View original document

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

Review
Vol. 69, No. 3




March 1987

5 Confronting M onetary Policy
Dilem m as: The Legacy o f
H om er Jones
9 Predicting Interest Rates: A
Com parison o f Professional and
Market-Based Forecasts
16 Changes in Wealth and the Velocity
o f M oney

The Review is published 10 times per year by the Research and Public Information Department o f the
Federal Reserve Bank o f St. Louis. Single-copy subscriptions are available to the public free o f charge.
Mail requests for subscriptions, back issues, or address changes to: Research and Public Information
Department, Federal Reserve Bank o f St. Louis, P.O. Bojc 44Z, St. Louis, Missouri 63166.
The views expressed are those o f the individual authors and do not necessarily reflect official
positions of the Federal Reserve Bank o f St. Louis or the Federal Reserve System. Articles herein may be
reprinted provided the source is credited. Please provide the Bank's Research and Public Information
Department with a copy o f reprinted material.




Federal Reserve Bank of St. Louis
Review
M arch 1987

In This Issue . . .




In the first annual Homer Jones Memorial lecture, reprinted in this Review,
Beiyl Sprinkel, chairman of the President’s Council of Economic Advisers, exam­
ines the dilemmas currently facing monetary policymakers. Sprinkel notes that
the aberrant behavior of M l velocity — the relationship between the M l money
stock and economic activity — has diminished its usefulness as an intermediate
target of monetary policy and made the consequences of monetaiy policy actions
less certain.
Sprinkel points out that, regardless of the current problems they face, monetaiy
policymakers must avoid policy actions that rekindle inflation. Moreover, mone­
tary policy must not be asked to resolve problems, such as the large federal deficit
and the large U.S. trade deficit, that monetary policy can not effectively address.
Sprinkel concludes by emphasizing that Homer Jones had a major influence on
monetary economics over the past several decades. Homer’s distrust of policy
fine-tuning and his introspective, no-nonsense, analytical approach is the legacy
for which he will be remembered.
*

*

*

Interest rate forecasts receive considerable attention in the financial and
popular press primarily because the public presumes that expert forecasters
possess superior information about forthcoming events and better knowledge
about how these events will affect credit markets. In the second article in this
Review, “Predicting Interest Rates: A Comparison of Professional and MarketBased Forecasts,” Michael T. Belongia compares the relative performance of
interest rate predictions of professional forecasters to those that are easily
obtainable from readily available market data. He finds that the market-based
interest rate forecasts are as accurate, on average, as those obtained from financial
experts.
*

*

*

In the third article in this Review, “Changes in Wealth and the Velocity of
Money,” G. J. Santoni describes an economic theory that shows how increases in
wealth relative to income can produce reductions in the income velocity of
money. He then considers whether the atypical behavior that income velocity has
exhibited in recentyears can be attributed to this cause. Santoni examines various
measures of wealth and shows that, with the sole exception of a stock market
measure of wealth, they did not increase significantly relative to current income
from 1982 through 1985. Moreover, even though the stock market wealth measure
has risen relative to current income since 1982, the behavior of this ratio over
longer periods does not appear to be related to the behavior of velocity. Thus, the
evidence suggests that the decline in the income velocity of money since 1981 can
not be attributed to increases in these measures of wealth relative to current
income.

3




MARCH 1987

FEDERAL RESERVE BANK OF ST. LOUIS

First Annual Homer Jones Memorial Lecture

Confronting Monetary Policy
Dilemmas: The Legacy of
Homer Jones
Beryl W. Sprinkel

I t is an honor to deliver this first annual lecture in
memory of Homer Jones. I first became acquainted
with Homer when writing my thesis at the University
of Chicago, and I found some of his writings to be
particularly useful. When Homer later became Direc­
tor of Research at the St. Louis Federal Reserve Bank, it
was — like many things in life — not particularly
momentous in itself, but the implications for mone­
tary economics were certainly important. In his price­
less style, Harry Johnson described Homer Jones as
"... an oasis in the desert that Keynesian economics
and concern with credit had made of the Federal
Reserve System, [and] the last outpost of classical
monetary civilization in a cancerous culture of barbar­
ian bumptiousness.” Only an academic, of course,
could say something like that — and about an era that
fortunately has long passed at the Federal Reserve.
Homer Jones should be remembered for many
things, not the least of which is the many people
whose intellectual development he shaped and whose
professional lives he fostered. He was one of Milton
Friedman's first teachers — not in economics, but in
insurance and statistics. Milton credits him for provid-

Beryl W. Sprinkel is the Chairman of the Council of Economic Advisers.
His speech, given on March 26, 1987, at Washington University in St.
Louis, is the first annual Homer Jones Memorial Lecture, which was
cosponsored by the Federal Reserve Bank of St. Louis, the Center for
the Study of American Business at Washington University, and the St.
Louis chapter of the National Association of Business Economists.



ing the inspiration that sparked his initial interest in
economics, as well as something more tangible —
getting him a scholarship to attend the University of
Chicago. And, of course, Homer had a strong influence
on the professional lives of the many economists who
worked for him in his years at the St. Louis Fed.
Homer had an intense respect for the market sys­
tem; that permeated both his economic analysis and
his views about economic policy. His basic policy
prescriptions in macroeconomics reflected this freemarket orientation: a distrust of the efficacy of finetuning and a fundamental belief in the inherent stabil­
ity of a free market economy. His reliance on the
market approach to problems also extended to inter­
national issues, labor market issues and regulatory
policy. From my perspective, the extent to which such
principles have become more generally accepted as a
basis for public policy decisions is remarkable, not
only in the United States, but in other countries as
well. Both as an Undersecretary at Treasury and as
CEA Chairman, I have been involved, along with of­
ficials from other governments, in policy discussions
on issues ranging from agriculture to tax reform. In
governments around the world, there is a greater
recognition of the efficiency of the market system in
pricing goods and allocating resources. While much
progress can still be made toward improving public
policy analysis and discussion, the movement toward
greater reliance on market forces is one I applaud, as I
am sure Homer would as well.

5

FEDERAL RESERVE BANK OF ST. LOUIS

One particular area where we have made substan­
tial progress by relying on market forces is in the de­
regulation of financial markets and institutions. Regu­
lations on interest rates paid by financial institutions
to their depositors have been eliminated. Restrictions
on competition within classes of financial institutions
and between different classes have been reduced. In
this area, however, more needs to be accomplished,
and I suspect that Homer would share my desire to
see rapid progress on the Administration’s proposals
for further financial market deregulation.
It was difficult to be around Homer without learning
a great deal from him. He had a remarkable ability to
focus on the practical issues and an impatience with
intellectual pretense and academic irrelevancy. His
technique was to put questions to you — always
pertinent questions, frequently penetrating questions,
sometimes relentless questions. In so doing, he forced
you to understand and articulate what you knew,
while discovering what you did not know. He had a
truly unusual ability to stimulate you to search for the
answers. In the St. Louis Fed Research Department, I
am sure that many promising ideas were hatched,
many empirical relations were tested, and many in­
fluential articles resulted directly from Homer’s in­
quiring mind and his ability to transmit that interest to
others.
The products of Homer Jones’ style and approach at
the St. Louis Fed are well known and well respected.
The weekly and monthly publications of the Research
Department, which have now become standard refer­
ences for eveiyone from undergraduates to White
House officials, were initially Homer’s products. The
St. Louis Fed Research Department became one of the
most prominent in the country and its monthly Re­
view became widely respected and earned the stature
of a professional journal. The metamorphosis of the
Research Department, its role in promoting policyrelated research and in providing an alternative point
of view within the System was what Karl Brunner has
labeled “a remarkable institutional event,” made
more remarkable and more influential because it oc­
curred within the System itself.
Given the nature of Homer Jones’ legacy, it is ironic
— and perhaps fitting — that we are gathered here to
honor his contributions at a time when there are so
many unanswered questions about the conduct of
monetary policy. The policy issues we face today are
different from those debated by Homer. Most analysts
now accept the important role of monetary policy in
economic performance. Most economists acknowl­

6 FRASER
Digitized for


MARCH 1987

edge an important relation between changes in
money growth and economic activity, although in
recent years there has been much more uncertainty
about the precise form of that once-reliable relation­
ship. Few doubt, at least in general terms, the long-run
link between money growth and inflation. Rather than
those fundamental issues that we debated in the 1950s
and 1960s, the policy challenges of today relate to the
changed environment in which monetary policy is
now conducted.
In the four years since this expansion began, there
have been substantial changes in both the institu­
tional and economic environment in which monetary
policy must be designed and implemented. These
developments are well known to this audience. The
inflation rate — excluding the effects of the oil price
declines in 1986 — has been cut to one-third the 1980
rate. Similarly, interest rates are one-third to one-half
their 1980 levels. Financial deregulation has changed
the institutional structure in which monetary policy is
conducted. In this decade, the introduction of NOW
and money market accounts has significantly altered
the composition of the monetary aggregates, and the
relaxation of restrictions on deposit interest rates has
led to the inclusion in M l of interest-bearing deposits
which pay market-determined rates of return.
These developments — and possibly others — ap­
pear to be affecting the basic relation between money
and nominal GNP growth as indicated by the behavior
of the “velocity” of money. Specifically, while there
have always been sizable fluctuations in velocity from
one quarter to the next, over longer periods velocity
rose at a reasonably predictable rate of about 3 per­
cent per year between 1947 and 1981. Since the cycli­
cal peak of 1981, however, velocity has declined at
more than a 3 percent annual rate.
There are a number of plausible explanations for
this decline in velocity. However, with the limited data
available, it is difficult to reach definitive conclusions.
To my knowledge, the most promising lines of em­
pirical research attempt to relate velocity declines to
the decline in inflation and interest rates and to their
effect on the interest-elasticity of the demand for
money. In the recent period of declining interest rates,
the opportunity cost of holding the highly liquid bal­
ances in M l has fallen, thereby raising desired M l
balances and suppressing velocity. As market interest
rates change, the public response in terms of moving
in and out of M l balances is difficult to predict. In part
this is because we have relatively little experience with
deregulated deposit rates and also because it is not

