourth, most controversially, he discussed oil price shocks. There is
a large and lengthy literature on the consequences of such shocks,
which contains three channels: energy as a factor input, energy as a
final consumption good, and the response of inflation expectations
to energy. Goodfriend focused on the factor input case. He noted that
applicable theory indicates that the central bank should be concerned
 n example is an increase in real government purchases of goods and services, as
A
had been studied in the fully articulated Richmond model and also in the framework
of McCallum and Nelson (1999), with the conclusion that it raised the natural rate of
interest.
65
Goodfriend (2004), p. 330.
66
Goodfriend (2004), p. 329.
64

376 |

Credibility and Inflation Targeting
with an index of sticky price goods, which he took as the core PCE.
A rise in the real price of oil would raise production costs for firms
producing sticky price goods and in turn PCE inflation. But such a rise
in PCE inflation could be offset if there is a reduction in employment
and wages, which he depicted as brought about by interest rate policy.
So, one can see the logic of flexible inflation targeting: the central bank
can avoid such labor market effects. But Goodfriend argued that these
are to be understood as a change in the economy’s time-varying potential, so that a central bank concerned with output gap stabilization
would not respond: once again there is no tension with strict inflation
targeting.67 While Goodfriend’s arguments in (1997) and (2004) were
based on sketches of model elements and intuitive implications of
these, the development of macroeconomic theory over this period
made it possible to be more precise. Aoki’s (2001) analysis of a model with one flexible price and one sticky price sector yielded results
consistent with Goodfriend’s intuition.68 Aoki showed that changes in
production conditions in the flexible price sector have no bearing on
the desirability of stabilizing inflation in the sticky price sector, which
is shown to be the appropriate argument in the utility-based objective
of the monetary authority, so there is also no impediment to strict
inflation targeting.
Goodfriend freely acknowledged that his strict inflation targeting
conclusion was based on a sticky price model in which labor market
quantities were the same as if nominal wages were flexible.69 A popular
framework with sticky wages and prices had been developed by Erceg,
Henderson, and Levin (2000), who applied monopolistic competition
and Calvo frictions to wage determination. This framework was later
extended by Bodenstein, Erceg, and Guerrieri (2008), who find that energy price shocks lead to small variation in “sticky price core inflation”
under optimal monetary policy: a 20 percent increase in the price of
energy leads to a 0.25 increase in PCE inflation measured at an annual
percent rate.
 oodfriend (2004), p. 330.
G
Aoki’s paper was not discussed at the conference by Goodfriend or others.
69
He summarized literature suggesting that this was a natural consequence of efficient
firm-worker arrangements.
67
68

| 377

Credibility and Inflation Targeting

King and Lu
Managing expectations: output and inflation
A crucial feature of modern models of optimal monetary policy with
full information and complete credibility is that the central bank seeks
to manage inflation expectations: this is implicit in the optimal policy
analysis of Clarida et al. (1999) discussed earlier and explicit in Woodford (2004). This perspective was hard wired into the chapters that
Giannoni and Woodford (2004) and Svensson and Woodford (2004)
contributed to the volume on “The Inflation Targeting Debate.”
One notable feature of the GW and SW studies was that they began
by determining the optimal behavior of inflation and real activity within a setting of full commitment and full information, exploiting convenient specifications with quadratic objectives and constraints. Each
manipulated the efficiency conditions from policy optimization to
derive a history-dependent target criterion, which the authors stressed
was conceptually appropriate in contrast to purely forward-looking
target criteria that had been employed at the Bank of England. Another notable feature was that these analyses investigated using the
short-term interest rate as the policy instrument to “support” these EIT
outcomes, while working to assure that there was a unique bounded
rational expectations solution.
Thus, SW and GW provide a considerable extension of the Richmond
approach of the mid-1990s discussed in the section “Evolving research
at FRBR”: the important new elements are inflation forecast targeting
and flexible inflation targeting. A natural question that Goodfriend
surely would have asked is whether these analyses provided a direct
extension of the Richmond rule discussed earlier.
Generalizing the Richmond rule
Our reading of Svensson and Woodford (2004) is that the unambiguous answer is “yes.” There are three components in SW that would have
been congenial to Goodfriend and perhaps unsurprising to him. First,
the Fisher equation expresses the nominal rate as the sum of a real
rate and expected inflation. With full credibility and commitment, the
central bank and the private sector have common beliefs and agree
on expected inflation. Second, the central bank’s optimization delivers
a unique solution for inflation and real activity, so that it also delivers
378 |

a real interest rate and expected inflation. While the real rate would
depart from the natural rate under flexible inflation targeting, the real
and nominal rate were nevertheless governed by Fisherian principles.
Third, as with the original Richmond rule, it is necessary to append a
nominal anchor to assure determinacy such as responding to departures of the price level from the path implied by optimal inflation.
We suspect, though, that Goodfriend would have been intrigued
and surprised by the consequences of imposing an “inflation forecast
targeting criterion” as in SW: the optimal nominal rate should be the
natural rate of interest plus the near-term optimal inflation forecast
with a coefficient of less than one.70 This implication of the benchmark
New Keynesian model is robust, in the sense of Giannoni and Woodford (2004), since it holds for a rich range of shocks and stochastic processes. But after some reflection on our part, we think that Goodfriend
would have found this implication congenial as well, given that he saw
commitment capability as enabling a central bank to pursue stabilization objectives using more modest variation in interest rate policy.71
Since the coefficient on the inflation forecast is less than one, an
econometrician studying outcomes under this forward-looking rule
(without taking into account the EIT regime) would be led to conclude
that there was an insufficiently aggressive response to assure determinacy.72
T his implication is displayed in SW (2004, section 2.2.3). We provide a derivation and
discussion in the context of considering interest rate policy and inflation targeting in
some companion research (King and Lu, 2022b). Conceptually, the revised rule does
require that the real rate deviate from the natural rate under optimal flexible inflation
targeting. The revised Richmond rule with a textbook IS curve depends only on one
period ahead central bank forecasts of optimal inflation and real output, which can
be consolidated into just a response to expected inflation using the approach of SW
and GW.
71
Woodford (2003), Chapter 6, considered the design of interest rate rules under optimal policy, emphasizing robustness and determinacy.
72
As with the earlier Richmond rule, determinacy is assured if one appends a small
response to deviations of the price level from the path implied by optimal inflation.
Giannoni (2014) reaches a similar conclusion for a simple ad hoc rule that is optimal
with respect to the parameter on the output gap. He describes his approach as a
“Wicksellian rule” following the terminology of Woodford (2003). In this essay, we
abstain from discussing the controversy over interest rate rules and indeterminacy
reignited by Cochrane (2011) but note his concerns applied to price level as well as
inflation rules.
70

| 379

King and Lu
September 1994 once again
We earlier described Al Broaddus’s conceptual comments and his
recommended policy action in the September 1994 FOMC meeting.73
The Fed — facing rising actual and expected inflation — was in the
midst of a tightening cycle that raised the policy rate by 3 percent
over the course of a year. Broaddus began by advocating instituting
inflation targets to communicate policy intentions and to provide
more flexibility in short-term policy actions. He noted that market
participants did not seem to be convinced that the Fed would take the
actions necessary to “achieve our own internal inflation forecast.”
From the perspective of the optimal policy rule under EIT just discussed, the presumption is that modest real rate movements are necessary to accomplish inflation objectives, in the specific sense that the
nominal rate can rise less than one-for-one with the near-term inflation
forecast. In settings with commitment and credibility, the forecasts of
the central bank and private agents are the same, although Broaddus
saw these as different in September 1994.
With the FOMC unwilling to move to explicit inflation targeting,
Broaddus supported the aggressive increases in the funds rate. A
retrospective analysis by Orphanides and Wieland (2008) shows that
a forward-looking Taylor-style rule, with a weight on the Fed’s internal
forecasts of greater than one, better captures this episode than does
the standard Taylor (1993) specification used by Poole (2007). Thus,
this episode is best understood as preemptive policy, matching the
description of FOMC participants at the time and Goodfriend’s (2002)
portrayal in his narrative account of the phases of monetary policy
1987-2001. Our reading of the September 1994 meeting, though, is
that Broaddus portrayed aggressive preemptive policy as necessary, at
least in part, due to the lack of explicit inflation targeting and imperfect Fed credibility.
Practical limitations of flexible inflation targeting theory
Goodfriend understood the power of expectations management in

