View original document

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

"

.

:

•-...-•

" .




CIS. Monetary Policy
and Financial Markets

By Ann-Marie Meulendyke
Federal Reserve Bank of New York




CIS. Monetary Policy
and Financial Markets

By Ann-Marie Meulendyke
Federal Reserve Bank of Hew York
Federal Reserve Bank of New York
33 Liberty Street
New York, N.Y. 10045




U.S. MONETARY POLICY AND FINANCIAL MARKETS
LIBRARY OF CONGRESS CATALOG CARD NUMBER 82-71781
FEDERAL RESERVE BANK OF NEW YORK
Designed by Howard Mont Associates Inc.

Forward
From time to time officials at the Federal Reserve Bank of New
York have prepared special works sharing with economists, market
participants, and others their own personal perspective on the monetary policy process. Of particular note in this regard was Robert
Roosa's 1956 essay entitled Federal Reserve Operations in the
Money and Government Securities Markets. In 1982, Paul Meek, in
U.S. Monetary Policy and Financial Markets, described a policy setting process that had changed considerably from that described by
Bob Roosa. The nature and functioning of financial markets continued to change in subsequent years, and the conduct of monetary
policy has evolved as well. Since Paul Meek's book was published,
the procedures of policy implementation that he described as "new"
have been transformed again.
In this volume, Ann-Marie Meulendyke, a manager and senior
economist assigned to the Bank's domestic trading desk, has offered
her personal perspective on the monetary policy process and financial markets of the 1980s. The new essay describes the recent evolution of Federal Reserve procedures and places them in the context
of the longer historical sweep of Federal Reserve policy. This new
essay should benefit students of the subject well into the 1990s.
E. Gerald Corrigan
President
Federal Reserve Bank of New York
New York City
December 1989







Acknowledgements
Paul Meek wrote the first edition of U.S. Monetary Policy and
Financial Markets, published in 1982. With its detailed descriptions
of the policy process at the Federal Reserve, it proved to be a valuable
and widely used resource for students and financial market participants in the CInited States and abroad. To date, the Federal Reserve
Bank of New York has distributed around 55,000 copies. The book
has also been translated into Japanese, Korean, and Portuguese.
During the years since the book was published, Federal Reserve
policy procedures and U.S. financial markets and institutions have
undergone substantial change. In order to maintain the usefulness of
Paul Meek's volume, it seemed appropriate to offer a new version of
the work. What has emerged is essentially a new book on similar
themes rather than simply an update of the earlier book. Nonetheless,
in doing the writing, I have been guided where possible by the structure
of Paul Meek's book and have made significant use of material from
it. I am deeply indebted to Paul not only for paving the way with his
book but also for encouraging me in my work during the years that
we both worked in the Open Market area at the New York Fed.
Special mention must go to Donald Vangel of the Open Market
Function and to Christine Cumming of the International Capital
Markets Group, who were primarily responsible for the preparation of
Chapters 3 and 9, respectively. Don brought to the topic his knowledge
of bank management and regulation and described some of the many
changes that have been taking place in the banking industry in recent
years. Chris drew upon her knowledge of international economic relationships to write about the international transmission of monetary
policy, expanding on a topic that received only brief attention in 1982.
This book also owes a great deal to many other colleagues in the
Federal Reserve System. Members of the Open Market Function
patiently read drafts, and offered valuable suggestions. In addition,
they graciously took over some of my normal responsibilities to
give me extra time to work on the book. Many thanks to Peter
Sternlight, Joan Lovett, Robert Dabbs, Kenneth Guentner, Sandra
Krieger, Cheryl Edwards, and Jeremy Gluck, as well as Don Vangel.
Jeremy, along with Lawrence Aiken, Deborah Perelmuter, and Mary
Vitiello, deserves thanks for assisting in the preparation of the material on financial markets in Chapter 4.
A number of members of the Bank's Research Department read
and made useful suggestions on various parts of the book, particularly
the discussion of monetary policy transmission in Chapters 8 and 9.
My thanks to Richard Davis, Gikas Hardouvelis, Bruce Kasman,
Charles Pigott, Lawrence Radecki, George Sofianos, Charles Steindel,

and John Wenninger. John Partlan and Judy Cohen of the monetary
and reserve projections unit helped with Chapter 6. Margaret Greene
and Willene Johnson of the Foreign Exchange Department reviewed
sections in Chapters 5, 6, and 9 on foreign exchange intervention.
At the Board of Governors, David Lindsey and Brian Madigan
read the chapters and offered helpful comments. Robert Hetzel at
the Richmond Federal Reserve Bank and Leland Crabbe at the
Board drew upon their knowledge of Federal Reserve history to
enhance Chapter 2.
Several people outside the Federal Reserve deserve special thanks
for their careful reading of the manuscript: Dana Johnson of the First
National Bank of Chicago, David Jones of Aubrey G. Lanston, and
Jeffrey Leeds of Chemical Bank. Lawrence DiTore of Fulton Prebon
provided insights into the functioning of the Federal funds market.
Several people who once worked in the Open Market area of the
Federal Reserve contributed oral history. They included Edward
Geng, John Larkin, Paul Meek, Robert Roosa, and Robert Stone.
Research assistance, editing, and typing are invaluable to any
such endeavor. Debra Chrapaty provided primary research assistance, spending long arduous hours reading historical documents
on microfilm. Two summer interns, Francisco Gonzales and
Jennifer Hof, provided additional research assistance. Robert
Hellmann provided research assistance on Chapter 9. Valerie
LaPorte patiently edited the book, always keeping me attentive to
the reader's desire for clarity. Peter Bakstansky, John Casson, and
Dan Rosen of the Public Information Department assisted with a
range of editorial and production jobs. Last but by no means least,
my secretary, Evelyn Schustack, earns many thanks for her patient
typing and retyping of drafts. Renee Legette also assisted with the
typing. While all of the people named helped to make the book
much more accurate and readable than it might otherwise have
been, I bear the responsibility for the remaining errors.
Ann-Marie Meulendyke
New York
December, 1989




CI.S. Monetary Policy
and Financial Markets

Monetary Policy and the U.S. Economy
Money and the Economy
Money and the Policy Process
Box: Money and Credit Definitions
The Tools of Policy
Plan of the Book
1. Evolution of Federal Reserve Procedures
2. The Depository Institutions
3. The Role of the Financial Markets
4. The Policy Process
5. The Economic Impact
6. International Dimensions of Monetary Policy
7. Reflections on the 1980s

18

48

3
4
5
7
10
11
11
12
13
14
16
17

The Federal Reserve and U.S. Monetary Policy: A Short History
The Beginning of the Federal Reserve and
World War I: 1914 to 1919
Adapting to a Changed Environment in the 1920s
Major Contraction: 1929 to 1933
Active Policy making by the Adminstration: 1933 to 1939
Accommodating War Finance in the 1940s
Resumption of an Active Monetary Policy
in the 1950s and 1960s
Targeting Money Growth and
the Federal Funds Rate: 1970 to 1979
Targeting Money and Nonborrowed
Reserves from 1979 to 1982
Monetary and Economic Objectives with Borrowed
Reserve Targets: 1983 to the Present

47

The Role of Depository Institutions
The Business of Banking
Banking Risks
Strategic Considerations
Tactical Considerations

49
54
59
60




18
20
25
28
30
32
38
43

Financial Intermediaries and the Financial Markets
Bank-related Financial Markets
1. The Federal Funds Market
2. The Market for Certificates of Deposit
3. Bankers' Acceptances
Box: Financing through Bankers' Acceptances
4. The Eurodollar Market
5. The Interest Rate Swap
Nonbank Financial Instruments
1. The U.S. Treasury Debt Market
a. The primary market
b. The secondary market
c. Derivative products
2. The Market for Federally Sponsored Agency Securities
3. Corporate Dept Instruments
a. Commercial paper
b. Corporate bonds
4. Municipal Securities

The Financial Markets
71
72
72
75
78
80
82
84
85
85
85
87
92
94
98
98
100
103

66

5
The FOMC Meeting: Developing a Policy Directive
Preparation
106
The Meeting
108
1. Reports of the Managers
109
a. The report on international developments
109
b. The report on domestic operations
111
2. Sizing Up the Economic Situation
112
a. The board staff presentation
112
b. Discussion of the economy
113
3. The Longer Run Monetary Targets
113
a. Presentation
113
b. Developing annual ranges
114
4. Short-Run Policy Alternatives
116
a. Presentation
116
5. Preparing the Directive
118
a. Discussion of the specifications
118
b. Writing the directive
119
Box: Operating Paragraphs from FOMC
Domestic Policy Directives
120



106

124

148

The Trading Desk: Formulating Policy Guidelines
Formulation and Operation of the Reserve Objective
1. Creating the Nonborrowed Reserve Objective
2. Adjustment by the Banking System to Policy Actions
Box A: Description of Reserve Measures
Total Reserves:
Required Reserves:
Excess Reserves:
Borrowed Reserves:
Nonborrowed Reserves:
Estimating Reserve Availability
Box B: Forecasting Factors Affecting Reserves
Currency:
Treasury Cash Balances:
Federal Reserve Float:
Foreign Exchange Intervention:
Foreign Central Bank Transactions:
Other Factors:

124
124
126
127
127
129
132
135
136
140
141
141
141
143
144
146
147

The Conduct of Open Market Operations
The Strategy of Reserve Management
Tools of Open Market Operations
1. Outright Operations
2. Temporary Transactions
A Day at the Trading Desk
1. Daily Dealer Meetings
2. The Treasury Call
3. Formulating the Day's Program
4. The Conference Call
5. Executing the Daily Program
Communications Within the System
Adjunct Desk Responsibilities

148
150
151
155
157
158
159
162
168
170
173
174




8
Evolving Views of Policy Transmission
Monetary Policy and Yield Curves
Responses to Operating Procedures
Policy's Effect on the Economic Sectors
1. The Household Sector
2. The Business Sector
3. State and Local Governments
4. The U.S. Government
The Role of the Fed Watchers
1. Projecting and Interpreting the
Behavior of Reserves and Desk Activity
2. Budgetary and Economic Forecasting

Responses to Federal Reserve Policy
179
185
188
190
191
193
194
195
197
198
200

International Aspects of Monetary Policy and Financial Markets
Key Changes in the International Financial System
202
International Channels of Transmission
of U.S. Monetary Policy
204
1. The Impact of Expectations and Financial Asset Prices
204
2. International Effects of Changes in
U.S. Real Activity and Prices
210
International Influences on U.S. Monetary Policy
212
The Impact of Changes in International Financial Markets
215
The Principal International Short-Term Markets
216
1. The Eurodollar Deposit Market
216
2. Mondeposit Money Markets
217
International Money Markets and U.S. Monetary Aggregates 218
The Principal International Long-Term Markets
219
1. The Eurobond Market
219
2. U.S. Treasuries and Other Government Bonds
220
Conclusion
221
Box: Foreign Exchange Market Intervention
222




178

202

w
Experiences of the 1980s

226




Monetary Policy and
the U.S. Economy
Few components of economic policymaking are as important to
the nation's economic well-being as monetary policy. This book
offers the reader information on monetary policy from the vantage
point of the Federal Reserve's domestic trading desk, the area
responsible for carrying out most monetary policy actions. It
emphasizes the process of formulating and implementing policy.
As the central bank for the United States, the Federal Reserve
has been entrusted by the Congress with the responsibility for conducting monetary policy. Monetary policy is concerned with the
terms and conditions under which money and credit are provided
to the economy. Money comprises currency and coin issued by the
Federal Reserve or the U.S. Treasury and various kinds of deposits
at commercial banks and other depository institutions. Credit
encompasses loans made by depository institutions and by other
types of financial or nonfinancial entities and includes loans evidenced by debt instruments such as notes or bonds.
Under the Full Employment and Balanced Growth Act of 1978,
usually referred to as the Humphrey-Hawkins Act for its primary
sponsors, the Federal Reserve must establish annual growth targets
for the monetary aggregates and explain how these targets relate to
goals for economic activity, employment, and prices. Monetary policy is carried out by the Federal Reserve through its regulations and
techniques for the issuance of currency and its provision of reserve
balances. The behavior of reserves can in turn influence deposit
behavior since some classes of deposits are partially backed by
them.7 The Federal Reserve can influence the rates and other conditions of credit extension by its monetary policy actions, although
it cannot directly control the quantity of credit or its price.2
In addition to having a mandate to carry out monetary policy in
a way that promotes sustainable economic expansion and reasonable price stability, the Federal Reserve also has responsibilities for
promoting the smooth functioning of the nation's financial system.
It tries to accommodate the substantial short-run variations in the
1 Currency outside the Federal Reserve, including cash in bank vaults, and total reserve
holdings at the Federal Reserve constitute the monetary base, sometimes called "highpowered money" or "outside money." The monetary base is often singled out as a potential target variable. At least conceptually, the central bank has the power to control the
issue of both components of the monetary base. Traditionally, the monetary base served
as backing for other forms of money, although currently some of the items contained in
the broader forms of money have little or no such direct backing.
2 The Federal Reserve does not set targets for credit growth, although it does announce
annual monitoring ranges for a particular indicator of total credit behavior called nonfinancial debt, defined in Box A.

demand for money and credit that inevitably arise in a complex
market economy. The Federal Reserve monitors a wide variety of
financial variables and responds when they seem to indicate that
credit conditions are out of step with System policy goals.
Determining the appropriate policy stance and balancing longand short-run objectives in the execution of policy have proved to
be very challenging. Decisions must be made as events are unfolding on the basis of data whose full significance is not yet clear. The
policy actions themselves become part of the dynamic economic
processes and may have effects that extend over considerable
periods of time.
The next three sections of this chapter provide information on
the role of money in the economy and examine the tools of policy.
These sections serve as background for the discussion of the financial system and policy process in later chapters.
Money and the Economy
Since money represents generalized purchasing power, it ought
to be reasonably well linked over time with the nominal value of the
total spending and output of goods and services in the nation's
economic system. Individuals and companies choose to hold
money because its use greatly simplifies a wide range of economic
transactions. On the other hand, they limit their money balances
because holding money has costs in the form of forgone opportunities for alternative investments in goods, services, or financial
instruments. The amount of money that is consistent with the goals
for prices and output depends upon the customs, practices, regulations, and political environment of the economy. If these are stable, the relationship between money and economic activity will
tend to be stable as well. Monetary growth in excess of that needed
to support sustainable growth in economic activity will be associated with generalized price increases.
The amount of money that people wish to hold will not, however, always bear a constant relationship to the level of economic
activity. The demand for money will also depend upon expectations of future price changes. For instance, if rapid inflation is
expected, people will seek to minimize holdings of those forms of
money that do not provide a return sufficient to offset the expected loss of purchasing power caused by rising prices. On the other
hand, if prices are expected to be steady, people will hold more
money because of its convenience in conducting transactions.
Another factor influencing the demand for money is the ease of




conversion between money and those nonmoney instruments that
provide a greater return than money.
As underlying economic conditions or expectations shift, the
behavior of money will also change. Usually these changes will be
so gradual that they will not seriously interfere with short-term policymaking, but on occasion they may be rapid enough to complicate policy choices. Even when the underlying conditions are
stable, demands for money will vary considerably from day to day
and week to week in response to seasonal and institutional payment conventions. The Federal Reserve attempts to sort out these
effects and to accommodate the short-run changes in money
demand without compromising its ability to influence money over
time to achieve long-term goals.
One factor complicating the process of determining the appropriate behavior of money is the absence of a good match between the
conceptual definition of money—given in textbooks as a medium of
exchange, a standard of value, a standard of deferred payments, and
a store of wealth—and the actual financial instruments that exist in
the United States. 3 Because financial instruments have varying
degrees of "moneyness," the Federal Reserve has set forth several
definitions of money, listed in Box A. The narrow measure of money,
Ml, comes closest to conforming to all the criteria of the textbook definition, but it omits items that have most of the characteristics of
money and are often better stores of value than Ml. The broader measures, M2 and M3, capture some of these close substitutes for Ml.
Money and the Policy Process
In the policy process, "money" has traditionally served as an
intermediate target or indicator, standing between the Federal
Reserve's ultimate goals of sustainable economic growth with price
stability and the operating targets used for day-to-day policy
implementation. Money occupies this position because its behavior is related both to the ultimate policy goals of the Federal
Reserve, which cannot be controlled directly, and to the potential
policy tools over which "the Fed" has direct control. Until the
1980s, empirical data supported the view that Ml growth was a
reasonably predictable leading determinant of nominal economic
activity. The Federal Reserve had only an imprecise ability to control Ml, but over several quarters, it could come close to achieving




3 See, for instance, Thomas Mayer, James S. Duesenberry, and Robert Z. Aliber, Money,
Banking, and the Economy, 3d ed. (New York: W.W. Norton and Company, 1987), p. 5.

Box: Money and Credit Definitions
Ml consists of currency in circulation outside of the Treasury,
Federal Reserve Banks, and depository institutions; travelers checks;
demand deposits at all commercial banks other than those due to
depository institutions, the U.S. government, and foreign banks and
official institutions, less cash items in the process of collection and
Federal Reserve float; other checkable deposits (OCD), including
negotiable order of withdrawal (NOW) and automatic transfer service
(ATS) accounts at depository institutions; credit union share draft
accounts; and demand deposits at thrift institutions.
M2 consists of Ml plus overnight and continuing contract repurchase agreements (RPs) issued by all commercial banks; overnight
Eurodollars issued to (J.S. residents by foreign branches of U.S.
banks worldwide; money market deposit accounts (MMDAs); savings and time deposits (including retail RPs) in amounts of less than
$100,000; and all balances in general purpose and broker-dealer
money market mutual funds. M2 excludes individual retirement
accounts (IRAs) and Keogh (self-employed retirement) balances at
depository institutions and in money market funds. Also excluded
are all balances held by U.S. commercial banks, money market
funds (general purpose and broker-dealer), foreign governments,
foreign commercial banks, and the U.S. government.
M3 consists of M2 plus time deposits and term RP liabilities in
amounts of $100,000 or more issued by commercial banks and
thrift institutions; term Eurodollars held by U.S. residents at foreign
branches of U.S. banks worldwide and at all banking offices in the
United Kingdom and Canada; and all balances in institution-only
money market mutual funds. M3 excludes amounts held by depository institutions, the U.S. government, money market funds, and
foreign banks and official institutions. Also excluded is the estimated amount of overnight RPs and Eurodollars held by institution-only
money market mutual funds.
Total nonfinancial debt is defined as outstanding credit market debt
of the U.S. government, state and local governments, and private
domestic nonfinancial sectors. Private debt includes corporate bonds,
mortgages, consumer credit (including bank loans), other bank loans,
commercial paper, bankers' acceptances, and other debt instruments.
The Federal Reserve Board's flow of funds accounts are the source of
domestic nonfinancial debt data expressed as monthly averages.







a desired rate of money growth by adjusting either the levels of the
banks7 reserve balances or short-term interest rates. Similarly, the
response of nominal gross national product (GNP) to changes in
Ml showed seasonal and cyclical variation, but it was also reasonably predictable over the long run.
In the 1970s, the Federal Reserve sought to take advantage of
the empirical regularities and to control money growth in order to
reduce inflation. For most of the decade, it adjusted reserves to
influence the interest rate on interbank transfers of reserves—the
Federal funds rate—as a means of changing money growth. However, persistent overshooting of the money targets and other forces
had pushed prices upward until, by 1979, inflation had reached
wholly unacceptable levels. Eager to halt and wind down the inflationary process of the 1970s, the Federal Reserve adopted a
reserve targeting approach to money control in October of that
year. The technique met with considerable success if judged by its
effect on average money growth and its impact on inflation. By
1982, the economy was in a deep recession and considerable
progress had been made in overcoming inflation. Nonetheless, Ml
was growing rapidly by recent standards. It appeared that the previous relationships between Ml growth and nominal economic
activity were not standing up well. Consequently, the techniques of
policy implementation were modified late in 1982 in a way that
deemphasized the money growth targets, especially those for Ml.
The causes of the shifts in money demand have gradually
become better understood, although even at this writing many
questions remain. For about three decades, the relationship
between money and income had been reasonably stable and predictable. Nominal GNP had grown faster than Ml, so the income
velocity of Ml, or its rate of turnover per income-generating transaction, had risen an average of 3 percent a year. But a series of factors combined to make people less reluctant to hold Ml balances,
and income velocity declined during the first half of the 1980s (see
Chart 1, page 7). The spread of interest-bearing NOW accounts
made individuals more inclined to hold some of their savings in
transactions form. Lower inflation made the loss in purchasing
power from holding money balances smaller, an outcome which
made holding money a more attractive option. When interest rates
began falling, forgone interest from holding money balances also
declined. Although the demand for money rose on average, it also
became more sensitive to short-run interest rate movements. With
components of Ml paying rates that were above zero but slow to

6

Chart 1 Trend of M1 Velocity
Velocity (log scale)
10

9
8

1953

'56

'59

'62

'65

'68

71

74

77

'80

'83

'86

'89

Velocity trend from 1953.1 to 1979.4 was 3.2 percent per year

change, there were large swings in the relative relationship between market rates and rates on money balances. The trend velocities of M2 and M3 did not shift as much, but for M2, the variability
of velocity increased (see Charts 2 and 3, page 8).
While reducing their reliance on the behavior of the monetary
aggregates as policy indicators, policymakers placed greater
emphasis on measures that might be termed intermediate indicators. These included commodity prices and monthly statistics on
employment, production, and trade. Such measures are not directly controllable but, taken together, they ought to suggest at least
the direction in which policy instruments should be adjusted to
achieve the ultimate policy goals.
The Tools of Policy

The Federal Reserve traditionally had three primary instruments of monetary policy: reserve requirements, the discount
rate, and direct open market purchases and sales of U.S. government securities. Using these tools, the Federal Reserve could
affect the cost and availability of reserves to commercial banks
and other depository institutions.




Chart 2 Trend of M2 Velocity
Velocity (log scale)
2

1.6

1.4
1953

'56

'59

'62

'65

'68

'71

'74

'77

'80

'83

'86

'89

'65

'68

'71

'74

'77

'80

'83

'86

'89

Velocity trend from 1953.1 to 1979.4 was -0.03 percent per year

Chart 3 Trend of M3 Velocity
Velocity (log scale)
2

1953

'56

'59

'62

Velocity trend from 1953.1 to 1979.4 was -0.07 percent per year




8

Reserve requirements play a role in establishing the banks'
demand for reserves and help determine the effects of the other
monetary tools on bank behavior. Commercial banks and other
financial institutions accepting deposits against which payments
can be made must maintain reserves in the form of cash held in
their vaults or deposits at Federal Reserve Banks. The existence of
reserve requirements underlies the relationship between the volume of reserves and the transaction deposit component of money.
The Depository Institutions Deregulation and Monetary Control
Act of 1980 (MCA) imposed uniform reserve requirements across
all depository institutions holding transactions deposits. It also
specified a schedule for implementing the new reserve requirements. Although the MCA gave the Board of Governors of the
Federal Reserve System authority to alter reserve requirements
without regard to the designated phase-in schedule if necessary for
monetary policy, the provision was not used during the phase-in
period. Indeed, changes in reserve requirements for the express
purpose of influencing the behavior of money or credit have not
been made since 1979.
The discount window provides depository institutions with the
means to borrow reserves from the Federal Reserve at a specified
rate. The Federal Reserve, relying on administrative procedures,
limits access to the facility by restricting frequency and amount of
use. Although the volume of borrowing is usually modest, the
terms for gaining access to the discount window are an important
part of the policy implementation process. The limitations on borrowing contribute to the seemingly contradictory result that
increases in the amount of reserves in the banking system, when
provided by the discount window, act to restrict reserve availability by putting banks under pressure to find other sources of
reserves to repay the loans.
Changes in either the discount rate or the rules and guidelines
for access to the facility can affect the costs to depository institutions of obtaining reserves to support deposit and credit growth.
Discount rate changes are initiated by the regional Reserve Banks'
boards of directors and are subject to final review and determination by the Board of Governors in Washington. The response of
depository institutions to the settings of the discount rate affects
short-term interest rates. Banks respond both to the rate itself and
to the implicit or explicit message about policy contained in the
announcement. These relationships make discount policy an integral part of monetary policy. Changes in Federal Reserve policies




toward use of the window have been rare. Normally they have consisted of temporary reductions on restrictions because of an
unusual strain on the banking system.
Open market operations are the primary tool used for regulating
the pace at which reserves are supplied to the banking system.
Open market operations consist of purchases and sales by the
Federal Reserve of financial instruments, usually securities issued
by the U.S. Treasury. Open market operations are carried out by
the domestic trading desk of the Federal Reserve Bank of New York
under directions from the Federal Reserve's principal monetary
policy making unit, the Federal Open Market Committee (FOMC).
The transactions are arranged through firms that act as dealers,
routinely buying and selling Treasury debt. Purchases by the desk
add reserves while sales drain reserves from the banking system.
Such purchases and sales may be made outright or they may be
made under a temporary arrangement in which the transaction is
reversed after a specified number of days.
Reserve requirements, discount policy, and open market operations can be used separately or in combination. Even though
they are under different jurisdictions within the Federal Reserve
System, their use can be coordinated to meet the needs of a particular situation. Of the three tools, open market operations provide the greatest flexibility and are the most actively employed
tool. Nevertheless, the FOMC must take account of the settings
of the other instruments in making its choices for open market
policy. The FOMC meets regularly and monitors a variety of
monetary and economic indicators in order to choose an appropriate policy mix.
Plan of the Book




The structure of the book largely follows that of its predecessor,
Paul Meek's 1982 volume on CI.S. monetary policy. This section
previews the principal topics in the sequence in which they are
treated in the book. Although a more detailed picture will emerge
in the following pages, it may be helpful to note here certain broad
divisions in the subject matter. Chapters 2-4 cover various aspects
of the institutional setting for U.S. monetary policy. They are followed by three chapters describing the policy process itself, then
two exploring the ways that policy affects the domestic and international economy. Chapter 10 assesses the record of monetary
policy in the 1980s and the economic and financial conditions that
accompanied it.

10

1. Evolution of Federal Reserve Procedures

The history of the policy process, the subject of Chapter 2,
reveals how the Federal Reserve has responded to new problems
and changing conditions by significantly modifying its primary
goals and the techniques for achieving them. Indeed, since the
Federal Reserve's beginnings in 1914, both Congress and the
Federal Reserve have substantially revised their views of the
Federal Reserve's mandate. In the early days, the gold standard
was expected to take care of stabilizing the price level. The Federal
Reserve saw its role as providing reserves to accommodate routine
variations in the needs for credit to finance trade and as providing
currency to avoid financial panics. But the experience of the Great
Depression altered priorities, and in the years following the Second
World War, the policymakers considered economic stabilization a
primary goal. Then, during the 1970s, the goal of price stability
acquired increased importance as inflation worsened.
The tools of policy evolved over time as well. In the System's
early years, discount window loans were the predominant means
of short-term adjustments to reserves while secular changes in
money and credit stemmed primarily from changes in monetary
gold. In more recent times, both secular growth in money and
accommodation of short-term variation in money and credit
demands have been provided through open market transactions,
primarily in U.S. Treasury securities.
2. The Depository Institutions
Monetary policy reflects continuing interactions among the
Federal Reserve, financial institutions, the financial markets, and
members of the nonbank public who deposit and borrow funds.
The functioning of the depository institutions, described in Chapter
3, plays a role in transmitting the Federal Reserve's policy impulses to the economy.
Each depository institution considers many factors in managing
its balance sheet. The balance sheet compares assets, which
encompass loans and investments, with capital and liabilities, the
latter consisting of deposits and other types of debt. When a depository institution makes loans or investments, it must weigh the
interest to be earned against risks incurred, and it must take
account of the cost of capital requirements on the asset acquired
and the return on the capital that the assets should generate. In
attracting deposit liabilities, it needs to take account of the direct




11

and indirect costs involved, including paying interest and account
management expenditures as well as any reserve requirements on
those deposits. If the maturities of the assets and liabilities differ, it
must consider the implications of changes in interest rates over
their lives. In recent years, the introduction of new regulations and
the greatly increased degree of interest rate volatility have complicated the tasks of asset and liability management.
While depository institutions set conditions for accepting
deposits and making loans, their customers choose how to
respond to the rates and terms established by the institutions.
Customer behavior in turn affects the mix of deposits and the
amount of lending that takes place. Customers choose among
deposit categories on the basis of interest rates and other features,
but they also act according to their incomes and assets, the timing
of their payments and receipts, and the ease of conversion between
money and near-money instruments. The amount of credit actually extended depends on the level of economic activity and perceived gains from investments, in addition to the interest rates
charged by depository institutions on their loans compared with
the cost of alternative sources of credit.
3. The Role of the Financial Markets
The effects of monetary policy actions are not limited to the
depository institutions. Indeed, as described in Chapter 4, governments at various levels, quasi-governmental agencies, private corporations, and individuals engage in extensive direct financial
market borrowing and lending. The United States has vast financial
markets where debt and equity are created and redistributed.
These markets are competitive and serve to direct capital to the
users with the most urgent demands.
Depository institutions, other financial firms, nonfinancial businesses, and governments all place funds in, or borrow from, the
money market—the term used for financial markets specializing in
instruments maturing in a year or less—to bridge differences in
timing between receipts and payments. They also use the market
to defer long-term borrowing or lending to a more propitious time.
They use the longer term capital markets to borrow for investment
purposes. Lenders may place funds for a long period, or they may
purchase a security with the intention of selling it when cash is
needed in what is called the secondary market.
The existence of active secondary markets facilitates transfers
of existing debt instruments before maturity and enables the New




12

York Fed's domestic trading desk to conduct open market operations efficiently. Open market operations take place in two segments of the markets: that for CI.S. Treasury securities and that for
temporary purchases and sales of government securities, referred
to as repurchase agreements (RPs) and matched sale-purchase
agreements (MSPs). The Federal funds market allows depository
institutions to exchange reserve balances at the Federal Reserve
among themselves, an arrangement which promotes the efficient
use of reserves and the building of a large volume of deposits and
credit on a relatively small reserve base. The Federal funds rate,
the rate for overnight exchanges of Federal funds, responds to
reserve availability and is regarded by money market participants
and observers as a useful, although not always reliable, indicator
of the stance of Federal Reserve policy.
4. The Policy Process
The formulation and execution of monetary policy, reviewed in
Chapters 5-7, occur in several stages. The process originates with
the actions of the FOMC, which typically meets eight times a year
in Washington, D.C. At these meetings, the 7 governors and the 12
presidents of the regional Reserve Banks evaluate the economic
outlook and develop monetary policy. The Chairman of the Board
of Governors presides over the meetings; the permanent voting
members of the Committee include the governors and the president of the New York Federal Reserve Bank. Four other Reserve
Bank presidents serve as voting members on a rotating basis for
one-year terms. At every FOMC meeting, instructions are adopted
and sent to the domestic trading desk at the New York Federal
Reserve. This "directive" indicates whether the Committee desires
to increase, maintain, or decrease the degree of reserve pressure.
Reserve pressure is produced by forcing depository institutions to
borrow rationed credit from the discount window. The directive
may also indicate that potential developments—for example, in the
monetary aggregates, economic activity, or prices—could call for
adjustments to the degree of reserve pressure during the period
between meetings.
The objectives set by the FOMC for the growth of various monetary and credit measures during the current calendar year are
reported by the Chairman every February to the banking committees of the Congress as required by the Humphrey-Hawkins Act. In
July, the Chairman reports any revisions in that year's objectives,
along with preliminary goals for the subsequent year.




13

The trading desk at the New York Fed provides reserves to the
banking system in a manner designed to be consistent with the
FOMCs desired level of discount window borrowing and sensitive to
the implications for short-term rates. In implementing the Committee's directive, the desk purchases or sells U.S. Treasury debt instruments to bring reserves into line with established objectives.
Changes in the policy stance of the FOMC usually are quickly
detected by banks and other financial market participants. When
the degree of pressure on bank reserve positions is increased,
banks will have to satisfy a larger share of their reserve needs at the
discount window. Since their access to the discount window is limited, they will bid up the Federal funds rate. Over time, depository
institutions will respond to the reserve restrictions by shifting the
rate structure of their assets and liabilities. Their actions help to
lower financial asset prices, raising market rates and providing
incentives for other economic participants to reduce their holdings
of money and their use of credit. Gradually, growth of money balances and credit should slow. At some point, the pace of real economic activity and of inflation will abate. Conversely, when pressure
on bank reserves is eased, the Federal funds rate falls, and over
time banks will be encouraged to acquire more assets. The resultant
portfolio adjustments will eventually work to spur monetary growth,
increase credit availability, and quicken economic activity.
5. The Economic Impact




What, then, are the channels through which monetary policy
impulses are transmitted to the economy? This question, addressed
in Chapter 8, is difficult to answer because lags and feedback
effects hamper efforts to trace all connections. Furthermore, a
complex economy operating in a wider world context will not
always react in a predictable way to a particular policy initiative.
Nonetheless, much has been learned over the years. Individuals and
businesses make decisions to buy or sell goods and services and to
borrow or lend on the basis of current and expected values of
income, interest rates, and prices. In addition, they respond to the
ease or difficulty of obtaining credit. It is the job of the Federal
Reserve to analyze these influences and to formulate a monetary
policy that appropriately responds to them.
Analysts of the monetary transmission process differ in the
importance they attach to the various channels. Some economists,
often referred to as Keynesians, emphasize the influence of interest
rates on economic decisions. They contend that a stimulative mon-

14

etary policy arises when sufficient reserves are provided to reduce
interest rates. Lower rates are viewed as stimulating borrowing for
investment and consumption expenditures. These expenditures in
turn will encourage an increase in output. If output rises to the point
where it is straining the productive capacity of the economy, then
competition for scarce resources will result in higher prices.
Other economists, often called quantity theorists or monetarists,
have emphasized the importance of adjustments in money supply
and demand in determining the state of the economy and the
behavior of the price level. They expect that money demand will
be reasonably stable over meaningful periods of time while recognizing that demand can be subject to short-run variations and that
shifts can arise out of institutional change. The quantity theorists
include interest rates, income, and prices in the list of factors determining the demand for money. They maintain that money demand
is fairly predictable over time, but they acknowledge that economic
activity and prices may be slow to respond to monetary growth
impulses since people do not fully adjust their spending patterns
as soon as their money balances change.
The impact of expectations on economic decisions has received
increased attention in the models of both groups of economists in
recent years. Expectations formulation has become an important
component of the analysis of the monetary transmission mechanism. In particular, much interest has centered on expectations
about inflation and their effect on the interpretation of interest
rates. Judging whether interest rates are high or low requires
knowing people's expectations about the degree to which inflation
will erode the purchasing power of money during the term the
funds are borrowed or lent.
Many analysts examine monetary policy transmission in the
context of the business cycle. In recent decades, the Federal
Reserve has attempted to take account of lags in the policy process and to anticipate economic responses in its use of policy to
counter cyclical developments. The idea is to take restrictive policy steps before the economy overheats and inflation follows, and
to initiate easing steps before a recession occurs. In practice, however, imperfect forecasts have meant that the Fed cannot always
correctly anticipate the best time for a shift in policy. Still, once an
economic turn has been recognized, prompt responses can mitigate extremes of business fluctuations.
Various sectors of the economy will respond differently over the
business cycle to monetary policy influences, in part because




15

interest rate changes have different implications for them. For
instance, consumers as a group are net creditors, while the federal
government is generally a net debtor. Moreover, within each sector
and income group, debt or credit positions will vary considerably.
Finally, the communication of economic and financial developments can be a factor in policy transmission. The speed with which
information is disseminated has increased markedly in recent
years. As prices and interest rates have become volatile over the
last two decades, firms that have particular needs to predict and
understand interest rate developments expanded the resources
they devote to monitoring the economy and Federal Reserve policy in order to avoid adverse surprises.
6. International Dimensions of Monetary Policy
In the United States, monetary policy is still largely conducted
with an eye toward domestic economic conditions and still based
on domestic monetary and financial aggregates. Nevertheless, it
has become increasingly apparent that the United States is far
from being a closed economy. As Chapter 9 shows, U.S. monetary
policy can have a significant impact on other countries' economies, and developments abroad can affect the U.S. economy to
a substantial degree. Moreover, foreigners use U.S. dollars as a
transactions medium, a store of value, and to establish value in
long-term contracts. In many dollar transactions, U.S. residents
are not participants, and the transactions do not enter into any U.S.
economic or financial statistics.
The increased awareness that the United States is an open economy that cannot operate in isolation from the rest of the world
reflects the rapid expansion of international trade and financial
transactions in the postwar period. As foreign trade has grown,
both absolutely and as a share of GNP, exchange rates have come
to have substantial bearing on U.S. income and production levels
and on the rate of inflation in the United States. Increased trade has
been accompanied by enlarged international capital flows, which
were facilitated by the dismantling of capital controls by many
nations in the 1970s. Looked at in isolation, the floating exchange
rates that replaced pegged rates in the early 1970s increased the
opportunities for each country to pursue its own monetary policy
goals independent of the actions of other nations. Nevertheless,
increased trade and financial flows worked to make exchange rate
changes—including those that stem from monetary policy actions
—important policy considerations. They also elevated the impor-




16

tance of coordinated policy procedures among major countries in
the world economy.
7. Reflections on the 1980s
In the 1980s, monetary policy contributed importantly to promoting economic expansion and reducing inflation. But the decade
also witnessed a mounting debt burden and many unsettling financial developments, including the apparent breakdown of traditional
policy relationships. Chapter 10 briefly reviews some of the developments that shaped monetary policy in the decade.




17

2
The Federal Reserve and
U.S. Monetary Policy:
A Short History1
The tools that the Federal Reserve uses today and its approach
to formulating and implementing monetary policy have evolved
considerably from what the framers of the Federal Reserve Act had
in mind in 1913. The economic consequences of two world wars,
the Great Depression, and the inflation of the 1970s have contributed to significant changes in Federal Reserve policy priorities
and in the techniques and tools used to pursue them. A System
that was decentralized at the outset has become much less so; the
goal of economic stabilization now figures importantly in the
Federal Reserve's objectives for policy; and open market operations, a procedure not even mentioned in 1913, has become the
primary tool of policy. This account focuses on the changing views
of the Federal Reserve's primary monetary policy responsibilities
and on the discovery and development of policy guidelines and
tools. It should provide some understanding of the roots of the current policy process that is the focus of much of the book.
The Beginning of the Federal Reserve and World War I: 1914 to 1919
The U.S. Congress passed the Federal Reserve Act in December
1913, creating a system consisting of the Federal Reserve Board in
Washington and 12 regional Federal Reserve Banks with main
offices and branches to serve the entire country. Congress acted
because the decentralized banking system of that time could not
respond adequately to variations in the cash and credit requirements of the economy. The Federal Reserve System was directed,
in the words of the preamble to the Federal Reserve Act, "to furnish
an elastic currency, to afford the means of rediscounting commercial paper, to establish a more effective supervision of banking in
the United States, and for other purposes." It was anticipated that
credit extended by the Federal Reserve Banks to commercial banks
would rise and fall with seasonal and longer term variations in business activity, thus providing a self-adjusting mechanism that would
prevent shortages of currency or runs on banks from leading to
financial panic and a breakdown in the economy. The framers did
not worry about the inflationary potential of such accommodative




1 This chapter draws heavily from policy records of the various Open Market committees,
annual reports prepared by the Open Market Function of the New York Federal Reserve; and
Milton Friedman and Anna Jacobson Schwartz, A Monetary History of the United States
1867-1960 (Princeton: Princeton University Press, 1963).
The first three sections make extensive use ofW. Randolph Burgess, The Reserve Banks and
the Money Market (Mew York: Harper and Brothers, 1936); and House Committee on
Banking and Currency, Subcommittee on Domestic Finance, Federal Reserve Structure and
the Development of Monetary Policy: 1915-1935, 92nd Cong., lstsess., December 1971.

18

credit provision, perhaps because long experience with the gold
standard had led them to expect that gold flows would limit tendencies to inflation or deflation.
From the beginning, the Federal Reserve was reasonably successful in accommodating the seasonal swings in the demand for
currency—in the terminology of the act, providing for "an elastic
currency." It thereby alleviated some of the troublesome strains on
the commercial banks that arose from the cyclical pattern of credit
demands in agriculture and from the year-end rise in currency
demand. Interest rates no longer exhibited seasonal fluctuations to
the degree that they had earlier.2 Other aspects of the System's
mandate developed more slowly and were subject to experimentation and controversy.
The act established a decentralized system. The regional Reserve
Banks were to have considerable authority to set the terms for credit
provision and regulate member banks in their districts. The Board
in Washington was assigned responsibility to oversee the activities
of the Reserve Banks. The Board consisted of a governor and four
other regular members; with the Secretary of the Treasury and the
Comptroller of the Currency designated as ex officio members. The
12 regional banks had considerable autonomy and power to
respond to financial conditions in their districts. They were headed
by governors, most of whom had been commercial bankers.
Between the outbreak of World War I in 1914 and the United States
entry into the war in 1917, gold flowed into the country from Europe
to purchase goods needed for the war effort. The Federal Reserve
found that it did not have the tools to offset the inflationary impact
of the inflows. It did not have the power to raise reserve requirements; indeed, the Federal Reserve Act mandated reductions in
reserve requirements for several years while reserve balances were
being consolidated at the Federal Reserve rather than scattered
among the large commercial banks. The Reserve Banks did not yet
have securities they could sell to absorb liquidity. Indeed, there were
only minimal amounts of Treasury debt outstanding, most of it backing national bank notes. During those years, the only tool potentially
available to offset the reserves provided by gold inflows was the discount window. Discount rates (or rediscount rates as they were then
called)—the rates at which the Reserve Banks made loans to the
member banks by discounting eligible paper—could have been
2 Jeffrey A. Miron, "Financial Panics, the Seasonality of the Nominal Interest Rate, and the
Founding of the Fed, "The American Economic Review, vol. 76, no. 1 (March 1986), pp.
125-40.




19

raised sufficiently to discourage banks from using the facility. While
the rates varied considerably, the governors left them low enough to
encourage banks to use the facility to obtain needed reserves. They
did discuss boosting the rates to discourage banks from borrowing,
but did not take that step. The rates differed among Reserve Banks
and according to the type of paper being discounted.
Once the United States entered the war, gold flows almost disappeared. The United States extended massive loans to its allies,
eliminating their need to make gold payments to the United States,
and it restricted exports of gold. The Federal Reserve also had to
cope with the large issuance of Treasury debt needed to finance the
war effort. The Secretary of the Treasury insisted that the Federal
Reserve hold down interest rates while the Treasury's Liberty Loan
issues were being sold. The Fed made it easy for member banks to
buy the issues by allowing them preferential rates for the discounting of Treasury securities. Expansion of Federal Reserve credit
took the place of gold inflows as a major source of inflationary
growth in money and credit.
After the war, the Federal Reserve struggled to sort out how to
operate in a climate that had changed greatly. The Treasury had
become an important participant in the credit markets. The discount rate was held down to support Treasury finance; deposits
expanded and inflation accelerated, prompting an outflow of gold.
Federal Reserve officials debated whether penalty discount rates
should be established or moral suasion used to discourage banks
from extending credit for speculation in commodities. Decisions
were deferred, however, until 1920, when the outflow of gold had
reached critical proportions, and the combination of currency
expansion and gold outflows had reduced the ratio of gold to
Federal Reserve notes to a level approaching the 40 percent legal
minimum then in effect. In that year, the Treasury dropped its
opposition to higher rates. Higher discount rates reversed the gold
outflows but contributed to dramatic declines in money and prices
and a short but severe economic contraction.3
Adapting to a Changed Environment in the 1920s
The 1920s were marked by ongoing discoveries about the effects
of the various monetary policy tools and considerable debate about
the role of the Federal Reserve. For much of the decade, the banks




3 Leland Crabbe, "The International Gold Standard and U.S. Monetary Policy from World
War I to the New Deal," Federal Reserve Bulletin, vol. 75, no.6 (June 1989), pp. 423-40.

20

made heavy use of the discount window. There was an understanding
that individual banks should not be continuously in debt to the
Federal Reserve, but on any given day, about one-third to one-half of
the member banks were likely to be borrowing. Large banks were
expected to repay their loans within a few days while smaller banks
could borrow for a couple of weeks at a time. Borrowed reserves often
met a significant portion of the banks' total reserve requirement.
The discount rates were usually kept modestly above the rate on
90-day bankers' acceptances and modestly below the rate on fourto six-month commercial paper. Occasionally the Fed made
attempts to discourage use of the discount facility for speculative
purposes. Multiple rates for discounting different types of paper
prevailed through 1921. Rates often differed among the regional
Reserve Banks until World War II. Discount rate changes had to be
approved by the Board, a requirement which sometimes precipitated disputes between the Board and the Reserve Banks. On average, the discount rate was changed about twice a year.
Federal Reserve thinking was influenced by the so-called real
bills doctrine, particularly in Washington, where Board member
Adolph Miller was its strongest advocate. This doctrine held that
credit used to finance commercial activity should expand and contract in line with the needs of trade. Accordingly, because shortterm commercial bills were issued to finance commercial transactions, it was believed they could not be issued in excessive
amounts and could not be inflationary. In contrast, other loans
might encourage speculation and thus could be excessive. This
reasoning led some to conclude that the Federal Reserve should
encourage financing conducted through commercial bills and discourage speculation.4 Other hypotheses were being developed at
the New York Federal Reserve and in academic circles. Inflation,
according to these alternative views, arose from excessive credit
expansion. Any provisions of Federal Reserve credit, regardless of
the original reason for the extension, would stimulate economic
activity and could potentially lead to inflation.5
From its founding, the Federal Reserve had promoted the creation and development of bankers' acceptances (BAs), a form of
4 The Tenth Annual Report of the Federal Reserve Board, released in February 1924, sets out
the Board's view of policy.
5 Benjamin Strong, "Federal Reserve Control of Credit," address delivered before students of
the graduate college, Harvard University, Cambridge, Massachusetts, November 28, 1922,
reprinted in Federal Reserve Bank of New York Quarterly Review, Special 75th Anniversary
Issue, May 1989, pp. 6-14.




21




commercial bill (described in Chapter 4). BAs were believed to be
a desirable means of promoting domestic and international trading
of goods. Federal Reserve Banks had purchased BAs before 1917
in order to provide earning assets to meet expenses and to encourage the growth of the instrument. The volume of purchases had fallen off after the United States entered the war, when earnings from
discount window loans covered expenses. Purchases of BAs were
resumed in the 1920s, initially to build up earnings of the Reserve
Banks and to help develop a secondary market for these instruments. To this end, Federal Reserve Banks also arranged repurchase agreements against BAs. In accord with the real bills doctrine, many officials initially did not believe that Federal Reserve
purchases of BAs could be inflationary. Purchases of Treasury certificates of indebtedness evoked more concern. By removing
Treasury securities from bank portfolios, they freed funds that could
be used for speculative purposes.6
Early in the 1920s, most Federal Reserve officials still regarded
open market purchases primarily as a source of revenue rather than
as a tool for regulating reserves in order to control money and credit.
Each regional Bank made its own purchases of both Treasury securities and BAs. It soon became apparent that these purchases had an
impact on short-term market interest rates. Benjamin Strong, the
influential Governor of the New York Reserve Bank, was one of the
first officials to recognize the power of open market operations to
affect reserve and credit conditions and, through them, economic
activity and prices. He argued that under a system with fractional
reserve requirements, increases in bank reserves, whether they
came from an inflow of currency to the banks or from Federal
Reserve provision, would support a multiple expansion of deposits
and credit. He wanted to use open market operations to offset undesired changes in gold holdings and to stabilize economic activity.
Beginning in 1920, Governor Strong sought to achieve better
coordination of open market operations. He preferred to have all
operations on behalf of the System conducted by the New York
Federal Reserve, but initially his goal was to coordinate open market operations among the regional banks. A series of committees
were formed to explore ways to achieve coordination and prevent
the Reserve Banks from bidding for securities against each other or
the Treasury. Gradually the policy implications of the operations
6 At the time, certificates of indebtedness were the primary short-term Treasury instrument.
Treasury bills were not introduced until 1929.

22

came to be considered. The efforts to study and coordinate Reserve
Bank operations led to the creation of the Open Market Investment
Committee (OMIC) in 1923, consisting of the Governors of the
Federal Reserve Banks in New York, Boston, Philadelphia,
Cleveland, and Chicago. None of the various open market committees during the 1920s had the exclusive power to approve the open
market operations of all regional banks either in BAs or in government securities. But they did receive reports on purchases and on
redemptions of maturing issues to guide the choices for System
operations. A trading desk at the New York Fed carried out operations for the System as well as for the New York Bank.
During the 1920s, the U.S. Treasury Department believed it had
some authority over Federal Reserve operations involving
Treasury debt issues. Indeed, in 1922, the Treasury expressed distress at the amount of its securities that had been purchased and
asked the Federal Reserve Banks to liquidate their holdings of its
debt to avoid inflation. Governor Strong acquiesced to the request
for portfolio liquidation because gold inflows to the United States
were financing credit expansion. Other governors, concerned that
sales of Treasury securities would reduce earnings, agreed only
reluctantly. Because of the gold inflows, discount window use
(another source of earnings) did not rise as the portfolio declined,
and Federal Reserve earnings reached critically low levels. The
Treasury then agreed that the Federal Reserve Banks could hold
sufficient securities to cover expenses.
The view that open market operations could serve as a countercyclical tool to influence reserve and credit conditions became better understood and gained adherents as the 1920s progressed.7
Nonetheless, there were ongoing disputes between those wanting
a procyclical policy based on the demand for credit for commercial
transactions (real bills) and those who wanted to make credit readily available when the economy was in a recession and make it
stringent when the economy was growing rapidly. The OMIC, with
Treasury approval, began to use open market operations as a
countercyclical policy tool during the 1924 recession.
The OMIC gauged whether credit was tight or easy by watching
short-term market interest rates and the amount of borrowing from
7 From 1926 to 1928, the Congress contemplated legislation that would direct the Federal
Reserve to keep prices steady, legislation which the Federal Reserve spokesmen generally
opposed. The testimony by various Federal Reserve officials and academic economists
revealed the range of thinking at the time. For more information see Robert L. Hetzel, "The
Rules versus Discretion Debate over Monetary Policy in the 1920s," Federal Reserve Bank
of Richmond Economic Review, vol. 71, no. 6 (November-December, 1985), pp. 3-14.




23




the discount window. A number of analysts observed that open
market purchases that did not offset gold outflows encouraged
banks to repay discount window credits. Similarly, open market
sales encouraged increased borrowing. Some people interpreted
this pattern to mean that open market operations had no effect on
reserve availability or on a bank's ability to lend. But others, including analysts at the New York Reserve Bank, argued that limitations
on prolonged discount window borrowing might make those banks
reducing borrowing feel more comfortable in extending additional
loans. Thus, open market purchases would have an expansionary
effect. Nonetheless, some analysts who conceded that open market
operations and discount rate changes could moderate business
cycles questioned the wisdom of countercyclical monetary policy
because they feared it might impart an inflationary bias to policy.
During much of the 1920s and 1930s, outright purchases and
sales of Treasury securities in the market were the only type of open
market operation regularly undertaken at the Federal Reserve's initiative. The OMIC at its regular meetings generally authorized the
New York Federal Reserve to undertake outright purchases or sales
of Treasury debt instruments for the consolidated System Account
in amounts up to a specified level.8 This "leeway" for portfolio
changes was available if needed to achieve the desired credit conditions. Decisions would be made by observing the behavior of borrowed reserves, especially borrowings by the money center banks,
and money market conditions, exemplified by the behavior of shortterm interest rates and the ease or difficulty encountered by securities dealers in obtaining financing. Operations were conducted with
recognized dealers and were negotiated on a case-by-case basis.
Other types of open market operations were generally carried
out at the initiative of the banks or dealers and were sometimes
referred to as passive open market operations. The Federal
Reserve Banks established rates at which they would buy BAs.
Through most of the 1920s, the rates were set close to market rates
and slightly below the discount rate. If the Federal Reserve Banks
were routinely buying more or fewer BAs than the OMIC wanted,
the offering rate would be adjusted. Repurchase agreements (RPs)
against both Treasury securities and BAs were arranged on behalf
8 The Committee members were kept informed of developments affecting the System Account
through written reports prepared in the open market operations area of the Mew York Fed.
These reports described reserve and money market conditions, trading desk operations, and
weekly lending patterns of large banks, and they also provided background information on
other securities markets. They were prepared at the end of each statement period and before
each Committee meeting. The reports, with modifications, have continued to the present.

24

of nonbank dealers at the dealers' initiative for periods of up to 15
days, with early withdrawals permitted. The Federal Reserve recognized that these passive operations affected bank reserves. But
because of their temporary nature (the average maturity of BAs
purchased was only about 15 days, and the BAs were redeemed at
maturity), passive operations were generally not seen as having
policy significance. Instead, they were believed to ease temporary
credit stringencies faced by dealers when reserves were drained by
Treasury cash management operations or some other noncontrolled factor. The Federal Reserve did, on occasion, deliberately
absorb reserves through what today would be called matched salepurchase transactions. When reserves were abundant because
Treasury cash positions were abnormally low before tax dates, the
Fed sometimes made temporary sales of short-term Treasury certificates of indebtedness bought directly from the Treasury.9
Meyor Contraction: 1929 to 1933
The absence of consensus concerning either the role or the
power of the Federal Reserve to respond to cyclical forces proved
to be a severe handicap during the 1929-33 contraction phase of
the Great Depression. Economic activity had already begun to
weaken at the time of the stock market crash in October 1929, but
the Federal Reserve had felt helpless to provide stimulus without
also feeding the speculative boom in stock prices. Governor
George Harrison, who had assumed leadership of the New York
Fed after Governor Strong's death in October 1928, had argued in
1928 for a sharp but short-lived increase in the discount rate, tempered by open market purchases. The Board turned down his
requests until August 1929, by which time Governor Harrison felt
that it was probably too late. Initially, the Fed tried, with limited
success, to use moral suasion to discourage banks from borrowing
funds from the discount window to invest in financial instruments.
Once it did raise the discount rate, it made only limited use of open
market operations to soften the pressure of high rates.
On the day the stock market crashed, the New York Fed bought
about $125 million of Treasury securities, five times the maximum
weekly purchase amount permitted by the OMIC authorization.
The purchases almost doubled total holdings of government securities by all Federal Reserve Banks, which stood at $260 million on
9 Once the Banking Act of 1935 proscribed such direct Federal Reserve loans to the Treasury,
the Federal Reserve regarded temporary sales as having been ruled out as well. Burgess,
op.cit.,p. 117.