FEDERAL RESERVE BANK OF ST. LOUIS

clear how depository institutions will adjust deposit
rates to changes in market rates. This implies contin­
ued uncertainty about the future behavior of velocity.
Over most of this expansion we have had monetary
growth — particularly in M l — that, based on the
historical relation with nominal spending and in­
flation, would be viewed as excessive. Yet, we have not
had the short-run surge in real growth and nominal
spending that would be expected from such high M l
growth. We are therefore left with a difficult dilemma
about the implications of recent M l growth for future
inflation. On that issue, a wide range of opinion exists.
Some forecasters — many of whom are long-time
friends of mine — foresee a major resurgence of in­
flation resulting from the monetary growth of the past
two years. Other analysts discount recent M l growth
as being the result of financial deregulation, disin­
flation, declining interest rates, or some combination
of such factors.
It is interesting to note, however, that even those
who rely most heavily on money growth as a forecast­
ing tool are not predicting an inflation as high as
would be implied by historical velocity behavior. The
Shadow Open Market Committee, for example, fore­
casts inflation and nominal GNP growth consistent
with the assumption that velocity growth remains well
below its postwar trend growth path. Neither the most
recent Blue Chip forecasts nor the Administration’s
economic projections reflect the expectation that re­
cent M l growth will be translated into spending and
inflation in accordance with historical velocity behav­
ior. In fact, I know of no serious, current forecast that
does not implicitly assume continued atypical veloc­
ity behavior, at least over the coming year.
These and related questions have made the con­
duct of monetary policy particularly difficult over the
course of this expansion. It is my judgment that in the
context of considerable uncertainty about velocity
growth, the Federal Reserve has done a reasonably
good job balancing the risk of renewed inflation
against the risks associated with too little money
growth. I do not believe, however, that we can afford to
be complacent about a long continuation of the
money growth we have experienced in recent years.
The Reagan Administration is committed not just to
reducing inflation, but to the ultimate goal of restoring
price stability. By distorting price signals and eroding
productive incentives, inflation is a powerful deterrent
to long-term real growth and job creation. Moreover,
high inflation ultimately brings the high costs of re­
ducing the inflation rate — costs that our economy



MARCH 1987

paid in the recession of 1981-82 and that are still being
paid in such sectors of our economy as agriculture
and energy. Given the inevitable costs associated with
reducing inflation and the importance to long-term
prosperity of keeping inflation under control, it would
be a policy blunder to allow inflation to reaccelerate.
In assessing monetary policy, it is important to
recognize what it can and cannot accomplish. It can­
not smooth out all short-term fluctuations in output,
employment, or the price level. Nor can it sustain real
growth rates that consistently exceed the economy’s
potential — as determined by underlying rates of
productivity and population growth and trends in
labor force participation. Monetary policy, however,
can deliver reasonable stability of the price level in the
longer run and can avoid being an additional impor­
tant cause of disturbances to output and employment
growth in the shorter run.
Monetary policy has contributed to the success we
have enjoyed in resolving the critical problems that
confronted the U.S. economy when President Reagan
assumed office. The annual inflation rate has been cut
by two-thirds — from double digit levels in 1979-80 to
about 4 percent for the past four years. Interest rates
generally have fallen to about one-third of the levels of
six years ago. The economy is now in the 52nd month
of what will soon become the longest peacetime ex­
pansion since World War II. As this expansion has
proceeded, in contrast with the experience in earlier
expansions, the inflation rate and interest rates have
shown no tendency to rise and to bring about the
strains that led to the ends of earlier expansions. Thus,
the destructive sequence of business cycles with pro­
gressively rising inflation and interest rates has been
broken, and the foundation has been laid for sustain­
able real growth with moderate inflation.
The problems that remain in the U.S. economy are
not primarily problems that can be addressed with
monetary policy — beyond its normal role in gradu­
ally moving toward the goal of long-run price stability,
while avoiding being a source of macroeconomic dis­
turbance. In particular, the critical and related prob­
lems of the large federal deficit and of the large U.S.
trade deficit cannot be resolved by monetary policy.
The federal government has a deficit because the
share of federal spending in GNP has risen well above
the average share that federal revenue has maintained
in GNP for three decades. The solution is to restrain
the absolute growth of federal spending, while eco­
nomic growth raises the absolute level of federal
revenues.

7

FEDERAL RESERVE BANK OF ST. LOUIS

The United States has a trade deficit because we as a
nation spend more than the value of what we produce.
To finance this excessive spending, we import capital
from the rest of the world in an amount that corre­
sponds to our current account deficit. Excessive fed­
eral spending and corresponding federal borrowing
are an important part of the problem — and reversing
them is an important part of the solution. So too are
stronger, internally generated growth and more open
trade policies on the part of our trading partners. We
require a coordinated approach to reducing interna­
tional payments imbalances in an environment that
maintains world economic growth.
I could discuss further the problems of our fiscal
and trade deficits, as well as other problems of the U.S.
economy. However, my experience even before I went
to Washington taught me brevity is a virtue — perhaps
a virtue even more appreciated by audiences than by
speakers.
Among the things that I have learned in Washington
— and there are many — one of the most important is
how simple things look from the outside, but how
much more difficult it is when you actually have to
take action and assume the responsibility for its effect
on peoples’ well-being. In policymaking, things are
seldom simple. Certainly in the macroeconomic field,
where policy tools are blunt and forecasts are fre­
quently wrong, there are risks associated with any
policy decision. Ultimately, policymakers must face
the question: What are the consequences if I am
wrong? If nothing else, it is a humbling experience.
No one in my memoiy had learned this lesson better
than Homer Jones. His humble and unpretentious
personal style was reflected in his professional ap­
proach: take nothing for granted and believe only what
can be justified by the data. So what would Homer
have to say about the current dilemma? I like to tell my
staff— some of them think I tell them too often — that
I’m from the “Show-Me” State. I want to see the data to
support a conclusion. While Homer wasn’t bom in
Missouri as I was, he certainly adopted the show-me
attitude about economic issues. Knowing his insis­
tence that policy be based on empirically tested rela­
tions, he surely would share the concerns about high

8



MARCH 1987

money growth over the past few years. He surely
would not easily discard long-term empirical relation­
ships. But I also doubt that he would counsel ignoring
current developments as they have varied dramati­
cally from historical patterns.
Given the aberrant behavior of velocity over the past
four years or so, policymakers have little alternative
but to supplement the information provided by the
monetaiy aggregates with other relevant data. To me,
this implies looking in addition at interest rates, ex­
change rates, sensitive prices such as gold and other
commodities, forward markets, and measures of real
economic activity for signals as to the meaning and
implication of money growth and monetaiy policy
actions. The limitations and deficiencies of these data
as guides to monetary polity are great and are well
known, and I will not recount them here. It is not an
ideal approach, but I see no workable alternative at the
present time. To date, I know of no completely satis­
factory explanation of what has happened to velocity.
When more time has passed in a deregulated and lowinflation environment, I am confident that reliable
relationships will re-emerge, which I trust can be
identified by appropriate empirical testing. In the in­
terim, policy decisions must be made that properly
balance the risks to the economy of alternatively too
much or too little money growth. As a nation, we
cannot afford the pain and disruption of allowing
inflation to resuige, nor can we afford to risk the
economic consequences of excessive monetary
restriction.
In a sense, the dilemmas and frustrations of today’s
policy issues lead those of us who knew Homer Jones
to plead, “Homer, where are you when we need you?”
For today, we surely could use his quiet, reasoned
assessment of the issues.
Many people accomplish important things in their
lives. I wonder whether there are not more important
things to be remembered for than what you invented,
discovered, wrote, or built. It may be a more lasting
legacy to be remembered for how you influenced the
thinking and accomplishments 'of others. Among
those of us who call ourselves monetary economists,
few can claim that legacy as readily as Homer Jones.

MARCH 1987

FEDERAL RESERVE BANK OF ST. LOUIS

Predicting Interest Rates:
A Comparison of Professional
and Market-Based Forecasts
Michael T. Belongia

In te re st rates have varied substantially in recent
years. Since 1981, for example, the monthly average
three-month Treasury bill rate has ranged between
5.18 percent and 16.30 percent while the Baa corpo­
rate bond rate ranged between 9.61 percent and 17.18
percent; the prime rate during this time reached a
high of 20.5 percent and fell to a low of 7.5 percent.
Interest rate movements are important, of course, be­
cause they affect the present value of streams of future
payments, that is, wealth. Moreover, the risk of interest
rate changes is related directly to the level of interest
rates.1During the 1980s, therefore, firms and individ­
uals have faced substantial exposure to interest rate
risk.
There are at least two approaches that can be taken
to reduce the magnitude of this problem. The first is to
hedge interest rate risk, which has been discussed at
length in this Review and elsewhere.2The second is to
forecast the likely course of interest rates. This article
investigates the reliability of such forecasts in general
and assesses the specific usefulness of forecasts by
professional economists.