Credibility and Inflation Targeting
the New Keynesian model employed in Svensson and Woodford (2004)
and Giannoni and Woodford (2004). In making his case for price stability at the ECB’s first research conference in 2000, he had employed one
period sticky price models capturing these ideas and discussed the
consequences of more elaborate dynamic pricing models.74
At the 2003 NBER IT conference, though, Goodfriend expressed
some reservations about these new concepts and tools. On GW, he “expressed concern about the degree of inflation control that the model
assumed the central bank had. This feature relied on the assumption
that the public was able to observe all shocks with precision. It would
be important to account in the analysis for the possibility that the
public might mistake movements in observed inflation for a change
in the central bank’s inflation target.” Further, he noted that the SW
framework “might be more valuable for analyzing future monetary
policy when central banks have acquired the degree of credibility
assumed in the paper. ... the central bank’s ability to fine-tune inflation
and inflation expectations assumed in the paper might be unrealistically high.” He also questioned “whether identifying cost shocks with
historical residuals from estimated Phillips curves may overstate their
importance, as some of those residuals may not reflect cost shocks, but
credibility problems.”
In his inflation targeting manifesto, Goodfriend highlighted three
reasons that made it “difficult ... for the Fed to manage inflation once
it moves outside its long-run target range.” First, “[t]he policy response
would depend on all information available to the Fed affecting the
conditional inflation forecast and the output gap forecast.” Second,
“[a]rguably, the inflation-generating process is the weakest part of the
macromodel. Among other things the cost, in terms of lost output relative to potential, of returning inflation to its long-run range depends
on the credibility of the Fed’s commitment to do so. The historical
record discussed [...] suggests that such credibility is sensitive to the
Fed’s actions themselves in the context of other aspects of the political
74

73

380 |

T he earlier section “Advocating inflation targets at the FOMC in 1994” provides the
specific quotations.

T his case, developed in Goodfriend and King (2001), is the subject of an essay in this
volume by Vitor Gaspar and Frank Smets.

| 381

King and Lu
economy in a way that is difficult to model.” Third, “the Fed may tend
to overstate the extent to which inflation has an inherent tendency to
persist after it has been shocked.”
Overall, he concluded that
	It is optimal for the monetary authority to vary its short-run inflation
target deliberately in response to some shocks in some macromodels. However, that optimal variation depends sensitively on the
details of the macromodel and on the size and type of shocks hitting
the economy. Given our uncertainty about the structure of the
economy, the difficulty in promptly and accurately identifying the
shocks hitting the economy, and the complications discussed above,
attempting to fine-tune the inflation target in the short run is more
likely to be counterproductive than not.... In any case, the historical
record suggests that the Fed’s ability to deliberately and systematically manipulate inflation in response to shocks is very limited. Moreover, such attempted manipulation would open the door to inflation
scares.75

Hence, he was led to recommend little short-run variation in explicit inflation targets. We see his main concerns about flexible inflation
targeting as fundamentally related to his recognition that the Fed’s
credibility for low inflation could be at stake during the return of inflation to its long-run target.
Goodfriend also had concerns about the degree of central bank
control of inflation assumed in theoretical models of flexible inflation
targeting. In recent work, discussed more fully in ”Credibility, inflation,
and real activity” below, we develop a variant of the standard New
Keynesian optimal policy analysis with both imperfect inflation control
and imperfect credibility. When the long-run goal is explicit, we show
that there is no conflict between flexible inflation targeting and maintaining credibility, in a sense we define more precisely below. However,
a committed central bank that begins with low credibility must alter
75

382 |

Goodfriend (2004), pp. 328-329.

Credibility and Inflation Targeting
the extent to which it is flexible as it works to build its credibility.

Credibility: explicit v. implicit inflation targets
Goodfriend (2004) asks: “[i]n what sense can monetary policy as
currently practiced by the Federal Reserve (Fed) be characterized as
inflation targeting? And what, if any, features of an inflation targeting
policy regime should the Fed adopt more formally?” (p. 311.)
He answered that the Fed monetary policy under Greenspan should
be viewed as implicit inflation targeting. He was not alone in this view,
which was shared by Bernanke and Gertler (1999) as well as others at
the time. But, he argued that the US should move toward an explicit
system, even though this had not been necessary for the conquest of
inflation in the 1980s and the stabilization of inflation in the 1990s.
As discussed above, according to Goodfriend, explicit inflation targeting is defined “by the announcement of an official target for the inflation rate and by an acknowledgment that low inflation is a priority for
monetary policy. Inflation targeting also involves enhanced transparency of the procedures and objectives of monetary policy.”76 At its core,
he argued, an implicit inflation targeting regime is different because
the central bank had chosen not to provide as much information to the
private sector about its plans for inflation and real activity.
Goodfriend had long advocated for greater Fed transparency, beginning with his well-known critique of the Fed’s own arguments for not
disclosing information about FOMC meetings.77 Like Black and Broaddus before him, he saw explicit inflation targets as important for the
Fed’s communication to the public, for its acquiring and maintaining
credibility, and for its accountability to the legislature and the public.78
 oodfriend (2004), p. 311.
G
In his essay for this volume, Lars E.O. Svensson describes the importance of Goodfriend (1986) to the evolution of transparency at the Fed.
78
He wrote “Over the long run, the Fed’s credibility must be based on an understanding of how inflation targeting works rather than being based in the leadership of the
Fed. Making the Fed’s inflation-targeting procedures explicit would help to achieve
these ends by securing the Fed’s commitment to low inflation and improving the
transparency and accountability of the Fed for attaining its monetary policy objectives,” citing Broaddus and Roger Ferguson.
76
77

| 383

King and Lu
Goodfriend drew lessons from three distinct subperiods of postwar
US monetary history to make the case for the importance of credibility
for low inflation. The go/stop period from the later 1950s to the late
1970s “illustrates the consequences of failing to make low inflation a
priority for monetary policy. The Volcker period illustrates the difficulty
in restoring credibility for low inflation after it has been compromised.
And the Greenspan era illustrates how and why the Fed has come to
target low inflation implicitly in recent years.”79
Depreciating credibility during the 1960s and 1970s
Goodfriend attributed the go/stop monetary policy of the ’60s and
’70s to “the Fed’s inclination to be responsive to the shifting balance
of concerns between inflation and unemployment.” The consequence
was that “the trend rate of inflation tended to ratchet up with each go/
stop policy cycles.”
He highlighted the interplay of Fed policy, inflation, and expectations. In the “go” phase of the policy cycle, “the Fed did not tighten
policy early enough to preempt inflationary outbursts before they
became a problem” and then “pricing decisions ... embodied higher
inflation expectations.” In the “stop” phase of the policy cycle, “the Fed
would need a recession to bring inflation and inflation expectations
back down” but once the unemployment rate began to rise, the lack of
public support for tighter monetary policy made it “difficult to reverse
rising inflation.”80
 e wrote “Over the long run, the Fed’s credibility must be based on an understandH
ing of how inflation targeting works rather than being based in the leadership of the
Fed. Making the Fed’s inflation-targeting procedures explicit would help to achieve
these ends by securing the Fed’s commitment to low inflation and improving the
transparency and accountability of the Fed for attaining its monetary policy objectives,” citing Broaddus and Roger Ferguson.
79
Goodfriend (2004), p. 313.
80
Goodfriend (2004), p. 314.
78

384 |

Credibility and Inflation Targeting
The central problem of this period “was that the Fed tended to justify
its periodic inflation-fighting actions against an implicit objective
for low unemployment. In doing so, the Fed made monetary policy
a source of instability and wound up worsening both inflation and
unemployment.”81
Adverse consequences of credibility depreciation
Goodfriend saw the 1970s increase in “the level and volatility of
inflation and inflation expectations” as leading to “a breakdown of
mutual understanding between the Fed and the public: the public
could no longer discern the Fed’s policy intentions, and the Fed could
not predict how the economy would respond to its policy actions.” The
Volcker Fed “experienced the adverse consequences of a near total
collapse of credibility for low inflation, and learned how difficult it is to
pursue interest rate policy to restore credibility for low inflation once
that credibility has been thoroughly compromised.”82
He stressed two major consequences of the credibility decline for
Fed policies: it increasingly faced inflation scares and it became more
costly to reduce inflation.
Inflation scares Central bank economists have long been very alert
to the potential for sharp changes in expected inflation to be reflected
in longer term yields, a phenomenon that Goodfriend (1993) famously
labeled as “inflation scares.” In his inflation targeting manifesto, Goodfriend attributed the “inflation scares” during the Volcker era to the
Fed’s credibility problems. He saw these episodes as posing a “costly
dilemma” for the Fed “because ignoring them would encourage even
more doubt about the central bank’s commitment to low inflation.
Yet raising real short rates to restore credibility for low inflation risked
precipitating a recession.”83 Goodfriend identified four examples of
 oodfriend (2004), p. 315.
G
Goodfriend (2004), p. 315.
83
Goodfriend (2004), p. 318.
81
82