25




October 30, 1929. w The New York Fed also indicated that its discount facility would be available to help the New York City banks
that were providing assistance to other banks facing cash needs.
However, the OMIC did not approve further purchases of securities until its next meeting, worrying that such an action would be
inflationary. It then approved only enough leeway to provide for the
normal seasonal increase in currency.
In 1930, the OMIC was replaced by the Open Market Policy
Conference (OMPC), composed of all 12 Federal Reserve Bank
governors and the members of the Federal Reserve Board. Power
to call and lead the meetings was transferred from the New York
Fed governor to the governor of the Board. The reorganization,
which had been in the works since 1928, had the effect of shifting
power from the New York Fed to the Board. An executive committee, consisting of a subset of the OMPC members, met more frequently than the whole Conference and worked closely with the
trading desk at the New York Fed on the specifics of operations.
The use of an executive committee was continued until 1955, when
improved transportation made frequent meetings of the full open
market committee relatively easy to arrange.
During 1930, the OMPC resisted using a countercyclical approach
to policy to offset the weakness of economic activity. Although
Governor Harrison several times asked the OMPC for authorization
to buy more Treasury securities to promote business recovery, he
was permitted to purchase only small amounts of securities. The
predominant sentiment was that with the economy weakening, the
needs of trade were declining. Thus, the contraction in money and
credit was appropriate. At least one governor viewed the economic
weakness as the inevitable consequence of the earlier "economic
debauch" of the speculative boom.77
The Federal Reserve did lower discount rates in several steps
until 1931 but at a pace that lagged behind the effects of the contraction in money, credit, and prices. Board member Adolph Miller
argued that further cuts in interest rates were desirable to counter
the depressed business conditions. To support his view that the discount rate cuts to date might not have been sufficient, Miller con10 The New York Federal Reserve did not violate the leeway provision because it booked
only $25 million of the securities purchased to the System special investment account. It
booked the rest of the securities to the Bank's own account, a practice that was permitted
until 1935. During the week ended October 30, discount window borrowing rose by
$195 million to $991 million.
11 Statement by Governor Norris, minutes of the OMPC meeting of September 1930.

26

tended at the September 1930 meeting that in times of depression,
a money rate is "a particularly imperfect indicator of the true state
of credit." Nonetheless, the OMPC remained cautious, hoping that
economic conditions would improve.
The OMPC was disturbed by the banking crises that took place
from October to December of 1930 and in March 1931. During
these periods, bank failures and runs on banks caused the demand
for currency to rise dramatically. The Federal Reserve provided the
currency demanded but did not fully offset the reduction in member bank reserves that the banks suffered as the currency was paid
out. Available records do not indicate that the OMPC members discussed the severe contractions of member bank reserves, money,
and credit resulting from the currency drains. The OMPC made no
adjustments to its routine instructions for open market operations,
which generally authorized net purchases (and in some instances
sales) of up to $100 million of Treasury securities between meetings if needed to stabilize money market rates. Much of the
Conference's discussion following the first banking crisis was about
supervisory issues, particularly as they applied to the Bank of the
United States, by far the largest failure.72
In contrast, the Federal Reserve raised rates promptly in October
1931 to stem gold outflows that occurred after Great Britain went
off the gold standard.73 The New York Fed raised its basic discount
rate from 1 1/2 to 3 1/2 percent. The action severely strained an
already weakened financial system. The higher rates did stem the
gold outflow, but they also led to a renewed increase in the rate of
bank failures and another depositor rush to currency. Although
banks used the discount window because they needed the reserves,
they were uncomfortable doing so, and some banks feared that
using the window would be viewed as a sign of weakness.
12 Most OMPC members were bankers who subscribed to the real bills view. They apparently
did not understand the contractionary mechanism at work during the banking crises,
even though the linkages had been understood at the New York Fed for some time.
Governor Strong had testified before a congressional committee in 1926 on the power of
open market operations (Hetzel, op. cit). Strong had described the expansion side of the
mechanism in detail, explaining how an increase in reserves leads over time to a multiple
increase in deposits. He noted that a drain of currency would reduce the expansionary
potential of an increase in reserves. This was the same mechanism through which the
stepped-up currency withdrawals were depriving the bank of needed reserves and causing a serious contraction of deposits during the banking crises.
13 The literature analyzing this period debates whether the requirement for gold collateral
against currency forced the Federal Reserve's hand. Most writers have argued that it was
not really a binding constraint. See David C. Wheelock, "The Fed's Failure to Act as Lender
of Last Resort during the Great Depression, 1929-1933," Proceedings of a Conference on
Bank Structure and Competition, Federal Reserve Bank of Chicago, May 1989.




27

In April 1932, the OMPC picked up another proponent of a more
active countercyclical policy, the new Treasury Secretary, Ogden
Mills. He found the Federal Reserve's failure to act "almost inconceivable and almost unforgivable. . . . the resources of the System
should be put to work on a scale commensurate with the existing
emergency."74 In the face of strong pressure from Congress and the
Administration, the OMPC did authorize $500 million of purchases
of Treasury securities. (The leeway had been increased gradually
from $120 million to $250 million between August 1931 and
February 1932.) The reserve impact of the initial purchases was
partially offset by gold outflows, but after a couple of months, gold
flowed back. Bank failures subsided, and people began to return
currency to the banks. The banks used some of the additional
reserves to reduce their use of the discount window and to build up
their holdings of excess reserves. But money and credit also grew,
and the economy showed some meager signs of recovery.
The OMPC members, however, believed that excess reserves
were rising because banks were not finding attractive lending
opportunities. More likely, banks simply wanted more excess
reserves in the wake of the banking crises. Burgess notes that during the depression, banks became increasingly strict in their lending practices and were not taking care of their regular customers.75
But in the face of the excess reserves, the Fed gave up on adding
reserves and did not make any more substantial open market purchases after August 1932. Indeed, in November the OMPC contemplated selling securities to eliminate the excess reserves, but
the Administration discouraged that policy course. Early in 1933,
the Fed again rejected suggestions to do something stimulative,
even though a third severe banking crisis began in January and
lasted into March.
Active Policymaking by the Administration: 1933 to 1939
When the Roosevelt Administration was installed in March 1933, it
very quickly instituted a universal bank holiday in the hope of resolving the crisis atmosphere and ending the series of runs and bank failures. Banking legislation in 1933 gave legal status to the bank holiday
and authorized orderly reopenings. It allowed for issuance of Federal
Reserve notes against government collateral and emergency
issuance against other collateral. The Board was given power to alter




14 Minutes of the joint meeting of the Federal Reserve Board and the OMPC, April 1932.
15 Op. cit, p. 65.

28

member bank reserve requirements within a fairly wide range that
included the existing ratios as lower bounds, and the OMPC, as then
constituted, was formally recognized. Finally, the legislation introduced federal deposit insurance and created the Federal Deposit
Insurance Corporation (FDIC). Temporary insurance began in
January 1934 while a more permanent plan was worked out.
The Banking Act of 1935 went further. It restructured the Federal
Reserve System, introducing the basic structure that exists today.
The Board became the Board of Governors of the Federal Reserve
System, with seven Governors, one of whom was designated
Chairman. The Treasury Secretary and the Comptroller of the
Currency no longer sat on the Board. The act formally charged the
Board with responsibility for exercising such powers as it possessed to promote conditions consistent with business stability.
The Reserve Bank governors were redesignated as presidents, and
membership of the renamed Federal Open Market Committee
(FOMC) was limited to five presidents at any one time. The act also
took away the power of individual Reserve Banks to buy or sell
government debt without permission of the FOMC, thereby formally ending one of the major controversies of the 1920s. Finally, it
made permanent the provision of deposit insurance.
The Roosevelt Administration generally supported activist government economic policies, and it took the lead in ending the pattern of money contraction. In 1934, Marriner Eccles was appointed
Governor of the Board (and later Chairman as the restructuring took
effect). He was a strong believer in an active Federal Reserve policy
to combat deflation and unemployment. Nevertheless, the Federal
Reserve actually made little use of either open market operations or
rediscounting, its traditional policy tools, between 1934 and 1939.
Instead, gold returned to center stage as the primary source of
money expansion. The Administration took the country off the gold
standard in April 1933. It allowed the price of gold to rise in the market until it established a new parity of $35 a troy ounce in January
1934, up from $20.67. The price was high enough to attract a large
gold inflow from abroad, which the Treasury monetized by issuing
gold certificates to the Federal Reserve. The Federal Reserve did
not offset the resulting rise in reserve balances. Furthermore,
because deposit insurance was increasing public confidence in the
banks and ending the runs, currency flowed back to the banks and
increased their reserves. Hence, even though the Federal Reserve
took no action, reserves and money grew rapidly between 1934 and
1937, and economic activity expanded.




29

The gold and currency flows did stimulate money growth, but
reserves grew even faster and the banks built up unprecedented
holdings of excess reserves. At the time, Fed officials were puzzled
by the buildup, and many of them interpreted it as a sign that there
was no loan demand from creditworthy customers. They worried
that the excess reserves could set off inflation at some point in the
future and consequently sought a way to eliminate them. Open
market sales of securities were contemplated, but the excesses
were so large that such sales would have reduced Federal Reserve
earnings to the point where covering expenses might have been
difficult. Discount window borrowing already was negligible, so
there was no scope for further reductions.
Instead, the Federal Reserve turned to its new tool, reserve
requirement ratios, and raised the ratios dramatically in several
steps in late 1936 and early 1937. To the frustration of Fed officials,
the banks built up their excess reserves again and, in the process,
contracted the money stock. At the same time, the Treasury
stopped issuing gold certificates to the Federal Reserve against the
gold inflows, thus halting the reserve injections from that source.
Economic activity contracted until 1938 when the Fed reduced
reserve requirements modestly and the Treasury resumed monetizing gold inflows.
The Federal Reserve made almost no use of open market operations to change the size of its portfolio, not even to offset seasonal
movements in currency and the Treasury balance. Variations in
excess reserves were permitted to absorb the seasonal swings in
those factors. The Fed merely replaced maturing issues and, to
achieve "orderly markets," made swaps that changed the composition of its holdings. Furthermore, even though the Fed cut the discount rate to 1 1/2 percent and then to 1 percent, the facility fell
into disuse after the banking crisis of 1933. Throughout the late
1930s, the discount rate almost always exceeded market rates on
short-term instruments. The combination of high excess reserves
and a slight penalty rate took away the incentive to use the window.
Outstanding discount window credit rarely exceeded $10 million in
the latter half of the 1930s.
Accommodating War Finance in the 1940s
Before the United States entered the Second World War, the
Federal Reserve made only very limited use of open market operations—most notably, some purchases of Treasury securities after
war was declared in Europe in 1939. Gold inflows continued to play




30

the major role in supporting reserve expansion through 1941. As
deficit financing of the war expanded, the Federal Reserve became
a more active purchaser of Treasury debt. The Treasury wanted to
keep its borrowing costs low and encouraged the Fed to hold down
interest rates. In April 1942, the Fed formally pegged the rate at
which it would buy 90-day Treasury bills at 3/8 of one percent, a
level held until 1947. It pegged rates for making purchases (or
sales) on longer term Treasury debt as well, although less formally.
Sales of Treasury bills and certificates of indebtedness to the
Federal Reserve were often substantial. Because the discount rate
was always at least 1/2 percent, banks that held Treasury bills
found it advantageous to sell them to the Federal Reserve when
they needed funding rather than to use the discount window. Hence
discount window borrowing was not important during the war.
With confidence in the banks rising and prosperous economic
times making banks more willing to expand loans and investments,
excess reserves fell. The drop in excess reserves was assisted in
November 1941 by an increase in reserve requirements. Measured
inflation picked up initially, but once the United States entered the
war late in 1941, it was very modest. Some inflation was disguised
by price controls, but the public also chose to hold more money balances and save more in a wartime economy with few consumer
goods available.
After the war, the nation's resolve to avoid another depression was
embodied in the Employment Act of 1946. The Federal government,
including the Federal Reserve System, actively sought to achieve
reasonably full employment of resources. The economy quickly
shifted resources to civilian production. In attempting to restrain
money and credit growth, the Federal Reserve was handicapped by
its commitment to stabilize interest rates on government securities.
By the late 1940s, inflationary pressures emerged as people spent
some of their accumulated wealth and reduced their money balances from the unusually high wartime levels. The government ran
large budget surpluses, but the debt outstanding was still substantial.
Accordingly, the Treasury resisted Federal Reserve requests to raise
interest rates to contain the inflationary pressures. In 1947, the
Treasury finally did agree to an upward adjustment of the rates on
the shorter maturities, creating a considerably flatter yield curve.
Federal Reserve purchases of securities were rather variable.
Despite the inflation, the 2 1/2 percent rate on long-term bonds was
above the market clearing rate, and the Federal Reserve actually
sold bonds. Money fell, and there was a mild recession in 1949.




31

Unlike most of its trading partners, the United States continued
to maintain a fixed price for gold, $35 an ounce, during and after
the war (although during the war gold exports were restricted).
Following the war, the United States ran large trade surpluses as
other countries began to rebuild. Gold flowed into the country and
thus did not constrain domestic policy. During the late 1940s, a
series of international negotiations resulted in the establishment of
a modified gold exchange standard. In addition, a new organization, the International Monetary Fund, was created to help countries reestablish pegged exchange rates and to ease the transition
to new exchange rates when currency imbalances created unacceptably large reserve flows at the existing rates. The founders of
the new system believed that it would be flexible enough to prevent
a recurrence of the international stresses of the 1930s. (In practice,
adjustments proved more difficult than had been anticipated and
were not often made.) The procedures took on the name of the
resort in New Hampshire where negotiators met, and came to be
known as the Bretton Woods System.
Resumption of an Active Monetary Policy in the 1950s and 1960s16
In 1950, inflation related to the Korean War convinced the FOMC
that the rates being pegged on Treasury securities were too low. The
trading desk attempted to discourage securities dealers from offering it Treasury issues. The desk often delayed processing offers for
several hours to induce dealers to find another purchaser. In the end,
however, if the dealers could not obtain reasonable bids from other
sources, the Fed generally bought the securities at the pegged rates.
The Treasury was reluctant to give up the ability to finance the
debt cheaply, and the Federal Reserve negotiated with the Treasury
for an extended period to gain the right to make its own monetary
policy decisions. By March 1951, an "Accord" was reached that
allowed the Federal Reserve to resume an active and independent
monetary policy. William McChesney Martin, who was soon to
become Chairman of the Board of Governors of the Federal Reserve,
handled the final stages of the negotiation for the Treasury.77




16 This section and those that follow draw heavily from Ann-Marie Meulendyke, "A Review
of Federal Reserve Policy Targets and Operating Guides in Recent Decades," Federal
Reserve Bank of New York Quarterly Review, vol. 13, no. 3 (Autumn 1988), pp. 6-17.
17 Allan Sproul, who participated in the negotiations as President of the Federal Reserve
Bank of New York, offered an interesting commentary on the process in "The 'Accord'—A
Landmark in the First Fifty Years of the Federal Reserve System," Federal Reserve Bank
of New York Monthly Review, November 1964, reprinted in Lawrence S. Ritter, ed.,
Selected Papers of Allan Sproul, Federal Reserve Bank of New York, December 1980.

32

After the Accord, the FOMC created a subcommittee, headed by
Chairman Martin, to investigate how best to carry out an active
monetary policy and to encourage the return of an efficiently functioning government securities market.28The FOMC adopted most of
the key recommendations of the subcommittee. It gradually withdrew its support of interest rates.79 Between 1953 and 1960, it
emphasized that it was no longer pegging interest rates by pursuing
what came to be known as a "bills only" policy. Open market operations were confined to the short maturity Treasury bill sector, leaving the longer maturity coupon securities to trade without Federal
Reserve interference. On only two occasions—when the coupon
market was widely perceived to be "disorderly"—were coupon
securities purchased. To create a climate in which the dealers could
make markets on an equal footing, the trading desk developed the
competitive "go around" technique, still in use today, in which all of
the dealers are contacted simultaneously and given the opportunity
to make bids or offers. The desk also increased the number of dealers with which it would trade and specified criteria that dealers had
to meet to qualify for a trading relationship.20
During the 1950s, the Federal Reserve developed open market
operations into the primary tool for carrying out monetary policy, with
discount rate and reserve requirement changes used as occasional
supplements. Margin requirements on stock purchases were adjusted
occasionally to encourage or discourage credit use. In establishing
open market policy, the FOMC took into account that the level of the
discount rate would influence interest rates and the banks' perception
18 "Federal Open Market Committee Report of Ad Hoc Subcommittee on the Government
Securities Market," 1952, reprinted in House Committee on Banking and Currency, The
Federal Reserve System after Fifty Years: Hearings before the Subcommittee on Domestic
Finance, 88th Cong., 2ndsess. (Washington, D.C.: GPO, 1964), vol. 3, pp. 2005-55.
19 Although it ended its routine support of interest rates in 1952, the Federal Reserve followed
a so-called even keel policy during Treasury financing periods until the early 1970s. In the
1950s and 1960s, most Treasury coupon securities were sold as fixed-price offerings.
Around the financing periods, the Fed avoided changes in policy stance and tried to prevent changes in money market conditions. Major financing operations occurred four times
a year, around the middle of each quarter, but extra unscheduled financing operations
occurred when the Treasury found itself short of money. In the 1970s, debt issuance was
put on a regular cycle on the recommendation of Treasury Secretary William Simon, and
coupon issues were generally sold at auction.
20 During the subcommittee hearings, several dealers had objected to the technique used by
the Trading desk to arrange an open market operation at its own initiative. The desk, on
a rotating basis, had chosen 1 of a group of 10 recognized dealers as a broker or agent to
handle its orders in the market. The dealers that were not part of that group complained
that they were unfairly excluded from dealings with the Federal Reserve. The dealers that
were part of the group were dissatisfied because they could not transact business with
the Fed for their own portfolios at times when they served as agent.




33




of reserve availability. It did not (and does not), however, have the
authority to change the discount rate, and it considered the rate to be
given within the context of short-term policymaking. The Board of
Governors approved periodic adjustments to the discount rate when the
rate got out of line with market rates. On other occasions, changes were
made in conjunction with adjustments in other tools when the Board
wished to emphasize a shift in policy stance. The window was administered to reinforce the banks' reluctance to borrow from the Federal
Reserve. The Board changed reserve requirements occasionally to signal a policy shift. The changes were far smaller in magnitude than those
of the 1930s, and the impact on reserves was generally cushioned with
open market operations that partially offset the reserve impact.
While FOMC members believed that interest rates played an important role in the economy, they felt it would be unwise to establish interest rate targets. The use of such targets, they reasoned, would increase
the difficulty of making a break with the strict rate pegging of the
1940s. In developing policy guidelines at its meetings, the FOMC considered a number of indicators. It gave special emphasis to the behavior of bank credit (commercial bank loans and investments) as an
intermediate policy guide. It sought to speed up bank credit growth in
periods when economic activity showed weakness and slow it down in
periods of rapid growth. It did not have direct control over bank credit,
however, or even timely information on recent performance, so bank
credit was not suitable for day-to-day operating guidance.
At the conclusion of each meeting the FOMC created a written
directive for the trading desk at the New York Fed. It was deliberately nonspecific, avoiding even a hint of targeting interest rates.
The Manager of the System Open Market Account surmised from
listening to the discussion at the FOMC meeting what policy steps
the Committee wanted.27
The desk's day-to-day operations focused on free reserves and
money market conditions. Free reserves are defined as excess
reserves less reserves borrowed from the discount window.22 Free
reserves were targeted in order to provide some anchor to the policy
guidelines. A relatively high level of free reserves was regarded as
21 At that time, the Trading desk was not authorized to modify its policy stance between
meetings without receiving additional instructions from the Committee. The executive
committee of the FOMC met frequently—generally every two weeks through the middle
of 1955. Subsequently, the full Committee met every three weeks. The Committee sometimes had telephone meetings between regular meetings.
22 Free reserves are referred to as net borrowed reserves when borrowed reserves are greater
than excess reserves. (Descriptions of the various measures of reserves appear in Chapter
6, Box A.)

34

representing an easy policy: the excess reserves available to the
banks were expected to facilitate more loans and investments. Net
borrowed reserves left the banks without unpledged funds with which
to expand lending and were consequently viewed as fostering a
restrictive policy environment. High rather than rising free reserve
levels were thought to foster expanding bank credit since banks
would perpetually have more excess reserves than they wanted and
would keep increasing their lending. High net borrowed reserve levels
would, in a parallel manner, encourage persistent loan contraction.23
Research staff members developed and refined techniques during
the 1950s and 1960s for estimating each day what free reserves
would be for the reserve maintenance period by forecasting both
nonborrowed and required reserves.24 The reserve factor estimates,
which affected nonborrowed reserves, were subject to sizable errors,
even though considerable resources were devoted to obtaining timely information about past and likely future behavior of the more
volatile factors. The reserve estimates and market conditions were
reviewed at a daily 11:00 a.m. conference call held with senior Board
staff officials and a president who was a voting FOMC member.25
The desk generally bought or sold Treasury bills when forecasts
suggested that free reserves were significantly below or above the
objective, especially if the free reserve estimates were confirmed by
money market conditions. RP operations were resumed in 1951. By
this stage, RPs in both government securities and BAs were generally being undertaken at Federal Reserve initiative "to provide temporary, but immediate, reserve assistance to the central money
market at times of unusual strain on that market."26 Until the 1970s,
RPs were only done with nonbank dealers at preannounced rates
23 See (Peter D. Sternlight), "The Significance and Limitations of Free Reserves," Federal
Reserve Bank of New York Monthly Review, November 1958, pp. 162-67; and "Free
Reserves and Bank Reserve Management," Federal Reserve Bank of Kansas City Monthly
Review, November 1961, pp. 10-16. A critique of free reserves and a survey of trie literature
is provided by A. James Meigs in Free Reserves and the Money Supply (Chicago:
University of Chicago Press, 1962).
24 Until 1968, maintenance periods were one week long for reserve city banks (member banks
with offices located in cities with Federal Reserve banks or branches) and two weeks long
for country banks (all other member banks). Computation and maintenance periods were
essentially contemporaneous. In 1968, the Board of Governors adopted a system of lagged
reserve accounting, under which reserve requirements were based on average deposit levels from two weeks earlier, with all member banks settling weekly. The change made it
easier to hit free reserve targets—ironically, shortly before free reserve targeting ended.
25 The daily conference call was introduced in 1954. Currently, it begins around 11:15 a.m.
26 Open Market Operations and Changes in Operating Procedures During 1954," Annual
Report of the Open Market Function, p. 18. The report went on to say that the introduction
of outright operations for same day "cash " settlement reduced the need for RPs.




35




jsually at or slightly below the discount rate—although beginning
in 1968, the RP rate was occasionally set slightly above the discount
rate. The practice of arranging RPs only with nonbank dealers was
a holdover from the earlier view that RPs served primarily to finance
dealer positions in securities. On occasion during the 1950s and
1960s, an RP would still be arranged at the request of dealers facing
difficulties in financing their positions in the markets. After the introduction of matched sale-purchase transactions (MSPs) in 1966, the
desk was also able to drain reserves temporarily.
Because the FOMC was also interested in money market conditions, the desk continued to watch the "tone and feel of the markets"
each day in deciding whether to respond to the signals given by the
free reserve estimates. The tone of the markets might suggest
whether the reserve estimates were accurate. If the banks were
short of reserves, they would sell Treasury bills, a secondary
reserve, and put upward pressure on bill rates. The banks would
also cut back on loans to dealers, thus making dealer financing
more difficult. Reading the tone of the markets was considered
something of an art. Desk officials monitored Treasury bill rates,
dealer financing costs, and comments from securities dealers concerning difficulties in financing their inventories of securities.
The rate on Federal funds played only a limited role as an indicator of reserve availability during these years although it gained
attention during the 1960s.27 The interbank market was not very
broad as the 1960s began, but activity was expanding. Until the
mid-1960s, the Federal funds rate never traded above the discount
rate. During "tight money periods," when the desk was fostering significant net borrowed reserve positions, funds generally traded at
the discount rate, and the funds rate was not considered a useful
indicator of money market conditions. When free reserves were
high, funds often traded below the discount rate and showed noticeable day-to-day variation. At such times, the funds rate received
greater attention as an indicator of reserve availability.
There was considerable surprise when the funds rate first rose
above the discount rate, briefly in October 1964 and more persistently in 1965. As large banks became more active managers of
21 There had been interbank exchanges of Federal Reserve funds (or Federal funds, as they
came to be called) as early as the 1920s; at that time, trades were mostly negotiated directly between two banks rather than through brokers. Burgess, op. cit, p. 152. For further
discussion of the expansion of the market in the 1960s, see Mark H. Willes, "Federal Funds
during Tight Money," Federal Reserve Bank of Philadelphia Business Review, November
1961, pp. 3-11, and "Federal Funds and Country Bank Reserve Management," Federal
Reserve Bank of Philadelphia Business Review, September 1968, pp. 3-8.

36

the liability side of their balance sheets, they borrowed funds in the
market in a sustained way. Banks had introduced large negotiable
certificates of deposit (CDs) in 1961. But CD borrowings were subject to reserve requirements and (until 1970) to interest rate ceilings under Regulation Q. Borrowings from other banks through the
Federal funds market were free of reserve requirements and interest rate ceilings. Furthermore, they were not subject to the restrictions on prolonged use that were applied to the Federal Reserve's
discount window. The changes in liability management techniques
meant that individual banks could expand credit even when they
did not have free reserves if they were willing to bid aggressively
for wholesale funding from other banks. Their actions were making
free reserves a less reliable predictor of bank credit growth.
In 1961, several developments led the FOMC to abandon its
"bills only" restrictions. The new Kennedy Administration was concerned about gold outflows and balance of payments deficits and,
at the same time, wanted to encourage a rapid recovery from the
recent recession. Higher rates seemed desirable to limit the gold
outflows and help the balance of payments, while lower rates were
wanted to speed up economic growth. To deal with these problems
simultaneously, the Treasury and the FOMC attempted to encourage lower long-term rates without pushing down short-term rates.
The policy was referred to in internal Federal Reserve documents
as "operation nudge" and elsewhere as "operation twist." The
Treasury engaged in maturity exchanges with trust accounts and
concentrated its cash offerings in shorter maturities. The Federal
Reserve attempted to flatten the yield curve by purchasing
Treasury notes and bonds while simultaneously selling Treasury
bills. The extent to which these actions changed the yield curve or
modified investment decisions is a source of dispute. The Federal
Reserve continued the procedure for another year and then allowed
it to lapse after short-term rates rose in 1963.
Through the middle of the 1960s, policymakers generally viewed
the basic policy process with some satisfaction. Reasonable price
stability had been reestablished, and recessions had been mild,
short-lived interruptions in a period of prolonged prosperity. In the
latter half of the 1960s, however, rising inflation began to accompany the prosperity. Primary blame was placed on the budget deficits
generated to finance CI.S. involvement in the Vietnam War and
"Great Society" social programs. But some people at the Federal
Reserve and in the academic community expressed the view that
expansionary monetary policy was also contributing to inflation.




37

Economists, both within and outside the Federal Reserve, questioned the assumptions underlying the existing monetary policy
procedures, including the connections of free reserves and bank
credit to the ultimate policy goals of economic expansion and price
stability. Quantitative methods were increasingly applied to test the
hypothesized relationships among operational, intermediate, and
ultimate policy objectives. Some studies suggested that more attention should be paid to money growth and to the behavior of total
reserves or the monetary base.28 In response, the FOMC expanded
the list of intermediate guides to policy. The directives continued to
focus on bank credit but added money growth, business conditions,
and the reserve base. Free reserves continued to be the primary
gauge for operations, although the Federal funds rate gained more
prominence as an indicator of money market conditions.
Although the FOMC met every three to four weeks, it was concerned that developments between meetings might alter appropriate reserve provision. Consequently, in 1966 it introduced a "proviso clause" that set conditions under which the desk might modify
the approach adopted at the preceding meeting. Bank credit data
still were available only with a lag. After some experimentation, the
FOMC adopted what it called the bank credit proxy, consisting of
daily average member bank deposits subject to reserve requirements. If the bank credit proxy moved outside the growth rate
range discussed at the FOMC meeting, the desk would generally
adjust the target level of free or net borrowed reserves modestly.29
Sometimes the proviso clause permitted either increases or
decreases in the objective for free reserves. Frequently it allowed
adjustments only in one direction.
Targeting Money Growth and the Federal Funds Rate: 1970 to 1979
The inflationary pressures that began in the late 1960s led to a
number of policy initiatives in the early part of the 1970s. Inflation
in the United States encouraged outflows of official gold holdings
and made the Bretton Woods system of pegged exchange rates pro-




28 Chapter 8 discusses the various hypotheses being developed to explain how monetary
policy works.
29 Logically, the bank credit proxy, which represented most of the liability side of the banks'
balance sheets, should have moved in a similar fashion to bank credit, which was a large
share of the asset side of their balance sheets. But they often differed, primarily because
of the growing use ofnondeposit liabilities to finance credit extension. In 1969, the bank
credit proxy was expanded to include liabilities to foreign branches, the largest nondeposit liability. Nonetheless, the proxy continued to deviate from bank credit when reserve
ratio changes made bank assets and liabilities diverge.

38

gressively less viable. In 1970, the Federal Reserve formally adopted monetary targets with the intention of using them to reduce inflation gradually over time. In August 1971, the Nixon Administration
froze prices and wages and suspended gold payments. 30 The
Administration's actions on gold effectively brought the Bretton
Woods exchange rate system and the last remnant of the gold standard to an end. Over the next two years, the industrialized countries
moved toward floating exchange rates. The official price of gold was
raised in two steps to $42.22 a troy ounce by 1973, but because the
Treasury did not make purchases or sales, the price ceased to have
any role in constraining growth in money or inflation.
The techniques for setting and pursuing money targets developed
gradually, with frequent experimentation and modification of procedures taking place in the first few years of the 1970s. Nonetheless,
until October 1979 the framework generally included setting a monetary objective and encouraging the Federal funds rate to move
gradually up or down if money was exceeding or falling short of the
objective. The Federal funds rate, as an indicator of money market
conditions, became the primary guide to day-to-day open market
operations, and free reserves took on a secondary role. An increasingly active market for Federal funds made the funds rate a feasible
target, and the passage of time reduced the association of interest
rate targeting with the rate pegging episode of the 1940s.
Bank credit and its proxy remained for a while in the list of subsidiary intermediate targets, but they received decreasing attention.
Free reserves served as an indicator of the volume of reserves
needed to keep the Federal funds rate at the desired level. The desk
used the forecasts of reserve factors to gauge the appropriate direction and magnitude for open market operations.
The FOMC selected growth targets for Ml—and to a lesser extent
for M2—that evolved into two-month growth rate ranges that used
the month before the FOMC meeting as a base.37 The FOMC directed
the staff to develop estimates of monetary aggregate growth aimed
at gradually reducing inflation. In 1972, it introduced six-month
30 During the next few years, the government imposed a variety of wage-price controls,
which had the effect of creating shortages and distorting various price indices. It also created a Committee on Interest and Dividends that restricted interest rate increases and thus
distorted financial market activities.
31 At the time, Ml consisted of currency and privately held demand deposits at commercial
banks. Other checkable deposits at commercial banks and transactions deposits at thrift
institutions were added to the definition in 1980. M2 consisted of Ml plus time and savings
deposits at commercial banks other than large CDs. Thrift institution deposits, overnight
RPs, Eurodollars, and money market funds were not included until 1980.




39




growth targets designed to achieve that goal. Econometric models,
supplemented by the judgments of the staff, were used to develop
the six-month and one-year estimates. The estimates assumed that
the demand for money depended on economic activity and interest
rate behavior. The weekly and two-month estimates were derived
judgmentally, allowing for a range of technical factors.
The FOMC instructed the desk to raise the Federal funds rate
within a limited band if the monetary aggregates were significantly
above the desired growth rates or to lower the funds rate within that
band if the aggregates were below them. The procedure required
the staff to estimate what funds rate would achieve desired money
growth. The funds rate worked by affecting the interest rates banks
both paid and charged customers and hence the demand for
money. The Board staff built models of money demand, as did
other Federal Reserve research departments.
In 1972, the FOMC addressed criticisms of its efforts to control
money from the demand side. It introduced a supplemental reserve
operating mechanism to influence money from the supply side.
The development of a reserve guideline was based on the reservemoney multiplier model. The model implied that controlling total
or required reserves would constrain money growth through the
operation of the reserve requirement ratio. The FOMC was concerned, however, that a pure reserve provision strategy would
cause undesired short-run volatility of interest rates.
The FOMC briefly tried reserve targeting in 1972. Because it
feared that reserve targeting would raise the volatility of interest
rates to unacceptable levels, it also constrained the funds rate. The
reserve measure targeted was called reserves on private deposits,
or RPD. It excluded reserve requirements on government and interbank deposits that were not in the money definitions.32 The linkage
between RPD and Ml was not very close because bank reserve
requirement ratios differed widely according to size and membership status. Using staff estimates of the various ratios, the FOMC
set two-month growth target ranges for RPD designed to be consistent with the desired growth in Ml; it then instructed the desk to
alter reserve provision in a way that was intended to achieve them.
As it turned out, relatively narrow funds rate constraints often dominated, and the desk frequently missed the RPD target. RPD targets
32 Government deposits at the time varied far more than they have in recent years. All tax
and loan account monies kept in commercial banks were subject to reserve requirements
until 1977, when a legal change introduced note option accounts that pay interest and
are not subject to reserve requirements.

40

were considered to be unachievable, although the funds rate constraint precluded a true test. In 1973, RPD changed from an operational target to an intermediate target, taking its place with Ml
and M2. Since information on the behavior of Ml was about as
good as information on RPD, RPD gradually fell into disuse. It was
dropped as an indicator in 1976.
The monetary targets were modified further in 1975 in response to
a congressional resolution. The Federal Reserve adopted annual target ranges and announced them publicly. A growth cone was drawn
from the base period, which was the calendar quarter most recently
concluded. Every three months, the target range was moved forward
one quarter. The procedure meant that by the time a given annual target period was completed, the original target had long since been
superseded. Frequently, the targets were overshot, and complaints
about upward "base drift" were legion. The Full Employment and
Balanced Growth Act of 1978, known as the Humphrey-Hawkins Act,
established the current procedure requiring the Federal Reserve to set
monetary targets for calendar years and to explain any deviations.
During most of the 1970s, the FOMC was reluctant to change the
funds rate by large amounts at any one time, even when staff estimates suggested that sizable modification was necessary to
achieve the two-month or annual monetary goals. Part of that
reluctance reflected a wish to avoid short-term reversals of the
rate. Keeping each rate adjustment small minimized the risk of
overdoing the rate changes and then having to reverse course.
These priorities meant that the FOMC was handicapped at times
when it sensed a large rate move might be needed but was uncertain about its size. The adjustments in the funds rate often lagged
behind market forces, allowing trends in money, the economy, and
prices to get ahead of policy.
At the FOMC meetings, the Committee frequently voted for a
funds rate range that surrounded the most recent rate target. The
Committee also put relatively narrow limits on the range of potential
adjustments that could be made between meetings if money growth
went off course. In the early 1970s, the width of the intermeeting
funds rate range was generally 5/8 to 1 1/2 of a percentage point.
By the latter part of the decade, the width narrowed to about 1/2 to
3/4 of a percentage point, and on a couple of occasions to only 1/4
of a percentage point. In addition, the specifications for the monetary aggregates were often set in a way that made it likely that the
funds rate would be adjusted in one direction only, effectively cutting the range in half.




41




In implementing the funds rate targeting procedure, the desk
became increasingly sensitive to preventing even minor short-term
deviations of the funds rate from target. It generally added reserves by
purchasing securities or arranging RPs in the market in a visible way
when the funds rate exceeded the objective even slightly, and it
absorbed reserves through sales or matched sale-purchase agreements when the funds rate fell short of the objective. It felt some constraint not to make reserve adjustments in an overt way when the
funds rate was on target. At times when reserve estimates suggested
that a large adjustment was needed but the funds rate did not confirm
it early in a statement week, the desk would worry about delaying the
reserve adjustment and having to make an unmanageably large open
market transaction late in the week. When the funds rate failed to confirm an estimated reserve excess or shortage, the desk often made the
reserve adjustments by arranging internal purchases or sales with foreign accounts that could not be observed by market participants. The
introduction in 1974 of customer-related RPs— agreements on behalf
of official foreign accounts—gave the desk a tool for adding reserves
when the funds rate was on target but a reserve need was projected.33
(Market participants routinely assumed that outright transactions in the
market for customers did not signal dissatisfaction with the funds rate,
and they initially regarded customer-related RPs the same way.)
If the estimated need to add or drain reserves was too large for
these techniques, the desk often pounced on very small funds rate
moves off target to justify an open market operation. For instance,
when estimates suggested that additional reserves were needed, the
desk would often enter the market to arrange an RP when the funds
rate rose 1/16 of a percentage point above the preferred level. But
if the funds rate fell despite the estimated need to add reserves, the
desk typically would allow a 1/8 percentage point deviation to
develop before it would arrange a small market operation to drain
reserves. If the funds rate continued to trade off target after the
desk's first entry of the day, the desk often arranged a second open
market operation. There were operational limits to how late in the
day transactions could be made to achieve a reserve effect on the
same day. The cutoff was around noon for outright bill operations.
It was supposed to be 1:30 p.m. for temporary transactions, but if
the desired funds rate move occurred just after that time, the desk
often responded if it was anxious to conduct an operation. The end
of its operating time was close to 2:00 p.m. by 1979.
33 Chapters 6 and 7 describe the various policy tools and how they affect reserves.

42

The desk's prompt responses to even small wiggles in the Federal
funds rate led banks to trade funds in a way that tended to keep the
rate on target. Except near day's end on the weekly settlement day,
a bank short of funds would not feel the need to pay significantly
more than the perceived target rate for funds. Likewise, a bank with
excess funds would not accept a lower rate. Rate moves during the
week were so limited that they provided little or no information
about reserve availability or market forces. Probably few, if any, in
the Federal Reserve really believed that brief, small moves in the
funds rate were harmful to the economy. The tightened control
developed bit by bit without an active decision to impose it.
Targeting Money and Nonborrowed Reserves from 1979 to 1982
In October 1979, Paul Volcker, who had recently become Chairman of the Board of Governors, announced far-reaching changes in
the FOMC's operating techniques for targeting the monetary aggregates. The acceleration of inflation to unacceptable rates over the
preceding decade inspired a change in priorities. Chairman Volcker
and other FOMC members realized that turning around these inflationary pressures, which had come to permeate economic relations,
would involve costs. Interest rates would have to rise significantly
beyond recent levels, although the extent of the increase could not
be determined in advance. Increased rate volatility was also likely to
accompany the efforts to halt inflation. The Federal Reserve's credibility with the public was low after previous efforts to slow inflation
had been followed by further price acceleration. Chairman Volcker
felt that only strong measures could rebuild public confidence.
Many analysts, both inside and outside the Fed, argued that using
the funds rate as the operational target had encouraged repeated
overshooting of the monetary objectives. They contended that
whatever measure was targeted was likely to be changed too slowly.
Partly in response to such arguments, the FOMC began to target
reserve measures derived to be consistent with desired three-month
growth rates of Ml. Reserve controls were expected to keep money
growth from persistently exceeding (or falling short of) the target
growth rate, although they would not prevent short-term deviations.
The limits on the Federal funds rate were applied only to weekly
averages, rather than to brief periods during the week as had been
common in the 1970s. A band 4 to 5 percentage points wide
allowed room for adjustments to achieve the monetary target.
Operationally, the FOMC chose desired growth rates for Ml (and
M2) covering a calendar quarter and instructed the staff to estimate




43




consistent levels of total reserves. The process resembled that used to
estimate RPDs. The staff estimated deposit and currency mixes to
derive average reserve ratios and currency-deposit ratios. The estimation technique employed a mix of judgment and analysis of historical
patterns. It was complicated by the wide range of reserve ratios
applied to Federal Reserve member bank deposits and by the absence
of reserve ratios, or even timely deposit data, from nonmember banks.
From the total reserve target, the desk derived the nonborrowed
reserve target by subtracting the initial level of borrowed reserves that
had been indicated by the FOMC.34 If money exceeded (or fell short of)
its path, total reserves would also exceed (or fall short of) their path.
Because the required reserves were predetermined, the desk had limited means to change total reserves within the reserve period. If the
desk only provided enough reserves to meet the nonborrowed reserve
objective, banks would have to increase (decrease) their borrowing
when money growth and total reserve demands were excessive (deficient).35 Because banks were still discouraged from making frequent
use of the discount window, the change in aggregate borrowing would
affect the ease of obtaining reserves and interest rates. It would
encourage the banks and the public to take actions that would accomplish the desired slowing or speeding up of money growth. If the pace
of adjustment implied by the mechanism did not seem appropriate,
instructions were occasionally given to accelerate or delay the adjustment to the borrowing objective. The FOMC could make alterations to
the basic mechanism at a meeting or direct the desk to make adjustments between meetings under specified conditions.
To reduce overweighting of weekly movements in money, the total
and nonborrowed reserve paths were computed for intermeeting
average periods or, if the intermeeting period was longer than five
weeks, for two subperiods. (In 1979 and 1980 the FOMC met 9 and
11 times, respectively; in 1981 it moved to the schedule of 8 meetings
a year in use today.) A consequence of this averaging technique was
that reserve target misses in the early part of the intermeeting period
had to be offset by large swings in borrowing in the final week.
Informal adjustments were sometimes made to smooth out these
temporary spikes or drops in borrowing that were deemed inconsis34 The Board staff made estimates of consistent combinations of borrowed reserves and
money growth for the given discount rate. The estimates were derived from modified versions of money demand models and borrowed reserve equations.
35 The scope for adjusting excess reserves was very limited since banks at the time held only
minimal levels of excess reserves. The relationships among reserve measures and the
effects of these measures on bank behavior and monetary growth are discussed more
extensively in Chapter 6.

44

tent with the longer term pattern. Although the adjustments were
considered necessary to avoid severe short-term swings in reserve
availability and interest rates, they gave the appearance of "fiddling"
and have caused considerable confusion for outside observers. Each
week the total reserve path and actual levels were reestimated using
new information on deposit-reserve and deposit-currency ratios.
In implementing the policy, the desk emphasized that it was targeting reserves and not the Federal funds rate by entering the market
at about the same time each day—usually between 11:30 and 11:45
a.m.—to perform its temporary operations. It confined outright purchases or sales to estimated reserve needs or excesses extending
several weeks into the future. It arranged outright operations early in
the afternoon for delivery next day or two days forward. The Federal
funds rate was not ignored; it was used as an indicator of the accuracy of reserve estimates, although it was not always very reliable.
On the margin, it could accelerate or delay by a day or so an operation to accomplish a needed reserve adjustment, but its role was
much diminished.
It had been anticipated that the new procedures would lead to
considerably wider short-term swings in the Federal funds rate,
although the actual changes exceeded most expectations and were
accompanied by greater variation in money growth rates as well.
In part, the sharp movements in both interest rates and money may
have reflected the difficulties in reversing strongly held beliefs that
inflation had become a permanent phenomenon. Expectations
about inflation and economic activity were in the process of being
reshaped, with many people uncertain whether a new lower inflation pattern would emerge or whether the inflation slowdown would
be a temporary pause on the way to even higher rates. In this environment, people evaluated new information and judged whether
the anti-inflation policies were likely to succeed. Some of the interest rate moves came in response to changes in expectations.
The control mechanism itself also appeared to play a role in the
variation of money growth. It forced borrowing to rise whenever
money was above the desired level. Consequently, following the
procedure caused borrowing to rise until money was back on target.
Since there were lags in the adjustment of money to borrowing pressures, money continued to weaken even after borrowing stopped
rising. The result appeared to be a "damped cycling process."
These years also saw major regulatory and legislative changes that
affected the climate for Federal Reserve policy. In 1980, the Congress
passed the Depository Institutions Deregulation and Monetary Control




45




Act (MCA), which simplified the structure of reserve requirements and
extended the requirements to nonmember commercial banks and
thrift institutions with transactions deposits.36 The MCA provided for
gradual elimination of interest rate ceilings on all but demand deposits.
(It permitted interest-bearing consumer transactions deposits, called
NOW accounts, to be extended to the states outside the Northeast.) In
1982, the Garn-St Germain Act modified the MCA reserve requirements and established money market deposit accounts (MMDAs),
which were free of interest rate and maturity restrictions. That same
year, the Federal Reserve Board announced plans to introduce contemporaneous reserve requirements on transactions deposits in 1984.
(These requirements are described in Chapter 6, Box A.)
Several motives underlay these changes. Deposit interest rate
restrictions and reserve requirements were particularly burdensome
when inflation and market rates were high. Numerous state-chartered banks had dropped their Federal Reserve membership, and
largely unregulated nonbank institutions were competing for consumer funds. Shifts in bank and customer behavior caused the
meaning and behavior of money to change. The Federal Reserve
attempted to deal with the behavioral and regulatory changes by
redefining the monetary aggregates in 1980 to include instruments
and depository institutions that were growing in importance. It also
used "shift adjusted" aggregates in setting Ml targets in 1981 in an
attempt to allow for effects on money demand from the introduction
of nationwide NOW accounts. These adjustments, however, did not
fully offset the money demand shifts.
As evidence mounted that the relatively close linkage between Ml
and economic activity had broken down, the FOMC suspended its
Ml target in late 1982. It had become apparent that the demand for
Ml had strengthened relative to income, so that growth within the
target range would have been more restrictive than seemed desirable. Some of the increase in the demand for money was attributed
to the popularity of NOW accounts included in Ml. In addition, the
maturing of a large volume of special tax-favored "all savers"
deposits that October was expected to add substantially to Ml holdings. The FOMC hoped that M2 would continue to be a reliable indicator, and for a few months at the end of 1982 it attempted to use it
as a guide to building total and nonborrowed reserve targets. But
MMDAs, first offered in December, proved very attractive, and the
demand for M2 rose sharply.
36 As Chapter 1 indicated, the structure of reserve requirements and a schedule for the transition were specified in the act

46

Monetary and Economic Objectives with Borrowed Reserve Targets:
1983 to the Present
In the absence of a stable relationship between money and economic activity, the FOMC modified its procedures for guiding reserve
provision. The FOMC focused on measures of inflation and economic activity and placed less weight on the monetary aggregates.
Instead of computing total and nonborrowed reserve levels linked to
one of the monetary aggregates and deriving a level of borrowing
that moved with the deviations of the aggregate from target, it targeted the borrowed reserve level directly. It intended to adjust it up
or down whenever money seemed to be deviating from the desired
growth path in a meaningful way. In deciding whether an adjustment
was appropriate, the FOMC would allow for any known distortions to
the aggregates and would consider supplemental indicators.
The monetary aggregates did not quickly resume their prior relationship with economic activity. Declining inflation made holding
money more attractive. Because rates on some components of Ml
were close to market rates but slow to change, interest rate sensitivity increased. Policy decisions continued to be guided by information
on economic activity, inflation, foreign exchange developments, and
financial market conditions. In time, money growth was moved from
a predominant position in the directive to join the list of factors shaping adjustments to the borrowing level.
The temporary policy procedures introduced in 1983 have persisted, with modifications, through most of the 1980s. The FOMC
has continued to set policy that is designed to be countercyclical,
but at the same time anti-inflationary. It has used a discretionary
approach that draws from some of the techniques developed in earlier decades. Particular tools of policy may have gained or lost
prominence, but all the tools now in use were developed and refined
over the years. The element of continuity in Federal Reserve policy
makes this brief history a fitting prelude to the extensive discussion
of current policy in Chapters 5 through 7. First, however, let us turn
our attention to two other subjects that bear on monetary policy at
the end of the decade—the structure of the U.S. banking system and
the financial markets.




47




The Role of
Depository Institutions
Depository institutions play a key role in the transmission of
monetary policy to the financial markets, to borrowers and depositors, and ultimately to the real economy. They hold a large share
of the nation's money stock in the form of various types of
deposits and provide for the transfer of those funds to effect the
payments that keep the economy functioning. Depository institutions also lend these funds directly to consumers and businesses
for a wide variety of purposes and lend them indirectly by investing in securities.
The United States has a wide variety of depository institutions —
commercial banks, savings banks, savings and loan associations,
and credit unions. Originally only commercial banks accepted
deposits upon which checks could be drawn, but during the late
1970s and early 1980s, checkable deposits developed at the other
institutions as well. The Depository Institutions Deregulation and
Monetary Control Act of 1980 (MCA) treated all depository institutions that accept checkable deposits similarly. Among depository institutions, commercial banks are still a major force in
commercial deposit-taking and lending activities although their
share of the business has dropped considerably.
The structure of the U.S. banking system, with many institutions
in a range of sizes, reflects U.S. banking traditions. Until recently,
depository institutions were permitted to have offices only in one
state. During the last few years, however, a majority of the states
have relaxed restrictions on interstate operations, prompting many
depository institutions to expand operations outside of their home
state. At the same time, institutions have merged, particularly thrift
institutions in areas of the country where regional problems or
overexpansion have created financial difficulties. In addition,
multistate or regional bank holding companies that are nearly as
large as the major money center banking organizations have been
formed. Despite these changes, the United States continues to
have many more depository institutions than other countries. At
the end of 1988 there were about 33,600 depository institutions, of
which 13,400 were commercial banks. Some of them are large
multifaceted organizations that attract deposits from and make
loans to a wide range of customers, while others specialize in corporate or retail activities.
For many years, commercial banks were unique in conducting
all types of banking. Thrift institutions — savings banks and
savings and loan associations — provided selected banking services for individuals, primarily savings accounts and mortgage

48

loans.7 Over time, the powers of thrift institutions have been expanded to overlap those of commercial banks. Nonetheless,
although thrifts have aggressively accepted checkable deposits
from individuals, to date most of them have entered the business of
commercial lending and deposit taking only in a very limited way.
Commercial banks still handle the bulk of the myriad commercial transactions that take place each day. They also hold most of
the reserve balances at the Federal Reserve Banks and play a
major role in intermediation. In addition, they are responsible for
the lion's share of large dollar payments over Fedwire, the Federal
Reserve's electronic transfer network.2 Thus, as the Federal
Reserve fashions and implements policy, it must stay closely
attuned to commercial bank behavior. Those involved in the policy
process must understand the dynamics of the financial system
generally, the activities of the major institutions involved, and the
manner in which these institutions respond to the day-to-day
demands made on them by the other participants in the economy.
Such an understanding of the role of banks prepares policymakers
to assess the linkages between monetary policy and growth in
money and credit. More specifically, a knowledge of the circumstances and behavior of individual institutions enables the open
market desk to evaluate the reserve situation knowledgeably as it
devises its operating strategy.
The Business of Banking
While the essence of banking—borrowing and lending money —
remains essentially what it was in ancient times, the nature of the
business in the United States has undergone dramatic change in
recent years. Deregulation is sometimes cited as the principal catalyst of the changes. Deregulation may be more appropriately
regarded, however, as an outgrowth of the competitive pressures
that have increasingly impinged on the banking franchise. Broader
access to the money and capital markets, information, and technology has irrevocably altered the competitive landscape.
Banks historically have had a comparative advantage in acquiring the information crucial to credit analysis and thus in making
informed credit judgments. This advantage has diminished in
1 Credit unions provide services similar to those offered by thrift institutions; their clientele
typically consists of members of affinity groups, such as people employed by a particular
corporation or members of a union.
2 "Large-Dollar Payment Flows from New York," Federal Reserve Bank of New York Quarterly
Review, vol. 12, no. 4 (Winter 1987-88), pp. 6-13.




49




recent years with increasingly broad dissemination of information.
Computer-aided analytical techniques for investors borrowing
directly in the money and capital markets have become accessible
to more and more businesses. Moreover, technological developments have greatly enhanced the cash management and funding
sophistication of the banks' traditional client base. So the enduring
relationships between banks and their customers that of necessity
used to form the basis for profitable banking have yielded to pricesensitive competition and a more fickle clientele.
Growing numbers of large corporations have turned to the commercial paper market for working capital financing. In response,
the banks have developed a role for themselves in that market by
providing backup credit lines to commercial paper issuers and
placing the paper as agent for the issuer. More recently, nonbank
affiliates of banks have begun underwriting commercial paper and
other corporate debt in competition with securities firms as a
result of expanded authority granted by the Federal Reserve, the
agency responsible for the supervision and regulation of bank
holding companies.
On the liability side of the bank's ledger has come competition
for both commercial and retail deposits. In the high interest rate
environment of the 1970s, competitors of commercial banks,
including investment banks and brokerage firms that managed
stock and bond mutual funds, began to offer savers an alternative
to time and savings deposits in the form of money market mutual
funds (MMMFs). The fund managers pooled small sums gathered
from many customers and placed them in short-term market
instruments—primarily commercial paper and Treasury securities—and in large CDs that were exempt from interest rate ceilings.
Consequently, the MMMFs were able to offer market-based rates
along with easy access through limited check writing privileges.
The volume of MMMFs expanded rapidly whenever interest rate
ceilings restricted the rates banks and thrifts could pay.
In response to this development, bank laws and regulations were
changed. Restrictions on interest rates that could be paid by
depository institutions on most types of deposits were removed
gradually, enabling the institutions to offer a directly comparable
product. In December 1982, banks and thrifts were allowed to offer
money market deposit accounts (MMDAs) paying competitive
interest rates on small sums that were immediately withdrawable
and federally insured. Depository institutions were thus able to
attract the funds that they needed by paying attractive rates. Many

50

Chart 1 Money Market Mutual Funds (Quarterly Average Assets)
Billions of dollars
350

292

233

175

117

58

1974 '75 '76 77

'78 '79 '80 '81 '82 '83 '84 '85 '86 '87 '88

large banks were able to reduce their issuance of wholesale
deposits as they acquired more consumer deposits; at the same
time, the volume of MMMFs fell sharply (Chart 1), reducing the
demand for such deposits. Once MMDAs grew to the volume the
banks desired, banks retreated from offering high promotional
rates, often dropping their rates well below those on MMMFs.
MMMFs then resumed their growth because they were able to
attract rate-sensitive liquid funds. Banks had the flexibility to compete when they desired to do so, but had to pay for the privilege.
Often they competed aggressively in their offerings of specific
maturities of consumer time deposits while holding down rates on
the more widely held MMDAs.
The one remaining area where the rates payable by banks and
thrifts are restricted is demand deposits, the only type of checkable deposit that may be offered to commercial customers. By
law, explicit interest may not be paid on demand deposits. Even
these funds involve costs to the banks, however, because banks
must compete for them by offering other services to corporate
customers. At the same time, demand deposits are a shrinking
share of bank balance sheets, because of the many substitutes
available and the development of techniques for closely managing
demand deposits.