Michael T. Belongia is a senior economist at the Federal Reserve Bank
of St. Louis. Paul Crosby provided research assistance.
'Interest rate risk, for a firm whose portfolio is composed of streams
of future receipts and payments, is measured by the interest elastic­
ity of the portfolio; for a single asset, this can be expressed as -n(i/
1 + i), where n is the term to maturity. A more general expression for
a portfolio of assets and liabilities is derived in Belongia and Santoni
(1987). In either case, the level of interest rate risk rises with the
interest rate.
2See Belongia and Santoni (1984,1985).



INTER EST RATE FORECASTS:
THEORY AND EVIDENCE
Given the popular attention that such forecasts
command, it is surprising to note what economic
theory says about them: they are unlikely to provide
accurate insights about the future. This argument is
stated clearly by Zarnowitz:
It might be argued that these are fo re c a sts of people
who study the economy (experts), which are quite
unlike the expectations of those who act in the econ­
omy (agents). On the one hand, the experts are usually
credited with more knowledge of the economy at large
than the agents have. On the other hand, the experts
are often charged with being less strongly motivated to
predict optimally than the agents who are seen as
having more at stake.3

Economists, at least on one level, lack sufficient incen­
tives to make forecasts that are more accurate than
information already available in the marketplace.
Moreover, previous studies have shown there is little
systematic difference among professional forecasts, at
least partly because they “use to a large extent the
same data, receive the same news, interact, and draw
upon a common pool of knowledge and techniques.”4
The key issue, however, really is not whether ex­
perts have more (or better) information than the pub­
lic, but whether individuals who consistently can fore­

se e Zarnowitz (1983), p. 2.
4See Zarnowitz (1986), p. 6, and the references cited therein.
9

FEDERAL RESERVE BANK OF ST. LOUIS

MARCH 1987

cast interest rates more accurately than the market are
likely to make their forecasts public. The reason has to
do with individual self-interest. Quite simply, why
would anyone reveal valuable insight about the future
when he could increase his wealth directly by appro­
priately trading in financial markets using this infor­
mation?

rent spot rates or prices in the interest rate futures
markets. This model, known as the efficient markets
model, states that the expected interest rate at some
specified future point in time, given all information
presently available, is equal to the current interest rate
plus whatever change in the interest rate is suggested
by currently available information.8

If, for example, a person knew that the three-month
Treasury bill rate would be 6.50 percent in December,
while the futures market currently priced it at 7.00
percent, the forecaster's wealth gain would be limited
only by his ability to buy December Treasury bill
futures; in this example, he would make a profit of
$1,250 on every contract he could buy.5 Certainly, he
has no incentive to make the same forecast public
without appropriate compensation, at least until he
had taken as large a position in the market as he could.
Of course, forecasters may have incentives to sell fore­
casts that are of no value to their wealth; it is not clear,
however, why other individuals would pay for such
predictions.

The driving force behind the efficient markets
model is the information available to traders in the
market and the incentives they have to use this infor­
mation. Current market rates and expectations of fu­
ture rates are influenced by changes in information
that affect expectations about the future. Because new
information is unknown until it actually is released,
success in predicting future interest rates depends
upon predicting both future changes in the informa­
tion and the market’s reaction to such “news.”

As a general rule, the accuracy of economic fore­
casts varies widely across variables. Previous research
has found that predictions of the three-month Trea­
sury bill rate six months into the future by major
commercial forecasters are within two percentage
points of the actual rate only 67 percent of the time.6
Thus, if in June, the three-month Treasury bill rate
was forecast to be 7 percent in December, there is only
a 0.67 probability that the actual December rate would
be somewhere between 5 percent and 9 percent.
Other studies have shown that error statistics often
double in size when the forecast horizon is extended
as little as from one to two quarters ahead.7

The Efficient Markets Hypothesis and
Interest Rate Forecasts
A model of interest rate determination demon­
strates why individuals are unable (as opposed to
unwilling) to forecast interest rates more accurately,
on average, than the forecasts already implied by cur-

5Treasury bill futures are priced by subtracting the Treasury bill
interest rate from 100. Thus, interest rates of 7.00 and 6.50 percent
imply contract prices of 93.00 and 93.50, respectively. Moreover,
each basis-point change in the interest rate is worth $25 on the
value of a contract. Buying one contract at 93.00 and selling at 93.50
would show a simple profit of 50 basis points x $25 = $1,250,
abstracting from commission and other costs.
6McNees, p. 11.
Typically, the criterion is root-mean-squared error (RMSE); see
McNees (1986). Also, see Zarnowitz (1983).
10



An Illustration o f the Efficient Markets
Model
One illustration of the efficient markets model ap­
plied to actual data is the change in interest rates that
follows the weekly Federal Reserve M l announcement
that usually occurs at 4:30 p.m. [EST] each Thursday.
The assumption is that the interest rate at 3:30 p.m.,
just prior to the announcement, fully reflects all cur­
rently available information relevant to the Treasury
bill rate, including various forecasts of the Fed’s yet-tobe-announced change in M l; thus, the available infor­
mation at 3:30 p.m. includes both actual and predicted
data.
When the Fed announces the M l change at 4:30
p.m., the market’s information set is revised with the
actual M l change replacing its predicted value. If no
other significant information is released until rates are
observed again at 5 p.m., the change in the Treasury
bill rate from 3:30 to 5 p.m. reflects the market’s reac­
tion to the news in the M l announcement. If the
actual and predicted M l values are different, the ef­
ficient markets model predicts that interest rates will
react to the new information in the Fed’s M l an­
nouncement; many studies have found this result
empirically.9

8The efficient markets model applied to interest rate determination
can be expressed as:
E(it+1in t) = it(1 + E(it+1-i,in,)),
where E is the expectations operator and fl, is the information
available to agents at the time forecasts are made. For more detail
on this model, see Fama and Miller (1972) or Mishkin (1983).
9See Sheehan (1985) and Belongia and Sheehan (1987) for a survey
and critique of these studies.

FEDERAL RESERVE BANK OF ST. LOUIS

This example demonstrates the major point of the
efficient markets model: changes in interest rates de­
pend on changes in information. A forecast that inter­
est rates will be higher six months from now than
what already is implied by the underlying term struc­
ture really is a forecast that new information will be
revealed which will cause market participants to raise
the rate of interest. Such forecasts are potentially use­
ful only if the forecasters consistently have better
information, on average, than the other market partici­
pants generally possess. Or, to state the proposition
differently, a useful forecast is not simply an accurate
one; it also must tell something about the future that is
not already reflected in current market interest rates.

A COMPARISON O F IN TEREST RATE
FORECASTS
A comparison of alternative interest rate forecasts is
essentially a comparison of information sets that fore­
casters possess. The futures market, as well as fore­
casts that simply assume the future will resemble the
present, provide useful alternatives to forecasts pro­
duced by specialized forecasting services. If all fore­
casts have similar accuracy, it would suggest that
market participants use essentially the same informa­
tion.

Survey Forecasts
The information content of economists’ forecasts is
intriguing for a variety of reasons. Presumably, their
specialized training gives them insight to the workings
of financial markets. In return for their services, the
economists involved earn relatively large salaries;
moreover, some command considerable public atten­
tion. The latter group should include those whose
forecasts are among the best of competing alterna­
tives.

Market Forecasts
The futures market offers an interesting perspective
on forecasts. At a given point in time, individuals may
enter into agreements to buy or sell interest-sensitive
assets, such as Treasury bills, at a date as much as two
years into the future. The collective actions of inves­
tors betting that interest rates will rise from today’s
level (who will sell Treasury bill futures short) and
investors betting that interest rates will fall (who will
buy, or go long in, Treasury bill futures) determine, at
each moment in time, the “market’s” expectation of
what interest rates will be at a specified future date.
Such forecasts are interesting for two reasons: they
reflect all available information held by market partici­



MARCH 1987

pants and these participants have a compelling rea­
son to forecast accurately. If they are wrong, the
money lost is their own!
A naive or no-change model is an interesting third
alternative because, as previously noted, predicting
interest rates really involves predicting changes in
information and the market’s reaction to this news. If
one believes it is impossible to predict actions by
OPEC, changes in macroeconomic policy, revisions in
economic data and other factors that affect expecta­
tions of future interest rates, the best strategy would
be to predict no change in information and, hence, no
change in interest rates. Certainly, as the length of the
forecast horizon grows shorter, the probability of large
changes in information (and interest rates) declines as
well.