| 385

King and Lu
inflation scares during the Volcker era and highlighted how increased
credibility for low inflation could help in resolving such a dilemma. In
particular, “the Fed’s responses to the first two scares in 1980 and 1981
precipitated recessions in those years.” But the third inflation scare
(1983-1984) “demonstrated that a well-timed and well-calibrated series
of preemptive interest rate policy actions could defuse an inflation
scare without creating a recession,” indicating that “the Fed acquired
enormous additional credibility for low inflation during this period.”84
He identified a fourth inflation scare as beginning in March 1987,
suggesting that “it may have occurred in part because Volcker was near
the end of his term as chairman and there was doubt about whether
the Fed under Volcker’s successor would continue to place a high priority on low inflation. In any case, the 1987 scare is particularly striking
evidence of the fragility of the credibility of the Fed’s commitment to
low inflation.”
Costs of restoring credibility Goodfriend described two rounds of
federal funds rate hikes by the Volcker Fed to bring down inflation.
The first round started from September 1979 and ended in April 1980.
The second round began in early 1981 and lasted until the summer of
1982. In both cases, he argued, the interest rate moves precipitated
recessions, but inflation remained high in 1980 and stabilized in 1982.
Goodfriend explained: “The difference is that in 1980 the Fed cut the
federal funds rate sharply by around 8 percentage points between
April and July to act against the downturn, [...] The lesson of 1980 was
that the Fed could not restore credibility for low inflation if it continued to utilize interest rate policy to stabilize the output gap.”85 This led
him to assert two aspects of costs of restoring credibility. One is: “When
the Fed’s credibility for low inflation is in question, the Fed loses the
flexibility to use interest rate policy to stabilize output relative to its
potential.” Another is: “the Fed needs a recession to restore credibility
for low inflation after it has been compromised.”86
 oodfriend (2004), p. 318.
G
Goodfriend (2004), p. 316.
86
Goodfriend (2004), pp. 316-317.
84

Credibility and Inflation Targeting
Benefits of higher credibility
Goodfriend began his description of the Greenspan Fed by describing how it had dealt with two inflation scares, one in 1987 and the
other in 1994. He viewed the Greenspan Fed’s policy response to 1987
inflation scare as insufficiently preemptive in containing inflation and
resulting in a minor loss of credibility for low inflation, but he indicated that the “successful preemptive policy action in 1994 brought the
economy to virtual price stability. Inflation and inflation expectations
were anchored more firmly than ever before.”87
After reviewing these historical experiences, he followed by listing
three benefits of high credibility for low inflation in the second half of
the Greenspan era. “First, credibility helped the economy to operate
well beyond the levels that might have created inflation and inflation
scares in the past. Second, [...] [h]aving attained price stability, the Fed
did not need a recession to bring inflation and inflation expectations
down. ... Third, ... the fact that inflation and inflation expectations were
well anchored enabled the Greenspan Fed to cut the nominal federal
funds rate aggressively from 6.5 percent to 1.75 percent in 2001 to
cushion (a) fall in aggregate demand and employment...without a hint
of an inflation scare.”

Systematic policy under Greenspan
During the middle of Greenspan’s chairmanship, two important forces appear to have been pushing the FOMC toward systematic policy.
These elements are important background to Goodfriend’s characterization of that period as an implicit inflation targeting regime.
The Taylor rule
One force was the Taylor rule as an input to FOMC meetings. Interest
by the FOMC and by the Board staff had been stimulated by Taylor’s
(1993) finding that there was a coincidence between his proposed
rule and the behavior of the Fed funds rate over the 1987-1992 period.
Soon afterward, the Board staff preparation for each FOMC meeting

85

386 |

87

Goodfriend (2004), p. 320.

| 387

King and Lu
included a memo on funds rate settings under the original rule and
various modifications.88 According to Taylor’s (1993) presentation, the
intercept in his rule was a combination of the implicit long-run inflation target π* and the long-run level of the real interest rate r*, so that
an assumption about the latter allowed an identification of the former.
However, investigations of empirical Taylor rules revealed considerable
uncertainty attached to such estimate of π*.
The FRB-US model
Another force for systematic policy was the new “consistent expectations” model FRB-US, which immediately allowed for a greater range
of monetary policy scenarios and later for calculation of optimal policy.
FRB-US forecasts for inflation and output at various horizons came into
presentations of monetary policy alternatives (the Blue Book) by the
January 1997 meeting. FRB-US was also used to explore consequences
of high productivity growth and a gradual disinflation from the “stable
inflation” scenario of 2 percent to “price stability” at 1 percent.89
Implicit policy and public misinterpretations
But even with the new model as a sharper tool and the Taylor approach as a rule of thumb, the FOMC and its economists had concerns
about how alternative actions could affect the public’s perception of
its inflation target. In the January 1997 meeting, the staff noted that
the 50 basis point hike in the fed funds rate under the disinflation scenario would surprise market participants and worried that “in view
 eaction functions had long been estimated by Federal Reserve economists,
R
notably Stephen McNees at the FRB Boston (1986, 1992). But in the wake of Taylor
(1993), economists investigated whether there was a similar rule for earlier periods, including Mehra (1997) in Richmond. In San Francisco, John Judd investigated
interest rate policies across various periods. Initially, with Trehan (1995), he studied
whether the Fed had gotten tougher after 1980. Later, with Rudebush (1998), he
showed how Taylor rule estimates evolved across the Burns, Volcker and Greenspan
chairmanships. At the Board, Williams (1999) explored various simple policy rules
and Athanasios Orphanides built on his experience in preparing the FOMC memos
on Taylor rules to study interest rate rules across chairmanships, including analysis of
the Volcker and pre-Volcker periods (2003, 2004).
89
The first five years of the disinflation saw a decline to under 1.2 percent, while the full
transition took somewhat longer.
88

388 |

Credibility and Inflation Targeting
of the shift in policy direction represented by such an action, intermediate- and long-term rates could rise appreciably, especially if market
participants thought the Federal Reserve now saw significantly greater
inflation risks than they had inferred from recent statements by FOMC
members.”90
That is, the Board staff was concerned about the confounding of
unobserved shifts in the long-run inflation goal and unobserved other
reasons for varying the policy rate. This idea is at the heart of Erceg and
Levin’s (2003) explanation of the costliness of the Volcker disinflation,
within a model in which agents must learn if a disinflationary shift has
taken place. It is thus important to explore whether an explicit inflation targeting system could avoid the sort of “breakdown in mutual
understanding” suggested by the Erceg-Levin analysis and the FOMC
discussion.

Credibility, inflation, and real activity
In his manifesto, Goodfriend advocated for making the “Fed’s inflation-targeting procedures explicit in order to secure the commitment
to low inflation, enhance transparency, and improve the Fed’s accountability for attaining its monetary policy objectives.”91 We also have seen
that Goodfriend stressed the importance of credibility for the behavior
of inflation and the consequences of Federal Reserve policy for real
activity. We now explore the link between explicit inflation targeting
and credibility from the perspective of basic macroeconomic models,
including some of our own work. We also describe some new research
questions stimulated by our close reading of Goodfriend (2004).

What is credibility?
Various areas of economic research have proposed definitions of
credibility: in this discussion, we use a definition that we think Goodfriend would have found congenial and that has been productive for
J anuary 1997 Blue Book, https://www.federalreserve.gov/monetarypolicy/files/
fomc19970205bluebook19970131.pdf.
91
Goodfriend (2004), p. 313.