51




Deregulation has not been the only response to changing conditions. Regulatory capital requirements have been strengthened in
recognition of the risks inherent in innovation and deregulation. At
least initially, the requirements may have increased the costs associated with banking for some institutions. Beginning in 1981, the
federal bank supervisory agencies (the Federal Reserve, the
Federal Deposit Insurance Corporation, and the Office of the
Comptroller of the Currency) have systematically raised the minimum requirements for bank capital-to-asset ratios. This development, in turn, encouraged banks to move business off their balance
sheets. The banks increased fee-generating activities and contingent exposures (for example, standby letters of credit) rather than
balance sheet assets against which capital had to be held. In
response, regulators have further modified capital requirements to
include some contingent liabilities. Banks have also begun to package and sell loans in the form of securities. This "securitization" of
assets — mortgages, auto loans, credit-card loans, and so forth —
shifts bank loans to permanent investors, leaving the banks to service the loans for a fee. Thus, origination, distribution, and servicing
capabilities have become increasingly significant elements of banking, while building up the balance sheet—once perceived as a measure of a bank's eminence—has diminished in importance.
In 1988, U.S. bank regulators moved to risk-based capital
requirements as part of an international agreement among the
Group of Ten (G-10) countries. The risk-based capital standards
differentiate among risk categories of assets, establish capital
requirements for each category, and require capital to be held
against off-balance sheet exposures. The new measures of capital
adequacy demand higher levels of capital, particularly for the large
banking organizations heavily involved in off-balance sheet business. Thus the shift to risk-based requirements is likely to alter further the nature of the business.
The upshot of these developments has been a greater premium
on flexibility and innovation. Banks have lost their essentially captive markets for "rate-controlled" deposits and for loans. Now they
must compete for market-priced liabilities and a wider variety of
lending and investment products and services. These changes call
for a more dynamic view of the balance sheet, increased levels of
capital, and the expansion of fee-based operations.
In response, the banking industry has become more diverse. The
cost of being all things to all people has grown more difficult to
finance with thinning profit margins, and most institutions have

52

looked to specialize or to diversify into potential growth areas. The
broad traditional distinction between "wholesale" and "retail"
banks has been further refined. Some banks have withdrawn
entirely from retail or branch banking to concentrate on serving
corporate clients. Others have seen their advantage in the consumer sector and have expanded that part of their business. Some
have pulled back from international operations while others have
expanded abroad, and several larger organizations have taken to
arranging and financing mergers and acquisitions in direct competition with investment banks. Many smaller institutions, on the
other hand, have managed to retain some of the traditional character of full-service banks by serving a geographically limited
clientele. In these cases, the banks' knowledge of their local communities and their relationships with depositors and borrowers are
what distinguish them from the competition.
The U.S. activities of foreign bank branches and agencies have
remained largely wholesale oriented, focusing for the most part on
the money markets, foreign exchange, and trade finance. But over
the past decade foreign banks have also tried to establish a more
broadly competitive presence in the U.S. corporate banking business. They have expanded their U.S. operations by establishing new
banks or by acquiring existing U.S. banks. Since 1980, the number
of foreign-owned U.S. banks has approximately doubled. In addition, foreign banks have increased their securities-related activities
in the United States alongside their American counterparts.
While the distinctions among banking firms have grown, banking more generally has increasingly overlapped other financial
industries such as the securities and insurance businesses in the
services it offers. Banking organizations in recent years have
established and marketed mutual funds (albeit technically limited
by existing law to trust customers), packaged their loans and sold
them in the form of securities to investors, entered the bond guaranty insurance and securities brokerage businesses, and begun to
underwrite and trade corporate debt through their nonbank affiliates. Currently, the Federal Reserve is reviewing whether to permit
banks to underwrite corporate stock as well.
Securities firms and insurance companies, conversely, have successfully offered deposit-like products to consumers and businesses and provided financing for corporate expansion. Like the banks,
they are testing the bounds of current law and regulation that generally prohibit affiliation between commercial banks and full-service
securities firms. Security firms have established or acquired spe-




53

cial-purpose banks such as Edge Act corporations (limited to an
internationally oriented business) and nondepository trust companies in order to obtain access to Federal Reserve services. These
developments represent significant inroads into banking's province
—the ability to maintain accounts at the Federal Reserve Banks and
to have direct access to the electronic payments system.
Banking Risks




Because of the crucial importance of depository institutions to
the economy, they are supported by a Federal "safety net," composed of the discount window, federal deposit insurance, and an
extensive framework of supervision and regulation. Other types of
financial firms, such as securities houses and insurance companies, are also heavily regulated and supervised, and their investors
and beneficiaries too are protected by pooled guaranty funds
against a company's failure. But only depository institutions have
access to central bank liquidity to guard against the risk that the
failure of one institution to settle its obligations on a given day will
cause other institutions to default in turn. The potential social costs
of a crisis of confidence in the banking system and the likely related money and credit dislocations are certainly large enough to
warrant such safeguards. In the extreme, a banking crisis could
bring the economy to a halt.
This framework of support does not imply, however, that the
business of banking is without risk. Indeed, as we have observed,
growing competitive pressures and deregulation have introduced
new types of risk and made the business much more complicated
in recent years. The safety net is intended to protect the system as
a whole, not individual banks. Still, these protections against systemic risk to some extent protect the participating institutions as
well, creating the "moral hazard" that banks will take excessive risk
in reliance on federal support. Bank supervision and regulation,
therefore, look to minimize these moral hazards while maximizing
systemic protection by letting the discipline of the marketplace
work as much as possible. Investors in bank or thrift stocks are certainly at risk, as are other creditors including uninsured depositors.
Banks perceived to be risky may find it difficult to raise capital. But
regardless of who in the end bears the cost — bank investors, the
federal support system or depositors — the basic risks to bank solvency remain.
These basic risks assume a variety of forms.3 Credit risk, per3 The rest of this section focuses on the diversified business of commercial banking, although
(continued on page 55)

54

haps the most notable form, centers on the possibility that a bank's
customer will be unable to meet the interest and principal payments on a loan. One of the principal functions of bank credit officers is to assess the borrower's financial condition and evaluate the
risk and return characteristics of the loan. To an extent, loan losses
are unavoidable. Among U.S. investments, only U.S. Treasury
securities are truly free from credit risk on the part of the issuer
since they are backed by the full faith and credit (that is, the taxing
power) of the federal government. However, since banks must pay
more than the (J.S. government for a large portion of their liabilities, an investment strategy free of credit risk would not be very
profitable. Rather, bankers generally look for the higher yields that
can be obtained from relatively riskier loans and investments.
In managing credit risk, banks attempt to maintain a diversified
portfolio priced both to absorb expected losses and to earn a satisfactory return on capital. The portion of a bank's capital that can
be lent to a single borrower is limited by law, and bank credit
departments typically establish even more restrictive internal limits for specific borrowers. Moreover, the financial condition of the
borrower is monitored on an ongoing basis as long as the loan or
commitment is outstanding.
Collateral also plays a role in the management of credit risk.
Important considerations are the liquidity of the collateral and the
coverage (margin) of collateral value in excess of the amount outstanding on the loan. Clearly, the ability to foreclose on a property
or a piece of machinery can be cold comfort to a bank if it cannot
readily sell the collateral in the market at a price that will cover the
balance due on the loan and the bank's related costs. In most
cases, the cost of managing, insuring, and maintaining the collateral pending its sale must also be considered, as must the risk that
a borrower will seek protection of the courts under the bankruptcy
laws. In the latter instance, the bank may not be allowed to liquidate the collateral at all.
A second form of risk facing banks is interest rate risk. Over the
past decade, this risk has grown in importance as interest rates
have become less regulated and more volatile. A bank's profitability can be heavily dependent on the direction of interest rate movements. In fact, because basic lending spreads have narrowed
substantially in recent years, a bank might find it difficult to cover
some of the discussion could be applied to thrift institutions. Because the thrifts specialize in
consumer deposit taking and real estate loans, their risk exposures and strategies for managing risk differ from those of commercial banks.




55




its overhead by running a "matched book" of assets and liabilities
with the same repricing dates or duration.4 In such cases, movements in interest rates will not affect the bank's profitability
because the rates paid on liabilities will change in lockstep with
those earned on the loans and investments they support. But the
earnings from such a strategy may not be sufficient both to cover
operating expenses and to make a profit for the shareholders.
Consequently, most banks mismatch or "gap" the repricing of
their assets and liabilities to some degree with a view toward profiting from changes in the level of rates or in the shape of the yield
curve. For instance, borrowing short and lending long may be profitable in an environment of falling interest rates because liabilities
can be repriced at lower rates while assets lock in relatively high
yields. Such a strategy may also be profitable if rates are stable and
the yield curve has a "normal" upward slope. Banks generally vary
their interest rate risk gaps in different maturity sectors if they
expect changes in rates over time, but "bets" on interest rates are
typically kept relatively small given the inherent difficulties of forecasting rates and the high cost of being wrong.
Liquidity risk, a third type of banking risk, turns on the bank's
ability to meet unexpected demands for cash in the form of withdrawals, funds transfers, or drawdowns of credit lines. In managing
its liquidity, a bank must balance the cost of holding cash and
money market instruments against its ability to obtain funding in
the market, or if necessary, from the discount window. Sales of
longer term assets are another possible source of liquidity.
Historically, banks have been reluctant to incur the capital losses
that at times may accompany such sales, while the realization of
capital gains has not always impressed regulators or bank analysts. In recent years, however, banks have turned increasingly to
the asset side of their balance sheets for liquidity, making growing
use of securitized assets and loan sales.
4 Duration is a more sophisticated measure than simple repricinggap measures for assessing
interest rate mismatches in that it takes into account the timing of the cash flows involved.
Duration weights the present value of annual cash flows by their term to maturity so that
near-term payments get proportionally greater weight and work to shorten the duration of
an instrument compared to its nominal maturity. Thus, a loan that pays interest and principal monthly always has a shorter duration than its maturity, while duration and maturity are identical for loans that pay both interest and principal only at maturity. The
duration concept is discussed extensively in Gerald O. Bierwag, Duration Analysis
(Cambridge, Mass.: Ballinger Publishing Company, 1987). It is also examined in a series of
American Banker articles by Sanford Rose. Key articles include those dated March 22,
1983; January 25, 1984; May 22, 1984; June 1, 1984; July 3, 1984; December 4, 1984;
December 11, 1984; February 5, 1985; and August 27, 1985.

56

Financial innovation has significantly affected the way in which
banks manage liquidity and interest rate exposures. Increasingly
banks can address liquidity and interest rate issues separately
because they can avail themselves of derivative instruments —
futures contracts, interest rate swaps, or options on (J.S. Treasury
securities, Eurodollars, and other primary instruments — and
dynamic hedging techniques that use these tools to alter hedges as
the rate relationships change. Through the use of such instruments
as futures and options contracts as well as interest rate swaps
(described in Chapter 4), banks can synthetically alter their interest rate exposures within a given funding profile, although they
may incur new risks in the process.
On the international side, banks face foreign exchange risk and
country risk (or sovereign risk). The first type of risk has long been
a major concern in international banking. Banks make markets in
foreign exchange and hold assets and liabilities denominated in
various currencies. Thus, they are exposed to gains or losses from
relative movements in exchange rates.
Country risk (or transfer risk) relates to the possible difficulties
in collecting from borrowers in another country as a result of some
development in that country. For example, a revolution or coup in
a foreign country may overthrow the government that took out a
(J.S. bank loan and bring in a successor government that repudiates the loan. Recently, the more common form of country risk has
arisen when public and private sector borrowers in less developed
countries (LDCs) borrowed heavily in dollars from the international
banking community and found it difficult to generate sufficient dollars to service their loans. In the mid to late 1970s the burgeoning
revenues of oil-exporting countries were recycled by banks in
industrialized nations in the form of loans to LDCs. Conventional
banking wisdom had been that countries do not default on their
obligations. Default would cut them off from future access to international credit and seriously hinder further development. But that
assumption did not recognize that the size of the country's debt
could overwhelm its ability to accumulate the dollars necessary to
service that debt. As foreign governments found it necessary to
defer loan payments, declare moratoria on debt service, and negotiate reschedulings that extended repayment terms, (J.S. banks'
cross-border exposures became a focus of attention for management, regulators, analysts, and investors.
Inherent in the management of country risk are assessments of
the local political situation, local economic conditions, and balance-




57




of-payments prospects. Beginning in the early 1980s, the Federal
Financial Institutions Examination Council, a joint body of the three
federal bank supervisory agencies, began to evaluate and monitor
the cross-border risk of public and private sector debt in particular
countries to permit consistent treatment of such debt in banks' loan
portfolios. The International Lending Supervision Act of 1983 also
provided a statutory basis for the Federal bank supervisors to direct
their information gathering and supervisory responses directly at
transfer risks. The supervisory agencies established criteria, comparable to those applied to domestic loans, classifying loans to foreign private or public sector borrowers according to their degree of
transfer risk. The agencies also stipulated that if a loan was particularly questionable, it might have to be accounted for on a cash
basis, meaning that interest would not accrue to the bank's income
statement as it was earned but would only be recorded when paid.
At some stage, the bank might also have to put aside reserves
against possible losses, in effect recognizing the lessened value of
the loan because of repayment risk.
Prompted by these concerns, the international banking community has sought to reduce its exposures to country risk while recognizing its interest in providing sufficient dollar funding to LDCs
to give the countries' internal economic adjustments an opportunity to work. Some of the smaller banking organizations have sold
off their entire LDC loan portfolios and exited the business, but the
largest international lenders have sold off only portions of their
loans. To protect against the risk inherent in the remaining loan,
the large lenders have raised additional capital and renegotiated
terms on existing debt while providing additional funding to the
LDCs in cooperation with multinational organizations such as the
International Monetary Fund and the World Bank.
Finally, with the increasing globalization of financial markets and
the rapid movements of huge volumes of funds and securities, payment and settlement risks have also emerged as key concerns. For
instance, an institution that fails to receive a wire transfer of funds
it had been expecting could be forced to acquire the funds in the
market or at the discount window. Alternatively, it might itself fail
to make a payment for which it had relied on the receipt of the
funds. Securities, too, may not be delivered to a buyer when
expected. The buyer, in turn, might not be able then to deliver them
to someone else. The implications of such problems for the liquidity of particular institutions or even of the system as a whole are
significant. Moreover, the fact that the underlying transactions

58

often occur across international boundaries raises the prospects of
dislocations in one market being transmitted globally.
Accordingly, there has been a concerted effort to manage such
risks more explicitly by monitoring exposures to particular counterparties and clearing systems and an increased interest in the
netting of transactions and in other mechanisms to reduce the
implications of a given problem.
Strategic Considerations
In most banks, the overall management of these various risks is
highly centralized. Central control is necessary to prevent fundamentally different strategies from offsetting one another to the
detriment of a bank's overall profitability. Typically, the central
body that sets the bank's strategic direction is the Asset and
Liability Committee or the Sources and Uses Committee. Senior
officers on this committee represent areas responsible for the
major elements of the bank's business—loans, investments, and
funding. In addition, the bank's chief economist normally plays an
important staff role on the committee, providing forecasts of the
real economy, interest rates, and monetary policy that are crucial
to the institution's strategic planning.
The committee meets periodically to review the bank's financial
position against the backdrop of the economic and market outlooks. Committee members give particular attention to recent
material changes in the consolidated global balance sheet and
expected future projects. The outlook for loan demand is reviewed,
both as it would flow from the firm's economic forecast and as it
would reflect particular business considered likely to develop over
the planning horizon. Upcoming maturities of assets and liabilities
are also reviewed, since they will generate funding needs and liquidity. Then committee members take up questions of pricing and
funding, considering in particular the implications for liquidity,
interest rate exposure, capital adequacy, and ultimately, expected
profitability. The members might, for example, decide whether the
bank should alter its asset allocation, enhance or reduce liquidity,
mismatch its book in particular maturity sectors, reduce its asset
size and hence its required capital, or raise equity or debt capital.
As noted earlier, banks generally take some position on the
direction of interest rates over periods of a few months or so. If, for
example, rates are expected to rise over three months and then
begin to fall, the bank might plan to be somewhat long funded up
to three months, so that its assets will reprice at increasing interest




59

rates while its fixed-term funding protects it against increasing
costs over the period. At the same time, it would try to be short
funded beyond three months to take advantage of expected
declines in financing rates while seeking to lock in higher returns
on its assets. Although some "gapping" of this sort is common at
most banks, such exposures are generally kept relatively modest
given the perils of interest rate forecasting and the attendant downside risk of "betting the store" on a particular outlook. On the other
hand, the relatively narrow profit margins inherent in simply
matching the maturities of assets and liabilities generally provide
an incentive to mismatch the book to some extent.
Tactical Considerations




Although the Asset and Liability Committee sets the bank's
overall strategy for managing assets and liabilities, the money desk
plays an important role in implementing the strategy in the market
day to day through their taking and selling of funds. In addition, the
money desk manages the bank's reserve position, making certain
that the bank's daily average reserve requirements are met on settlement day (every other Wednesday) with as little uninvested or
"wasted" excess reserves as possible. This task necessarily includes monitoring the flow of customer activity that affects the
bank's reserves. For instance, a wire transfer of funds for a customer or a deposit withdrawal that was not anticipated by the desk
could force the bank to replace those reserves in the Federal funds
market to avoid being overdrawn in its reserve account at day's
end. Or an unexpected inflow of reserves could provide the bank
with unwanted (and costly) excess reserves.
The money desk can vary the terms it sets on repurchase agreements (RPs) with dealers in government, municipal, and corporate
securities.5 Changing the posted dealer loan rate has long been
used by banks to adjust their asset positions. The money desk typically uses the bank's portfolio of U.S. government securities to
offer corporate and other customers overnight or term RPs.
Moreover, it sells certificates of deposit (CDs) for the bank and
keeps track of commercial paper sales by the parent bank holding
company. Other responsibilities include directing the acquisition of
Eurodollar funds for the head office through foreign branches and
carrying out the funding operations booked at branches in Nassau
and the Cayman Islands for tax or other reasons. In addition, the
5 The various financial market instruments mentioned are described in more detail in
Chapter 4.

60

money desk handles the funding of the bank's (J.S.-based international banking facilities. Through these facilities, the bank can conduct offshore business without incurring the cost of reserve
requirements or U.S. taxes.
In recent years, the funding process has witnessed tremendous
innovation. For instance, options, futures, and forward contracts
have become widely used to hedge exposures in the cash market,
and interest rate and currency swaps have emerged as vehicles for
synthesizing a particular profile. "Caps," "collars," and "floors"
have evolved in the derivative product markets. They allow interest
rate exposures to be shaped as desired. In all, the funding operation has become more complex; the depository institutions have
faced variable interest rates on an increased share of their balance
sheets and have had to manage the risks.
The officer supervising the money desk directs daily tactics
within a biweekly reserve strategy. Under the contemporaneous
reserve accounting regime in place since 1984, reserve requirements for transactions deposits over the two-week statement period ending on Wednesday are based on daily average deposits held
for the two weeks ending two days earlier.6 For time and savings
accounts, requirements are lagged, based on deposits held a
month earlier. Vault cash held a month earlier is counted toward
meeting requirements. Modest reserve excesses from the previous
period can be carried over into the current period, while modest
deficiencies from the previous period must be covered.
As a general matter, the bank's objective is to end each day with
roughly enough reserves to keep pace with requirements to that
point. To do this, it needs to make short-run interest rate forecasts.
The expected trend in rate movements is one element examined.
Technical factors affecting reserves of the banking system and the
money desk's assessment of how the Federal Reserve will respond
to those factors must also be considered. Short-run interest rate
expectations may suggest to the money desk that it might be profitable to run the reserve position somewhat long or short of
requirements for a given day or maintenance period. Both strategies involve risks. If the bank holds more excess reserves early in
the period than it can eliminate before the period ends, it suffers an
opportunity cost for holding non-interest-bearing reserves. On the
other hand, if it has a deficiency, it risks running an overdraft at the
Fed or having to borrow at the discount window.
6 Chapter 6, Box A offers a detailed description of reserve measures and accounting techniques.




61




Even when the intention is to run the position short of required
levels, the money desk typically targets a sufficiently large positive
reserve balance at the end of the day to guard against a last-minute,
unexpected outflow that could leave the account overdrawn. If a
bank realized at day's end that it was overdrawn, it would have to
cover the overdraft with a visit to the discount window. If the bank
did not discover the overdraft until the next day, it would face a
penalty, and it would have to make up the reserve deficiency.
The officers or committee in charge of the money desk will make
use of projections of asset and liability maturities and estimates of
additional sources and uses of funding in the coming weeks. They
will try to anticipate unusual loans or deposit buildups. A large
bank will also predict patterns of activity by small banks that place
excess funds with it and rely on it for other services. The big bank
will use this information to devise an approach to be used in rolling
over wholesale deposits and RPs and in making loans to securities
dealers. Out of this plan will emerge estimates of the bank's net
overnight funding needs. The bank will plan either to borrow or
lend in the interbank Federal funds market. The outlook for the
Federal funds rate over the current maintenance period and the
ensuing one is considered, and the officer in charge of the money
desk directs the Federal funds trader whether to err on the long or
short side in managing the bank's reserve position.
The Federal funds trader of a major bank usually starts each day
with an estimate of the bank's closing position at its Federal Reserve
Bank the night before. The trader may not be confident of that position until noon, but active trading in the Federal funds market begins
by around 8:30 a.m. For a large bank that is a net borrower in the
Federal funds market, funds borrowed the previous day are typically
returned early the following morning, a practice which lowers the
bank's reserve balance significantly and puts the bank temporarily
into an overdraft position in its reserve account that will have to be
covered over the day. A bank that had loaned overnight Federal funds
will see a rise in its reserve balance that morning when the loans are
repaid. The bank has at hand estimates of the major transactions that
will go for or against it that day from newly issued and maturing RPs,
CDs, and Eurodollars, as well as from customer and other transactions. It has some control of these items, but unexpected transactions
do occur. The net inflow from direct transactions in Federal funds
from smaller client banks is reasonably predictable; hence the funds
trader has a fairly good estimate of the remaining net amount of
Federal funds that must be bought or sold in the brokers' market.

62

Federal Reserve open market operations may affect the bank's
reserve calculations. For example, if the Fed's trading desk executes an RP, the volume of reserves for the banking system will
rise. The individual bank can choose whether to participate directly
in the operation; if it does so successfully, it will increase its reserve
balances. Even if it does not execute an RP transaction directly, its
customers may participate. In that case, it will receive some of the
reserves added in the operation. Otherwise, it will be affected only
indirectly—reserves will be more plentiful in the banking system,
making additional reserves less costly to obtain or making surplus
reserves less profitable to sell.
In planning the day, the trader tries to gauge from broker comments, the bank's own direct Federal funds trades, and projections
of aggregate reserve supplies made by money market economists
whether the Federal funds rate is firm and likely to rise, or whether
it may ease, allowing the bank to cover its reserve shortage 1/16
or 1/8 of a percentage point less expensively. The money desk
manager will generally prefer not to be both a borrower and a
lender of wholesale funds in the brokers' markets, even though
there may be profit in undertaking both transactions when rates
change through the day. Capital requirements on gross asset positions discourage extensive use of two-way operations. As the day
goes on and the trader's own picture becomes clearer, the object
is to buy enough funds to come out about on target.
On settlement Wednesday, the trader has to bring the Fed balance to the level needed to meet the average level required for the
two-week maintenance period, after allowance for any excesses or
deficiencies carried in from the previous period. It is an especially
tricky and often volatile day. Reserves in the system are likely to
be either overly abundant, encouraging the Federal funds rate to
fall, or in short supply or maldistributed, producing upward rate
pressure. The trader's success in contributing to bank profits
depends importantly on winding up on settlement Wednesdays in
need of reserves when the funds rate is low and with adequate
reserves when the funds rate is high. On those occasions when the
bank finds itself suddenly short because of unexpected transactions, or the trader cannot find funds in sufficient volume late in the
day before Fedwire closes, the remaining option is to turn to the
Federal Reserve discount window. In some instances, the bank's
officers may have concluded that they can risk borrowing at the
discount window because of their limited recourse to it in the recent
past. In other cases, they may go to extraordinary lengths to avoid




63




borrowing at the window if they have used it often and are likely to
be counseled against further borrowing.
In addition to managing the day's flows with a view toward producing a desired end-of-day position, the money desk must also monitor
its intraday position in accordance with Federal Reserve policies
regarding payment system risk. The roughly $1 trillion of daily payments over Fedwire and private payment systems such as the
Clearing House Interbank Payments System (CHIPS) and the payments arising out of the Federal Reserve's Book Entry Securities
System currently produce about $40 billion daily on average of free
intraday credit (that is, "daylight overdrafts"), with peak amounts well
in excess of $100 billion. These overdrafts reflect transfers made with
the expectation that sufficient inflows of funds will cover them before
the day is over. The risks in this process have received a great deal of
attention from the Federal Reserve and the financial community as a
whole in recent years: the Federal Reserve has established policies to
limit daylight overdrafts and the banks have put procedures in place
to monitor and manage their exposures. A systemic risk is inherent
in this extension of daylight credit—the risk that one institution's
inability to settle its net debit position on a private large dollar settlement system will in turn force other participants to default. Moreover,
the failure of a bank to settle Fedwire overdrafts poses the risk of
direct loss to the Federal Reserve. Consequently, depository institutions are expected to limit their daylight overdrafts to specified multiples of their capital on each separate payment system, such as
Fedwire, and across all payment systems. They are also expected to
monitor those positions in real time. Federal Reserve proposals to
price Fedwire daylight overdrafts and require collateralization of certain overdrafts are currently pending; if adopted, they would further
add to the money desk's intraday funds management responsibilities.
(Overdrafts, of course, could occur on the books of private institutions
as well as on the Federal Reserve's books.)
In addition to adjusting positions minute by minute, the money
desk must also take account of the longer term perspective contained in the Asset and Liability Committee's strategy. A significant
change in the economic outlook and expectations for monetary
policy may, for example, suggest lower rates over the relatively
near term. This development could prompt the Asset and Liability
Committee to decide to shift out of a long-funded position to minimize potential losses. In such circumstances, the money desk
might seek to unwind some long liability positions, presumably at
a cost, and put some liquid funds to work in longer maturity assets.

64

Flexibility is a key element in the development and implementation of asset and liability strategies. The inability to react quickly
and effectively to changing circumstances can eat into hardearned profits. As the competitive landscape grows more difficult
to negotiate and the financial marketplace becomes increasingly
complex, the one constant for banks is that nothing is really likely
to stay as it is for very long. And it is within this continually changing framework that monetary policy must do its work.
The Federal Reserve's trading desk routinely observes how the
large money center banks and various groupings of smaller banks
and thrifts are managing their reserve positions. Members of the
desk staff speak with money desk managers of large banks and
monitor daily statistics on reserve positions of groups of other institutions by size and type. What they learn can clarify the behavior
of aggregate excess reserves and of the Federal funds rate. More
specifically, it may help desk officials understand instances when
reserve and Federal funds rate behavior do not seem consistent.
This knowledge often proves helpful to desk officials in planning a
strategy for reserve management.




65




The Financial Markets
The existence of broad-based, active financial markets in the
United States is very important to Federal Reserve policy implementation. The markets provide a place where the Federal Reserve
can buy and sell Treasury debt instruments in carrying out open
market operations. The Federal Reserve uses such transactions to
make large-sized reserve adjustments quickly. If active markets in
financial instruments did not exist, the Federal Reserve would not
be able to make open market operations its primary policy instrument, and a very different, less efficient set of monetary policy procedures would have developed. Moreover, without large-scale
financial markets, the economic conditions addressed by Federal
Reserve policy would barely resemble the complex system that
has evolved in the United States, since the variety and efficiency of
means of borrowing and lending have affected the course of economic development.
The financial markets encompass a vast array of techniques and
instruments for borrowing and lending that facilitate investment,
consumption, saving, and the convenient timing of purchases and
sales of goods and services. The borrowers are mostly businesses,
individuals, and governmental units with a variety of needs for
funding. Lenders are businesses and individuals with savings to
invest. Many entities fall into both categories. Financial institutions,
including commercial banks, investment banks, insurance companies, and others, intermediate between borrowers and lenders. In
addition, a wide variety of financial instruments have been developed that permit borrowers to sell their own securities, usually with
the assistance of investment banks, without relying on the intermediary services of commercial banks.
Active financial markets help potential borrowers and lenders
find the most advantageous terms and interest rates. The marketmaking processes allocate savings to the uses offering the highest
return and search out the interest rates that bring supplies and
demands into balance. The determination of the overall level and
the structure of interest rates according to the maturity of the instrument is a complex process (see the discussion in Chapter 8). For
any maturity, rates will differ among instruments if they are perceived to have different risk, tax, or marketability characteristics, or
1 This chapter draws on the following sources: Marcia Stigum, The Money Market, rev. ed.
(Homewood, Illinois: Dow Jones-Irwin, 1983); Timothy Q. Cook and Timothy D. Rowe,
eds., Instruments of the Money Market, 6th ed., Federal Reserve Bank of Richmond, 1986;
and First Boston Corporation, Handbook of Securities of the United States Government and
Federal Agencies, and Related Money Market Instruments, July 1988. It also draws heavily from the 1982 version of this book that was written by Paul Meek.

66

are available to different classes of purchasers (lenders). The
spread between interest rates of two financial instruments of the
same maturity may change if perceptions about such characteristics change.2
The highly developed nature of financial markets in the United
States and the wide range of choices for borrowing and lending have
facilitated a massive expansion of outstanding debt. The large volume of debt can be seen as a sign of economic and financial vigor,
but at times it can also be worrisome. Servicing the debt could be a
problem in a period of economic retrenchment when corporate
profits and personal income tend to weaken. In addition, with market development has come increased integration among the various financial instruments, an outcome that may speed the transfer
of credit problems from one part of the financial markets to another.
Market participants often make a distinction between financial
instruments with maturities of a year or less and those with longer
initial maturities. The market in which the shorter instruments are
issued and traded is referred to as "the money market." The money
market is really a market for short-term credit, or the option to use
someone else's money for a period of time in return for the payment of interest. The money market helps the participants in the
economic process cope with the financial uncertainties they face
in daily life. It assists in bridging the differences in the timing of
payments and receipts that arise in a market economy. Borrowers
rely on it for seasonal or short-term cash requirements; lenders use
it to offset uneven flows of funds. By providing a place for funds to
be placed temporarily, the money market also permits borrowers
to time their issuance and lenders to time their purchases of bonds
and equities in accordance with their forecasts of stock prices and
long-term interest rates. (Table 1, page 68, lists characteristics of
several money market instruments.)
Markets dealing in instruments with maturities that exceed one
year are often referred to as capital markets, since credit to finance
investments in new capital would generally be needed for more
than one year. The time division is somewhat arbitrary. A longterm project can be started with short-term credit, with additional
financing arranged at a later date. Furthermore, two- or three-year
credit instruments may need to be renewed before a project is
completed. Debt instruments that differ in maturity share other
2 Ray C. Fair and Burton Malkiel, "The Determination of Yield Differentials between Debt
Instruments of the Same Maturity," Journal of Money, Credit, and Banking, vol. 3, no. 4
(November 1971), pp. 733-49.




67

Table 1 The Money Market
Instruments

Typical
Maturities

Principal
Borrowers

Secondary
Market

Federal funds

Chiefly 1 business day

Depository institutions

None

Negotiable certificates
of deposit (CDs)

1 to 6 months and longer

Depository institutions

Modest activity

Bankers acceptances

90 days

Financial and
business enterprises

Active

Eurodollars:
Time deposits
(nonnegotiable)

Overnight, 1 week,
1 to 6 months and longer

Banks

None

CD (negotiable)

1 to 6 months and longer

Banks

Moderately active

Treasury bills

3 to 12 months

U.S. government

Very active

Repurchase agreements

1 day, 1 week, 3 to 6 months

Banks, securities dealers,
other owners of securities

None, but very active
primary market for
short maturities

Federal agencies
Discount notes

30 to 360 days

Coupon securities

6 to 9 months

Commercial paper

Municipal notes




1 to 270 days

30 days to 1 year

Federally sponsored
agencies:
Farm Credit System
Federal Home Loan Banks
Federal National Mortgage
Association
Financial and
business enterprises
State and Local governments

Limited
Active

Very limited
Moderately active
for large issuers

characteristics. Hence, the term "capital market" could be—and
occasionally is—applied to some shorter maturity transactions.
(Table 2, page 69, gives examples of capital market instruments.)
A distinction is also made between primary and secondary markets. The term "primary market" applies to the original issuance of

68

Table 2 The Capital Market
Instruments

Typical
Maturities

Principal
Borrowers

Secondary
Market

Treasury notes
Treasury bonds

2 to 10 years
Recently, 10 to 30 years

U.S. government
U.S. government

Active
Active

Federal agencies:
Bonds

3 months to 10 years

Farm Credit System,
Federal Home Loan Bank,
and related institutions

Moderately active for
recent issues,
less active for
older issues

Debentures

2 to 30 years

Federal National Mortgage Assn;
Federal Home Loan Mortgage Assn.

Moderately active
depending on
maturity

Master notes

Up to 10 years - negotiable

Federal National Mortgage Assn;
Student Loan Marketing Assn.

Active

Zero coupon

Long-term

Federal National Mortgage Assn;
Student Loan Marketing Assn.

Limited

Fixed/floating rate
Swaps

2 to 10 years

Student Loan Marketing Assn.

Active
[see swaps below]

Corporate bonds

2 to 30 years

Financial and business enterprises

Active (OTC market)

Municipal bonds

2 to 30 years

State governments

Active (OTC market)

Contracts mature every
3 months out to 2 years

Dealers, banks (users)

Active arbitrage with
cash market

Options

Exercise at strike price on or
before prearranged expiration
date

Dealers, banks, nonbanks

Very active

Swaps

Exchange of interest streams
over the lives of underlying
debt issues

Dealers, banks, nonbanks

Active
(sales termination,
reverse swaps)

Strips

Semiannually on each coupon
date and bond maturity date
out to 30 years

U.S. government (indirectlystripping done by dealers)

Active

Derivative products:
Futures contracts




69




a credit market instrument. There are a variety of techniques for
such sales, including auctions, posting of rates, direct placement,
and active customer contacts by a salesperson specializing in the
instrument. Once a debt instrument has been issued, the purchaser
may be able to resell it before maturity in a "secondary market."
Again, a variety of techniques are available for bringing together
potential buyers and sellers of existing debt instruments. In many
cases, the same firms that provide primary marketing services
help to create or "make" secondary markets. Sometimes, brokers
play a key role in the market process. The development of active
secondary markets has increased the attractiveness of debt instruments to potential purchasers. Firms can keep some of their liquid
working balances in short-term instruments, which they can then
sell before maturity if they need cash. This source of liquidity has
affected money and bank credit because it has reduced firms'
needs to keep funds on deposit and to obtain short-term loans from
the banks.
In addition to making outright purchases and sales in the secondary market, entities with money to invest for a brief period can
buy a security temporarily, and holders of debt instruments can
borrow short term by selling securities temporarily. The two types
of transactions are repurchase agreements (RPs) and reverse RPs.
In the wholesale market, banks and government securities dealers
offer RPs at competitive rates of return by selling securities under
contracts providing for their repurchase from one day to several
months later. Finally, a variety of derivative instruments such as
futures and options contracts on various financial instruments can
be used for hedging interest rate risk or for speculating.
Increasingly, the financial markets have become international in
scope. Banks of many nations bid for deposits and make loans
throughout the world. Foreign borrowers may raise funds in the U.S.
credit markets and (J.S. borrowers can raise money abroad by issuing securities denominated in U.S. dollars or in other currencies.
Foreign central banks and others hold (J.S. dollar securities in large
volume as part of their reserves. U.S. Treasury securities trade virtually around the clock in major financial centers in Europe and
Asia as well as in the United States. The U.S. dollar is the main international currency, but some financial instruments are denominated
in currencies other than the local currency or occasionally in a basket of currencies. The international markets provide facilities for
managing currency risk by allowing investments in many currencies and by providing hedging facilities for managing exchange risk.

70

Financial Intermediaries and the Financial Markets
The development of financial markets has allowed large creditworthy commercial entities to avoid traditional intermediaries and
borrow directly from investors, either through dealer departments
of investment banking firms or by direct placement. Business and
financial corporations and municipalities often borrow by issuing
unsecured commercial paper. A corporation with a strong credit
rating can capitalize on the rating and obtain short-term loans from
investors at a rate lower than that charged by a bank.
Commercial banks, nonetheless, play several important roles in
the financial markets. In addition to providing traditional deposit
transfers and loans, they create and deal in financial market instruments. The 15 to 20 largest U.S. banks are particularly active in
the money market. The banks figure importantly in the markets for
Federal funds, bankers' acceptances (BAs), certificates of deposit
(CDs), Eurodollars, and RPs. Their holding companies issue commercial paper. Money center banks are typically the principal
domestic traders in the worldwide foreign exchange market. They
also furnish the transfer, record keeping, and credit facilities needed by nonbank participants. Many banks act as dealers in money
market securities, while others service customer investment needs
through a short-term investment desk. A few banks in New York
City serve as clearing agents for dealers in money market instruments, receiving and delivering securities for them and insuring
that payments are made. A number of large banks meet residual
financing needs of nonbank money market dealers.
Some of the larger domestic regional banks and affiliates of foreign banks are active, too, in trading Federal funds and issuing
CDs and BAs. They serve both area businesses and nationally
based firms, providing their customers with access to the international money market. American affiliates of foreign banks provide
access to the U.S. money market for their head offices abroad, for
their global branch networks, and for the U.S. operations of their
overseas clients. The affiliates also are active participants in the
foreign exchange markets.
For most depository institutions, participation in the capital
markets consists of purchases and sales of government securities
for their investment portfolios. A few serve as dealers in U.S. government securities and underwriters of state and local government
securities. Their authority to engage in other types of underwriting
and market-making activities is limited by statute, but as noted in




71

Chapter 3, the regulatory agencies have recently been expanding
the limits administratively.
Thrift institutions—savings and loan associations, savings banks
and credit unions—have participated in a wider range of financial
activities in recent years and have made use of market instruments
to manage cash and risk positions. They have also sold mortgage
loans and purchased mortgage-backed securities to manage risk.
Most of them, however, have a smaller role in the financial markets
than the commercial banks have.
Bank-related Financial Markets
1. The Federal Funds Market3
The Federal funds market is the market for immediately available reserve balances at the Federal Reserve.4 Depository institutions that maintain accounts at the Federal Reserve, either directly
or through a correspondent, can borrow (buy) or lend (sell) a portion of those reserve balances.5 Depository institutions hold
reserve balances at the Federal Reserve to meet their reserve
requirements — on average over a two-week maintenance period
— and to handle clearings of transactions among banks while
avoiding overnight overdrafts that may arise from those transactions. Because the Federal Reserve does not pay interest on
reserve accounts, depository institutions have an incentive to hold
their reserve balances to the minimum levels consistent with meeting their various needs.
Regular flows of business to a bank are unlikely to leave it with
the desired level of reserves. A bank that is short of reserves has a
number of adjustment options, including the purchase of enough
Federal funds to offset the shortage. Such borrowings are not classified as deposits, so they are subject neither to reserve requirements nor to the statutory prohibition against paying interest on




3 Lawrence DiTore of Fulton Prebon and Jamie Paterson of Chemical Bank provided helpful
information for this section.
4 The various reserve concepts are described in more detail in Chapter 6, Box A.
5 The other eligible participants in the Federal funds market are federally sponsored agencies
that provide banking services such as the Federal Home Loan Banks, and certain official
international banking organizations, such as the International Bank for Reconstruction
and Development.
The term "Federal funds transaction" is sometimes applied to payment in immediately
available funds. Such payments would normally use the Federal Reserve's funds transfer
facilities. Any nonfinancial entity can arrange with its bank to have such a payment made
on its behalf. But this type of payment is not a Federal funds transaction in the sense that
the term is used above because it does not involve the borrowing and lending of reserve
balances at the Federal Reserve.

72

demand deposits. A bank with reserve balances in excess of its
needs may lend them in the Federal funds market.
Most banks tend to be routinely either net buyers or net sellers
of funds although some shift back and forth. The large banks are
more often buyers than sellers while the smaller institutions tend to
be net sellers, but in both cases there are exceptions. Typical sellers include small domestic commercial banks, thrift institutions,
and credit unions.6 The institutions that are routine sellers often
view the monies sold in the Federal funds market as part of their
liquidity. Data collected by the Federal Deposit Insurance Corporation (FDIC) on its call report of all the institutions it insures indicated that the institutions' daily purchases of Federal funds were
around $145 billion at the end of 1988. Sales were $100 billion.
(Data on transactions by institutions that were not insured by FDIC
are not included in these figures.)
There are two methods for buying and selling Federal funds.
First, depository institutions can deal directly with each other.
Second, brokers can bring together financial institutions with
shortages and those with excesses of reserves. Direct transactions
most commonly consist of sales by small- to medium-sized institutions to larger correspondent banks. Small institutions rarely
generate reserve excesses large enough to allow them to participate in the brokers' market. Instead, they arrange to have a correspondent bank buy from them directly. Often these transactions
take place on a regular basis: if the respondent institution routinely
generates more reserve balances in its business than it needs, it
may make daily sales to its correspondents. Usually, the transaction takes place either at the opening rate or the average effective
rate set in the brokers' market less a fraction. Some direct transactions also occur between two large institutions and may, at
times, be of substantial size.
A substantial share of large transactions are arranged in the brokers' market. Trades through the brokers are typically for $25 million or more, although trades of around $10 million are arranged
routinely.7 Brokers provide an essential service to the one thousand
or so financial institutions that are regular participants. The major
Federal funds brokers do not take positions themselves but bring
together potential buyers and sellers. They take bids and offers
6 The Federal Home Loan Banks sell reserve balances in the funds market on behalf of the
member savings and loan associations. The U.S. Central Credit Union sells funds on behalf
of member credit unions.
7 Trades as small as $1 million are occasionally arranged as odd lots.




73

Diagram 1 Fed Funds Transaction (with Broker)

The transaction isjeversed
the following day.




notifies

Selling Bank's
District
Reserve
Bank

Wires funds to
buying bank's
account at
its district
reserve bank

Debits selling
bank's reserve
account*

from banks by phone, charging each party to the trade a commission of 50 cents per $1 million. Generally either 1/16 or 1/8 percentage point separates the bid from the offer. If the market is very
one-sided or rates are changing rapidly, the spread may be greater.
Since most loans are unsecured, depository institutions establish
credit limits for each potential buyer. Once the terms of the
exchange are agreed upon, the selling institution notifies its District
Reserve Bank to debit its account and wire the funds to the buying
bank. The banks entering into the contract, rather than the broker,
are responsible for making sure the transactions are completed.
Typically, the transaction is reversed and the interest is paid the
next day (Diagram 1).
Participants in the Federal funds market can get an idea of the
rates at which funds are trading by looking at on-line information
screens provided for a fee by various financial service firms. Several
brokers report the current rates bid and offered for Federal funds
and the rate at which the most recent transactions took place.
Participants phone the brokers to get their views on the market and
to place bids or offers. Brokers will indicate whether the market is
"better bid" or "better offered." They will try to get bidders to step

74

up their rate or sellers to accept a lower rate when they observe a
concentration of bids or offers. Staff members at the trading desk of
the New York Reserve Bank also watch the news screens and telephone the brokers from time to time during the day to keep abreast
of the rates, the volume of activity, and the balance between supply
and demand. In the latter part of 1988, the daily volume of Federal
funds trades arranged through the brokers reporting to the Federal
Reserve Bank of New York averaged around $50 billion.
Most activity in the market involves purchases and sales for one
business day, but trading for future delivery and for an extended term
also takes place. Trading for future delivery is most common ahead
of quarter ends when heavy flows of funds through the banking system may inflate cash needs. The market for "term" Federal funds is
a wholesale market in unsecured interbank lending. Maturities
range from a few days to more than a year, although most transactions mature in six months or less. The term funds market is considerably smaller than the overnight market; the volume of activity
varies but the amount of term Federal funds outstanding is likely to
be on the order of one-tenth of the amount of overnight funds
arranged on a given day. The term market is less liquid than the
overnight market. On occasion the broker may need hours or even
days to find a counterparty willing to meet the rate bid or offered.
For a bank with an extended need for funding, buying funds for a
specified term is equivalent to issuing a CD, except that such borrowing does not carry reserve requirement and deposit insurance
costs. Banks can thus afford to pay a higher rate than they would
be willing to pay if they were issuing CDs. The sellers in the term
market are members of the same group that participates in the
overnight Federal funds market. Resident foreign banks often place
funds raised abroad in this way when the rate spread is favorable.
Some banks situated abroad lend Federal funds whenever the rate
is sufficiently above that available on RPs to compensate them for
the lack of collateral against the loan. Savings and loan associations
and the supervising Federal Home Loan Banks also use the term
funds market to invest liquid reserves. Term Federal funds transactions are not subject to early termination except in unusual circumstances when both parties agree.
2. The Market for Certificates of Deposit
After its introduction in 1961, the large negotiable bank CD grew
rapidly in importance and often served domestic banks as a major
source of funds. Banks borrow by issuing CDs, principally to non-




75

Chart 1 Large Time Deposits (Semiannual Averages)




Billions of dollars
600

1959 '61 '63 '65 '67 '69 71

73

75

77

79

'81 '83 '85 '87 '88

banks. The CD, like a U.S. Treasury bill, can be sold before maturity. Its secondary market, however, is not nearly as liquid as the
bill market, and a CD carries some credit risk. Consequently, CDs
must offer investors a rate of interest that exceeds that on a
Treasury bill of the same maturity. The success of the domestic CD
was followed by the growth of an active market for Eurodollar CDs,
or dollar-denominated CDs issued by banks or branches located
outside the United States, primarily in London. Secondary markets
developed first in London and later in New York. Dollar-denominated CDs are also issued by foreign banks located in the United
States and are known as Yankee CDs. Large domestic savings and
loan associations are also issuers.
The domestic CD market grew rapidly from its inception, with
only a few notable interruptions (Chart 1). During the 1960s, CD
rates were subject to interest rate ceilings specified under Federal
Reserve Regulation Q. When market rates rose above the ceiling
rates in 1966 and again in 1969, demand for domestic CDs
dropped. In both instances, the Eurodollar market, which was
exempt from the ceilings, got a boost. Then in 1970, the collapse of
the Penn Central Transportation Company caused a crisis in the
commercial paper market. To ease the resultant liquidity problems,
the Federal Reserve took the first in a series of steps to remove
interest rate ceilings: it immediately eliminated ceilings on shortterm time deposits of $100,000 or more in value. Growth in large

76

CDs resumed; the volume of large domestic time deposits, the bulk
of which were negotiable CDs, rose from $21 billion at the end of
1969 to $334 billion in November 1982. In December 1982 depository institutions were able to begin issuing money market deposit
accounts (MMDAs), which paid unrestricted interest rates on consumer deposits with no minimum maturity. MMDAs and other
deregulated consumer deposits were very popular. Banks—particularly the large money center banks that had been heavily dependent on wholesale funding—reduced CDs outstanding for a number
of months. Expansion resumed in the second half of 1983, partly
because brokers sold $100,000 CDs for regional financial institutions, thus offsetting the decline in money center bank wholesale
CDs. In recent years, with the reduction in wholesale CD issuance,
the secondary CD market became much less active.
Most primary market sales of large CDs are negotiated between
banks and their customers. Most banks still post the rates at which
they are prepared to accept deposits for the most popular maturities—one, two, three, and six months—although they will only post
attractive rates when they are anxious to issue CDs. Major banks
usually have salespeople who keep up with customer needs and
call around if the bank decides to make more sales than are
achieved through routine contacts. A bank also may place CDs
with a dealer who will market them at a concession, although the
decline in volume in 1983 reduced the importance of this method
of issue. In many cases, dealers will act as brokers, finding customers for a bank's CDs but not taking them into their own positions. Sales handled through dealers tend to be in round lots of $5
million or more, although $1 million pieces are placed.
In addition to issuing short-term CDs, primarily with fixed interest rates, banks offer a considerable volume of longer term variable-rate CDs. CDs with an initial maturity of 18 months or more
are popular with banks because they are exempt from the 3 percent reserve requirement imposed on shorter term "nonpersonal"
CDs. They are priced off a variety of short-term interest rate indexes such as the London interbank offer rate (LIBOR) and the Federal
funds rate.
Banks also issue what are known as deposit notes or CD notes,
a hybrid between ordinary CDs and corporate bonds. Most of these
notes mature in 18 months to about 10 years. Like deposits, they
are free of the Securities and Exchange Commission (SEC) registration requirements that apply to bonds. Banks must pay insurance premiums on deposit notes, although some of the notes do




77

not use the term "deposit" and thus avoid the insurance premium.
The FDIC, however, is seeking to collect insurance on all such
notes. The notes would also be subject to reserve requirements
except that their initial maturities exceed 18 months. Banks must
report their deposit note volume to the Federal Reserve as part of
their total large time deposits. Like bonds, deposit notes pay interest semiannually, and they are often purchased by traditional bond
buyers. Sizable issuance began in 1985, the year CD notes were
first rated by major bond rating services.
The secondary CD market became less active as the primary
dealer market contracted. In the 1970s and early 1980s, CDs often
traded through a brokers' market in which bids and offers were listed on a "run" of about 10 to 12 banks, with the individual bank
names not reported. Beginning in 1982, credit concerns about certain banks "on the run" complicated such trading. As credit problems increased, tiered CD rates for different banks developed and
issuance of wholesale CDs diminished. Consequently, trading
through brokers disappeared. Currently 10 dealers participate in
the secondary market in CDs, although only about 6 are active.8
Because regional and foreign banks (especially Japanese banks)
have become more important as issuers in recent years, they have
taken on an expanding role in making markets. With the markets
becoming so thin, a dealer that buys a CD from a customer cannot
count on reselling it quickly. The dealer therefore will demand a
price concession or liquidity premium to protect it in case it has to
hold the CD to maturity. It faces financing rates that are generally
higher than those on Treasury securities in recognition of the CDs'
lesser marketability and somewhat greater credit risk. The CD
futures market, which once provided a means of hedging position
risk, has become moribund.
3. Bankers' Acceptances




The Federal Reserve Act authorized U.S. banks to engage in
acceptance financing of the domestic and foreign trade of their customers. Nurtured by the Federal Reserve, the market in BAs burgeoned to the point where it now finances a significant share of
trade denominated in dollars in the United States and foreign countries (see Box A, pages 80-81). Federal Reserve regulations continue to govern the issuance of most acceptances, limiting their use to
short-term, self-liquidating commercial transactions. Acceptances
8 A dealer that is owned by a bank is forbidden by Federal Reserve regulations to buy back
CDs issued by that bank.

78

are subject to reserve requirements if they exceed 200 percent of
the bank's capital and surplus. Since 1982, when the limit was liberalized, no bank has come close to its limit.
The BA available from banks or in the dealer market is a prime
short-term investment because both the bank and its customer are
legally obligated to pay it at maturity. Acceptances are written in
varying amounts based on the underlying transaction, but they are
put together for sale in round lots of $1 million to $5 million. The
odd lots remaining, in pieces down to about $25,000, are sometimes sold to individual investors and sometimes held by the
accepting bank. About 15 to 20 firms make active markets in these
instruments, buying acceptances from the accepting banks and
retailing them to corporations, government agencies, foreign
investors, banks, and other financial institutions.
Secondary market trading of the larger pieces has grown significantly; dealers typically quote a bid-offer spread of 2 to 4 basis
points. Dealers also use brokers to facilitate trading with other dealers while preserving their own anonymity; the cost is one basis point
to the originator of the trade. Two "runs" of acceptances trade on
brokers' screens, and within each, acceptances trade interchangeably. The domestic run consists of BAs of about 9 to 10 large domestic banks, while the Japanese run consists of BAs of about a dozen
Japanese financial institutions. Dealers finance their positions with
bank loans or RPs arranged with a wide variety of investors.
BAs trade in a tiered market at rates reflecting the size of the
accepting banks, market perceptions of the banks' creditworthness, and the perceived liquidity of the paper in the market.
Membership in the tiers changes from time to time as market conditions and perceptions of credit risk and liquidity are altered. The
first tier generally consists of some of the large U.S. banks that
constitute the domestic run. The next group comprises 10 to 12
large regional banks and the other names on the domestic run.
After that are the Japanese banks. The final tier, a group of perhaps
20 to 25 banks, includes non-Japanese foreign branches and
agencies of foreign banks. This tier also includes Edge Act subsidiaries of U.S. banks, institutions established to transact foreignrelated business outside of a bank's home state. The spread
between the top names and the final group is 10 basis points or
more and depends on market conditions. (A basis point is 1/100
of a percentage point.)
The trading desk in New York was an active buyer of BAs for the
System's own account from the beginning of the Federal Reserve




79

Box: Financing through Bankers' Acceptances
If an importer and an exporter
arrange a delayed payment
transaction, the seller must
assume the risk that the buyer
may be unable to pay....

But if the same basic transaction is arranged with a bank
guaranteeing payment, the risk
is transferred to the bank.

GOODS
(NOW)

GOODS
' (NOW) '

Importer

mmmm Substantial credit risk
CASH
(LATER)

Minor credit risk
*

DOCUMENTS
(NOW)
ACCEPTED
TIME DRAFT
(NOW)

Typically an acceptance is purchased
(discounted) first by the
accepting bank and
then resold (rediscounted) to another
investor

ACCEPTANCE
(NOW)

CASH
(NOW)

CASH
(LATER)

Chart adapted from William Melton and Jean Mahr,
"Bankers'Acceptances,"Quarterly Review Federal
Reserve Bank of New York, Summer 1981, p. 40.




Exporter

MATURED
ACCEPTANCE
(LATER)

System through the early 1930s and again after World War II until
the mid-1970s (although in the postwar years BAs met only a
small proportion of reserve needs since Treasury issues had taken
over the dominant role). In 1977, the FOMC decided that its active
support of the market was no longer necessary and it discontinued
its purchases. In 1984, the Fed discontinued its purchases of
acceptances under RPs because the volume of government securities available was sufficient to meet its reserve management
objectives.
Although the Federal Reserve no longer participates in the
acceptance market for its own account, it does purchase and sell

In the course of international trade, importers and exporters find
it advantageous to organize transactions in such a way that the
importer does not pay until a specified date in the future. However,
in the absence of information on the creditworthiness of the
importer, the exporter may be reluctant to extend the credit.
Bankers' acceptances are financial instruments that bridge this difficulty; they substitute the bank's creditworthiness for that of the
importer. The bank acts as intermediary by guaranteeing to make
the payment for the goods on the specified date. The bank guarantees payment by "accepting" a time draft (or order to pay)
drawn on it by the exporter. The bank charges the importer a fee,
which is usually about 50 basis points but may be larger or smaller,
depending on competitive conditions and the creditworthiness of
the importer. The accepting bank either sells the acceptance at a
discount to a dealer in the secondary market or holds it in its portfolio as an investment. In either case the exporter receives immediate payment for the shipment. At maturity the bank receives
payment from the importer and pays the holder of the acceptances. Dollar acceptances are also used to finance trade and other
transactions between two foreign countries even though the United
States is not a party to the transactions.

acceptances in the secondary markets on behalf of its foreign official customers. It also retains a direct regulatory role, determining
the eligibility of acceptances as collateral at the discount windows
of the Reserve Banks. Eligibility at the window is limited to transactions involving original maturities of six months or less for foreign and domestic trade or for storage of goods in transit. Also
eligible are dollar-exchange acceptances drawn by approved
countries for three months or less to finance seasonal needs.
Reserve requirements apply to bills originally written to mature in
more than six months and to finance bills issued to raise working
capital. These requirements have made such acceptances rare.