Sources o f Forecasts: Professional and
Market Data
The six-month-ahead forecasts of the three-month
Treasury bill rate by nine economists surveyed regu­
larly by the Wall Street Journal were collected over the
period December 1981 through June 1986. These fore­
casts, which are published on or about each January 1
and July 1, yielded 10 forecast periods and 90 predic­
tions to be evaluated. Each forecast was assumed to be
made the day before publication.10
Comparable forecasts from the futures market were
derived by observing on June 30 the three-month
Treasury bill rate implied by the December Treasury
bill futures contract and on December 31 the rate
implied by the June contract. A larger sample to be
used later also employed observations on the March
futures contract from the previous September 30 and
on the September contract from March 31. These data
were compared with actual Treasury bill rates on the
day the relevant futures contract ceased trading." The
procedure yielded 40 observations, of which 10 coin­
cided with dates of the economists’ forecasts. The
naive or no-change forecast was obtained by observing
the spot Treasury bill rates on the last business days of
March, June, September and December and predict­
ing that same rate would exist on the last day of the
month six months hence. Again there are 40 observa-

,0The full Wall Street Journal survey includes many more economists,
but only nine individuals have responded consistently since the
initial survey in December 1981.
"Treasury bill futures contracts usually are liquidated in the third
week of their terminal months, not the last day of the month as with
the economist forecasts.
11

FEDERAL RESERVE BANK OF ST. LOUIS

MARCH 1987

C hart 1

Treasury Bill Rates: Actual a n d Predicted
Forecast from
6 months pr eviou s

Fo rec as t from
6 months previous
(Percent)

(Percent)

13

12
11

10
9
8

7

Date

tions over the 1977-86 interval with 10 coinciding with
dates of the economist survey. Although this sample of
market-based forecasts includes only 10 observations
that coincide with the economists’ forecasts, it serves
as the basis for the first comparison. Subsequent anal­
ysis uses the entire sample back to 1977 for a stronger
test of forecast accuracy.

Forecasts o f Direction o f Change
A first assessment about the accuracy of the profes­
sional forecasts was made against a relatively weak
criterion, the predicted direction of change. That is, if
rates were forecast to increase (or decrease), did they?
The individual forecasts relative to subsequent actual
values are plotted in chart 1.
Digitized for
12FRASER


The 90 individual expert predictions correctly fore­
cast the direction of change on 38 occasions, or 42
percent of the time. If interest rate movements are
random, a 50 percent record of accuracy would be
expected.12 Only one of the nine forecasters guessed

,2This type of performance — the strategies of professional investors
yielding returns inferior to those of simple rules — is common. For
example, the mean equity fund managed by professional institu­
tional money managers rose 16.7 percent in 1986 compared with an
18.7 percent rise in the S&P 500 index. Moreover, more than 67
percent of the money managers produced returns in 1986 smaller
than the general increase in market values, as measured by the
S&P 500; see Wallace (1987). For a more extensive discussion of
this result and a similar finding of inferior performance by mutual
fund managers over time, see Malkiel (1985), pp. 147-82, and the
references to his chapter 7.

FEDERAL RESERVE BANK OF ST. LOUIS

MARCH 1987

Table 1
Summary Statistics for Errors from Alternative Forecasts: June 1982-December 1986
MAE
RMSE
Mean error
2.056
Economist individual forecasts
1.625
-0.406
90
Economist mean forecast
1.889
10
1.550
-0.406
Futures market forecast
1.466
-1.132
2.253
10
-0.543
1.859
10
Naive forecast
1.321

the direction of change correctly more than one-half
of the time; he was correct on six of 10 occasions.
Three others guessed the correct direction of change
on five of 10 occasions. The worst individual perfor­
mance was two correct predictions.
For the 40 quarterly predictions derived from fu­
tures market observations, 22, or nearly 55 percent,
correctly forecast the direction of change. Over the
shorter 1982-86 sample, five of 10 directions of change
were predicted correctly by the futures market. On the
simple criterion of direction of change, the futures
market outperforms the economists surveyed.13

Point Forecasts
A different criterion by which to evaluate forecasts is
a comparison of the point estimates of the predicted
changes in interest rates with the actual changes.
These comparisons were analyzed several ways. First,
forecasts by the nine experts provided 90 individual
predictions of the Treasury bill rate. These individual
predictions also could be aggregated to form a con­
sensus, or average, prediction for the nine economists
at a specific moment in time. The performance of the
experts relative to the futures market and naive fore­
casts first was judged over the short 1982-86 sample
that coincided with the economist survey. Differences
between actual Treasury bill rates and, respectively,
the economist, futures market and naive forecasts
were calculated to generate values for forecast errors.
All errors were calculated as actual minus predicted
values. Table 1 shows the summary statistics for these
errors.

13There is no meaningful way to construct a direction-qf-change
criterion for the naive forecast.



The entries in table 1 represent the mean absolute
error (MAE), mean error and root-mean-squared error
(RMSE) from forecasts for the three-month Treasury
bill rate six months into the future. The first two rows
are associated with the individual and consensus fore­
casts from the survey of experts. The third row is
based on the differences between the actual Treasury
bill rate and the futures market prediction. The fourth
row is based on the naive predictions, the differences
between current and previous actual rates.
The most interesting aspect of these summary sta­
tistics is their remarkable similarity. Of course, this
result was predicted by the earlier theoretical discus­
sion, which emphasized that all available information
would be reflected in current market rates. The mean
errors for all forecasts are negative, indicating that
these methods tended to overestimate the interest
rate; the futures market, however, tended to be the
most bearish forecaster on this account by overpre­
dicting the Treasury bill rate an average of 1.132 per­
centage points. MAE statistics also are similar, with a
range of about 30 basis points between the best (naive)
and worst (individual economist). The RMSE statistic,
which is a measure of the dispersion of forecast errors,
shows the naive and economist consensus to perform
best.14

14The likely explanation for the futures prediction having the highest
RMSE is the method of calculation. The RMSE will tend to be lower
for forecasts that made many errors of a similar size relative to
forecasts that had smaller errors, on average, but had several very
large errors. This result occurs, of course, because calculating the
RMSE involves squaring the forecast errors. The effects of random
variation in small samples also is a potential source of distortion.
Thus, two very large futures market errors offset a record of gener­
ally accurate forecasts as indicated by other statistics.
13

MARCH 1987

FEDERAL RESERVE BANK OF ST. LOUIS

Table 2
Market-Based Forecasts Over a Longer Horizon: March 31,1977 - December 31,1986
Futures market forecast
Daily data
Weekly averages
Naive forecast
Daily data
Weekly averages

MAE

Mean Error

RMSE

n

1.676
1.702

-0.163
-0.141

2.589
2.634

40
40

1.740
1.788

0.035
0.027

2.578
2.695

40
40

Longer Sample Results fo r
Market-Based Forecasts
Error statistics from the longer 10-year sample of
quarterly observations described earlier are reported
in table 2. Because daily interest rate changes are
volatile and a large, one-day change could affect the
results, forecasts for a specific date also were com­
pared with the average Treasury bill rate for the week
in which that date occurred.
Belative to the previous results, the futures market
average errors declined substantially to near 15 basis
points, compared with the shorter sample mean error
of about 113 basis points. MAE and RMSE values in­
creased slightly, however, for the longer sample. The
forecast errors do not appear to vary with the use of
daily or weekly average values for the terminal period
spot rate. The naive forecast also shows slight in­
creases in MAE and RMSE values but its mean error
falls about 50 basis points to near zero. Again, while
these statistics are not directly comparable with the
economist forecasts because of the different sample
periods, nothing in them suggests superior perfor­
mance by the economists.

Market Reaction to Forecasts
As a final check on the information content of the
expert forecasts, daily Treasury bill rates were divided
into two groups: those for days when the experts’
forecasts were published and those for other trading
days. (Recall that the forecasts are useful to the market
only if they add to the existing pool of market informa­
tion.) To test whether this is true, equation (1) was
estimated:
(1 ) TB, = 0.015 + 0.998 TBm + 0.049 ANNOUNCEMENT + e„
(1.02) (657.2)
(0.95)
R2 = 0.99
DW = 1.77
Digitized for14
FRASER


where the daily value of the Treasury bill rate (TB,) is
regressed on the previous day’s value (TB,.,) and a
dummy variable (ANNOUNCEMENT) that takes a value
of one on the 11 days that the expert forecasts were
released.15 If the expert forecasts add to the market’s
information, the coefficient for the ANNOUNCEMENT
variable should be significantly different from zero; as
the t-statistic of 0.95 reveals, however, we cannot reject
the hypothesis that the forecast announcements have
no effect on Treasury bill rates. Apparently, the Trea­
sury bill market had already incorporated the infor­
mation underlying these forecasts prior to their public
release.

SUMMARY
Interest rate risk has been substantial in the 1980s,
and, by no coincidence, the demand for interest rate
forecasts has increased. There are strong theoretical
reasons to believe, however, that such forecasts are
subject to large errors. Moreover, anyone who could
predict interest rates more accurately, on average,
than other market participants would have no reason
to make his forecasts publicly. Comparisons of inter­
est rate forecast errors support the notion that several
market-based forecasts, using information easily ac­
cessible to the general public, predict the Treasury bill
rate six months into the future as well as a panel of
prominent forecasters.
Why, then, do economists make public forecasts of
interest rates and seemingly earn large salaries for
doing so? Several explanations related to other pri­
mary functions of corporate economists seem plausi­
ble. First, economists may serve an advertising func­
tion for their firms: they are paid, in part, to get the

15lt is possible to use the January 3,1987, survey for this estimation.

FEDERAL RESERVE BANK OF ST. LOUIS

firm’s name mentioned in the press often, and fore­
casting interest rates is one way to achieve this end.
Second, economists may provide a managerial insur­
ance function. If a business decision has the potential
to cause large losses, managers who have relied on the
input of economists cannot be held negligent, in the
sense of acting without seeking “the best information
available at the time.’’ Finally, forecasting interest rates
may be a trivial portion of an economist’s overall
function; his compensation may be based primarily
on analytical performance in other areas. It is unlikely,
however, that economists are employed primarily for
their ability to predict interest rates more accurately
than the market.