90

| 389

King and Lu
us in research, while recognizing that the literature on sustainable
plans92 and loose commitment93 provide other useful approaches.
One key ingredient for us, as for Blinder (2000), is that credibility
“involves matching deeds to words: a central bank is credible if people believe it will do what it says.” Making this definition operational
requires measures of a central bank’s actions and its statements about
these actions. It also requires measures of private sector beliefs. We
have seen that Goodfriend viewed the extent of “credibility for commitment to low inflation” as varying widely over time and responding
to central bank actions and macroeconomic outcomes. To capture
such evolving partial credibility of central bank announcements and
actions, our work employs a reputational state variable that governs
the extent of credibility at a point in time: this is a private sector likelihood that the central bank is of a type that can commit and matches
deeds with words. This reputational state variable is governed by
Bayesian learning.
Our definition of credibility also accords with another important
aspect of Goodfriend’s inflation targeting manifesto. He views explicit
inflation targeting as a means of protecting the economy against both
deflation and inflation, as a result of specifying a 1 to 2 percent range
for inflation. Thus, it is credibility for policy consistency with a publicly
announced framework rather than to a specific outcome such as “low
inflation” that is central to Goodfriend and to us.
 ne prominent line of research essentially requires perfect credibility of central
O
bank choices and macroeconomic outcomes in order to focus on expectations as
disciplining a central bank that cannot commit. Central bank actions are a part of a
“sustainable plan,” in the terminology of Chari and Kehoe (1990) when these will be
carried out even though the central bank cannot commit to its future actions and
when it would renege on other plans. Like Goodfriend, these authors highlight the
importance of credibility to low inflation, but studies adopting this approach typically do not feature time-varying credibility.
93
Another important line of research assumes that the current central bank can commit but faces a time-invariant probability of a future regime change in which a new
committed central bank will reoptimize its inflation plans (Schaumburg and Tambalotti, 2007; Debortoli, Maih, and Nunes, 2014). This approach highlights the idea that
imperfect credibility limits expectations management by a committed central bank
but does not incorporate time-varying credibility that is influenced by its decisions.
92

390 |

Credibility and Inflation Targeting
The importance of understanding macroeconomic equilibria when
a central bank cannot commit, of course, came to the forefront with
the work of Kydland and Prescott (1977) and Barro and Gordon (1983a,
1983b), henceforth KPBG. In a basic equilibrium without commitment,
an inflation bias arises when the central bank objective makes real
stimulus desirable and the central bank (appropriately) treats inflation
expectations as beyond its control. The temptation to inflate leads
to excessive inflation.94 We have found that models with such 1980s
linkages between inflation and real activity are very tractable, and we
therefore use them as starting points in thinking about imperfect credibility and its consequences.
Modern New Keynesian macro models with forward-looking price
dynamics generally possess such an inflation bias, but also contain
stabilization biases in equilibria without commitment (see, for example, Clarida, Gali, and Gertler, 1999). Under commitment, by contrast,
optimal monetary policy in such models generally leads to low and
relatively stable inflation. In the basic textbook NK model and direct
elaborations of it, optimal policy more specifically leads to a long-run
price level path that is not much affected by various shocks.95 Although
these intertemporal models are more complicated, we study them
because they are arguably more realistic and certainly more akin to
policy models actually in use at many central banks.

Why does credibility evolve?
The analysis of monetary policy with evolving imperfect credibility
was initiated by Alex Cukierman in the late 1980s and early 1990s using an elaboration of the KPBG model.96 In our early work on managing
expectations with imperfect credibility and evolving reputation, we
I nterestingly, the 1975 Economic Report of the President discusses the inflation process in these terms. Presumably, it was written by either Alan Greenspan (CEA chief )
or William Fellner, the Yale professor who was one of the first economists to explore
combining rational expectations with Keynesian mechanisms, as well as seeing a
central role for policy credibility.
95
See, for example, the appendix to Clarida, Gali, and Gertler (1999), Khan, King, and
Wolman (2003), and Woodford (2005).
96
The best single example is Cukierman and Liviatan (1991), and his work is collected
in Cukierman (1992).
94

| 391

King and Lu
used a KPBG-style model to examine the interplay of announcements,
actions, outcomes, reputation and credibility.97 Our more recent work
features a forward-looking New Keynesian Phillips curve and two types
of central banks, one that can commit and another that cannot.98 Crucially, although only one type of central bank is present in any period,
private agents do not know the type that is in place but must learn
about it from macroeconomic outcomes that central banks control
only imperfectly.
Central bank announcements play an important role in such theories
but in a subtle manner. We assume that the committed central bank
can accurately communicate its intended inflation to the private sector
via an announcement, which is usefully interpreted as an inflation
target.99 Since the announcement is important for the beliefs of the
private sector, a central bank that cannot commit must also make the
same announcement or its type will be disclosed leading to adverse
shifts in inflation expectations. The key state variable — reputation —
rises and falls as inflation outcomes differ from announced targets.
The literature offers some specific definitions of credibility in this
context. Cukierman and Meltzer (1986) define it as “the absolute distance between the policymaker’s plans and agent’s beliefs about those
plans.” Given that inflation is imperfectly controllable by the central
bank, another definition is the likelihood that actual inflation will be
within a band around planned inflation target.100 In basic models, each
of these credibility measures is dependent on central bank reputation
for commitment: when reputation is low, there are large absolute gaps
between actual and perceived plans as well as low likelihood of small
inflation deviations from announced targets.
 ing, Lu, and Pasten (2008).
K
Lu, King, and Pasten (2016), and King and Lu (2021).
99
Using a similar model with two types of policymakers, one who can commit and
another who cannot, Lu (2013) proves that the unique equilibrium announcement is
the optimal policy for the committed type.
100
This latter credibility measure seems closer to Goodfriend’s narrative and is consistent with the perspective of King (2005), who writes that “credibility is not an all-ornothing matter. Policy is neither credible nor incredible. It is, as we say in economics,
a continuous variable.”
97

Credibility and Inflation Targeting

Managing expectations with imperfect credibility
Goodfriend saw credible disinflations as relatively costless based on
both the 1980s New Classical and the 1990s New Keynesian models
of inflation dynamics. Yet, in describing the conquest of inflation after
the Great Inflation of the 1970s, he portrayed the Fed as “need(ing) a
recession to bring inflation and inflation expectations back down,” as
in the “the Volcker disinflation from 1979 to 1987.”101 He described the
Volcker-led Fed as having “experienced the adverse consequences of
a near total collapse of credibility for low inflation” and the process of
restoring its reputation (“credibility for low inflation”) as difficult “once
that credibility has been thoroughly compromised.” Looking at the
Volcker and Greenspan chairmanships, Goodfriend also portrayed the
Fed as delicately balancing the benefits of inflation reduction with the
costs imposed on the real economy as the Fed sought to lower inflation and increase its credibility.
With a committed central bank having an imperfect reputation for
commitment, our models only feature imperfect reputation as the
reason for partial credibility. Our theoretical models102 deliver three
implications related to these key ideas in Goodfriend (2004). A first
implication is that its announced policies have a lower leverage over
inflation expectations, because individuals attach some likelihood to
these not being carried out. This leads to a second implication that
imperfectly credible disinflations are costly even when credible disinflations are not, because implementing disinflation amid high inflation
expectations precipitates a recession. The third implication of these
models is that it is costly to build reputation if the initial reputation is
low. This is consistent with Goodfriend’s observation that “(t)he lesson
of 1980 was that the Fed could not restore credibility for low inflation if
it continued to utilize interest rate policy to stabilize the output gap.”103

98

392 |

T homas Sargent reviews Goodfriend’s work (with King, 2005) on “The Incredible
Volcker Disinflation” in an essay for this volume.
102
King, Lu and Pasten (2008), Lu, King, and Pasten (2016), and King and Lu (2021).
103
Goodfriend (2004), p. 316.
101

| 393

King and Lu

Is evolving credibility relevant?
As we discussed in the earlier section “Goodfriend’s inflation targeting manifesto,” Goodfriend’s historical narrative depicts US inflation
history from the mid-1960s to the early 2000s, highlighting evolving
credibility in various episodes. Our most recent work (King and Lu,
2021) provides a model-based perspective on inflation over this period.
As discussed previously in general terms, we consider a minor variation on the standard New Keynesian model with two types of central
banks: one that can commit and another that cannot. More specifically,
the central bank that cannot commit is myopic, responding to inflation expectations and shifts in the NK Phillips curve. By contrast, the
committed type is an explicit inflation targeter that recognizes private
sector skepticism about its type and takes actions with an eye to managing its reputation. We assume a single switch in type, with commitment starting around January 1981.
To make the model quantitative, we extract two state variables
(reputation and shifts to the NK Phillips curve) from one-quarter and
three-quarter expected inflation based on the Survey of Professional
Forecasters. Although the procedure does not make use of actual inflation data, we find that the framework gives a reasonable account of
the rise, fall, and stabilization of inflation over 1965-2005. Notably, the
inflation of the 1960s occurs even though the central bank has a very
small amount of intrinsic inflation bias, defined as the extent to which
it seeks to stimulate inflation and real activity with expected inflation
held fixed. But, as an initially high reputation declines and inflation
expectations rise, a large equilibrium inflation bias occurs, leading
to an ultimate inflation peak in the late 1970s. When the Volcker Fed
moves to reduce inflation in the framework, it faces low reputation and
cannot effectively manage expected inflation. Overall, the evolution
of reputation, with implications for the credibility measures discussed
above, is a first order feature of the period.104

Credibility and Inflation Targeting
Of course, such a basic exercise in quantitative theory inevitably
raises as many questions as it answers, but it suggests that evolving
credibility may ultimately be key to understanding the US inflation
experience in a more detailed manner.