81

4. The Eurodollar Market9




Eurodollars are U.S. dollar deposits at banking offices in a country other than the Gnited States. Eurodollars came into existence
in the 1950s when Soviet bloc governments placed dollar deposits
in London in order to conduct transactions in Europe and avoid the
potential risk that the G.S. government might, for political purposes, freeze deposits held in the Gnited States. Eurodollar deposits
proved attractive to a wide range of depositors because, unlike
U.S. deposits, they were not subject to interest rate ceilings,
reserve requirements, or FDIC insurance premiums. The Eurodollar market—the process through which banks solicit these
deposits and place the proceeds—grew spectacularly in the 1960s.
Although the interest rate restrictions were gradually removed
from domestic deposits beginning in 1970, reserve requirements
and insurance premiums have remained in effect. Furthermore,
because the dollar finances international trade and investment,
investors have found it convenient to hold deposits in the time
zones where trade-related dollar transactions are taking place. For
these reasons, Eurodollars have continued to be popular.
Eurobanks—which include foreign branches and international
banking facilities of U.S. banks—now operate not only in London
but also in Tokyo, Hong Kong, Singapore, Bahrain, Western European financial centers, several Caribbean islands, and other parts
of the world. Eurodollar deposits may be either nonnegotiable time
deposits or negotiable CDs. In contrast to the large time deposits
in the Gnited States, nonnegotiable deposits predominate in the
Eurodollar market.70 Both types come in a wide range of maturities,
from the next day to five years or more in the future. There are no
Eurodollar transactions deposits. The bulk of deposits mature
within one year, but multiple-year maturities are considerably
more common than in the domestic market. The banks bid for the
deposits of international corporations, investors, and governmental units to fund the loans being made to businesses and governments. They also bid for the deposits of other banks or place funds
with them, using the huge interbank market to manage the balance
between the maturities of their assets and their liabilities.
9 The role of the Eurodollar market in the international transmission of policy impulses is
discussed in Chapter 9.
Win some countries, there are legal obstacles to issuance and clearance of negotiable instruments. Most Eurodollar deposits are arranged through brokers. Rates are quoted relative
to LIBOR.

82

G.S. banks and resident foreign banks help keep Eurodollar
rates closely parallel to rates in the domestic money market.
Changes in Federal funds and other short-term G.S. rates rapidly
affect Eurodollar rates. Interest rate differentials between Eurodollar and domestic funds that are not based on regulatory differences are quickly eliminated through arbitrage. Same-day
settlement of Eurodollar transactions, introduced in the 1980s
through the clearing house interbank payments system (CHIPS),
reduced arbitrage costs.
CJ.S. banks may place domestically generated funds in the
Eurodollar market for varying terms when interest rate relationships
favor such actions. They may also borrow Eurodollars to use in their
domestic banking operations. The borrowing of repatriated
Eurodollars in excess of the volume that the bank has placed abroad
subjects the institution to a 3 percent reserve requirement. The banks
closely monitor their net positions and will only take Eurodollars in
excess of their placements if the interest rate is sufficiently below the
Fed funds rate to compensate for the reserve requirement.
(J.S. banks are among the major participants in the Eurodollar
market, both through their overseas branches and through their head
offices, which in most cases control the global "dollar book." The
head office generally conducts the funding side of offshore operations booked in the Nassau, Cayman Islands, and Panama branches
— branches which do business on New York time. Since 1981, CJ.S.
and foreign banks have been able to conduct international operations in CJ.S.-based international banking facilities, free from G.S.
reserve requirements, taxation, and FDIC insurance premiums.
Negotiable Eurodollar CDs grew in popularity from their introduction in 1966 until the early 1980s among G.S. and, to a lesser
extent, continental investors. G.S. money market mutual funds
(MMMFs) were major purchasers of Eurodollar CDs during their
period of greatest expansion in the late 1970s and early 1980s but
cut back on their takings when the funds shrank in 1983. Patterned
after the domestic instrument, Eurodollar CDs are usually issued
for maturities of one year or less in minimum pieces of $1 million.
They are delivered and held in London, and paid at maturity in the
Gnited States by telegraphic transfer from the London issuers. The
negotiability feature enables most banks to offer Eurodollar CDs at
rates below those prevailing on Eurodollar time deposits of similar
maturity. Rates on Eurodollar CDs are generally slightly higher
than those on CDs issued domestically by the same bank. The differential exists because physical delivery of Eurodollar CDs is




83

more complex, the roles of both the home and host countries* central banks as lenders of last resort are uncertain, and the host government could conceivably restrict withdrawals. Banks are willing
to pay the spread because they do not face the costs of reserve
requirements and deposit insurance.
London branches of U.S. investment firms and British merchant
banks and discount houses provide the main secondary market.
U.S. banks and corporations purchase a large share of Eurodollar
CDs—reportedly above three-quarters of the total. The quoted bidask spread in the secondary market has recently been about 5
basis points, with $1 million as the basic trading unit. Settlement is
made in New York in clearing house funds two days after the trade,
but delivery and custody of the CDs are in London. U.S. dealers
also make an active market in New York for these Eurodollar CDs.
5. The Interest Rate Swap




The interest rate swap was developed in the early part of the
1980s. Swaps permit two bond issuers to exchange commitments
to make interest payments over the lives of the debt instruments
that they issue, although each retains the obligation to pay off its
own principal borrowing at maturity. One borrower issues fixed-rate
debt while the other issues floating-rate debt with similar maturities.
Under the swap, the borrower that issued the fixed-rate debt will pay
the floating-rate interest over the life of the instrument, while the
party that sold the floating-rate debt will pay the fixed-rate interest.
Swaps can be profitable because of anomalies between fixed- and
floating-rate debt markets. Potential lenders atfloatingrates may differ
from potential lenders at fixed rates in their credit evaluation of potential borrowers. Sometimes borrowers find it cheaper to borrow in the
fixed-rate market while the revenue streams they will use to service the
debt are more closely related to a floating rate; in other instances, the
reverse may be the case. Swaps bring together borrowers with opposite revenue patterns. They allow each to borrow in the sector permitting the lower rate option and to hedge the interest rate exposure.
A commercial bank is often the intermediary in a swap, acting
as counterparty to two borrowers with opposite mismatches in
their borrowing and cash flow structures. In this role, the bank
assumes potential credit risks, which become actual risks if interest rate changes unmatch the payment commitments in the offsetting deals. Banks can avoid this interest rate risk by astute
offsetting of swap agreements, but sometimes they do not achieve
precise matches and therefore assume some residual rate risk.

84

According to call report data, commercial banks had $857 billion
of swaps on their books as of December 1988.
Nonbank Financial Instruments
1. The U.S. Treasury Debt Market
a. The primary market
The U.S. Treasury is the dominant issuer of debt instruments in
the financial markets. It sells both marketable and nonmarketable
debt, the former representing the larger share of its issuance. It
sells bills that mature within a year, notes that mature in 2 to 10
years, and bonds with maturities out to 30 years. The Treasury's
regular issuance of securities is an important part of its program
for managing the U.S. public debt, which was about $2.7 trillion at
the end of 1988. Of this amount, just over $1.8 trillion was in the
hands of the public, while a little over $0.2 trillion was held by the
Federal Reserve and about $0.6 trillion was held in Treasury trust
accounts. Treasury debt issues are purchased by a wide range of
investors who are attracted by the securities' freedom from credit
risk, ready marketability, exemption from state and local taxes,
and wide range of maturities. Banks, thrift institutions, foreign central banks, and a range of other financial and nonfinancial businesses in the United States and abroad buy marketable Treasury
securities. As of December 1988, the Treasury estimated that of
the $1.8 trillion of Treasury debt held by the public, 43 percent was
held by banks and other financial institutions, 22 percent by private
nonfinancial businesses and individuals, 19 percent by foreigners,
and 17 percent by state and local governments.
The Treasury has sold bills at competitive auctions since bills
were introduced in 1929. Beginning in the early 1970s, auctions
became the predominant sales technique for notes and bonds as
well. Nonmarketable debt is sold to specific purchasers on prearranged terms.77 The Treasury auctions bills most frequently, offering 91- and 182-day bills (generally referred to as 3- and 6-month
bills) each Monday (Tuesday, if Monday is a holiday). It sells 52week bills (referred to as year bills) every fourth week. Bills are dis11 Nonmarketable debt includes savings bonds, which are sold to the public as requested.
They are sold at a discount and pay the face value at maturity. Special securities are sold
to state and local governments when they want to invest the proceeds of a tax-exempt
security sale. To keep the local governments from making arbitrage profits from their taxexempt status by selling low-cost debt and purchasing Treasury debt paying higher rates,
the Treasury sells these governments special issues (often referred to as SLGs, or "slugs")
that pay rates equal to the municipalities' cost of funds. Some special issues are also sold
to Treasury trust funds.




85




count instruments, for which the purchaser pays an amount below
the face, or par, value. The Treasury repays the face value at maturity. The interest earned, referred to as the rate of discount, is computed approximately as the amount below the face value divided
by the fraction of the year that the bill is outstanding.72
To obtain bills at an auction, bidders must submit tenders on a
timely basis to the Treasury Department or to any Federal Reserve
Bank or branch serving as fiscal agent for the Treasury. Currently,
tenders are due before 1:00 p.m. eastern time on the day of the auction. The minimum tender size is $10,000, with additional amounts
permitted in $5,000 increments. Tenders can be either competitive
or noncompetitive. Competitive tenders must show both the amount
being tendered for and the rate of discount that the bidder is willing
to accept. There is no maximum bid size, but the Treasury limits its
issuance to any one bidder to 35 percent of the amount of the auction available to the public (exclusive of awards to the Federal
Reserve and foreign official institutions). This restriction is designed
to prevent one or two parties from taking so much of an issue as to
create a situation where the price could be manipulated artificially.
Once the tenders are received, each Federal Reserve office
arranges them in ascending rate order. The offices report the
amounts at each rate to the Treasury, which then combines the figures. The Treasury will accept all tenders at the rates that were bid,
starting with the lowest rate, until it covers the preannounced
amount of the auction. If there are more tenders at the highest
acceptable rate than needed, the Treasury will make partial
awards, proportionate to the sizes of the bids. Bidders seeking only
limited amounts, currently $1 million or less, are permitted to bid
noncompetitively. They receive the full amount of their tender at
the average rate that emerges in the competitive bidding.
Treasury notes and bonds pay interest in the form of a semiannual coupon. They are auctioned in the same way as bills, except
the bidder indicates a yield to maturity rather than a rate of discount
on the tender. The minimum tender size is $5,000 for maturities of
three years and less and $1,000 for longer maturities. The Treasury
generally sets the coupon rate at the nearest 1 /8 percentage point
that produces an average auction price slightly below par.
12 The formula for the rate of discount on a bill is: d=((F-P)/F)x(360/t), where d is the rate of
discount, F the face value, P the price paid, and t is the number of days to maturity. For
bills maturing in six months or less, bond equivalent yields, which are higher, are computed as follows: y=((F-P)/P)x(365/t), where y is the bond equivalent yield. Calculations
are more complex for longer time periods since they must account for the semiannual
coupon payments made on coupon securities.

86

The Treasury announces the results of the auction as soon as
they are computed, generally about two to three hours after the
bidding deadline. Depository institutions and primary dealers must
pay the full amount on the delivery date, usually a few days later.
Others must either have a bank or dealer guarantee their payment,
or they must submit full payment with the tender.
Dealers can judge what rates to bid for a new issue by talking to
customers and by trading in the secondary market. Trading begins
in new Treasury securities as soon as the Treasury announces an
upcoming auction. Dealers trade the securities prior to the issue
date in what is called the "when-issued" market. Instead of the
usual settlement in a day or two, settlement of such trades will take
place on the day that the Treasury delivers the security. The whenissued market allows dealers to sell "short" to customers ahead of
the auction date and to cover the sale in the auction.0
b. The secondary market
The secondary market is an over-the-telephone dealer market
with a large number of participants. Until recently, the secondary
market for Treasury securities was essentially unregulated. The
Government Securities Act of 1986 introduced regulation setting
financial responsibility and custody rules for brokers and dealers
in government securities. The rules were designed to preserve the
efficiency of the market and to encourage wide participation. As
Chapter 7 explains in more detail, the Federal Reserve maintains
a list of "primary dealers" from which it selects its trading counterparties, but other firms also trade, subject to rules specified in
the act.
In the most actively traded Treasury bills, competition is keen.
Bills are quoted in terms of discount rates. The spread between the
bid and asked rates quoted to customers is often only 1 to 2 basis
points—$25 to $50 per million dollars on a three-month maturity.
Coupon issues trade on a price basis (except for pre-auction whenissued trading, which is on a yield basis). Prices are quoted relative
to the par value of 100, and in increments of 1/32 of a point. Thus
a price of 99 31/32 means the issue is 1/32 point below par. As the
price falls, the yield rises; the amount of yield increase associated
with a 1/32 drop in price is largest for short maturities. It is close
to 2 basis points for a note maturing in two years. A 1/32 drop in
price lifts the yield on a 30-year bond by only 1/3 of a basis point.
13 When dealers sell short, they sell securities that they do not own on the assumption that
they can acquire them, either through purchase or loan, in time for delivery.




87




Bid-ask spreads on coupon issues depend on how actively the
issue trades and when it matures. Market spreads tend to widen
with maturity because the risk of price fluctuation increases.
Spreads generally range from 1/32 to 1/2 point or so, with small,
older issues at wider spreads. The spreads also depend somewhat
on recent market volatility. Trades can be for any size, although
transactions smaller than $1 million face value are considered to
be odd lots and are subject to an extra charge. Most dealers are
prepared to "make markets" to customers on the telephone for
amounts that are routine in size at that time. Generally, large orders
will be accommodated, but occasionally the dealer may need time
to assess the market before quoting a price.
The dealers trade actively with each other as a means of achieving inventories consistent with customer demands and with interest
rate expectations. Most interdealer trading is arranged through half
a dozen brokers specifically serving the dealer community. Dealers
post anonymous bids and offers through the brokers on issues they
wish to trade. Even after the trade is completed, the dealers do not
know their counterparties; they know only that they must be members of the group that has access to the broker. The broker is compensated by the dealer who hits a bid or takes an offering.
Trades are normally for the next day (regular delivery) or two
days forward (skip-day delivery), although same-day transactions
(cash delivery) can be arranged for bills in the morning. All bills
and most coupon issues are held in computerized "book entry"
accounts. The transfer of ownership is effected by depository institutions through the Federal Reserve's Fedwire transfer network.
Other owners must arrange to have a depository institution, generally a large money center bank, make transfers for them.
Securities are transferred in one direction and reserve balances are
transferred in the other direction simultaneously so that the party
selling the securities does not give up possession until payment is
assured and the party buying the securities does not give up the
money until the securities are transferred.
In the last few years, participation by foreigners in the U.S.
Treasury debt market has mushroomed. Interest by Europeans in
owning and trading securities encouraged the expansion of trading
in London. Japanese participation fostered a market in Tokyo.
Trading also occurs to a lesser extent in Australia, Singapore, and
many western European centers. The increased international trading in U.S. Treasury issues has led to expanded participation by foreign-based dealers and lengthened trading hours. Even before the

88

extensive foreign involvement, trading hours were never strictly
controlled because trading took place over the telephone rather
than on an organized exchange. Convention held that normal trading in the United States would take place between 9 a.m. and 4 p.m.
eastern time, but it was common for trading to range outside those
hours when there was significant news. In recent years, however,
trading hours have expanded to the point where securities are traded almost around the clock. Brokers operate during the Asian and
European trading days to serve those markets. U.S. firms can make
trades through the brokers by way of their Tokyo or London operations and will often trade through the London market when important U.S. economic data are released at 8:30 a.m. eastern time.
Government securities dealers perform a variety of tasks. In
addition to buying or selling securities at the request of customers,
they provide information, analysis, and advice to stimulate trading
activity and customer loyalty. To meet customer needs, they maintain inventories of government and other securities, financing them
through RPs arranged with corporations or other lenders or, if necessary, with relatively expensive bank loans. They manage their
securities positions with a view to profiting from both short- and
long-term swings in interest rates. They also engage in "arbitrage"
transactions by making offsetting purchases and sales to take
advantage of price disparities. For example, dealers can capitalize
on the price differences between securities of varying maturities,
or on price differentials between cash markets and futures and
options markets (discussed below).
Profitability for a dealer firm potentially arises from several sources.
A firm can realize a financing or "carry" profit when it earns a return
on securities owned that exceeds its costs to finance the securities. A
firm may make a position profit from having sold short (sold securities they did not own and borrowed securities to make delivery) in
falling markets and having gone long (held inventories of securities)
in rising markets. A firm may, in principle, make a trading profit from
the spread between bid and offer prices in trading with customers and
other dealers, although the business is sufficiently competitive that
bid-ask spreads are generally too narrow to serve as a significant
source of profit. Arbitrage transactions can be in a source of profit or
a protection against losses. Such transactions are often quite complicated, involving offsetting transactions in the cash, futures, and
options markets. Dealers generally have in-house traders who specialize in arbitrage. Such transactions are often kept separate from the
trading positions of those making markets to customers.




89




Achievement of substantial profits involves taking risks, since
competition limits the returns from risk-free operations. Hedging
strategies can be used to manage that risk, but implementing those
strategies can be costly. Consequently, dealer operations inevitably
show sharp fluctuations in returns that require a firm to be well capitalized if it is to succeed through a range of market conditions.
For most dealers, maintaining a sizable customer base is essential to success in the business. Knowing what customers prefer,
what securities they hold, and what they are doing, or thinking of
doing, enables the dealer to make markets intelligently, to judge
the likely market impact of prospective news developments, and
to manage the firm's own positions profitably. The key people in
the effort are the traders, who bid and offer close enough to the
competition to do business; the sales staff, who keep the customers in touch with the market and the firm in touch with its customer base; and the money market economist, who keeps the
traders informed of recent and prospective economic developments and the likely implications for the market.
Many dealer firms have branches in important domestic and international centers to maintain close personal contact with both large
and small customers; some of the major stock brokerage firms also
draw in retail customers through registered representatives in their
large network of stock-oriented branches. Other nonbank dealers
and most of the banks rely principally on direct telephone or telex
contacts, followed up with periodic personal visits. Leased wire information systems, which keep the customer abreast of the latest market and news developments, have greatly reduced the need for
routine informational calls by sales staff. The sales effort has shifted
toward providing computerized information on trading spreads and
arbitrage possibilities, as well as up-to-the-minute analyses of economic developments and the Federal Reserve's policy posture. The
rapid availability of information and analysis has eroded the dealers'
comparative advantage in day-to-day trading.
The financing of dealer positions has developed a life of its own.
Years ago, the dealers searched out the cheapest source of financing to increase the positive interest rate carry earned on their positions. (The carry is the difference between the interest earned on
the security held and the cost of borrowing the money used to purchase the security.) The dealers would try to minimize the negative
carry in periods when short-term financing rates were higher than
the longer term rates being earned on the security. To improve their
returns, both bank and nonbank dealers developed the sale of gov-

90

ernment and federal agency securities to corporations and other
lenders under agreements to repurchase the securities a day, a
week, or several months later at the same price plus an agreed rate
of interest for the period. Such RPs allowed investors to earn a
return on very short-term lending. (Until 1980, banks were not
allowed to pay interest on deposits of less than 30 days maturity.)
The financing cost to the dealer was typically below the Federal
funds rate or the dealer loan rates posted by the major banks—usually by an appreciable amount. Most lenders allowed the dealers
the right of substitution of collateral, so that the dealer could sell
securities on demand, replacing them with others.
Over the last 15 years, bank and nonbank dealers have run
matched books as well. They buy government securities for an
extended period under a reverse RP from a holder who needs
funds. Then they lend the securities on RP for an equivalent period
at an interest rate lower than the one they charge the seller. The
matching of maturities minimizes risks from price fluctuations. In
effect, dealers have gone into the banking business, taking care
that the credit standing of both customers assures the reversal of
the transaction. Dealers also protect themselves by taking a
greater margin of collateral on the securities acquired than they
give when lending the securities. Recent changes, some of them
prompted by the failures of a few dealers that had too little collateral and some introduced by the Government Securities Act,
increased protection of customers against inadequate collateral in
RP transactions. Dealers may also run an "unmatched book"—for
example, financing securities acquired under 60- or 90-day
reverse RPs with shorter term RPs to increase the interest rate
spread earned. Such activity runs the risk, of course, that financing
costs may rise in the interim and result in a loss rather than a profit.
Just as dealer position-taking is basically a bet on the future
course of interest rates, the unmatched book is a bet on future
financing costs. The resale value of the securities is fixed in the original contract, the reverse RP. In a straightforward position play, a
dealer may purchase six-month Treasury bills at auction, expecting
to finance at a positive carry for three months and then sell the bills
at three months to maturity for a gain that, over time, should equal
the average difference between three- and six-month bill rates over
the cycle. If interest rates were to fall over the interval, both the
carry earned and the yield-curve-based sales gain would be larger.
But if interest rates were to rise sharply, the carry could become
negative at the same time the price of the bill was declining. All




91

straight position decisions involve weighing the expected behavior
of both dealer financing costs and government securities prices.
Government securities dealers are extremely sensitive to the interest rate outlook because their positions at risk can be very large relative to their equity. A multiple of securities held to capital as large
as 50, aside from the matched book, is not uncommon for a nonbank dealer expecting a decline in interest rates. A 1 percent rise in
the price of securities held in such a situation would increase the
dealer's capital by 50 percent; a similar drop would wipe out half of
present capital. In practice, dealers tend to make substantial gains
by acquiring and financing an outright position during recessions,
when rates decline. When interest rates are choppy or rise persistently, however, dealers often encounter moderate-to-large losses, in
part because it is difficult to maintain effective markets for customers while holding a sizable net short position. Moreover, borrowing securities to sell short requires using scarce capital to execute
reverse RPs or pledging other securities and paying a borrowing fee
of 1/2 of a percentage point or more. This strategy requires securities prices to fall sharply for the dealer just to break even.
c. Derivative products




A number of innovations during the past 15 years have provided
new means for hedging interest rate risk or speculating on the
future course of interest rates. The new instruments have helped
dealers to manage their positions and have enabled a wide variety
of businesses to lock in costs or returns consistent with expected
cash flows.
Financial futures markets began to develop in the mid-1970s
and expanded in the early 1980s as interest rate volatility rose. The
growth of futures market activity has spawned so much arbitrage
and trading between the cash and futures market that they function
as a single market most of the time. Futures markets provide a
means of hedging against the effects of volatility, but by making
speculative bets easier, they may contribute to volatility at times.
Treasury bill and Eurodollar futures trade on the International
Monetary Market in Chicago. Futures on Treasury notes and bonds
and government-backed mortgage securities trade on the Chicago
Board of Trade. These contracts help increase liquidity and flexibility. They allow dealers to offset the positions they must maintain
to service customers—or to establish short positions—by entering
futures contracts to deliver the specified securities at a limited
number of specified dates over two years. The commission cost is

92

very small—around five dollars per contract on a "half turn" or a
single side of the futures transaction. The futures exchanges, which
are private corporations of exchange members, issue contracts to
buyers and sellers, each of whom must meet the low initial margins
set by the exchange. Initial margins are in a range of 1 to 5 percent
of the value of the instrument to be delivered. A clearing corporation marks each contract to market daily and requires additional
margin when a margin drops below the required maintenance
level, which is somewhat lower than the initial margin.
Options on Treasury securities and options on Treasury futures
contracts have been available since the latter part of 1982. They
expand the range of possible hedging strategies that can be used
to manage interest rate risk. Call options give the purchaser the
right but not the obligation to purchase from the seller the indicated
security or futures contract at a specified (strike) price at any time
before the maturity of the contract (a process known as exercising
the option). The purchaser benefits if the security or contract price
rises above the contract strike price, while the risk from price
declines is limited to the price of the option contract itself. Put
options give the purchaser the right to sell the security or the
futures contract at a set price within the period of the contract; thus
they benefit the purchaser in a falling market. Put options are like
a short sale but with limited downside risk for the purchaser.
Options on futures contracts are much more actively traded than
the straight options on securities. The writers of options contracts
take open-ended risks from a price rise in the case of a call or a
price fall in the case of a put. Writers may hedge this risk through
diversification or other techniques, but of course these techniques
may have costs that offset the gains from writing options.
Another form of derivative product based upon Treasury debt
instruments is the stripped security. Stripped notes and bonds are
zero-coupon instruments created by separating the coupons from
the "corpus," or principal, of a security and trading them separately.
Like Treasury bills, zero coupon debt instruments are sold at a discount. The return to the investor comes from increases in price until
maturity, when they pay the face amount. (As interest rates rise and
fall, their actual price will fluctuate around a rising trend line.) With
no periodic interest payments to reinvest, these securities have an
assured yield to maturity that is not dependent upon a reinvestment
return on intervening interest payments. They can be attractive to
pension funds and other entities with known future payment commitments. On the other hand, because all the return is deferred to




93

the maturity date, larger price changes will result from a given
change in the general level of interest rates than would occur if the
security returned its interest periodically. Consequently, stripped
securities may be attractive as a vehicle for speculation.14
Stripping of Treasury notes and bonds began during the 1970s.
Initially, dealers physically removed the coupons from the corpus,
since at the time coupon issues could be bought in definitive
(paper) form. Because stripping reduced tax revenues, the
Treasury discouraged the practice until 1982, when the tax laws
were changed. The new tax laws forced holders of zero coupon
and stripped Treasury securities to pay taxes each year on the
portion of the accrual representing the movement toward the par
value to be paid at maturity. The changes also required new
coupon debt to be sold only in book entry and not definitive form.
Physical stripping of older issues expanded once the practice was
no longer discouraged. Holding of stripped issues mostly attracted
entities that were not heavily taxed because the revised laws made
the tax burdensome.
Since the new book entry securities could not be stripped, a
number of government securities dealers created derivative instruments; they purchased Treasury issues, then placed them with a
custodian and sold separate rights to the various coupons and the
corpus. These receipts, called by a variety of proprietary names,
were popular for a time. In 1985, the Treasury began what is known
as the STRIPS program (Separate Trading of Registered Interest
and Principal of Securities). It permitted separate registration of the
coupons and corpus of the book entry securities and thus allowed
dealers to sell them to different purchasers. Later, the Treasury
provided the means to reconstitute a complete security if a party
had accumulated all the needed pieces. The STRIPS form soon
came to dominate the zero coupon market. Its popularity has
reflected the interest in zero-coupon products generally, which has
varied with perceptions of the future course of interest rates.
2. The Market for Federally Sponsored Agency Securities
A number of special purpose agencies with varying degrees of
federal government sponsorship sell debt to finance their support
of designated sectors of the economy, primarily agriculture and
housing. As of December 1988, regular debt outstanding (excluding pass-through securities, described below) totaled about $350




14 Sean Becketti, "The Role of Stripped Securities in Portfolio Management," Federal Reserve
Bank of Kansas City Economic Review, May 1988, pp. 20-31.

94

billion. The principal agencies are the Farm Credit System (FCS),
the Federal Home Loan Bank System (FHLB), the Federal Home
Loan Mortgage Corporation (FHLMC, also called "Freddie Mac"),
the Federal National Mortgage Association (FNMA, also called
"Fannie Mae"), the Government National Mortgage Association
(GNMA, also called "Ginnie Mae"), and the Student Loan
Marketing Association (SLMA, also called "Sallie Mae"). Except for
securities of GNMA and some other mortgage-backed securities as
well as some special issues noted below, agency debt obligations
are not explicitly backed by the full faith and credit of the U.S. government, even though the agencies are federally sponsored. There
are other government agencies that have access to the Federal
Financing Bank (FFB), which is funded by direct Treasury borrowing. Though these agencies have raised funds through the FFB
since 1974, some of them, such as the Postal Service, still have
securities outstanding that were sold in earlier years.
The FCS consists of a number of different regionally based institutions that provide credit to farmers. The system has undergone
restructuring in recent years as a result of financial difficulties in the
agricultural sector that left individual member institutions financially weakened or even insolvent. What had previously been a group
of largely independent regional institutions was combined into a
more centralized organization in January 1989 when the National
Bank for Cooperatives was formed. The problems that led to the
restructuring arose when the farm economy deteriorated in the
1980s. Institutions that were members of the FCS suffered dramatic
losses. To provide financial assistance to the system, Congress
passed the Agricultural Credit Act of 1987. The act, which went into
effect early in 1988, authorized the FCS to issue up to $4.0 billion
of 15-year bonds through the Financial Assistance Corporation
(FAC), with interest to be partially paid by the Treasury, and provided a $1.5 billion line of credit with the Treasury. The obligations
issued under the plan are fully guaranteed by the Treasury. In 1988,
the FAC sold two issues totaling $690 million. Yield spreads on
other FCS debt over comparable-maturity Treasury debt which had
risen dramatically relative to those of debt issued by other Federally
sponsored agencies when the financial problem developed, gradually declined during the year following the assistance package.
The FHLB, supervised by the Federal Housing Finance Board,
provides loans to member savings and loan associations and other
thrift institutions as a means of fostering the flow of funds into
home mortgages; the Federal Home Loan Banks are owned by the







member associations. The FHLB Board used to oversee the
Federal Savings and Loan Insurance Corporation (FSLIC), which
insured deposits of up to $100,000 at thrift institutions, but FSLIC
was abolished in 1989 and the thrift insurance fund transferred to
the FDIC As the thrift industry weakened in the early to mid-1980s
and FSLICs resources dwindled to the point where it was technically insolvent by the end of 1986, Congress and the Administration approved a plan under which the newly created Financing
Corporation (FICO) could issue up to $10.8 billion of debt to recapitalize FSLIC. Principal payments on FICO issues are backed by
zero-coupon Treasury bonds, while interest payments are secured
by a first lien on insurance premia paid to FSLIC. In 1989, the
Financial Institutions Reform, Recovery, and Enforcement Act
(FIRREA) was passed to resolve the thrift situation. It created the
Resolution Trust Company (RTC) and the Resolution Funding
Corporation (REFCORP). The RTC manages the liquidation of
assets of failed thrift institutions. REFCORP issues debt obligations
to fund the rescue operations. The first REFCORP sale was a 30year bond offering auctioned in October 1989.
FNMA, whose stock is traded on the New York Stock Exchange,
operates with guidance from the Secretary of Housing and Urban
Development. It buys government-insured and government-guaranteed mortgages and conventional private mortgages in the secondary market. FNMA issues its own debentures and notes and
guaranteed pass-through securities backed by pools of mortgages.
GNMA and the FHLMC also operate in the long-term credit market
to help finance housing. GNMA is a government corporation that
functions principally by guaranteeing pass-through securities. These
securities pass through to the purchaser the interest and dividends
earned on pools of government-guaranteed mortgages. The holder
of the securities receives a pro rata share of the principal and interest
payments earned on the mortgages. The FHLMC buys conventional
residential mortgages to foster a secondary market for them; it sells
pass-through securities and other bonds to finance its activities. The
FHLMCs voting capital stock was held solely by the FHLB, but under
the terms of the FIRREA, its voting stock will be publicly issued.
SLMA provides a variety of support services to institutions making loans to students. Before 1982, it borrowed directly from the
FFB, but since then it has borrowed in the market under its own
name. It issues primarily floating-rate debt. Occasionally it sells
fixed-rate debt, then converts its payment stream to a floating-rate
obligation through the use of swaps.

96

Federal agencies generally use a designated fiscal agent to manage sales to investors, although FICO, FAC, and REFCORP debt
has been sold in auctions. The fiscal agents sell their coupon securities to the public through separate selling groups composed of
about 25 to 120 banks and securities firms. Members of each
group are allocated securities on the basis of their past performance as distributors; fees paid to each member range from $0.30
to $3.00 per $1,000, and are scaled according to the maturity of
the security being sold.
The fiscal agents rely on major members of their selling group for
advice in choosing the maturities to be offered and the interest
coupons necessary to sell the securities. Because of the long-term
profitability of this relationship, members of the group characteristically take up the securities even when they think the pricing is
aggressive. The farm credit agencies sell 3-, 6-, and 12-month
paper monthly, while intermediate-maturity issues are sold at
longer and less regular intervals. The FHLB system sells short- and
intermediate-term issues each month. FNMA offers intermediateterm debt approximately monthly and occasionally sells long-term
issues. The fiscal agents also sell discount notes maturing in a range
of 5 to 360 days through designated dealers, who receive a commission of 5 basis points (0.05 percent) on notes they distribute.
Such notes are increased or reduced to reflect cash needs between
regular financings; rates are adjusted flexibly to attract funds as
needed. Most of the paper sold is in the 30- to 60-day area.
During periods when their credit quality has not been a particular
source of concern, agency issues have attracted wide investor participation. Because of their government sponsorship and supervision, the securities of the sponsored agencies generally trade at
yields only modestly above those on comparable maturity Treasury
issues. Although the yield differentials have usually been small
—occasionally as low as 5 basis points in the shorter maturities—
they have risen to 100 basis points or more during periods in which
financial difficulties of the issuing agencies have raised questions
about liquidity and creditworthiness. The yield differentials also
reflect the tax treatment. The FCS, FHLB, SLMA, FICO, and FAC
issues enjoy exemption of their income from state and local taxation, while those of FNMA, GNMA, and the FHLMC do not.
Most dealers in government securities make secondary markets
in these issues, although trading in many outstanding issues is
inactive. The size of some issues is small—as little as $200 million.
Perhaps only a dozen dealers are willing to make bids and offers in




97

most of the over 200 outstanding issues. Bid-ask spreads are related to the amount of activity in the secondary market. They are
generally wider than those on Treasury securities of corresponding
maturity. Federal agency discount notes, like commercial paper,
do not enjoy much of a secondary market. A dealer who distributes
the notes is usually prepared to make a bid to a customer who has
a pressing need to sell notes, but customers are generally expected
to hold the notes to maturity.
3. Corporate Debt Instruments
a. Commercial paper
One of the most rapidly growing sectors of the money market in
the last 15 years has been the market for the short-term promissory notes of creditworthy financial and other business enterprises.75
Corporate issuers are attracted by borrowing costs below those
available from banks; investors, by the yield premium offered over
Treasury issues. To be exempt from registration with the SEC, such
notes must mature in 270 days or less and be issued for working
capital purposes, such as financing inventories and accounts
receivable. The most popular maturities depend somewhat on the
rate structure at the time of issuance, with 15- to 60-day paper
generally the most common.
Commercial paper is sold to money market investors either
directly by a firm's own sales force or through a dealer that makes
sales on behalf of many borrowers. Direct placement is characteristic of large finance and credit companies, which are often affiliates of automobile and other manufacturers, and of bank holding
companies. Around 40 to 45 percent of the approximately $450
billion in commercial paper outstanding in the latter part of 1988
was placed directly by such firms with customers. The remainder
was placed by a small number of major dealers that have specialized sales forces. Among the approximately 1,000 companies
issuing through dealers are several hundred industrial companies
and public utilities, and more than 100 each of bank holding companies and smaller finance companies. Foreign banks and a few
foreign government agencies also borrow in the market.
Most commercial paper is sold by companies that have good
credit ratings. Some small- to medium-sized firms may obtain a
letter of credit from a bank—in most instances a foreign bank—that
will allow the firm to obtain a strong credit rating. The credit rating




15 Municipalities may occasionally issue commercial paper.

98

companies—Standard & Poor's (S&P), Moody's, and Fitch's—assign
numerical ratings to a company's debt after a careful review of the
company's balance sheet and operations. S&P and Fitch use ratings of A-l, A-2, or A-3 and Moody's uses ratings of P-l, P-2, or
P-3. Perhaps three-quarters of all paper sold is in the top A-l/P-l
grade. Most of the remainder is graded A-2/P-2. Investors have
generally shied away from lower graded paper, although some
issues of speculative paper have been purchased by investors willing to take a risk in return for a high yield.
Paper issuers generally establish bank credit lines to provide a
liquidity backup that will cover the amount of paper they expect to
have outstanding. In contrast to the irrevocable letters mentioned
above, these more standard lines do not provide a guarantee of the
outstanding paper. The rating agencies do not deem full coverage
necessary for financial institutions with liquid portfolios or ready
access to a central lending agency. In theory, bank credit lines
require maintenance of deposits with the bank equal to 10 percent
of the credit line, with an additional compensating balance equal
to 10 percent when the line is drawn on; in practice, issuers rarely
maintain a balance at such a level. Some issuers pay a fee rather
than maintain compensating balances.
Allowing for the cost of the liquidity backup, issuers of commercial paper can usually save between 1 and 2 percentage points
over the cost of borrowing from a bank at a rate tied to the prime.
Recently, domestic and foreign banks have recaptured some of this
business by making loans for 30 days or less at a markup over the
cost of funds to the bank.
Commercial paper, which continues to be in definitive (paper)
form, is sold at a discount and is redeemed at par at maturity. The
smallest denomination for dealer-placed paper is $100,000; blocks
of $5 million or $10 million are more common, especially on
directly placed paper. Paper is lodged by the company with a New
York bank, which countersigns and delivers the notes to the commercial paper dealer for payment that same day.
Dealers usually take down paper as it is sold at quoted rates, but
they also are prepared to buy paper at a concession for inventory
when an issuing company's needs are pressing. Dealers are usually prepared to buy back paper they have sold to an investor, but in
fact only a small percentage comes back. Some dealers carry very
little inventory for their own account; inventories of others may run
around $400 million and on occasion have reached $1.5 billion,
typically financed at rates above those prevailing for Treasury




99

issues. Inventories tend to be largest when financing rates are
below the return on paper. When financing rates exceed the return,
dealers attempt to minimize inventories. Spreads between the rates
at which paper is bought and sold are around 10 basis points.
Holdings of commercial paper by MMMFs grew very rapidly in
the late 1970s and early 1980s; however, the pace of accumulation
slowed after 1984, and the MMMF share of total commercial paper
outstanding declined from a peak of around one-third in that year
to about one-fourth by the end of 1988. Some of the slack has been
taken up by households and pension funds. Other important
investors are insurance companies, business corporations, and
commercial bank trust departments that manage investments for
the account of corporate customers.
b. Corporate bonds




While commercial paper helps satisfy the short-term borrowing
needs of many firms, corporate bonds are issued to provide longer
term financing. They are often classified by the type of issuing firm:
public utility, transportation, industrial, financial, or real estate.
Because the volume of corporate equities, an alternative source of
long-term financing, actually fell in the 1980s, net new issuance of
corporate bonds grew very rapidly. Nonfinancial firms accounted
for around one-half to three-quarters of net new corporate debt.
Unlike Treasury securities, corporate bonds may carry a significant risk of default. The risk for a particular issue depends in part
on how it is secured; mortgage bonds are secured by a first lien on
property or equipment, collateral bonds by the holding of securities, and debentures by whatever unpledged assets remain at the
time of liquidation.76 But regardless of the type of security, investors
rely on the credit ratings assigned by the major ratings agencies
such as Moody's and S&P. These ratings range from Moody's Aaa
(or S&P's AAA) for prime-grade issues down to C for the poorest
prospects (or S&P's D for issues actually in default). More highly
rated issues are naturally offered at lower yields.
Federal Reserve data show that typical corporate bond maturities have fallen in recent years; the proportion of new domestically
offered issues in the under-10-year range rose from around 35 percent in 1983 to 55 percent in 1988 as the share of new offerings in
the 20-or-more-year range fell from about 35 to 15 percent. The
longer maturity issues can normally be called by the issuer at a
16 The term "bond" is often used genetically to refer to debentures and notes, which have
no specific pledged collateral.

100

prearranged price after an initial period of 5 to 10 years. The issuer
will usually call a bond if interest rates have fallen far enough to
allow refinancing at lower yields. Even if an issue has no call feature, some of it may be retired before the bond reaches its nominal
maturity by means of a "sinking fund" provision. The provision
requires the issuer to retire gradually a specified portion of the
issue each year; in some cases, the fund requires retiring all of the
issue by maturity, but in other cases, a single "balloon" payment at
the end may be necessary to retire the remaining debt. Sinking
fund provisions are characteristic of industrial bonds but are
almost never attached to financial issues. Most corporate bonds
pay interest semiannually, though a relatively small volume of
zero-coupon corporates has been issued as well.
Corporate bonds are usually sold to the public through underwriting syndicates formed by investment banking institutions that
have corporate bond divisions. In 1989, a Federal Reserve Board
ruling gave commercial banks limited authority to underwrite corporate debt through their securities subsidiaries. The firm acting as
lead manager recommends the maturities and types of issues
believed to be consistent with the issuer's financial needs and tests
market appetite through conversations with potential large buyers.
Each member of the syndicate will be allocated securities to place
with its customers. In some cases, a whole issue may be placed
privately with a large investor, generally an insurance company,
and never be sold publicly.
Public offerings must be registered with the SEC, which requires
the corporation to report actual and potential obligations that
might affect the ability of the corporation to repay the debt. Since
1982 the commission has permitted "shelf registration," allowing
corporate issuers to register their intent to issue debt without specifying issuing dates or amounts (Rule 415). Consequently, issuers
can bring the debt to market relatively quickly once the decision is
made to offer it. Valid for a period of three years, shelf registration
reduces the average cost per offering and allows greater flexibility
than issue-by-issue registration. The provision has given rise to a
market for medium-term notes, issues with maturities of 5 to 10
years that can be offered whenever market conditions are favorable, much like commercial paper. Insurance and pension funds
dominated net purchases of corporate bonds in the early 1980s;
life insurance firms were particularly active. Households began to
step up net purchases significantly in 1984 and were joined by
commercial banks and foreign investors a couple of years later.




101




Nearly all secondary market trading of corporate bonds takes
place in the over-the-counter market, with the residual occurring on
organized exchanges such as the New York Stock Exchange. The
over-the-counter market is made by securities dealers who trade
directly with other dealers and with large institutional investors.
Given the vast number of outstanding corporate issues, the market
for most individual issues is illiquid, although the market for particular types of bonds may be fairly broad. Since corporate bonds are
less liquid than Treasury securities, the bid-ask spreads quoted by
dealers normally exceed those on Treasuries; investment-grade
corporate spreads typically range from 1/8 to 3/8 percent, while
spreads for lower rated issues are larger. Issues listed on an
exchange are more liquid than otherwise similar unlisted issues.
An important development in the corporate bond market in the
1980s has been the explosive growth of relatively risky bonds, generally referred to as "high-yield" or "junk" bonds. These issues offer
nominal returns which have generally been about 2 to 7 percentage points above those on Treasury issues of comparable maturity
(higher if the issue is considered to be in imminent danger of
default). They carry Moody's ratings of Bal or less and S&P ratings
of BB+ or less. Some investors, such as commercial banks, cannot
hold high-yield bonds but are limited to investment-grade issues
carrying ratings of Baa (Moody's) and BBB (S&P), or higher. FIRREA stipulates that thrift institutions will have to divest their holdings of lower rated bonds over five years.
Junk bonds appealed to many investors who were attracted by
their high yields. From the issuer's point of view, high-yield debt
was an attractive means of financing risky corporate acquisitions;
the intent was to sell newly acquired assets to provide funds to pay
interest and principal on the debt. To some extent, the expansion
of junk bond issuance represented a substitute for private placements of unrated bonds primarily with insurance companies.
(Unrated private placements remain a popular means of finance.)
To a greater extent, it represented conversions, from equity finance
to debt finance, undertaken at least partly for the tax benefit. In
some cases, the company's management bought outstanding
common stock, financing the purchase through sales of bonds. In
other cases the conversions were accomplished in a takeover by
an outside interest. From 1982 to 1986, new issuance of junk
bonds expanded at about an 85 percent annual rate,77 reaching a
17 From data reported by Morgan Stanley.

102

point where the bonds accounted for about a quarter of new corporate debt issuance. New issuance leveled off after 1986.
The market for junk bonds has received a number of setbacks,
each of which led to widened spreads and a reduction of issuance.
In 1986, it faced a default by LTV, a big steel producer, and a protracted insider-trading scandal; in October 1987, the stock market
crash reduced the attractiveness of issues to finance takeovers and
leveraged buyouts; and in 1989 Campeau Corporation, a real estate
and large retail firm with an extremely heavy volume of debt outstanding, struggled to raise cash to meet debt service payments and
avoid bankruptcy. Its troubles made investors more sensitive to the
relationship between debt service costs and revenues, and more
selective in their purchases.
The higher yields offered by junk bonds relative to investmentgrade issues should compensate for the higher risk of default. It is
an open question whether the spreads between the two types of
bonds are appropriate. Some studies have concluded that a diversified junk bond portfolio will earn enough over time to more than
compensate for the higher level of defaults. Other studies, using
different methodology, have reached the opposite conclusion.78 In
any case, the substantial increase in volume of junk bonds in the
1980s means that studies drawn from an era when the market was
smaller may not be totally applicable to the future.
4. Municipal Securities
Municipal securities are issued by state and local governments
and by special authorities providing services such as housing,
education, transportation facilities, and industrial development.
Issues maturing in one year or less are generally referred to as
"notes," while longer term obligations are known as "bonds"; the
great bulk of funds raised in the municipal market take the form
of bond offerings.79
Until recently, the distinctive feature of municipal bonds had
been their exemption from federal income taxes as well as from
state and local taxes for investors residing in the state in which the
18 The classic study supporting the first view is Walter Braddock Hickman, Corporate Bond
Quality and Investor Experience (Princeton: Princeton University Press, 1958). For a more
recent analysis, see Jerome S. Fons, "The Default Premium and Corporate Bond
Experience, "The Journal of Finance, vol. 42 no. 1 (March 1987), pp. 81-97. For a contrary
view, see Paul Asquith, David W. Mullins, Jr., and Eric D. Wolff, Original Issue High Yield
Bonds: Aging Analysis of Defaults, Exchanges, and Calls," The Journal of Finance, vol.
44, no. 4 (September 1989), pp. 923-52.
19 See Public Securities Association, Fundamentals of Municipal Bonds, 3rd ed., 1987.




103




securities were issued. Because of the bonds' unique status, municipalities could issue high-grade debt at yields that were usually
below those on taxable bonds of comparable maturity. But
changes in the Federal tax code, particularly the Tax Reform Act
of 1986, have eroded the tax-exempt status of municipal bonds.20
Among the more important changes are: 1) treatment of private
purpose (industrial development) bonds issued after August 7,
1986 as a "preference item" under the alternative minimum tax, so
that the tax-exempt status of municipal issues now varies from
investor to investor, 2) limitations on the volume of partially taxed
private purpose issues that can be offered annually in each state,
3) elimination of the deductibility of the interest cost incurred by
financial institutions in borrowing to buy municipal bonds, and 4)
a prohibition against the practice of issuing low-yielding municipal
bonds to invest in higher yielding debt. After the reforms restricted
the ability of municipalities to issue tax-exempt debt, a few municipalities began to offer fully taxable issues.
Like corporate debt, municipal bonds carry some risk of default.
Investors were reminded of that risk in 1975, when New York City
temporarily defaulted on a note issue, and again in 1983, when
Washington Public Power Supply System defaulted on $2.25 billion
of bonds. Substantial help in assessing the likelihood of default is
offered by Moody's and S&P.2i Before assigning ratings, the firms
must consider the security for each issue. Generally, municipal
bonds can be secured in one of two ways: "revenue" bonds are
issued to finance specific projects, and the proceeds of those projects, normally in the form of user fees, are used to service and
retire the debt; "general obligation" bonds are backed by the full
faith and credit of the issuer, which can use its taxing authority to
raise funds to pay interest and principal on the bonds. Some issues
are hybrids of the two types, and a sizable proportion of new debt
is independently guaranteed by firms that specialize in municipal
bond insurance.
Investors in municipal bonds were traditionally drawn by the taxexempt feature. The three major groups of investors have been
households (including mutual funds), commercial banks, and
property and casualty insurance firms. The tax reforms encouraged households, the group of investors with the largest holdings,
20 The word "bond" should be understood to mean both notes and bonds throughout this
section, unless otherwise indicated.
21 The ratings schemes for municipal bonds are similar to those for corporate bonds. For
municipal notes, Moody's uses the symbols M1G 1 to M1G 4, while S&P uses SP-1 to SP-3.

104

to expand that share to around 50 percent by the end of 1988,
according to Federal Reserve statistics. Although lower marginal
tax rates and the alternative minimum tax have reduced the attractiveness of municipal bonds, other tax shelters available to households have been curtailed as well. By contrast, the elimination of
the deducibility of carrying costs has greatly diminished the aftertax yields on municipal bonds for commercial banks; their share of
holdings, which was already falling in the 1980s, declined further
to about 25 percent by the end of 1988. Property and casualty
insurers have generally invested in municipal securities when they
have profits to shelter. These firms reduced their holdings of
municipal debt during several unprofitable years through the mid1980s. But as tax reform increased accounting profits by reducing
the extent to which insurers could carry losses forward, some firms
again found "munis" a profitable investment.
New public offerings of municipal bonds may be marketed either
by competitive bidding among underwriters or through directly
negotiated under writings. Underwriting is done by investment and
commercial banks. Most general obligation issues are competitively offered, while revenue issues may be underwritten through either
method. Once distributed, issues trade in a fairly active secondary
market. Though municipal issues are not listed on formal exchanges, the secondary market is maintained by hundreds of dealers nationwide. Transactions may be carried out by phone, and
issues are advertised both through the Bond Buyefs "munifacts"
teletype system and through S&P's Blue List publication. Typical
bid-ask spreads quoted by dealers for retail investors are about 2
points, while spreads for institutional investors tend to run around
1/2 point or less.
This review does not cover some other major components of the
U.S. financial markets, such as the mortgage market (although
touching on some of the Federally sponsored agency issues that
are backed by mortgages), and the market in equities.




105

The FOMC Meeting:
Developing a Policy Directive
At each of its eight scheduled meetings a year, the Federal Open
Market Committee (FOMC) develops its policy priorities and prepares instructions for carrying them out. At the February and July
meetings, the FOMC focuses explicitly on annual growth ranges for
specified monetary and credit aggregates and sets forth the range
of expectations among FOMC members and other Reserve Bank
presidents for real GNP, inflation, and employment. This review
provides the Chairman with information for the semiannual testimony before the House and Senate Banking Committees required
by the Humphrey-Hawkins Act.7 At the other six meetings each
year, the longer run money and credit targets may be reviewed if
special conditions so require. But any such discussion is likely to
be shorter and less formal.
At all of its meetings, the FOMC develops policy specifications
to guide the domestic trading desk at the Federal Reserve Bank of
New York. The FOMC discusses the outlook for economic activity,
the monetary aggregates, inflation, and financial market conditions. It weighs information from a variety of sources and considers
the likely consequences of alternative policy prescriptions. Table
1-5, page 107, presents a typical FOMC meeting agenda.
Preparation




In advance of each FQMC meeting, documents are prepared and
circulated to those who will attend and to other staff members at
the Reserve Banks who brief their presidents. Three of these documents are described by the colors of their covers as the green
book, the blue book, and the beige book.
The green book presents the Board staff's detailed appraisal of
the forces currently at work in the major economic sectors and in
the financial markets and summaries of the outlook for domestic
economic activity, prices, and the international sector. Tables present quantitative forecasts of a number of key economic and financial variables for the current and upcoming calendar years.
Generally, the forecasts take into account the longer run monetary
growth rate ranges most recently adopted. While the forecasts
make use of structural econometric models, the end product relies
heavily on the judgment of the senior staff members.
The blue book provides the Board staffs view of recent and
prospective developments related to the behavior of money, bank
reserves, and interest rates. In the blue book prepared for the
1 Formally titled the "Full Employment and Balanced Growth Act of 1978."

106

Table 1-5 Federal Open Market Committee Meeting
Sample Agenda
1.

Approval of minutes of actions taken at the last meeting of the Federal Open
Market Committee.

2.* Foreign currency operations.
A.

Report on operations since the last meeting.

B.

Action to ratify transactions since that meeting.

3.* Domestic open market operations.
A.
B.
4.

Report on operations since the last meeting.
Action to ratify transactions since that meeting.

Economic situation.
A.
B.

5.

Staff report on economic situation.
Committee discussion.

Longer run ranges for monetary policy (February and July meetings).
A.

Staff comments.

B.

Committee discussion and actions on longer run ranges.
1.
2.

6.

Review of ranges for year in progress.
Establishment of tentative ranges for following year (July meeting).

Current monetary policy and domestic policy directive.
A.
B.

Committee discussions.

C.
7.

Staff comments.

Action to adopt directive.

Confirmation of date for next meeting.

*At the February and July meetings, reports on operations in foreign currencies and the
domestic securities market and their discussion are sometimes deferred until after the
longer run ranges are developed.