REFERENCES
Belongia, Michael T., and Gary J. Santoni. "Hedging Interest Rate
Risk with Financial Futures: Some Basic Principles,” this Review
(October 1984), pp. 15-25.
_________“Cash Flow or Present Value: What’s Lurking Behind
That Hedge?” this Review (January 1985), pp. 5-13.
_________“ Interest Rate Risk, Market Value, and Hedging Finan­
cial Portfolios,” Journal of Financial Research (Spring 1987), pp.
47-55.




MARCH 1987

Belongia, Michael T., and Richard G. Sheehan. “The Informational
Efficiency of Weekly Money Announcements: An Econometric
Critique,” Journal of Business and Economic Statistics (forthcom­
ing).
Fama, Eugene F., and Merton H. Miller. The Theory of Finance,
(Holt, Rinehart and Winston, Inc., 1972).
Granger, C. W. J. “Comment,” Journal of Business & Economic
Statistics (January 1986), pp. 16-17.
Litterman, Robert B. “A Statistical Approach to Economic Forecast­
ing,” Journal of Business & Economic Statistics (January 1986), pp.
1-4.
Malkiel, Burton G. A Random Walk Down Wall Street, 4th edition,
(W. W. Norton and Company, Inc., 1985).
McNees, Stephen K. “Forecasting Accuracy of Alternative Tech­
niques: A Comparison of U.S. Macroeconomic Forecasts,” Journal
of Business & Statistics (January 1986), pp. 5-15.
Mishkin, Frederic S. A Rational Expectations Approach to Ma­
croeconometrics: Testing Policy Ineffectiveness and Efficient-Markets Models (University of Chicago Press, 1983).
Sheehan, Richard G. “Weekly Money Announcements: New Infor­
mation and Its Effects,” this Review (August/September 1985), pp.
25-34.
Wallace, Anise C. “Funds That Put The Pros to Shame,” New York
Times, February 8,1987.
Zarnowitz, Victor. “ Rational Expectations and Macroeconomic
Forecasts,” NBER Working Paper No. 1070 (January 1983).
_________"The Record and Improvability of Economic Forecast­
ing," NBER Working Paper No. 2099, Cambridge, MA (December
1986).

15

FEDERAL RESERVE BANK OF ST. LOUIS

MARCH 1987

Changes in Wealth and the Velocity
of Money
G. J. Santoni

O
NE long-standing view among economists is
that the quantity of money in circulation and aggre­
gate income are closely related in the long run.1This
relationship, known as the income velocity of money,
is particularly important because it makes it possible
to determine the effect of changes in money growth on
income over extended periods.2
Unfortunately, this relationship has not behaved
well in recent years. Various investigators have at­
tempted to identify the reasons for this unusual be­
havior, focusing on institutional changes that allowed
the payment of interest on transaction deposits, the
rise in the trade deficit, and changes in tax rates.3 In
general, however, their results have been inconclusive.
This article discusses how changes in wealth can
affect velocity and considers whether the atypical be­

G. J. Santoni is a senior economist at the Federal Reserve Bank of St.
Louis. Thomas A. Poiimann provided research assistance.
'See, for example, Fisher (1963) who notes that: “This theory,
though often crudely formulated, has been accepted by Locke,
Hume, Adam Smith, Ricardo, Mill, Walker, Marshall, Hadley, Fetter,
Kemmerer, and most writers on the subject.” (p. 14) See also pp.
157-59 and 296-97. More recent examples are Friedman and
Schwartz (1963 and 1982). Thornton (1983) presents a nontechni­
cal discussion of the theory.
2Using monetary policy to hit short-run stabilization objectives is
problemmatical if not impossible. See Thornton (1983) and Mankiw
and Summers (1986), p. 419, for discussions of this point.
3Rasche (1986), Mankiw and Summers (1986), Tatom (1983), Taylor
(1986), Siegel and Strongin (1986) and Kopcke (1986) represent
some of the recent attempts to resolve the issue.
Digitized for16
FRASER


havior that velocity has exhibited in recent years can
be attributed to changes in wealth.4

VELOCITY: A MEASURE O F THE
RELATIONSHIP BETWEEN MONEY
AND SPENDING
The most commonly used measure of the relation­
ship between income and the stock of money is the
income velocity of money. It is the ratio of GNP to M l
(the sum of currency in the hands of the public and
checkable deposits).
Chart 1 plots the income velocity of money from 1/
1959 through III/1986. As indicated, this measure has
risen fairly steadily throughout most of the period.
Before 1982, the growth rate of income velocity was
remarkably stable, averaging about 3 percent per year.5
Hence, the average annual growth rate of GNP ex­
ceeded the average annual growth rate of the quantity

4See Knight (1941), Friedman (1956) and Meltzer (1963) for exam­
ples of this argument.
5The average annualized growth rate was 3.13 percent with a stand­
ard deviation of 4.03. The standard deviation is a measure of the
variation in velocity growth around its average. Short-run changes in
velocity growth have been attributed to cyclical factors, changes in
the pattern of receipts and payments, financial innovations and
changes in the nominal interest rate. See, for example, Fisher
(1963), pp. 58-73, Tatom (1983) and Thornton (1983), p. 10.

MARCH 1987

FEDERAL RESERVE BANK OF ST. LOUIS

Chart 1

Income Velocity

of money by about 3 percentage points.6 In recent
years, however, velocity growth has changed consid­
erably. As chart 1 indicates, velocity generally has
declined at an annual average rate of about 3.0 percent
since the end of 1981.7
If monetary policymakers were certain that the
long-run average growth in velocity had changed per-

manentlyfrom + 3 percent per year to —3 percent per
year, they could, once again, determine the impact of
any given long-run growth in M l on GNP. There is
considerable uncertainly, however, about whether the
change in velocity’s average growth is permanent or
only temporary. Identifying the reasons for the recent
declines might help resolve this issue.

VELOCITY AND MONEY DEMAND
6lf V is the income velocity of money while Y and M are GNP and the
quantity of money, then V = Y/M. This equation can be written in
growth rate form as below. The dots over the variables indicate
compounded annual growth rates.
V = Y- M
Since V averaged about 3 percent before 1982, Y exceed M by
about 3 percentage points on average.
The break in velocity growth has been dated at the end of 1981. See
Rasche (1986), pp. 2 and 8. The average annualized growth rate in
velocity during 1/1982-111/1986 was -2 .9 percent with a standard
deviation of 5.67.



Velocity relates the equilibrium level of income to
the equilibrium quantity of money; the latter depends
importantly on the quantity of money demanded.8
Momentarily ignoring other things that may influence
people’s choices, money demand theory states that
the demand for money is proportionally related to

'See Friedman (1956), p. 4.
17

MARCH 1987

FEDERAL RESERVE BANK OF ST. LOUIS

some scale variable, either income or wealth. The
transaction theory of money demand uses the flow of
current income as the scale variable while the portfo­
lio approach to money demand uses the stock of
wealth.9

The Transaction Approach to
Money Demand
The transaction approach presumes that money is
held to support current spending and that current
spending is closely related to current income. This
theoiy relates the demand for money (MD) to current
income (Y) by some proportion (k). In equilibrium,
since the quantity of money demanded is equal to the
quantity supplied (MD= Ms= kY), the ratio of income to
the quantity of money (income velocity) is equal to the
inverse of this proportion (V = Y/Ma= Y/kY = 1/k). If cur­
rent income rises, desired spending rises in propor­
tion; consequently, people will want more money to
facilitate their increased spending.
The transaction approach has an important advan­
tage from an empirical point of view. The data on
income are relatively good and readily available.
What’s more, numerous empirical tests of the theory
have been conducted, and the empirical relationships
between money and income have performed well dur­
ing certain time periods.10There have been occasions,
however, when they have broken down. Various ana­
lysts have pointed to breakdowns in the mid-1960s,
when velocity fell unexpectedly, in the mid-1970s,
when it rose unexpectedly, and, in recent years, when
it has fallen again unexpectedly.11

A Portfolio Approach to Money Demand
An alternative theoiy of money demand suggests
that the quantity of money balances that people hold
is related more closely to their wealth than their cur­
rent income.12Money is simply one of many assets in

which wealth may be held. The desired mix of assets
that make up wealth depends on both the net benefits
of holding wealth in the various forms and risk prefer­
ences. The portfolio theoiy states that an increase in
wealth is associated with an increase in the quantity of
money people want to hold and vice versa.
Again, ignoring other factors that may influence
choices, this theoiy says that the demand for money is
a constant proportion (0) of wealth (W).13 In equilib­
rium, the quantity of money demanded is equal to the
quantity supplied; thus, the ratio of wealth to money
(wealth velocity) is equal to the inverse of this propor­
tion (W/M = W/0W = 1/0).

The Difference between the I\t o
Both theories of money demand agree that certain
variables, such as short-term interest rates, popula­
tion, the pattern of receipts and payments, the tech­
nology of the payment system and risk preferences,
are important for money demand. They differ, how­
ever, in regard to the scale variable.
If current income were always a constant propor­
tion of wealth, there would be no substantive em­
pirical difference between the two theories. In this
case, 0 and k would differ by a constant factor that
reflects the ratio of income to wealth [V = 1/k = Y/M =
(W/MMY/W) = (1/0MY/W)]. If income is not a constant
proportion of wealth, however, and if the portfolio
theoiy of money demand is correct, income velocity
will fluctuate whenever current income changes rela­
tive to wealth. If current income rises relative to
wealth, for example, the income velocity of money will
rise also, other things the same. The reverse move­
ment in the income velocity of money would occur if
wealth rises relative to current income.