Perils of implicit targeting
Under an implicit inflation targeting system, the central bank does
not announce its inflation and output intentions as it does in an
explicit targeting regime. Private agents therefore face uncertainty
about policy: it could be an unobserved long-run inflation goal (as in
the analysis of Faust and Svensson [2001]) or an unobserved plan for
returning inflation to the long run goal. Such uncertainty is relevant for
private agents directly but also for the central bank (recall the earlier
discussion of the Board staff concerns about misinterpretations in the
January 1997 Blue Book).
Our close reading of Goodfriend (2004) for this volume has led us to
confront how very different an implicit targeting regime is as well as
the perils that it could represent. To investigate elements of his analysis, we were therefore led to explore basic concepts and to develop a
simple model.
To fix ideas, it is useful to start with thinking about how a committed
central bank chooses optimal policy in the KPBG model. It acts before
private agents set their expected price level (inflation rate) and knows
how they will respond to its alternative policy actions. Hence, it chooses low inflation and abstains from seeking a positive output or employment gap. In the sense familiar from microeconomics, the central bank
is a Stackelberg leader and the private sector is a follower. By contrast,
without commitment, the inflation bias solution obtains. As Sargent
and Soderstrom (2000) have stressed, this may be viewed as a Nash
equilibrium of the simultaneous game between the central bank and
the private sector.105
105

104

394 |

 otably, the two credibility measures discussed above (one that captures a credibilN
ity gap and the other that captures an extent of credibility) vary dramatically but are
highly correlated. Each depends positively on the reputation state variable.

I t is more frequently represented as a game in which the central bank moves after
the private sector. Notably, Barro and Gordon (1983a) describe inflation expectations as set “at the start of the period” and the inflation action as chosen “during the
period” (see pp. 595-596).

| 395

King and Lu
An extreme version of peril
An important, but underappreciated, literature in game theory thus
becomes relevant to thinking about implicit inflation targeting. Bagwell (1995) observes that “the advantage from commitment [springs]
from a combination of two assumptions. First, moves in the game
are sequential, with some players committing to actions before other
players select their respective actions, and, second, the late-moving
players perfectly observe the actions selected by the first movers.
These assumptions are so frequently combined that it is easy to forget
that they are not equivalent.”
To break apart commitment and observability, Bagwell develops a
noisy leader game, in which one player moves first and then a second
player observes a signal of the first mover’s actual selection before
making his own move.” He establishes a striking result: “With even the
slightest degree of imperfection in the observability of the first-mover’s selection, therefore, the strategic benefit of commitment is lost.” An
implication of Bagwell’s analysis is thus an implicit inflation targeting
regime implemented by a committed central bank — but one that
cannot accurately communicate its intended inflation — would give
rise to the same outcomes as if the central bank cannot commit.106
This is an extreme peril from implicit inflation targeting: it formally captures a complete breakdown of the mutual understanding
between the Fed and the private sector that marks a commitment
regime.

Credibility and Inflation Targeting
We therefore next explore the possibility that such limited reports
are noisy signals about the Fed’s planned actions that would be reported more precisely and more frequently under an explicit targeting
regime. We make use of an important paper by Maggi (1999), who
developed a leader follower game with two types of randomness: (i)
noise as in Bagwell; and (ii) a privately observed shock to the leader’s
objectives. Maggi establishes that a small amount of noise is no longer
totally destructive to the first-mover advantage in the same duopoly
game studied by Bagwell. In an example with linear decision rules and
normal shocks, he shows that a reduction in the amount of noise pushes the outcome toward the Stackelberg outcome in the duopoly game.
Analyzing the basic KPBG model using a similar approach, we have
been able to study the effects of changing the extent of noise about
the central bank’s planned actions when its long-run inflation target is
subject to a privately observed shock.107 In examples when the equilibrium is unique, low noise cases resemble the commitment solution
and high noise cases resemble the solution without commitment.
Thus, an implicit inflation targeting system works better when it
more closely resembles an explicit inflation targeting system. This
conclusion dovetails nicely with increasing transparency of instrument
choices under Greenspan and also with ideas in Goodfriend’s manifesto:
	Openly clarifying the priority for price stability would reinforce the
Fed’s commitment to low inflation and enhance the credibility of that
commitment. It would balance the recently increased transparency
of the Fed’s interest rate instrument with greater transparency of its
low-inflation goal. And it would act to defuse further the idea that secrecy has any role to play in monetary policy (see Goodfriend 1986).
In this regard, the Fed could go further and publicly acknowledge its
quantitative working definition of long-run price stability. If a 1 to 2
percent range for core PCE inflation is it, then the Fed could acknowledge that it intends to keep core PCE inflation in or near that range
indefinitely.108

Noisy signals about Fed intentions
Explicit inflation targeting regimes frequently have featured high visibility, quarterly “Inflation Reports” that were not part of the Greenspan
regime. But, in his comments on Goodfriend’s manifesto, Kohn (2004)
stressed that the Fed did provide considerable information about
monetary policy in its January and July “Humphrey Hawkins” reports to
Congress.
106

I t was to avoid this implication that King, Lu and Pasten (2008) assumed that the
committed type could make an accurate announcement about its inflation intention (see footnote 24, p. 1650) with a relevant game theory reference.

107
108

396 |

 e report them in greater detail in King and Lu (2022a).
W
Goodfriend (2004), p. 332.

| 397

Credibility and Inflation Targeting

King and Lu
Another finding of our simple KPBG example with noisy signals is
that multiple equilibria are present when the Phillips curve is flat or the
central bank places high weight on output losses. The nature of these
equilibria is quite intuitive: if private agents place little weight on the
noisy signals about central bank planned actions, their expectations
are less manageable; but if the central bank views expectations as
less responsive to its planned actions, its optimal actions become less
responsive to shocks in its long-run inflation target and, in turn, the
signals are less informative for the private sector. Such equilibria with
low information transmission appear to capture a less extreme “breakdown of mutual understanding between the Fed and the public.” There
is a decline in the credibility of central bank statements and weakened
effect of its actions even when the central bank has full commitment
capacity.
Scares and misinterpretations
A framework with noisy policy messages also appears to provide a
simple explanation of inflation scares, which were introduced in Goodfriend (1993) and are much discussed in his manifesto. In the simple
model outlined earlier, we assume that the private sector receives a
message that is the policymaker’s planned inflation plus a zero mean
random noise. That random noise could be given an economic interpretation as vagueness in Humphrey-Hawkins testimony; accidents
of language by a Fed chairman, governor or regional bank president;
misinterpretation of the Fed communication by market observers; etc.
When such a noise occurs, it will appear to the Fed that private sector
inflation expectations have become irrationally scared. But, fundamentally, the problem arises from the Fed’s own lack of explicitness.
The problem is not just “inflation scares.” Goodfriend (2004, p. 326)
sees the potential for “destabilizing deflation scares,” suggesting that
“announcing an explicit lower bound on inflation would make the
public more confident that the Fed will not allow the United States to
fall into a Japanese-style deflation, zero-bound trap.”
If one were to move beyond a simple one-period model, then it
would become possible to confront the concerns that the Board’s

economists expressed in the January 1997 Bluebook, which was that
a rise in the funds rate — a tightening of monetary policy at the start
of a disinflation — would be misinterpreted as reflecting heightened
Fed concerns about near-term inflation that had to be preempted.
An explicit targeting mechanism could well reduce or eliminate such
misinterpretations.

Explicit targeting and credibility, again
Our theoretical review of Goodfriend’s inflation targeting manifesto
has led us to conclude that noisy policy messages in an implicit inflation targeting regime have similar, if not more serious, detrimental
effects as other forms of imperfect credibility, by reducing the central
bank’s leverage over inflation expectations and the effectiveness of its
stabilization policies.
Explicitly committing to inflation targets — including flexible inflation targeting — avoids the perils of implicit targeting and helps the
central bank to acquire and maintain credibility for attaining its monetary policy objective. This is a form of central bank credibility that both
Goodfriend and our recent work have shown to play an important role
in the history of US monetary policy and, very likely, in the challenging
times of today.