107

February meeting, the staff presents monetary growth scenarios for
the year just beginning. In July, the blue book offers a review and outlook for the year in progress and a preliminary look at the following
year. Both the February and July blue books set out alternative
annual ranges for growth of the monetary aggregates to be presented in the Humphrey-Hawkins testimony later in the month. Also
noted are developments that might lead the Committee to reconsider its earlier target ranges for the aggregates for the current year.
All eight blue books provide the Board staffs view of monetary
and financial developments for the few months surrounding the
meeting in question. The staff typically offers three alternative combinations of reserve pressures and three-month growth rates for
those monetary aggregates being targeted, along with amounts of
discount window borrowing and short-term interest rates expected
to be associated with each alternative. It identifies the likely effects
of the suggested combinations on future economic activity. The
Committee traditionally targeted Ml, M2, and M3, but the demand
for Ml shifted so much in the 1980s that the FOMC stopped setting
targets for that aggregate. To aid its judgment, the staff draws on
simulations of money demand in monthly and quarterly econometric models. In addition, recent special studies of factors affecting the
demand for money may be employed, particularly at times when
changing institutions or changing perceptions of inflation seem to
be altering the previous relationships.
The beige book, made available to the public shortly before each
FOMC meeting, presents reports on regional economic conditions
in each of the 12 Federal Reserve Districts. The reports are compiled from conversations with local business leaders and analyses
of statistical reports for the area. A summary of the conditions
described by the 12 Banks leads off the report.
Before every FOMC meeting, a series of briefings and discussions are held. Members of the Board staff review their economic
forecasts with the governors. Each Reserve Bank president discusses with staff members policy options that will be addressed at
the FOMC meeting. Reserve Bank research officers may also present their own review of economic and financial developments,
delineating any differences they have with the Board staffs monetary or economic outlook.
The Meeting




The FOMC meetings take place in the boardroom of the Board of
Governors in Washington. The seven governors and 12 Reserve

108

Bank presidents gather around a long conference table, along with
the Secretary of the FOMC, Board staff members serving as advisers
to the FOMC, and two officers of the New York Reserve Bank—the
Manager of Foreign Operations and the Manager of Domestic
Operations. Senior research officers of the Reserve Banks, other
senior Board officials, and an officer from the New York Reserve
Bank's domestic trading desk sit around the sides of the room.
1. Reports of the Managers
a. The report on international developments
The Chairman generally opens the meeting by seeking approval
of the minutes of the previous meeting. The next order of business
usually involves reports by the Managers for foreign operations
and for domestic open market operations. At the February and
July meetings, however, these reports are occasionally deferred
until after the discussion of the longer term policy specifications.
The Manager for foreign exchange operations reviews developments in the exchange markets during the period and describes
any intervention by the foreign trading desk to buy or sell dollars
against foreign currencies. The Manager gives reasons for such
intervention, highlighting information contained in a more detailed
written report submitted previously. Following questions and discussion of international developments, including the response of
the exchange markets to U.S. economic statistics, the actions of
foreign trading desk generally are ratified by the Committee.
The FOMC and the Foreign Exchange Department of the New
York Bank are important players in the conduct of foreign
exchange policy, but responsibilities in this area are shared by the
Federal Reserve and the U.S. Treasury. The Treasury has broad
responsibilities for the country's international financial policy while
the Fed plays a key policy and technical role in matters touching
the foreign exchange markets. Exchange rate policies and operations are the subject of continuing conversation between the
Treasury and the Federal Reserve, and the actual operations are
carried out on behalf of the Federal Reserve System and for the
Treasury by the Federal Reserve Bank of New York. The Chairman,
other FOMC members, the Manager for foreign operations, and the
Board staff periodically exchange views with senior Treasury officials on U.S. policy in the foreign exchange markets.
The FOMC adopts general guidelines and monitors activity with
respect to foreign exchange operations by the Federal Reserve
System. The Federal Reserve operates for its own account in the




109




foreign exchange markets through the foreign trading desk at the
New York Federal Reserve, which also serves as agent for the
Treasury in the Treasury's foreign exchange operations.2 As the
nation's central bank, the Federal Reserve has close working relationships with foreign central banks, and the New York Reserve
Bank may execute foreign exchange transactions on their behalf in
the CJ.S. market.
During the meeting, the Manager for foreign operations may
comment on the views expressed by the central bank governors of
major industrial countries at their monthly meeting at the Bank for
International Settlements in Basle, Switzerland. At times, the
Chairman or one of the Committee members reports on discussions of current issues with foreign officials or with the Treasury.
Committee members routinely question the Manager about foreign
central bank policies and developments in the exchange market.
They may ask the Manager to predict how the exchange market
would react to a change in the Federal Reserve discount rate or in
the approach to supplying nonborrowed reserves. Members often
ask the Board staff about the U.S. balance of payments and other
international economic developments.
The Committee is also responsible for monitoring operations executed under the Federal Reserve swap network—reciprocal credit
lines established between the Federal Reserve and other central
banks. From modest beginnings in 1962, the network grew to $30.1
billion of standby credit in October 1989, negotiated between the
Federal Reserve and 14 foreign central banks plus the Bank for
International Settlements.3 The swap network is occasionally used
by the Federal Reserve or one of its partners to supplement its holdings of foreign currency for intervention in the foreign exchange
markets (although the frequency of use has diminished in recent
years because the Federal Reserve and other central banks hold
foreign exchange balances outright). From time to time, swaps
have also been used temporarily by countries in the process of
negotiating major borrowings to refinance outstanding loans. Swap
drawings can be arranged quickly by telephone or other means,
2 The effects of intervention on bank reserves are discussed in Chapter 6, Box B. On occasion,
the United States has expanded its foreign currency resources by drawing down its reserve
position at the International Monetary Fund through sales of Special Drawing Rights, and
it has issued securities denominated in foreign currencies in the capital markets of other
countries. These operations are conducted by the U.S. Treasury. Intervention has occasionally been financed with swap lines, described below.
3 See Sam Y. Cross, "Treasury and Federal Reserve Foreign Exchange Operations," Federal
Reserve Bank of New York Quarterly Review, Autumn 1989.

no

with the central bank supplying the currency needed. Such drawings, which pay interest, are typically arranged for three-month
periods and can be renewed for additional three-month terms by
mutual consent. Over the years the FOMC has maintained the principle that swap drawings are to be fully repaid within a year.
The Manager for foreign operations reports to the Committee on
any outstanding swap drawings that are approaching maturity and
outlines current plans for the drawing party to repay. The Committee,
aided by its staff, also reviews and approves any additions to individual country swap lines and any changes in the instructions under
which the Manager conducts foreign exchange operations.
b. The report on domestic operations
Once the Committee has considered the foreign exchange issues
before it, the Manager for domestic operations reports on the implementation of the Committee's directive since the last meeting. The
Manager typically reports whether the policy instruments have been
held as initially specified by the FOMC at its last meeting. If they
have been changed, the Manager relates the reasons for the adjustment, commenting on the monetary and economic conditions that
precipitated the change. The Manager then describes the behavior
of nonborrowed and borrowed reserves relative to the objectives set
for them and explains any special factors that might have led to
either deliberate or unintentional misses. Such misses may occur
when the behavior of excess reserves deviates from normal assumptions or when banks modify their willingness or reluctance to borrow
at the discount window. Large errors in projecting the behavior of
technical factors affecting reserves can also cause deviations. If in
the period since the last meeting, the normal relationships between
nonborrowed and borrowed reserves and reserve pressures had
seemed to shift, the Manager, in consultation with senior Board staff
and the Chairman, would have had to make choices as to what, if
any, modifications were appropriate in the desk's immediate objectives. The oral comments at the FOMC meeting would indicate how
such situations had been handled. These comments supplement
material in a written report submitted in advance that covers daily
domestic operations and changes in the System portfolio over the
intermeeting period.
In the presentation, the Manager discusses financial market
developments in the intermeeting period, with special emphasis on
the Treasury market and the Federal funds market—the overnight
market in bank reserves. The Manager reports any changes in




111

market sentiment about the interest rate outlook, shifting perceptions of the pace of economic activity and inflation, and market
views of the likely course of Federal Reserve policy. Attention may
be given to legislative initiatives, budgetary developments, or
changes in the market response to developments in the foreign
exchange markets.
If unusually large needs to add or drain reserves are expected to
develop in the upcoming intermeeting period, the Manager may
ask the Committee to amend the authorization for domestic operations to permit a larger than normal net change, or "leeway," in
the System portfolio over the period. As of 1989, the normal intermeeting leeway was $6 billion.4 (Chapter 7 gives a more detailed
description of the authorization for domestic operations.)
Following the Manager's report, FOMC members may comment
or raise questions about operations or market developments. They
may ask about market expectations concerning future policy.
Finally, the Committee is asked to ratify the operations conducted
over the interval.
2. Sizing Up the Economic Situation
a. The Board staff presentation
Members of the Board staff then review current and prospective
economic and financial developments, summarizing the material
presented in the green book. Typically, the forecast horizon
includes both the current year and the following year. At the
February and July meetings, the presentation is more extensive
than at other meetings and includes numerous charts and reports
from specialists in several fields. The staff presentations explain
the green book's forecasts and comment on its estimates of a number of measures, including prospective output, employment, and
prices. The staff members examine factors underlying the forecasts such as expected personal consumption expenditures, business fixed investment, home sales, government spending, trends
in labor productivity, wage settlements, and unit labor costs. The
analysis may touch on the extent to which tax policy or inflationary
expectations are influencing consumer spending and home construction. Business capital spending plans may be evaluated.
The international staff presents its views of the output, growth, and
price performances expected abroad in relation to U.S. performance




4 On occasion, the need for greater leeway is not foreseen at the time of a meeting, but shifting
conditions introduce a need between meetings. The Committee members are advised by the
Chairman of the need during the intermeeting period and are canvassed for their votes.

112

and the implications for trade and current account balances and the
dollar's exchange rate. The staff goes on to outline expected credit
flows and interest rates associated with established policy priorities.
Following the staff presentation on the economy, Committee
members generally ask a number of specific questions about the
assumptions underlying the forecasts. They may inquire how economic forecasts would change if certain factors behaved differently. They often ask senior staff members whether a forecast is more
likely to err on the high or the low side.
b. Discussion of the economy
The Committee members (and the nonvoting Reserve Bank presidents) then present their views on the economy, including the outlook. At the February and July meetings, they will have submitted in
advance their individual estimates of economic growth, inflation, and
unemployment. They may defend or modify those forecasts as a
result of thet discussion. The ranges and central tendencies of these
forecasts are included in the Humphrey-Hawkins report.
In the commentary, the Board staffs latest forecasts for the
economy are used as a benchmark. In giving their assessments of
the economy, the governors and presidents typically note the areas
of agreement or disagreement with the staff. The speakers employ
a range of analytical approaches to the topic when they comment.
Some build their conclusions about the total economy from the
spending dynamics they observe in the consumer, business, and
government sectors. Others may focus on particular factors that
they believe play a key role in indicating the direction in which the
economy is headed. A policymaker may cite developments in one
or a few key industries or single out recent interest rate movements, commodity prices, or the exchange value of the dollar as a
key indicator of future economic activity and inflation. Others put
a heavier emphasis on past monetary or credit expansion. Some
Reserve Bank presidents may remark on regional developments,
but typically they stress that their view of appropriate monetary
policy is not based on merely local factors.
3. The Longer Run Monetary Targets
a. Presentation
Generally at this stage of the February and July meetings, the
Committee turns its attention to the longer run ranges for money
and credit growth. At the start of the discussion of the annual rates
for money growth, a senior Board staff member, usually the Director




113

of the Division of Monetary Affairs (the key staff member responsible for policy implementation), highlights the significance of the
material covered in the blue book. This speaker describes the alternative annual growth rate ranges and the reasoning behind them.
The Director may offer comments on income velocity—the rate
of turnover of money relative to GNP. As noted in Chapter 1, until
the early 1980s, Ml velocity had followed a reasonably predictable
trend, growing at an average annual rate of about 3 percent between
1960 and 1980. Velocity generally accelerated during expansions,
especially when inflationary pressures built up and interest rates on
alternative liquid assets rose. It increased more slowly or declined
during recessions, when interest rates fell and inflationary pressures
subsided. Accordingly, descriptions of velocity during those years
had focused mostly on the stage of the business cycle.
The tenor of the presentation on longer run monetary targets
gradually changed after it began to appear in 1982 that the traditional velocity relationships were breaking down. As Chapter 1
indicated, a combination of interest rate deregulation, the creation
of new types of deposits, and sharp shifts in actual and perceived
inflation was changing both the demand for money and its interest
elasticity and thus the relationship between money and GNP. The
staff presentations during the 1980s often included comments on
demand shifts observed to date and cited recent research results
in an attempt to shed light on the changes. For instance, the presentation might have focused on how money demand was adjusting to the removal of an interest rate ceiling from a particular type
of deposit. Alternatively, it might have examined how banks were
changing their methods of establishing compensating balance
requirements in the face of declining or rising interest rates.
While cognizant of sharp variability in velocity, the staff and the
FOMC have retained the belief that the behavior of money over
time has an important influence on inflationary pressures.
Accordingly, they have increased the attention given to the broader
aggregates, M2 and M3, in the wake of the distortions to Ml behavior. M2 and M3 have not been immune to the changes in financial
practices taking place, but generally they have seemed less
severely distorted.
b. Developing annual ranges
Following the staff presentation, the Chairman calls for a general
discussion of the ranges of money and credit growth to be adopted for
the current calendar year and, at the July meeting, tentative growth




114

ranges for the next year. As the governors and presidents speak, they
may spell out the growth objectives they consider appropriate.
A starting point for the Committee's deliberations is the blue
book formulation of two or three families of annual growth rates for
the monetary and credit aggregates being targeted or monitored,
and the economic, price, and interest rate patterns expected to be
associated with each option. Recently the aggregates have been
M2, M3, and nonfinancial debt. Each participant either signifies
agreement with one of these options or proposes different ranges.
A speaker may expect that the relationships among the growth
rates of the monetary aggregates will differ from the staff's estimates. Therefore, he or she may want to change the range for one
of the aggregates while concurring in the staffs suggestions for the
others. Some Committee members may express a preference for
widening the growth ranges to encompass the staff forecast for
nominal GNP while allowing for a wide range of behaviors of velocity. Others may suggest narrower ranges as a means of imposing
a tighter discipline on the policy process. All are concerned about
the message that will be conveyed if the new annual objectives do
not seem consistent with the past performance or the desired
future performance of the economy.
In making policy choices, Committee members interpret the staffs
recommendations in light of their own views. For example, while
acknowledging that the staffs projection of the economy's performance is both reasonable and desirable, a member may still conclude that money will have to grow more rapidly than the staff judges
necessary in order to accommodate a larger-than-predicted shift in
the money demand function. Another member may believe that the
rise in interest rates that the staff associates with achieving the money
growth objective will in fact keep the economy from attaining its projected level of performance. At times, members may essentially
agree with the staffs formulation of interest rate, monetary, and GNP
relationships but wish to see faster monetary growth. They may reason that excess capacity is sufficient to permit more rapid economic
expansion than the staff believes feasible without creating undue
inflationary pressures. At other times, individual members may desire
slower growth of the monetary aggregates. Rapid money growth is
most likely to be troubling when other precursors of inflation such as
speculative increases in precious metal or other commodity prices
have been observed. At other times, members may fear that allowing
shortfalls in money growth to persist will lead to a recession, which in
turn may encourage an overly stimulative fiscal policy.




115

The Committee's wide-ranging discussion underscores the complexities and uncertainties affecting decentralized decision making
in a flexible and efficient market economy. Still, the Committee
must set monetary and credit objectives in this uncertain environment, recognizing that it may neither achieve them nor the economic performance its members wish. When the voting members
give the actual growth ranges they favor for the calendar year, their
views are often closer than one might expect from the diversity of
their analyses. After the discussion, the Chairman can usually present ranges for the monetary and credit aggregates that will command majority support. Further discussion indicates whether a
modification of the Chairman's proposal will pick up additional
support. Prompted by a strong desire to be as united as possible in
setting the year's objectives, the Committee members reach their
decisions with considerable give and take along the way. Even after
the agreement on the growth rate ranges is achieved, however,
members of the voting majority may well entertain different expectations of what the fulfillment of those objectives will mean to the
economy's performance. (Members are free to modify their estimates of QNP growth and price performance following Committee
discussion.) If a consensus is emerging, a vote on the longer run
specifications may be taken at this time. Sometimes the vote may
be postponed in the hope that subsequent discussion of the shortrun parameters will clarify the options.
Once a set of long-run specifications has been approved, the Committee considers how to express its choices in the policy directive.
The blue book contains suggested wording for the paragraph on
the longer run aggregates. Usually, the Committee uses that wording, but occasionally it may make some modifications to highlight
a particular concern.
4. Short-Run Policy Alternatives
a. Presentation
The Monetary Affairs Division Director then discusses alternatives for the short-run growth paths for the monetary aggregates,
along with associated amounts of discount window borrowing and
ranges for the Federal funds rate. The presentation draws upon and
amplifies the material presented in the blue book. At the February
and July meetings, the Director gets right to the specific policy
alternatives, since the presentation on long-run ranges will have
included general comments about factors affecting the behavior of
the monetary aggregates. At other times, this portion of the meet-




116

ing starts with a review of the recent behavior of the aggregates,
with attention to any special factors that have been upsetting the
normal relationships.
At all meetings, the Director refers to alternative specifications
(typically three are provided, labeled A, B, and C) for three-month
growth rates of the monetary aggregates spanning a calendar quarter. The central alternative, labeled B, normally contains the monetary growth rates estimated by the staff to be consistent with the
prevailing degree of reserve pressure and amount of discount window borrowing. If financial market participants expect reserve pressures and the discount rate to remain unchanged, then the Director
will probably suggest that money and bond market rates would hold
steady if the Alternative B specifications were adopted. But if holding pressures steady would surprise market participants or if special
seasonal factors were expected to distort rate relationships, the
Director could indicate that some other rate behavior was likely.
Alternative A usually encompasses reduced reserve pressures
and higher monetary growth rates relative to the Alternative B
assumptions. Normally, a decrease in reserve pressures would be
achieved by lowering the assumed level of discount window borrowing incorporated in the reserve paths. As a rule, the blue book
does not discuss possible discount rate changes as an alternative
approach to reducing pressure because the FOMC does not have
the power to change the discount rate and consequently no definitive decision could be reached on that score. If borrowing were
already about as low as it was likely to go, however, a discount rate
cut might be suggested as an alternative. (The presidents and governors at the meeting would know if there had been initiatives for
cuts made by any of the Reserve Banks. The governors would also
know if they were inclined to approve them.) The Director is likely
to suggest that adopting the Alternative A specifications will lead
to lower short-term interest rates, while the behavior of long-term
rates will depend upon how market participants regard such a
move. If participants see the policy as a desirable stimulus to
financing investment, they may help to bring down rates. But if
they believe such a policy to be inflationary, they might react by
pushing up rates on long-term financial instruments.
Alternative C usually encompasses an increase in reserve pressures relative to the B alternative. Its adoption would be expected
to slow money growth relative to its current course. The Director
will note that market participants would probably respond to the
adoption of this alternative by pushing up short-term rates and that




117

long-term rates might also rise if the action were seen as restricting
credit flows. On the other hand, if the move were seen as a welcome attack against rising inflationary forces, long-term rates
would be expected to remain steady or even decline.
Once the Director's presentation is completed, Committee members may ask technical questions about the behavior of the aggregates or other specifications. Members may also ask the Director
about the likely market reaction to a particular policy course.
5. Preparing the Directive
a. Discussion of the specifications
Next, the Chairman turns to the specifications to be used in writing the section of the policy directive that will guide the New York
domestic trading desk in its execution of policy until the next
FOMC meeting. Opening remarks are designed to give some focus
to what follows. The Chairman may express particular concerns
about policy or comment that Committee members appear to be
in general agreement about the appropriate direction of policy.
The Chairman then asks the other 11 voting members of the
FOMC and the 7 nonvoting presidents to offer their policy preferences. The discussion is generally couched in terms of the blue
book alternatives for the short-run aggregates and the initial borrowing assumption. The speakers use a shorthand, saying, for
instance, that they favor "B" to indicate that they favor the set of
specifications contained in Alternative B. Some people may agree
with only part of an alternative. For instance, a speaker may favor
the slower growth rates for the monetary aggregates contained in
Alternative C but prefer initially to specify a level of discount window borrowing associated with Alternative B. Often, speakers suggest specifications that fall somewhere between two of the blue
book alternatives. For instance, they may want to see a more subtle degree of easing in reserve pressures than contemplated in
Alternative A, because they are only slightly worried about signs of
slack in the economy. Such a preference might be expressed as
support for an "A-" or "B+" option. Or, individual members may
prefer a policy stance outside the range presented in the blue book
—either to the easier or to the more restrictive side.
Once the round of policy comments has been completed, the
Chairman summarizes the results of an informal tally of voting
members. If there is a clear preponderance of support for a particular approach, it is noted, and the Committee proceeds directly to
the language of the directive. However, if a range of views is




118

expressed, with members pushing for different policy prescriptions, the Chairman may have to explore several alternatives to see
if any intermediate positions can command broad support.
Most policymakers view the monetary policy process as one of
evolutionary adjustments and therefore are willing to accept a prescription that moves in the direction they favor even if at a modestly faster or slower pace than they would prefer. Consequently, it
is usually possible to come up with some shading that captures the
support of most members. On rare occasions, however, the divisions may be deep and the views strongly held. In such cases, a
number of combinations must be tried before a majority of members give their support to a particular policy prescription.
b. Writing the directive
Once the general outlines of the specifications for money growth
and reserve conditions for the near term have been established, the
Committee addresses the wording of the directive that will guide
the Open Market Function of the Federal Reserve Bank of New
York. The blue book presents suggested language that normally
follows the pattern of the previous directive unless there is a special
reason to make some modification.5
The conversation focuses on the final, operational paragraph of
the directive. The Box contains two samples. The first sentence
indicates the degree of pressure on reserve positions desired in the
immediate future. The usual phrasing is "maintain the existing
degree of reserve pressure" or "increase" or "decrease reserve pressure." A preference for Alternative B with no change in discount
window borrowing is conveyed by the first expression. The alternative phrasings are associated with the selection of Alternative C or
A, respectively. The degree of change in pressure desired can be
indicated with the modifier "slightly" or "somewhat." The directive
goes on to indicate the three-month growth rates for the aggregates
that are expected to be consistent with the specified degree of
reserve pressure, either immediately following the introductory sentences or in a later part of the paragraph.
The Committee also writes a passage explaining the adjustments
it might wish to see made to the degree of reserve pressure before
5 The early paragraphs of the directive review recent developments in the economy, the
exchange markets, and the monetary aggregates. They state the Committee's fundamental
goals of price stability and sustainable economic expansion and report the long-run money
and credit growth ranges. Except for the paragraphs on the long-run objectives constructed
at the February and July meetings, these paragraphs are rarely discussed at meetings,
although they can be discussed if any member wants to suggest a change.




119

Box: Operating Paragraphs from FOMC Domestic Policy Directives
Meeting Held July 5-6, 1989
In the implementation of policy for the immediate future, the
Committee seeks to decrease slightly the existing degree of pressure on reserve positions. Taking account of indications of inflationary pressures, the strength of the business expansion, the
behavior of the monetary aggregates, and developments in foreign
exchange and domestic financial markets, somewhat greater
reserve restraint or somewhat lesser reserve restraint would be
acceptable in the intermeeting period. The contemplated reserve
conditions are expected to be consistent with growth of M2 and M3
over the period from June through September at annual rates of
about 7 percent. The Chairman may call for Committee consultation if it appears to the Manager for Domestic Operations that
reserve conditions during the period before the next meeting are
likely to be associated with a federal funds rate persistently outside
a range of 7 to 11 percent.
Meeting Held August 22, 1989
In the implementation of policy for the immediate future, the
Committee seeks to maintain the existing degree of pressure on
reserve positions. Taking account of progress toward price stability, the strength of the business expansion, the behavior of the monetary aggregates, and developments in foreign exchange and
domestic financial markets, slightly greater reserve restraint might
or slightly lesser reserve restraint would be acceptable in the intermeeting period. The contemplated reserve conditions are expected to be consistent with growth of M2 and M3 over the period from
June through September at annual rates of about 9 and 7 percent,
respectively. The Chairman may call for Committee consultation if
it appears to the Manager for Domestic Operations that reserve
conditions during the period before the next meeting are likely to
be associated with a federal funds rate persistently outside a range
of 7 to 11 percent.




120

the next meeting if certain factors behave significantly differently
than expected. For a number of years, this passage gave prominence to deviations in the behavior of the monetary aggregates. But
as the demand for money became more variable in the 1980s, the
Committee instructed the Open Market officials to respond to a
range of factors in addition to the monetary aggregates. The factors
to be considered for a potential change in reserve pressures generally include the pace of business expansion, the extent of inflationary pressures, developments in the foreign exchange markets, and
conditions in the domestic and international credit markets.
The Committee identifies those factors that should be given the
most weight when the Open Market officials evaluate whether to
make modifications to reserve pressures before the next meeting.
For instance, if there is particular concern about inflationary forces
and signs of acceleration in economic activity attributed to recent
rapid money expansion, the Committee may be especially sensitive to the pace of money growth. It can indicate that it wants to be
particularly responsive to overshoots of the monetary goals, while
only responding to monetary shortfalls if the economy appears to
be weakening significantly. In composing the directive, the
Committee will make money growth prominent in the list of factors
to consider in deciding whether to change reserve pressures. In this
case, it will probably say that "somewhat" greater reserve restraint
"would" be acceptable while "slightly" lesser reserve restraint
"might" be acceptable. If, on the contrary, the concerns are
focused on weakness in the economy and imminent inflationary
pressures appear unlikely, the Committee can reverse the use of
"somewhat" and "slightly" and of "would" and "might." It can also
rearrange the list of considerations affecting reserve pressure
changes to give more weight to economic activity. In the extreme,
it may allow scope for adjustments to be made only in one direction. When the risks are deemed essentially balanced, the
Committee typically chooses symmetrical phrasing.
The final topic addressed in writing the directive is the Federal
funds rate range. If a borrowing adjustment causes funds to trade
persistently outside the specified range, the Committee will want to
be consulted. Since the 1982 modifications to procedures, the
amount of adjustment in reserve pressures between meetings has
tended to be smaller than before, and the band on the Federal
funds rate has rarely served as a constraint. The band has been
adjusted occasionally when the central tendency of Federal funds
trading has shifted toward one end of the band.




121




Once the directive's wording has been completed, the Chairman
calls for a vote. All voting members will indicate their approval or
disapproval. By that point, majority support is virtually assured, as
the directive has been constructed to encompass most views that
have been expressed. The Secretary of the FOMC records the
votes and reports the results. The Secretary then confirms the next
regular meeting date, and the meeting is adjourned.
Following the meeting, copies of the directive are sent to the participants. Later, the Secretary's office prepares a more extensive policy record that reports the highlights of the meeting. This record is
sent to Committee members for review and possible revision and then
is published, along with the directive, a few days after the following
meeting. Any member who voted against the directive includes with
the policy record an account of his or her reasons for dissenting.

122




The Trading Desk:
Formulating Policy Guidelines
Formulation and Operation of the Reserve Objective
Upon returning from an FOMC meeting, the Manager of the
System Open Market Account confers with the other officers and
senior staff members of the Open Market area charged with actually
implementing the policy directive. The Manager describes the principal topics of discussion at the meeting and reports on the actions
taken by the Committee regarding the degree of reserve pressure
to be used initially for building reserve paths, the monetary targets,
and the likely trading range for Federal funds. The conditions under
which the Committee would want the degree of reserve pressure
modified during the period before the next meeting are also reported. A written text of the directive, prepared by the secretary of the
FOMC, will arrive during the day. Implementation of the directive
will require translating the general reserve specifications established at the meeting into operating parameters.
1. Creating the Nonborrowed Reserve Objective
The policy procedures focus on establishing targets for reserve
measures consistent with the directive's instructions concerning
reserve pressure. Reserve pressure is interpreted in the context of
a series of interrelated reserve concepts. Reserve paths are developed based upon these concepts and the FOMCs specifications.
The paths exploit relationships between the supply of and demand
for total reserves (TR) and the forces that bring them into equilibrium (see Diagram 1, page 125). Box A describes the reserve
components and their determinants in some detail. The demand
for total reserves can be viewed as a demand for required reserves
(RR) plus a demand for excess reserves (ER). Depository institutions are required to hold reserves against certain deposit liabilities, as specified in Federal Reserve Regulation D. They hold
excess reserves for precautionary reasons because it is not always
possible to achieve a level of total reserves precisely equal to
required reserves. Excess reserves provide a cushion so that a
reserve shortfall will not leave depository institutions with an
unwanted reserve deficiency for the maintenance period or an
overnight overdraft in their reserve account, either of which could
be subject to penalties.
The supply of total reserves essentially comes from two different
sources, nonborrowed reserves (NBR) and borrowed reserves
(BR). Nonborrowed reserves are supplied from additions to
Federal Reserve assets (other than loans at the discount window)
and from reductions in liabilities. The Federal Reserve can change




124

Diagram 1 Total Reserves
Secured loans
from Federal Reserve

Buffer against
clearing needs
and uncertainties

Comfi onents and Determinants
Tot Reserves
al
'Demand"

Total Reserves
"Supply"

Adjustment \

Total Reserves
"Components"

Seasonal

> credit

Extended J

i
Transactions
deposits in current
computation
period

Reserve
ratios
0-3-12%
=
reserves
3%

transaction
deposits

Nonborrowed

|||

-*•
Nonpersonal
time deposits
and other reservable
liabilities from
earlier computation
period

Balances
at
Federal
Reserve

RR+ER

= NBR+BR

the level of nonborrowed reserves through purchases, sales, and
redemptions of government securities. Also affecting nonborrowed
reserves are changes in so-called market factors, which are not
under the direct control of the desk. These factors include currency
in circulation, float, and the Treasury's cash balance at the Fed.
Discount window borrowing, the alternative source of total
reserves, is initiated by the depository institutions but is subject to
the rules and guidelines set by the Federal Reserve for discount
window use. Since the banks face restrictions on borrowing,
reductions in nonborrowed reserves relative to demand will
increase reserve pressures. The relationship between the demand
for and the supply of reserves can be represented in equation form
as: RR + ER = TR = NBR + BR.
To construct the nonborrowed reserve objective at the start of a
two-week reserve maintenance period, staff members at the Board
of Governors in Washington, D.C. and the New York Federal
Reserve develop estimates of the demand for total reserves. They
make estimates of required reserves by using forecasts of deposit
behavior and reserve requirement ratios. Average excess reserve
levels change slowly, so a "normal" allowance is generally used.




125

Federal Reserve
portfolio of
Treasury and Federal
agency securities

Other balance
sheet items:
Total Federal
Reserve assetsother Federal
Reserve liabilities

Variations from the norm are forecast through the use of econometric models and staff judgment. At times when excess reserves
are expected to be substantially different from the norm—for
example, in a reserve period that includes a quarter end or when
unusually large excesses or deficits have been carried into the
period—a different allowance might be used. Otherwise, informal
adjustments are made to the nonborrowed reserve path when a
deviation in excess reserves appears likely.
The projection staff adds together the estimated demands for
required reserves and excess reserves to obtain an estimated
demand for total reserves. From that, it subtracts the FOMCs preferred level for borrowed reserves to form tentative nonborrowed
reserve objectives for the current and two succeeding maintenance
periods. Staff members at the Board and New York Fed will make
revisions to the nonborrowed reserve objective during a maintenance
period as incoming data call for modifications to estimates of required
reserves or excess reserves, in order to keep the likely amount of borrowed reserves in line with the FOMCs intentions. On occasion, the
desk may make a technical adjustment to the borrowing allowance
—for instance, if a computer problem or other operational difficulty
during the period has resulted in a large amount of borrowing. At
other times it may choose not to achieve the borrowing allowance if
doing so would cause misleading movements of the Federal funds
rate away from the expected trading range.
2. Adjustment by the Banking System to Policy Actions
The demand for, and supply of, total reserves must come into
balance over the maintenance period. The full response of the
banks and the public to the provision or withdrawal of nonborrowed
reserves will depend upon many factors, including the underlying
institutional and regulatory structure. Because of the large number
and variety of financial institutions in the United States that participate in the creation of reservable deposit liabilities, the adjustment
process is inevitably complex. How a particular action will affect
money, credit, and interest rates can be described in a general way,
but the magnitude and timing of the responses to a monetary policy action can at best be roughly estimated.
Suppose the FOMC had voted for increased reserve pressures.
The desk, under the new specifications, would meet a smaller
share of the estimated demand for total reserves through injections




Continued on page 137

126

Box A:
Description of Reserve Measures
Total Reserves:
Total reserves consist of depository institutions' reserve balances
held at day's end at the Federal Reserve and applied vault cash
(defined below). Depository institutions must hold total reserves to
meet reserve requirements, specified as averages over two-week
reserve maintenance periods that end every other Wednesday.7
Applied vault cash is defined as that portion of total currency held
by depository institutions that is used to meet reserve requirements.
The vault cash that is applied during a two-week reserve maintenance period was held during a two-week computation period that
ended on a Monday almost two and one-half weeks before the reserve
maintenance period began. Large banks apply all of their vault cash
toward meeting their requirements since their required reserves
exceed their vault cash. However, many small banks and thrift institutions hold more vault cash than they need to meet their reserve
requirements. Hence, applied vault cash for those institutions is equal
to their required reserves. Their total vault cash is known at the beginning of the reserve maintenance period, but the portion applied to
meeting reserve requirements cannot be computed until after the
reserve maintenance period ends and their required reserves are
known. This excess vault cash arises because depository institutions
base the amount of vault cash they will hold on expectations of customer demands rather than on reserve requirements. Many smaller
institutions need more cash for conducting business than they need
to meet requirements. On average over 1988, applied vault cash
amounted to $24.9 billion while excess vault cash, not needed to
meet reserve requirements, was $1.7 billion.
Reserve balances at the Federal Reserve are also held for purposes
beyond meeting requirements. They provide the means for transferring funds among banks by check and by wire. As checks clear
through the Federal Reserve, reserves are transferred from the paying
bank to the receiving bank. Private clearing services also clear checks
and arrange for the net transfers among reserve accounts arising from
the settlement of the checks. Many transfers are made when a depository institution directs the Federal Reserve to make a wire transfer for
1 For days on which depository institutions are closed, all deposit, reserve, and vault cash
levels are counted as equal to the previous day's levels.




127

Chart 1 Measures of Systemwide Fedwire Activity
(Daily Averages)

Thousands
700

Billions of dollars
700

600
500

300

200

200

100

100

nfl

left axis)
Dollar volume

(

400

300

(

500

400

Number of transfers

600

right axis)




0
1977

'78

0
'79

'80

'81

'82

'83

'84

'85

'86

'87 '88

itself or a customer over what is called Fedwire. Treasury and some
Federal agency securities that are held in book entry form at the
Federal Reserve are also transferred through the Fedwire system. The
volume of reserve transfers over Fedwire is very high. In 1988, for
example, it averaged $640 billion a day (Chart 1).
In 1988, reserve balance accounts, which averaged $37 billion,
turned over around 25 to 30 times a day, more often at large banks.
Balances in the reserve accounts facilitate such transfers. The high
turnover rates often leave some banks' reserve accounts overdrawn
at some point during the day. Such daylight overdrafts are permitted, but their size is restricted according to the capital of the bank
and the bank's own assessment of its creditworthiness. The Federal
Reserve is currently exploring ways to limit daylight overdrafts such
as introducing specific charges for them. By day's end, a bank must
cover all overdrafts or be subject to a significant penalty, even if it
has already met its requirements for the period.2
2 The penalty for an overnight overdraft is at a rate of 10 percent or 2 percentage points above
the effective Federal funds rate that day, whichever is higher. In addition, the banks have
to make up the overdraft by holding the same amount of extra reserves on other days in
that maintenance period.

128

Required Reserves:
Reserve requirements can be satisfied by holding either or both of
the two forms of total reserves—vault cash from an earlier computation period and end-of-day reserve balances at the Federal
Reserve. Depository institutions must come close to meeting their
requirements on average over a maintenance period; they are
allowed to carry forward for one maintenance period an excess or
deficiency of up to 2 percent of their requirements or $25,000,
whichever is greater.3 Once these carryovers are taken into account,
a depository institution that fails to meet its requirement will be
assessed a penalty on the deficiency at a rate that is 2 percentage
points above the discount rate (although the penalty may be waived
if there are extenuating circumstances). If a depository institution
frequently fails to meet requirements, its senior management will be
contacted to discuss the problem and reminded that repeated failure
to comply with this important obligation would put the institution
under scrutiny.
Reserve requirements are computed as various fractions of deposit
levels. Requirements are specified in Federal Reserve Regulation D
according to rules and guidelines established in the Depository
Institutions Deregulation and Monetary Control Act of 1980 (MCA)
and the Garn-St Germain Depository Institutions Act of 1982.4 The
guidelines distinguish between nontransaction deposits and transaction deposits. Of the nontransaction deposits, primarily time and savings deposits, only those that are classified as "nonpersonal" and
carry initial maturities of less than 18 months are subject to reserve
3 A bank can use excess reserves carried forward in the next period by running a deficiency
equal to the excess carried forward. If it does not use the excess in that period, the carryover is lost. A bank must cover in the next period a deficiency carried forward by holding
excess reserves in a volume at least equal to the deficiency or the bank will be judged to
have failed to satisfy its requirement.
4 Schedules of reserve requirements were contained in the MCA. At the completion of the
phase-in period, institutions had to hold reserves on transactions deposits equal to 3 percent
up to an indexed amount and reserves equal to 12 percent for additional amounts. The GarnSt Germain Act of 1982 exempted the first $2 million of transactions deposits from reserve
requirements and indexed the exempted amount. Member commercial bank reserve requirements were gradually reduced between 1980 and 1984. Nonmember institutions, on the
other hand, had not been subject to the Federal Reserve's reserve requirements before passage of the MCA. They became bound by requirements phased in between 1980 and 1987.
By the latter date, the same reserve requirements applied to member and nonmember institutions. At the same time, the MCA eliminated reserve requirements on "personal" time and
savings deposits and all deposits with original maturities of 18 months or more. It cut reserve
requirements on shorter term "nonpersonal" time deposits to a flat 3 percent.




129

Diagram 2 Timing of Contemporaneous Reserve Accounting System
Contemporaneous Reserve Accounting System
Weeki
Week 2
T\W\T\F\S\S\M\T\W\T\F\S\S\M
Two-week computation period
for all reservable liabilities
other than transaction deposits

Average vault cash in this
two-week period counts as
reserves in the maintenance
period ending 30 days later

Week 3
Week 4
T\W\T\F\S\S\M\T\W\T\F\S\S\M
Reserve Accounting
for Liabilities

Week 5
Week 6
T\W\T\F\S\S\M\T\W\T\F\S\S\M

T\W

Two-week computation period
for transaction deposits

Accounting for Reserves

Two-week reserve
maintenance period

Note: A "reserve maintenance period" is a period over which the daily average reserves of a
depository institution must equal or exceed its required reserves. Required reserves are based
on daily average deposit liabilities in "reserve computation periods."




requirements. In most cases, the reserve requirement on these
deposits is 3 percent. Cinder the reserve accounting structure introduced in February 1984, it is calculated on a lagged basis, using the
same computation period as vault cash (Diagram 2). Thus, when
constructing the objective for nonborrowed reserves, staff members
do not need to estimate required reserves for nontransaction deposits
for the maintenance period in progress, since these requirements are
known before the period starts.
For required reserves against transaction deposits, the computation period is the two weeks ending on the Monday two days before
the maintenance period ends.5 Thus, required reserves are not known
during the period and must be estimated. The projection staffs at the
New York Fed and the Board make estimates of both the underlying
deposit trends and of special seasonal or technical factors. The staffs
estimate the underlying behavior by looking at recent trends in the
monetary aggregates (primarily Ml) and considering how interest
rate movements and economic developments are likely to affect
them. For instance, deposits may be growing rapidly because interest
5 Very small depository Institutions are exempt from reserve requirements and only report
their deposits annually. Institutions of the next larger size only report quarterly. For that
group, required reserves on transaction deposits are lagged. Most of these institutions meet
their entire requirement with vault cash.

130

rates on market instruments are falling faster than the rates on
deposits. A prominent seasonal factor affecting deposits is the payment of social security benefits on the third of every month. Most
recipients allow their cash balances to rise initially, then gradually
work down the deposits as they pay their bills. Before seasonal adjustment, total deposits in the various money measures move in a pattern
that reflects the swings in social security recipients' balances. The
Treasury's cash balance might show offsetting movements, but
Treasury cash is not counted in the monetary aggregates.
Once the projection staffs have developed forecasts of total
transaction deposits, they must estimate the appropriate average
required reserve ratios to use in deriving required reserves.
Transaction deposits are divided into three tranches, with indexed
cutoffs that change slightly each year. In 1989, the first $3.4 million
of transaction deposits are exempt from reserve requirements.
Then deposits up to $41.5 million are subject to a 3 percent reserve
ratio. All deposits in excess of that amount are subject to a 12 percent reserve ratio. Between the enactment of the MCA in 1980 and
February 1984, new and generally lower reserve requirements were
phased in for member banks of the Federal Reserve System.
Between 1980 and September 1987, higher requirements were
phased in for nonmember institutions. During those periods, the
staffs had to take account of differing required reserve ratios
between member and nonmember institutions. In September 1987,
the distinction ceased to exist. Nonetheless, the staffs still have to
estimate changes in average reserve ratios as deposits move
between large and small institutions.6
During the maintenance period, the Board and New York staffs
make frequent updates of their forecasts of required reserves as
information on actual deposit levels becomes available. Usually,
they modify their forecasts during the middle part of the period —
on Tuesday, Wednesday, and Thursday—in response to incoming
data on deposit levels for the first week of the period. They make
further adjustments on the following Tuesday and Wednesday as
they receive deposit data for the second week of the period. They
will continue to revise required reserves after the period is over, taking into account the more complete information received, but these
6 The staffs abo have to estimate reserve requirements on interbank and Treasury demand
deposits, which are not part of the monetary aggregates. Interbank deposits have no pronounced trend, although they show both a seasonal pattern and residual volatility.
Treasury demand deposits are small and reasonably predictable.




131

changes will not, of course, affect the desk's reserve provision for
the period that has concluded.
Excess Reserves:




Excess reserves consist of total reserves that are not needed to
meet reserve requirements. All excess reserves take the form of
reserve balances at the Fed rather than applied vault cash, since
the vault cash included in total reserves is defined to exclude that
portion of vault cash that does not satisfy reserve requirements.
Excess reserves arise because depository institutions do not have
perfect control over the level of their reserve balances. They hold
excess reserves when the cost of eliminating them is estimated to
exceed the interest lost by investing the reserves.
Federal Reserve strictures have given depository institutions a
strong sense of obligation both to meet their requirements and to
avoid end-of-day overdrafts. Large banks devote considerable
resources to monitoring reserve and deposit flows so as to achieve
these ends without having to hold excess reserves. These banks
usually are able to keep excess reserve levels within the relatively
narrow band established by their carryover allowance. Occasionally, around quarter ends or at other extraordinary times, they are
left with excess reserves that exceed their carryover limits, but
large overshoots happen infrequently.
For many small and medium-sized commercial banks and thrift
institutions, the flows through their reserve accounts each day are
large compared with the relatively small amount of reserve
balances needed to meet requirements. For some of these institutions, it costs less to hold reserve balances in excess of requirements than to engage in the close management of reserve positions
necessary to eliminate excess reserves.
Excess reserves grew during the 1980s (see Chart 2, page 133)
after the MCA and the International Banking Act of 1978 mandated
that reserve requirements be extended to more institutions. The
MCA made nonmember commercial banks and thrift institutions
subject to Federal Reserve rather than state requirements; the
International Banking Act applied to branches and agencies of foreign banks operating in the United States. Many of these institutions
previously had met their state's reserve requirements by holding
vault cash and deposits at correspondent banks, neither of which
were counted by the Federal Reserve in any of its reserve measures.

132

Chart 2 Average Excess Reserve Levels
Millions of dollars
1200

1977 78

79

'80

'81

'82

'83

'84

'85

'86

'87

'88

When reserve requirements were phased in, some of these institutions became subject to reserve requirements in excess of vault
cash. As they opened reserve accounts to meet requirements and
to provide funds for wire transfers and check clearing through
the Federal Reserve, they often found that the reserve balances
needed for clearing were greater than the reserve balances needed
to meet requirements.7
Excess reserves also rose between 1980 and 1984 because banks'
needs for reserves for clearing purposes were rising but required
reserve balances were falling (see Chart 3, page 134). During those
years, MCA-mandated cuts in reserve requirement ratios were being
7 Depository institutions that routinely need excess reserves for clearing purposes either
because their required reserve balances at the Federal Reserve are low or because vault cash
fully meets requirements may establish "required clearing balances." An institution may
choose the amount of reserve balances it expects it will need for clearing and make a commitment to hold that amount on average. The Federal Reserve then compensates the institutions for those balances by extending credits to cover fees for priced services. An institution's
maximum required clearing balance level is determined by the amount of priced Federal
Reserve services it uses. Often that level is below the reserve balance needed to avoid overdrafts. In many cases depository institutions have chosen not to establish required clearing
balances since that would entail paying increased attention to reserve management.
Required clearing balances are not treated as part of total reserves (or excess reserves).




133

Chart 3 Composition of Total Reserves (Quarterly Averages)
Billions of dollars (not seasonally adjusted—not adjusted for reserve requirement changes)
70
60
50
40
30
Reserve balances at the Fed.
20
Total reserves

10
Applied vault cash

Required reserves




12 3 4 12 3 4
1980 1981

2 34
1982

2 34
1983

2 34
1984

2 34
1985

2 3 4 12 3 4 12 3 4
1987
1986
1988

phased in for Federal Reserve member banks; the reduction in member bank requirements exceeded the increase in reserve requirements for nonmember institutions. At the same time, the demand for
vault cash was rising because of the spread of automated teller
machines and because deregulation of interest rates was encouraging
many banks to expand their consumer business. Even some fairly
large banks found themselves in a position where vault cash was
meeting a substantial share of their reserve requirements. Some
increase in excess reserves at those institutions accompanied the
need for larger balances to cover clearings. Required reserve balances began to rise again after 1984, but excess reserves continued
to increase as the ongoing phase-in of requirements for nonmember
institutions resulted in increasing numbers of institutions maintaining
reserve accounts at the Federal Reserve. Furthermore, the volume of
reserve balances needed to avoid overdrafts continued to grow, since
the number of transactions clearing through the Federal Reserve was
continuing to rise. These factors combined to raise the routine excess
reserves of all but the largest member banks.

134

Borrowed Reserves:
Three basic types of collateralized credit may be made available
to banks and other depository institutions at the discount window:
adjustment credit, seasonal credit, and extended credit borrowing. A
bank can use adjustment credit when it comes up short in its efforts
to meet its reserve requirement or when it would otherwise run an
overnight overdraft. Banks are instructed to make a good faith effort
to obtain the reserves from other sources before borrowing from the
discount window; they could, for example, purchase Federal funds
from another bank. The Federal Reserve extends adjustment credit
for one or at most a few business days at the basic discount rate and
discourages prolonged or frequent use. Banks also must demonstrate that they are not relending the borrowed funds to other banks.
Thrift institutions eligible to borrow from a Federal Home Loan Bank
may also use adjustment credit, but normally they would meet anticipated needs through Home Loan Bank advances.
Cinder the seasonal borrowing program, small banks that have a
significant seasonal pattern to their lending can borrow modest
amounts for a more lengthy period during that portion of the year
when their lending is regularly high. Because these banks are small,
the Federal Reserve assumes that they have limited ability to obtain
funds from the national money markets. The discount window officers will not require the same justifications that apply to requests
for adjustment borrowing. Seasonal borrowing has a true seasonal
pattern because many of the users are agricultural area banks that
face their strongest credit demands over the spring and summer.
Nevertheless, seasonal borrowing shows a sensitivity to reserve and
interest rate pressures that is similar to that of adjustment borrowing
although smaller in magnitude.
For reserve path purposes, adjustment and seasonal credit together constitute borrowed reserves. Adjustment credit borrowing
is included because the limitations on banks' use of this type of
credit will force them to bid up the Federal funds rate before they
turn to the discount window and will consequently encourage them
to restrict the extension of credit. Seasonal borrowing is treated as
part of borrowed reserves because of its tendency to respond to
changes in reserve pressures in a fashion similar to adjustment
credit. For a number of years no clear seasonal pattern to the sum
of adjustment and seasonal borrowing was evident, despite the




135

strong pattern to seasonal borrowing, apparently because adjustment borrowing was dominant. But in recent years, when seasonal
borrowing has often constituted a significant share of the sum, the
FOMC has sometimes had to adjust the borrowing objective to
allow for a seasonal pattern. Otherwise, borrowing relative to the
spread between the Federal funds rate and the discount rate would
go off track as the seasons progressed.
Extended credit borrowing represents loans to depository institutions that are having unusual difficulties. An institution in this program does not generally have the means to borrow additional funds
from normal market sources. Hence, its dependence on Federal
Reserve credit is likely to last until its basic problems are resolved
either through an acquisition, an infusion of additional capital, or
some other action by its insurer. While in the program, banks are
permitted to borrow without the normal pressures to repay promptly. Soon after their borrowing begins, the interest rate they pay is set
above the basic discount rate so as to make the cost to the troubled
institution commensurate with market rates. In formulating reserve
paths, the desk treats extended credit borrowing as if it were nonborrowed reserves because of its special characteristics.
Monborrowed Reserves:




Nonborrowed reserves can most easily be described as the portion of total reserves provided to depository institutions through any
means other than the discount window. The primary source of nonborrowed reserves is the purchase of Treasury securities by the desk.
Nonborrowed reserves can also be provided or withdrawn when
other Federal Reserve balance sheet items change. Box B reviews
the major factors affecting reserves. The formal definition of nonborrowed reserves does not include extended credit borrowing,
although the definition used for reserve path construction includes it.

136

of nonborrowed reserves or, if reserves were above desired levels,
it would drain more reserves than otherwise.8 If the desk sells
Treasury securities, the purchasers' banks will initially lose
reserves. But the resultant reserve shortages will quickly shift from
the banks of firms participating in the open market operations to
other depository institutions as the institutions encountering the
shortages take steps to cover them. A similar quick redistribution
of the additional reserves would follow an open market purchase.
Within a single two-week reserve maintenance period, the
banks' adjustment options to a reserve excess or shortage are in
actuality quite limited. Banks confronting a shortage would have
several possible options, but most of the options would be impractical to carry out in such a short time period. In principle, banks
could take actions to reduce their demand for reserves by trying to
lower their required reserves within the period. Such a strategy
would require reducing transaction deposits (a decrease in time
deposits would not lower required reserves until a future maintenance period). To this end, banks could increase lending rates and
lower transactions deposit rates, thereby encouraging customers
to reduce loans and deposits.9 Such prompt adjustments by many
banks and by their customers, however, are unlikely. They would
require very quick recognition that a change in policy toward
reserve provision had occurred and very fast responses to the
change. In practice, the depository institutions and their customers
have less disruptive ways to respond to a reserve shortage, so
most of the initial adjustment to reduced provision of nonborrowed
reserves will take place in categories other than required reserves.
Another adjustment option available in theory to the banks is to
cut back on excess reserve holdings. But as indicated in Box A, the
banks' holdings of excess reserves are effectively already low. Large
banks generally manage to keep their excess reserves close to zero
on average, so they cannot economize on excess reserves more than
they already do. Some small banks may be able to economize slightly on excess reserves, but the scope for further reductions is small.
8 Raising the discount rate instead of reducing the amount of nonborrowed reserves supplied
would also increase reserve pressures. The discount rate is not a tool of the FOMC, although
the Board of Governors, in approving a discount rate change, takes account of its effect on
reserve pressures. Discount rate changes have a strong announcement effect, which may
or may not be desired.
9 Banks are prohibited by law from paying explicit interest on demand deposits. However,
they can adjust the implicit return by changing fees charged and services provided. They
are free to adjust rates on consumer NOW accounts, although NOW account rates have
changed very little since the rate ceilings were removed in 1986.




137




Thus, when nonborrowed reserves within a reserve maintenance
period are insufficient to meet the demand, the banking system as
a whole has no practical alternative to borrowing more reserves at
the discount window. Because of the limitations on frequent borrowing, however, each individual bank will explore other ways to
adjust its reserve position before it turns to the discount window. A
bank can sell assets such as short-term securities or loans, it can
bid for wholesale deposits such as large CDs or Eurodollars, or it
can purchase reserves from other banks directly or through a
Federal funds broker. None of these actions increases the total
reserves of the banking system but all do redistribute the shortages. When banks attempt to acquire an increased share of the
pool of nonborrowed reserves, they will bid up the Federal funds
rate. As the funds rate rises, more banks will be induced to borrow
at the discount window until total reserves available are brought
into line with the demand.
Although these actions do not slow the growth in the monetary
aggregates within the maintenance period, they will begin a process that will lead to a slowdown in both money and credit growth
over time. How much the funds rate will increase as a result of the
induced increase in discount window borrowing and how quickly
the other responses will occur depend upon a number of factors.
Bank perceptions of likely and appropriate Federal Reserve policy
actions are important. If banks initially see the increased reserve
pressures as an anomaly—as, for example, the result of some seasonal or other transitory development—they will be unlikely to
make any basic adjustments to their approach to lending and
deposit taking. Likewise, participants in the securities markets initially may not adjust market rates if a change in the funds rate
appears to be an aberration. But once participants perceive that
the higher funds rate is part of a deliberate policy action, they will
take steps that will lift other short-term rates as well.
As the banks come to believe that the reduced reserve availability is a deliberate policy move, they will begin to make basic
adjustments to the pricing of loans and deposits. They may also
act to save their access to the discount window for future times
when reserve conditions may be even firmer than currently. These
adjustments will gradually work to alter money and credit growth.
Banks may reevaluate the structure of their lending rates and make
upward adjustments. They may also find that they will have to pay
higher rates on some types of deposits to discourage depositors
from shifting their funds to market instruments. It is likely that both

138

transaction deposits and time and savings deposits will shrink initially as market rates adjust upward more rapidly. In the current
environment, with unrestricted interest rates on all but demand
deposits, there will be recovery in those classes of deposits on
which the banks raise their rates enough to bring them into line with
the higher market rates. Thus, M2 and M3 will weaken more in the
months right after a restrictive policy action is recognized than they
will once the rate adjustments are complete. By contrast, Ml
growth, affected by continued rate restrictions on some types of
deposits and slow adjustment of NOW account rates by banks, will
likely be held down for a more extended period.
The details of the adjustment to a change in reserve provision
will vary according to initial conditions and expectations. Banks'
degree of reluctance to borrow will not always be the same. For
instance, when confronted with reserve shortages, banks that have
borrowed recently will be particularly reluctant to borrow and are
likely to bid more aggressively in the Federal funds market to cover
the shortages than would banks that had not borrowed recently.
Although the Federal funds market functions well, there is enough
friction that the initial distribution of the shortages may also influence how much the rate rises before borrowing occurs. Banks may
become more reluctant to borrow when financial difficulties at
other institutions with similar characteristics have received some
publicity. Even though the Federal Reserve does not release the
names of the banks that borrow, there is a risk that the borrowing
might be discovered through some other channel and interpreted
by depositors as a sign of weakness. Furthermore, although the
Federal Reserve tries to keep the standards for use of the discount
window constant over time and across Federal Reserve districts,
an institution may interpret a question or comment from an official
at the discount window as implying a change in the willingness of
the Federal Reserve to lend.
The public's response to the firmer policy will arise from whatever
steps the banks take to change their pricing of deposits and loans
and from the public's perception of future interest rate developments. If banks raise their rates on loans, customers may cut back
on their use of bank credit, either by substituting other, less costly
types of credit or by reducing their overall dependence on credit. If
customers expect this rise to be the first of many increases in interest
rates, however, borrowers may rush to get fixed-rate loans before
they become even more expensive, thus initially accelerating rather
than reducing loan demand and related deposit expansion.




139

Estimating Reserve Availability
Once the desk has an objective for nonborrowed reserves for the
two-week reserve maintenance period, it must develop a strategy
for bringing actual nonborrowed reserves into line with the objective. The first stage of the process is getting estimates of nonborrowed reserves for the period in progress and for future periods.
Past open market operations, which will have established the size
of the System's portfolio of Treasury and Federal agency securities, will be the primary determinant of nonborrowed reserves. But
nonborrowed reserves can also be provided or absorbed by a number of factors besides changes in the portfolio. Most of these consist of balance sheet items not under direct Federal Reserve
control. Several are subject to considerable period-to-period or
even day-to-day variation. For some factors, the variability is reasonably predictable. Others may have a predictable component,
but they may also change in a way that is hard to forecast. The
desk undertakes a substantial share of open market operations to
offset the unwanted reserve impact of swings in these factors.
Each morning, members of the New York Fed's monetary projection staff make estimates of the likely behavior of all of the factors affecting reserves and then present their estimates to the desk.
Their counterparts at the Board perform a similar exercise to provide the desk with a second set of estimates. As information flows
in, projection staff members learn how the factors turned out the
day before and what developments may affect these factors in the
future. They interpret any deviations from the expected behavior of
the factors and decide how to modify their forecasts. The major
factors that they examine are described in Box B, page 141.10
These forecasts form the basis of the estimate of reserve supplies.
They will be compared with the reserve paths in developing a plan
for carrying out policy operations.




10 An extensive discussion of these reserve factors and the techniques for forecasting them
can be found in John C. Partlan, Kausar Hamdani, and Kathleen M. Camilli, "Reserve
Forecasting for Open Market Operations/' Federal Reserve Bank of New York Quarterly
Review, Spring 1986, pp. 19-33.

140

Box B:
Forecasting Factors Affecting Reserves
Over time, providing for currency demands requires the largest net
reserve injections of any of the factors because currency has a strong
growth trend. Every time a bank needs to replenish its currency holdings following cash withdrawals by the public, it obtains currency
from the Federal Reserve, which debits the bank's reserve account.
Most short-term movements in currency follow recurring seasonal
patterns and are therefore predictable (see Chart 4, page 142). There
is residual variation, however, which is not always forecast accurately.