Two Important Conditions
The above discussion indicates that two conditions
must hold if changes in wealth relative to income are
important in explaining the decline in velocity since

9The transaction approach includes other variables besides current
income in the money demand function. These are mentioned below.
See Laidler (1985), pp. 49-97, for a more complete discussion of the
various approaches to money demand.
10See Laidler (1985), pp. 117-34.
"Few people dispute the importance of the recent breakdown. There
is some disagreement, however, regarding the significance of the
earlier breaks. See Judd and Scadding (1982), Laidler (1985), pp.
135-51, and Rasche (1986), p. 7, for a discussion of the earlier
breaks.
,2See Knight (1941) and Friedman (1956), pp. 4-5. Knight, for exam­
ple, argues that “The economic process in a pecuniary economy
involves the holding or owning, by somebody, of wealth — all the
Digitized for18
FRASER


wealth of the economy — and also the entire stock of money. Hence
every property owner has the alternative either of holding money up
to the amount of his fortune or of choosing the concrete kind of
wealth other than money he will hold.” (p. 210)
13The portfolio theory does not necessarily imply a constant propor­
tional relationship between money and wealth. This is an empirical
question. See Laidler (1985), p. 58. There is evidence that the
relationship is proportional for at least some wealth proxies. See, for
example, Mankiw and Summers (1986) and Meltzer (1963). It is
assumed to be proportional above for illustrative purposes.

FEDERAL RESERVE BANK OF ST. LOUIS

the end of 1981: 1) the demand for money must be
more closely related to wealth than to current income
(that is, wealth is the appropriate scale variable); 2)
wealth must have risen relative to current income
since 1981.
Although the choice of the appropriate scale vari­
able is still an unresolved issue among economists, it
is useful to determine whether the second condition
holds. If wealth has not risen relative to current in­
come since 1981, whether the first condition holds or
not is irrelevant; we can conclude that changes in
wealth do not help explain the recent decline in veloc­
ity. On the other hand, if wealth has risen relative to
current income, resolving the first issue becomes
more important.

WHAT IS WEALTH?
To see why the ratio of current income to wealth
may vaiy, it is helpful to understand how they are
related.
An individual’s wealth is the market value of his net
assets; this market value is found by adding together
the present values of all his assets and subtracting the
sum of the present values of all his liabilities. This
difference is equal to the present value of the expected
stream of net receipts (income minus expenses). In the
simplest case, it is the expected net income flow
divided by the long-term interest rate.14Thus, an indi­
vidual’s wealth at any time depends on both the ex­
pected future flow of net income and the relevant
long-term interest rate.

The Effect o f Current Income on Wealth
and Vice Versa
Current income is the actual amount of income
received each period. Because unanticipated events
influence the income actually received, current in­
come generally differs from the income expected for
any period. The difference between current and ex­
pected income is called transitory income.
Since wealth is the present value of the expected
future income flow, transitory income has only a small

,4See Fisher (1954), pp. 12-13, and Friedman (1956), pp. 4-5.
Fisher, for example, defines wealth (or capital value) as “simply
future income discounted or, in other words, capitalized. The value
of any property, or rights to wealth, is its value as a source of income
and is found by discounting that expected income... The bridge or
link between income and capital is the rate of interest." (emphasis in
original)



MARCH 1987

effect on wealth. For example, suppose a person re­
ceives a surprise Christmas bonus of $2,000. If the
person’s annual income is $20,000, the bonus is 10
percent of current income, a fairly large percentage. If,
however, the person does not associate the bonus
with a change in his future income prospects, and his
wealth before the bonus was $200,000, the effect of the
bonus on his wealth is relatively small (1 percent of
wealth).
Another way to view this is to note that the individ­
ual’s "permanent income” is not much affected by the
bonus. Permanent income is the amount of consump­
tion that can be sustained without changing wealth.
Permanent income and wealth are closely related.15In
the above example, if the person consumed $22,000 in
the year the bonus is received, he would necessarily
have to reduce his consumption in the following year
or draw down his wealth, other things the same.
Suppose the person in this example is promised an
increase in his annual salary beginning some time in
the future. His wealth increases immediately upon
learning of the prospective raise, other things the
same, and his expected stream of future income is
now higher, even though his current income does not
yet reflect the raise. A decline in the interest rate
induces a similar increase in wealth, other things the
same, because it increases the present value of the
unchanged stream of expected future income; current
income, again, is unchanged.
While these examples refer to individuals, the argu­
ment applies to the whole community as well. Unex­
pectedly good harvests, favorable relations between
unions and management, or tranquil foreign relations
can produce positive transitory components of in­
come for the whole community. A reduction in trade
barriers or changes in the tax laws that result in more
productive use of resources can raise the expected
future flow of aggregate income and, thus, raise wealth
relative to current income. Finally, a decline in the
level of interest rates can raise the present value of a
given expected future flow of aggregate income; this
increase in wealth occurs without affecting the cur­
rent or expected future levels of income.

15See Friedman (1956), p. 5. For a perpetual stream of income that is
expected to increase at a constant rate (p), wealth is
w = y0 (1 +p)/(r-p),
where y0 is the initial income receipt and r is the real interest rate
(r>p). Permanent income is y* = rw. Wealth and permanent income
are constant across time as long as r and p are constant and saving
is zero. Expected income in period n, however, is y„ = y0 (1 +p)".
19

FEDERAL RESERVE BANK OF ST. LOUIS

As the above discussion suggests, the ratio of cur­
rent income to wealth generally is not a constant.
Thus, if wealth is the appropriate scale variable for
money demand, velocity will vary as the ratio of cur­
rent income to wealth varies.

MEASURING NATIONAL WEALTH
Since we are concerned with the relationship be­
tween wealth and society’s demand for money, we
need to establish a concept of national wealth. Na­
tional wealth is simply the aggregate wealth of the
nation’s residents.16There are two theoretically equiv­
alent methods of measuring national wealth: the in­
come and balance sheet methods.

The Incom e Approach
National wealth, in theory, can be measured by
discounting the expected stream of net national in­
come by the appropriate interest rate. Some practical
problems must be dealt with, however, when applying
this method of measuring national wealth. One obvi­
ous problem is that the expected stream of income is
not directly observed. Only current and past incomes
are known. Thus, practical applications of the income
method must depend on good estimates of the ex­
pected stream of net national income.
Many studies have used univariate time-series
methods to estimate the expected stream of future
income.17 Roughly, time-series models account for
patterns in past movements of a particular variable
(national income, in this case) and use the information
contained in the pattern to predict future values of the
variable. In a sense, a time-series model is a sophisti­
cated method of extrapolation.18
While these models are a useful estimating tool,
they have a serious drawback. When using them, the
investigator must assume that the underlying eco­
nomic structure that generated the observations will
remain unchanged during the period of analysis. For
example, a time-series model is not designed to fore­
cast changes in the stream of future income that are
produced by significant technological changes, insti­
tutional changes such as a major shift in the tax law, or

16See Goldsmith (1968), p. 51.
17See Laidler (1985), pp. 88-90.
18See Pindyck and Rubinfeld (1981), p. 470.
Digitized for20
FRASER


MARCH 1987

significant changes in relative supplies such as pro­
duced by OPEC production quotas.
Another problem concerns the interest rate that is
appropriate for discounting expected national in­
come. National income is the sum of wages, rents and
profits. Wage income, which accounts for about 75
percent of national income, is produced by human
capital while nonhuman capital is the source of rents
and profits. Unfortunately, the interest rate that is
relevant in discounting the expected stream of wage
income (iH) is not observable; moreover, because hu­
man capital is not as liquid as nonhuman capital, iHis
probably higher than the interest rate that applies to
income produced by nonhuman capital (ik).19 If ex­
pected wages are discounted at the lower rate ik,
national wealth will be overstated.20 Of course, per­
centage changes in the wealth estimate will not be
distorted as long as the ratio of iHto ikdoes not change.
Empirical estimates that depend on a constant rela­
tionship between these two interest rates, however,
will produce misleading results whenever this ratio
changes substantially.

The Balance Sheet Approach to
Estimating Wealth
Some investigators have estimated national wealth
using the balance sheet approach.21 This measure is
obtained by summing the present values of the assets
owned by U.S. residents and subtracting the sum of
the present values of all the liabilities owed by U.S.
citizens.
When these assets and liabilities are aggregated, all
claims of one U.S. citizen against another U.S. citizen
cancel out. Since most liabilities of U.S. citizens are
owed to other U.S. citizens, these liabilities and their
asset counterparts disappear from the aggregation.
What remains is national wealth. It includes nonreproducible and reproducible tangible assets such as
land, buildings, structures, machinely, vehicles, con­
sumer durables, inventories (of raw materials, work in

,9See Friedman (1956).
“ The only interest rates that are directly observable are those that
apply to financial instruments. Consequently, neither iH or ik are
directly observable. The capital asset pricing theory developed in
the finance literature, however, can be used to produce estimates
of ik.
21See Goldsmith (1968). in theory, the balance sheet and income
approaches to measuring wealth are equivalent. But there are
numerous practical problems that arise. These are discussed
below.

