View original document

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

Current Challenges for
U.S. Monetary Policy
J. Alfred Broaddus, Jr.

I

t is a pleasure and an honor to be invited to participate in this conference.
I last visited Vienna in 1962, when I was a Fulbright scholar at the University of Strasbourg in France. Needless to say, Vienna has maintained its
appearance much more successfully in the intervening years than I have, but I
am very happy to have this opportunity to return nonetheless.
Let me offer a few of my views regarding the challenges facing U.S. monetary policymakers currently. Notice that I said challenges we’re confronting
“currently” rather than “in the new economy” or “in the new economic paradigm.” In this regard, some of you may have seen the comments about
paradigms by my friend and colleague Bob McTeer, president of the Dallas
Fed, in his Bank’s current Annual Report. Bob points out that if you want to
cook a frog, which I gather some people do, you don’t just throw it into a
pot of boiling water because it will jump out. Instead, you put it into a pot of
cold water and slowly increase the heat, since it won’t realize its paradigm is
shifting.
I don’t know whether Bob had me specifically in mind when he told that
story, but I suspect he had in mind people who think about this issue the way
I do. I confess to being very skeptical about the view that the macroeconomy
functions—if that’s the right word—in a systematically different way now from
the past, requiring a markedly different approach to conducting policy.
I do, however, recognize that some of the U.S. economy’s key parameters,
like the sustainable longer-term GDP growth rate, may have changed, and that
the Fed and other central banks facing similar changes need to take this into

This article is the text of an address given by J. Alfred Broaddus, Jr., president of the
Federal Reserve Bank of Richmond, before the 28th Economics Conference sponsored by
the Oesterreichische Nationalbank in Vienna, Austria, on June 15, 2000.

Federal Reserve Bank of Richmond Economic Quarterly Volume 86/1 Winter 2000

1

2

Federal Reserve Bank of Richmond Economic Quarterly

account in their efforts to optimize the contribution of policy to economic
performance. Where I might differ from some new paradigm advocates is that
I believe we can do this effectively using analytical models that have evolved
from the rational expectations revolution of the 1970s. Specifically, my own
approach to policy analysis currently draws heavily on new neoclassical synthesis models, which integrate real world phenomena like price stickiness that
many would think of as Keynesian with modern real business cycle theory. My
colleague Marvin Goodfriend and several other members of our Bank’s staff
have made important contributions to the development of these models and to
our appreciation of how they can be used to help guide monetary policymakers
in making policy decisions in a changing environment.
This is not the place for a detailed discussion of these models, and I am
certainly not the one to deliver it in any case. But let me briefly describe one of
their key features, which will be useful when I turn in a minute to the U.S. economy and the immediate monetary policy challenges we face. In these models,
the real interest rate (presented in the models as a single, representative rate)
plays a central stabilizing role. Basically, the real rate serves as an intertemporal rate of substitution. In simple language, the real rate establishes how much
households and business firms have to give up in terms of future consumption if
they choose to consume and invest today. An unsurprising corollary is that the
level of the rate directly affects the strength of the aggregate current demand for
goods and services—the lower the rate, the stronger demand, and vice versa.
In what follows I hope to show how this quite straightforward framework can
be useful in analyzing current policy options in the U.S. and elsewhere.
Before doing this, let me briefly review a few of the main features of
recent U.S. economic developments. As you may know, the U.S. economy
recently entered its tenth consecutive year of economic expansion; indeed, we
are enjoying the longest continuous expansion in our history. GDP growth during the early years of the expansion was somewhat below average compared
to the corresponding phases of earlier post-World War II expansions. Growth
equaled or exceeded 4 percent in each of the last four calendar years, however,
and was about 5.5 percent at an annual rate in the first quarter of this year.
These are exceptionally high growth rates at such an advanced stage of an
expansion. Moreover, domestic demand grew at a 5.1 percent annual rate over
this same time period. Most economists believe growth at this rate exceeds
the sustainable growth in aggregate domestic supply, a supposition supported
by the steady recent increase in the U.S. current account deficit. Beyond this,
labor markets are exceptionally tight, and the national unemployment rate—at
4.1 percent—is close to its lowest level in a generation. Despite these signs
of domestic macroeconomic imbalance, U.S. inflation has remained reasonably
well contained up to now. The core consumer price index rose 1.9 percent
in 1999, and the core personal consumption expenditures price index rose 2.1
percent. Most recently, however, core inflation has shown signs of accelerating.

J. Alfred Broaddus, Jr.: Current Challenges for U.S. Monetary Policy

3

The core CPI, for example, rose 2.2 percent in the 12 months ended in April
compared to only 1.9 percent in the 12 months ended last December.
There are some signs in the most recent monthly economic data that the
growth of demand may be moderating. These signs are hopeful but at this point
must still be considered tentative.
In this situation, as you know, the Federal Open Market Committee has
increased its federal funds rate operating instrument on six occasions recently,
from 4.75 percent last summer to 6.5 percent currently. In a world where central
bank transparency is increasingly valued, it is essential that the American public
understand clearly the rationale for Fed actions, particularly tightening actions
such as these. In this instance, while the increases have been reasonably well
received by many Americans, they have not been accepted by all, at least in
part because the increases seem counterintuitive to some in the context of the
new economy-new paradigm idea. Specifically, many “new economy” adherents
apparently believe that rising labor productivity growth has restrained increases
in labor costs and hence reduced the risk of a renewal of inflation and reduced
the need for preemptive monetary restraint by the Fed.
It is true that accelerated productivity growth temporarily limits labor cost
increases in the interval before increased demand for workers forces wages up,
and the initial increase in the output of goods and services can temporarily
restrain price increases. I don’t believe, however, that new economy advocates
have thought this matter through fully. The analytical framework I mentioned
earlier suggests exactly the opposite policy conclusion. It indicates that higher
interest rates are required to restore macroeconomic balance and ensure sustained higher growth over the longer term.
Some background information on recent U.S. productivity growth trends is
required to appreciate this result. U.S. hourly labor productivity grew at about
a 2.25 percent average annual rate over the 80-year period between 1890 and
1970. This persistent and healthy growth had an enormously positive impact
on income and living standards. At this rate, output per worker doubled approximately every 30 years and increased nearly eight-fold over the period as
a whole.
Around the mid-’70s, however, trend productivity growth decelerated noticeably to about a 1.5 percent annual rate, at which rate per worker output
doubled only about every 45 years, and the reduced growth persisted until the
mid-’90s. We still don’t fully understand the cause of the slowdown, although
it is reasonable to suspect that it was related in part to the oil shocks of the midand late ’70s and the high inflation of that period. It may also have reflected
changes in the composition of the workforce, particularly the entry of a large
number of young workers with less than average work experience and therefore
lower productivity.
Whatever its causes, the key point is that most Americans perceived the
slowdown, although they did not think of it analytically in terms of a reduced

4

Federal Reserve Bank of Richmond Economic Quarterly

trend productivity growth rate. Rather, they thought of it in personal terms as
reduced economic opportunities both currently and prospectively. It was during
this period that, for the first time in recent U.S. history, many workers concluded
that their living standards would be no higher than those of their parents.
As you undoubtedly know, there is now considerable evidence that trend
productivity growth in the U.S. has revived since the mid-’90s. It is of course
much too early to verify this statistically, but the persistently higher-thanexpected real growth in the U.S. economy over the last four years or so without
a reacceleration of inflation would be consistent with higher trend productivity
growth. Many U.S. economists now estimate that this trend growth has increased 1 to 1.5 percentage points from the reduced mid-’70s-to-mid-’90s rate
to the vicinity of 2.5 to 3 percent currently. With trend labor force growth at
approximately 1 percent, trend productivity growth at this higher rate would imply that the economy’s “speed limit”—its maximum sustainable, noninflationary
growth rate—is now in the neighborhood of 3.5 to 4 percent, an appreciable
increase from the commonly perceived 2 to 2.5 percent limit in the early ’90s.
Just as the earlier slowdown in trend productivity growth was perceived, at
least intuitively, by the public, so, too, the apparent recent acceleration in trend
growth is perceived. Evidence of this perception is widespread. The long bull
market in U.S. stocks reflects higher expected future business earnings growth.
And I can assure you that my two grown sons and their friends and associates
expect lifetime incomes and living standards well above those of their parents.
Again, neither my sons, other households, nor business firms typically think
explicitly of their expected higher future income as the result of an increase
in trend productivity growth. But their expectations and—as I will indicate
momentarily—the actions they take based on these expectations make it clear
that they perceive the increase implicitly.
What do all these developments in the “real” economy have to do with
monetary policy? The answer is that U.S. households are now borrowing quite
liberally against their higher expected future incomes to consume today. They
are buying new homes, adding on to existing homes, and buying consumer
durables such as new cars, furniture, and electronic equipment. Similarly, firms
are borrowing against their higher expected future earnings to invest in new
plant and equipment.
The problem posed for monetary policy by all this is that the higher
expected future income driving the increased current demand for goods and
services is not yet available in the form of increased current output of goods
and services. This mismatch between expected future resources and currently
available resources, in my view, is the principal factor creating the present
aggregate demand-supply imbalance in the U.S. economy I discussed earlier.
The excess demand has been satisfied to date by imports and progressively
tighter labor markets. But demand is now rising more rapidly here in Europe
and elsewhere around the world, which may soon put upward pressure on the

J. Alfred Broaddus, Jr.: Current Challenges for U.S. Monetary Policy

5

dollar prices of imports. And labor shortages are now widely reported in a
number of sectors and industries. On their present course, U.S. labor markets
will eventually tighten to the point where competition for workers will cause
wages to rise more rapidly than productivity, which sooner or later would
induce businesses to pass the higher costs on in higher prices. As I suggested
earlier, there is evidence in some of the latest U.S. price and labor cost data
that an inflationary process of this sort may now be beginning.
The implication of this analysis, as I indicated at the outset, is that the
apparently higher trend productivity growth in the U.S. economy—whether
one labels it a “new paradigm” or not—requires higher real interest rates to
maintain macroeconomic balance. In order to prevent a reemergence of inflationary pressures and, in doing so, to sustain the expansion, U.S. monetary
policy must allow short-term real interest rates to rise to induce households and
business firms to be patient and defer spending until the higher expected future
income is actually available, in the aggregate, in the form of higher domestic
output.
This necessity presents the Fed with several challenges. First, while the
need for rate increases seems clear, how do we decide on the magnitude and
timing of the increases? In principle, of course, we want to allow rates to
rise to the level where the growth in aggregate current demand equals the
sustainable growth in productive capacity. In the technical language I noted
earlier, ideally we would like to establish an equilibrium intertemporal rate of
substitution consistent with aggregate demand-supply balance. Identifying this
equilibrium level is difficult, because it is continuously responding not only
to the apparent trend productivity growth increase but also to any number of
other shocks hitting the economy. Taylor-type rules may offer some operational
help in setting the appropriate federal funds rate level, but in the absence of a
stronger professional consensus regarding how to use these rules, policymakers
in practice will have to apply judgment based on their interpretation of current
economic data and forecasts.
As you know, we have in fact been allowing real rates to rise. (I am deliberately avoiding the misleading terminology that the Fed is “raising rates.”)
In the spirit of the increased emphasis on transparency in monetary policy,
perhaps the principal challenge for the Fed currently is making it clear to the
public that these actions have not been the misguided result of “old economy”
thinking, but steps that are essential for maintaining balance and maximizing
long-term growth in the economy, whether one regards it as new, old, or simply
evolving.

6

Federal Reserve Bank of Richmond Economic Quarterly

REFERENCE
Goodfriend, Marvin, and Robert King. “The New Neoclassical Synthesis and
the Role of Monetary Policy” in Ben S. Bernanke and Julio J. Rotenberg,
eds., NBER Macroeconomic Annual 1997. Cambridge, Mass.: MIT Press,
pp. 231-82.

The Role of a Regional Bank
in a System of Central Banks
Marvin Goodfriend

A

modern central bank seeks to maintain a financial environment within
which competitive markets support the efficient use of productive
resources. 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 indirect 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.
The principle that markets should be given free reign wherever possible
creates three difficulties of understanding that a central bank must overcome
in order to carry out its policies effectively. The presumption that monetary
and banking policies are best when they are as unobtrusive as possible creates
the first difficulty. Inevitably, central banks seem shadowy and distant from
the public’s point of view. Yet, to work well, central bank policies need to
shape the expectations of households and businesses. Monetary policy encourages economic growth and stabilizes employment over the business cycle by
anchoring inflation and inflation expectations. Bank supervision and regulation
aims to promote confidence in the banking system.

This article is reprinted as it appeared in the Carnegie-Rochester Conference Series on Public Policy, volume 51, number 1, with permission from Elsevier Science. It also appeared
in the Federal Reserve Bank of Richmond 1999 Annual Report. The article benefited from
presentations at the European Central Bank, the Federal Reserve Board, the Bank of Finland,
and at the Bank of Italy, IGIER, Paolo Baffi conference on “Monetary Policy of the ESCB:
Strategic and Implementation Issues.” Ignazio Angeloni, Al Broaddus, Bennett McCallum,
and Mark Wynne provided helpful comments. The views are the author’s and not necessarily
those of the Federal Reserve Bank of Richmond or the Federal Reserve System.

Federal Reserve Bank of Richmond Economic Quarterly Volume 86/1 Winter 2000

7

8

Federal Reserve Bank of Richmond Economic Quarterly

The need to influence expectations and promote confidence puts a premium on credibility, a commitment to goals, and a central bank’s perceived
independence and competence to achieve its objectives. Thus, a central bank
must create in the public’s mind an understanding of the methods by which its
objectives can be sustained. This formidable problem has to be overcome in
spite of the fact that a central bank operates in the background, with obscure
methods and procedures.
The second and third difficulties arise because central bankers must understand markets. Dynamic markets introduce evermore efficient productive
technologies and create new goods and services to better satisfy consumer
wants. Economic dynamism complicates the measurement of macroeconomic
conditions. A central bank seeks to understand the latest market developments
in order to implement monetary and banking policies appropriately. Policy
actions are inevitably benchmarked against historical correlations in data. Yet
a central bank must be prepared to question its interpretation of data in light of
anecdotal and other information that suggests behavior different from historical
averages.
The third difficulty of understanding is in the area of economic analysis.
Because policies influence economic activity indirectly, central bankers must
use economic analysis to think about how their policies are transmitted to
the economy. Some sort of quantitative theoretical model must be used to
think about how markets respond to monetary and banking policies, and how
monetary and banking policies ought to react to the economy.
The role of regional banks in a system of central banks is about creating understanding in the three senses described above. For example, decentralization
enhances credibility because the diffusion of power makes it more difficult for
outside pressures to be brought to bear on a central bank. The regional presence
helps a central bank to get its policy message out and to gather anecdotal and
specialized information on regional economies. Information gathering and dissemination are particularly important for central banks such as the Eurosystem
and the Federal Reserve System, whose currency areas span large and populous
regions. For this reason, the Central Bank of the Russian Federation and the
Peoples Bank of China might profitably restructure themselves as a system of
regional central banks.1
A regional presence also benefits a central bank with responsibilities for
bank supervision and regulation, and the power to extend emergency credit
assistance to troubled financial institutions. Specialized knowledge of local
economies, industries, and businesses is of use to bank examiners and helpful
in determining whether a troubled bank deserves emergency credit assistance.

1 In

branches.

late 1998, the Peoples Bank of China announced its intention to establish nine provincial

M. Goodfriend: The Role of a Regional Bank

9

Likewise, central banks that play a role in the provision of payments services
run far-flung operations through their regional offices.
Last but not least, the diversification of research within a system of central
banks brings a variety of analytical perspectives to policy deliberations that
is invaluable in our increasingly complex economy. Moreover, a system of
regional banks led by the center institution harnesses competitive forces to
encourage innovative thinking within the central bank.
The first half of this article, which includes Sections 1 through 4, highlights the role played by the Reserve Banks in the Federal Reserve System.
The remainder of the article, Sections 5 through 8, offers some observations on
the new Eurosystem based on the experience of the Federal Reserve System.
There is a short concluding section.
Having spent 20 years as an economist at the Federal Reserve Bank of
Richmond, I welcome the opportunity to clarify my thinking on these matters.
I hope that my discussion of the Federal Reserve System helps the European
national banks and the European Central Bank to think about their respective
roles in the Eurosystem. Early in the century the Federal Reserve System looked
to European central banks for guidance in designing its institutional structure
and operating procedures. The Federal Reserve will be pleased if it can now
return the favor.

1.