^'

Treasury Cash Balances:
Although the Treasury's balance at the Federal Reserve changes
little over the year as a whole, it is the reserve factor that shows the
most variation from one reserve maintenance period to another.
Increases in Treasury cash balances at the Federal Reserve absorb
reserves since they involve a transfer of funds from the banking
system to the Federal Reserve, while declines in the Treasury's
Federal Reserve balances supply reserves to the banks. The
Treasury attempts to keep a steady working balance at the Federal
Reserve for making its payments," and it places additional cash in
so-called Treasury tax and loan note option, or TT&L, accounts at
depository institutions that have agreed to accept them.72 Each
morning, the Treasury evaluates the estimated flows. It may decide
to transfer funds to the Fed by making a "call" on the TT&L
accounts if its balance would otherwise be below the target balance
or to transfer funds to the TT&L accounts by making a "direct
investment" to the TT&L accounts if the balance would otherwise
be higher than desired.'3
/1 The standard working balance target was raised from $3 billion to $5 billion in October
1988.
12 Depository institutions receive Treasury funds when their customers make payments to
the Treasury. Those that do not wish to participate in the TT&L program remit all such
funds to the Federal Reserve the next day. All institutions hold the funds for one day as
non-interest-bearing but reservable government demand deposits.
13 Same-day or next-day calls generally are only made on large banks (referred to as C
banks). Calls also are made on smaller institutions (A and B banks), but they are usually
made with longer lead times and are not used for marginal adjustments to the balance.
Calls are calculated as fractions of the book balance in each TT&L account on the previous
day. Direct investments will be sent to all depository institutions that have chosen to participate. They are computed as a share of the available capacity of each institution.




141

Chart 4 Currency in Circulation (Maintenance Period Averages)




Billions of dollars (not seasonally adjusted)
250 i

240

230

220

210

200

190

Jan. Feb. Mar.

Apr. May June

July Aug. Sept. Oct. Nov.

Dec.

The banks must pay interest on the TT&L accounts at a rate 1/4
of a percentage point below the weekly average Federal funds rate
and must hold collateral against them. Because of these requirements, the participating banks place caps on the amount of
Treasury balances that they will accept. At times when the
Treasury is particularly flush with cash, such as after some of the
major tax dates—those in the middle of January, April, June, and
September—its cash balances may exceed the capacity of the
TT&L accounts to a considerable degree. The excess cash will lift
the balance at the Federal Reserve (see Chart 5, page 143). As the
funds flow from the commercial banks to the Fed, they drain
reserves. Once the Treasury spends the money, the Treasury's balance at the Fed falls back to normal levels, adding to reserves.
Errors occur in the day-to-day forecasts of the Treasury balance
because it is not possible to estimate precisely the level or timing
of the myriad receipts and expenditures of the federal government.
Most of the time the errors have only a modest effect on the average level of nonborrowed reserves over the two-week reserve
maintenance period because the Treasury will adjust the size of the

142

Chart 5 Weekly Average Treasury Balances at Federal Reserve Banks
Billions of dollars
24
mill

22
20
18
16

*

14

i\

12
10
8
6
4

VI
I

K ,>IIU

£

1
I 1
\
V

A

L ^ 5

/Vs A

/I

is

3
I"

7987

2

1988
Jan.

Fed.

Mar. Apr. May

June

July

Aug.

Sept.

Oct. Nov. Dec.

next day's calls or direct investments in order to bring the balance
back to the normal target level. When total Treasury cash exceeds
the capacity of the TT&L accounts, however, changes in flows,
such as higher or lower receipts than expected, will affect the level
of the Treasury's balance at the Federal Reserve not just for a day
or two but until the balance drops below the TT&L capacity again,
a development which may take a couple of weeks. The resulting
reserve effect will be magnified.
Federal Reserve Float:
Federal Reserve float is generated when checks are processed
more slowly than specified in a preset schedule for crediting the
banks presenting the checks. When the presenting bank's reserve
account is credited before a corresponding debit is made to the
account of the bank on which the check is drawn, two banks will
simultaneously have the same reserves credited to their respective
accounts. Thus, float is a source of reserves. Float has declined
dramatically in recent years as the Fed has worked to discourage
it under the terms of the MCA (see Chart 6, page 144). In 1983,
the Federal Reserve started charging the banks explicitly for the




143

Chart 6 Annual Average Federal Reserve Float (Including As-of Adjustments)
Billions of dollars

7

1973

'74

'75

'76

'77

'78

'79

'80

'81

'82

'83

'84

'85

'86

'87 '88

float they generate. Float has also become more predictable as
forecasters have been able to gather more information about delivery and processing of checks. Float occasionally jumps unexpectedly, however, most commonly when bad weather interrupts
normal check delivery. Interruptions to the Fed's wire transfer systems can also create or reduce float. Errors introduced by incomplete or misdirected wire transfers are corrected after the fact with
so-called as-of adjustments. If the problem is not completely
resolved before the end of the maintenance period, the adjustment
may be made in a later period, thus affecting each period's reserve
availability; efforts have been made to minimize the unpredictable
component of as-of adjustments.
Foreign Exchange Intervention:
In the United States, foreign exchange transactions are not
undertaken for the purpose of affecting reserves. Nonetheless, the
reserve impacts of various foreign exchange developments have to
be taken into account along with other market factors. When the
Federal Reserve intervenes in the foreign exchange markets, it




144

either buys dollars—draining reserves—or sells dollars—adding
reserves. The reserve absorption or provision from the purchase or
sale of dollars usually occurs two business days after the intervention takes place. Generally, intervention is split between the
Federal Reserve and the Treasury. The Federal Reserve's portion
of the intervention will add or drain reserves when the payments
are made.'4 The reserve impact of the Treasury portion will depend
upon how the Treasury pays for it. If the Treasury pays out either
dollars or foreign currencies from the Exchange Stabilization Fund,
the intervention will generally have no reserve impact. The potential injection or withdrawal of dollars will be offset by adjustments
to the size of the call or direct investment made by the Treasury.
(When the TT&L accounts are at capacity, however, the intervention will change the Treasury's Federal Reserve balance and therefore will affect reserves until the balances fall back below the TT&L
capacity.) If the Treasury finances its portion of intervention to sell
dollars with dollars acquired from a sale to the Federal Reserve of
Special Drawing Rights (SDRs) obtained from the International
Monetary Fund, this "monetization" of SDRs adds reserves. If the
Treasury acquires dollars to sell from the Federal Reserve by placing foreign currencies with the Fed (with an agreement to buy them
back at the same exchange rate) in what is known as a warehousing technique, the warehousing will add reserves as the the
Treasury invests the funds.'5 The Treasury will reduce its TT&L
calls when it receives funds from SDR monetization or warehousing
that exceed its immediate intervention needs.
Reserve levels are also affected by monthly revaluation of
Federal Reserve holdings of foreign currencies to incorporate
changes in exchange rates. The Federal Reserve's profits from this
source rise when its foreign currency balances appreciate. The Fed
turns over its profits to the Treasury each week, allowing the
14 The reserve effect occurs if the Federal Reserve finances the dollar purchases by drawing
down its foreign currency balances or if it adds to its foreign exchange holdings when it
makes a dollar sale. If it finances a purchase by drawing on a swap line that it has
arranged with another central bank, however, it will credit dollars to the central bank that
provided the currency. When that bank invests the proceeds, the investment will offset
the reserve drain. If the Fed uses foreign exchange acquired from a dollar sale to pay down
a swap drawing, the reserve impact of the intervention will again be offset when the central bank pays back the dollars it acquired in the swap. Intervention has generally not
been financed with swaps in recent years.
15 If the Treasury intervenes to buy dollars and uses the dollars acquired to retire SDRs or
reverse a warehousing transaction, reserves will be drained.




145

Treasury to call in less cash from the banks. Losses from depreciation of foreign currency balances reduce its payments to the
Treasury and increase the need for TT&L calls.
Foreign Central Bank Transactions:
Many foreign central banks hold transactions balances at the
Federal Reserve for use in executing a variety of dollar-denominated
transactions. Transfers of funds into their accounts at the Federal
Reserve from commercial banks drain reserves. Because the central
banks cannot earn interest on these accounts, however, they hold
essentially steady working balances. The ultimate reserve effect of an
inflow of dollars to a central bank account depends on how the central
bank invests the receipts. If the funds stay within the Federal Reserve,
then the inflow drains reserves. The most common way for the funds
to stay within the Fed is for the foreign account to arrange a repurchase agreement (RP) with the Fed acting as counterparty. From the
Federal Reserve's perspective, this transaction is a matched sale-purchase agreement (MSP). The Fed arranges MSPs with the foreign
accounts when their cash buildup is expected to be temporary. The
reserve forecasts routinely allow for the drain from the inflow of funds
to the Federal Reserve and their arrangement as MSPs, since they
occur every business day. If the desk instead passes part of the orders
through to the market as customer-related RPs, the RPs add reserves
since the drain has already been factored into the assumptions about
reserve levels. The desk must estimate how large the foreign RP
orders will be in coming days. While the central banks attempt to predict any large flows into or out of their accounts, their estimates are
sometimes wide of the mark. Unexpected variations in the flow of
central bank RP orders cause errors in the reserve forecasts.
If a central bank expects a rise or fall in its cash holdings to persist,
it may ask the Federal Reserve to make an outright purchase or sale
of Treasury securities on its behalf. In contrast to the RP orders, these
operations are routinely arranged in the market, so the Federal
Reserve does not serve as counterparty. As long as the funds come
into the Federal Reserve and get paid out on the same day, there is no
reserve impact. Therefore, the projection staffs do not need to forecast
the outright transactions. There are times when the desk will be the
counterparty if it serves the desk's reserve management purposes. In
these situations, the transactions will have the same reserve impact as
an outright purchase or sale in the market, discussed in Chapter 7.




146

Other Factors:
Also affecting reserves are a number of other balance sheet
items such as interest accruals and remittances of profits to the
Treasury. For the most part, however, forecasting these factors is
reasonably straightforward. Gold holdings of the Federal Reserve
affect reserves, but the volume of reserves would change only if the
Treasury changed the amount of U.S. gold for which certificates
have been issued to the Federal Reserve or if the official price of
gold were changed. (The official price has been $42.22 per troy
ounce since 1973.)




147

The Conduct of Open
Market Operations
The Strategy of Reserve Management




The Manager of the System Open Market Account and other
trading desk personnel use the estimates of both the objective for
and the forecasts of nonborrowed reserves to develop a strategy for
bringing actual nonborrowed reserves into line with the objective
over the two-week reserve maintenance period. Working out plans
for adding or draining reserves to achieve the nonborrowed reserve
objective is partly an "art" that requires skill and experience in taking into account the many diverse factors affecting reserves. In
working out a plan, desk personnel assess the prospective as well
as the current estimated reserve situation. They also attempt to
make needed reserve adjustments in a way that does not conflict
with the thrust of policy. In seeking to maintain the desired degree
of pressure on bank reserves, desk personnel concentrate on achieving the nonborrowed reserve objective on average over the reserve
maintenance period. But they also are sensitive to the day-to-day
distribution of reserve shortages or excesses.
The Manager's task is to generate the desired degree of pressure
on bank reserve positions despite uncertainty about how bank
reserves are behaving. The short-term variability in reserve supplies
and demands arising from factors other than open market operations
is substantial and far outstrips the longer term needs for reserves
(see Chart 1, page 149). In 1987 and 1988, for example, the System's
securities portfolio rose an average of about $700 million per twoweek reserve maintenance period. The bulk of the increases supported currency, which expanded an average of $680 million per
period. Nonborrowed reserves rose about $65 million per period.
These moderate net changes in reserves and the portfolio mask the
rather substantial changes from one maintenance period to the next.
During those same two years, the average absolute change in
reserve availability attributable to operating factors (including currency) from one reserve maintenance period to the next was $2.7billion. A large portion of open market operations served to offset this
variability. The operations reduced average period-to-period
changes in nonborrowed reserves to about $900 million. Much of the
residual change in nonborrowed reserves accommodated seasonal
variation in required reserves (see Chart 2, page 149).
Further complicating open market strategy are the difficulties in
estimating reserve factors and the inevitable forecast errors that
result. (Box B in the preceding chapter described the major reserve
factors.) Average absolute errors—the averages of forecast misses
without regard to sign—of operating factor estimates made at the

148

Chart 1 Nonborrowed Reserves (Plus Extended Credit Borrowing)
(Biweekly Averages)

Billions of dollars
70

Including open
market operations
Excluding open
market operations
Note: Data are not
seasonally adjusted.
J F M A M J J A S O N D J F M A M J J A S O N D
1987

1988

Chart 2 Selected Reserve Measures
Billions of dollars

(Biweekly Averages)

65
Required reserves seasonally adjusted
Required reserves not seasonally adjusted
Nonborrowed reserves and
extended credit borrowing not seasonally adjusted

J F M A M J J A S O N D J F M A M J J A S O N D
1987




Note: Required reserve data
are not adjusted for changes
in reserve requirements.
Vertical lines indicate dates
when reserve requirements
where changed.

1988

149

start of each maintenance period were about $1.2 billion for 1987
and 1988. Comparable errors for required reserve estimates averaged about $200 million.
Each day the Manager of the System Open Market Account studies the various reserve estimates, allowing for possible revisions.
The Manager must decide whether to buy or sell securities outright
or to provide or drain reserves on a temporary basis, reversing the
impact on reserves sometime in the next few days. To achieve the
reserve objectives smoothly without generating undue confusion in
the market, the reserve management strategies of depository institutions are also taken into consideration. Depository institutions
may be more or less inclined to use borrowed reserves and may
occasionally show a preference for excess reserves that are higher
or lower than usual. These preferences will affect the relationship
between the demand for reserves and the behavior of the Federal
funds rate. The Manager weighs the various factors affecting
reserve supplies and demands and tries to cope with the large
swings in reserves in a manner that will be readily understood by
the depository institutions and others trying to duplicate the Federal
Reserve's projections of factors affecting reserves. Each day the
desk weighs both the action indicated by the reserve projections
and the wide margins of error in the projections. The desk's main
concern is to maintain a degree of reserve pressure on banks consistent with the nonborrowed reserve objective for the period.
The lengthening of the reserve maintenance period from one to
two weeks, a change that accompanied the switch to quasi-contemporaneous reserve requirements in February 1984, has had the
effect of giving the desk greater choice in timing the reserve adjustments within a maintenance period. The desk must plan its operations with sensitivity to both the overall reserve picture and the
day-to-day pattern of reserve availability. To continue targeting borrowed reserves, it became necessary to forecast required reserves
within maintenance periods and allow for significant revisions to
those forecasts.
Tools of Open Market Operations
The desk uses the System's portfolio to achieve its reserve
objectives. The FOMC spells out the Manager's authority in a special directive, which is reviewed at the February FOMC meeting
(previously March) and may be amended as necessary. (It is published in the Federal Reserve Bulletin with the Record of Policy
Actions for that meeting.) It authorizes outright transactions at mar-




150

ket prices in Treasury and federal agency securities with securities
dealers and with official foreign and international accounts maintained at the Federal Reserve Bank of New York. The FOMC also
authorizes the desk to make repurchase agreements (RPs) involving the same types of securities for periods of up to 15 days for the
New York Reserve Bank's account. When conducting RPs in the
market on behalf of official foreign and international accounts, the
desk interposes the New York Reserve Bank's account between the
foreign accounts and the market. Finally, as Chapters 2 and 5
explained, the FOMC sets a limit on the net change permitted in
the System's outright portfolio in the interval between meetings—
routinely $6 billion during 1989.7 This leeway for portfolio changes
between meetings has long since ceased to be the key indicator of
Committee policy preferences that it was in the 1920s and early
1930s. If the portfolio leeway appears to be insufficient to counter
seasonal swings in reserve factors, the Manager seeks an increase,
and the FOMC generally approves it with at most a discussion of
the technical factors behind the request.
7. Outright Operations
In buying and selling securities, the Manager of the System Open
Market Account functions within a framework of Federal ReserveTreasury relations that has evolved to keep monetary policy and
debt management separate. Currently, the degree of separation
between Treasury operations and the Federal Reserve far exceeds
that of the early years of the System. As indicated in Chapter 2, until
the Treasury-Federal Reserve Accord of 1951, the Federal Reserve
felt obliged to provide stable market rates for CJ.S. Treasury securities when the Treasury asked it to do so. During and immediately
after both world wars, the Fed was unable to use open market operations for policy purposes because the Treasury wanted rates held
steady to support its heavy debt issuance.
1 The authorization continues to permit aggregate holdings of prime bankers' acceptances up
to $100 million should circumstances call for it. In 1977, however, the Federal Reserve discontinued operations in bankers' acceptances for its own account when the FOMC concluded that the market was sufficiently developed to make Federal Reserve participation
unnecessary. The desk bought acceptances under RP until 1984, when it discontinued the
practice, although the authorization still permits such operations. By that time, the massive
volume of government debt outstanding usually provided sufficient collateral to achieve
the needed reserve injections without using bankers' acceptances.
The authorization also permits the desk to lend securities from the System portfolio against
collateral to dealers to smooth the clearing operations in the securities markets and ensure
effective conduct of open market operations. The loans have no direct reserve impact,
although the interest paid by the dealers results in a small reserve drain.




151




The Accord freed the Federal Reserve from the obligation to support prices in the secondary market, but for many years thereafter
the Fed continued to follow an "even-keel" policy. This meant that
the Fed operated in a way designed to hold rates steady in the period immediately surrounding major Treasury coupon financings. In
the 1970s, however, Treasury financings increased in frequency,
and interest rates became more volatile as inflation became a much
more serious problem. The Treasury gradually shifted its financing
techniques away from presetting interest rates to holding auctions
to determine rates. That change made it possible for the Federal
Reserve to move away from the even-keel restrictions and for the
Treasury to sell mounting quantities of debt efficiently.
The Federal Reserve makes all additions to its portfolio through
purchases of securities that are already outstanding. The Federal
Reserve Act does not give the System authority to tender for new
Treasury issues that are sold for cash. 2 Moreover, in a refunding,
the System cannot subscribe for a larger amount of the issues
offered than it holds of the maturing securities. Normally the
Manager will decide to roll over the System's maturing securities
by submitting a noncompetitive tender at the auction and receiving
the average auction rate. The Manager can, however, reduce the
System's portfolio by redeeming a part of the maturing holdings—
bidding at a higher rate than the Treasury is likely to accept for the
amount it wants to redeem. The desk occasionally redeems maturing three- and six-month Treasury bills as a means of absorbing
reserves. In 1989, when it had already reduced bill holdings substantially to offset increases in the foreign exchange portfolio, it ran
off some maturing coupon securities. It has also redeemed securities when no replacement issue was being delivered when the old
issue matured. That has occurred occasionally when the Treasury
has temporarily run out of legislative authority to issue new debt
and has paid down maturing issues.
In the secondary market, the Manager's key choices involve the
timing and type of security and the counterparties of the securities
transaction. The timing of outright activity depends principally on
the outlook for nonborrowed reserves. When the forecasts show
large reserve needs stretching several weeks ahead, the Manager
may decide to buy a sizable volume of Treasury securities in the
2 Over the years, a variety of provisions had permitted the Treasury to borrow limited
amounts directly from the Federal Reserve. Beginning in 1979, the length of the loan and
other conditions for use of the facility were also restricted. Authority for any kind of direct
loans to the Treasury lapsed in 1981 and has not been renewed.

152

market.3 Since the operations are geared to addressing extended
needs, they normally will be undertaken to coincide with the projected rise in the reserve need. Typically, they are arranged early in
the afternoon for delivery of the securities within the next day or two.
Desk officers also take market conditions into account in choosing
the day to arrange an outright operation. They try to avoid conducting operations in rapidly rising or falling markets, not wishing either
to add to market volatility or to impede price adjustments.
Another choice the desk makes in arranging outright transactions is between operating in Treasury bills and operating in
Treasury coupon securities (notes and bonds). In general, both
markets are sufficiently broad to accommodate the desk's purchase or sale of several billion dollars of securities at a time with
no undesirable effects.4 The desk has not tested the capacity of the
coupon market to handle sales for the System Account. (It does
make periodic sales on behalf of official foreign accounts, usually
for modest quantities of relatively short maturity issues.) In most
years, the need to sell securities arises only occasionally, and can
be accomplished through sales and redemptions of bills. As indicated above, the need to reduce the portfolio expanded in 1989
because of large scale purchases of foreign currencies, leading the
desk to redeem some maturing coupon securities.
The Federal Reserve prefers to have a portfolio that contains the
full range of maturities of securities that the Treasury has issued.
In recent years, the Treasury has auctioned on a weekly basis bills
that mature in 3 and 6 months; every four weeks, it has auctioned
bills that mature in a year. It sells coupon issues that mature at various intervals ranging from 2 to 30 years. Most coupon securities
are sold quarterly, although 2-year notes are sold monthly. To
some extent, the decision to buy bills or coupon issues will be
affected by market availabilities of different types of issues. For
example, in 1987 and to a lesser extent in 1988, outright desk purchases were weighted toward coupon issues when the Treasury
was reducing supplies of bills and adding to outstanding coupon
issues. The desk is, nonetheless, conscious of the greater liquidity
normally present in the bill market and seeks to make sure that it
3 The Federal Reserve also has authority to purchase securities of federally sponsored agencies. Typically, if the agency is offering a new issue, the Fed will acquire it in exchange for
its maturing issues. Market purchases have been infrequent; the desk last bought such
securities in the market in 1981 and last sold them in 1979.
4 Through October 1989, the largest single bill purchase (in 1986) and coupon purchase (in
1987) were both around $4 billion. The largest bill sale (in 1989) was about $4 1/2 billion.




153




holds a sufficient volume of bills to meet any plausible contingency
requiring substantial sales within a short time period.
When the desk executes outright purchases or sales, it selects
from among the dealers' offers or bids in order to achieve the highest or lowest yields to maturity in relation to the prevailing yield
curve. When buying Treasury coupon issues, the desk purchases
offerings throughout the maturity spectrum, planning in advance
the approximate amounts it will take from each maturity area.
Purchases reflect market availability rather than any desk view of
the interest rate outlook. At the margin, the System tends to buy
issues within each maturity area that are in excess market supply,
since those are likely to be offered at the highest relative yields.
As long as the operations are considered to be a routine part of
seasonal or secular reserve adjustments, the dealers will normally
have anticipated and prepared for the operation in their positioning
of securities. A desk purchase on a particular day may run around
10 percent of that day's market volume of bill or coupon trades
between customers and dealers, although a large bill transaction
may be above 20 percent. If the operation does not suggest a
change in policy has occurred, only minimal rate moves are likely
to result from the supply changes generated by the operation.
In addition to its transactions in the market, the trading desk has
the option to buy Treasury securities from foreign official accounts
on any day the accounts are selling or to sell issues from its own
portfolio to meet the buy orders of such accounts. These trades give
the desk a means of adding or draining relatively modest amounts
of reserves in a way that is quick, operationally simple, and unobtrusive. A market operation, in contrast, is rather cumbersome. While
a market go-around is in progress, it may be difficult for dealers to
trade since they do not know the results of their bids or offers with
the Fed. Consequently, the desk prefers to limit such operations to
times when the prospective reserve excess or shortage is relatively
large. On occasion, it may be an advantage to the desk that purchases from or sales to foreign accounts do not have the announcement
effects that accompany a market entry for the System Account.
During periods when persistent increases or decreases in reserve
needs are projected, the desk may be buying securities from, or
selling them to foreign accounts rather consistently, if orders permit.
Because the foreign orders are generally modest in size—from a few
million dollars to several hundred million dollars—the desk will usually buy all the bills being sold by the foreign accounts or sell all the
bills being bought. Transactions with foreign accounts are put

154

through at the middle of the latest bid and asked rates in the market.
Occasionally, if the order size is large relative to the needed reserve
adjustment, the desk will buy or sell only a portion of the foreign
orders. Foreign buy and sell orders that do not match the desk's
reserve needs may be arranged between two foreign accounts that
have purchase and sale orders of matching maturities. The desk will
execute the balance of the orders in the market.
2. Temporary Transactions
In managing bank reserves, the Manager finds it very helpful to
put reserves in or take them out in large volume for one day to a
few days at a time. Suppose the forecast suggests nonborrowed
reserves will be in short supply for the remainder of the current
maintenance period but will be about as desired for the next period.
One possibility would be to provide the needed reserves by buying
Treasury bills outright. The bills could then be sold from the portfolio in the next period when the additional reserves were no longer
needed. But such a course would involve substantial transactions
costs. The System could instead buy securities under RP, which
obligates dealers to return the cash plus interest and reacquire the
securities on the desired date. Eligible collateral includes not just
bills but also Treasury coupons and federal agency securities held
by both dealers and their customers. 5
The enlarged collateral and reduced market risk mean that the
desk can conveniently undertake a larger operation through an RP
than through an outright purchase. 6 To some extent, collateral
offered to the desk reflects the availability of securities being
financed from day to day. This factor in turn depends partly on
floating supplies from recent Treasury sales and on positioning
strategies of dealers and others active in the RP markets. If market
participants anticipate that interest rates will soon decline, they
tend to add to their positions. Accessible collateral becomes plentiful and the RP rate rises toward the Federal funds rate. When higher rates are anticipated, dealer positions are cut back sharply and
the RP rate may fall well below the Federal funds rate.
When the Manager wants to absorb reserves for one or a few
days, matched sale-purchase transactions (MSPs) with dealers
5 Those federal agency securities that are eligible for outright purchase are also eligible for
purchase under RP.
6 Through October 1989, the largest volume ofRPs arranged on a single day was almost $16
billion (May 1989), and the largest volume of contracts outstanding was about $22 billion
(May 1987).




155




provide a convenient mechanism.7 In an MSP, the desk sells
Treasury bills from the System's Account for immediate delivery
and simultaneously buys them back for delivery on the date
desired. This procedure provides securities to be financed for one
or a few days, making it unnecessary to add to dealer positions at
risk of loss from a price drop. While MSPs are just the reverse of an
RP in their effect on reserves, their form is different. Technically
they encompass two separate outright transactions.
The desk relies heavily on temporary reserve operations in dealing with the uncertainties that affect bank reserves. Even when the
reserve forecasts on the first day of the maintenance period indicate
no need for System action, reserves may actually turn out substantially higher or lower than projected. RPs or MSPs enable the desk
to respond quickly when reserves fall short of desired levels or
prove excessive. Furthermore, a need or an excess may be concentrated on certain days within the period, and temporary transactions
can help to smooth the daily pattern of reserve positions.
As Chapter 6 indicated, many foreign official and international
accounts place a portion of their dollar holdings in a daily RP
investment facility provided by the Federal Reserve Bank of New
York. How the desk handles these orders will determine their
reserve effect. It can arrange the RP orders internally, executing an
overnight MSP using its portfolio, or it can pass them through to
the market as a customer-related RP. Reserves are drained under
the first option because the funds received by the foreign account
from a commercial bank remain with the Fed. The reserve projections allow for this drain on the assumption that the investment
orders will be arranged internally. Consequently, if some of the
orders are instead passed through to the market as customer-related RPs, they add reserves relative to the projected level.8
The choice between a System RP and a customer-related RP
depends largely on the magnitude and duration of the reserve injection that the desk wants to accomplish. Customer RPs are normally
arranged to mature on the next business day, since participation in
the foreign RP pool changes each day. Customer RPs are limited in
volume, since they must not exceed the total funds available to for7 Through October 1989, the largest MSP operation made in the market amounted to $7 3/4
billion (March 1979). The largest balance of such contracts outstanding was about $10 1/2
billion (February 1989).
8 In a technical sense, customer-related RPs merely offset the drain of reserves that results
when the foreign central banks receive money into their Federal Reserve accounts. Since
foreign investments occur every day, the Federal Reserve builds an estimate of the reserve
drain into its reserve forecasts.

156

eign accounts for investment. The choice of customer-related RPs
may be interpreted as indicating that the estimated reserve need is
modest or uncertain in size. When the reserve need is large and
expected to persist for a number of days, the Manager is likely to
choose System RPs in the market. Sometimes the choice between
System and customer RPs may be influenced by concerns about
how the operation will be interpreted by market participants.
A Day at the Trading Desk
The working day at the trading desk has a regular rhythm, which
flows from the task at hand. The morning is filled with a number of
information-gathering activities and meetings that prepare the
Manager to make the day's policy decisions. Using direct telephone lines, traders at the desk talk to dealers selected by the
Federal Reserve for a trading relationship.9 The dealers are asked
how they expect the day's activities to unfold in the securities markets and how the task of financing their securities positions is progressing. Traders talk to the large money center banks about their
reserve needs and their plans for meeting them. Meanwhile, the
reserve forecasters in the New York Bank's Research Department
and at the Board of Governors in Washington will begin gathering
data on bank reserves for the previous day and on factors that
could change their projections for future days. By midmorning, the
information is used in determining what, if any, open market operations to arrange that day. The plan will be reviewed at a conference call and then carried out.
The officers and senior staff that participate in the policy process
begin their day by gathering information about various financial and
economic developments and how they are being interpreted by
analysts and market participants. U.S. government securities trade
in Tokyo and other Asian centers and in Europe while it is nighttime
in the United States. Thus, part of the preparation involves reading
headlines from various news services concerning developments
abroad. U.S. economic data are often released at 8:30 a.m. If any
such economic reports are made that day, the officers and staff will
read them on the news screens along with any descriptive information concerning special factors that may clarify the data. A number
of financial services whose findings are available on information
screens provide analyses of recent economic and monetary releases
9 The desk chooses its trading partners from the list of primary government securities dealers
that report daily data to the New York Federal Reserve. Criteria for inclusion on that list are
described in the final section of this chapter.




157

and offer their forecasts of coming data. This background information helps the officers and staff to understand what the representatives of the government securities dealers have in mind when they
arrive beginning at 9 a.m. for a series of meetings.
1. Daily Dealer Meetings




Each day, one or two members of the Open Market Function,
usually including the junior officer or senior staff member who will
be speaking on the conference call later that morning, will hold a
series of 15 minute meetings with representatives of government
securities dealers. Normally, three or four such meetings are
scheduled each morning. The meetings with dealers that are located near the Federal Reserve in the New York City financial district
are held in person at the Reserve Bank. For the dealers located in
midtown Manhattan and in other cities, the meetings are conducted over the telephone. Over a three-week period, each of the dealers with which the Federal Reserve has a trading relationship will
have the opportunity to talk with the representatives from the desk.
These discussions with leading members of the dealer community help the people from the desk keep abreast of the forces at
work in financial markets. The conversations are free flowing, covering a wide range of subjects. They are full of market jargon and
shorthand references, puzzling to outsiders but well suited to conveying information efficiently to the Federal Reserve. The desk
participants do not comment on current policy although they may
answer questions on technical matters.
A dealer representative may open the meeting by giving the firm's
operating assumptions about the desk's current policy objectives.
The dealer usually describes how rates are expected to change in
coming weeks. The firm's money market economist often outlines
the economic analysis and interest rate outlook the firm is presenting
to clients. Particularly useful are dealer comments on what their customers actually are doing in the market—sitting on the sidelines or
clearly favoring a particular type of security. The dealers will often
give their own estimate of the technical position of "the Street" by
identifying shortages in particular maturity areas or excess supplies.
This information helps to explain the way the market has been
behaving in recent days. The succession of meetings each day, day
after day, keeps the desk in close touch with the changing analyses,
moods, and concerns of the market makers themselves.
Treasury financings provide a recurring theme for the dealer
meetings. The Manager or an associate may ask how much debt

158

the market expects the Treasury to sell in the current or following
quarter. They get running reports on the progress in distributing
recently sold issues, and news of potential customer interest in
coming offerings. To dealers, bidding in Treasury auctions of bills
and coupon securities is an extension of the daily task of making
markets in outstanding issues.
Dealers associated with banks may comment on bank trading
activity, the strength of business loan demand, and any issuance of
wholesale market instruments to fund lending. Some dealers will
offer information on mortgage-market developments and the liquidity and earning pressures experienced by thrift institutions.
Desk participants also gain insights into prospective corporate and
municipal bond flotations by quizzing the dealers with large investment banking operations.
After concluding the dealer meetings around 10 a.m., the people
from the desk return to the trading room. They will check the information screens for the latest quotes on the most actively traded
government securities, for news headlines, and for updated commentary from money market analysts. They will peruse any important news developments that traders will have clipped from several
news tickers.
2. The Treasury Call

A daily conversation with the Treasury takes place around 10:30
a.m. In preparation for this call, the Manager and other officers and
staff of the Open Market Function gather in an office near the trading room and review data on the previous day's member bank
reserves and borrowing. Soon the New York staff reserve forecasters arrive and present a preliminary estimate of what nonborrowed
reserves will be on average over the reserve maintenance period in
the absence of any further open market operations. Their estimate
gives desk personnel an idea of the operations that will be needed
to achieve the nonborrowed reserve objective.
The first step in refining this estimate is to examine the assumptions about the Treasury balance at the Fed on that day and the
next two days. As noted in Box B of Chapter 6, the Treasury balance is often the biggest source of uncertainty about reserve levels.
After a review of the figures, a desk official will place a call to
Treasury Department personnel who make their own estimates of
Treasury cash flows. The Treasury attempts to keep its balance at
the Fed relatively steady so as to minimize its impact on bank
reserves. When the Treasury and New York staffs suggest that the




159

A Day at the Trading Desk
8:30 am

9:00 am

Watching the market react to data

Discussing market developments
with securities dealers



10:30 am

10:45 am

Consulting the Treasury by phone about
its balances at the Federal Reserve Banks

Developing a plan of action for the day

11:15 am

11:40 am

5:00 pm

• MHfif

The daily conference call with
FOMC representatives

Reviewing how the day turned out

Contacting the dealers to announce an open market operation



balance is likely to move away from desired levels, the Treasury
will, if possible, take action to bring the balance back into line by
transferring funds to or from depository institutions in the form of
direct investments or calls.
On occasion, the Manager may ask that the balance be shaded
somewhat to the high or low side to assist in the management of
reserves. Such requests are most common when the Treasury balance is expected to rise or fall sharply, perhaps in conjunction with
tax flows. If the Treasury adjusts its actions to achieve a more gradual rise or fall over several days, it can make managing the reserve
adjustments less difficult.
Other topics come up at the Treasury conference call as well.
Desk officers may pass along information about foreign central
bank subscriptions to forthcoming Treasury issues, subscriptions
which often affect the projected cash position. The timing of future
Treasury offerings and the associated payment dates also may be
discussed, since these affect reserves.
After the Treasury call, the forecasters will, if necessary, revise
their reserve estimates in light of the actual Treasury actions. About
that time, one of the traders will phone to report the preliminary estimates of outright orders and the size of the foreign RP pool for the
day that are prepared from information provided by the New York
Reserve Bank's Foreign Exchange Function. The reserve forecasters will incorporate the RP pool into their estimates. They will then
depart for their offices to make final refinements to their forecasts.
3. Formulating the Day's Program
The Manager and other members of the open market area will
then begin a preliminary discussion of the appropriate plan of
action for the day. Some of the subjects discussed will be the same
each day, while others will depend upon whether the two-week
maintenance period is just beginning, partially completed, or
drawing to a close. Early in the period, the options for making any
estimated reserve adjustments will be greatest. But the chances of
substantial revisions either to the objective for nonborrowed
reserves or to reserves provided by market factors are also greater
than they will be later in the period.
Throughout the maintenance period, the starting point of the
discussion will be the estimates of the need to add or drain reserves
for the period in progress and coming periods. A telephone call to
the Board will provide a preliminary reading on the Board staffs
reserve estimates as well as an opportunity to provide senior Board




162

Table 1 Sample Reserve Data (in millions of dollars)
Reserve
Period
Ending

RR
Estimate

ER
Assumption

BR
Assumption

NBR
Objective

Projected
Supply NBR

Open Mkt.
Oper. needed

Aver. ER
to Date

Aver. BR
to Date

2/28

60,250

1,000

500

60,750

NY 60,150
Bd 59,850

+600
+900

900

400

3/14

60,850

1,000

500

61,350

NY 61,450
Bd 61,400

-100
- 50

—

—

3/28

60,400

1,000

500

60,900

NY 60,600
Bd 60,200

+300
+700

—

—

If these were the figures presented to the desk on February 22, midway through a maintenance period that would end on February 28, it would need to add $600 to $900 million of
reserves on average for the period to reach the NBR objective. Since half of the period has been
completed, it would need to add $1.2 to $1.8 billion of reserves on average over the remaining
days to reach the NBR objective. The ER and BR to date are close enough to the path assumptions to suggest the NBR objective can be obtained without causing an unduly high or low
Federal funds rate. (Abbreviations are defined on page 124.)
The desk knows that the forecasts are uncertain. NBR could end up being higher than
$60,150 million or lower than $59,850 million by several hundred million dollars.
The desk would probably not choose to make an outright purchase of securities since the
estimates suggest that there will not be a need to add reserves in the following period.
Given the forecast uncertainty, the desk would probably arrange an overnight customerrelated RP for about $1 1/2 billion, which would keep pace with the remaining daily average
need (midway between The Board and New York staff estimates).

staff with preliminary indications of what sort of action the Manager
may be contemplating. (Table 1 illustrates one possible situation
the desk might face.) Generally New York and Board staff forecasts
will be given similar weight in the deliberations. If the forecasts suggest that nonborrowed reserves are not close to the objective, the
participants will discuss ways to bring them into line. They will consider whether the expected reserve adjustments are going to be
persistently in one direction. If so, they would discuss whether to
undertake an outright operation in the market. As an alternative,
they may be able to accomplish the reserve adjustment using
modest outright operations with foreign accounts, supplemented
at times with temporary operations. If an extended need to drain
reserves is expected, the Manager also has the option of allowing
securities to mature without replacing them.




163




Another factor considered each day is the expected reserve
excess or deficiency on that day. If reserves are especially short,
some banks may face overdrafts if additional reserves are not provided through an open market operation. To avoid ending the day
overdrawn, those banks short of reserves are likely to borrow at the
discount window. Before they borrow, however, they will probably
bid up the Federal funds rate. Just how short reserves have to be
before borrowing will rise above amounts that the desk desires will
depend upon how much routine borrowing the desk is seeking. It
will also depend on the volume and pattern of flows through the
financial system expected that day. For example, the high volume
of transactions through the Federal Reserve's "Fedwire" funds
transfer system that accompanies deliveries of new Treasury
issues or social security payments will increase the need for
reserve balances. The desk could probably arrange overnight RPs
to ease a shortage on the day without much risk of making
reserves overly abundant for the period as a whole, even if the
overall need were modest. An overnight RP would add to the average reserve level for the two-week maintenance period only onefourteenth of the dollars paid out.
The desk will watch the Federal funds rate each day. The
Manager estimates the range of rates at which funds will most likely
trade given the discount rate and the amount of borrowing being
sought. The funds rate may move outside the anticipated range for
a number of reasons. There may be a reserve need or excess, which
may or may not be captured in the projections. The funds rate may
also reflect expectations of a policy change or a poor distribution of
reserves among banks. The desk officials will try to read the rate's
significance. If the level of the funds rate is inconsistent with the
reserve estimates, the desk officials must decide how much weight
to give to the reserve estimates as compared with the funds rate. A
number of factors influence that decision. Sometimes the desk may
fear that the funds rate could mislead the banks about the Fed's policy intentions, particularly if a shift in policy stance is widely expected. Moreover, the desk may on occasion wish to give more weight
to the signal from the funds rate because its confidence in the
reserve forecasts is low. For example, there are times, such as
around tax dates, that reserve estimates may be especially uncertain. At other times, however, the desk may feel more comfortable
in taking guidance from the reserve estimates, possibly when a
reserve distribution pattern suggests an explanation for the level of
the funds rate. Moreover, there is a desire to avoid such close atten-

164

tion to the funds rate as to suggest that even minor variations from
some expected central tendency should be resisted.
On occasion, the FOMC may have indicated that it wants to be
especially sensitive to conditions in the foreign exchange markets.
At such times, reports from the foreign trading desk suggesting disturbing movements in exchange rates may temper the Manager's
response to reserve projections and funds rate signals. The main
effect of exchange rate developments would tend to be on timing of
open market operations within the maintenance period, rather than
on the reserve objective itself or the decision to meet it.
Early in the maintenance period, the discussion will give particular attention to the expected daily pattern of reserve excesses and
deficits for the period. Other things equal, the Manager prefers to
add reserves, if that is the action needed, on days when the reserve
deficiencies are expected to be particularly large. Sometimes the
pattern of reserves may be so skewed that the Manager will plan to
both add and drain reserves within the maintenance period. The
estimates at the start of the period only rarely show the extremely
uneven pattern of reserves that would warrant such an approach,
although subsequent revisions to market factors or the nonborrowed reserve objective may lead to that result.
On the first day of the period, the reserve forecasters will have
presented formal estimates of the demand for excess reserves
based on econometric models and some individual judgment. The
initial estimates are computed without any figures on the reserve
excesses or deficiencies carried into the period, although estimates
of the previous period's excess reserves may suggest whether
reserves carried in will be positive or negative. More complete
information will be available on the second Thursday, at which time
the forecasters will revise their estimates.
After the first day, the discussions will include a review of the
planned approach to see if it needs to be modified. Small changes
are the norm, since reserve factors rarely behave exactly as expected. Major shifts in the outlook for reserves within a maintenance
period generally occur a few times a year. As a reserve period progresses, the Manager will consider the distribution of reserves so far.
For instance, have the small banks been holding unusually large
amounts of excess reserves that might mean actual excess reserves
will run higher than forecast or have the large money center banks
accumulated abnormally large reserve deficiencies that will have to
be covered, possibly creating extra money market pressures? The
traders at the desk have daily conversations with the large banks




165




during which they may obtain useful clues about the bank's reserve
management strategies. Finally, are banks borrowing actively relative to the planned average? Usually, the money center banks borrow only on the last day of the period unless unusual reserve
distributions or wire or computer transmission problems leave them
facing an overdraft at day's end when routine means of covering it
are no longer available. If the large banks have not borrowed, then
the day-to-day borrowing tends to run a bit below the planned average until the final day of the period.
Near the end of the maintenance period, the desk will have to
take a very close look at borrowed reserves and excess reserve
behavior to date, to see if it is still appropriate to achieve the nonborrowed objective. For instance, banks may have borrowed so
much at the discount window early in the period that it becomes
mathematically impossible to achieve the planned borrowing level,
since borrowing can never be less than zero. In practice, there is
always some borrowing since many small institutions use the seasonal program for extended periods and often use the adjustment
program for several days at a time. If large amounts of borrowing
have occurred and the Manager still decides to meet the nonborrowed reserve objective, total reserves would be high and consequently excess reserves would be large relative to probable bank
demands. The excesses would most likely foster easy money market conditions near the close, which could convey a misleading
impression as to policy intentions. Alternatively, the Manager could
opt to allow nonborrowed reserves to fall short of the path, accepting the inevitable overshoot of borrowing in order to prevent an
overabundance of excess reserves. If low borrowing had occurred
through most of the period such that borrowing would have to rise
sharply on the last day, the funds rate would be subject to upward
pressure. Again, the Manager will have to weigh the consequences
of meeting or of deliberately overshooting the nonborrowed
reserve objective. Either course of action could leave the financial
markets with mixed signals about policy intentions.70 Relative sensitivities to the reserve measures as against the funds rate tend to
shift as policy priorities change.
The amount of excess reserves finally reported for the period will
largely be determined by what has already happened. If small institutions have built up a large reserve excess, they will only work off
a limited portion of it. Many never run reserve deficiencies and oth10 Federal Reserve actions are studied carefully by banks, dealers, and other market participants. 'Ted watching" is described in Chapter 8.

166

ers only run small ones because the reserve balances that they are
required to hold at the Federal Reserve are modest. Large banks
usually meet their reserve goals, although occasionally a large
reserve miss during the period may have locked one or two of them
into excess reserves positions that they cannot eliminate without
ending one of the remaining days with their accounts overdrawn.
Since excess reserves are not a target in themselves, the Manager
is likely to make an allowance for expected deviations from the
assumption used in the objective.
Not all of these issues must be readdressed at each day's morning strategy session, however. The actual conversation among
open market officials will usually last only 10 to 15 minutes,
depending on the number of issues present. Around 11 a.m. the
reserve forecasters will deliver computer spread sheets containing
their final reserve estimates as well as those of the Board staff. If
the estimates differ significantly from those made earlier, some
modification of the plan of action may be needed. Usually, however, the forecasters have made at most minor adjustments to the
earlier estimates.
As the discussion progresses toward a conclusion, one of the
junior officers will write a program of action for the day. The program will review the reserve situation and any other considerations
that have influenced the approach taken. On Thursdays, the program will also summarize the forecasts of the monetary aggregates, which will have been revised the night before on the basis of
an additional week's data and reviewed earlier that morning. The
Thursday program will describe the degree of reserve pressure
being sought. If the approach to reserve provision is being
changed, the reasons will be indicated. Once the program is drafted, the Manager and other officers will review it.
Toward the end of the strategy session, the person who will be
speaking on the conference call will go to the nearby trading
room, to be briefed by the traders who follow developments in the
various financial markets. This person will first be given a rundown on developments in the Treasury market. One trader summarizes price moves since the previous close, beginning with
developments in Asia and Europe. The trader then reports on
market reactions to indexes of economic data and describes other
factors that the dealers have mentioned. Another trader will report
on the corporate and municipal markets, noting new issues that
have been priced that day and highlighting other topics that dealers have mentioned. Even though the desk does not deal in these




167

issues, it likes to have a sense of what is happening, because
developments in other markets and the economy often impinge
on the Treasury market.
Someone else will report on short-term financing in the Federal
funds and RP markets and will note conditions relating to the new
issuance of bank paper. The trader who has been talking with the
money center banks will offer updates on how the banks see their
own reserve needs and will report their perceptions of overall
reserve availability. In the few remaining minutes before the conference call begins, the person who will do the call consults the
information screen to see where the dollar is trading relative to several key currencies and then telephones the foreign trading desk.
The foreign desk will provide a review of foreign exchange trading
abroad and locally, a report on intervention by the Federal Reserve
or foreign central banks, and an enumeration of factors behind
movements in the exchange rates.
4. The Conference Call




Around 11:15 a.m., a secretary will report "the call is in." The
person who will be talking will gather up his or her notes, along with
sheets of paper with information about reserves and the securities
markets, and head for the office near the trading room where the
call will take place. The Manager, other officers, and senior staff
members will already have gathered. The conference call links the
desk with the office of the Director of the Division of Monetary
Affairs at the Board where several Board staff members will have
assembled, and with one of the four Reserve Bank presidents outside of New York serving on the FOMC. The call provides the Board
staff with the desk's review of market and reserve developments
and the Manager's plan for open market operations. The Board
staff condenses this into a brief report, given to each governor by
early afternoon and wired to each Reserve Bank president. The call
also enables the Manager to consult daily with one of the Committee members concerning the desk's execution of FOMC instructions. The Reserve Bank president on the call not only has an
opportunity to comment daily on the desk's approach, but also
gets a sense of the circumstances, including uncertainties and difficulties faced by the desk between meetings.
The call itself usually runs 15 to 20 minutes. The person speaking
for the desk draws upon the findings of the morning's informationgathering sessions to give a review of price and rate movements in
the principal securities markets and to explain the influences mar-

168

ket participants see at work. Mention is made of current developments in the gold and foreign exchange markets. The speaker will
describe the state of the Federal funds market, giving the latest
information on rates and on the progress made by the major banks
in covering their reserve needs. The speaker then discusses the
reserve outlook and explains revisions to the figures. If the reserve
forecasts of the Board and New York staffs differ significantly, the
call provides a convenient opportunity for discussing the reasons
for the discrepancy. During the course of the maintenance period,
the estimates usually come closer together, but they may still be
one or two hundred million dollars apart on the Wednesday settlement date. Finally, the speaker reads the proposed program of
action and asks the Reserve Bank president for comments.
(Occasionally, a governor will sit in at the Board and will also be
asked to comment.)
Usually the president will concur in the Manager's approach.
Occasionally, the president may ask the Manager whether an alternative approach has been considered. Such a question will elicit further
elaboration of the reasoning behind the proposed program. If more
discussion of recent developments seems needed, the Manager usually calls the president after the day's operations are launched.
The Director of the Monetary Affairs Division at the Board uses
the call to give the other participants the latest information on the
monetary aggregates. Usually by Tuesday morning the Director's
staff has a fairly good fix on bank deposits for the week ended on
Monday, eight days earlier. These figures will be incorporated into
the weekly levels of Ml, M2, and M3 to be published on Thursday.
Often the Director will discuss Ml for that week and give a first estimate of the Ml data for the week that ended the previous day,
drawing on reports for part of the week by the large institutions and
a sample of other institutions. Deposit data are subject to further
refinement on Wednesday; by Thursday morning the Director will
report the final monetary data for the week ended 10 days earlier
to be published that afternoon and will review the revised projections of the aggregates for the current month.
The Chairman of the Board of Governors does not generally sit
in on the call but is kept fully informed of all significant matters
relating to open market operations. The Manager and Director
make sure the Chairman knows in advance about large outright or
other significant open market operations. If necessary, the Manager, the Director, and the Chairman will discuss unfolding economic and monetary data that may bear on how the Committee's




169

directive is to be carried out—particularly whether a change in the
desired degree of reserve pressure may be appropriate. The Chairman may decide that a consultation of the full Committee, or perhaps a formal telephone meeting, is in order.
5. Executing the Daily Program
The call typically ends shortly after 11:30 a.m.; the officers usually carry out any temporary transactions authorized in the program fairly soon afterwards.77 In the go-around, traders contact the
dealers quickly so that the announcement time is kept short. At the
same time, an officer will inform the public information area of the
Federal Reserve Bank of New York of the action to enable its press
officers to respond to inquiries from the news media.
When arranging RPs, either for the System or for foreign official
customers, the desk notifies each of the dealers that it wants to do
RPs for a specified period, most commonly ranging from overnight
to one week. Within about ten minutes, dealers will begin to call in
offerings of the par amount that they and their customers want to
arrange and the interest rate they are willing to pay on each lot.
Typically, traders on the desk will have to round up the stragglers
to complete the array of competitive propositions over the next 10
minutes. The traders consolidate the offering amounts by rate and
then inform the Manager of the totals at each rate. The Manager
has in mind an approximate volume of RPs to be arranged, but
there is some flexibility, especially early in the period when further
reserve adjustments can be made on subsequent days. A somewhat greater volume may be arranged if large offerings or a stringent Federal funds market suggests a larger-than-forecast need for
reserves. If offerings are skimpy or unattractively priced, a smaller
volume than was contemplated initially may be executed. Once
the Manager decides how low a rate to accept, the traders quickly
notify all dealers of the propositions accepted and rejected. Each
accepted proposition will be arranged at the rate the dealer offered.
The whole operation is over in about a half hour.
After an RP operation is completed, those dealers whose propositions were accepted notify the traders on the desk of the specific
securities that they or their customers are selling to the Federal
Reserve under RP. Since the specific issues identified may trade




11 On rare occasions, such as in the tumultuous market conditions that accompanied the
stock market crash in October 1987, the desk may operate before the normal time.
Expectations of collateral shortages or an early closing of the markets ahead of a major
holiday have also led to early entries.

170

significantly above or below par depending on whether their
coupon interest rates are higher or lower than current rates, the
collateral value of each security must be assessed. A price is set
somewhat below the bid price currently being quoted in the market
plus any accrued interest on the issue. The dealer receives that
price in return for the securities. This pricing procedure affords the
Federal Reserve protection against declines in market price during
the term of the contract and potential losses should the dealer not
return the money when the contract matures. The task of valuing
the securities is in the process of being automated. Computers will
soon be used to do the necessary calculations. The accounting
division of the Open Market Function puts the needed information
into the computer to start the delivery process through the
Securities Clearance Department. Procedures allow payment to be
made only when the securities are received.
RPs are of short duration and dealers generally may withdraw
from multiday transactions before maturity, a possibility which
complicates the accounting. Hence, RPs are made for the account
of the Federal Reserve Bank of New York rather than for the
System Open Market Account because the System Account must
be divided each business day among the 12 Federal Reserve
Banks. RPs are subject to the terms of a comprehensive collateral
agreement, which affirms the right of the Reserve Bank to sell the
securities in the event the dealer fails to repurchase them.
Both the desk and the dealer retain the option to terminate routine multiday RP contracts before maturity. In practice, the desk has
not exercised its option. Dealers often do, repurchasing their securities because they wish to sell them to another party or because
they have arranged an RP from another source at a lower rate. The
right of withdrawal means that the desk is uncertain at the time it
makes multiday RPs what the effect will be on average reserve levels for the period. The early withdrawal option can sometimes work
to the desk's advantage. If nonborrowed reserves are too high, the
funds rate is likely to slip, encouraging dealers to repurchase their
securities early and absorb reserves in the process.
Nevertheless, the withdrawal feature also can lead to difficulty in
maintaining desired reserve levels—for example, when a strong
demand for securities rather than an abundance of reserves causes
the withdrawals from RPs. Furthermore, although withdrawals are
permitted until 1:00 p.m., the desk enters the market earlier than
that. If the desk wishes to replace the RPs withdrawn, it must do so
without knowing total withdrawals. When the desk wants to be sure




171




that its reserve injection sticks, usually when the need for reserves
is especially large, it may offer nonwithdrawable RPs. Of course,
the absence of an option to terminate the contract early can lead
the dealers to offer less or offer lower rates than might have been
the case otherwise.
If the day's action is a matched sale-purchase agreement (MSP),
the entry time is the same but the operational procedures are
slightly different. The sale side is an outright sale of a specific
Treasury bill (or specific bills) from the System's portfolio. The
purchase side is a contract to buy back the bill for delivery at a particular future date. In executing MSPs in the market, desk traders
notify dealers of the market rate at which the System will sell the
particular bill. Dealers then indicate the amount they are willing to
buy and the rate at which they will reoffer. The rate of discount set
by the desk determines the price realized by the System on its sale,
while the competitively set reoffering rate determines the prices at
which the System reacquires the same bill one or more days later.
MSPs are not subject to withdrawal since they are literally matched
outright transactions with specified delivery dates.
When the desk executes an outright transaction in bills or coupon
securities it usually enters the market in the early afternoon. The
transactions are for delivery the next day or two days forward.
Outright operations are more time-consuming than temporary transactions, both because dealers contact a larger number of customers
and because a larger number of bids or offers have to be recorded and
reviewed. Choosing among the propositions is also more complicated because a yield curve involves more choices than the single interest rate comparison typical of RPs and MSPs. Generally, an hour or
more is required for a go-around in Treasury bills. A similar operation
in Treasury coupon securities can require two hours or more, largely
because of the increased number of issues and the operational time
required by desk personnel to convert prices to yields and to compare
offering yields relative to the yield curve. (Plans to automate these
operations are in the development stage.)
As dealers call back, traders record dealer bids or offerings on
strips preprinted with two dealer names at the top of each strip. The
issues involved have been posted on each strip so that the rates for
bills, or prices for coupons, and the amounts can be listed quickly.
Desk traders read back each dealer's propositions to guard against
communication errors. The officers assemble the go-around strips
used by the traders on one or more boards. The rates or prices being
quoted by different dealers for each issue line up horizontally. The

172

different maturities run vertically. The officers then choose the better propositions (based on maximizing yield to maturity) from the
array before them, bearing in mind predetermined parameters for
the size and distribution of purchases desired. Subsequently, they
return the strips to the individual traders so that they may let the
dealers know which propositions were accepted and which rejected.
Next, the traders must record the trades on tickets that will, in
turn, be entered onto the computer for subsequent processing.
When a dealer has served as an intermediary for a customer, the
dealer or its clearing bank tells the Federal Reserve Bank the name
of the institution that will deliver the securities and receive payment
over Fedwire. Purchases of the dealer's own securities generate
credits to the dealer, if a bank, or to the dealer's clearing bank.
Regardless of which bank receives the initial reserve credit,
banks across the country will soon benefit from the infusion of
reserves, although they may not distinguish it from all the other
bank and customer transactions that are flowing through their
reserve accounts. In a similar fashion, sales of securities reduce the
reserves of individual banks and the effects quickly spread through
the banking system.
Communications Within the System
One of the desk's functions is to keep others in the System informed, not only about how operations are carrying the Committee's thrust forward into the financial arena but also about the
kind of feedback that its policy is receiving. Regular desk reports to
the Committee and visits by System personnel to the desk are
important means of maintaining a clear understanding of these key
aspects of the monetary policy process.
The desk's main channels for communicating with the rest of the
System are the daily wires to the Reserve Banks and the written
reports sent to the Board and the Reserve Banks. Wires sent from
the trading desk describe each day's developments in the financial
markets. As noted earlier, the Board staff's wire describing the
11:15 a.m. call gives a full view of reserve data, the markets, and
the desk's program for the day.
Every other Friday the analysis division of the Open Market
Function mails a comprehensive report on operations for the maintenance period ended on the preceding Wednesday. This report
describes the daily conduct of operations and tracks the behavior
of the reserve measures in relation to the objectives. It also conveys the latest data and projections of the monetary aggregates.