MARCH 1387

FEDERAL RESERVE BANK OF ST. LOUIS

Should Government Bonds and the Money Stock Be
Included in Wealth?
Government Debt
The market value of domestic government debt
that is held by domestic citizens is sometimes in­
cluded in national wealth estimates. Presently,
however, there seems to be no agreement among
professional economists about whether it belongs
in the wealth aggregate.1 On one side of the argu­
ment, government bonds are like other financial
instruments, assets to the creditor and liabilities to
the debtor. The creditor’s wealth in government
bonds is equal to the present value of the expected
stream of future receipts of interest and principal.
The debtors are the taxpayers who must pay the
future taxes obligated by the bond interest and
principal payments. In theoiy, the present value of
the taxpayers’ liability offsets the creditors’ wealth
in the bonds. It is unclear, however, whether taxpayer&actually take account of the future tax liabil­
ity obligated by the bonds. If these liabilities are
wholly or partially ‘ignored, an increase in the
present value of outstanding government bonds
causes people to behave as if their wealth has
increased.

and Federal Reserve notes.2Outside money is typi­
cally included in national wealth since it is an asset
to the holder and is not offset by any domestic
taxpayer liability.
Inside money is the sum of checkable deposits
issued by depository institutions. These deposits
represent about 75 percent of M l balances. Though
these deposits are part of the wealth of the people
who hold the deposits, they must be redeemed by
the bank on demand of the depositor. Hecause of
this provision, some economists view these de­
posits as liabilities for the owners of depository
institutions. Consequently, in this view, inside
money is like other domestic financial instruments:
it represents a claim by one citizen on another, and
these claims cancel in computing national wealth.3

The Money Stock

Some investigators have argued that the distinc­
tion between inside and outside money is irrele­
vant. They suggest that checkable deposits are the
output of the banking system in the same sense that
autos are the output of the motor vehicle industry.
In their view, inside money, like autos, is a compo­
nent of national wealth.4

Outside money is money issued by sovereign gov­
ernments; it consists of notes and coins issued by
foreign governments, as well as U.S. Treasury coin

While economists generally agree that outside
money is part of wealth, there is no general agree­
ment about inside money.

'See Barro (1974), Kormendi (1983), and Thompson (1967).

2See Patinkin (1965), p. 15.
3See Patinkin (1965), pp. 295-96.
4See Pesek and Saving (1967).

process and finished goods), military assets, works of
art, human capital and net claims on foreigners.22
Most investigators who estimate national wealth by
the balance sheet method agree that the above items
belong in national wealth. There is some disagree­
ment, however, about which other things should also
be included. These controversial items are discussed
in the shaded box above.

“ See Goldsmith (1968), p. 52.



The balance sheet method of estimating national
wealth is costly because it requires an extensive inven­
tory of the nation's assets and liabilities. As a result,
estimates of national wealth that employ this method
are available only on an annual basis.
There are several other problem s with this
method.23 It requires that assets and liabilities be val­

“ See Goldsmith (1968), pp. 52-54, for a further discussion of the
practical problems of applying this method.
21

MARCH 1987

FEDERAL RESERVE BANK OF ST. LOUIS

Chart 2

Expected Income and Physical W ealth

ued at their market prices. Many assets, however, are
traded infrequently, if at all. For example, the dis­
counted value of people’s wages is never traded in
markets, yet this makes up a substantial part of eveiyone’s wealth. There are many privately held busi­
nesses, unique pieces of recil estate and personal
property that are infrequently traded; consequently, it
is difficult to obtain accurate assessments of their
market values.
In practice, measurements of national wealth based
on the balance sheet method depend on estimates of
market values. Reproducible assets are valued at their
replacement cost net of straight-line depreciation,
land holdings are valued at assessed market prices
and no estimates are made of the value of human
capital.24The exclusion of human capital means that
estimates of national wealth seriously understate the
actual wealth of the nation.

Z4See for example, “Balance Sheets for the U.S. Economy 1946-85,”
p. i.
Digitized for 22
FRASER


HAS WEALTH RISEN RELATIVE TO
CURRENT INCOME?
Empirical studies that test the portfolio theory have
used various empirical measures for national wealth.
In most cases, the studies have based the wealth
estimates on either the income or balance sheet ap­
proach to measuring wealth.25
Various empirical estimates of money demand have
used three different scale variables as proxies for na­
tional wealth: expected income, permanent income

“ Some studies have used “financial wealth” as an empirical measure
of national wealth. Financial wealth is the sum of "household and
business deposits and credit market instruments.” This measure
bears little theoretical relationship to the measures of national
wealth discussed above. It is largely composed of claims by one
U S. citizen on another, claims that cancel when national wealth is
calculated. As a result, it is difficult to interpret empirical estimates
that employ this wealth proxy, so they are ignored in the following
analysis. See, for example, Kopcke (1986), p. 19, and Friedman
(1978), p. 625, noteb.

FEDERAL RESERVE BANK OF ST. LOUIS

MARCH 1987

Table 1
Growth of Expected Income, Permanent Income Wealth and Current Income: 1959-85
(average annual percentage rates of change)
W
Yc
(Y«-Y°)
Period
Y*
Y"
(YP-Y')
(W —Yc)
8.56%
8.26%
- .02%
.30%
- .03%
1959-81
8.24%
8.23%
5.38
6.75
.04
-1.37
.90
1982-85
6.79
7.65
-.0 2
1959-85
8.02
8.01
8.08
8.03
-.01
.05
Ye s
Y>> =
W =
Yc =

Growth rate of expected income from the time series model
Growth rate of personal consumption expenditures — a proxy for permanent income
Growth rate of physical nonhuman wealth
Growth rate of current income

and physical nonhuman wealth taken from balance
sheet data.26These proxies are examined to determine
whether they have risen relative to current income
since 1981.
Chart 2 plots the logarithms of expected income (Ye),
permanent income (Yp) and physical nonhuman
wealth (W). The time-series model used to estimate
expected income is given in the appendix. Personal
consumption expenditures are used as a proxy for
permanent income.27 Finally, physical nonhuman
wealth is estimated from balance sheet data and in­
cludes an estimate of the market value of federal
government debt.2®
Chart 2 indicates that all three wealth proxies be­
have in much the same way over 1959-85. Each vari­
able rises in a smooth fashion at about the same
growth rate.

*See, for example, Meltzer (1963), Brunner and Meltzer (1963),
Chow (1966), Laidler (1966), Laumas and Spencer (1980), Mankiw
and Summers (1986) and Rasche (1986).
27lt has been suggested recently that “consumption is an ideal proxy
(for permanent income) since it is proportional to this unobserved
variable. Indeed, it has often been noted that the procyclical behav­
ior of the velocity of money is evidence for a permanent income view
of money demand, since the ratio of GNP to consumption is also
procyclical.” Mankiw and Summers (1986), p. 416. See, in addition,
Friedman and Schwartz (1982) who note that “income as measured
by statisticians may be a defective index of wealth because it is
subject to erratic year-to-year fluctuations, and a longer-term con­
cept, like the concept of permanent income developed in connection
with the theory of consumption, may be more useful.” (p. 38)
“ The data source for the wealth estimate is “Balance Sheets for the
U.S. Economy 1946-85.” It is Total Consolidated Domestic Net
Assets (line 10) minus U.S. Gold and SDR’s (line 8) plus the market
value of federal debt. The market value of federal debt is calculated
by the method suggested by Butkiewicz (1983). See also Seater
(1981).



The average annual growth rates of these variables
are presented in table 1 along with the average annual
growth rate of current income. The growth rates
across all four variables are virtually identical for the
1959-85 period.29
Changes in wealth help explain a decline in velocity
if wealth rises relative to current income. The data
presented in table 1 give no indication that this oc­
curred during 1982-85, the period of declining veloc­
ity. In fact, the growth rate of physical nonhuman
wealth actually fell relative to the growth of current
income from 1982-85. Thus, changes in this estimate
of national wealth do not help explain the decline in
the income velocity of money that began in 1982. Other
investigators have found similar results.30
The average growth rates of expected income (Ye)
and permanent income (Yp) are somewhat greater
than the average growth rate of current income (Yc)
during 1982-85. The averages of the quarterly differ­
ences between these growth rates and the growth rate
of current income, however, are not significantly dif­
ferent from zero in a statistical sense.31 The small
positive differences that are observed are likely the
result of chance variation in the data.
On net, then, table 1 indicates that none of the three
wealth measures rose significantly relative to current

“ This result is expected for the growth rates of Ye and Yc over long
periods. Recall that Ye is generated from a time-series model of Yc
which is a sophisticated technique for estimating the trend of Yc.
“ See, for example, Rasche (1986), pp. 50 and 94.
31The t-ratio is .33 for the average of the differences between Ye and
Yc for the 1982-85 period. The t-ratio is 1.12 for the average of the
differences between Yp and Yc for the 1982-85 period. Both are in­
significant at the 5 percent level.
23

FEDERAL RESERVE BANK OF ST. LOUIS

MARCH 1987

Chart 3

Velocity and Stock M arket M easu re s of Y / W
V, Y‘/W

income during the 1982-85 period. Consequently,
changes in these measures of wealth do not help
explain the decline in the income velocity of money
that began in 1982.