THE FEDERAL RESERVE BANK PERSPECTIVE

The improvement over time in communication, information, and transportation
technologies has enhanced the role of Reserve Banks in the Federal Reserve
System. The United States has seen a deconcentration of metropolitan employment that appears to be the result of urban congestion and technologies
that make it increasingly possible to locate businesses away from traditional
urban centers.2 The tendency is toward an equalization of regional economic
activity.3 Think of the growth of California, Florida, and Texas, and the tremendous growth in the South and Southwest. Atlanta, Georgia, has become a major
commercial center; Charlotte, North Carolina, is a major banking center; Seattle
is the home of aircraft and software production.
The growing dispersion of economic activity increases the value of local
information that Reserve Bank presidents bring to the Federal Open Market
Committee. The presence of Reserve Banks in the midst of the various regional economies makes possible a deeper understanding of these than can be
acquired from Washington. Personal contacts built up over time create trusting
relationships that facilitate the timely acquisition of information about local
2 See,

for example, Chatterjee and Carlino (1998).
and Sala-´-Martin (1992) present evidence of convergence within the United States.
ı

3 Barro

10

Federal Reserve Bank of Richmond Economic Quarterly

businesses and markets. Personal contacts are particularly valuable in periods
of financial stress when it is especially difficult to know what is happening
in certain sectors. Reserve Banks tend to specialize in knowledge concerning
industries concentrated in their respective districts. For instance, the New York
Fed follows financial markets generally, the Chicago Fed follows commodity
markets and heavy manufacturing, the Dallas Fed follows oil production and
developments in Mexico, etc.
Thanks to the progress in information and communication technology, Reserve Banks are no longer at an information disadvantage relative to the Federal
Reserve Board or the New York Fed with respect to general market information. All receive news and data instantaneously from everywhere. Reserve
Bank presidents, in turn, contribute to policy discussions with speeches and
articles transmitted instantaneously around the world by wire services and by
the Internet.
Reserve Bank officials are familiar with both their regional private sector
world and the world of the Federal Reserve Board. Reserve Banks help bridge
the two worlds. Responsibilities and pressures at the Board create a culture
very different from the private sector. The Board staff relies on aggregate
data and abstract concepts to think about the whole economy. Thinking at the
Board reflects consensus beliefs and attitudes, and is cautious in adopting and
even considering new ideas. Because the Board has ultimate responsibility for
much that is done in the System, it has little trouble attracting hard-working,
dedicated, and highly skilled employees. Yet because of the responsibility, the
pressure, the need for consensus, and the need to focus on abstractions and
aggregates, the Board staff can be distant from the private sector. This is
a manifestation of the remoteness described in the introduction that plagues
central bankers.
With important exceptions there is less ultimate responsibility for System
matters at Reserve Banks. On the other hand, there is opportunity for distinguishing one’s Reserve Bank from the others. This is a manifestation of the
competitive innovation, described in the introduction, that a system of central
banks promotes.
One of the Federal Reserve Board’s most important duties is to manage
relations with Congress. The Board also handles international relationships
and deals directly with large financial institutions and national interest groups.
Board members testify and give speeches frequently. While these are critically
important responsibilities, such communications are nevertheless rather abstract
and remote.
Because of its regional presence and focus, the staff at Reserve Banks is
more engaged with the rank and file public. Much of what Reserve Banks
do involves direct relations with people in the private sector. For instance,
Reserve Bank officials manage relations with their Boards of Directors made
up of private citizens. Officials speak to local groups about Federal Reserve

M. Goodfriend: The Role of a Regional Bank

11

policies and current economic conditions. Staff members supervise and examine banks, collect data on banking and regional business conditions, provide
financial services, promote economic education, and help facilitate community
development. The staff at Reserve Banks understands core policy, regulatory,
and operational issues and knows how to explain these to its constituencies.
In short, Reserve Banks keep the central bank from becoming disembodied,
isolated, and out of touch.

2.

FEDERAL OPEN MARKET COMMITTEE MEETINGS4

The Federal Open Market Committee (FOMC) meets every six weeks on average at the Federal Reserve Board in Washington. The meetings are attended by
the seven governors of the Federal Reserve System, the twelve Reserve Bank
presidents, and research directors and other staff members from the Reserve
Banks and the Board. The Chairman of the Board of Governors sets the agenda,
leads the discussions, shapes the policy decisions, and develops the consensus
to support the Committee’s policy actions.
The meetings routinely include a report from the open market desk at the
Federal Reserve Bank of New York, a briefing by the Board staff on current
economic and financial conditions in the United States and abroad, a couple of
“go arounds” in which the governors and presidents present their views on the
economy and policy, and a discussion and vote on the intended federal funds
rate. Normally, an FOMC meeting lasts four to five hours, but twice a year
the Committee meets for two days to set annual target ranges for the monetary
aggregates and to consider longer-run procedural and strategic issues.
Even though all Reserve Bank presidents but the New York Fed president
vote on a rotating basis, all 19 members of the Committee participate on equal
terms at every meeting. The time for discussion among the members is, accordingly, limited. More often than not, Committee members influence each other
incrementally by revisiting issues as time passes, rather than by exchanging
views at any particular meeting. Economic conditions usually do not call for a
change in the intended federal funds rate. The Committee uses such occasions
to prepare itself for possible future policy actions. Such “down time” affords
ample opportunity to consider strategic and procedural questions. All in all,
there is time for Committee members to educate and influence each other, and
to reach consensus. But, again, much of the back and forth among Committee
members takes place over time. In this regard, the verbatim written transcript
that is prepared and circulated after each FOMC meeting (but released with
a five-year lag) is of great help in enabling members to review each other’s
statements in detail.
4 See

Meyer (1998).

12

Federal Reserve Bank of Richmond Economic Quarterly

The deliberative process works reasonably well in practice. The repeated
interaction creates a mutual understanding that enables a variety of geographical
and professional perspectives (academic economist, banker, business economist, businessperson, financial market professional, government administrator,
lawyer, and regulator) to be brought to bear in making policy decisions.
Two related pitfalls have the potential to weaken the FOMC. First, the
bonding that takes place as a consequence of repeated meetings can cause
Committee members to begin to think alike. As a result, the FOMC could be
blindsided by a risk or side effect of a policy stance that it had not taken into
account. To some extent, that risk is diminished by the external community of
“Fed watchers” offering professional advice on monetary policy.
The sheer size of the FOMC reduces the likelihood that Committee members will think alike. One of the great strengths of policy made by representatives from a system of regional central banks is the diversity and number
of points of view brought to the table. But the size of the FOMC actually
creates the second potential pitfall: a free rider problem. Recognizing that their
influence in the Committee may be small, members may be inclined to freeride on the preparations of others more interested, expert, or responsible for
monetary policy, such as the Chairman and the Board staff.
The free rider problem is dangerous because it has the potential to make
the effective size of the FOMC much smaller than the full Committee. Even
worse, free riding is hard to detect because free riders can continue to participate
with thoughtful-sounding statements. Widespread free riding would weaken the
Committee in much the same way as the tendency to think alike.
The Chairman of the Federal Reserve Board
Even though the Chairman has only one vote in the FOMC, he is preeminent
for a number of reasons. The Chairman and the other Board members are
appointed by the President of the United States, and the Chairman is named by
the President to lead the Federal Reserve System. The Chairman has command
of the large staff at the Federal Reserve Board. Most importantly, only he is
involved in every key central bank operation (monetary policy, bank supervision and regulation, financial services, foreign exchange operations, relations
with Congress and the Treasury, and public relations). The Chairman is the
only member of the FOMC fully aware of all the potential interconnections in
what the Federal Reserve does. Consequently, no major decision can be taken
without the Chairman’s assent for fear of not having all the facts. For all these
reasons it is difficult to challenge the Chairman’s leadership.
By the same token, a good Chairman is aware of the risks of excessively
centralizing power in his hands. For the reasons discussed above he must encourage diverse points of view in the FOMC. Central bankers worry about a
variety of risks to the economy and the Chairman must encourage Committee

M. Goodfriend: The Role of a Regional Bank

13

members to bring their concerns to the table. The Chairman must help prioritize
the concerns and suggest a course of action to achieve the central bank’s goals.
Finally, the Chairman must mobilize the Committee to action. All in all, the
Chairman must use his preeminence to make the most of the diversity in the
FOMC while preserving the decisiveness needed to make monetary policy.
Reserve Bank Presidents at the FOMC
Broadly speaking, Reserve Bank presidents contribute to FOMC meetings in
two important ways. They make regular reports on their respective regional
economies, and they provide their own analysis of the national economy and
the policy options.
Regional information compiled by Reserve Banks for the FOMC in the
Beige Book is of great importance.5 But information in the Beige Book can be
stale by the time of an FOMC meeting. Presidents bring more timely information to the meeting, including confidential information from personal or other
sources not included in the Beige Book. Anecdotal information brought to the
FOMC can signal changing sentiment before it becomes evident in aggregate
data. Mutually supportive signals from various regions may help to identify or
confirm a change in trend or a turning point in the aggregate data. It is particularly important that a central bank recognize and react promptly to turning
points in inflation and employment trends.
Besides the Chairman, the Board staff presents the most influential economic analysis at FOMC meetings. The staff’s analysis is primarily presented in
two briefing documents with which Committee members’ views are invariably
compared. The Greenbook summarizes national and international economic
conditions and presents a forecast; the Bluebook lays out the policy alternatives.
Although the briefing books are comprehensive, the analysis of individual
members provides essential perspective. Governors and presidents alike contribute substantively to the interpretation of current economic conditions and
the analysis of alternative policy options. Many important possibilities such as
the risk of an inflation or deflation scare or the chance of a crisis of confidence
in financial markets are particularly difficult to assess and take account of in
econometric models. The state of consumer and business confidence is also
difficult to assess formally. Such issues are addressed in the statements of
Committee members themselves.
Economic analysis is a great equalizer among members of the FOMC. An
argument based on economic reasoning that can be challenged and debated
in the language of economics is ultimately more influential than an intuitive
assertion about the economy or policy, no matter who expresses it and how
strongly it is held.
5 See

Balke and Petersen (1998).

14

3.

Federal Reserve Bank of Richmond Economic Quarterly

ECONOMIC RESEARCH AT
FEDERAL RESERVE BANKS

Reserve Bank research departments are staffed with an average of 15 or so
research economists (except for the New York Fed, which has more than twice
as many). Economists graduate from top schools where they acquire the latest
analytical skills and an appreciation of how to think about macroeconomics,
monetary policy, and banking policy. For the most part, there is a belief in the
power and practical value of economic theory and empirical work, and a drive
to use economics to make good policy.
Reserve Banks are able to attract and retain good economists because they
offer a unique combination of opportunities. Above all, there is the opportunity
to prepare the bank president for FOMC meetings. In their role as policy advisors, Reserve Bank economists acquire an intimate empirical understanding
of the macroeconomy and a broad understanding of policy issues. Economists
produce policy essays for the Bank’s Economic Review and may be encouraged to publish articles in professional economics journals. The best of these
essays may influence the way that the Federal Reserve, other central banks,
and academic economists think about policy. It is possible for a Reserve Bank
economist to become increasingly effective as a policy advisor while acquiring
a research reputation in the economics profession at large.
Reserve Bank research departments need not specialize. The expression of
alternative points of view is an important strength of a system of central banks.
Nevertheless, Reserve Bank research departments often develop a specialization. A Reserve Bank president may encourage research of one type or another;
or a particularly skillful economist may happen to make a department strong in
a particular sort of research. A Bank may also exploit a feature of its regional
economy or its operational responsibilities to develop a research advantage.
Differences of opinion among Federal Reserve economists are discussed at
regular System research meetings. From time to time, there are differences of
opinion involving essays in a Reserve Bank Economic Review. Reserve Banks
send review articles to the Board for a prepublication review. Ordinarily essays
benefit from comments by the Board staff. On occasion, the Board staff may
recommend against publication because an article is thought to be technically
flawed or because the article takes a position regarded as inconsistent with
System policy. Conflicts arise because the Board staff prepares speeches and
testimony for the Chairman and other Board members in which the Federal
Reserve explains current policies to Congress and others. Policy essays published by a Reserve Bank that implicitly or explicitly question current policies
may be a nuisance or worse from the perspective of the Board.
Obviously, Reserve Bank economists could be prevented from publishing
essays critical of current policy. But that would deny the public the work
of economists most knowledgeable about central banking. It would leave the

M. Goodfriend: The Role of a Regional Bank

15

field wide open to others less familiar with the subject. Besides, policy essays
reveal a healthy open debate within the Federal Reserve System. In keeping
with the mission of a central bank to worry about the economy and policy, it
is helpful to have policy questioned by enterprising economists at the Reserve
Banks. Furthermore, the best essays facilitate policy advances by suggesting
alternatives.
Ultimately, a Reserve Bank has both the incentive and the ability to discipline the output of its economists. The Reserve Bank itself has the most
to lose by publishing a poor essay in its Review. Reserve Bank research is
regularly presented at Federal Reserve System committees and at academic
conferences and seminars. Research directors have ample opportunity to judge
the professional reception of a particular piece of research prior to publishing
it in the Bank’s Review.

4.

PUBLIC INFORMATION

The modern era of monetary policy at the Federal Reserve began when Chairman Paul Volcker took responsibility publicly for inflation in the early 1980s,
and subsequently brought it down. This was a watershed event because before
that Federal Reserve officials and much of the public, too, generally blamed
inflation on a variety of causes beyond the central bank’s control. Since then, the
public has come to understand that Federal Reserve monetary policy determines
the trend rate of inflation over any substantial span of time.
The acceptance of the responsibility for low inflation by the Federal Reserve greatly elevated the importance of public information and communication
in the policy process. Previously, the Federal Reserve preferred to operate in
the background and out of the limelight. The public thought that important
economic policy decisions were made elsewhere, and the Fed felt relatively
little need to communicate with the public about its policy intentions. All that
changed after the disinflation initiated by Chairman Volcker, for two reasons.
First, the Fed thrust itself into the limelight with inflation-fighting policy actions that raised interest rates and weakened economic activity in order to
bring down inflation. Second, the Fed realized that bringing down inflation and
maintaining price stability would be easier if the Fed had credibility for low
inflation. Thus, the public became more interested in what the Fed was doing,
and Fed officials came to see communication with the public as a tool useful
for building credibility.
The Fed has two primary public information objectives with respect to
monetary policy.6 A consensus has emerged among monetary economists and
central bankers that some sort of explicit mandate for low inflation is beneficial.
6 See

Goodfriend (1997).

16

Federal Reserve Bank of Richmond Economic Quarterly

Yet, Congress has not mandated in a clear way that the Fed place a priority on
low inflation. Consequently, Fed officials bear the burden of responsibility for
educating the public about the benefits of low inflation. Second, the guiding
tactical principle of monetary policy is to preempt inflation, or deflation, for
that matter. A well-timed preemptive increase in the intended federal funds rate
is nothing to be feared. For instance, the 1994 monetary tightening was almost
certainly necessary to keep inflation from ending the business expansion. If the
Fed is to successfully maintain price stability, it must create an understanding
of the need for policy to be preemptive; and the Fed must build a consensus
for specific preemptive policy actions when they are needed.
The regional presence of the Reserve Banks is a great advantage in getting
the Fed’s message out to the public. The participation of Reserve Bank presidents in the FOMC puts them in great demand as speakers in their districts.
Economists and other staff members at the Reserve Banks also carry the Fed’s
message to the public. Reserve Banks produce a variety of literature aimed at
educating the public about the Federal Reserve. There are extensive economic
education programs through which the staff at Reserve Banks explains monetary
policy to schoolteachers and college professors.
Sometimes market participants complain that speeches by members of the
FOMC complicate the business of understanding the Fed’s current thinking. As
mentioned above, the great strength of the Federal Reserve System is that it
brings a number of different points of view to the FOMC. There is no reason
why the public should not hear these diverse views.
Markets know that the Chairman, and only the Chairman, speaks for the
whole FOMC, and the Chairman’s rhetoric is understood to represent the current
consensus thinking of the FOMC on policy. The Chairman makes use of his
numerous appearances before Congress and elsewhere to update or elaborate
upon the current thinking of the FOMC. Moreover, the FOMC announces any
change in its intended federal funds rate immediately after any meeting in
which the rate is changed. Minutes of each FOMC meeting, released shortly
after the following meeting, give a fairly comprehensive idea of the concerns
and inclinations of Committee members, though without individual attribution.
Included with the minutes is the policy directive from the FOMC to the open
market desk. The directive contains “symmetry language” that indicates any
inclination on the part of the Committee as a whole to be more concerned with
the risk of inflation or recession over the next few weeks. The minutes also
contain the voting record and any statements of dissent expressed by members
of the FOMC.
The public does not seem to mistake the personal views of individual members for information about the FOMC as a whole. Transparency of a Committee
member’s views, rather than secrecy, seems more likely to build understanding
and credibility for the Federal Reserve over time. Not to air differences among
Committee members would deprive markets of useful information, and it would

M. Goodfriend: The Role of a Regional Bank

17

put the public at a permanent disadvantage in understanding monetary policy.
It is worth emphasizing that the Federal Reserve’s most effective voice is
that of its Chairman. The great respect accorded the Fed Chairman is largely
due to his own analytical ability and experience, and the informational and
analytical support of the capable Board staff. A good measure of credit is no
doubt due to recent monetary policy successes. But an important source of the
Chairman’s personal credibility probably comes from the fact that he represents
the views of the diverse members of the FOMC. If the public were to believe
that the Chairman was acting alone, the public would be more inclined to worry
that the Chairman could be co-opted, i.e., that he might take policy actions for
political rather than economic reasons. The Chairman’s credibility and influence would suffer accordingly. Even here, the regional nature of the Federal
Reserve System plays an important role. The Federal Reserve Chairman needs
the FOMC as much as the Committee needs its Chairman.

5.