Concluding thoughts
In the nearly two decades since Marvin Goodfriend’s inflation
targeting manifesto, there have been important changes in the Fed’s
monetary policy framework and practice. Continuing the trend toward
improvement in communications about policy actions and intentions,
the Fed in 2003 began forward guidance with respect to the funds
rate, with an eye to influencing the short end of the term structure of
interest rates.109 In the fall of 2007, after Ben Bernanke became chair,
the Fed started to publish quarterly summaries of policy projections
made by FOMC members, providing their views about how output,
inflation, and the funds rate would evolve over three years under their
preferred policy actions. This was a welcome and important step
109

398 |

See Poole (2007), p. 10.

| 399

King and Lu
toward the communication practices of inflation targeting, consistent
with Marvin Goodfriend’s call for increased transparency, although
the FOMC did not consolidate these “straw votes” into a committee
consensus. Fed officials also stressed that the projections were “not a
commitment” at the time and have continued to do so more recently.
Rather, as Bernanke stressed in 2016, “If the FOMC as a whole is going
to make a commitment or provide explicit guidance about future rate
policy, it will do that in its post-meeting statement, or the chair will
communicate it.”110 In 2012, the Fed formally introduced a 2 percent
inflation target, not too different from the median longer-run PCE inflation projections by FOMC members over 2009-2011.111 112
During the global financial crisis and its aftermath, with the funds
rate near zero, the Fed used forward guidance about future interest
rates with the objective of raising real activity, inflation, and inflation
expectations. Bernanke (2020) provides an overview of the application
of “new tools of monetary policy” during this period. Despite extensively applying these new tools, the Fed’s expansionary policies led to
inflation that tracked below the long-run goal of 2 percent for much
of a decade. The theory of optimal monetary policy at the zero lower
bound as developed in Eggertsson and Woodford (2003) highlighted
the desirability of price level (path) targeting, as inflation lower than
target would need to be matched with higher future inflation.113 But,
combining theory and evidence, Bodenstein, Hebden, and Nunes
(2012) found that the benefits were substantially curtailed by credibility difficulties in both the US and Sweden. In December 2012, the Fed
moved to “threshold-based forward guidance” that specified that the
policy rate would remain near zero until unemployment was reduced
T hese quotations are taken from a 2016 Brookings blog post by Bernanke, https://
www.brookings.edu/blog/ben-bernanke/2016/11/28/federal-reserveeconomic-projections/.
111
Shapiro and Wilson (2021) estimate that the FOMC had an implicit inflation target of
1.5 percent over a baseline sample of 2000-2011.
112
Lacker (2020) providers an insider’s account of the FOMC adoption of the target.
113
Federal Reserve economists had earlier explored interest rate rules of the form
suggested by their analysis, including Wolman (1998) using a forward-looking staggered pricing model and David Reifschneider and Williams (2000) using the FRB-US
model.

Credibility and Inflation Targeting
below 6.5 percent, so long as near-term inflation expectations did not
run much above the 2 percent long-run inflation target and long-term
inflation expectations continued to be well anchored.114
In August 2020, the Fed moved to flexible average inflation targeting (FAIT).115 As it moved toward the new regime, the Fed had a very
open process in which some noted economists and well-known former
senior Fed staffers called for a permanently higher inflation target to
decrease the likelihood of zero lower bound events. The consideration
of a higher long-run target for inflation was mainly motivated by declining estimates of r*, as the work of Laubach and Williams (2003) was
updated by Fed and other economists.
Ultimately, though, the new policy framework reaffirmed that 2
percent PCE inflation in the long run was “most consistent with the
Federal Reserve’s statutory mandate.”116 Yet, the Fed also announced
that “in order to anchor longer-term inflation expectations at this level,
the Committee seeks to achieve inflation that averages 2 percent over
time, and therefore judges that, following periods when inflation has
been running persistently below 2 percent, appropriate monetary
policy will likely aim to achieve inflation moderately above 2 percent
for some time.” At the time, senior Fed officials also signaled that there
would no longer be a preemptive approach to inflation management
and that there would be an increased priority attached to “mitigating
short falls of employment from its maximum level.”117
The description of the new policy regime made clear that the Fed
would seek or tolerate above 2 percent inflation for some time, after
periods in which inflation ran below 2 percent, as it frequently did between 2008 and 2020. However, the Fed’s policy announcement did

110

400 |

 ttps://www.federalreserve.gov/monetarypolicy/timeline-forward-guidance-abouth
the-federal-funds-rate.htm.
115
Nessen and Vestin (2005) introduce average inflation targeting, which provides an
intermediate policy between basic inflation targeting, with target misses being bygones, and price level (path) targeting, with misses being fully offset subsequently.
116
FRB statement on revisions, https://www.federalreserve.gov/monetarypolicy/guideto-changes-in-statement-on-longer-run-goals-monetary-policy-strategy.htm.
117
See, for example, a September 2020 speech by Lael Brainard, https://www.federal
reserve.gov/newsevents/speech/brainard20200901a.htm.
114

| 401

King and Lu
not specify how much higher than 2 percent inflation might run or
for how long it would tolerate inflation above that level. In advance
of the first FOMC meeting of the new regime in September 2020, a
well-informed economics team at IHS Markit wrote in their pre-FOMC
briefing: “[The Statement] was vague in key respects, beginning with
the definition of what constitutes a ‘moderate’ overshoot of 2 percent
inflation. Is 2¼% sufficient? 2½%? 3%? For how long? How much of the
previous inflation undershoot does the committee intend to offset? In
other words, how far does the look-back period extend? What is the
horizon for making up past inflation misses?” Later, in January 2021,
the IHS Markit team summarized a clarification of the new regime’s
nature by Vice Chairman Richard Clarida: it was “temporary price-level
targeting (TPLT) at the effective lower bound that reverts to flexible
inflation targeting (FIT) once the conditions for lift-off have been
reached.”118 They also highlighted the part in Clarida’s speech that “Inflation averaging 2% over time is an ‘ex ante aspiration’, not an ex post
commitment.”
The lack of explicit commitment to a path of inflation leading back to
the long-run goal and the fuzziness of the average inflation targeting
mechanism are sources of ambiguity and confusion that will be reflected in market expectations about real activity and inflation. In turn,
these features of the new policy framework open the door to erosion
of credibility, especially in light of recent calls for a higher long-run
inflation target.

Credibility and Inflation Targeting
instruments and their inflation-indexed counterparts, on the other
hand, are standing at 2.85 percent for the 10-year rate, 2.81 percent
for the 20-year rate, and 2.49 for the 30-year rate. Relative to their
December 2021 levels,120 the 10-year rate is up by 39 bps, the 20-year
rate by 30 bps, and the 30-year rate by 22 bps. We view this as likely an
increase in long-run inflation expectations by market participants that
differs from the stable long-run projection by the Fed.
We end this essay with another quote from Goodfriend’s inflation
targeting manifesto: “if inflation moves outside its long-run target
range, [...] the cost, in terms of lost output relative to potential, of returning inflation to its long-run range depends on the credibility of the
Fed’s commitment to do so. The historical record [..] suggests that such
credibility is sensitive to the Fed’s actions themselves [...] In any particular case the Fed must judge the extent to which drawing out the
return of inflation to its long-run target might be counterproductive
by reducing the credibility of its intention to bring inflation all the way
back down. That consideration must be balanced against attempting
to bring inflation down before the credibility for doing so has been
built up. An error in either direction would increase the output cost of
restoring price stability.”121

As we write in spring 2022, the most recent Summary of Economic Projections shows that the median FOMC member sees a path of declining
PCE inflation under their preferred policy: 4.3 percent for 2022, 2.7
percent for 2023, 2.4 percent for 2024, and a longer run goal of 2 percent.119 The most recent breakeven inflation rates on standard Treasury
T his led the IHS team to refer to the new regime as asymmetric flexible average
inflation targeting. See also, speech by Clarida, https://www.federalreserve.gov/
newsevents/speech/clarida20201116a.htm.
119
The December 2021 projections were 2.6, 2.3, 2.1, and 2.0. https://www.federal
reserve.gov/monetarypolicy/files/fomcprojtabl20220316.pdf.
118

402 |

120
121

The December 2021 levels were 2.46, 2.51, and 2.27.
Goodfriend (2004), pp. 327-328.

| 403

King and Lu

References
Akerlof, George A., William T. Dickens, George L. Perry, Robert J. Gordon, and N. Gregory Mankiw. 1996. “The Macroeconomics of Low
Inflation.” Brookings Papers on Economic Activity 1: 1, no. 27: 1-76.
Aoki, Kosuke. 2001. “Optimal Monetary Policy Responses to Relative-Price Changes.” Journal of Monetary Economics 48, no. 1 (August):
55-80.
Bagwell, Kyle. 1995. “Commitment and Observability in Games.” Games
and Economic Behavior 8, no. 2: 271-280.
Barro, Robert J., and David B. Gordon. 1983a. “A Positive Theory of Monetary Policy in a Natural Rate Model.” Journal of Political Economy 91, no.
4 (August): 589-610.
Barro, Robert J., and David B. Gordon. 1983b. “Rules, Discretion and
Reputation in a Model of Monetary Policy.” Journal of Monetary Economics 12, no. 1 (July): 101-121.
Bernanke, Ben, and Mark Gertler. 1999. “Monetary Policy and Asset
Price Volatility.” Federal Reserve Bank of Kansas City Economic Review
84, no. 4 (Fourth Quarter): 17-51.
Bernanke, Ben S. 2020. “The New Tools of Monetary Policy.” American
Economic Review 110, no. 4 (April): 943-983.
Blinder, Alan S. 2000. “Central-Bank Credibility: Why Do We Care? How
Do We Build It?” American Economic Review 90, no. 5 (December): 14211431.
Bodenstein, Martin, Christopher J. Erceg, and Luca Guerrieri. 2008.
“Optimal Monetary Policy with Distinct Core and Headline Inflation
Rates.” Journal of Monetary Economics, Contributions to Macroeconomics
in Honor of John Taylor, 55 (October): S18-S33.
Bodenstein, Martin, James Hebden, and Ricardo Nunes. 2012. “Imperfect Credibility and the Zero Lower Bound.” Journal of Monetary Economics 59, no. 2 (March): 135-149.
404 |