173

Separate sections examine bank reserve management, money
market developments, and developments in government, corporate, and municipal securities markets. Statistical appendixes
summarize all transactions in the System Account and contain an
array of information on the financial markets.
Before each FOMC meeting, the analysis division prepares a
summary report of operations and financial market developments
since the last Committee meeting. The officers and staff most
closely involved with the reports also prepare a comprehensive
annual report analyzing policy implementation and financial market developments for the year. A modified version of this report is
published, generally in the spring issue of the New York Federal
Reserve Bank's Quarterly Review.
The Trading desk also has a regular flow of visitors from the
staffs of the Board and other Reserve Banks. They observe operations so that they can brief their principals and do monetary
research. Usually visitors participate in the desk's daily routines for
a week and spend time with government securities dealers, a bank
money desk, and a Federal funds broker. Desk officers and staff
lead the visitors through the morning dealer visits, the Treasury
call, and the 11:15 a.m. conference call, explaining the array of
data that influence daily decision making. Interviews with research
personnel who project both reserves and money supply acquaint
visitors with the current state of the projector's art, while the observation of daily meetings at which revisions to the reserve outlook
are presented drives home the uncertainties attending the conduct
of open market operations.
Adjunct Desk Responsibilities
Trading desk officers and staff members perform a number of
other duties. As mentioned above, they carry out securities transactions for customers, mostly for central banks but occasionally
for Treasury trust funds. Desk officers are in daily contact with
senior Treasury officials responsible for Treasury cash and debt
management. They also play a role in official surveillance of the
primary dealers in government securities. Finally, officers and staff
undertake studies relating to monetary policy and to a broad range
of other financial developments.
Purchase or sale orders are executed for foreign customers at the
customers' initiative. The Bank's Foreign Relations Department
keeps in contact with the more than 150 official foreign institutions
maintaining accounts with the Fed and compiles their requests for




174

transactions. Any orders that the desk does not choose to meet
using the System Account as counterparty must be either crossed
between foreign accounts, if buy and sell orders coincide, or
arranged in the market on a competitive basis. When the customer
orders are small relative to ordinary market transactions, the desk
will seek bids or offers from a handful of dealers. For large orders,
the desk will arrange a go-around of all dealers to get competitive
bids or offerings. Market entries to make purchases or sales for
customer accounts may either precede or follow the operations in
RPs or MSPs for reserve management purposes. (Since customer
RPs are used for reserve management, the desk does not arrange
them on the same day it does a System RP or MSP.) While the desk
does not publicly announce the magnitude of its own operations, it
does announce the approximate size of customer transactions
(including RPs) at the time of market entry. The bulk of customer
operations are in Treasury bills and short-dated coupon issues.
The desk will also purchase or sell non-Treasury instruments such
as bankers' acceptances and large certificates of deposit when
asked to do so. Most of the excess foreign cash is placed in the foreign RP pool, but at day's end, a limited amount of uninvested
funds that arrived too late to be arranged as RPs may be placed
overnight in the Federal funds market.
The Trading desk's relationship with the Treasury has many
facets. The traders on the desk provide information several times
each day on secondary market activity in government securities—
both through oral reports and market quotations. Desk personnel
also convey to the Treasury the dealers' comments on new offerings that investors might find attractive. Once the offerings have
been announced, the desk reports on the apparent extent of
investor interest. Traders monitor the market's estimates of rates
to be established at each auction of Treasury issues. Officers
supervise the opening of competitive tenders submitted in New
York, which usually account for 80 percent or more of the national
awards of new issues.
The desk's advisory role stems naturally from its daily involvement in the U.S. government securities market. The Manager is in
frequent contact with the Deputy Assistant Secretary of the
Treasury for Debt Management and the Fiscal Assistant Secretary
to discuss the Treasury's cash needs and its plans for meeting
them. Desk staff regularly inform the Treasury about foreign official interest in new issues. Once each quarter, Treasury officials
come to New York to obtain the suggestions from primary dealers




175




on structuring the midquarter financing and on meeting remaining
cash needs in the months ahead. In the following week, the
Manager and an associate typically attend the briefing sessions
held by the Treasury in Washington. There, the Treasury obtains
financing recommendations from a special advisory committee of
the Public Securities Association that includes representatives of
government securities dealers and other market participants.
Trading desk representatives participate in the Treasury's debt
management group discussions, which lead to the Treasury's final
decisions on what to offer.
Open market personnel have a special responsibility for relations with the primary dealers by virtue of the desk's role in the
market. Members of the Open Market and Dealer Surveillance
Departments of the New York Federal Reserve Bank work closely
together on matters relating to primary dealers. Surveillance procedures have been somewhat modified as a result of the Government Securities Act of 1986, which provided for regulation of what
had previously been an essentially unregulated market. Nonetheless, the desk still wants to be assured that the parties with which
it does business are financially sound and have the ability to carry
through on a commitment.
The Federal Reserve publishes a list of reporting primary dealers. The number of primary dealers grew from 38 in 1985 to 46 in
late 1988. Subsequently, however, some firms that did not find participation profitable withdrew. At the end of July 1989, the primary
dealer list included 42 dealers. Firms are usually added to the list
when they have satisfactorily demonstrated the adequacy of their
capital, the experience of their management and trading personnel, and the capability to remain market makers. They must also
show that they have achieved a significant volume of trading activity in government and federal agency securities with a broad range
of customers and other dealers.
A surveillance team conducts an on-site review of each firm's
policies, management controls, and reporting procedures before
the firm is accepted. The primary dealers and the aspiring dealers
report daily on their trading activities, their cash and futures positions in Treasury and other securities, and their means of financing
positions. Such reports supply open market and dealer surveillance personnel with up-to-date information on the functioning of
each dealer and the market as a whole. In addition, the firms provide monthly and annual profit reports, which enable the Federal
Reserve to keep close watch on their financial soundness. The

176

surveillance team schedules periodic visits to the primary dealers
to strengthen the New York Fed's understanding of a dealer's business approach, review management controls, and test adherence
to reporting procedures. Occasionally, senior officials from the
Open Market and Dealer Surveillance Departments meet with the
principal officers of a dealer firm to discuss the dealer's recent performance and review any outstanding issues. The Federal Reserve
admits a firm on the primary dealer list to a trading relationship
after determining that it has a good track record and would help
the desk perform its own functions. Generally most of the primary
dealers have a trading relationship with the desk.
Finally, open market personnel engage in a wide variety of
reporting and analytical assignments. Reporting to the FOMC itself
is one important task. Beyond the routine reports, described
above, special studies are undertaken. They may involve proposed
modifications in the Committee's approach to reserve management or to the choice of intermediate targets. Technical matters
involving reserve management, such as the behavior of excess
reserves or the relationship between discount window borrowing
and interest rates, may be analyzed. Open market staff may also
examine issues of Treasury financing and the Treasury's tax and
loan accounts to find ways to reduce their disruptive effects of
Treasury cash flows on reserves. Market developments, such as
the emergence of markets for derivative financial instruments, are
of particular interest as well.




177




8
Responses to
Federal Reserve Policy
Monetary policy has a wide range of direct and indirect effects on
economic activity, prices, and interest rates. Although there is considerable agreement among economists about the channels through
which monetary policy impulses are transmitted to the economy, the
relative importance of each of these channels remains controversial.
Most economists believe that monetary policy influences economic
activity and prices by affecting the availability and cost of money and
credit to producers and consumers. People make spending and
investment decisions based upon current and expected wealth,
income, and prices, all of which are influenced by past, current, and
expected future monetary policy actions. Interest rates respond to the
current and prospective economic climate and to monetary policy.
Monetary policy actions in turn enter into decisions about consumption, savings, and investment. The effects of policy will both influence
and depend upon the underlying economic environment, including
perceptions about federal government expenditures and receipts, the
strength of credit demands and supplies in the United States and
abroad, and the outlook for the exchange rate of the dollar.
The four major sections of this chapter explore some of the ways
in which domestic monetary policy can affect the U.S. economy.
The first section examines how economists over the last 50 years
have interpreted the transmission of monetary policy. Considerable
understanding has been gained from the experience and analysis of
recent decades; most analysts currently believe that monetary policy influences prices and incomes by affecting both interest rates
and wealth. Nonetheless, the significance of particular channels of
transmission is still debated, and questions persist about the way
people respond to new information. No single, comprehensive,
widely accepted view has emerged of the role of monetary policy in
determining economic activity and price behavior.7
The second section considers the impact of policy on the cyclical
behavior of the interest rate maturity structure, commonly called
the yield curve. The third section examines how the desk's procedures may influence interpretations of policy and interest rate determination. The potential effect of policy on the individual sectors of
the economy is the subject of the fourth section; the discussion
demonstrates that the sectors' different compositions and constraints shape their responses to interest rates and income. The final
section discusses the communication of policy through the activi1 Stanley Fischer, "Recent Developments in Macroeconomics," The Economic Journal, vol.
98, no. 391 (June 1988), pp. 294-339, provides an overview of the literature on a number
of policy issues.

178

ties of "Fed watchers"—economists who forecast and interpret economic behavior, interest rates, and Federal Reserve actions.
Evolving Views of Policy Transmission
Economists' understanding of the transmission of monetary policy to the financial markets and to the economy has been modified
extensively during the past several decades. One school of thought
begins with J. M. Keynes' General Theory, written in 1935. Influenced by the Great Depression, Keynes had argued that monetary
policy had limited power to promote economic expansion; once
interest rates fell close to zero, as they did during the depression, no
further declines to stimulate investment would be possible. Wages
were believed to be sticky and particularly resistant to falling, even
in the face of high unemployment. Although A. C. Pigou, a member
of the earlier classical school, objected that Keynes' arguments
depended upon an implausible failure of workers to respond to
changes in their purchasing power, the predominant belief in academic and policy making circles during the 1940s was that monetary
policy, in practice, had little power to promote economic expansion.
In the postwar period, however, economic expansion and inflation
were the dominant conditions. Interest rates were no longer close to
zero and "Keynesian" economists modified Keynes' models. They
came to expect that monetary policy would work primarily by affecting interest rates, which in turn would affect investment.2 Economists
who continued to follow the prewar classical tradition were more
likely to focus on long-run equilibrium conditions and generally
believed that over time money would determine the price level.
Monetary policy was frequently analyzed in the context of business
cycles, both by Keynesians and by economists who were extending
and modifying the classical analysis. Some economists suggested
that business cycles resulted, at least in part, from the policy process
itself.3 Their arguments ran as follows: Monetary authorities would
respond to weakness in the economy with an easier monetary policy
that would lower interest rates by making loanable funds more plentiful. (Deliberate fiscal stimulus might make a difference in long con2 Fora further description of the evolution ofKeynesian thought in the 1950s and 1960s, see
Paul A. Samuelson, "Money, Interest Rates, and Economic Activity: Their Interrelationship
in a Market Economy," Proceedings of a Symposium on Money, Interest Rates, and
Economic Activity, American Bankers Association, New York, 1967. Reprinted in Robert
C Merton, ed., The Collected Scientific Papers of Paul A. Samuelson, vol. 3 (Cambridge:
MIT Press, 1972), pp. 550-70.
3 Gottfried Haberler, Prosperity and Depression (London: George Allen and Unwin, 1964).




179




tractions, but lags in recognizing the recession and in changing policy
would preclude its use in countering short recessions.) Lower rates
and increased credit availability would encourage more expenditures
on investment and consumption, which in turn would support economic expansion as long as there was excess capacity to absorb the
increased demand. When capacity constraints were felt, increased
demand would merely lift prices, producing "demand pull" inflation.
Interest rates would begin to rise, and investment would be choked off
as credit became more expensive. If ceilings were then placed on the
interest rates that banks could pay or charge, or if banks became
reluctant to make loans that they perceived to be risky, credit availability would be reduced. The "credit crunch" would constrain economic activity, bringing the business expansion to its close.
During the 1950s and 1960s, economists struggled to explain inflation that arose when the economy seemingly was not at full employment. It was generally described as a "cost push" phenomenon and
was attributed to structural distortions in the labor markets rather than
to aggregate monetary or fiscal policies. A related hypothesis was
developed by A. W. Phillips.4 He observed that increasing levels of
employment seemed to be associated with rising nominal wage rates.
Graphical relationships between the unemployment rate and changes
in wages or prices came to be referred to as Phillips curves.
These economic models relied on nominal wage rigidities, which
meant that the prevailing descriptions of the policy transmission
process could at best hold only in the short run. At some stage,
people would change their behavior as prices and wages were
observed to rise (or fall) persistently. In economic parlance, people
would not suffer permanent money illusion but would, in time, recognize the decline in their wages defined in "real" terms—adjusted
to take account of price changes. In particular, as prices rose,
workers would demand higher wages for a given amount of work
to offset the loss of purchasing power, or alternatively would work
less for the same wages. The Phillips curve would not represent a
long-run equilibrium. Once inflation expectations rose, the curve
would shift outward.5
4 "The Relation Between Unemployment and the Rate of Change of Money Wage Rates in
the United Kingdom, 1861-1957," Economica, vol. 25, no. 100 (November 1958), pp.
283-99.
5 Milton Friedman, "The Role of Monetary Policy," Presidential Address at the 80th Annual
Meeting of the American Economic Association, December 1967, reprinted in American
Economic Review, vol. 58, no. 1 (March 1968), pp. 1-17; and Edmund S. Phelps, "Money
Wage Dynamics and Labor Market Equilibrium," Journal of Political Economy, vol. 76,
no. 4 (July-August 1968), pp. 687-711.

180

During these years, Milton Friedman and several other economists argued that the standard view of the monetary transmission
process gave too little emphasis to the role of money balances.
Consumers who found themselves with larger amounts of money
when an increase in the aggregate quantity worked its way through
the economy would feel wealthier and increase their spending.
Sellers would respond to the increased purchases by ordering more
goods and by raising prices; the increased total quantity of money
available for purchases would sustain the higher prices.
The theory did not define how the increase in aggregate demand
stimulated by the higher money balances would be divided
between output and prices. Drawing on the classical model, the
economists who emphasized the role of the quantity of money
argued that increasing money balances would not affect output
indefinitely but in the end would only lift prices. The short-run
dynamics of a response to a series of increases in money balances,
however, would affect both economic activity and prices. The pattern followed would depend upon how the increases affected
expectations and how quickly people adjusted their behavior when
their expectations changed.
When inflation was expected to pick up, nominal interest rates
—those observed directly—would increase. If the change were
merely offsetting the expected decline in purchasing power over the
term of the loan, the increase in nominal rates would not raise the
inflation-adjusted or real cost of borrowing or the return to lending.
Consequently, using nominal rates as a gauge of the tightening or
easing of policy could prove misleading and could induce perverse
results. For instance, the monetary authorities might think they
were providing for a steady cost of credit by holding interest rates
constant, but if the rate of inflation was expected to accelerate, they
would really be fostering easier money and credit conditions.6
It became increasingly important to deal with distortions in nominal
interest rates as inflation gained force in the late 1960s and 1970s.
Economists made use of the concept, introduced years earlier by
Irving Fisher,7 of real interest rates—approximated as the nominal
rate less the expected rate of inflation. This concept held that borrowers and lenders made their decisions based on expected real rates.
These rates, however, could not be observed but had to be derived
from expectations of inflation. Fisher suggested that the real rate was
6 Friedman, op. cit.
7 The Theory of Interest (New York: MacMillan, 1930).




181




related to underlying economic conditions as they affected investment and savings opportunities and therefore might be reasonably
constant. If that were the case, changes in nominal interest rates
could be used as an indication of changes in inflationary expectations. Using that hypothesis, Fisher and subsequent observers found
that interest rates appeared to adjust to changes in inflation with long
lags, possibly because it took time for borrowers and lenders to realize
that a change in the rate of inflation would be sustained.
Fisher's hypothesis is difficult to test because ex ante real rates
cannot be measured directly. Ex post rates may or may not be a good
proxy. If real rates are computed ex post by subtracting actual inflation from nominal rates—even averaged over fairly long periods—we
find that real rates have been far from constant (see Chart 1, page
183). Ex post real returns have often been low and occasionally negative during periods of accelerating inflation such as the 1970s; they
were well above levels of the previous decades in the early part of the
1980s when inflation was slowing.8 The apparent variability in real
rates may mean either that expectations of inflation are subject to
substantial errors or that ex ante, real rates are not constant.9 Real
rates may well change as part of the price adjustment process.
With the distinction between real and nominal measures attracting
increased attention, analysts considered how to measure inflationary
expectations. Most work done in the 1960s estimated expectations of
future inflation by extrapolating from past inflation. During the 1970s
economists argued that this "adaptive" expectations approach was
incomplete. They suggested that borrowers and lenders would make
use of all of the relevant information available, not just that on past
inflation. In particular, they would consider current monetary and fiscal policies and likely future actions by the policymakers. The
approach has been termed rational expectations.70
8 A. Steven Holland, "Real Interest Rates: What Accounts for Their Recent Rise?" Federal
Reserve Bank of St. Louis Review, December 1984, pp. 18-29.
9 A number of different techniques have been used to estimate the ex ante real rate of interest
For instance, see Frederick Mishkin, "The Real Interest Rate: An Empirical Investigation, "The
Cost and Consequences of Inflation, Carnegie-Rochester Conference Series on Public Policy,
vol. 15 (Autumn 1981), pp.151-200; and Charles R. Nelson and William G. Schwert, "ShortTerm Interest Rates as Predictors of Inflation: On Testing the Hypothesis That the Real Rate of
Interest is Constant," American Economic Review, vol. 67, no. 3 (June 1977), pp. 478-86.
10 The concept was introduced in John F. Muth, "Rational Expectations and the Theory of
Price Movements," Econometrica, vol. 29, no. 3 (July 1961), pp. 315-35. Rational expectations analysis was applied to monetary questions by Robert E. Lucas Jr. in "Expectations
and the Neutrality of Money," Journal of Economic Theory, vol. 4, no. 2 (April 1972), pp.
103-24; and by Thomas J. Sargent and Neil Wallace in "'Rational' Expectations, the
Optimal Monetary Instrument, and the Optimal Money Supply Rule," Journal of Political
Economy, vol. 83, no. 2 (April 1975), pp. 241-54.

182

Chart 1 Ex Post Real Rate of Return: One-Year Horizon
Interest rate

CPI-percent change

lt=Nominal one-year
Treasury yield at
start of period
Pt=Percent change in
consumer price index
over previous four quarters

1960 '62 '64 '66 '68 70

72

74

76

78

'80 '82 '84 '86 '88

Real interest rate

Rt=Ex post real
rate of return
(Rt=lt-Pt)

1960 '62 '64 '66 '68 70

72

74

76

78

'80 '82 '84 '86 '88

The basic rational expectations concept has attracted a wide audience. Its proponents have argued that people will incorporate all of
the predictable consequences of available information about monetary policy into their decision making, including the future effects of
changes in policy priorities. Hence, if the Federal Reserve enacted a
change in policy procedures", people would alter the way they
11 The observation that a change in policy procedures will change the structure of the transmission mechanism was made in Robert E. Lucas Jr., "Econometric Policy Evaluation: A
Critique," in Karl Brunnerand Allan H. Meltzer, eds., The Phillips Curve and Labor Markets,
Carnegie-Rochester Conference Series on Public Policy, vol. 1 (Amsterdam: North-Holland,
1976), pp. 19-46.
Carl Walsh applied the hypothesis of the "Lucas critique" to the monetary policy transmission mechanism in "The Effects of Alternative Operating Procedures on Economic and
Financial Relationships,"Monetary Policy Issues in the 1980s, a symposium sponsored
by the Federal Reserve Bank of Kansas City, August 1982, pp. 133-63.




183




responded to observed monetary variables in keeping with their
understanding of the revised operating guidelines.
Some writers, particularly in the academic community, have followed this line of analysis to the conclusion that people will adjust
their behavior to prevent any anticipated monetary policy actions
from having an effect on the real economy. Thus, it is argued, only
when a policy action is a surprise can it affect real interest rates and
economic activity. This version of rational expectations theory figures importantly in real business cycle theory, which holds that
business cycles can be caused by exogenous shocks to the economy or by unexpected monetary policy developments but not by
predictable monetary policy actions.72
Many economists, however, have rejected the notion that anticipated monetary policy does not affect real economic activity in
the short run, partly because the argument does not seem to be
supported by empirical work. Even within the rational expectations
framework, a case can be made that predictable monetary policy
can affect real activity because of the substantial costs of acquiring
and interpreting all of the potentially available information about
past, current, and future monetary policy. In addition, although
money data are available with short lags, their significance will not
always be clear, particularly when the demand for money seems to
be shifting. Furthermore, there may be institutional rigidities, such
as long-term contracts, that are not indexed for inflation. Rigidities
will slow down the responses to a policy impulse, permitting a
more conventional adjustment process to occur. Hence, most
economists, while recognizing some difficulties with the standard
models, still view business cycles in more or less traditional terms
as arising from a combination of real and monetary forces.73
As a group, Federal Reserve policymakers continue to regard
monetary policy as a powerful tool, although different individuals
will place different emphases on the various transmission routes
12 Michael Dotsey and Robert G. King, "Rational Expectations Business Cycle Models: A
Survey," Federal Reserve Bank of Richmond Economic Review, vol. 74, no. 2 (MarchApril 1988), pp. 3-15.
13 For more detailed discussions of the business cycle, see the essays in Robert J. Gordon,
ed., The American Business Cycle (Chicago: University of Chicago Press, 1986). In "The
Mechanisms of the Business Cycle in the Postwar Era," pp. 39-122, Otto Eckstein and
Allen Sinai identify five stages to the business cycle: "(1) recovery/expansion; (2) boom;
(3) precrunch period/credit crunch; (4) recession/decline; and (5) reliqueftcation."
In the same volume, Olivier J. Blanchard and Mark W. Watson question whether business
fluctuations are sufficiently similar to give analytical value to the concept of a business
cycle ("Are Business Cycles All Alike?" pp. 123-79).

184

and on the appropriate role of monetary policy in the short-run stabilization of economic activity. Some policymakers believe
response lags are sufficiently short to allow policy actions to be
used for short-run stabilization. Others are concerned that long lags
that are not very predictable will keep monetary policy from being
a good tool for short-run stabilization. Policymakers generally
agree, however, that long-term relationships exist between price
and money behavior and that these relationships must be kept in
mind if price stability is to be achieved.
Monetary Policy and Yield Curves
Monetary policy works most directly by changing reserve availability. Such changes affect the overnight Federal funds rate and
other short-term rates which in turn affect the monetary aggregates.74 How short-term rates affect longer term rates has long been
a source of controversy among both academic and market analysts.
Perhaps the most commonly accepted view is that the pattern of a
default-free yield curve (such as for Treasury debt) reflects both
expectations of future short-term rates and preferences for liquidity
that lead investors to favor shorter term maturities to varying
degrees. (Liquidity in this context means the ability to turn an asset
quickly into cash without significant loss of nominal value. Hence,
short-term instruments with their smaller price fluctuations have
greater liquidity even though there is an active market for long-term
instruments.) This view of yield curve determination is described in
the literature as the liquidity-augmented expectations hypothesis.75
Expectations of future interest rates influence the shape of the
yield curve as long as potential investors and borrowers have
choices about the maturity of the instruments they purchase or
14 Because the Federal Reserve alters reserves through purchases and sales of Treasury securities, it changes the amount of securities held by the public, a development which also affects
rates. However, Federal Reserve purchases or sales usually represent such a small part of the
total Treasury market that the rate impact, separate from the effect on reserves, is very small.
15 Edwin J. Elton and Martin J. Gruber, in Modern Portfolio Theory and Investment Analysis
(New York: John Wiley and Sons, 1987), pp. 458-67, offer a helpful review of common
hypotheses of the yield curve. Fora review of the theoretical and empirical literature on the
yield curve, see Robert J. Shillerand J. Houston McCulloch, 'The Term Structure of Interest
Rates," National Bureau of Economic Research, Working Paper no. 2341, August 1987.
The major alternative hypothesis of yield curve determination is that markets for securities
of different maturities are segmented. Parties that normally borrow or lend in a particular
maturity range are assumed to be very reluctant to alter their actions in response to changes
in the shape of the yield curve. Therefore, there need be no relationship between yields on
different maturity instruments. While such segmentation may apply in limited circumstances, it breaks down ifparties on the margin are willing and able to change the maturities
at which they borrow or lend in response to changes in actual or expected rate relationships.




185




issue. For instance, potential investors can compare their expected
returns from buying either a long-term security or a succession of
short-term securities. They will buy the longer term security if their
expectations about the course of interest rates over the security's
lifetime support their view that the longer term instrument is more
attractive. Investors will continue to switch to longer term issues until
those rates fall relative to shorter rates by enough to remove the
expected rate advantage of the longer term issues. Investors form
their expectations on the basis of the outlook for inflation and for real
interest rates, which will, in turn, be influenced by expectations about
economic activity, monetary policy, and fiscal policy.76
If preferences for liquidity are imposed on the expectations
model, interest rates will rise as the term to maturity lengthens
when expectations of steady rates would otherwise call for a flat
yield curve. The "normal" upward slope provides investors with a
higher expected yield on longer term obligations in return for giving
up the greater liquidity of shorter term instruments. The longer the
time before maturity, the greater will be the change in price that will
accompany a given change in interest rates and the greater the
likelihood that rates will change substantially during the security's
remaining life. The degree of liquidity preference may shift with
changing perceptions of the potential for rate variability.77
The yield curve for Treasury securities serves as the reference
standard. Such securities are free of the credit risk attached to
other securities, generally are not subject to early redemption, and
enjoy a broad and active secondary market. Debt instruments
issued by the federally sponsored credit agencies and by private
business typically have higher yields for corresponding maturities.
The sizes of the rate differentials between Treasury securities and
16 Since the slope of the yield curve embodies the market expectations of future interest rates,
it can be used as a summary statistic of the market participants' predictions of future interest rate changes. Examination of interest rates out to six months has suggested that, historically, the predictive power of the slope of the yield curve appears to have varied
inversely with the degree to which the Federal Reserve smoothes interest rate fluctuations. See Gregory N. Mankiw and Jeffrey A. Miron, "The Changing Behavior of the Term
Structure of Interest Rates, "Quarterly Journal of Economics, May 1986, pp. 211-28, for an
analysis of the period 1890-1979. See Gikas A. Hardouvelis, "The Predictive Power of the
Term Structure during Recent Monetary Regimes," Journal of Finance, June 1988, pp.
339-56, for an analysis of the 1970s and 1980s.
17 For a different perspective on the slope of the yield curve, see John H. Wood, "Do Yield
Curves Normally Slope Up? The Term Structure of Interest Rates, 1862-1982," Federal
Reserve Bank of Chicago Economic Perspectives, July-August 1983, pp. 17-23. Wood's
examination of 19th-century data leads him to question the widely held view that an
upward slope to the yield curve is normal. He suggests that it arose in the 20th century
because the change in monetary standard introduced an inflationary bias.

186

Chart 2 Yield Curves over a "Normal" Business Cycle
Interest rate

Time to maturity

other debt instruments of comparable maturity reflect the market's
allowance for lesser marketability and the risk that the borrower
may default, as well as differences in tax treatment and in other
characteristics such as the potential for early redemption.
When business cycles follow what is considered the traditional pattern, the shape of the yield curve may do so as well, because the traditional business cycle is accompanied by cycles in inflation and
credit demands. A "normal" upward-sloping yield curve would
emerge when a major acceleration or deceleration in inflation was not
expected (Chart 2, Figure A). Toward the latter part of a recession,
the yield curve would be expected to slope upward more sharply than
"normal" (Chart 2, Figure B); short-term rates would fall more than
long-term rates, since the recession would not be expected to persist
throughout the life of the long-term debt instruments.
As a recovery got underway, the yield curve would initially remain
steep. At some stage it would be likely to flatten somewhat as shortterm rates rose, but it would generally remain upward sloping (Chart
2, Figure C). If significant inflationary pressures became evident, however, the Federal funds rate and other short-term rates would be likely
to rise substantially, prompted by tighter monetary policy and changes
in the market's expectations about inflation. The yield curve would
then tend to shift upward and flatten (Chart 2, Figure D). An inverted
or downward-sloping yield curve (Chart 2, Figure E) would reflect




187

market views that short-term interest rates were already high enough
to reduce real GNP growth and inflation and that a slowing in economic
activity and inflation should lead to a decline in interest rates.

Responses to Operating Procedures
Federal Reserve actions generally have had a strong direct
impact on short-term rates. When the Federal funds rate was the
primary target of desk operations during the 1970s, the Federal
Reserve controlled that rate within a narrow range. Other shortterm interest rates were influenced by the current and expected
behavior of the funds rate. Decisions about the management of
money and credit were influenced by estimates of how rapidly the
monetary authorities would raise or lower the funds rate; the judgment often depended as much on assessing policymakers' willingness to allow interest rates to change as on observing the behavior
of money growth and the economy. Most forecasting focused on
the factors believed to affect the timing and magnitude of funds
rate changes. Changes in the direction of the funds rate were relatively infrequent and generally only occurred after considerable
evidence of a turnaround in economic activity had accumulated.
Since the funds rate was targeted directly, changes in the desired
rate were easy to recognize once they had taken place.
The switch to nonborrowed reserve targeting in 1979 changed
the ways that the banks and the financial markets interpreted
Federal Reserve actions. Observers gradually incorporated expectations that significant deviations of money from its target would
bring immediate countervailing pressures on reserves and hence
on interest rates. Market participants had to shift their emphasis
when making interest rate forecasts. The Federal funds rate was
less reliable as a day-to-day or week-to-week guide to the thrust of
policy.78 Analysts had to forecast near-term money behavior in
order to project the timing of interest rate changes, since money
growth significantly affected the degree of pressure put on reserve
positions and hence on short-term interest rates. Interest rates
became very sensitive to current policy and to the perceived
chances of subduing inflation.79 Analysts looked at the economy's
performance to assess likely money demand as well as credit




18 Dana Johnson, "Interest Rate Variability under the New Operating Procedures and the
Initial Response in Financial Markets," New Monetary Control Procedures, vol. 1, Board
of Governors of the Federal Reserve System, February 1981.
19 Financial innovation, encouraged by rapid inflation and efforts to avoid restrictive interest
rate ceilings, also affected the demand for money.

188

requirements, resource deployment, and inflationary pressures.
Economists devoted considerable effort to predicting weekly
movements in Ml; they made surveys of large depository institutions, gathering information on Ml deposits, and became experts
in anomalies in the seasonal factors.
The procedures adopted early in 1983 focused most directly on
borrowed reserves. Consequently, the relationships between policy
actions and market response shifted once more. The amount of
discount window borrowing being sought by the Federal Reserve
became less variable; the objective no longer depended primarily
on movements in Ml but was changed only when a group of indicators suggested that a change was needed. For a given discount
rate and set of rules of access to the discount window, the amount
of borrowing served as a key determinant of the Federal funds rate.
Other factors influencing the funds rate included seasonal pressures, concerns about the banking system, and expectations about
Federal Reserve policy and its likely effects on future rates.20
The relatively limited moves in the objective for borrowed
reserves made the Federal funds rate considerably less likely to
change significantly from week to week than during the 1979-82
period. Short-term fluctuations in the funds rate, nonetheless, were
somewhat greater than in the 1970s when the rate had been targeted directly. In observing the behavior of the funds rate, market
participants had to decide whether a move in the rate occurred for
policy or for other reasons. Since the Federal Reserve was basing
its borrowed reserve objectives on a variety of indicators, including
the outlook for economic activity and inflation as well as the
behavior of the monetary aggregates, participants forecast those
variables in order to estimate the next move in the policy stance.
Under any of the mechanisms used to determine short-term
rates, Federal Reserve policy actions affected longer term rates
through their impact on expectations of money growth, the outlook
for economic activity, inflation, and at times, the exchange value of
the dollar. In making decisions affecting longer term interest rates,
market participants weighed the appropriateness of policy actions
in relation to their own forecasts of inflation and economic activity.
If policymakers seemed reluctant to allow short-term rates to rise
20 For a review of the borrowed reserves operating procedures and a comparison with the
earlier nonborrowed reserves procedures, see Henry C. Wallich, "Recent Techniques of
Monetary Policy," Federal Reserve Bank of Kansas City Economic Review, May 1984, pp.
21-30; and Brian F Madigan and Warren T. Trepeta, "Implementation of U.S. Monetary
Policy," in Changes in Money Market Instruments and Procedures: Objectives and
Implications, Bank for International Settlements, March 1986.




189

sufficiently in periods of economic growth to restrain money and
credit growth, at some stage longer term rates would increase to
allow for the possibility of higher inflation. During the 1979-82 period, longer term rates showed much more short-run volatility than
they had previously, although the Fed's policy procedures were only
one source of this volatility. Participants were not sure whether the
Federal Reserve would ultimately succeed in halting inflation or
whether its efforts would be overwhelmed, allowing another burst of
price acceleration. Much depended upon how the economy adapted to the adjustment pressures. The uncertainty made long-term
rates unusually sensitive to short-run developments.27
Policy's Effect on the Economic Sectors
Monetary policy influences spending and investing decisions in
all sectors of the economy. The various economic sectors will
respond in different ways to changes in income and interest rates
depending on the extent to which they are borrowers or lenders.22
Over the 1982-88 period, the Federal Reserve's flow of funds
statistics show that the net funds raised annually by the nonfinancial sectors were equivalent to about one-sixth of GNP. Households
took up about one-third of that flow and business another third. The
U.S. government has accounted for nearly one-quarter of the total.
State and local governments accounted for the remainder. Of the
funds advanced, the household sector supplied a little less than
half and the business sector around 10 percent. Close to threequarters of household funds were placed with depository institutions or money market funds, while the remainder was invested
directly in credit market instruments.
Because of the large current account deficits that the United
States amassed in recent years, foreigners have also lent large
sums to U.S. residents—almost 5 percent of total funds advanced
over the 1982-88 period. When real interest rates in the United
States were perceived to be relatively high through the middle of
the decade, private foreigners were willing to provide the financing.
When real rates did not seem attractive, particularly when the dollar was falling, foreign central banks often intervened in the ex-




21 For a discussion of the relation of long-term rates to policy, see Gikas A. Hardouvelis,
"Market Perceptions of Federal Reserve Policy and the Weekly Monetary
Announcements," Journal of Monetary Economics, September 1984, pp. 225-40.
22 For more discussion of the interest sensitivity of household and business spending, see M.
A. Akhtar and Ethan S. Harris, "Monetary Policy Influence on the Economy—An
Empirical Analysis," Federal Reserve Bank of New York Quarterly Review, vol. 11, no. 4
(Winter 1986-87), pp. 19-34.

190

change markets. They purchased dollars, which they then invested
in the United States, largely in Treasury securities.
1. The Household Sector
Monetary policy's cumulative impact on the household sector
can be substantial. Policy can influence household spending
through numerous channels:
• income and employment
• wealth
• division of income between saving and consumption
• expectations of inflation
• cost and availability of credit.
To the extent that monetary policy affects the level of overall
business activity, it affects households' incomes and employment.
These factors, in turn, have strong effects on consumer spending.
Changes in interest rates are likely to influence household spending by affecting wealth and income and by shifting the relative
returns to future savings and investment. The impact of interest
rate changes on different households has been altered by the
expanded use of variable rate loans and deposits.23 Households will
respond to changes in interest rates according to the forms of
wealth they hold, the types of debt they have incurred, and their
propensities to save the changes in income brought about by
changes in rates. Households as a group are net savers, so their
incomes should be increased by a rise in rates. The net stimulus to
spending from higher rates will be partially offset, however, by
declines in existing wealth held in the form of fixed-rate investments, since their prices fall when rates rise. Furthermore, if rates
rise, those households with variable rate loans will have to make
higher payments on existing loans; these obligations will take away
resources that might otherwise be directed to consumption.
Household responses to interest rate changes will also depend
upon how those changes are interpreted. For instance, if consumers
regard a sharp rise in nominal interest rates as a sign that greater
economic uncertainty and rising unemployment lie ahead, they will
tend to save more and borrow less. If at the same time, however,
23 For more detail, see Changes in Money-Market Instruments and Procedures: Objectives
and Implications, Bank for International Settlements, March 1986; Financial Innovation
and Monetary Policy, Bank for International Settlements, March 1984; Thomas D. Simpson,
"Developments in the U.S. Financial System since the Mid-1970s," Federal Reserve
Bulletin, vol.74, no.l (January 1988), pp. 1-13; and John Wenninger, "Financial
Innovation—A Complex Problem Even in a Simple Framework," Quarterly Review,
Federal Reserve Bank of New York, vol. 9, no. 2 (Summer 1984), pp. 1-8.




191




inflationary expectations shift upward such that real interest rates
fall, consumers may reduce saving and increase their borrowing in
order to step up current spending before prices increase further. In
recessions, a sharp fall in interest rates could lower borrowing costs,
a development that might temper the decline in consumer spending
that results from the reductions in income.
Household investment in housing can be sensitive to the cost
and availability of credit. However, the removal of ceiling rates on
loans and consumer deposits over the last decade has made the
adjustments less sudden than they were in the past when interest
rate ceilings on loans and deposits became binding. Currently, if
rising interest rates outstrip the effects of rising incomes from an
expanding economy, some prospective home buyers may not
make purchases either because they are unwilling to take on such
a large debt burden or because lenders are reluctant to extend the
needed credit. Unless consumers are convinced that future home
prices will provide a good hedge against inflation, they will curtail
their purchases as mortgage rates rise. Rising mortgage interest
rates also reduce the attractiveness of refinancing existing homes
to provide additional resources for spending. In addition, the higher rates should discourage the use of home equity loans, which are
often on a floating-rate basis. Households may feel less wealthy
when homes become harder to sell or the prices of existing houses
actually decline. When interest rates fall as a recession progresses, the activation of deferred housing demand can contribute to
economic recovery.
Monetary policy also affects the use of consumer credit such as
installment and credit card loans. Consumers are heavy users of
such credit, especially for purchases of automobiles and other
durable consumer goods. Historically, use of such credit has not
been very sensitive to the level of interest rates charged. And during the 1980s, the automobile manufacturers, by providing direct
financing to consumers at attractive below-market rates, may have
lessened the impact of high market rates. Changes made to the tax
law in 1986 are gradually eliminating the deductibility of interest
on all personal loans, other than those secured by owner-occupied
housing, on a five-year timetable. Consequently, consumers are
likely to rely increasingly on home equity loans and mortgage refinancings as sources of credit.24 Because most home equity loans
and many mortgages are made at variable rates, consumers feel
24 Glenn B. Canner, James T. Fergus, and Charles A. Luckett, "Home Equity Lines of Credit,"
Federal Reserve Bulletin, vol. 74, no. 6 (June 1988), pp. 361-73.

192

the effects of higher interest rates not only on the debt incurred for
new purchases but also on their stock of existing debt.
2. The Business Sector
Monetary policy affects business decisions concerning production levels, inventories, and new investments. Businesses produce
most of the goods and services consumed by the other sectors. To
be successful, they must anticipate and respond to the demands of
consumers, other businesses, governmental units, and foreign
buyers. These demands will be influenced by monetary policy as
it responds to changes in general business conditions.
Businesses depend heavily on the credit markets to finance the
inventories and productive capacity needed to meet customer
demands.25 The Federal Reserve's flow of funds statistics show that
during the 1980s, corporate cash flow from retained earnings and
depreciation allowances provided about three-fourths of the funds
used for capital spending, the extension of trade credit, and the acquisition of other financial assets. The remainder came chiefly from borrowing in the credit markets. Moreover, business credit demand
typically grows more rapidly than the economy once the expansion
phase of the business cycle is well under way. If business managers
expect strong sales, they try to keep inventories growing ahead of
demand so that sales are not lost because of shortages. Similarly,
expectations of strong demand lend greater urgency to plans for
increasing capacity that may require several years to accomplish.
When the economy approaches capacity, short-term interest
rates rise and business managers have to weigh the increasing cost
of financing inventories against the possible sales gain from having
ample supplies. Price expectations also enter their calculations. If
inflation is widely expected to accelerate, managers are likely to
increase purchases of raw material and other inputs to raise
planned levels of finished inventories. Expectations can then
become self-fulfilling. If monetary policy allows short-term interest
rates to rise rapidly during expansions, making inventory building
a costly strategy, managers will respond by keeping inventories
under tight control. Otherwise, there could be speculative inventory building. At the same time, rising rates reduce the buoyancy of
household demand for business output through the channels
already noted. As a result, business profitability begins to suffer
25 In "Relative Prices and Inventory Investment," Journal of Monetary Economics, vol. 4,
no. 1 (January 1978), pp. 85-102, John Bryant discusses the effects of financing costs on
inventories and investment.




193

both from high borrowing costs and a softening of demand. At
some point, inventories become heavy in relation to sales, requiring a reduction in current production to bring them back in line.26
Managers, encountering sustained pressure on profit margins,
step up their efforts to cut costs. If their businesses are heavy credit
users, a review of capital spending may prompt them to trim present plans or to defer completion of particular projects—especially
when lenders are reluctant to provide additional financing. The rise
in long-term rates may itself reduce the attractiveness of projects
under consideration by increasing the rate at which projected
income streams are discounted. Often the spreads widen between
yields on bonds of well-capitalized firms that are given investmentgrade ratings and yields on lower rated bonds. The increased
spreads may force potential borrowers contemplating risky uses of
the funds to turn to less conventional financing; higher financing
costs may also crowd some borrowers out of the credit markets
altogether.27 Aggregate capital spending tends to be sustained well
into a recession, diminishing only as the rising margin of available
capacity reduces the desirability of further additions.
During the 1980s, a number of techniques were developed to
allow corporations to hedge some of the interest rate risk involved
in their operations. Corporations employing these techniques can
reduce their sensitivity to changes in market rates. For example,
they can make use of financial futures and other derivative products, described in Chapter 4, to lock in a specified interest rate cost
or to cap future increases. They can use swaps to change a variable-rate commitment into a fixed-rate one or to do the reverse,
depending upon their expected pattern of revenue streams.
Although these devices cannot be used without cost and cannot be
expected to insulate a firm from all effects of interest rate changes,
they can reduce costs of rate changes for those who use them.
3. State and Local Governments
Most units of government below the federal level operate essentially by balancing current spending with receipts from taxes and grantsin-aid from a higher level of government. Since persistent deficits are




26 The relationships among monetary policy actions, interest rates, and inventories are discussed in Alan S. Blinder and Stanley Fischer, "Inventories, Rational Expectations and the
Business Cycle," Journal of Monetary Economics, vol. 8, no. 3 (November 1981), pp. 277304; and M.A. Akhtar, "Effects of Interest Rates and Inventory Investment in the United
States," American Economic Review, vol. 73, no. 3 (June 1983), pp. 319-28.
27 The volume of "junk" bonds, or bonds below investment grade, has risen dramatically
since the 1982 recession. The response of those new bonds to a recession is untested.

194

not permitted, state and local spending is heavily dependent on the
current condition of the economy. As the economy expands, revenues increase, encouraging new spending initiatives. Conversely,
disappointing revenues in times of recession often lead state and local
governments to scale down their capital expenditures fairly quickly
because of the need for balancing income and expenditures.28 The
credit market effects of monetary policy on governmental units work
chiefly through capital spending, but total capital outlays have recently been around 13 to 14 percent of state and local government expenditures, so that reducing these outlays when rates rise will affect only
a modest portion of total spending. Major capital projects that depend
heavily on bond financing include building and repairs of schools,
roads, water systems, sewers, and transportation systems. A general
rise in interest rates increases the rates that governmental units have
to pay on their bonds. At the margin, rate increases may lead to some
reduction in, or postponement of, capital spending programs. Some
issuers may be unable to borrow because rates rise above ceilings
established by state law on what they can pay.
General obligation debt of governmental issuers is secured by the
taxes that can be levied. For most purchasers, their interest is
exempt from federal income taxation. The current expense budget
of the issuing body usually provides for interest and authorization
charges on such debt. The states have also established separate corporations to run enterprises, which charge for the services they render. Their services include financing mortgage lending and building
and maintaining bridges, turnpikes, utilities, and hospitals. The capital spending of such corporations is usually financed by revenue
bonds, that is, bonds secured by the stream of revenues expected
from the facilities they finance. These bonds will be tax-exempt if the
proceeds are used for public purposes. Private purpose financing is,
with a few limited exceptions, no longer exempt from federal taxes.29
4. The U.S. Government
Monetary policy's direct impact on federal spending and revenue
decisions generally is limited. The changes in interest rates that accompany economic expansion and recession do affect the interest
28 Donald Phares, "The Fiscal Status of the State-Local Sector: A Look to the 1980's," in
Norman Walzer and David L. Chicoine, eds., Financing State and Local Governments in
the 1980's (Cambridge, Mass.: Oelgeschlager, Gunn, andHain, 1981).
29 John King, "State and Local Debt," in J. Richards Aronson and John L. Hilley, eds.,
Financing State and Local Governments, 4th ed. (Washington, D.C.: Brookings Institution,
1986) pp. 160-92.




195




cost of refinancing outstanding debt and issuing new debt. The budgetary impact can be sizable, since a significant portion of the outstanding debt must be refinanced annually.30 In the 1980s, however,
the Treasury's heavy reliance on financing with intermediate and
long-term issues has reduced the share of the marketable Treasury
debt that was refinanced each year from around 50 percent at the
beginning of the decade to around 35 percent more recently.
Changes in Treasury interest costs over the business cycle affect
the size of the budget deficit and hence the degree of concern about
it. However, they do not appear to have influenced federal fiscal policy very much. Furthermore, the Treasury chooses the maturity
structure of its debt on the basis of longer term objectives and does
not alter its debt management strategy in response to changes in
the shape of the yield curve.
The Treasury is a major force in financial markets, competing
with other borrowers for funds and for command over real
resources. Real federal credit demands have tended to rise relatively more in recessions than in expansions. Thus, over the cycle,
they have generally run counter to demands of other borrowers.
Spending on unemployment compensation and other income-sustaining programs generally falls during expansions and rises during
contractions. The 1980s have seen considerable modifications to
that pattern as deficits remained high during the long period of
expansion. In the early stages of the expansion, changes in the
cyclical elements of government spending were more than offset
by a buildup in defense spending.
The pattern of Treasury revenues has also been modified in the
1980s. Revenues generally rise faster than GNP during expansions
because tax receipts rise more than proportionately as incomes
increase. The procyclical pattern to revenues has been reduced
during the 1980s through changes in the tax code. The 1981 tax law
introduced indexing of tax brackets; receipts have ceased to be
accelerated by bracket creep during periods of inflation. Nevertheless, under the 1981 law the progressivity of tax rates, particularly
for individuals, still provided for an increased rate of tax collection
as real incomes rose. In the 1986 tax reform, the progressivity of the
tax rates was reduced considerably, and some of the tax burden was
shifted from the consumer to the corporate sector. The new tax
structure will likely cut back somewhat on the cyclical variation in
30 The impact of interest rates on the deficit is discussed in T.M. Holloway, "Measuring the
Sensitivity of Net Interest Paid to the Business Cycle and to Inflation," Public Finance
Quarterly, uol.15, no. 3 (July 1987), pp. 235-58.

196

the ratio of taxes to income, although it will still provide for rising
tax collections as incomes increase. In any case, it will require some
experience with tax revenues over a range of business conditions to
obtain a complete picture of the cyclical effects of the changes in
the tax laws.
Since the Treasury can always satisfy its credit needs, some
observers have questioned whether its heavy borrowing during the
long expansion of the 1980s might have adversely affected the flow
of funds to other potential borrowers. Despite the high deficits,
however, nominal rates fell during much of the decade's expansion
because of the sharp decline in inflation and inflationary expectations from the unusually high levels reached previously. Real interest rates, nonetheless, remained quite high from 1982 to 1985
even as nominal rates fell. While the high interest rates may have
crowded out some borrowers, most firms that were willing to pay
for credit seemed to be able to get it. Over this period, high real
interest rates and a growing economy attracted large capital
inflows to the United States. For a few years, these factors also promoted a stronger dollar. The strong dollar in turn constrained
growth in the U.S. export sector, a development which offset some
of the stimulus coming from the budget deficit. Beginning in 1985,
the dollar declined substantially because of investor concern about
the longer run implications of the large U.S. budget and trade
deficits and as a result of a change in foreign exchange market
intervention policies. At the same time, slower growth in the U.S.
economy and a narrowing of the spreads between interest rates in
the United States and abroad also contributed to a weaker dollar,
which encouraged a pickup in exports.37
The Role of the Fed Watchers
The participants in the money and capital markets watch the
actions of the Federal Reserve. They try to understand the basic
thrust of policy and to detect any signs that objectives are changing
by predicting the variables that the Federal Reserve follows and anticipating future policy developments.32 Their activities may help speed
changes in expectations about policy and hence the rate at which
economic activity adjusts to new information. The increased sensitiv31 Gerald P. Dwyer, Jr., "Federal Deficits, Interest Rates, and Monetary Policy," Journal of
Money, Credit and Banking, vol. 17, no. 4, pt. 2 (November 1985), pp. 655-81.
32 David M. Jones, Fed Watching and Interest Rate Projections: A Practical Guide, New York
Institute of Finance (New York: Simon and Schuster, 1989); and William W. Melton, Inside
the Fed: Making Monetary Policy, (Homewood, Hi: Dow Jones-Irwin, 1985).




197

ity of markets to information may add to interest rate volatility. The
financial firms employ economists to help them anticipate the effects
of policy moves on interest rates and on the demand for credit
because such information is important to the firms' trading and positioning strategies. In addition, The economists track the activities of
the trading desk, the FOMC and the economy.
Analysts who work for banks or dealer firms provide regular
briefings to their own managers and are close at hand to analyze
current developments during the day. Sales personnel at most
firms circulate rapidly the current views of their in-house experts.
The economists also meet with clients and are available to customers for telephone consultation. Their expertise, or boldness, in
making interest rate forecasts helps them achieve high visibility
through the financial press. The independent entrepreneurs often
present their basic analyses to clients through a market letter, usually a weekly report on recent and prospective developments.
Many of the analysts provide daily commentary on prospective
and actual Federal Reserve operations through computer information systems, which can be accessed by paying customers. The
popularity of these services has led to their introduction in Europe
and the Far East through news wire and computer screen systems.
Some of the services report on surveys made of their fellow forecasters so that subscribers can learn the range of the estimates of
various monetary and economic variables.
1. Projecting and Interpreting the Behavior of Reserves and Desk Activity
In tracking and anticipating the desk's actions, financial market
economists begin with a close reading of the most recently released
FOMC directive and policy record; these documents are made available a few days after the subsequent FOMC meeting. The economists try to understand the Committee's concerns and the balance of
opinion among its members. They will identify those factors that the
FOMC chose to emphasize in its guidelines to the desk in order to
predict a change in reserve pressures that might be made between
meetings. Before the policy record is released, the economists must
interpret developments in light of their understanding of the
Committee's primary emphasis—be it the monetary aggregates,
economic activity, inflation, financial market conditions, or the
exchange rate. Then they must gauge the likely behavior of those
variables that the Fed appears to be following most closely.33




33 See H. Robert Heller, "Implementing Monetary Policy," Federal Reserve Bulletin, July
1986, pp. 419-29, especially pp. 428-29.

198

When money is thought to be the focus, then developing a view
of money market conditions over the coming month or two requires
estimating how the aggregates are likely to behave compared to the
growth assumed acceptable to the FOMC. For instance, if money
growth thus far in the year has been below targeted growth, the
economists will consider whether there are good reasons for the
shortfall that might prompt the Committee to accept the slow
growth. Should they not find such reasons, they may expect the
Committee to encourage faster growth until money gets back on
path. When the economists believe that the Fed is giving primary
weight to the economy, inflation, or the exchange value of the dollar,
they will put particular effort into forecasting these elements.
In tracking desk operations and evaluating whether reserve
pressures have been modified, analysts try to distinguish between
the defensive and dynamic aspects of open market operations. An
open market operation may represent a change in stance toward
the policy objective or it may merely be designed to offset the
movement in some other balance sheet item or address normal
seasonal movement in required reserves. To interpret open market
actions properly, Fed watchers must analyze a variety of statistics.
They essentially duplicate the Fed's daily estimation of reserve
supplies and demands. (For a description of this process, see
Chapter 6.) The outside forecasters operate under a handicap during the period since they do not have the daily flow of reserve information available to their Federal Reserve counterparts. While they
can estimate the scale of daily desk operations in the open market
and learn what the Treasury's cash balances were with a one-day
lag, they can only speculate on the behavior of Federal Reserve
float or the size of the desk's transactions with foreign accounts.
Each Thursday afternoon, after the Federal Reserve's 4:30
release of a variety of statistics including weekly data on the
Federal Reserve balance sheet, the Fed watchers analyze the borrowing at the discount window, excess reserves, and other factors
to try to assess the Fed's policy stance. They will question whether
the borrowing level achieved was the one that was intended, or
whether some unexpected movement in the demand for excess
reserves or some balance sheet item pushed borrowed reserves off
track. Under most circumstances, the analysts will expect borrowing and reserve pressures to continue near the level recently prevailing. They will also estimate what range for the Federal funds
rate appears to be consistent with such borrowing. Sometimes,
however, the behavior of recent economic or monetary data and




199

their reading of desk behavior will lead them to believe that a
change in reserve pressures has occurred.
The analysts are keen observers of the trading desk's actual
operations and the Federal funds rate. They use their expectations
about policy and their knowledge of seasonal pressures on the
funds rate to sort out whether a change in the funds rate results
from a change in reserve pressure engineered by the desk or from
skewed distribution of reserves or a seasonal development. Fed
watchers have to be on guard against overestimating the accuracy
of the desk's own information on demands and supplies of
reserves. Given the uncertainties with which they contend, these
economists usually perform well in providing relevant counsel to
their clients or their own firm's trading operations.
2. Budgetary and Economic Forecasting
The financial economists also project and interpret developments other than Federal Reserve policy that are likely to affect
future economic conditions and interest rates. Recognizing the federal budget's economic significance, its importance in determining
the volume of marketable securities sold to the public, and its
impact on Treasury cash positions and on reserve flows, the analysts are often skillful interpreters of the continuing adjustments to
the budget data. They pore over the fine print in the budget documents and then estimate what new spending commitments and tax
actions are likely to emerge from the congressional mill. Since the
Treasury and off-budget agencies are by far the largest borrowers
in the financial markets, their future activities influence the outlook
for interest rates.
The hunger of market participants for information on the variables
tracked by the Federal Reserve has placed new demands on these
economists. They produce advance estimates of key economic
statistics to give market traders a benchmark for evaluating the
statistics when they are released. Statistics published monthly on the
economy—retail sales, production, employment, and prices—are all
forecast. There are fashions in what data receive the most attention.
Whenever the Federal Reserve is perceived to be shifting its focus,
market attention and forecasting efforts shift as well.
To forecast economic behavior, financial economists examine
recent trends and consider components that might be changing.
Some analysts make comprehensive forecasts of the supply and
demand for funds associated with different sectors of the economy
—consumer, business, government, and foreign. The modeling




200

involved relies heavily on both individual judgment and econometric techniques.34
As this chapter has conveyed, the economy responds to Federal
Reserve policy in a variety of ways. It reacts to interest rates, to
monetary impulses, and to expectations of future developments in
these measures. Because of the important role of expectations,
almost any factor that affects the formation of expectations can
enter into the transmission process. This chapter has concentrated
on the domestic side of policy transmission. The next chapter
looks at the international dimensions.