THE EUROSYSTEM7

The Eurosystem shares the basic structure of the Federal Reserve System. The
Eurosystem consists of the European Central Bank (ECB) headquartered in
Frankfurt am Main, more or less the equivalent of the Federal Reserve Board,
and 11 national central banks (NCBs), which are like the 12 Federal Reserve
Banks. Monetary policy in the Eurosystem is made by the Governing Council
(the equivalent of the FOMC). The Governing Council includes six members
of an Executive Board housed at the ECB (the rough equivalent of the sevenmember Board of Governors of the Federal Reserve System) and the governors
of the 11 national central banks. The President of the ECB chairs the Governing
Council, playing a role similar to the Chairman of the Board of Governors.
Power in the Eurosystem is more decentralized than in the Federal Reserve
System. First of all, the governors of the NCBs all vote on policy matters in
the Governing Council on each occasion. The seven members of the Board of
Governors and the New York Fed president vote all the time in the FOMC,
but the other 11 Reserve Bank presidents have only four votes on a rotating
basis. As is the case in the FOMC, policy decisions in the Governing Council
require a simple majority vote.
Secondly, the Board of Governors exercises more power in the Federal
Reserve System than the ECB does in the Eurosystem. For instance, the Board
of Governors exercises general supervision over the Reserve Banks: the Board
approves Reserve Bank budgets, approves the appointment of Reserve Bank
7 European Union (1995) contains the Maastricht Treaty, which, in turn, contains the language governing the structure, administration, and objectives of the Eurosystem. Wynne (1999)
summarizes the documentation authorizing the establishment of the Eurosystem.

18

Federal Reserve Bank of Richmond Economic Quarterly

presidents, and appoints three of nine directors at each Reserve Bank, including
the chairman. In contrast, the Maastricht Treaty gives the NCB governors control over the terms and conditions of employment of the staff at the ECB. The
NCBs are financially independent of both the ECB and their respective national
governments. Decentralized control, the so-called principle of subsidiarity, is
enshrined in the preamble of the Maastricht Treaty.
Even the ECB itself is more decentralized than the Board of Governors.
For instance, the Economic and Research Directorates, which employ the bulk
of the ECB’s professional economists, do not report to the President of the
ECB but to another member of the Executive Board. The fact that there is
no Chief Executive of Europe to give his assent to the President of the ECB
and other Executive Board members, as in the United States, probably makes
for a weaker ECB within the Eurosystem. The NCB governors are appointed
by their respective national governments, without approval of the Executive
Board.
On the objectives for monetary policy, the Maastricht Treaty states unambiguously that the primary objective of the Eurosystem shall be to maintain
price stability. Although the treaty obliges the Eurosystem to support the general
economic policies of the European Union, that support is to be without prejudice to the objective of price stability. Accordingly, the Eurosystem mandate
is considerably more definite than the objectives given in the Federal Reserve
Act.
The Maastricht Treaty safeguards the independence of the Eurosystem. The
Eurosystem charter is an international treaty that cannot be revoked without
unanimous consent of the signatories. Moreover, the treaty itself actually tells
the Eurosystem not to take instructions from other institutions in the European
Union. The greatest threat to the Eurosystem’s independence and the pursuit of
price stability could come from the ambiguity in the treaty on exchange rate
policy, which is to be established by the European Council. It is not completely
clear how a conflict between exchange rate and price stability objectives would
be settled.
On transparency, the Maastricht Treaty mandates that the ECB publish
quarterly and annual reports. Executive Board members have signaled their
willingness to testify regularly before the European Parliament. The ECB intends to keep the public informed of its policy actions and thinking through
press conferences, speeches, and other regular publications. The President of the
ECB holds a press conference to discuss monetary policy immediately after one
of the two Governing Council meetings held each month. Notably, the treaty
specifies that the proceedings of the meetings shall be confidential, but that the
Governing Council may decide to make the outcome of its deliberations public.
For now, the Eurosystem does not coordinate and centralize bank supervision and regulation, or emergency credit provision. NCBs carry on in these
areas according to their respective national policies. This, of course, differs from

M. Goodfriend: The Role of a Regional Bank

19

Federal Reserve practice, where the Board exercises control over emergency
credit assistance and over the supervision and regulation of banks.

6.

DECENTRALIZATION IN THE EARLY FEDERAL
RESERVE: IMPLICATIONS FOR THE EUROSYSTEM

The decentralized Governing Council described above is reminiscent of the
early Open Market Committee of the Federal Reserve System. Established informally in 1922 with 5 of the 12 Reserve Banks represented, the Committee’s
membership was broadened to include all 12 banks in 1930. The FOMC took
its modern form with the Banking Act of 1935, which gave the seven members
of the Federal Reserve Board a vote in open market policy for the first time,
and reduced the Reserve Bank votes to five.
As is well known from the account by Milton Friedman and Anna
Schwartz, the decentralized structure of the Open Market Committee in the
1920s depended for its decisiveness on the leadership of Benjamin Strong,
Governor of the Federal Reserve Bank of New York.8 Governor Strong’s powers of persuasion, personal courage, and good judgment gave coherence and
purpose to Federal Reserve policy. After Governor Strong died in October 1928,
the Open Market Committee became unworkable. Without Strong’s leadership
the decentralized Open Market Committee made for drift and indecisiveness in
Federal Reserve policy.
The Governing Council of the Eurosystem appears to be susceptible to the
same indecisiveness as was the early Open Market Committee. A closer look,
however, shows why this is not likely to be the case.
First, the objectives of Federal Reserve monetary policy in the early years
were ambiguous. The United States was on a gold standard, and the Fed was
committed to defend the dollar price of gold. Yet for much of the 1920s Governor Strong sterilized gold flows and instead tried to stabilize the price level.9
In large part, Strong’s personal discretion substituted for the lack of an agreed
objective. The Eurosystem’s price stability mandate should go a long way
toward preserving the decisiveness of the Governing Council.
Second, it will take some time for the Eurosystem to develop and become
familiar with euro-area data. But on the whole, much better macroeconomic
data exist today than were available to the early Fed. This, too, should make
the Governing Council more decisive than the early Open Market Committee.
Third, today’s central banks can draw on the considerable theoretical and
practical knowledge that economists have accumulated since the early years of
the Fed. Central bankers have accumulated a good deal of practical knowledge
8 See
9 See

Friedman and Schwartz (1963).
Hetzel (1985).

20

Federal Reserve Bank of Richmond Economic Quarterly

themselves. The early Fed had little experience in managing monetary policy
and very little in the way of analytical skills at its disposal to help guide policy.
Fourth, professional central bank watchers today provide external advice
and discipline.10 This, too, should act against policy indecision. Fifth, the Fed
did not yet have the tradition of making the Chairman of the Board of Governors
the Chairman of the FOMC. In effect, the Fed then lacked an institutional leader
designated by the President of the United States. This was a great weakness in
a decentralized structure such as the Open Market Committee. The President of
the ECB is the designated leader. He is appointed by the European Council and
confirmed by the European Parliament. In any case, it should be pointed out
that centralization of power in the FOMC such as occurred with the Banking
Act of 1935 did not guarantee good monetary policy, as the Great Inflation
from the late 1960s to the early 1980s showed.
To sum up, the analogy with the early Fed is far from conclusive. With the
help of the support systems described above, the Governing Council should be
able to strike a reasonable balance between decentralization and decisiveness.

7.

SUBSIDIARITY AND ECB STAFFING

One problematic issue facing the Eurosystem is the nature of the control that
the NCBs will exercise over the staffing budget of the ECB according to the
principle of subsidiarity. This is critical because, as the discussion of the Federal Reserve System makes clear, the Eurosystem cannot function effectively
without a sufficiently strong ECB. The ECB must perform certain tasks. For
instance, the ECB must represent the Eurosystem in its external relationships.
Presumably, only the President of the ECB can speak for the Governing Council. Also, the ECB is the natural home for economists following the euro-area
economy as a whole. The ECB is a natural repository for euro-area data, and
its economists will assume primary responsibility (though by no means an
exclusive one) for following and interpreting these data for the Eurosystem.
In addition, the ECB needs a staff with analytical capabilities sufficient to
support the President in his role as leader of the Eurosystem. Among other
things, the ECB’s staff, working with the staff at the NCBs, must devise an
analytical framework that can help the President of the ECB guide the members
of the Governing Council in their monetary policy deliberations.
The funding of the ECB staff must be authorized by the NCB governors.
Yet the NCBs lack the experience to judge the ECB’s priorities and needs.
The problem is twofold. First, NCBs know relatively little about managing
independent monetary policy. Second, NCBs have little experience as regional

10 See,

for example, Begg et al. (1998).

M. Goodfriend: The Role of a Regional Bank

21

banks in a system of central banks. The division of labor between the NCBs
and the ECB will have to be worked out gradually over time.
One hopes that the NCBs will agree to build up staff at the ECB fast
enough to provide the leadership that the Eurosystem needs. The analogy with
the Fed system makes clear that critical responsibilities should be borne by the
ECB. NCBs have responsibilities and comparative advantages of their own that
they should exploit for the benefit of the Eurosystem.11

8.

NATIONAL CENTRAL BANKS AND THE
CREDIBILITY OF THE EUROSYSTEM

The Eurosystem will establish full credibility for low inflation over time by
satisfying three conditions. First, the Eurosystem must manage monetary policy
competently. Second, the NCB governors and Executive Board members on the
Governing Council must learn to work together. Third, the Eurosystem must
build on its price stability mandate to broaden the public’s support for price
stability and the preemptive policy actions necessary to sustain it. The NCBs
play a central role in seeing that these three conditions are satisfied.
Competence
It seems fair to say that the Eurosystem’s expertise in maintaining price stability
derives in large part from the Bundesbank, which has had a long and successful
track record in managing independent monetary policy.12 Other NCBs have less
experience because for the most part they have chosen to fix their exchange
rates to the Deutsche Mark. The Eurosystem adopted many of the Bundesbank’s
operational procedures to facilitate the transfer of the Bundesbank’s monetary
policy credibility to the Governing Council.
One significant difference between the Eurosystem and its fixed exchange
rate system predecessor led by the Bundesbank is that monetary policy will
now take account of euro-area aggregate data. Since those data are only recently being created, little is known about their historical behavior or their
relationship to euro-area monetary policy. Until the Eurosystem becomes more
familiar with the new area-wide aggregates, the Governing Council needs to
rely on anecdotal regional information and the intimate knowledge that NCBs
possess of their own country’s data.
Finally, the NCBs have relatively large research departments compared to
the ECB and extensive operational experience in financial and banking markets.
The competence of the Eurosystem will depend on the ability of the ECB to
11 See,
12 See

for instance, Liebscher (1998).
Deutsche Bundesbank (1999).

22

Federal Reserve Bank of Richmond Economic Quarterly

draw on the talents of staff at the NCBs, as need be, for the good of the system
as a whole.
Working Relationships on the Governing Council
Despite the safeguards in the Maastricht Treaty, the independence of the Eurosystem is at risk because the regional members of the Governing Council
represent countries. Members could be influenced by their governments. Votes
on the Governing Council could be traded for those on other governing bodies
of the European Union. As mentioned above, the ambiguity on exchange rate
policy opens the door to political interference in monetary policy. Politically
motivated disputes could greatly complicate the business of the Governing
Council. Such conflicts could cause indecisiveness, inconsistent policy actions,
and a loss of credibility.
FOMC experience suggests a number of additional measures to prevent
the politicization of the Governing Council. First, a macroeconomic framework
should be developed to guide policy deliberations. The framework should be
rich enough to encompass a wide variety of views and sufficiently coherent to
provide the basis for prioritizing concerns and building a consensus for policy
actions. The Governing Council should utilize economic arguments disciplined
by the price stability objective to smoke out and defuse political rhetoric. Economic reasoning is, to repeat, a great equalizer.
Second, the ECB President’s role in the Governing Council should be
strengthened so that he can guide the debate within the agreed upon framework.
The ECB President should act against free riding by encouraging members of
the Governing Council to prepare thoroughly and to participate actively. The
effectiveness of members would be enormously enhanced if each were allowed
to bring an economist advisor to the meetings. A verbatim transcript of the
meetings should be produced, if only for internal use, to facilitate the give and
take that must occur over time.
Third, the macroeconomic framework should be explained to the public
in some detail so that Eurosystem watchers can more readily exercise professional discipline on the internal debate.13 Minutes without individual attribution,
published shortly after each Governing Council meeting, would help focus Eurosystem watchers on issues of concern to policymakers. Over the long run,
greater transparency can serve as a powerful safeguard against political interference.
Admittedly, the FOMC never had the potential for internal international
disputes that exists in the Governing Council. However, FOMC experience
suggests that the above-mentioned practices would facilitate the development
of productive professional working relationships in the Council.
13 See

Issing (1998).

M. Goodfriend: The Role of a Regional Bank

23

Broadening Public Support for the Eurosystem
The Bundesbank has an admirable monetary policy record in large part because
it always had the full support of the German public for its price stability objective. That support was there because the Bundesbank was associated in the
public’s mind with the postwar economic miracle that began in the late 1940s
at the time that the Deutsche Mark and the Bundesbank were created.
The European public has little natural affinity for the new Eurosystem. As
was the case for the Federal Reserve System, the Eurosystem will have to earn
the public’s confidence. If anything, public relations will be more difficult for
the Eurosystem than they have been for the Federal Reserve System because
the euro area is made up of 11 different countries whose citizens speak many
different languages. The Eurosystem should make extensive use of the regional
presence of its NCBs to broaden the understanding of its mission and methods,
much as the Fed uses the Reserve Banks.
The Eurosystem has one big advantage over the Fed in explaining itself
to the public. In contrast to the Fed, whose mandate only exists in the Federal
Reserve Act and is ambiguous at that, the Eurosystem’s price stability mandate
is unambiguous and part of one of the founding documents of the European
Union.

9.

SUMMARY

The main message of this paper is that regional (national) banks play an especially important role in central banks whose currency areas span a continent,
such as the Eurosystem and the Federal Reserve System. A regional presence
facilitates the acquisition of specialized information on the economy and positions the staff to reach out to the public with an explanation of the central
bank’s policy objectives and practices. Presidents (governors) of regional central banks bring analytical diversity to the monetary policy committee. Above
all, a system of central banks promotes a healthy competition that stimulates
innovative thinking on operational, regulatory, research, and policy questions.
Federal Reserve experience teaches that a decentralized system needs a
strong center. Staff at the center needs to be large enough to support a strong
Chairman (President) of the system. The Chairman must be strong enough to
encourage diverse views in the policy committee and to build a consensus for
decisive and timely policy actions. The Chairman should exploit diversity and
promote decisiveness.
The key to success in the Eurosystem, in addition to the above-mentioned
points, is to establish good working relationships on the Governing Council.
To facilitate this, the staff at the center should take the lead in developing a
macroeconomic framework within which diverse policy views can be expressed
and debated productively. Personal advisors should accompany members to the

24

Federal Reserve Bank of Richmond Economic Quarterly

policy meetings. Verbatim transcripts should be prepared for internal use to facilitate an exchange of views over time. Minutes without individual attribution
should be published to present opposing views clearly, to focus central bank
watchers, and to guard against the potential for politically motivated policy
mistakes.
The Eurosystem and the Federal Reserve System will succeed in the long
run by broadening the public’s understanding and support for low inflation and
the preemptive policy procedures to maintain price stability. The way to do
that is to involve the Reserve Bank presidents (national central bank governors)
and their advisors fully in the policymaking process, and to utilize the system’s
regional presence to take the central bank’s monetary policy message to the
public.

REFERENCES
Balke, Nathan, and D’Ann Petersen. “How Well Does the Beige Book Reflect
Economic Activity?,” Federal Reserve Bank of Dallas Working Paper
98-02, June 1998.
Barro, Robert, and Xavier Sala-´
i-Martin. “Convergence,” Journal of Political
Economy, vol. 100 (April 1992), pp. 223-51.
Begg, David. et al. “The ECB: Safe at Any Speed?” London: Center for
Economic Policy Research, October 1998.
Chatterjee, Satyajit, and Gerald Carlino. “Aggregate Employment Growth and
the Deconcentration of Metropolitan Employment,” Federal Reserve Bank
of Philadelphia Working Paper 98-6, March 1998.
Deutsche Bundesbank. Fifty Years of the Deutsche Mark: Central Bank and
the Currency in Germany Since 1948. London: Oxford University Press,
1999.
European Union: Selected Instruments taken from the Treaties, Book 1,
Volume 1. Luxembourg: Office for Official Publications of the European
Communities, 1995.
Friedman, Milton, and Anna Schwartz. A Monetary History of the United
States, 1867-1960. Princeton: Princeton University Press, 1963.
Goodfriend, Marvin. “Monetary Policy Comes of Age: A 20th Century
Odyssey,” Federal Reserve Bank of Richmond Economic Quarterly, vol.
83 (Winter 1997), pp. 1-22.
Hetzel, Robert. “The Rules versus Discretion Debate over Monetary Policy in
the 1920s,” Federal Reserve Bank of Richmond Economic Review, vol. 71
(November/December 1985), pp. 3-14.