Credibility and Inflation Targeting
Broaddus, J. Alfred, and Marvin Goodfriend. 1985. “Base Drift and the
Longer Run Growth of M1: Experience from a Decade of Monetary
Targeting.” Federal Reserve Bank of Richmond Working Paper 85-1,
January 1.
Calvo, Guillermo A. 1983. “Staggered Prices in a Utility-Maximizing
Framework.” Journal of Monetary Economics 12, no. 3 (September):
383–398.
Chari, V. V., and Patrick J. Kehoe. 1990. “Sustainable Plans.” Journal of
Political Economy 98, no .4 (August): 783-802.
Clarida, Richard, Jordi Gali, and Mark Gertler. 1999. “The Science of
Monetary Policy: A New Keynesian Perspective.” Journal of Economic
Literature 34, no. 4 (December): 1661–1707.
Cochrane, John H. 2011. “Determinacy and Identification with Taylor
Rules.” Journal of Political Economy 119, no. 3 (June): 565-615.
Cook, Timothy, and Thomas Hahn. 1989. “The Effect of Changes in the
Federal Funds Rate Target on Market Interest Rates in the 1970s.” Journal of Monetary Economics 24, no. 3 (November): 331-351.
Cukierman, Alex. 1992. Central Bank Strategy, Credibility, and Independence: Theory and Evidence. Cambridge: MIT Press.
Cukierman, Alex, and Nissan Liviatan. 1991. “Optimal Accommodation
by Strong Policymakers under Incomplete Information.” Journal of
Monetary Economics 27, no. 1 (February): 99-127.
Cukierman, Alex, and Allan H. Meltzer. 1986. “A Theory of Ambiguity,
Credibility, and Inflation under Discretion and Asymmetric Information.” Econometrica 54, no. 5 (September): 1099-1128.
Debortoli, Davide, Junior Maih, and Ricardo Nunes. 2014. “Loose Commitment in Medium-Scale Macroeconomic Models: Theory and Applications.” Macroeconomic Dynamics 18, no. 1 (January): 175–198.
Eggertsson, Gauti B., and Michael Woodford. 2003. “The Zero Bound
on Interest Rates and Optimal Monetary Policy.” Brookings Papers on
Economic Activity 1: 139–211.
| 405

King and Lu
Erceg, Christopher J., Dale W. Henderson, and Andrew T. Levin. 2000.
“Optimal Monetary Policy with Staggered Wage and Price Contracts.”
Journal of Monetary Economics 46, no. 2 (October): 281-313.
Erceg, Christopher J., and Andrew T. Levin. 2003. “Imperfect Credibility and Inflation Persistence.” Journal of Monetary Economics 50, no. 4
(May): 915-944.
Evans, Charles L. 2012. “Monetary Policy in a Low-Inflation Environment: Developing a State-Contingent Price-Level Target.” Journal of
Money, Credit and Banking 44, no. s1 (February): 147-155.
Faust, Jon, and Lars E.O. Svensson. 2001. “Transparency and Credibility: Monetary Policy with Unobservable Goals.” International Economic
Review 42, no. 2 (May): 369-397.
Giannoni, Marc P., and Michael Woodford. 2004. “Optimal Inflation-Targeting Rules.” In The Inflation-Targeting Debate, edited by Ben S. Bernanke and Michael Woodford, 93-172. Chicago: University of Chicago
Press.

Credibility and Inflation Targeting
Goodfriend, Marvin. 1997. “A Framework for the Analysis of Moderate
Inflations.” Journal of Monetary Economics 39, no. 1 (June): 45-65.
Goodfriend, Marvin. 1998. “Using the Term Structure of Interest Rates
for Monetary Policy.” Federal Reserve Bank of Richmond Economic
Quarterly 84, no. 3 (Summer): 13-30.
Goodfriend, Marvin. 2002. “The Phases of U.S. Monetary Policy: 1987 to
2001.” Federal Reserve Bank of Richmond Economic Quarterly 88, no. 4
(Fall): 1-16.
Goodfriend, Marvin. 2004. “Inflation Targeting in the United States?” In
The Inflation-Targeting Debate, edited by B. Bernanke and M. Woodford,
311-352. Chicago: The University of Chicago Press.
Goodfriend, Marvin, Gary Anderson, Anil Kashyap, George Moore, and
Richard D. Porter. 1986. “A Weekly Rational Expectations Model of the
Nonborrowed Reserve Operating Procedure.” Federal Reserve Bank of
Richmond Economic Review January/February): 11-28.

Giannoni, Marc P. 2014. “Optimal Interest-Rate Rules and Inflation Stabilization versus Price-Level Stabilization.” Journal of Economic Dynamics and Control 41 (April):110-129.

Goodfriend, Marvin, and Robert G. King. 1997. “The New Neoclassical
Synthesis and the Role of Monetary Policy.” In NBER Macro Annual 1997,
Volume 12, edited by Ben S. Bernanke and Julio Rotemberg, 231-296.
Cambridge: MIT Press.

Goodfriend, Marvin. 1982. “A Model of Money Stock Determination
with Loan Demand and a Banking System Balance Sheet Constraint.”
Federal Reserve Bank of Richmond Economic Review (January/February): 13-16.

Goodfriend, Marvin, and Robert G. King. 2001. “The Case for Price Stability.” In Why Price Stability? Proceedings of the First ECB Central Banking
Conference, edited by A. Garcia-Herrero, V. Gaspar, L. Hoogduin, J. Morgan, and B. Winkler, 53-94. Frankfurt, Germany: European Central Bank.

Goodfriend, Marvin. 1986. “Monetary Mystique: Secrecy and Central
Banking.” Journal of Monetary Economics 17, no. 1 (January): 63-92.

Gürkaynak, Refet S., Brian Sack, and Eric Swanson. 2005. “The Sensitivity of Long-Term Interest Rates to Economic News: Evidence and
Implications for Macroeconomic Models.” American Economic Review
95, no. 1 (March): 425-436.

Goodfriend, Marvin. 1987. “Interest Rate Smoothing and Price Level
Trend-Stationarity.” Journal of Monetary Economics 19, no. 3 (May): 335348.
Goodfriend, Marvin. 1993. “Interest Rate Policy and the Inflation Scare
Problem: 1979–1992.” Federal Reserve Bank of Richmond Economic
Quarterly 79, no. 1 (Winter): 1-23.
406 |

Judd, John, and Bharat Trehan. 1995. “Has the Fed Gotten Tougher on
Inflation?” Federal Reserve Bank of San Francisco Economic Letter 95-13,
March 31.
Judd, John P., and Glenn Rudebusch. 1998. “Taylor’s Rule and the Fed,
| 407

King and Lu
1970-1997.” Federal Reserve Bank of San Francisco Economic Review 3:
3-16.
Kerr, William, and Robert G. King. 1996. “Limits on Interest Rate Rules in
the IS Model.” Federal Reserve Bank of Richmond Economic Quarterly
82, no. 2 (Spring): 47-75.
Khan, Aubhik, Robert G. King, and Alexander L. Wolman. 2003. “Optimal Monetary Policy.” Review of Economic Studies 70, no. 4 (October):
825-860.
King, Mervyn. 2005. “Credibility and Monetary Policy: Theory and Evidence.” Bank of England Quarterly Bulletin (First Quarter): 84-91.
King, Robert G., and Yang K. Lu. 2021. “The Rise, Fall and Stabilization
of US Inflation: Shifting Regimes and Evolving Reputation.” National
Bureau of Economic Research Working Paper 29585, December.
King, Robert G., and Yang K. Lu. 2022a. “Notes on Interest Rates, Inflation Targeting and Optimal Monetary Policy.”
King, Robert G., and Yang K. Lu. 2022b. “Notes on The Perils of Implicit
Inflation Targeting.”
King, Robert G., Yang K. Lu, and Ernesto S. Pasten. 2008. “Managing Expectations.” Journal of Money, Credit and Banking 40, no. 8 (December):
1625-1666.
King, Robert G., and Alexander L. Wolman. 1996. “Inflation Targeting
in a St. Louis Model of the 21st Century.” Federal Reserve of St. Louis
Review 78, no. 3 (May): 83-107.
King, Robert G., and Alexander L. Wolman. 1999. “What Should the
Monetary Authority Do When Prices Are Sticky?” In Monetary Policy
Rules, edited by John B. Taylor, 349-404. Chicago: University of Chicago
Press.
Kohn, Donald L. 1996. “Commentary: What Operating Procedures
Should Be Adopted to Maintain Price Stability?—Practical Issues.” Federal Reserve Bank of Kansas City Economic Policy Symposium August
29-31.
408 |