34 Short-term economic forecasting is difficult and even the most skilled practitioners' forecasts are often wide of the mark. Furthermore, preliminary published figures may be substantially different from the final figures. Accuracy of private forecasts is discussed by
Steven Strongin and Paula S. Binkley in "A Policymakers' Guide to Economic Forecasts,"
Federal Reserve Bank of Chicago Economic Perspectives, May-June 1988, pp. 3-10.




201

International Aspects of Monetary
Policy and Financial Markets.
The increasing interdependence of U.S. economic activity and
developments abroad has made the U.S. economy more open today
than it was in the 1950s. The United States has become increasingly
subject to direct influences from abroad in the form of substantial
cross-border flows of goods, services, and capital. In addition, foreign exchange rates, foreign interest rates and prices figure in the
economy's response to developments in other countries. These
changes reflect a worldwide increase in the freedom of goods and
capital to move across borders. As a result, developments in the
United States, including policy moves, can have part of their effect
on international flows and foreign developments, with subsequent
effects back on the domestic economy.
Barriers to trade and capital flows have fallen substantially and
world trade volumes have grown twice as fast as real GNP in the
industrial countries over the last four decades. The sum of U.S.
imports and exports, one measure of openness, has expanded from
less than 8 percent of GNP in the 1950s to 16 percent in 1988 (see
Chart 1, page 203). In the 1980s, cross-border capital flows have
become a complex web of banking, securities, and direct investment
transactions. They are denominated not only in the traditional currencies, U.S. dollars and U.K. pounds, but also in German marks,
Japanese yen, and other currencies. The United States was the recipient of net foreign capital inflows averaging more than $125 billion a
year from 1985 to 1988, the result of its large current account deficit.
Because of the U.S. economy's size and the dollar's roles as major
reserve currency and medium for trade and financial transactions,
U.S. monetary policy has long had an important influence on interest rates, inflation, and economic growth in much of the world. While
U.S. monetary policy has primarily been determined by domestic
economic and financial developments, the greater openness of
world trade and finance has also meant that events abroad can at
times influence the formulation of U.S. monetary policy and the
effects of policy actions on the U.S. economy.
Key Changes in the International Financial System
The current degree of economic interdependence was fostered
by two sets of developments in the 1970s and 1980s. The first was
the abandonment in the early 1970s of the Bretton Woods system
of pegged exchange rates and the move to floating exchange rates.
The change occurred in an environment in which there was both a
growing belief in allowing markets to adjust and a series of strains
in the old system arising from persistent U.S. balance of payments




202

Chart 1 U.S. Nominal Merchandise Trade Flows as a Proportion of Nominal GNP




1950 '52 '54 '56 '58 '60 '62 '64 '66 '68 70 '72 74 76 78 '80 '82 '84 '86 '88

deficits, worsened by the emergence of inflation in the United States.
The floating-rate system that supplanted the Bretton Woods regime
relied on a highly visible price adjustment mechanism, the dollar's
exchange rate against other major currencies, to eliminate imbalances in international payments. Under floating rates, the world
economy was expected to escape from persistent payments problems. At the same time, central banks were expected to have greater
independence in choosing their domestic policy course, since they
would be freed of the obligation to intervene to support fixed
exchange rates.
The second set of changes was the progressive dismantling of
restrictions on international capital flows. Foreign exchange and
capital flow controls had restrained—with ever less effectiveness —
the potential flow of capital. The United States removed most capital
controls with the advent of floating exchange rates. Other countries
reduced capital controls gradually under the combined influences
of free-market philosophies and pressures from shifting payments
imbalances that proved wider than those under Bretton Woods.'
The two sets of changes have had differing implications for the
effectiveness of monetary policy. Domestic monetary policy actions
1 Fora fuller treatment of these issues, see Federal Reserve Bank of New York, Research Papers
on International Integration of Financial Markets and G.S. Monetary Policy, December 1987,
a compendium of papers on increased capital mobility and its effects.

203

can be more effective under flexible exchange rates than they were
under relatively fixed exchange rates in part because policy is less
constrained by official balance of payments settlements. In addition,
policy-induced exchange rate changes reinforce the effects of monetary policy on the domestic economy. A tightening of monetary
policy, for example, tends to restrain (J.S. economic growth and
inflation and also tends to drive up the dollar's exchange rate. A
higher exchange rate in turn shifts production from the United States
to other countries and contributes to slowing U.S. economic growth.
Yet at the same time that flexible exchange rates seem to have made
monetary policy potentially more effective, freer capital movements have made the policy transmission mechanism more complex than it was in the 1950s and 1960s. Freer capital movements
promote a wave of rapid and widespread financial adjustments to
a U.S. monetary policy action in advance of the slower and more
complicated real and price adjustments.2
International Channels of Transmission of U.S. Monetary Policy
Domestic open market operations, as well as other monetary policy measures, have international effects that feed back to the U.S.
economy through their influence on financial and economic conditions in other countries. The following sections describe the series
of changes that might be set in motion by a Federal Reserve decision
to ease monetary policy.
1. The Impact of Expectations and Financial Asset Prices
The international money and capital markets and the foreign
exchange markets represent the most immediate channel of transmission of U.S. monetary policy to other countries. The financial
market response to a U.S. monetary policy action depends on an
interplay of domestic and foreign influences.
A U.S. monetary easing that produces a reduction in nominal
and real interest rates on short-term dollar investments relative to




2 Open economy macroeconomics under varying assumptions about capital mobility and
exchange rate flexibility is covered in Rudiger Dornbusch and Stanley Fischer,
Macroeconomics, 4th ed. (New York: McGraw-Hill Book Company, 1987). Cinder fixed
exchange rates and moderate to low capital mobility, monetary policy has relatively more
effect on the external balance while fiscal policy has relatively more effect on output and
employment. Cinder floating rates and high capital mobility, the effects of both policies are
more pervasive. Current models of an open economy under floating rates and high capital
mobility are descendants of the Mundell-Fleming model. See Jacob Frenkel and Michael
Mussa, "Asset Markets, Exchange Rates and the Balance of Payments," in R.W. Jones and
R Kenen, eds., Handbook of International Economics, vol. 2 (Amsterdam: North-Holland
Books, 1985).

204

those abroad should cause investors to shift from dollar to foreign
currency assets, thereby placing downward pressure on the exchange value of the dollar and on foreign interest rates. Investor
and borrower decisions take into account the broad range of factors
discussed in Chapter 8, but they also reflect the exchange rate risk
involved in cross-border transactions.3
This last consideration comes into play as investors and borrowers compare expected rates of return across currencies expressed
in the home currency. A U.S. investor would compare expected
nominal returns from making a U.S. dollar-denominated investment
to the return denominated in dollars of making a foreign currency
investment.4 The nominal dollar return on the foreign currency
investment consists of its nominal interest rate plus the expected
change in the exchange value of foreign currency, less an adjustment for risk. For two instruments of comparable credit risk and
maturity, the main additional risk of a foreign currency instrument
is uncertainty about the future exchange rate.5 In some circumstances there may also be uncertainty about possible changes in
exchange and capital controls either by the United States or a foreign country. A U.S. borrower would compare the nominal interest
cost of borrowing in U.S. dollars to the cost of borrowing in a foreign
currency, expressed in dollars.6 A foreign investor would compare
the nominal return in the currency of the home country with the
return on U.S. dollar investments expressed in the home currency.
The foreign investor would add the expected change in the exchange
value of the dollar to the nominal dollar interest rate and subtract
3 Other factors play an important role in investor decisions about cross-border transactions,
but they tend to change less frequently than interest rates and market expectations and
thus are reflected in persistent interest rate differentials. These factors include the investment
and tax climates, perceptions of economic risk (that is, the extent to which economic performance is perceived to be variable or unsatisfactory) and political risk. Another factor is
the perceived riskiness of expected real rates, which depends on the expected volatility of
interest rates and, fora foreign asset, the expected volatility of the exchange rate.
4 The return calculations would allow for taxes, default risk, and any other factors that would
be considered in choosing among dollar-denominated investments. Expected returns on
long-term assets include expected capital gains and losses at the end of the investment horizon; these can be quite important. At times in the 1980s, foreign investors have purchased
U.S. bonds while long-term rates were falling because expected capital gains exceeded
expected exchange losses. The purchases in fact helped to bid up the dollar temporarily.
5 For short-term investments it is possible to avoid this risk by engaging in a forward purchase
of dollars for the maturity date.
6 For a further explanation of the reasons that interest rates differ across currencies and a
look at the evidence in the postwar period, see Bruce Kasman and Charles Pigott, "Interest
Rate Divergences among the Major Industrial Nations," Federal Reserve Bank of New York
Quarterly Review, Autumn 1988, pp. 28-44.




205




an adjustment for risk. A foreign borrower would make a similar
comparison of borrowing costs in the home currency and in U.S.
dollars expressed in the home currency.
These straightforward comparisons by investors and borrowers
contribute to a more fundamental economic process. Because
investors and borrowers ultimately make decisions to save and borrow based on the ex ante real interest rate level in their home country, their comparisons of international returns adjusted for exchange
rates lead to an economic mechanism that compares real interest
rates across countries adjusted for the expected change in the real
exchange rate over the holding period. A change in the U.S. real
exchange rate is the change in the nominal exchange rate of the
U.S. dollar in units of foreign currency plus an adjustment for
changes in the dollar's purchasing power relative to foreign currency, represented by the difference between foreign inflation and U.S.
inflation. In effect, the real exchange rate expresses the relative
value of U.S. goods in terms of foreign goods. Changes in the ex
ante real rate cannot be measured directly and are commonly measured by changes in the ex post real exchange rate, a rather imperfect proxy. The ex post real rate has shown large swings in recent
years. The dollar appreciated 40 percent in real trade-weighted
terms from late 1980 to its peak in early 1985, then depreciated
dramatically through mid-1988 (see Chart 2, page 207).7
A decrease in U.S. ex ante real interest rates relative to those
abroad tends to lower the foreign exchange value of the U.S. dollar.
A decrease in U.S. real interest rates encourages domestic and foreign investors to purchase foreign currency financial assets.
Borrowers turn to the U.S. dollar markets from markets with higher
real interest rates. The incipient flows bid down the U.S. dollar
exchange rate and thereby alter supply and demand across national
7 In the international economics literature, the Fisher parity condition states that the nominal
interest rate on a U.S. financial asset is equal to the foreign nominal interest rate plus the
expected appreciation of the foreign currency exchange rate (an expected depreciation
would be subtracted). This condition can be written:
RUS = RF+E,
where R(j$ is the nominal U.S. interest rate, ft/r is the nominal foreign currency interest
rate, and E is the expected rate of foreign currency exchange rate appreciations. After adjusting for expected price inflation in the United States and in the foreign country, P(j$ and Pf,
respectively, we obtain the Fisher condition expressed in terms of real interest rates:
R

US - pUS = IRF- pFl + lE + (pF- pUS)l
The second term on the right hand side is the expected appreciation in the real exchange
rate of the foreign currency (the nominal exchange rate plus the difference between foreign
inflation and U.S. inflation).

206

Chart 2 U.S. Dollar Effective Exchange Rates
Weighted against Sixteen Industrial Countries
Index 1980 =100
160

/

150

\
\

140
130

fir"

120
110

Nominal

100

Real

90

1975

76

77

78

79

'80

'81

'82

'83

'84

'85

'86

'87 '88
Source: International Financial Statistics

capital markets to bring real interest rates into closer alignment
(see Charts 3 and 4, page 208). 8
The importance of real interest rates points to one situation in
which a fall in nominal interest rates will have little effect: when the
fall in U.S. rates just matches a drop in CJ.S. inflationary expectations. A drop in inflationary expectations leads U.S. and foreign
investors to raise their expectations of the future path of the U.S.
dollar exchange rate, so that the path reflects the dollar's expected
gain in relative purchasing power. The lower nominal return on dollar assets largely compensates investors for higher expected dollar
appreciation (or less expected depreciation). In this case, lower
8 This tendency holds even if ex ante real interest rates adjusted for expected real exchange
movements are not equalized, that is, even if foreign and domestic assets are not perfect
substitutes. The ex post real interest rates in Chart 4 can abo diverge because the expected
inflation and exchange rate movements can differ substantially from the actual movements.
Some economists believe that real interest rates are equalized across countries. This real
interest rate parity hypothesis, a fairly strong condition, is usually rejected by empirical
evidence. A weakness of such tests is that direct measures of ex ante real rates cannot be
made. The concept and measurement of real interest rates are discussed in Chapter 8 of
this volume.




207

Chart 3 Nominal Three-Month Interest Rates
Percent

U.S. commercial paper rate
Japanese Gensaki rate
German interbank rate

17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
1980

1981

1982

1983

1984

1985

1986

1987

1988

Chart 4 Ex Post "Real" Three-Month Interest Rates
Percent

^ ^ ^
U.S. commercial paper rate

10
9
8
7
6
5
4
3
2

Pi




i

(
i

#

V"

* I-#

iV

7986

7987

V

-1
-3

7980

208

1982

1983

1984

V

WO

#

\

1

-2
German interbank rate

A

r

0
Japanese Gensaki rate

1

"A/

1985

x

f

7988

nominal interest rates do not induce investors to switch their assets
to other currencies from U.S. dollars.
While the U.S. easing sets in motion a process that tends to lower
the dollar, the extent of the dollar's fall depends in part on the monetary
policies pursued by foreign central banks. If economic and inflationary pressures abroad are weak or the foreign central bank is stabilizing
the foreign exchange value of its currency, the foreign central bank
may also ease its short-term interest rates. In this case, the dollar's
exchange rate may undergo little or no change. However, if a foreign
central bank takes steps to maintain its interest rate levels for domestic policy reasons while (J.S. rates are lower, the dollar may fall.
The effects depend further on market perceptions of the appropriateness of the (J.S. and foreign monetary policy stances. Central
banks, if they choose to, can have considerable influence over shortterm interest rates. Hence, market perceptions may be reflected primarily in changes in the spot exchange rate and in relative long-term
interest rates, changes which themselves reflect reactions to the policy moves and expected longer term relative price performance. If
the (J.S. monetary easing is viewed as inflationary but the foreign
central bank's stance is not, U.S. long-term interest rates may rise
while foreign rates remain unchanged. The relatively higher U.S. longterm rate compensates investors for lower expected future U.S. dollar
purchasing power and thus for likely expected future exchange losses.
In practice, since interest rates and exchange rates are simultaneously determined, predicting with precision how they will change
after a monetary policy action is impossible. The broad tendency,
however, is for a reduction in U.S. nominal and real interest rates
eventually to lower the exchange value of the dollar unless other
central banks are reducing interest rates at the same time.9
In some countries, intervention in the foreign exchange markets
may be used for domestic monetary policy purposes (see Box, page
222). In the United States, intervention is not used to adjust reserves
or interest rates and is generally not permitted to do so; any reserve
9 This discussion has assumed a fairly close relationship between inflation, interest rate, and
exchange rate expectations, on the one hand, and on the other, the term structure of domestic
interest rates, foreign interest rates and forward foreign exchange rates. The purchasing
power parity hypothesis posits a very close relationship between inflation differentiab and
exchange rate movements. That is, at any time, a dollar will buy the same amount of goods
in the United States as it does in Germany after conversion to marks; the real exchange
rate is constant. Rational investors would therefore expect future exchange rate changes to
mirror expected inflation differentials exactly. Purchasing power parity is nearly always
violated in the short run, but some economists believe it may hold in the long run. The
analysis here assumes that inflationary expectations play a straightforward role in determining the term structure of interest rates. The actual interrelationships are more complex.




209

impact is offset with other policy tools. U.S. foreign exchange intervention is designed specifically to modify foreign exchange market
conditions. On occasion, the economic forces prompting the intervention will encourage an adjustment in the stance of domestic policy that works in the same direction, but two separate decisions
would be made.
2. International Effects of Changes in U.S. Real Activity and Prices
The reduction in Ci.S. interest rates under a monetary easing triggers responses in the real economy and in the prices of goods and
services in the United States that ultimately affect the U.S. trade
account. If the easing stimulates U.S. domestic demand, it also
increases U.S. demand for imported goods and services over the
next few quarters. If the easing at the same time raises U.S. prices
of goods and services, both foreign and domestic customers have
an incentive to shift from U.S.-produced goods to foreign goods.
Together, these factors increase U.S. import demand and may
reduce foreign demand for U.S. exports somewhat. Thus, the trade
surplus will decline or the deficit widen.70
If the monetary easing also produces a fall in the foreign exchange
rate, the stimulative effects of the easing on the U.S. economy are
reinforced. A lower dollar exchange rate makes foreign goods more
expensive relative to domestic goods and U.S. goods more attractive in overseas markets. Both U.S. and foreign customers have an
incentive to shift to relatively less expensive U.S.-produced goods
and services from those produced abroad, offsetting part of the
impact on trade of higher U.S. growth.77
While higher U.S. demand and a lower U.S. dollar are both stimulative to the U.S. economy, they have offsetting long-run effects on
foreign economies. The expansionary effects initially predominate.
From the foreign country's perspective, the increased U.S. demand
for its goods resulting from higher U.S. growth and prices provides
a stimulus to production, while the lower dollar channels demand




10 See the analysis of trade balance determinants in Robert A. Feldman, "Dollar Appreciation,
Foreign Trade, and the U.S. Economy," Federal Reserve Bank ofNew York Quarterly Review,
Summer 1982, pp. 1-9.
11 If the easier U.S. monetary policy is accompanied by a decline in the dollar, changes in
price competitiveness and the terms of trade may, in part, offset the effects of stronger U.S.
growth on the trade balance. "Terms of trade" describes the price of imports relative to the
price of export goods. When the dollar falls, U.S. consumers lose some purchasing power
through the terms of trade effect because imported goods are now relatively more expensive.
The larger the trade flows are relative to domestic economic activity, the larger the potential
effect. The lower dollar can also contribute to a higher price level in the United States since
foreign goods are more expensive.

210

back to the United States by making G.S. goods and services relatively less expensive. Changes in G.S. real income fairly quickly
affect the G.S. trade flows, while several quarters can pass before
the major effects of changes in prices and the exchange rate become
apparent.72 Hence, a G.S. monetary policy easing is usually initially
stimulative to foreign countries. Policymakers in those countries can
try to offset the stimulative effects, but they may dislike some of the
implications of doing so. A move to tighten monetary policy tends
to drive up the country's interest rates and its exchange rate, reducing
its trade competitiveness. Since international trade is a large part of
GNP in many countries, choosing an appropriate level of policy
response to offset a stimulus from the Gnited States is difficult.
The initial stimulus to foreign economies from a G.S. monetary
policy easing eventually feeds back—albeit weakly—to the Gnited
States through higher demand for G.S. exports and through changes
in the price competitiveness of G.S. goods and services. When combined with the rapid and continuous adjustments in the financial
markets and with changes in real activity and prices in foreign countries, the G.S. economy's response to a change in monetary policy
can quickly develop a complex and sustained dynamic.
The type of cross-border interaction between policy moves and
economic performance just described strengthens the role of medium-term interdependence between the Gnited States and the rest
of the world, while still leaving a substantial role for purely domestic
factors. The mix of interdependent and independent factors can be
seen in the inflation process in three representative industrial countries—the Gnited States, Germany and Japan—during the 1970s and
1980s (see Chart 5, page 212). Rapid growth in the early 1970s,
fueled in many industrial countries by monetary expansion, put
strong upward pressure on commodity prices and set the stage for
OPECs first round of oil price increases. These developments contributed to inflation rates in 1974 that were high relative to those in
the preceding 20 years in most industrial countries.
The magnitude of the acceleration varied considerably among
countries, however, because of differences in domestic policy and
structural features of the economy. Monetary tightening to resist
inflation in all three industrial countries helped produce a sharp
12 See Ellen E. Meade, "Exchange Rates, Adjustment, and the J-Curue," Federal Reserve
Bulletin, vol. 74 (October 1986), pp. 633-44; Catherine L Mann, "Prices, Profit Margins,
and Exchange Rates," Federal Reserve Bulletin, vol. 72 (June 1986), pp. 366-79; and Paul
R. Krugman and Richard E. Baldwin, "The Persistence of the U.S. Trade Deficit," Brookings
Papers on Economic Activity, 1987: 1, pp. 1-43.




211

Chart 5 Consumer Price Inflation

United States
Japan
Germany
Source: International
Financial
Statistics

1950 '52 '54 '56 '58 '60 '62 '64 '66 '68 '70 '72 '74 '76 '78 '80 '82 '84 '86 '88

recession in 1974-75 and a general slowdown in inflation in 1975
and 1976. Subsequently, the inflation trend in the United States
diverged from that in Germany and Japan because U.S. monetary
and fiscal policies remained relatively more expansive. The reemergence of strong growth and inflationary pressures in the United
States and some other industrial countries in the late 1970s contributed to a second sharp runup in commodity prices, a second
round of oil price increases, and rising inflation—although from differing levels—in all three countries by 1980. The simultaneous
efforts of the industrial countries to combat inflation brought about
an extended period of disinflation in the first half of the 1980s.
International Influences on U.S. Monetary Policy
International developments have often had an impact on the ultimate goals of U.S. monetary policy in the form of sustainable U.S.




212

economic growth and price stability. For example, they have contributed to large swings in the value of the dollar, the prices of oil
and other commodities, and the state of the G.S. trade and current
accounts. The Federal Reserve generally does not directly adjust
its policy in response to international developments, but it may do
so indirectly when the developments affect the intermediate and
ultimate goals of policy. For example, a political development that
reduces the value of the dollar or a supply restriction that raises
internationally traded commodity prices tends to raise import
prices and can create domestic inflationary pressures. In formulating domestic policy, the FOMC monitors changes in the dollar
and commodity prices; it will respond to them only if the movements appear to signal changes in G.S. inflationary pressures or
G.S. real growth.
International developments can on occasion move countries substantially away from both full employment and price stability, producing a policy dilemma. The oil shocks of the 1970s are good
examples: they raised inflationary pressures and weakened domestic demand (by reducing consumer purchasing power), outcomes
which called for opposite monetary responses. International developments can also move countries away from both domestic goals
and external goals. The oil shocks of the 1970s illustrate this potential as well: the policy stimulus to offset the contractionary effects
of the oil price increases would worsen the already large current
account deficits arising from the higher cost of oil imports. Indeed,
past experience has shown that countries with large current account
deficits cannot long escape assigning considerable weight to the
external goal—even though it conflicts with domestic objectives.
Because of the size of the G.S. economy and the dollar's role as a
reserve currency and medium for international transactions, the Gnited
States has perhaps more latitude than most countries to place domestic above external policy considerations. In the postwar period, clearcut instances in which domestic goals had to be sacrificed to external
concerns have been rare. Such conflicts are possible, however. In the
mid-1980s, many economists worried that such high dollar interest
rates would be required to attract private financing for the large G.S.
current account deficit that domestic growth would slow sharply.
As a reflection of growing interdependence, international monetary and fiscal policy coordination has often had an important place
in the formulation of G.S. monetary policy. The extent of policy
coordination has been greater at some times than others; disagreements sometimes arise about the distribution of the costs and the




213

Chart 6 U.S. Current Account and Merchandise Trade Balance
Billions of dollars
20

Current account
Trade balance




1973 74 75 76 77 78 79 '80 '81 '82 '83 '84 '85 '86 '87 '88

risks involved in any coordination effort. The emergence of large
U.S. current account deficits in 1977-78 and in the 1980s (Chart
6) prompted intense debate among the major industrial countries
about the relative magnitude of adjustment to be undertaken by
each as they shaped a collaborative approach to reducing the international imbalances.
The difficulty of coordinating macroeconomic policies in the
1960s was a major reason fixed exchange rates were abandoned.
The experience with large capitalflowsand wide and often persistent
payments imbalances in the 1970s and 1980s suggests that policy
coordination is still desirable under the floating exchange rate
regime. Whilefloatingexchange rates have increased the potential
independence of monetary policy, the growing openness of world
trade and finance has increased interdependence among countries'
economies. Floating exchange rates allow the central bank to set its
nominal interest rates (or reserve base) independently of other countries, but capital mobility means that a change in the real interest
rate relative to other countries is offset by a change in real exchange
rate expectations. Thus, central bank freedom to set interest rates
is only as great as its tolerance of the foreign exchange rate movements associated with a change in interest rates.

214

The Impact of Changes in International Financial Markets
The initial financial adjustments triggered by a U.S. monetary policy
action involve the huge dollar and nondollar financial markets outside
the United States. The rapid growth and changing character of these
financial markets do not so much alter the qualitative effects of a U.S.
monetary policy action on U.S. inflation or output as add to the complexity of tracing the timing and ultimate magnitude of those effects.
Reductions in the cost of information and of transactions and
improvements in trading system technology have accelerated the
interaction between U.S. and overseas financial markets. International
trades can be executed quickly. For a few instruments, trading can
occur at nearly any time of the day or night. Foreign investors are able
to react almost immediately to news in the United States; U.S. investors
can respond quickly to developments abroad. Furthermore, market
participants seek out and exploit arbitrage opportunities across an
increasing number of markets, reducing market segmentation.
The underlying character of many national financial markets has
changed over the last two decades as well. Changes in regulatory
philosophy and economic forces lay behind the extensive deregulation of domestic financial markets in the United States and abroad.
The experience of rapid and persistent inflation undermined regulated or administered interest rate systems. In many industrial countries, the emergence of large government budget deficits opened
up alternatives to bank deposits for domestic investors and prompted government debt market reform.73
Competitive forces and the price mechanism gained a much larger role in foreign financial markets, particularly in the 1980s. Access
to foreign and offshore markets, falling information costs, and
increased investor sophistication reduced the effectiveness of regulating deposit and lending rates and imposing quantitative credit
controls. Many countries liberalized and developed their money
markets and stock and futures exchanges. They substantially modified the separation of powers of financial institutions and eased the
entry of foreign financial firms.
These changes in turn induced foreign central banks to change
their approaches to conducting monetary policy. In the late 1970s,
very few foreign monetary authorities conducted open market operations in domestic securities as the principal instrument of monetary
policy. Most relied primarily on administered interest rates, discount
window quotas, and credit and capital controls. Freer financial mar13 The forces behind innovation and their impact on the financial markets are discussed in
Bank for International Settlements, Financial Innovation and Monetary Policy, Basle, 1964.




215

kets required more flexible monetary policy tools, however. Now
monetary authorities in most larger countries carry out some form
of open market operations in domestic securities, operations which
not only affect domestic money market conditions but also are
closely watched in the exchange markets. The shift in operations
led to more professional analysis of foreign monetary policy along
the lines of "Fedwatching."
The internationalization of markets was further assisted by the
development of markets for off-balance sheet instruments such as
interest rate and currency swaps, financial futures and options, and
forward rate agreements.74 These instruments allow financial market participants to hedge interest rate and exchange rate risks, to
adjust market risk positions quickly, and to separate to a considerable extent the management of price risks from the management
of credit and liquidity. Markets for instruments that transfer credit
and liquidity risks, such as letters of credit, sureties, and committed
lines of credit, also grew substantially in the 1980s.
Changes in the ability of financial market participants to manage
exposures to foreign currency and interest rate risk and to tap foreign credit sources have raised concern that the influence of monetary policy on macroeconomic performance has waned. Analysts
have shown, however, that the evidence for an erosion is not compelling.75 Still, the growth of the U.S. economy after 1982 despite
unusually high nominal and real interest rates raises questions, even
when the strong fiscal stimulus is taken into account. While it was
widely appreciated that the repeal of usury laws and the phaseout
of Federal Reserve Regulation Q restrictions on interest rates would
alter one of the channels of transmission of a monetary policy tightening—that from credit rationing to higher credit cost—few observers
probably anticipated the large potential supply of credit from overseas investors at high interest rates.
The Principal International Short-Term Markets
1. The Eurodollar Deposit Market
The large Eurodollar deposit market provides convenient access
to dollars for non-CJ.S. residents who use them for commercial and




14 These instruments are described in Bank for International Settlements, Recent Innovations
in International Banking, Basle, 1986.
15 See M.A. Akhtar and Ethan S. Harris, "Monetary Policy Influence on the Economy—An
Empirical Analysis/' Federal Reserve Bank of New York Quarterly Review, Winter 1987.
Further evidence is contained in Lawrence Radecki and Vincent Reinhart, "The Globalization
ofFinancial Markets and the Effectiveness ofMonetary Policy Instruments," Federal Reserve
Bank of New York Quarterly Review, Autumn 1988, pp. 18-27.

216

investment activities. The gross volume of Eurodollar deposits
amounted to $4.6 trillion at the end of 1988. Interbank deposits
make up about two-thirds of the total.
As noted in Chapter 4, the term Eurodollar is applied to dollar
deposits of all maturities booked at offices of financial institutions
outside the United States. Eurocurrency deposits, more generally,
are deposits in currencies other than the domestic currency of the
country in which the deposit is booked. The Euromarket is predominantly a wholesale market based in London, although significant
Euromarket activity occurs in Luxembourg, Hong Kong, Singapore,
and other financial centers. The core of the Euromarket is the shortterm deposit market. Although deposits exist in all major currencies,
the dollar component of the market, at about 50 percent, continues
to be the largest. The interbank component plays a part in directing
credit to areas of greatest demand, even though it does not add to
total money or credit available.
Just as the Eurodollar deposit market has become closely integrated with the U.S. domestic markets, the Eurodeposit markets
are closely tied to the foreign exchange markets. Spot and forward
foreign exchange rates for most currencies against the dollar are
determined jointly with Eurodeposit rates; the percentage gap
between the spot and forward exchange rates is very close to the
interest rate differential prevailing in the respective deposit markets.
Transactions costs almost fully explain any remaining difference.76
2. Nondeposit Money Markets
New markets in Eurocommercial paper and Euronotes developed
in the 1980s. The largely dollar-denominated Eurocommercial paper
market, like the U.S. commercial paper market, competes for the
funds of institutional investors, but Eurocommercial paper is generally offered in longer maturities (three or six months) and has a liquid
secondary market. The major issuers are foreign governments and
their agencies and nonfinancial corporations based in the United
States and abroad. Euronotes are short-term promissory notes
issued under arrangements that allow the borrower to obtain credit
from banks if the notes' spread over a reference rate exceeds a specified level. With around $70 billion in issues outstanding at year-end
1988, the Europaper market is still relatively small.
Until recently, few national markets in money-market securities
existed overseas. Active Treasury bill markets are still limited to a
16 See Lawrence Kreicher, "Eurodollar Arbitrage," Federal Reserve Bank of New York
Quarterly Review, Summer 1982, pp. 10-22.




217

handful of countries, among them France, Italy, and Canada.
Activity in the large commercial bill market overshadows the
Treasury bill market in the United Kingdom. Commercial paper was
introduced during the 1980s in Japan, France, and a number of
smaller countries. International participation in these markets so
far appears small.
International Money Markets and U.S. Monetary Aggregates
Short-term instruments in the international markets can substitute for domestic deposits and thus have the potential to change
the behavior of U.S. monetary aggregates. The relationship
between nominal income, interest rates, and the monetary aggregates in the United States will remain stable only if other financial
assets cannot easily be substituted for components of the U.S.
money supply.
For U.S. wholesale depositors, Eurodollar deposits have proved
to be nearly perfect substitutes for domestic time deposits. As the
Euromarkets grew, the broader monetary aggregates became less
representative of the purchasing power available in liquid form to
the U.S. economy. In 1980, the definitions of the monetary aggregates were revised to include U.S. resident deposits in overnight
Eurodollars in M2 and both overnight and term Eurodollars in M3.77
Foreign currency Eurodeposits and other nondollar short-term
instruments represent potential substitutes for U.S. domestic
deposits in two respects. They can be paired with an off-balance
sheet instrument such as a short-dated swap or a foreign currency
future or forward to create a synthetic dollar instrument. Or dollar
deposit holders may prefer the yield and appreciation potential of
a short-term foreign currency instrument to dollar deposits. Economists are in substantial disagreement about the extent and the
effects of such substitution.18
As foreign investors have made more adjustments to their crossborder asset holdings by buying and selling medium- and long-term
foreign securities rather than by shifting their Eurodeposit holdings,
international substitutes for domestic deposits should have receded
as a source of instability in the monetary aggregates. Nevertheless,
analysts have recently inferred from changes in the money-income




17 Thomas Simpson, "The Redefinition of Monetary Aggregates," Federal Reserve Bulletin,
February 1980, pp. 97-114.
18 See M.A. Akhtarand Kenneth J. Weiller, "Developments in International Capital Mobility:
A Perspective on the Underlying Forces and the Empirical Literature," in Research Papers
on International Integration of Financial Markets and U.S. Monetary Policy.

218

relationship that changes in international capital flows may have
affected the behavior of Ml in the 1980s.79
Others have argued that distortions in the monetary aggregates
appear to be partly related to currency expectations. When the
domestic currency is expected to appreciate, foreign demand for
domestic money increases, so that at any interest rate, money
demand is higher than before and money growth faster.20 This may
be one reason that money demand appears unstable in many countries.21 Potentially at least as important for the aggregates are money
demand shifts stemming from political instability and high inflation.
The Principal International Long-Term Markets
1. The Eurobond Market
The Eurobond market, centered in London, is the principal longterm offshore dollar market, a source of capital for multinational
corporations and for foreign governments. It developed after the
Onited States instituted the Interest Equalization Tax in 1963 to stem
capital outflows that were inspired by relatively low U.S. interest
rates.22 The tax gave European corporations an incentive to issue
bonds in Europe rather than in New York and to draw on the dollar
funds that had accumulated in Europe in the postwar years. Efforts
in the 1960s to limit U.S. direct overseas investment prompted U.S.
19 See John Wenninger and Thomas Klitgaard, "Exploring the Effects of Capital Movements
on Ml and the Economy," Federal Reserve Bank of New York Quarterly Review, Summer
1987, pp. 21-31.
20 A compelling quantitative case for the United States has not been made. Some cases that
seem to support this line of analysis involve market expectations of a European Monetary
System (EMS) realignment. See, for example, Deutsche Bundesbank, Annual Report 1986,
for a description of the large capital flows surrounding the January 1987 EMS realignment
and their impact on German monetary aggregates. When foreign investors hold securities
rather than deposits, deposit rates may still be bid up to provide clearing balances and
finance dealer inventories. See Bank for International Settlements, "International Banking
and Financial Market Developments," July 1987.
21 Ronald McKinnon has advocated targeting a world Ml money aggregate made up ofindustrialcountry money supplies to get around shifts in currency preferences. See An International
Standard for Monetary Stabilization, vol. 8, Institute for International Economics, Policy
Analysis in International Economics (Cambridge: MITPress, 1984). Although the concept of
currency substitution has attracted attention, most economists reject a world Ml target. Some
economists are skeptical because they oppose money targets generally, others because they
see targeting a world monetary aggregate as stabilizing only if the relative changes in national
money demands reflect pure financial shocks, not real side shocks such as those experienced
in the early 1980s. Finally, the very low share of cross-border Ml deposits in international
assets raises questions about the practicality of such targets.
22 A brief history of the early years of the Eurobond market can be found in Frederick G. Fisher
III, The Eurodollar Bond Market (London: Euromoney Publications, 1979). The Interest
Equalization Tax was especially burdensome to the bond markets since it took the form of
an up-front fee based on the principal amount and the maturity.




219

corporations to raise capital for overseas operations in the
Euromarkets as well. Moreover, bonds issued by subsidiaries of U.S.
corporations chartered outside the United States were exempt from
the U.S. withholding tax on interest paid to foreigners. The Eurobond
market was not limited to dollars: the first Eurobond issues in foreign
currencies occurred in the mid-1960s.23
The Eurobond market was well established when the Interest
Equalization Tax was removed in 1974 and faltered only briefly
before resuming its growth. From 1985 to 1988, an average of $175
billion a year was issued in the Eurobond market, much of it in nondollar currencies. While issuance of foreign currency Eurobonds
picked up at times in the 1970s, it soared after the dollar's exchange
value began to decline in 1985. The liberalization of national markets and the growth of currency swaps added momentum to the
surge of foreign currency Eurobond issuance. A currency swap
allows a U.S. borrower, for example, to issue an Australian dollar
bond in the Euromarket and transform the exposure to U.S. dollars.
The major foreign currencies, especially the yen, accounted for
most of the growth in nondollar Eurobonds, but currency swaps
have promoted issuance in several other currencies.
2. U.S. Treasuries and Other Government Bonds
In the 1980s, government securities became prominent longterm international investments. The large stock of public debt that
accumulated in major currency countries since 1975 and the reform
of government bond markets abroad contributed to more liquid
markets. Around 1984, withholding taxes on interest income earned
by foreigners were repealed in many countries, effectively raising
returns to foreign investors. Institutional investors who prefer liquid
instruments with low credit risk also favored bonds issued by governments over the less liquid corporate Eurobonds.
As Chapter 4 revealed, a secondary market in U.S. Treasury debt
issues functioning outside the United States has grown rapidly in the
1980s. U.S.- and foreign-based dealers and U.S.-based brokers operate actively in London and Tokyo, with more limited activities in other
financial centers. The internationalization of the market for U.S.
Treasuries has set a pattern for other government bond markets.
International trading in the major government bonds has supplemented and to some extent replaced the international flows between




23 Although most of the Eurobond market is based in London, an agreement with the domestic
authorities requires DM and Swiss franc Eurobonds to be issued in the foreign bond markets
in Germany and Switzerland, respectively.

220

deposits in different currencies that were the chief source of
exchange rate pressures in the 1970s. These flows involve the bond
markets of the major currency countries and of a number of smaller
industrial countries as well.
Conclusion
The shift to floating exchange rates and international capital
mobility in the 1970s changed the framework of monetary policy
analysis for the United States and its trading partners. The greater
openness of trade and finance has increased the international influences on the U.S. economy, heightened the sensitivity of U.S. financial markets to foreign developments, and altered the timing and
path of the transmission of policy.
These changes have affected both the transmission and formulation of U.S. monetary policy. While the basic concerns of U.S. monetary policy—sustainable economic growth and price stability—are
unaltered, greater openness has changed the interpretation of domestic indicators, introduced new risks in monetary strategy, and
underscored the importance of international considerations in the
creation of policy. The techniques of monetary policy implementation, however, have been little affected by international flows.
Theoretical and empirical insights into the international aspects
of U.S. monetary policy need to be developed more fully. Economists
are still assessing the impact of changing foreign financial markets
and foreign monetary policy techniques on domestic financial indicators and still exploring the real and financial linkages between the
domestic economy and the rest of the world. Thinking on such central
issues as the determination of exchange rates has yet to crystallize
into a consensus or a clear rivalry among competing theories. As
these issues are clarified, they should help us to understand how
U.S. policies affect conditions outside the United States and in turn
are affected by them—in the commodity markets, in other industrial
countries, and in the less developed countries—and how current economic problems are likely to evolve in the future.




221

Box: Foreign Exchange Market Intervention
Intervention in the foreign exchange markets by the U.S. Treasury
and the Federal Reserve is undertaken to restore orderly conditions
in the exchange markets and at times to influence exchange rates.
As explained in Chapter 5, the U.S. Treasury is primarily responsible
for foreign exchange policy, but developments in the foreign
exchange markets have important ramifications for U.S. financial
and economic conditions and therefore for monetary policy and the
Federal Reserve. Decisions to intervene in the foreign exchange
market are made jointly by the U.S. Treasury and the Federal
Reserve, and the Foreign Exchange Desk at the Federal Reserve
Bank of New York conducts the operation.24 Intervention policy in
the second half of the 1980s has been influenced significantly by
coordination efforts of the governments and central banks of the
Group of Seven countries.25
Foreign exchange intervention by a central bank primarily affects
exchange rates by affecting the psychology of the foreign exchange
market. It may lead participants to reassess their assumptions about
the relative risks of short or long positions in foreign currency. It
may also have a direct effect on supplies and demands in the short
run, but to do so the volume of the intervention would generally
have to be relatively large. The enormous expansion of transactions
volume in the foreign exchange market (upward of $500 billion a
day in 1989) has greatly enhanced the importance of tactics and
timing in achieving a desired impact. In some cases, intervention
may also signal a willingness to alter monetary policy to achieve
an exchange market objective, and in some countries, it can serve
as a policy tool for altering short-term interest rates.
As Chapter 6 explained, purchases and sales of foreign currencies
by the central bank generally involve an exchange of domestic currency reserves with the banking system and thus add or drain
reserves. If the central bank (or treasury) offsets the full change in
the monetary base produced by foreign exchange market intervention, then the intervention is sterilized. Sterilized intervention is a tac-




24 A good discussion of foreign exchange interuention practices is contained in Roger M.
Kubarych, Foreign Exchange Markets in the United States, Federal Reserve Bank of New
York, 1983. Since substantial exchange reserves accumulated in the early 1980s, most
exchange intervention has been financed by drawing down or building up reserves rather
than by activating the swap network.
25 The Group ofSeven countries are the United States, Canada, France, Germany, Italy, Japan,
and the United Kingdom.

222

tic used to influence market psychology and to signal central bank
concerns. Its success depends on the readiness of market participants to interpret it as an indication of central bank policy resolve.
If the central bank (or treasury) offsets only part of the change
in the monetary base, then the remainder is unsterilized intervention. Unsterilized intervention is a joint policy action involving both
foreign exchange intervention and a monetary policy change. Since
unsterilized intervention induces changes in the money supply and
short-term interest rates of the intervening country, private residents
and nonresidents have additional incentives to alter their investment
and borrowing decisions.
The monetary effects of foreign exchange intervention by the
United States are routinely offset under Federal Reserve operating
procedures (see Box B of Chapter 6). The FOMC can and occasionally does change its monetary policy stance in response to the
same factors that inspired the exchange market intervention, but
two separate decisions are involved. The intervention is never passively permitted to change reserves. Nor is intervention undertaken
as a way of changing reserves, since domestic open market operations can be arranged when needed.
In many other countries, however, the central bank does not automatically offset intervention in the exchange markets. On a technical level, open market operations in domestic securities are often
limited by thin domestic financial markets. Hence, the operations
cannot be as large or as frequent as they are in the United States.
Some central banks may not have the domestic tools to achieve
domestic goals and must operate in the exchange markets. Indeed,
many foreign central banks use foreign exchange operations as an
alternative to domestic operations for monetary policy purposes,
although they generally only do so when the likely foreign exchange
effects are consistent with their policy objectives. They may choose
to operate in the exchange markets under some circumstances
because larger, more flexible operations may be feasible or because
the sectoral and inflation consequences of intervention may be preferred to a domestic money market operation.26
Because foreign exchange intervention can be an important signal
of central bank intentions, market participants try to detect and interpret intervention when it occurs. While the United States reports peri26 See Bank for International Settlements, Exchange Market Intervention and Monetary Policy,
Basle, 1988.




223




odically on its foreign exchange intervention, few other countries
release much information. As a result, market participants estimate
dollar-related intervention from the growth of foreign exchange
reserves abroad or the size of investments by foreign monetary
authorities in the United States. These indicators can be misleading.
Increases in official reserves include nondollar reserves. For
example, central banks participating in the exchange rate mechanism of the European Monetary System (EMS) have at times intervened substantially in EMS currencies to maintain agreed-upon
exchange rate relationships. Most countries do not disclose the currency breakdown of their reserves, although the International
Monetary Fund publishes periodic estimates in its Annual Report.
The proportion of official financing of the U.S. current account
deficit changes not only as foreign central banks accumulate dollar
reserves, but also as foreign central banks shift the composition of
investments of existing reserves among instruments. Official
financing is the increase in the dollar reserves invested in the United
States plus reductions in the reserves of the United States. Central
bank investment of dollars tends to be confined to a relatively narrow
but expanding spectrum of high-quality, highly liquid instruments.
These instruments have traditionally consisted of Treasury securities, deposits or custody accounts at commercial banks in the
United States, and usually a minimum working balance at the
Federal Reserve, but increasingly they include Eurodeposits and
other eligible Eurodollar instruments such as Eurocommercial
paper or Eurobonds issued by governments or supranational agencies. Foreign central banks may shift their reserves between investments in the United States and those in the Eurodollar markets for
pure portfolio considerations even in periods of little intervention.

224







10
Experiences of the 1980s
United States monetary policy during the 1980s was by some
measures extraordinarily successful. Inflation was brought down
from the frightening levels that had developed at the end of the previous decade to more manageable, though still disquieting, levels.
After a major recession early in the decade, the economy grew
without significant interruption for a number of years, achieving the
longest recorded peacetime expansion.7
Although gratified by the slowdown in the rate of inflation, monetary policymakers were troubled by the breakdown of the traditional relationships among money, economic activity, and prices
during the 1980s. Money growth had served as a policy guide and
as an advance indicator of inflation during the 1970s and the early
part of the 1980s. When it became apparent during the 1980s that
the demand for money was shifting, the Federal Reserve began to
look for alternative measures to help direct its policy actions. The
search, still going forward, has yielded many possible guides,
including specific economic, employment, and price reports,
movements of M2 around trend, and the shape of the yield curve.
These indicators may sometimes help policymakers forecast turns
in economic activity or increases in inflation, but none of the indicators has proved to be completely reliable.
In retrospect, the breakdown of the reasonably reliable relationship between Ml and economic activity in the 1980s is in some
ways understandable. At the beginning of the decade, many
observers had predicted reductions in the demand for money as
sophisticated electronics made close monitoring of money balances easier and encouraged daily investment of excess cash.
Instead, money demand increased. Indeed, Ml growth rates
between 1982 and 1986 were similar to those that had produced
accelerating inflation in the 1970s, but they were accompanied by
dramatic slowing of the rate of inflation. A portion of the rapid
growth of money in those years stemmed from the greater willingness of people to hold wealth in forms included in the money measures. Because both actual and prospective inflation had been
reduced by the tight monetary policy pursued between 1979 and
1982, money was not losing value so quickly. At the same time,
interest rate deregulation made it possible to earn reasonably
attractive returns on some types of liquid money balances,
increasing the inducement to hold savings in the form of money
and reducing the incentives to manage cash balances intensively.
1 The dating of business cycles by the National Bureau of Economic Research begins in 1854.

226

After 1986, money growth slowed even as inflation accelerated
moderately. The increase in money demand resulting from the earlier disinflation had about run its course. Market interest rates once
again began to rise. Depository institutions often delayed increasing their offering rates on some types of deposits, thus reintroducing incentives to economize on cash balances. Meanwhile,
corporations still had the means to monitor and adjust cash positions carefully, by using the techniques that had been developed a
decade earlier when inflation and market interest rates were high.
Recent experience suggests that money demand will continue to
be sensitive to interest rates and expected inflation.
At this stage, it is an open question whether the increased understanding of the factors affecting the demand for money will allow
any of the monetary aggregates to become more reliable guides to
appropriate policy once again. M2 appears at this writing to have
the most stable track record over time. Short-run sensitivities to
changes in market interest rates are substantial, however, even
though M2 displays little sensitivity to rate levels over extended
time periods. Furthermore, M2 is more difficult than Ml for the
Federal Reserve to control.
The absence of a clear and widely recognized advance indicator
of inflation in the wake of the shifting relationships among money,
economic activity, and prices has challenged Federal Reserve
monetary policymaking. The Federal Reserve has to maintain its
vigilance against frequent pressures to produce a more inflationary
policy than is desirable over the long run. In an era when the public
wants more from the government than it is willing to pay for
through taxes, there may be considerable clamor for policies that
seem to promote expansion and thus reconcile diverse and possibly inconsistent demands placed on the political system. In the
end, however, such policies are only likely to achieve higher rates
of inflation.
As long as memories of the costs of the rapid inflation of the
1970s are relatively fresh, the Federal Reserve should receive substantial support for holding to a course aimed at promoting price
stability. But the risk will grow that the Federal Reserve will be pressured to follow more inflationary policies as memories of the 1970s
experience fade. A clear and widely recognized leading indicator of
inflation in the form of a definition of money or some other variable
could help to anchor the process of controlling inflation and make
it easier for the Federal Reserve to resist pressures for inflationary
actions. Nonetheless, even in the absence of a good leading indi-




227




cator of price developments, the Federal Reserve so far has met
with reasonable success in feeling its way toward price stability.
The Federal Reserve was able to achieve reduced inflation in the
1980s and to help promote a long period of economic expansion
despite a number of adverse financial developments that sometimes complicated its task. The total amount of public and private
sector debt outstanding grew during the decade to levels beyond
what many observers believed to be wise or sustainable. The rising
debt of the CJ.S. federal government led the increase, and a number
of other forms of debt also grew rapidly. Many observers have been
especially troubled by growth in areas that may be hard hit by
defaults in an economic downturn. Corporations with credit ratings
low enough to have limited their access to the debt markets in earlier times issued substantial amounts of debt in the 1980s.
Another source of concern has been persistent U.S. trade
deficits. To date, political stability in the United States and the perception that the country offers many good investment opportunities have eased the financing of those deficits. But many observers
believe that the cumulative buildup of foreign ownership of U.S.
debt could mean that even a modest reduction in confidence in the
U.S. economy could lead to disruptive capital flows.
A number of banking and financial crises developed during the
decade, requiring prompt Federal Reserve action to contain them
and restrict their impact to the immediate area of difficulty. The
breadth of the financial markets, both within the United States and
across international borders, has had the positive effect of encouraging the efficient use of resources. At the same time, however, it
has meant that a disturbance in one financial market can be quickly transmitted to others both nationally and internationally.
Policymakers also had to cope with the repercussions of the
debt crises in developing countries. Excessive debt burdens taken
on by a number of countries in the 1970s when credit was easily
obtained and inflation was believed to be headed ever higher
proved unmanageable when dollar prices slowed their climb and
prices of oil and some other commodities fell. By the middle of
1982, many of these countries were suffering serious difficulties in
meeting their financial obligations, putting pressure on large
banks in the United States and other industrialized countries and
jeopardizing prospects for growth of the world economy. The
unfolding Mexican debt crisis in 1982, which was followed by similar crises in other developing countries, was a factor contributing
to the Federal Reserve's decision to move to a more accommoda-

228

tive monetary policy stance at that time—although domestic economic developments remained the chief factors determining the
monetary policy actions. While the crises have abated, the problems remain, requiring ongoing broad-based international cooperation throughout the decade.
Further, a number of banks and thrift institutions in the United
States suffered serious losses from their domestic operations during the 1980s. The collapse of Penn Square Bank of Oklahoma in
1982 had a wider impact than would normally follow the failure of
a modest-sized bank because Penn Square had arranged loan participations with several large banks; Continental Illinois National
Bank, one of the largest banks in the country, was seriously weakened. In 1984 it faced runs in the form of withdrawals of large uninsured deposits, a development which set in motion the biggest
rescue ever by the Federal Deposit Insurance Corporation. Further
oil price declines in 1986 and a related collapse in real estate prices
in some of the oil-producing areas in the United States weakened
a number of the regions' banks to the point where assisted mergers
and takeovers had to be arranged.
In 1985, fraudulent practices by a securities firm brought down
an Ohio thrift institution that was insured by a state-sponsored
fund. The fund could not handle the claim, an event which set off
runs on privately-insured thrift institutions in Ohio and in Maryland.
The Ohio insurance fund collapse proved to be a prelude to the
troubles of the Federal Savings and Loan Insurance Corporation
(FSLIC). The real estate loan crisis in the Southwest that followed
the second set of declines in oil prices in 1986, together with mismanagement and alleged fraud at some institutions, created insolvencies at many thrifts. Forced mergers, restructurings, and
closings weakened FSLIC, bringing its own solvency into question,
and it had to be restructured by the federal government.
In 1987 the crash in stock prices brought back fears of a global
economic and financial collapse. The Fed, however, took forceful
actions in response, temporarily making more credit available to
the banking system until the crisis had clearly passed. Hence, both
the policy actions and the outcome stood in sharp contrast to the
unfortunate events that followed the 1929 stock market crash.
The banking and financial problems that developed in the 1980s
underscore the need for steady vigilance on the part of the Federal
Reserve. Like all central banks, the Federal Reserve must promote
stability in the financial markets in order to limit the possibility of
financial crisis and must discourage excessive risk taking by those




229




responsible for making credit judgments. It must also take seriously its role as lender of last resort when crises do develop.
If the decade has presented the Federal Reserve with a number of
difficult problems, it has also demonstrated the effectiveness of the
System's primary monetary policy tool. Open market operations
have repeatedly proved to be an extraordinarily useful and flexible
device. As long as there are active markets in high-quality financial
instruments, the Federal Reserve should have the means to make
purchases and sales to achieve desired reserve adjustments.
A second tool, required reserve changes specifically designed to
make policy "tight" or "easy," has been absent from the Federal
Reserve's repertoire in the 1980s. Its absence has gone virtually
unnoticed, suggesting it was not much missed.
The System's third tool, discount window borrowing, has been a
source of controversy for many years. The current system of management holds the discount rate below market rates while strongly
discouraging banks from borrowing. One advantage of present
procedures is that imposing administrative guidelines enables the
Federal Reserve to indicate to banks that have borrowed that it is
attentive to their reserve management practices. Moreover, the
procedures have usually functioned reasonably well as part of the
mechanism for imposing varying degrees of monetary restraint.
Many economists, however, have questioned features of the current discount window procedures. They have criticized the implicit
subsidy in a rate that is generally below short-term market rates—
even though the amounts borrowed under the seasonal and adjustment credit programs typically have been quite modest in recent
years—and the consequent need to limit access to the window
through nonprice rationing. They have been concerned because
the administrative guidelines may at times be more or less restrictive than intended. Observers within the Federal Reserve have
recently noted apparent shifts in the demand for borrowing that
have not been entirely understood, as well as the short-run variability in the relationship of borrowing to the spread between the
Federal funds rate and the discount rate. It would be helpful to the
policy process if borrowing were more predictable in that key relationship. Nonetheless, changing the guidelines for discount window borrowing to address one set of problems could create new
problems for other aspects of the policy process. Any such
changes should not be considered without careful study.
Of course, taking a long view, further changes in many aspects
of policy will inevitably occur, for the U.S. financial system is

230

dynamic and the Federal Reserve must adapt to a changing world.
Some of the information in this book will no doubt become obsolete before long; a few points may even be outdated before the
book rolls off the presses. Nonetheless, much of the information
should remain relevant for several years. The book has been
designed to highlight the full range of factors that enter into the task
of making and executing monetary policy. It is hoped that the reader will have gained some sense of how the policy process operates
and how it has evolved in the United States during the Federal
Reserve System's first 75 years.




231




Notes