M. Goodfriend: The Role of a Regional Bank

25

Issing, Otmar. “Monetary Policy in EMU.” Speech delivered at the 1998
Annual Meeting of the International Monetary Fund and the World Bank
Group in Washington, D.C., October 1998.
Liebscher, Klaus. “The Role of a National Central Bank within the ESCB as
Illustrated by the OENB.” Vienna: Austrian National Bank, May 1998.
Meyer, Laurence. “Come with Me to the FOMC,” Federal Reserve Bank of
Minneapolis The Region, vol. 12 (June 1998), pp. 6-15.
Wynne, Mark. “The European System of Central Banks,” Federal Reserve
Bank of Dallas Economic Review, (First Quarter 1999), pp. 2-14.

The Business Cycle and
Industry Comovement
Andreas Hornstein

T

he U.S. economy, as of the writing of this article, is in its longest
postwar expansion. This expansion has prompted various proponents to
declare a “new” economy and the death of the business cycle. These
pronouncements may well turn out to be premature, as similar announcements
have proved in the past. In this case we can already say that the current expansion shares one feature with all previous business cycles, namely that all
parts of the economy take part in the expansion, although possibly to different
degrees. In particular, although the symbol of the “new” economy appears to
be the Internet, a general expansion of all industries in the manufacturing and
the service sector accounts for the growth in GDP. Indeed, it is the general up
and down movement of all parts of the economy that defines the business cycle
in Burns and Mitchell’s (1946) early work.
In contrast to this earlier work, modern business cycle research has focused for the most part on the comovement of aggregate variables, like output,
employment, consumption, investment, the price level, interest rates, etc. In
part, the focus on the aggregate economy has been justified by the observed
comovement, which is supposed to indicate the presence of common aggregate disturbances to which all parts of the economy respond in a similar way.
The argument for aggregate shocks as the source of business cycles proceeds
as follows (Lucas 1977). Suppose the economy is subject to a large number
of industry-specific disturbances which are unrelated to each other. Then we
would expect that these disturbances change the relative productivities of various inputs such as labor. This change in relative productivities, in turn, should
lead to a reallocation of inputs. That is, input use should decline in industries with falling relative productivities and should increase in industries with
Andreas.Hornstein@rich.frb.org. I would like to thank Yash Mehra, Pierre Sarte, and John
Weinberg for helpful comments. The views expressed are the author’s and not necessarily
those of the Federal Reserve Bank of Richmond or the Federal Reserve System.

Federal Reserve Bank of Richmond Economic Quarterly Volume 86/1 Winter 2000

27

28

Federal Reserve Bank of Richmond Economic Quarterly

rising relative productivities. What we actually observe, however, is the opposite outcome; therefore we should conclude that the business cycle is not due to
unrelated industry disturbances, but rather to aggregate disturbances that affect
all sectors of the economy. One natural candidate for an aggregate disturbance
is, of course, monetary policy. Given the current economic expansion, which
appears to be driven to some extent by the widespread application of computer
technology, aggregate productivity shocks are also a possibility.
In this article, I argue that industry comovement is an important defining
characteristic of the business cycle, and that current economic theory has difficulties accounting for this characteristic. I first document the pattern of industry
comovement for inputs and outputs. I then discuss a simple extension of the
standard aggregate business cycle model to make two points. First, I formalize
the argument against unrelated industry disturbances as the cause of business
cycles. Second, I point out that even if there are only aggregate disturbances,
one should not necessarily expect that all sectors of the economy respond in
the same way to these disturbances. Finally, I provide some evidence on the
extent to which the economy is subject to aggregate productivity disturbances.

1.

COMOVEMENT IN THE U.S. ECONOMY

Industry comovement over the business cycle means that the level of activity
in different industries increases and decreases together. There are various ways
to measure the activity level of an industry. One method is to ask how many
inputs are used or how many goods are produced in an industry. For this
article, I use data from Jorgenson, Gollop, and Fraumeni (1987), who provide
annual series on inputs and outputs at the two-digit industry level. Their data
set covers prices and quantities for industry gross output and use of capital services, labor, materials, and energy for the years 1950-1991.1 I will show that,
for almost all measures of activity, industries move together over the business
cycle. This result confirms previous work by Christiano and Fitzgerald (1998),
who study the comovement of quarterly two-digit industry employment, and
Murphy, Shleifer, and Vishny (1989), who study annual one-digit industry value
added and employment.
In addition to short-term business cycle fluctuations, most economies are
characterized by substantial structural change over time. This change means
that some industries are growing, and their production and use of resources is
increasing over time relative to other industries such as services or the financial
industries. Likewise, other industries’ contribution to the economy is declining,
1 The data used here are taken from Jorgenson’s Web page at http://www.economics.harvard.
edu/faculty/jorgenson/data.html. All industries of the data set are included, except agriculture
(1) and government enterprises (36).

A. Hornstein: The Business Cycle and Industry Comovement

29

such as textiles. Since I am not interested in the long-run secular changes of
industries, I remove this trend component by using a band pass filter.2
I study the comovement of industries using two different measures. For
the first measure, I consider the comovement of industry variables with their
corresponding aggregate counterparts, for example the comovement of industry
employment with aggregate employment. For the second measure, I consider
cross-industry comovement directly, for example the pairwise industry employment correlations. I find that in almost all industries employment is positively
correlated with aggregate employment and that this relationship is quite tight.
Furthermore, the same positive comovement of industry variables with aggregate variables occurs for all other output and input measures, such as gross
output, value-added, capital services, employment, and intermediate inputs. For
pairwise cross-industry correlations, positive correlations are also much more
frequently observed than negative correlations. Finally, the positive comovement pattern does not only apply to the manufacturing sector but also to the
service sector and the construction industry. Only the mining sector has several
industries which do not move in step with the rest of the economy.
In order to study the comovement of industry series with aggregate series,
I construct aggregate quantities as Divisia indices using the price and quantity
industry series. A Divisia index is a way to weight the contribution of individual series to the aggregate series. Suppose we have a collection of goods
with prices and quantities for different time periods {qit , pit : i = 1, . . . , N and
t = 1, . . . , T}; then we define the growth rate of the aggregate quantity index
between periods t and t + 1 as the weighted sum of the growth rates of the
individual series
N

ωit ∆ ln qit ,
¯

∆ ln qt =
i=1

where an individual series’ weight is its average value share ωit = 0.5(ωi,t+1 +
¯
ωi,t ) and ωi,t = pi,t qi,t / j=1,N pj,t qj,t . I use this method to construct aggregate
input and output series from the industry series and to construct for each industry an intermediate goods index from the materials and energy use series.
For each industry, I also construct a value-added quantity index (Sato 1976).
Value added of an industry is the total value of payments that goes to primary
factors of production: capital and labor. Alternatively, value added represents
the industry’s value of production after deducting payments for inputs, which
have been purchased from other industries in the current accounting period,
namely intermediate inputs.
2I

identify the components of a time series with periodicity less than or equal to eight
years with the business cyle. The band pass filter which extracts the business cycle component
is approximated by a symmetric moving average with four leads and lags. For a description of
band pass filters, see Hornstein (1998) or Christiano and Fitzgerald (1998).

30

Federal Reserve Bank of Richmond Economic Quarterly

Comovement of Sectoral Variables with Aggregate Variables
The results for the comovement of industry variables with aggregate variables
are displayed in Tables 1a and 1b. Table 1a displays whether an industry series
increases or decreases when its corresponding aggregate series increases. Most
industry series move contemporaneously with their aggregate counterpart, but
I want to allow for the possibility that an industry series is leading or lagging the aggregate series. For this purpose, Table 1a displays the correlation
which is maximal in absolute value among the contemporaneous, once-lagged
and once-led correlations. In Table 1b, I provide a measure of how tight the
relation between the industry and the aggregate economy is. For this purpose
I regress the industry series on one lagged value, one leading value, and the
contemporaneous value of the aggregate series. Table 1b then displays the R2 of
this regression, that is the variation of the industry series explained by variation
of the aggregate series through this regression equation. The higher is the R2 ,
the tighter is the fit between the industry and the aggregate series.
Industry employment in the manufacturing sector moves with aggregate
employment, as Table 1a demonstrates. The correlation between industry and
aggregate employment in the manufacturing sector (industries 7 through 27)
are all positive, and almost all of them are contemporaneous and quite high, at
least 0.4 or higher. Furthermore, as we can see from Table 1b, the relationship
between the industry and the aggregate series are quite tight with R2 s of at least
0.4. The main exceptions are tobacco (8), petroleum and coal (16), and food (7),
industries where employment is not closely related to the aggregate economy.3
Notice that these are industries which are subject to shocks exogenous to the
aggregate economy, like weather or world oil markets, and whose contribution
to the aggregate economy is limited.
The close relation between industry and aggregate variables also holds for
other inputs and outputs. With few exceptions, industry gross output, value
added, use of intermediate goods, and capital services are all positively correlated with the corresponding aggregate variables. The exceptions concern
tobacco (8), leather (18), apparel (10), lumber and wood (11), petroleum and
coal (16), primary metals (20), and transportation equipment (25). To the extent
that an industry variable declines when the aggregate increases, the relationship
tends to be quite weak, with R2 s less than 0.2. Only the use of capital services
in primary metals (20) has a strong negative correlation with a high R2 . These
results are consistent with Murphy, Shleifer, and Vishny (1989).
The evidence for industry comovement with aggregate variables is not
limited to the manufacturing sector. We also find strong evidence for the service
sector and the construction industry. Employment in service sector industries
3 This

evidence confirms Christiano and Fitzgerald’s (1998) analysis of employment in the
manufacturing sector with monthly data.

A. Hornstein: The Business Cycle and Industry Comovement

31

Table 1a Maximal Correlation of Industry Series with Aggregate Series
Sector
2 Metal mining
3 Coal mining
4 Oil and gas extraction
5 Non-metallic mining
6 Construction
7 Food
8 Tobacco
9 Textile mill products
10 Apparel
11 Lumber and wood
12 Furniture and fixtures
13 Paper and allied
14 Printing
15 Chemicals
16 Petroleum and coal
17 Rubber and misc. plastics
18 Leather
19 Stone, clay, and glass
20 Primary metal
21 Fabricated metal
22 Machinery, non-electrical
23 Electrical machinery
24 Motor vehicles
25 Transportation equipment
26 Instruments
27 Misc. manufacturing
28 Transportation
29 Communications
30 Electric utilities
31 Gas utilities
32 Trade
33 FIRE
34 Services

q
0.53
−0.31+
0.47
0.66
0.74
0.45
−0.21+
0.78
0.72
0.72
0.93
0.78
0.70
0.92
0.53
0.88
−0.30−
0.92
0.90
0.93
0.79
0.86
0.78
0.40−
0.70
0.73
0.88
0.43
0.71
−0.29+
0.79
0.41+
0.69

y
0.44
0.51
0.73
0.72
0.61
0.44
0.38
0.46+
0.67
−0.30
0.90
0.75
0.69
0.77
0.62+
0.78
−0.39
0.89
0.75
0.88
0.79
0.84
0.77
0.49−
0.73
0.65
0.75
−0.45+
0.66
−0.57+
0.82
0.22+
0.71

k

l

x

−0.13+ 0.44
0.41
−0.13
0.30 −0.51+
0.30 −0.48+ −0.56+
0.25− −0.31+ 0.28
0.32
0.70
0.74
0.31
0.29+ 0.20−
−0.24+ 0.19 −0.32
0.58
0.66
0.82
0.55
−0.43− 0.52
0.43
0.77
0.75
0.70
0.84
0.87
0.62
0.69
0.62
0.67
0.50
0.62
0.72
0.77
0.73
0.50
0.37+ 0.50
0.31
0.85
0.84
0.60
0.46
0.63
0.75
0.84
0.87
0.89
−0.67+ 0.65
0.76
0.86
0.89
0.75
0.86
0.75
0.72
0.88
0.86
0.66
0.79
0.77
0.64− 0.62
0.33−
0.76
0.67
0.55
0.46
0.47
0.65
0.61
0.84
0.90
0.33
0.55
0.56
0.17+ 0.58− 0.18
0.38
0.66
0.41+
0.82
0.81
0.62
0.75
0.24− 0.49+
0.72
0.74
0.47

m

e

0.39
0.57
−0.52+ −0.35+
−0.57+ 0.19+
0.29
0.47
0.74
0.67
0.21− 0.26−
−0.33 −0.30−
0.83
0.64
0.54
0.57
0.74
0.68
0.87
0.76
0.62
0.62
0.62
0.54
0.72
0.50
−0.31+ 0.70
0.85
0.60
0.63
0.65
0.87
0.75
0.90
0.69
0.89
0.77
0.75
0.69
0.86
0.73
0.77
0.65
0.32− −0.22+
0.55
0.54
0.64
0.65
0.91
0.78
0.56
0.35
0.15
0.45
−0.24− 0.55
0.62
0.57
0.47+ 0.37+
0.46
0.58

Note: The industry series are gross output q, value-added y, capital k, employment l, intermediate input
aggregate x, materials m, and energy e. A correlation is the maximal correlation in absolute value of the
contemporaneous, one-period lagged, and one-period leaded correlation between the industry variable zi
and the corresponding aggregate variable z, corr[zi,t , zt+s ] with s = 1, 0, −1. A plus (minus) superscript
¯
¯
denotes that the industry variable is leading (lagging) the aggregate variable, that is s = 1 (s = −1). No
superscript indicates that the contemporaneous correlation is maximal.

(28-34) and construction (6) tends to show a strong positive correlation with
aggregate employment above 0.5, and the relationship tends to be quite tight,
with R2 above 0.5. Finance, insurance, and real estate (FIRE) (33) is the only industry where employment is not tightly correlated with aggregate employment.

32

Federal Reserve Bank of Richmond Economic Quarterly

Table 1b R2 from Regression of Industry Series on Aggregate Series
Sector
2 Metal mining
3 Coal mining
4 Oil and gas extraction
5 Non-metallic mining
6 Construction
7 Food
8 Tobacco
9 Textile mill products
10 Apparel
11 Lumber and wood
12 Furniture and fixtures
13 Paper and allied
14 Printing
15 Chemicals
16 Petroleum and coal
17 Rubber and misc. plastics
18 Leather
19 Stone, clay, and glass
20 Primary metal
21 Fabricated metal
22 Machinery, non-electrical
23 Electrical machinery
24 Motor vehicles
25 Transportation equipment
26 Instruments
27 Misc. manufacturing
28 Transportation
29 Communications
30 Electric utilities
31 Gas utilities
32 Trade
33 FIRE
34 Services

q

y

k

l

x

m

e

0.50
0.20
0.34
0.60
0.78
0.25
0.12
0.85
0.65
0.79
0.90
0.69
0.61
0.85
0.35
0.81
0.19
0.86
0.92
0.93
0.88
0.82
0.74
0.34
0.67
0.60
0.82
0.34
0.56
0.17
0.76
0.25
0.52

0.33
0.28
0.61
0.68
0.60
0.41
0.22
0.21
0.50
0.12
0.83
0.61
0.55
0.65
0.40
0.64
0.21
0.80
0.64
0.81
0.82
0.83
0.73
0.37
0.70
0.47
0.62
0.25
0.47
0.36
0.79
0.07
0.60

0.04
0.05
0.15
0.12
0.09
0.06
0.08
0.35
0.31
0.19
0.60
0.51
0.41
0.49
0.20
0.33
0.43
0.61
0.50
0.53
0.63
0.52
0.50
0.42
0.57
0.32
0.37
0.20
0.06
0.18
0.72
0.56
0.53

0.23
0.18
0.35
0.14
0.58
0.16
0.07
0.57
0.43
0.70
0.80
0.57
0.25
0.74
0.12
0.83
0.53
0.74
0.44
0.76
0.82
0.81
0.80
0.54
0.60
0.33
0.75
0.59
0.54
0.60
0.65
0.16
0.60

0.26
0.25
0.34
0.16
0.81
0.08
0.20
0.89
0.48
0.84
0.84
0.51
0.55
0.62
0.36
0.78
0.57
0.84
0.89
0.87
0.83
0.77
0.71
0.31
0.45
0.51
0.84
0.35
0.06
0.24
0.51
0.35
0.29

0.24
0.27
0.37
0.14
0.80
0.09
0.19
0.89
0.46
0.83
0.83
0.50
0.56
0.62
0.21
0.78
0.58
0.84
0.90
0.88
0.83
0.76
0.71
0.30
0.45
0.49
0.85
0.35
0.03
0.13
0.50
0.35
0.29

0.38
0.15
0.05
0.31
0.65
0.09
0.13
0.48
0.40
0.63
0.61
0.39
0.33
0.29
0.60
0.50
0.49
0.65
0.54
0.61
0.63
0.65
0.54
0.24
0.43
0.51
0.62
0.26
0.20
0.33
0.32
0.24
0.38

Variables other than employment also increase and decrease with their aggregate counterparts. Only for communications (29) and gas utilities (31) do we
observe some negative comovement, but the relationship is not very strong, as
the low R2 s indicate. Mining is the only sector which does not always increase
and decrease with the rest of the economy. In particular, coal mining (3) and
oil and gas extraction (4) are not synchronized with the aggregate economy.
Construction, manufacturing, and services contribute about 95 percent to
private sector value added and employ almost all labor in that sector. Thus, for
the majority of the U.S. economy’s industries, gross output, value added, the

A. Hornstein: The Business Cycle and Industry Comovement

33

use of capital services, employment, and intermediate inputs tend to increase
and decrease with their aggregate counterparts.
Comovement of Variables Across Sectors
Not only do individual industries move with the aggregate economy, but there
is also strong evidence that industries move together individually.4 Tables 2a
through 2d and Figures 1 through 4 display some of the evidence on the pairwise comovement between industries. Even for a small number of industries,
there exists a large number of possibilities to pair any two of these industries.
I represent this information in two ways. In Tables 2a through 2d, I show the
quartile and average values for the pairwise maximal correlations. In Figures
1 through 4, I show the histograms for the maximal and contemporaneous
pairwise correlations.
Consider the manufacturing sector. As seen in Figure 1 and Table 2a, the
pairwise correlations for industry inputs and outputs are predominantly positive
and quite high. Gross output, employment, and energy use display a consistent and strong positive correlation across industries. The average correlation
coefficient is about 0.5, and more than three-fourths of all industries are positively correlated with each other. Capital services are less strongly correlated,
and there is a relatively high number of negative correlations for value added
and material use, especially for maximal correlations. The average correlation
coefficient for these variables remains positive, about 0.3.
As noted previously, the manufacturing sector industries that produce durable goods tend to be more closely related than those that produce
nondurable goods (Christiano and Fitzgerald 1998).5 Figures 2 and 3 and
Tables 2b and 2c confirm this observation. For more than three-fourths of all
industries in the durable goods manufacturing sector, we find that all output and
input measures are positively correlated across industries, the average correlation coefficient being about 0.5. In the nondurable goods manufacturing sector,
employment, energy, and gross output display consistent positive correlations
across industries, while value added, capital use, and especially material use
show a number of negative correlations. The negative correlations are mainly
due to one industry, tobacco (8), which as already noted above is not that
tightly related to the aggregate economy.
Finally, consider the cross-industry correlation pattern for the private business sector, excluding mining, summarized in Figure 4 and Table 2d. As we can
4 It is useful to study the comovement of individual industries for the following reason. An
aggregate series is the sum of sectoral series. Therefore, even if the sectoral series are independent of each other, we would observe that each individual series is positively correlated with the
aggregate series since it is perfectly correlated with its own contribution to the aggregate series.
5 Another difference between the nondurable goods and the durable goods sector is that
output and input use tends to be more volatile for industries in the latter sector.