Credibility and Inflation Targeting
Kohn, Donald L. 2004. “Comment on ‘Inflation Targeting in the United
States?’” In The Inflation-Targeting Debate, edited by B. Bernanke and M.
Woodford, 337-350. Chicago: University of Chicago Press.
Kydland, Finn E., and Edward C. Prescott. 1977. “Rules Rather than Discretion: The Inconsistency of Optimal Plans.” Journal of Political Economy 85, no. 3 (June): 473-492.
Lacker, Jeffrey M. 2020. “A Look Back at the Consensus Statement.” Cato
Journal, 285-319.
Laubach, Thomas, and John C. Williams. 2003. “Measuring the Natural
Rate of Interest.” Review of Economics and Statistics 85, no. 4 (November): 1063-1070.
Leeper, Eric M. 1991. “Equilibria Under ‘Active’ and ‘Passive’ Monetary
and Fiscal Policies.” Journal of Monetary Economics 27, no. 1 (February):
129-147.
Lindsey, David E., Athanasios Orphanides, and Robert H. Rasche. 2004.
“The Reform of October 1979: How It Happened and Why.” Federal
Reserve Bank of St. Louis Working Paper 2004-033A, October
Lu, Yang K. 2013. “Optimal Policy with Credibility Concerns.” Journal of
Economic Theory 148, no. 5 (September): 2007-2032.
Lu, Yang K., Robert G. King, and Ernesto Pasten. 2016. “Optimal Reputation Building in the New Keynesian Model.” Journal of Monetary Economics 84 (December): 233-249.
Maggi, Giovanni. 1999. “The Value of Commitment with Imperfect
Observability and Private Information.” The RAND Journal of Economics
30, no. 4 (Winter): 555-574.
Mankiw, N. Gregory, and David Romer, eds. 1991a. New Keynesian Economics, Vol. 2: Coordination Failures and Real Rigidities. Cambridge: MIT
Press.
Mankiw, N. Gregory, and David Romer, eds. 1991b. New Keynesian Economics, Vol. 1: Imperfect Competition and Sticky Prices. Cambridge: MIT
Press.
| 409

King and Lu
McCallum, Bennett T. 1981. “Price Level Determinacy with an Interest
Rate Policy Rule and Rational Expectations.” Journal of Monetary Economics 8, no. 3: 319-329.
McCallum, Bennett T. 1989. Monetary Economics: Theory and Policy.
New York: Macmillan.

Credibility and Inflation Targeting
St. Louis Review 90, no. 4 (July/August): 307-324.
Orphanides, Athanasios, and David W Wilcox. 1996. “The Opportunistic
Approach to Disinflation.” Board of Governors Finance and Economics
Discussion Series 1996-24, May.

McCallum, Bennett T. 1994. “Topics in Monetary Theory and Policy.”
Lectures at the Institute for Advanced Studies, Vienna, Austria.

Parkin, Michael. 1978. “A Comparison of Alternative Techniques of
Monetary Control Under Rational Expectations.” The Manchester School
46, no. 3 (September): 252–287.

McCallum, Bennett T., and Edward Nelson. 1999. “An Optimizing IS-LM
Specification for Monetary Policy and Business Cycle Analysis.” Journal
of Money, Credit and Banking 31, no. 3 (August): 296-316.

Poole, William. 1970. “Optimal Choice of Monetary Policy Instruments
in a Simple Stochastic Macro Model.” Quarterly Journal of Economics 84,
no. 2 (May): 197-216.

McNees, Stephen K. 1986. “Modeling the Fed: a Forward-Looking Monetary Policy Reaction Function.” New England Economic Review November: 3–8.

Poole, William. 2007. “Understanding the Fed.” Federal Reserve Bank of
St. Louis Review 89, no. 1 (January/February): 3-13.

McNees, Stephen K. 1992. “A Forward-Looking Monetary Policy Reaction Function: Continuity and Change.” New England Economic Review
November: 3–13.
Mehra, Yash P. 1997. “A Federal Funds Rate Equation.” Economic Inquiry
35, no. 3 (July): 621-630.
Nessén, Marianne, and David Vestin. 2005. “Average Inflation Targeting.”
Journal of Money, Credit and Banking 37, no. 5 (October): 837-863.
Orphanides, Athanasios. 2002. “Monetary-Policy Rules and the Great
Inflation.” American Economic Review 92, no. 2 (May): 115-120.
Orphanides, Athanasios. 2003. “The quest for prosperity without inflation.” Journal of Monetary Economics 50, no. 3 (April): 633–663.
Orphanides, Athanasios. 2004. “Monetary Policy Rules, Macroeconomic
Stability, and Inflation: A View from the Trenches.” Journal of Money,
Credit and Banking 36, no. 2 (April): 151–175.
Orphanides, Athanasios, and Volker Wieland. 2008. “Economic Projections and Rules of Thumb for Monetary Policy.” Federal Reserve Bank of
410 |

Reifschneider, David, and John C. Williams. 2000. “Three Lessons for
Monetary Policy in a Low-Inflation Era.” Journal of Money, Credit and
Banking 32, no. 4 (November): 936-966.
Rogoff, Kenneth. 1985. “The Optimal Degree of Commitment to an
Intermediate Monetary Target.” Quarterly Journal of Economics 100, no.
4 (November): 1169-1189.
Sargent, Thomas J., and Ulf Soderstrom. 2000. “The Conquest of American Inflation: A Summary.” Sveriges Riksbank Economic Review 3 (November): 12–45.
Sargent, Thomas J., and Neil Wallace. 1975. “‘Rational’ Expectations, the
Optimal Monetary Instrument, and the Optimal Money Supply Rule.”
Journal of Political Economy 83, no .2 (April): 241-254.
Schaumburg, Ernst, and Andrea Tambalotti. 2007. “An Investigation of
the Gains from Commitment in Monetary Policy.” Journal of Monetary
Economics 54, no. 2 (March): 302–324.
Shapiro, Adam Hale, and Daniel J Wilson. 2021. “Taking the Fed at its
Word: A New Approach to Estimating Central Bank Objectives using
Text Analysis.” Review of Economic Studies, forthcoming.
| 411

King and Lu
Svensson, Lars E.O. 1997. “Inflation Forecast Targeting: Implementing
and Monitoring Inflation Targets.” European Economic Review 41, no. 6
(June): 1111-1146.
Svensson, Lars E. O. 1999. “Inflation Targeting: Some Extensions.” Scandinavian Journal of Economics 101, no. 3 (September): 337-361.
Svensson, Lars E.O., and Michael Woodford. 2004. “Implementing Optimal Policy through Inflation-Forecast Targeting.” In The Inflation-Targeting Debate, edited by Ben S. Bernanke and Michael Woodford, 19-92.
Chicago: University of Chicago Press.
Taylor, John B. 1993. “Discretion versus Policy Rules in Practice.” Carnegie-Rochester Conference Series on Public Policy 39 (December): 195-214.
Williams, John C. 1999. “Simple Rules for Monetary Policy.” Board of
Governors Finance and Economics Discussion Series 1999-12 (February).
Wolman, Alexander L. 1998. “Staggered Price Setting and the Zero
Bound on Nominal Interest Rates.” Federal Reserve Bank of Richmond
Economic Quarterly 84, no. 4 (Fall): 1-24.
Woodford, Michael. 2003. Interest and Prices: Foundations of a Theory of
Monetary Policy. Princeton: Princeton University Press.
Woodford, Michael. 2004. “Inflation Targeting and Optimal Monetary
Policy.” Federal Reserve Bank of St. Louis Review 86, no .4 (July/August):
15-42.
Woodford, Michael. 2005. “Central Bank Communication and Policy Effectiveness.” Proceedings - Economic Policy Symposium - Jackson Hole,
Federal Reserve Bank of Kansas City, August: 399-474.

412 |

| 413

414 |

| 415

416 |