A Stock Market Wealth Measure
The conclusion that wealth has not increased rela­
tive to current income since 1981 appears to conflict
with the recent behavior of the values of common
stock. This narrow measure of wealth has received an
increasing amount of attention, especially as various
indexes of stock market values have risen to record
high levels.
Common stock values of publicly traded firms are a
precise measure of the capital values of the firms.
Chart 3 examines the behavior of one measure of the

24


V,

y ‘/ w

ratio of current income to the capital values of publicly
traded firms and compares this to the behavior of
income velocity over 1959-85. The measure used in
chart 3 is the ratio of current income to the market
value of stocks included in the Standard and Poor’s
500 composite stock index.32
Chart 3 indicates that both the ratio of current
income to stock market wealth and velocity have de­
clined from 1982-85. At first glance, it appears that the
recent decline in velocity may be the result of an
increase in wealth relative to current income.

“ Hamburger (1977 and 1983) uses the ratio of dividends paid (a
measure of the current income generated by the capital of the firms)
to the market value of stocks in his estimates of money demand.
Hamburger’s ratio behaves similarly to the ratio shown in chart 3.

FEDERAL RESERVE BANK OF ST. LOUIS

Looking at the entire period shown in chart 3, how­
ever, there does not appear to be a close relationship
between V and this income/wealth ratio. For example,
the ratio did not change much from 1959 to 1969 or
from 1978 to 1981, yet velocity rose. Except for the
years 1973-74 and 1977-78, the income to wealth ratio
appears to move "sideways” while velocity continu­
ously rises over the whole period until 1981. Although
other things that influenced velocity over the period
shown in chart 3 were no doubt changing, perhaps
markedly, it is still interesting to note that the simple
correlation coefficient between changes in the
income-wealth ratio and changes in velocity is .17,
which is not significantly different from zero.

SUMMARY
The income velocity of money — the ratio of GNP to
M l — has behaved differently since 1981 than it had
over the previous 30 years. This paper discusses the
portfolio approach to money demand, which suggests
that money demand is more closely related to wealth
than to current income. The portfolio theory implies
that, when wealth increases relative to current in­
come, income velocity falls, other things the same.
Therefore, if the theory is valid, a substantial increase
in wealth since 1981 would serve as a possible expla­
nation of the recent fall in velocity.
The paper examines the behavior of current income
relative to alternative measures of wealth. With one
exception, a stock market wealth measure, the wealth
measures examined here did not increase signifi­
cantly relative to current income during 1982-85.
Moreover, while the ratio of current income to the
stock market measure of wealth declined after 1982,
the behavior of this ratio over longer periods does not
appear to be related to the behavior of velocity. Thus,
the evidence suggests that the decline in the income
velocity of money since 1981 cannot be attributed
solely to an increase in these measures of wealth.

REFERENCES
Alchian, Armen, and William R. Allen. Exchange and Production:
Competition, Coordination and Control, 2nd ed., (Wadsworth,
1977), pp. 69-70,160-61, 163, 165.
“Balance Sheets For the U.S. Economy 1946-85” (Board of Gover­
nors of the Federal Reserve System, Washington, "D.C.).
Barro, Robert J. “Are Government Bonds Net Wealth?” Journal of
Political Economy (November/December 1974), pp. 1095-1117.
Brunner, Karl, and Meltzer, Allan H. “Predicting Velocity: Implica­
tions for Theory and Policy,” Journal of Finance (May 1963), pp.
319-54.



MARCH 1987

Butkiewicz, James L. “The Market Value of Outstanding Govern­
ment Debt: Comment," Journal of Monetary Economics (May
1983), pp. 373-79.
Chow, Gregory C. “On the Long-Run and Short-Run Demand for
Money,” Journal of Political Economy (April 1966), pp. 111-31.
Fisher, Irving. The Theory of Interest (Kelley and Millman, 1954), pp.
3-58, 61, and 206-27.
_________The Rate of Interest (Macmillan, 1907), p. 213.
_________The Purchasing Power of Money (Augustus M. Kelley,
1963).
Friedman, Benjamin M. “Crowding Out or Crowding In? Economic
Consequences of Financing Government Deficits,” Brookings Pa­
pers on Economic Activity (3/1978), pp. 593-641.
Friedman, Milton. “The Quantity Theory of Money — A Restate­
ment,” Studies in the Quantity Theory of Money (University of
Chicago Press, 1956), pp. 3-21.
Friedman, Milton, and Anna J. Schwartz. A Monetary History of the
United States, 1867-1960 (Princeton University Press, 1963).
_________Monetary Trends in the United States and United King­
dom, 1867-1975 (Chicago University Press, 1982), chapters 2, 5
and 6.
Goldsmith, Raymond W. "National Wealth,” International Encyclo­
pedia of the Social Sciences (Macmillan Co., 1968), pp. 50-59.
_________A Study of Saving in the United States (Princeton Uni­
versity Press, 1955).
_________The National Wealth of the United States in the Post- War
Period (Princeton University Press, 1962).
Hafer, R. W., and Scott E. Hein. “ Evidence on the Temporal Stabil­
ity of the Demand for Money Relationship in the United States,”
this Review (December 1979), pp. 3-14.
Hamburger, Michael J. “The Demand for Money in an Open Econ­
omy: Germany and the United Kingdom,” Journal of Monetary
Economics (January 1977), pp. 25-40.
_________“ Recent Velocity Behavior, The Demand for Money
and Monetary Policy,” Monetary Targeting and Velocity (Federal
Reserve Bank of San Francisco, 1983), pp. 108-28.
Johnson, Harry G. Essays in Monetary Economics (Harvard Univer­
sity Press, 1969), pp. v-viii and 161-78.
Judd, John P., and John L. Scadding. “The Search for a Stable
Money Demand Function: A Survey of the Post-1973 Literature,”
Journal of Economic Literature (September 1982), pp. 993-1023.
Knight, Frank H. “The Business Cycle, Interest, and Money: A
Methodological Approach,” Review of Economic Statistics (May
1941), pp. 202-26.
Kormendi, Roger C. “Government Debt, Government Spending,
and Private Sector Behavior,” American Economic Review (De­
cember 1983), pp. 994-1010.
Kopcke, Richard W. “How Erratic is Money Growth?" New England
Economic Review, Federal Reserve Bank of Boston (May/June
1986), pp. 3-20.
Laidler, David E. W. The Demand for Money: Theories, Evidence,
and Problems, 3rd ed., (Harper and Row, 1985).
_________"Some Evidence on the Demand for Money,” Journal of
Political Economy (February 1966), pp. 55-68.
Laumas, G. S., and David E. Spencer. "The Stability of the Demand
for Money: Evidence From the Post-1973 Period,” The Review of
Economics Statistics (August 1980), pp. 455-59.
Mankiw, N. Gregory, and Lawrence H. Summers. “Money Demand
and the Effects of Fiscal Policies,” Journal of Money, Credit and
Banking (November 1986), pp. 415-29.
25

MARCH 1987

FEDERAL RESERVE BANK OF ST. LOUIS

Meltzer, Allan H. “The Demand for Money: The Evidence from the
Time Series,” Journal of Political Economy (June 1963), pp. 21946.
“M1 Loses Status at the Fed,” New York Journal of Commerce
(June 23,1986).
Patinkin, Don. Money, Interest, and Prices, 2nd ed. (Harper and
Row, 1965).
Pesek, Boris P., and Thomas R. Saving. Money, Wealth, and Eco­
nomic Theory (Macmillan, 1967), pp. 72-102.
Pindyck, Robert S., and Daniel L. Rubinfeld. Econometric Models
and Economic Forecasts, 2nd ed. (McGraw-Hill, 1981).
Rasche, Robert H. “M1 -Velocity and Money Demand Functions: Do
Stable Relationships Exist?” Carnegie-Rochester Conference on
Public Policy, processed (November 21,1986).
Seater, John J. “The Market Value of Outstanding Government

Debt, 1919-1975,” Journal of Monetary Economics (M y 1981), pp.
85-101.
Siegel, Diane F., and Steven Strongin. “Can the Monetary Models
be Fixed?” Economic Perspectives, Federal Reserve Bank of
Chicago (November/December 1986), pp. 3-14.
Taylor, Herb. “What Has Happened to M1?” Business Review,
Federal Reserve Bank of Philadelphia (September/October 1986),
pp. 3-14.
Tatom, John A. “Was the 1982 Velocity Decline Unusual?” this
Review (August/September 1983), pp. 5-15.
Thompson, Earl A. “Debt Instruments in Both Macroeconomic The­
ory and Capital Theory,” American Economic Review (December
1967), pp. 1196-210.
Thornton, Daniel L. “Why Does Velocity Matter?” this Review (De­
cember 1983), pp. 5-13.

Appendix
Time Series Forecasts
A time series forecast of the GNP growth rate is used
as a proxy for the expected percentage change in GNP.
The model, which uses quarterly data, was estimated
over the period I/1959-IV/1985.
GNP appears to be a first-order homogeneous pro­
cess. The estimated time series model is reported
below. Calculated t-statistics appear in parentheses,


26


and B is a backward shift operator, that is, (1 —B)x, =
x, - x,_,.
(1 - .248B) ALnGNP = 5.93 + et.
(2.63)
Chi-square (2, 24) = 23.53
This equation forecasts the growth in GNP. These
forecasts were integrated to generate a forecast of the
level of GNP.