34

Federal Reserve Bank of Richmond Economic Quarterly

Figure 1 Frequency Distribution of Cross-Industry Correlations for
All Manufacturing Industries

see, cross-industry correlations for employment, capital services, energy use,
and gross output are consistently positive, whereas the pattern is somewhat
weaker for value added and materials use. Again, the negative correlations we
observe can be attributed to a small number of industries. For gross output,

A. Hornstein: The Business Cycle and Industry Comovement

35

Table 2a Manufacturing Sector: Maximal Cross-Correlations
q
Minimum
1st Quartile
Median
3rd Quartile
Maximum
Average

y

k

l

x

m

e

−0.64
0.30
0.56
0.74
0.94
0.44

−0.58
−0.15
0.44
0.61
0.90
0.30

−0.59
0.15
0.40
0.56
0.87
0.30

−0.60
0.28
0.47
0.61
0.91
0.41

−0.62
−0.26
0.48
0.66
0.91
0.29

−0.63
−0.30
0.43
0.66
0.91
0.25

−0.40
0.35
0.55
0.68
0.90
0.51

Note: For notation see Table 1a.

Table 2b Manufacturing Sector: Nondurable Goods, Maximal
Cross-Correlations
q
Minimum
1st Quartile
Median
3rd Quartile
Maximum
Average

y

k

l

x

m

−0.29
0.25
0.47
0.66
0.89
0.40

−0.58
−0.25
0.29
0.49
0.90
0.19

−0.44
−0.15
0.25
0.40
0.67
0.16

−0.58
0.22
0.33
0.49
0.91
0.32

−0.60
−0.35
−0.11
0.60
0.84
0.10

−0.62
−0.40
−0.21
0.59
0.84
0.05

e
0.09
0.32
0.42
0.60
0.88
0.46

Note: For notation see Table 1a.

Table 2c Manufacturing Sector: Durable Goods, Maximal
Cross-Correlations
q
Minimum
1st Quartile
Median
3rd Quartile
Maximum
Average

y

k

l

x

m

e

−0.53
0.55
0.67
0.78
0.91
0.53

−0.42
0.40
0.59
0.71
0.89
0.45

−0.56
0.42
0.54
0.62
0.87
0.48

−0.45
0.49
0.61
0.70
0.88
0.57

−0.57
0.42
0.60
0.72
0.88
0.43

−0.56
0.41
0.60
0.72
0.88
0.43

−0.40
0.52
0.62
0.74
0.90
0.58

Note: For notation see Table 1a.

most of the negative correlations are accounted for by tobacco (8) and gas
utilities (31). For value added, most of the negative correlations are accounted
for by leather (18), lumber and wood (11), and gas utilities (31).

36

Federal Reserve Bank of Richmond Economic Quarterly

Figure 2 Frequency Distribution of Cross-Industry Correlations,
Nondurable Goods Only

A. Hornstein: The Business Cycle and Industry Comovement
Figure 3 Frequency Distribution of Cross-Industry Correlations,
Durable Goods Only

37

38

Federal Reserve Bank of Richmond Economic Quarterly

Figure 4 Frequency Distribution of Cross-Industry Correlations for
All Industries, Except Mining

A. Hornstein: The Business Cycle and Industry Comovement

39

Table 2d All Industries Except Mining: Maximal
Cross-Correlations
q
Minimum
1st Quartile
Median
3rd Quartile
Maximum
Average

y

k

l

x

m

e

−0.64
0.25
0.51
0.67
0.94
0.39

−0.72
−0.21
0.39
0.57
0.90
0.24

−0.72
0.14
0.32
0.49
0.87
0.25

−0.64
0.27
0.44
0.59
0.91
0.38

-0.62
0.13
0.41
0.59
0.91
0.30

-0.63
−0.15
0.38
0.58
0.91
0.26

-0.49
0.28
0.45
0.62
0.90
0.40

Note: For notation see Table 1a.

2.

COMOVEMENT IN TWO SIMPLE DYNAMIC
GENERAL EQUILIBRIUM MODELS

In this section, I discuss why the observed comovement of inputs and outputs
across industries is difficult to reconcile with the basic business cycle model.
First, I describe a simple model where labor is the only input to production,
and where permanent changes in productivity do not affect employment. For a
two-sector version of this economy, I then formalize the argument that changes
in relative productivities cause sectoral employment to move in opposite directions. Finally, I discuss a two-sector interpretation of the neoclassical growth
model. In this model, employment in the consumption and investment goods
sectors move in opposite directions following an aggregate shock that affects
production equally in the two sectors.
A Simple Model of Production and Employment
Consider the following simple economy. There are two goods, consumption c
and labor n. Labor is used to produce the consumption good
c = znα ,
with 0 < α ≤ 1, and z is labor productivity. Output is produced under conditions of constant returns to scale if α = 1. A representative agent has a fixed
labor endowment of 1 which can be supplied as labor or used as leisure l,
n + l = 1. Preferences over consumption and leisure are
[clγ ]1−σ − 1
,
(1)
1−σ
with σ, γ ≥ 0. The competitive equilibrium of this economy is Pareto-optimal.
For this setup, Pareto-optimality means that the equilibrium allocation of consumption and labor maximizes the utility of the representative agent subject to
being feasible. An allocation is optimal, if at the margin, the utility loss from
u (c, l) =

40

Federal Reserve Bank of Richmond Economic Quarterly

using one more unit of labor in production is equal to the utility gain derived
from the additional consumption produced by that unit of labor:
∂u
∂u ∂c
·
=
∂c ∂n
∂l

(2)

We can use this condition to solve for the optimal labor supply:6
α
.
n=
α+γ
Notice that optimal employment is independent of productivity z. An increase
of productivity raises the marginal product of labor and thereby the real wage.
Because a higher real wage makes leisure relatively more expensive, the agent
consumes less leisure and supplies more labor. This result is called the substitution effect of the real wage increase. A rise in real wages also increases the
income of the agent, thereby increasing the demand for leisure and reducing the
labor supply. This result is called the income effect of the real wage increase.
For the class of preferences defined by (1), the income and substitution effect
cancel each other and employment does not depend on the productivity level
(King, Plosser, and Rebelo 1987). This property of preferences is desirable
if we want to match the long-run behavior of employment in industrialized
countries. Relative to the increase in labor productivity over the last hundred
years, per capita employment has scarcely moved.
Changes in Relative Productivities
Now consider a two-sector version of the economy described above. To keep
things simple, I will treat the two sectors symmetrically. Essentially, the two
sectors will only differ with respect to their relative labor productivities. In this
economy, aggregate employment also does not depend on labor productivity.
Furthermore, employment in the two sectors will always move in opposite
directions if relative labor productivity changes. Thus there is negative comovement of employment if productivity changes in the two sectors are not
perfectly correlated.
Production of each consumption good is
ci = zi nα ,
i
with 0 < α ≤ 1. The agent’s preferences for the two consumption goods and
the labor supply for the two sectors are defined in two stages. First, there is a
utility index for aggregate consumption:
c = (cρ + cρ )1/ρ ,
1
2
6 Using the definition of the production and utility function and substituting for marginal utilic
ties and marginal product of labor yields c−σ (1 − n)γ(1−σ) · α n = γc1−σ (1 − n)γ(1−σ)−1 ,
which can be solved for n.

A. Hornstein: The Business Cycle and Industry Comovement

41

with ρ ≤ 1. If ρ = 1, then the two goods are perfect substitutes. If ρ = 0, then
the elasticity of substitution is unitary, and the agent spends constant and equal
shares of income on the two goods. There is also a disutility index for labor
supply in the two sectors:
n = (nψ + nψ )1/ψ ,
1
2
with ψ ≥ 1. Labor supplied to the two sectors is a perfect substitute when
ψ = 1. The agent’s utility is again a function of the consumption and leisure
l = 1 − n as defined in (1). For each consumption good, an optimal allocation
equates the marginal utility gain from consuming one more unit of the good
with the marginal utility loss from producing this good,
∂u ∂n
∂u ∂ci
·
·
=
.
∂ci ∂ni
∂l ∂ni
This optimality condition, after some algebraic manipulation, simplifies to
α

ci
c

ρ

=γ

n
1−n

ni
n

ψ

for i = 1, 2.

The ratio of the two optimality conditions yields an expression for the relative
employment as a function of relative productivities:
n1
=
n2

z1
z2

ρ/(ψ−αρ)

.

If the two goods are substitutable in consumption, ρ > 0, then employment
in sector one increases relative to employment in sector two if the relative
productivity of sector one increases. On the other hand, if the two goods are
complementary in consumption, ρ < 0, then employment is shifted from the
relatively more productive sector to the less productive sector, because the agent
tries to maintain the same consumption ratio. With some additional algebraic
manipulations, we can show that aggregate employment is again independent
of productivity, that is, the percentage increase of employment in one sector
is always balanced by the same percentage reduction of employment in the
other sector. This result, of course, implies that employment in the two sectors
always moves in opposite directions if relative productivities change.
Changes in Aggregate Productivity
I now reinterpret the standard neoclassical growth model as a two sector economy with a consumption goods sector and an investment goods sector. In
contrast to the findings of the previous section, this example demonstrates that
even without any change in relative sectoral productivities in the two sectors,
employment in one sector can move opposite to that in the other simply because

42

Federal Reserve Bank of Richmond Economic Quarterly

the two sectors respond differently to the same shock.7
Consider the representative agent described before but now assume that
the agent is infinitely lived and has the utility function (1) for every period.
The agent discounts future utility at rate 0 < β < 1 and the utility from the
consumption-labor sequence {ct , nt } is given by
∞

β t u(ct , 1 − nt ).
t=0

There is now one consumption good and an investment good. The investment
good xt is used to augment the capital stock kt according to
kt+1 = (1 − δ)kt + xt
and 0 < δ < 1 is the capital depreciation rate. In the standard neoclassical
growth model, capital and labor are used to produce a homogenous output good
that can be used for investment or consumption. For the present interpretation
of the model, I instead assume that the investment and consumption good are
produced in two distinct sectors with the technologies
1−α
1−α
ct = zt kct nα and xt = zt kxt nα ,
ct
xt

where 0 < α < 1. Notice that relative productivity in the two sectors does not
change. There are only aggregate productivity changes. The total amount of
capital and labor used has to satisfy
kt = kct + kxt and nt = nct + nxt .
Again the competitive equilibrium is Pareto-optimal. Furthermore, the equilibrium allocations of this economy are the same as in the standard growth
model.
How does this economy respond to a productivity increase? In general, we
cannot derive analytical solutions for the behavior of equilibrium allocations for
this economy, rather we have to derive numerical solutions. (See for example
King, Plosser, and Rebelo [1987] or Benhabib, Rogerson, and Wright [1991]).
It is straightforward, however, to interpret the economy’s equilibrium response
to the productivity increase. In this economy, output, consumption, investment,
and employment all increase. Consumption increases because the representative agent prefers more consumption to less, and higher productivity enables
the economy to produce more goods for consumption with the same amount
of resources. Investment increases because the household accumulates capital

7 This

observation appears in Benhabib, Rogerson, and Wright (1991). See also Christiano
and Fitzgerald (1998).

A. Hornstein: The Business Cycle and Industry Comovement

43

in order to smooth consumption over time. Employment increases because the
higher productivity increases the real wage and labor supply.8
What are the implications of the model for sectoral comovement? The
model clearly captures the sectoral comovement of output. But I will now show
that even though both consumption and investment increase, employment in the
two sectors move in opposite directions. Consider the intratemporal optimality
condition for the allocation of labor to the production of consumption goods,
which is essentially the same as (2) above. At the margin, the utility gain from
the production of one additional consumption good has to be balanced by the
utility loss from the additional labor supply,
∂ut
∂ut ∂ct
·
=
.
∂ct ∂nct
∂lt

(3)

Using the definitions of the utility and production functions, we can simplify
this expression9 to
α
γ
=
.
nct
1 − nct − nxt
This equation clearly shows that following the productivity increase, employment in the consumption goods sector falls since total employment increases. Furthermore, employment in the investment goods sector has to increase because total employment increases and employment in the consumption
goods sector declines. Thus, employment in the two sectors moves in opposite
directions.10 To understand this behavior of sectoral employment, note that
higher employment in the consumption goods sector implies two things. It implies a decline of the marginal product of labor in the production of consumption
goods. Likewise, it implies a decline of the marginal utility of consumption
since consumption increases. For the optimality condition (3) to be satisfied,
the marginal utility of leisure has to decline, that is, the consumption of leisure
has to increase. But leisure can only increase if employment in the investment
goods sector declines, since employment in the consumption goods sector is
assumed to increase.

8 In

this model, as in the previous static models, a change in productivity has both substitution and wealth effects. The higher productivity increases the marginal product of labor, which
induces the agent to substitute from leisure to work time. It also increases wealth, which induces
the agent to consume more leisure. In the long run, the two effects cancel each other, but during
a transitional period, the substitution effect dominates.
c
9 The expression is c−σ (1 − n )γ(1−σ)
α n t = γc1−σ (1 − nt )γ(1−σ)−1
t
t
t
ct
10 Christiano and Fitzgerald (1998) discuss the comovement problem for a somewhat more
general specification of the growth model.

44

3.

Federal Reserve Bank of Richmond Economic Quarterly

TOTAL FACTOR PRODUCTIVITY COMOVEMENT

As shown in the last section, simple multisector extensions of the neoclassical
growth model have difficulties accounting for industry comovement in the presence of aggregate or sector-specific disturbances. The question remains whether
the economy is mainly driven by aggregate or sector-specific disturbances.
In the introduction, I alluded to monetary policy shocks as a possible
aggregate shock. Here I look at whether we should think of productivity disturbances as aggregate or sector-specific shocks. For this purpose, I study the
comovement of measures of total factor productivity (TFP) across industries. I
find that TFP in different industries move together over the business cycle, but
that comovement appears to be weaker than for outputs or inputs. This finding
seems to indicate that industry changes in productivity are not dominated by
aggregate productivity changes.
Consider an industry where output is produced using capital, labor, materials, and energy as inputs to a constant returns-to-scale technology
qt = zt f (kt , nt , mt , et ),

(4)

and z represents industry TFP. Changes in output can be attributed to corresponding changes in inputs and TFP, and a first order approximation of the
change in output is
dqt = zt [fk,t dkt + fn,t dnt + fm,t dmt + fe,t det ] + ft dzt ,
where fk,t = ∂f (kt , nt , mt , et )/∂kt and similarly for the other inputs. Dividing the
equation by output yields an expression for output growth as a weighted sum
of input growth rates and the TFP growth rate:
fk,t kt dkt
fn,t nt dnt
fm,t mt dmt
fe,t et det
dzt
dqt
=
+
+
+
+
,
qt
ft kt
ft nt
ft mt
ft et
zt
where each input’s weight is equal to the elasticity of output with respect to
that input. For given weights, we can use this expression to solve for the TFP
growth rate.
Solow’s (1957) important insight was that, in a competitive economy, input elasticities can be measured through observations on factor income shares.
Suppose that the industry is competitive in input and output markets and that
everybody has access to the technology represented by (4). Consider a firm
which maximizes profits, sells the output good at a price pt , and hires, or
purchases the services of, inputs capital, labor, materials, and energy at prices
wkt , wnt , wmt , and wet . In order to maximize profits, a firm will hire labor until
the marginal revenue from the last unit of labor hired equals its price, that is,
pt fn,t = wnt .

A. Hornstein: The Business Cycle and Industry Comovement

45

Multiplying each side of the equation with nt /pt qt shows that the elasticity of
output with respect to labor is equal to the share in total revenues that goes to
labor:
wnt nt
fnt nt
=
= ωnt .
ft
pt qt
The same applies to all other inputs. We can therefore measure productivity
growth using observations on output growth, input growth, and revenue shares
of inputs. This measure of TFP growth is the Solow residual:
dqt
dkt
dnt
dmt
det
dzt
=
− ωkt
+ ωnt
+ ωmt
+ ωet
.
zt
qt
kt
nt
mt
et

(5)

The Solow residual provides an accurate measure of disembodied technical
change as long as we are willing to assume that all markets are competitive.
Table 3 characterizes the business cycle comovement of industry TFP for
the Jorgenson, Gollop, Fraumeni data set used in Section 2.11 For the TFP
calculations, I consider two production/output concepts: TFP of all inputs with
respect to gross output and TFP of primary inputs capital and labor with respect
to value added. The qualitative features of TFP comovement are similar to
those of input and output comovement. First, there is some evidence that TFP
in the different industries increases and decreases together. Second, industries
in the manufacturing sector appear to move closer together than do industries
in the rest of the economy, and within the manufacturing sector we see more
comovement in the durable-goods-producing sector than in the nondurablegoods-producing sector. These observations apply to both gross output based
and value-added-based measures of TFP. However, there appears to be less
comovement of industry TFP than of industry output and labor input, as seen
from the lower average and median correlation coefficients for TFP measures.
From this I conclude that there is no strong evidence for an aggregate TFP
shock.12
11 I have calculated TFP growth rates for all industries using equation (5). After nomalizing
TFP at one in the initial year, TFP levels are calculated as the cumulative sum of the growth rates.
The business cycle component is then obtained by detrending industry TFP with the bandpass
filter discussed in footnote 2.
12 In recent work, for example Basu and Fernald (1999), it has been questioned whether
Solow residuals accurately measure TFP movements. The issue is whether the assumption of
perfect competition and constant returns to scale is appropriate and whether there is substantial
unmeasured input variation. It is unlikely that these objections substantially affect the results on
comovement of industry TFP. First, in the absence of perfect competition in the product markets
or constant returns to scale, one would have to adjust the scale of the measured TFP movements,
which would affect the volatility of measured TFP but it is unlikely to affect the industry comovement pattern. Second, most of the empirical work which tries to account for unmeasured
input variation, uses some measured input as a proxy. Since we observe positive comovement
for measured industry inputs, this correction removes a component from TFP measures that is
positively correlated across industries, and corrected TFP measures are likely to display even less
comovement.

46

Federal Reserve Bank of Richmond Economic Quarterly

Table 3 Maximal Industry Cross-Correlations for Total
Factor Productivity
Gross Output Based

Value-Added Based

Mft
Minimum
1st Quartile
Median
3rd Quartile
Maximum
Average

4.

NDR
Mft

DUR
Mft

All

Mft

NDR
Mft

DUR
Mft

All

−0.70
−0.25
0.32
0.47
0.88
0.17

−0.61
−0.30
0.18
0.51
0.88
0.13

−0.70
−0.31
0.39
0.49
0.68
0.20

−0.70
−0.27
0.26
0.43
0.88
0.13

−0.68
−0.26
0.30
0.47
0.89
0.16

−0.62
−0.30
0.15
0.52
0.89
0.12

−0.68
−0.31
0.38
0.49
0.69
0.19

−0.68
−0.27
0.27
0.43
0.89
0.12

CONCLUSION

I have documented that, over the business cycle, activity in almost all industries
of the economy simultaneously increases and decreases. This comovement can
be observed for a wide variety of activity measures, such as gross output, value
added, employment, the use of capital services, or intermediate inputs. Based
on this finding one might conjecture that aggregate disturbances to which all
sectors of the economy respond in the same way account for the business
cycle. Indeed, one might even suggest that this evidence points to a particular
disturbance, namely monetary policy, as a major source of the business cycle.
Monetary policy shocks arguably affect all sectors of the economy, while evidence of other aggregate shocks, namely aggregate productivity disturbances, is
quite weak. However, as shown here, it is by no means clear that all industries of
an economy will respond in the same way to an aggregate shock. Explaining the
comovement of industries then appears to be an important task for any theory of
the business cycle. Initial attempts to study this problem use natural extensions
of the growth model such as the inclusion of the input-output structure of the
economy (Hornstein and Praschnik 1997 or Horvath 1998) and limited sectoral
mobility of labor (Boldrin, Christiano, and Fisher 1999). Other explanations
consider the effect of various frictions in standard multisector growth models,
for example credit market imperfections as in Murphy, Shleifer, and Vishny
(1989). There has been some progress, but the problem clearly has not been
addressed successfully.

A. Hornstein: The Business Cycle and Industry Comovement

47

REFERENCES
Basu, Susanto, and John G. Fernald, 1999. “Why is Productivity Procyclical?
Why Do We Care?” International Finance Discussion Paper 638. Board of
Governors of the Federal Reserve System, 1999.
Benhabib, Jess, Richard Rogerson, and Randall Wright. “Homework in
Macroeconomics: Household Production and Aggregate Fluctuations,”
Journal of Political Economy, vol. 99, (December 1991), pp. 1166–87.
Boldrin, Michele, Lawrence J. Christiano, and Jonas D.M. Fisher. “Habit
Persistence, Asset Returns and the Business Cycle,” draft, Northwestern
University, 1999.
Burns, Arthur and Wesley C. Mitchell. Measuring Business Cycles, Studies in
Business Cycles No. 2. New York: NBER, 1946.
Christiano, Lawrence J. and Terry J. Fitzgerald. “The Business Cycle: It’s Still
a Puzzle,” Federal Reserve Bank of Chicago Economic Perspectives, vol.
22 (Fourth Quarter 1998), pp. 56–83.
Hornstein, Andreas. “Inventory Investment and the Business Cycle,” Federal
Reserve Bank of Richmond Economic Quarterly, vol. 84 (Spring 1998),
pp. 49–71.
Hornstein, Andreas and Jack Praschnik. “Intermediate Inputs and Sectoral
Comovement in the Business Cycle,” Journal of Monetary Economics,
vol. 40 (December 1997), pp. 573–95.
Horvath, Michael. “Sectoral Shocks and Aggregate Fluctuations,” Journal of
Monetary Economics, vol. 45 (February 2000), pp. 69–106.
Jorgenson, Dale W., Frank M. Gollop, and Barbara Fraumeni. Productivity
and U.S. Economic Growth, Harvard University Press: Cambridge, Mass.,
1987.
King, Robert G., Charles I. Plosser, and Sergio T. Rebelo. “Production,
Growth, and Business Cycles: I. The Basic Neoclassical Model,” Journal
of Monetary Economics, vol. 21 (March/May 1988), pp. 195–232.
Lucas, Robert E. “Understanding Business Cycles,” in Karl Brunner and
Allan H. Meltzer, eds., Stabilization of the Domestic and International
Economy, Carnegie-Rochester Conference Series on Public Policy, vol. 5,
Amsterdam: North-Holland, (1977).
Murphy, Kevin K., Andrei Shleifer, and Robert W. Vishny. “Building Blocks
of Market Clearing Business Cycle Models,” in Olivier J. Blanchard and
Stanley Fisher, eds., Macroeconomics Annual 1989. Cambridge, Mass.:
MIT Press, (1989), pp. 247–87.

48

Federal Reserve Bank of Richmond Economic Quarterly

Sato, Kazuo. “The Meaning and Measurement of the Real Value Added Index,”
Review of Economics and Statistics, vol. 58 (November 1976), pp. 434–42.
Solow, Robert. “Technical Change and the Aggregate Production Function,”
Review of Economics and Statistics, vol. 39 (1957), pp. 312–20.

Alternative Monetary
Policy Rules: A Comparison
with Historical Settings
for the United States, the
United Kingdom, and Japan
Bennett T. McCallum

R

ecent years have witnessed an upsurge of interest among monetary
policy analysts in the topic of simple and explicit rules for monetary
policy. In this recent work it is presumed that such rules would not be
followed literally and slavishly by central banks, but that they could be consulted for indicative purposes—perhaps by providing a starting point for policy
discussions. Tangible evidence of this interest appears in publications based
on two 1998 conferences, both entitled “Monetary Policy Rules,” sponsored
by the National Bureau of Economic Research (NBER) and by the Sveriges
Riksbank in collaboration with Stockholm University’s Institute for International Economic Studies (IIES).1 Most of the work in these papers is based
on some variant of the now-famous Taylor rule. Introduced in Taylor (1993),
this rule specifies settings of a nominal interest rate instrument in response to
observed or predicted values of inflation and the output gap (i.e., the percentage difference between output and its reference value2 ).3 Some of the studies

For comments, suggestions, or assistance I am indebted to Miguel Casares, Martin Gervais,
Marvin Goodfriend, Robert Hetzel, Yash Mehra, Allan Meltzer, Athanasios Orphanides, Edward Nelson, and Jeffrey Walker. The views expressed are the author’s and not necessarily
those of the Federal Reserve Bank of Richmond or the Federal Reserve System.
1 Proceedings

of the NBER conference have been published in Taylor (1999b); papers
from the Riksbank-IIES conference appear in the June 1999 issue of the Journal of Monetary
Economics.
2 This reference value is defined variously, in different studies, as the trend or capacity or
potential or natural-rate or market-clearing value of output.
3 In many of the studies, a lagged value of the interest rate is also included as a determinant
of the current value, thereby reflecting interest rate smoothing behavior.

Federal Reserve Bank of Richmond Economic Quarterly Volume 86/1 Winter 2000

49

50

Federal Reserve Bank of Richmond Economic Quarterly

consider alternative instrument or target variables,4 and very recently some
criticisms of the Taylor rule have been expressed by Orphanides (1998, 1999),
Meltzer (1999), and others. Accordingly, the purpose of the present paper
is to conduct counterfactual historical analysis of the type used by Stuart
(1996) and Taylor (1999a), and to compare and consider the messages provided by Taylor’s rule with others featuring alternative instrument and/or target
variables.
The type of analysis developed by Stuart (1996) and Taylor (1999a) consists of contrasting actual settings of instrument variables during some historical
time span with the values that would have been specified by particular rules in
response to prevailing conditions.5 Discrepancies or agreements between rulespecified and actual values can then be evaluated, in light of ex-post judgements
concerning macroeconomic performance during the span studied, to yield tentative conclusions concerning the merits of the various rules. Of particular interest
is whether major policy mistakes, judged ex-post, would have been prevented
by adherence to some of the candidate rules. Stuart (1996) conducted such
comparisons for Taylor’s rule and also one promoted by McCallum (1987,
1993) that features a monetary base instrument and a nominal-income growth
target.6 The sample period utilized by Stuart was 1985.1 through 1996.2 for
the United States and the United Kingdom. This article will examine the years
spanning the early 1960s through 1998.4 for those two countries and the 1970s
through 1998.4 for Japan. The investigation will also extend the range of rules
considered by combining interest rate and monetary base instruments with both
Taylor-type and nominal-income target variables.
It should be said that no suggestion is intended to the effect that historical analysis of the Stuart-Taylor type represents the only useful approach to
policy-rule evaluation. Most of my work, in fact, has involved simulations with
quantitative structural macroeconomic models (e.g., McCallum, 1988, 1993;
McCallum and Nelson, 1999a, b). The premise is merely that the Stuart-Taylor
type of study can also be useful, in addition to simulations with structural models. In this regard, it is important to be clear about the nature of the exercise
involved, i.e., to appreciate its limitations and strengths—both of which are
considerable. Accordingly, these will be reviewed in Section 2, immediately
following the paper’s first application of the Stuart-Taylor procedure.
The article proceeds as follows. In Section 1 the alternative rules are specified, notation is established, and some general issues are discussed. Applications
to the United States, the United Kingdom, and Japan are then conducted in
Sections 2– 4. Issues concerning the specification of target variables are taken
4 See,

for example, McCallum and Nelson (1999a, 1999b).
examples of such analysis include Judd and Rudebusch (1998) and Kozicki (1999).
6 In its growth-rate version, considered exclusively here and by Stuart (1996), McCallum’s
rule is similar (though not identical) to one promoted by Meltzer (1987).
5 Other

B. T. McCallum: Alternative Monetary Policy Rules

51

up in Section 5 and others related to instrument variables in Section 6. Section
7 presents a brief conclusion.

1.

SPECIFICATION OF RULES

The well-known Taylor rule can be expressed as follows:
Rt = r + ∆pa + 0.5(∆pa − π ∗ ) + 0.5˜ t .
¯
y
t
t

(1)

Here Rt is the short-term nominal interest rate that the central bank in question
uses as its instrument or “operating target,” i.e., the interest rate over which it
exerts control at a daily or weekly frequency. Next, r is the long-run average
¯
real rate of interest, ∆pa is an average of recent inflation rates (or a forecast
t
˜
value), and π ∗ is the central bank’s target inflation rate. Finally, yt is a measure
of the output gap, the percentage difference between actual and capacity output
values. In Taylor’s original application (1993), the values r = 2 and π ∗ = 2
¯
were specified, expressing the belief that 2 percent per annum is an approximation to the long-run average real rate of interest in the United States, and that 2
percent per annum is a reasonable specification for the Federal Reserve’s target
inflation rate.7 Also, in Taylor (1993) the measure used for ∆pa is the avert
age of GDP deflator inflation rates over the past four quarters, while capacity
output is represented by a linear trend for the log of real GDP fit to quarterly
observations for the years 1985–1992. In Taylor (1999a), the Hodrick-Prescott
(1997) filter is used instead to generate residuals from “trend” that are taken
to represent yt . The rule suggests, of course, that monetary policy should be
˜
tightened (by an increase in Rt ) when inflation exceeds its target value and/or
output exceeds capacity.
Subsequent applications of the Taylor rule have modified or extended formula (1) in several ways. Some have used proxies for expected future inflation
˜
˜
in place of ∆pa while others have done something similar for yt or used yt−1
t
instead. A common and major change is to include Rt−1 on the right-hand
side as a determinant of Rt ; this adjustment is intended to reflect the practice
of interest rate smoothing, which is widely believed to be prevalent in the
behavior of many central banks.
An important line of investigation has been pioneered by Orphanides (1998,
1999), who has attempted to base rule calculations on values of ∆pt (inflation)
and yt that were actually available to central bank policymakers at the time
˜
that historical instrument settings were chosen. Orphanides (1998) recognizes
that current-period values for yt could not be known until after the end of
˜
7 It

is not necessary that constants be used for these values, but they are in Taylor (1993)
and for additional postwar periods in Taylor (1999a).

52

Federal Reserve Bank of Richmond Economic Quarterly

period t8 and also emphasizes that macroeconomic data is often substantially
revised after its initial reporting. In Orphanides (1999) it is argued that these
problems are so severe that adherence to the Taylor rule would not have prevented the inflation of the 1970s, as claimed by Taylor (1999a). Partly for
reasons to be mentioned below in Section 5 and partly because of the difficulty
of doing otherwise, the present study will be based on data available in June
1999, not on real time data of the type recommended by Orphanides.
The rule proposed by McCallum (1987, 1988, 1993) can be expressed as
follows:
∆bt = ∆x∗ − ∆va + 0.5(∆x∗ − ∆xt−1 ).
t

(2)

Here ∆bt is the change in the log of the adjusted monetary base, i.e., the
growth rate of the base between periods t − 1 and t. The term ∆x∗ is a target
growth rate for nominal GDP, ∆xt being the change in the log of nominal
GDP. This target value ∆x∗ is specified as π ∗ + ∆y∗ , where ∆y∗ is the longrun average rate of growth of real GDP. The second term on the right-hand
side of (2), ∆va , is the average growth of base velocity over the previous 16
t
quarters, vt = xt − bt being the log of base velocity. This term is intended to
reflect long-lasting changes in the demand for the monetary base that occur
because of technological developments or regulatory changes (presumed to be
permanent); it is not intended to reflect cyclical conditions. These conditions
are responded to by the final term, which prescribes that base growth is adjusted
upward (i.e., policy is loosened) when ∆xt−1 falls short of ∆x∗ . In McCallum
(1988, 1993), values other than 0.5 are considered for the coefficient attached
to ∆x∗ − ∆xt−1 , and variants of (2) that respond to discrepancies of the level
type, rather than the growth rate type, are investigated. Here, however, we shall
limit our attention to the particular formulation given in (2).
A bit of discussion needs to be given to the topic of units of measurement.
In previous studies by McCallum, growth rate variables such as ∆xt have
been measured as changes in logs. Therefore such variables reflect quarterly
changes, instead of annualized changes, and are presented in fractional rather
than percentage units. Accordingly, such variables need to be multiplied by
400 to be commensurate with similar variables as measured by Taylor and
in most papers on policy rules. Similar comments pertain as well to interest
rate measures. To maintain consistency among the different rules considered,
we shall here report all results as annualized percentages, rather than in the
quarterly fractional units previously used in the work of McCallum.
Another detail of rule specification concerns timing. In (2), both of the
variables on the right-hand side are based on variables realized in period t − 1
or earlier; i.e., current-period values are not utilized. The reason, as suggested in
8 This

type of operationality issue has been emphasized by McCallum and Nelson (1999b)
and McCallum (1999).

B. T. McCallum: Alternative Monetary Policy Rules

53

footnote 6, is to make the rule specification realistically operational. In Taylor’s
studies, the inflation variable ∆pa is typically measured as referring only to
t
previous-period values, but it is assumed that yt pertains to period t. Since
˜
current-quarter values of real GDP cannot be observed until well into the next
quarter, in the present study yt will be measured as the value of the output gap
˜
variable (however measured) pertaining to the previous quarter.
Clearly, the Taylor and McCallum rules differ in regard to both instrument
and target variables.9 It is not obvious, however, why these should be paired in
any particular combination. It would be quite natural, that is, to consider a rule
with an interest rate instrument and a nominal income growth target. Similarly,
it would be reasonable to consider a rule with a base growth instrument and a
Taylor-style target specification. Therefore, the investigation that follows will
also consider, in addition to (1) and (2), rules of the form
¯
Rt = r + ∆pa − 0.5(∆x∗ − ∆xt−1 )
t

(3)

∆bt = ∆x∗ − ∆va − 0.5ht ,
t

(4)

and

where we define the “hybrid” target variable ht = (∆pa − π ∗ + yt ).10 Thus rule
˜
t
(4) features responses to the same macroeconomic conditions as in Taylor’s
rule (1) but with a base instrument. Examination of the results involving (1)–
(4) should then enable one to determine whether differences in policy advice
offered by (1) and (2) are due primarily to their different instruments or targets.

2.

UNITED STATES

We begin with the case of the United States. For xt , yt , and pt we use the
logarithms of nominal GDP, real (chain-linked) GDP, and their ratio. The monetary base is the series computed by the St. Louis Fed, which incorporates
adjustments for changes in reserve requirements. In addition, an adjustment for
sweep accounts has been made for 1994–1998.11 Finally, Rt is the federal funds
rate averaged over the quarter. All variables except Rt are seasonally adjusted.
The series are taken from the FRED data base of the Federal Reserve Bank
of St. Louis. In what follows, ∆pa = 0.25(∆pt−1 + ∆pt−2 + ∆pt−3 + ∆pt−4 )
t
9 Here I am using the term “target variable” to mean a variable that the policy rule responds to
in a manner designed to reduce its deviations from some reference path. Svensson (1999) objects
to this usage, preferring to reserve the word “target” for variables appearing in loss functions. For
a brief discussion see McCallum and Nelson (1999a).
10 The term “hybrid” was used for this variable by Hall and Mankiw (1994).
11 Specifically, 0.10 times the cumulative total of sweeps of transaction deposits into MMDAs, reported by FRED, are added to the adjusted base series. Here 0.10 represents the marginal
reserve requirement ratio.

54

Federal Reserve Bank of Richmond Economic Quarterly

while for yt we report the percentage excess (in period t − 1) of output over
˜
a “trend” reference value provided by the Hodrick-Prescott (HP) filter, as in
Taylor (1999a). The effect of the latter choice will be discussed below, in
Section 5.
Figure 1 plots values of Rt implied by the Taylor rule (1) with π ∗ = 2
and r = 2 together with actual values over the period 1960–1998.12 From this
¯
figure, it can be seen that the actual interest rate was lower than the rule-implied
value throughout the 1970s, indicating that monetary policy was too loose,
according to the rule. Beginning in 1981, policy was too tight until 1987, the
first year considered in Taylor’s original study (1993). Over 1987–1995 policy
was about right, according to the figure, but since 1996, it has been somewhat
too tight.
As mentioned above, it is important to recognize the limitations and virtues
of the type of comparison provided by Figure 1. If in a particular period the
actual value of Rt was lower than the rule-specified value, then the rule’s indication is that policy was too loose in that period, given the prevailing conditions.
There is no suggestion that the actual setting of Rt in that period was too low
unconditionally. Indeed, the presumption of Stuart (1996) and Taylor (1999a),
which we adopt here, is that prevailing inflation would have been lower during
the 1970s if Taylor’s rule had been followed in practice. So the Rt settings that
would have been appropriate, according to rule (1), would have been lower
than those indicated by the solid line in Figure 1.13 Thus the solid line in this
figure does not pretend to represent an optimal or even desirable path for Rt
over the period. But that does not prevent the comparison of the two lines from
indicating that, conditional upon prevailing conditions, actual Rt values were
set lower than the rule would have specified in virtually every period during
the ’70s. From the standpoint of rule (1), therefore, monetary policy was too
loose during the ’70s. That is the only type of conclusion provided by Figure
1, and other such plots presented below.
Thus the principal weakness of this type of comparison is that it does not
indicate what “optimal” policy settings would have been or even what time
path crucial variables would have followed under the rule being examined.
But there is an offsetting virtue. Any designation of optimality—indeed, any
specification of how Rt or other variables would have evolved historically under
any specified policy rule—is necessarily dependent upon the specific model of
˜
the economy used to predict how ∆pt and yt would have responded to Rt
settings. The Stuart-Taylor procedure, by contrast, does not require adoption
of any specific model. This feature is definitely advantageous, because there is
12 Since our data base is for 1960.1–1998.4, rule-implied values begin with 1961.2 because
lagged values are needed to determine ∆pa .
t
13 If ∆pa had been lower in each period, the R values prescribed by (1) would have been
t
t
lower.

B. T. McCallum: Alternative Monetary Policy Rules

55

Figure 1 U.S. Interest Rates, Actual and Implied by Rule (1)

no professional agreement concerning the proper specification of the “correct”
model of the economy.14
We now return to the main line of analysis and move on to rule (2). For its
application, we take ∆x∗ = 5, combining a 2 percent inflation target with an
assumed long-run average output growth rate of 3 percent per year.15 The comparison of base growth values implied by rule (2) with actual historical values
is presented in Figure 2.16 There it will be seen that policy was too loose—
actual base growth was greater than specified by the rule—during the second
half of the ’60s and much too loose throughout the ’70s. This discrepancy was
gradually reduced between 1981 and 1987. Then policy became slightly too
loose during 1990–1992 and too tight during 1994–1995, according to the rule.
Since 1995 it has been about right, on average, although the final observation
of 1998 suggests slightly excessive base growth at that date.
14 For an elaboration on this last point, see McCallum (1999, pp. 1490–1). As mentioned in
the introduction, the purpose of the present digression is not to object to counter-factual simulation studies, based on specific models, but only to argue that different procedures have different
strengths and weaknesses.
15 The value of 3 percent for output growth was used in McCallum (1987, 1988)—together
with an inflation target of 0 percent—and in subsequent studies. The actual average over 1960–
1998 was 2.97 percent.
16 Rule values begin with 1964.2 because of the lags needed to calculate va .
t

56

Federal Reserve Bank of Richmond Economic Quarterly

Figure 2 U.S. Base Growth, Actual and Rule (2)

Figure 3 gives results for rule (4), which combines the base instrument (as
in (2)) with the hybrid target variable (as in (1)). Somewhat surprisingly, the
overall characterization of the results can be described with the same words
used for the base rule (2). The main difference is that the rule-indicated path
of ∆bt has less quarter-to-quarter variability than in Figure 2. The reason,
evidently, is that (4) does not respond to quarter-to-quarter movements in the
growth rate of output (real GDP), which are quite volatile. The basic message
provided by the hybrid target variable is much the same as that provided by
nominal GDP growth because the HP filter yields quite small values for the
output gap, as will be illustrated below. Therefore, ht and ∆xt behave alike
except for the volatility introduced into the latter by the yt component.
Results with rule (3), featuring the interest rate instrument with a nominal
GDP growth target variable, are shown in Figure 4. Here the broad overall
signals are much like those of Figure 1, which features the Taylor rule, except
with a more erratic path because of the output growth component of the target
variable. So the comparison among the four figures suggests that the choice
of an instrument variable matters more for the trend of monetary policy than
the choice of a target variable. It should be emphasized, however, that this
preliminary conclusion pertains only to the nominal GDP growth and hybrid
variables, with the output gap component of the latter determined by the HP
filter method.

B. T. McCallum: Alternative Monetary Policy Rules
Figure 3 U.S. Base Growth, Actual and Rule (4)

Figure 4 U.S. Interest Rate, Actual and Rule (3)

57

58

3.

Federal Reserve Bank of Richmond Economic Quarterly

UNITED KINGDOM

Before delving more deeply into the comparisons among the rules, let us look
at the basic cases using data for the United Kingdom. Again xt , yt , and pt are
logs of nominal GDP, real GDP, and their ratio. For the monetary base bt we
use the Bank of England’s M0 measure, seasonally adjusted, which requires no
adjustments for reserve requirements because the latter are small enough to be
negligible.17 The interest rate used is a one-month Treasury-bill rate, averaged
over the quarter. For the United Kingdom, we use a value of 2.25 for r, 2
¯
for π ∗ , and 2.25 for ∆y∗ . Thus ∆x∗ = 4.25. The output gap measure is the
percentage departure of real GDP from trend, obtained from the residuals from
a regression of the log of real GDP on a linear trend fitted over the years 1960–
1998.
Results using the Taylor rule (1) are shown in Figure 5. The indication
there is that monetary policy was much too loose during the 1970s, with the
rule calling for an interest rate of 38 percent in 1975.3, as compared with an
actual value of 10.4.18 From 1983 through 1987 policy was slightly too tight,
according to the rule, and since 1987 it has been just about right, except perhaps
in 1994.
The McCallum rule (2) presents a somewhat different story, as can be seen
in Figure 6. It agrees that policy was much too loose during the ’70s, but
suggests that it stayed too loose most of the time until 1990 (when the U.K.
entered the European Union’s exchange rate mechanism in October, dropping
out in September 1992). Since 1992, policy was slightly loose, according to
Figure 6, until 1997 when it became just about right. The main difference in
the messages of Figures 5 and 6 is that the latter suggests that policy was too
loose during the mid–1980s. Ex-post, this suggestion seems correct, as U.K.
inflation rose to excessive heights prior to 1990—probably as a consequence
of the episode of “shadowing the D-mark” that occurred during 1986–1988.
As in the case of the United States, the messages from rules (3) and (4)
tend to agree when the instrument, not the target variable, is the same. Thus
in Figure 7 we have base growth figures implied by rule (4), with the hybrid
target variable, and the policy messages are much the same as in Figure 6,
but with less quarter-to-quarter variability of the indicated ∆bt values. Also, in
Figure 8, plotted for an interest instrument and a ∆xt target, we find substantial
agreement with the indications of Figure 5, which pertains to the Taylor rule.
Agreement is incomplete, however, since this rule does not call for looser policy
in the mid–1980s.

17 Data for M0 are published by the Bank of England for 1969.3–1998.4. Earlier values were
obtained from Capie and Webber (1985) and spliced on.
18 Of course if rule (1) had been followed throughout, actual inflation would probably have
been much less severe and the values of Rt indicated by the rule would have been much lower.

B. T. McCallum: Alternative Monetary Policy Rules
Figure 5 U.K. Interest Rate, Actual and Rule (1)

Figure 6 U.K. Base Growth, Actual and Rule (2)

59

60

Federal Reserve Bank of Richmond Economic Quarterly

Figure 7 U.K. Base Growth, Actual and Rule (4)

Figure 8 U.K. Interest Rate, Actual and Rule (3)

B. T. McCallum: Alternative Monetary Policy Rules

4.

61

JAPAN

In the case of Japan, our rules will be applied only to the period 1972.1–
1998.4, rather than a time span beginning in the early ’60s. The reasons are
that Japanese data for constructing a monetary base series does not exist prior
to 1963; that Japan kept a fixed exchange rate with the U.S. dollar prior to
1971; and that there was a marked break in the growth rate of Japanese real
¯
GDP around 1971 or 1972.19 For the subsequent period we use r = 3, a higher
value than that for the United States or the United Kingdom, because real
output growth was higher in Japan. Nevertheless, for ∆x∗ we adopt a value of
5, corresponding to an average long-run real output growth rate of 3 percent
and, again, a target inflation rate of π ∗ = 2 percent. In measuring the output
gap yt we cannot use either the HP filter or a linear trend because output in
˜
1998 was quite far below capacity, in the judgement of most observers. Instead,
we have measured the fractional gap over 1972.1–1992.2 as the residual from
a regression of the log of real GDP on a linear trend (fitted to 1972.1–1992.4
observations), and have assumed that trend or capacity output grew at a rate
of 2.5 percent per annum since 1992.2. This procedure yields a gap that grows
to a figure of 11.2 percent for 1998.4.
The Bank of Japan now publishes four monthly data series on the monetary base, beginning in 1970, with and without adjustments for seasonality and
reserve requirement changes. The monthly series with both adjustments was
averaged to generate values for 1970.1–1998 and data from McCallum (1993)
was spliced on to cover 1963.1–1969.4. (Values prior to 1967.4 were not used
in the study, however.) For Rt the overnight call rate (uncollateralized) was
used and official GDP statistics provided the basis for the remaining variables.
Application of Taylor’s rule (1) to Japan for 1972.1–1998.4 is depicted in
Figure 9. There the indications are that policy should have been much tighter
during 1973–1974 and somewhat tighter over 1975–1978. Policy was slightly
too tight most of the time over 1982–1987, and then about right until 1994.
Since then it has been too tight most of the time, but not in 1997. At the end
of 1998, the call rate was almost 4 percentage points too high, the Taylor ruleindicated value being −3.6 percent. Of course, the latter value is not feasible,
but it indicates that the rule calls for much more stimulative policy than actually
prevailed in late 1998.
This last message is also provided by the base rule (2), as shown in Figure
10, but to an even greater extent. Indeed, this rule suggests that monetary policy
has been too tight most of the time since the middle of 1990. Like (1), it points
to a too-loose stance over 1972–1978. Interestingly, in light of the “asset-price
bubble” of the late 1980s, Figure 10 indicates that monetary policy was slightly
too loose during 1986–1988.
19 For

these reasons McCallum (1993) begins its rule study with the quarter 1972.1.

62

Federal Reserve Bank of Richmond Economic Quarterly

Figure 9 Japan Interest Rate, Actual and Rule (1)

Figure 10 Japan Base Growth, Actual and Rule (2)

B. T. McCallum: Alternative Monetary Policy Rules
Figure 11 Japan Base Growth, Actual and Rule (4)

Figure 12 Japan Interest Rate, Actual and Rule (3)

63

64

Federal Reserve Bank of Richmond Economic Quarterly

We now turn to rules (3) and (4). Figure 11 shows that in the case of Japan,
rule (4) again gives much the same signals as does the other rule with the ∆bt
instrument, rule (2). Also, as in previous cases, the hybrid target variable yields
a smoother path for base growth than does (2). As for the interest rate rule with
a nominal GDP growth target, rule (3), Figure 12’s results are more similar to
those in Figure 9 than in Figure 10. The extent to which rule (3) calls for added
stimulus in recent years is even less than in Figure 9, however. The rule does
call for easier policy in the last half of 1998, but finds policy about right during
1995–1997.

5.

ISSUES CONCERNING TARGET VARIABLES

One of the main preliminary indications of our previous discussion is that rules
with ∆xt and ht target variables give rather similar policy signals, provided that
the instrument variable is the same. This notion needs to be strongly qualified,
however, as follows. The main point is that the similarity of ∆xt and ht signals
observed in Sections 2– 4 depends upon the use of output gap measures that do
not yield large numerical magnitudes over the time span studied. In the case
of the United States, the measure used was based on residuals from the HP
filter. The standard deviation of these values over 1960–1998 is only 1.63, in
percentage points. If instead the output gap measure was based on residuals
from a linear trend (for the log of real GDP), the standard deviation would be
4.15 and the impact of the gap measure would be significantly greater. In that
case the Taylor rule vs. actual comparison, comparable to Figure 1, would be
as shown in Figure 13. Here the monetary policy message is not drastically
different from that of Figure 1 for the subperiod 1966–1990, although the need
for tighter policy during the ’70s would be more clearly indicated. But for the
early ’60s and the late ’90s the message would be quite different, with lower
interest rates indicated by the gap based on the log-linear detrending. According
to the Figure 13 version of the Taylor rule, the federal funds rate was too high
by about 300 basis points throughout 1995–1998!
It is my belief that reliance of a policy rule upon any output gap measure is
risky, for different measures give quite different values and there is at present no
professional consensus on an appropriate measure—or even a concept. Linear
detrending depends rather sensitively on the time period selected for fitting of
the trend, as is illustrated in Figure 14, where gap measures based on loglinear trends fitted over 1960–1998 and 1980–1998 are shown, together with
values based on the HP filter. One might suggest that quadratic detrending could
alleviate this problem, but quadratic trends are themselves rather sensitive to
the time period selected for fitting. This claim is supported by Figure 15, which
shows gap measures based on quadratic trends for the log of real U.S. GDP
fitted over the time periods 1960.1–1998.4 and 1980.1–1998.4. As can be seen
there, these measures often differ by as much as 3 percentage points.

B. T. McCallum: Alternative Monetary Policy Rules
Figure 13 U.S. Rule (1) with Output Gaps Based on
Log-Linear Detrending, 1960–1998

Figure 14 U.S. Output Gaps Based on HP Filter and
Log-Linear Detrending, 1960–1998 and 1980–1998

65

66

Federal Reserve Bank of Richmond Economic Quarterly

With respect to the HP filter, the problem is that this procedure produces
a “trend” that is so flexible that it follows the time path of actual GDP rather
closely. Consequently, measures of the output gap are generated that would
appear to underestimate (in absolute terms) the economically relevant gap
values.20 To illustrate this point, Figure 16 shows how the HP filter handles
U.S. observations on real GDP during the ’20s and ’30s. According to this
figure, U.S. output had fully returned to “trend” by 1934 and the incidence of
above-trend output was approximately the same as below-trend output during
the ’30s, suggesting that the Great Depression actually did not occur.
More fundamentally, McCallum and Nelson (1999b) argue that any gap
measure based on an output detrending procedure, which excludes the effects
of current shocks from the measured gap, is conceptually inappropriate. The
point is that (e.g.) positive technology shocks serve to increase the capacity
or natural-rate value of output, not the value of actual output relative to the
latter; but many univariate detrending procedures presume just the opposite.
To overcome this difficulty, McCallum and Nelson (1999b) propose a measure
based on the assumption of a Cobb-Douglas production function and utilizing
values of manhours employed per member of the civilian workforce. This measure treats technological change appropriately, at least arguably, but relies upon
debatable assumptions about labor supply and does not have a well-defined zero
value.
As mentioned above, the recent work of Orphanides (1998, 1999) has
attracted considerable attention. In the earlier of the cited papers, Orphanides
constructed data series for 1987–1992 reflecting values of macroeconomic variables that were actually available at the time of (FOMC) policy decisions in
the past. These series do not, accordingly, reflect data revisions and measurements that have taken place after the FOMC meetings at which instrument
settings (usually values of the federal funds rate) were actually decided. In this
context, the measurement of “potential” or “natural-rate” output is especially
problematic. This study indicated that the magnitude of the informational problems were serious enough that “real-time policy recommendations differ widely
from those obtained with the revised published data employed [by researchers]
later on” (Orphanides, 1998, p. 3). The broad overall policy messages offered
by the Taylor rule for 1987–1992 are not overturned, however, by the results
of the 1997 study.
The results in Orphanides (1999) are more drastic. In this later work, the
time span studied goes back to 1966.1 and so includes the major inflationary
buildup and continuation that Taylor (1999a) refers to as “The Great Inflation.”
Orphanides’ dramatic conclusion is that adherence to Taylor’s rule throughout

20 The

present discussion presumes adoption of the standard value of 1600 for the HP filter’s
smoothing coefficient in work with quarterly data.

B. T. McCallum: Alternative Monetary Policy Rules
Figure 15 U.S. Gap Measures Based on Log-Quadratic Detrending,
1960–1998 and 1980–1998

Figure 16 U.S. Real GNP and HP Trend, 1921–1939

+

67

68

Federal Reserve Bank of Richmond Economic Quarterly

the period would not, in constrast to Taylor’s (1999a, pp. 338–39) contention,
have prevented the Great Inflation.
Developing an appropriate evaluation of Orphanides’ (1999) analysis is not
a trivial undertaking. Certainly Orphanides’ reconstruction of the data represents a major contribution to economic policy analysis; moreover, his arguments
are carefully constructed and exposited. In my judgement, they do not imply
that simple monetary policy rules cannot be useful when constrained by realtime data availability. Instead, I believe, the (1999) study shows in a striking
fashion that reliance on an inappropriate concept of potential (or reference)
output can be ruinous, i.e., can result in a monetary policy rule that is counterproductive. But some rules do not rely upon such measures, as the examples
of Sections 2 through 4 above illustrate.21 Also, the truly dramatic results in
Orphanides (1999) stem from a potential output concept revision, rather than
a data revision. Thus, although the data-revision problem is not of negligible
significance, it is not as profound as a quick reading of Orphanides (1999)
might suggest. In my view, it seems satisfactory to abstract from that problem
for the purposes of the present study.
Given the foregoing argument and the findings of Sections 2– 4, a natural
step would be to investigate the performance of rules that use inflation as
the target variable, i.e., the cyclical variable responded to by the instrument.
Accordingly, we now present figures based on the two policy rules
¯
Rt = r + ∆pa + 0.5(∆pa − π ∗ )
t
t

(5)

∆bt = ∆x∗ − ∆va − 0.5(∆pa − π ∗ ),
t
t

(6)

which can be compared with (1) and (4). The results are shown in Figures
17-22. For the United States, the interest rate rule in Figure 17 calls for Rt
values quite close to those of Figure 1. Also, the base growth rule in Figure
18 yields settings for ∆bt that are rather close to those shown in Figure 3,
in which the rule responds to the hybrid target variable. Likewise, the plots in
Figures 19 and 20 for the United Kingdom are fairly similar to those in Figures
5 and 7. For Japan, however, the policy advice for recent years provided by
the inflation-target rule in Figure 21 is quite different than that in Figure 9. In
particular, in the absence of an output gap signal, rule (5) calls for Rt settings
somewhat higher than actual values during 1997 and 1998. The base rule (6)
results in Figure 22 remain more stimulative than the actual record for recent
years, but to a lesser extent than in Figure 11.
21 This conclusion is basically consistent with Orphanides’ warning against “activist” policy
rules, by which he means rules that place emphasis on measures of the level of an output gap
concept. Orphanides finds that a rule featuring “natural growth targeting,” which is rather similar
to nominal income growth targeting as in rules (2) or (3) above, is not strongly subject to the
difficulties that he emphasizes.

B. T. McCallum: Alternative Monetary Policy Rules
Figure 17 U.S. Interest Rate, Actual and Rule (5)

Figure 18 U.S. Base Growth, Actual and Rule (6)

69

70

Federal Reserve Bank of Richmond Economic Quarterly

Figure 19 U.K. Interest Rate, Actual and Rule (5)

Figure 20 U.K. Base Growth, Actual and Rule (6)

B. T. McCallum: Alternative Monetary Policy Rules
Figure 21 Japan Interest Rate, Actual and Rule (5)

Figure 22 Japan Base Growth, Actual and Rule (6)

71

72

Federal Reserve Bank of Richmond Economic Quarterly

One attractive aspect of the inflation target variable ∆pa relative to the
t
nominal income variable ∆xt−1 is that the former features smaller quarter-toquarter movements and therefore imparts a smoother, less choppy path to the
instrument variables in (5) and (6), in comparison to (2) and (3). One reason
for this, certainly, is that ∆pa reflects four-quarter averaging while ∆xt−1 does
t
not. Accordingly, it should be of some interest to see how the nominal income
variable would perform if averaged over periods t − 1 through t − 4. Results
with that modification are presented in Figures 23-28. In these plots, it can be
seen that the choppiness of rules with a nominal GDP growth target is reduced
substantially, although the implied instrument settings remain slightly more
variable than with the inflation target. Are there any compensating advantages
of the averaged ∆x values relative to the averaged ∆p values? For the United
States and the United Kingdom, the policy advice seems to be basically the
same in Figures 23–26 as in Figures 17–20. In the case of Japan, however,
the nominal income targets in Figures 27–28 give more stimulative signals
than with inflation targeting (Figures 21–22), which seems desirable. But the
magnitude is not very large.

6.

ISSUES CONCERNING INSTRUMENT VARIABLES

One of the more surprising aspects of the results in Sections 2– 4 is that the
policy diagnoses provided by the various rules seem to be more dependent
upon the instrument variable used than upon the choice of target variable.
This indication seems inconsistent with most analysts’ beliefs about monetary
policy design. Reflection upon the role and nature of the rules makes this
finding understandable, however, in the following manner. First, the way in
which the rules are used in a study such as the present one implies that the
rule-specified instrument settings are actually serving as magnitudes of indicator variables, not instruments. That is, one could view the resulting values
for quarterly settings of Rt or ∆bt as intermediate targets to be obtained by
day-to-day or week-to-week manipulation of other variables that serve as the
central bank’s instrument.22 Second, the policy stance—i.e., degree of tightness
or ease—represented by rule-specified settings of Rt or ∆bt depends upon the
magnitude of those variables relative to some reference value that can vary
from period to period. In the case of the Taylor rule (1) the reference value
is r + ∆pa , which serves to convert Rt movements into movements in a real
¯
t
r
¯
interest rate measured relative to r, since Rt − (¯ + ∆pa ) = (Rt − ∆pa ) − r. With
¯
t
t
the McCallum rule (2), the reference value for ∆bt is ∆x∗ − ∆va . In this case,
t
22 A

study that proceeds in this fashion is McCallum (1995), which considers how the U.S.
federal funds rate could be manipulated on a week-to-week basis to hit quarterly intermediate
targets for monetary base growth with the latter set so as to keep nominal income growth close
to a specified target value.

B. T. McCallum: Alternative Monetary Policy Rules
Figure 23 U.S. Interest Rate, Actual and Rule with
Averaged Nominal Income Growth

Figure 24 U.S. Base Growth, Actual and Rule with
Averaged Nominal Income Growth

73

74

Federal Reserve Bank of Richmond Economic Quarterly

Figure 25 U.K. Interest Rate, Actual and Rule with
Averaged Nominal Income Growth

Figure 26 U.K. Base Growth, Actual and Rule with
Averaged Nominal Income Growth

B. T. McCallum: Alternative Monetary Policy Rules
Figure 27 Japan Interest Rate, Actual and Rule with
Averaged Nominal Income Growth

Figure 28 Japan Base Growth, Actual and Rule with
Averaged Nominal Income Growth

75

76

Federal Reserve Bank of Richmond Economic Quarterly

∆bt − (∆x∗ − ∆va ) reflects the difference between ∆bt and the value of base
t
growth that would yield an inflation rate of π ∗ if output growth were equal to
its long-run average value and base velocity growth were equal to its average
over the past 16 quarters (a value that is implicitly being used as a forecast of
the average over the indefinite future).
In each case, in other words, a necessary reference variable must be specified to convert raw values of Rt or ∆bt into measures of monetary ease or
tightness.23 Therefore the precise specification of these reference variables is
of considerable importance to a rule’s performance. If rules are to be relatively simple, it is necessary that the specification of these reference values be
simple—hence Taylor’s specification of a constant “equilibrium” real rate of
interest or McCallum’s constant “long-run growth rate of output.” Evidently,
however, the properties of any rule will depend critically upon how these reference values are specified. Consequently, it would appear that future research
should perhaps devote more attention to this aspect of policy rule specification.
To date, researchers have instead directed most of their attention to the choice
among target variables, details of their specification, and the magnitude of
coefficients attached to them.

7.

CONCLUSION

Let us close with a brief summary of the findings developed above, based on
historical policy-rule studies for the United States, the United Kingdom, and
Japan. The basic results in this study come from comparisons of actual values
with rule specifications involving either interest-rate or monetary-base instrument settings and nominal GDP growth or Taylor-style hybrid (inflation plus
output gap) target variables. For the United States, all of the rules considered
would have called for tighter monetary policy during the ’70s, although the
base-instrument rules would have done so more strongly than those with the
Fed’s actual funds-rate instrument. There is some disagreement among the rules
concerning the ’80s and ’90s, although all of the candidate rules indicate that
policy has not been highly inappropriate since 1987. For the United Kingdom,
the various rules agree regarding the excessive inflation of the ’70s, but the
base-instrument rules suggest that policy was too loose during the middle and
late ’80s whereas the interest-instrument rules do not. In the case of Japan,
interest centers on the record since 1990. Most of the rules indicate that policy
was too tight in 1998, but the base rules suggest excessive tightness for the
entire period 1990–1998, while the interest rate rules do not. All in all, the
recommendations provided by the base rules seem somewhat more appropriate
from an ex-post perspective.

23 This

statement applies to all of the rules, of course, not just (1) and (2).

B. T. McCallum: Alternative Monetary Policy Rules

77

Some of the study’s suggestions are methodological, rather than substantive. In particular it is argued that reliance on output gap measures is risky,
because various measures of potential or natural-rate output levels differ widely
and there is no professional consensus on the most appropriate measure or even
concept to be used. Most univariate detrending procedures, which are frequently
utilized, would seem to be fundamentally inappropriate, because they assign
the effects of technology shocks primarily to the gap between output and its
reference value, rather than to the latter variable itself. Omitting the output gap
term from a rule with the hybrid target converts it into an inflation targeting
rule; we show that such rules give good advice in most of the episodes. So,
too, do nominal income growth rules that average recent values.
A somewhat surprising finding is that rules’ messages are evidently more
dependent upon which instrument rather than which target variable is used.24
This finding can be understood as resulting from the necessity of specifying a
reference value, relative to which instrument settings are implicitly compared,
in representing policy tightness or ease. For rules to be sufficiently simple,
these reference-value specifications must themselves be simple, but different
implicit assumptions about macroeconomic behavior are thereby built into the
rule. The paper suggests, consequently, that investigation of these implicit
assumptions could be an important topic for future research on alternative
monetary policy rules.

REFERENCES
Capie, Forrest, and Alan Weber. A Monetary History of the United Kingdom,
1870–1982, Vol. 1. London: George Allen and Unwin, 1985.
Hall, Robert E., and N. Gregory Mankiw. “Nominal Income Targeting,” in N.
Gregory Mankiw, ed., Monetary Policy. Chicago: University of Chicago
Press, 1994.
Hodrick, Robert J., and Edward C. Prescott. “Postwar U.S. Business Cycles:
An Empirical Investigation,” Journal of Money, Credit, and Banking, vol.
29 (February 1997), pp. 1–16.
Judd, John P., and Glenn D. Rudebusch. “Taylor’s Rule and the Fed: 1970–
1997,” Federal Reserve Bank of San Francisco Economic Review (No. 3,
1998), pp. 3–16.
Kozicki, Sharon. “How Useful Are Taylor Rules for Monetary Policy?” Federal
Reserve Bank of Kansas City Economic Review, vol. 84 (Second Quarter,
1999), pp. 5–33.
24 Provided

that strong dependence upon an output gap measure is avoided.

78

Federal Reserve Bank of Richmond Economic Quarterly

McCallum, Bennett T. “Issues in the Design of Monetary Policy Rules,” in John
B. Taylor and Michael Woodford, eds., Handbook of Macroeconomics.
Amsterdam: North-Holland Publishing Co., 1999.
. “Monetary Policy Rules and Financial Stability,” in Kuniho
Sawamoto, Zenta Nakajima, and Hiroo Taguchi, eds., Financial Stability
in a Changing Environment. New York: St. Martin’s Press, 1995.
. “Specification and Analysis of a Monetary Policy Rule for Japan,”
Bank of Japan Monetary and Economic Studies, vol. 11 (November 1993),
pp. 1–45.
. “Robustness Properties of a Rule for Monetary Policy,” CarnegieRochester Conference Series on Public Policy, vol. 29 (Autumn 1988),
pp. 173–203.
. “The Case for Rules in the Conduct of Monetary Policy: A
Concrete Example,” Federal Reserve Bank of Richmond Economic
Review, vol. 73 (September/October 1987), pp. 10–18.
McCallum, Bennett T., and Edward Nelson. “Nominal Income Targeting in an
Open-Economy Optimizing Model,” Journal of Monetary Economics, vol.
43 (June 1999a), pp. 553–78.
. “Performance of Operational Policy Rules in an Estimated
Semiclassical Structural Model,” in John B. Taylor, ed., Monetary Policy
Rules. Chicago: University of Chicago Press for NBER, 1999b.
Meltzer, Allan H. “The Transmission Process,” in The Monetary Transmission
Process: Recent Developments and Lessons for Europe. London: Macmillan Publishers, Ltd., for Deutsche Bundesbank, 2000 (forthcoming).
. “Limits of Short-run Stabilization Policy,” Economic Inquiry, vol.
25 (January 1987), pp. 1–14.
Orphanides, Athanasios. “The Quest for Prosperity Without Inflation,” Working
Paper, Board of Governors of the Federal Reserve System, May 1999.
. “Monetary Policy Rules Based on Real-Time Data,” Finance and
Economics Discussion Series 1998–03, Board of Governors of the Federal
Reserve System, 1998.
Stuart, Alison. “Simple Monetary Policy Rules,” Bank of England Quarterly
Bulletin, vol. 36 (August 1996), pp. 281–87.
Svensson, Lars E. O. “Inflation Targeting as a Monetary Policy Rule,” Journal
of Monetary Economics, vol. 43 (June 1999), pp. 607–54.
Taylor, John B. “A Historical Analysis of Monetary Policy Rules,” in Monetary
Policy Rules. Chicago: University of Chicago Press for NBER, 1999a.

B. T. McCallum: Alternative Monetary Policy Rules

79

, ed. Monetary Policy Rules. Chicago: University of Chicago Press
for NBER, 1999b.
. “Discretion Versus Policy Rules in Practice,” Carnegie-Rochester
Conference Series on Public Policy, vol. 39 (November 1993), pp. 195–
214.


Federal Reserve Bank of St. Louis, One Federal Reserve Bank Plaza, St. Louis, MO 63102