View original document

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


Address by
Lawrence K. Roos
Federal Reserve Bank of St. Louis

The New York Society of Securities Analysts
New York, N. Y.
March 19, 1981

It is a pleasure, once again, to have the opportunity to address this distinguished
assemblage. Meetings of the New York Society of Securities Analysts have a well-deserved
reputation as a forum for the discussion of vital economic issues, and I consider it an
honor and privilege to have been invited to a repeat performance before you.

When I previously appeared here on February 5, 1979, I gave a talk entitled "Monetary P o l i c y . . . A Better Way."

In that speech I was sharply critical of the manner in

which monetary policy was then being conducted, and I made specific reference to how
interest rate stabilization stood as a major obstacle in the way of reducing inflation. I
suggested that consistent monetary restraint offered a much better means of dealing
with inflation than continued efforts to fine-tune interest rates.

As you may recall, on October 6, 1979, some ten months after I spoke to you, the
Federal Reserve announced a significant change in the manner in which it intended to
conduct monetary policy. It announced that henceforth it would focus greater emphasis
on controlling the monetary aggregates and less on constraining movements in interest

I remind you of this in no way to imply that the Fed's change of heart was a direct
consequence of the irresistible logic of my 1979 speech. Nor do I infer that my colleagues
at the Fed suddenly "saw the light" and, in a moment of blinding inspiration, rushed to
embrace the monetarist precepts of Milton Friedman. Rather, the Federal Reserve found
itself compelled to change its procedures simply because it was increasingly apparent that
the old methods of policymaking had proved ineffective in combating inflation, and that
something better was needed if financial disaster was to be averted.

In order to appreciate the seriousness of the situation leading up to the October
pronouncement, I would ask you to think back for a moment to that particular period of

By the end of September 1979, interest rates were rapidly rising and the foreign

exchange value of the dollar was plummeting. In both 1977 and 1978, money growth
 had exceeded

the Federal Reserve's intended targets, and, in the six months immediately

preceding October 1979, money had grown at an annual rate in excess of 10%. It had
become obvious that, unless some significant actions were taken, monetary growth in
1979 would exceed the Federal Reserve's target ranges for the third year in a row. Against
this backdrop there was widespread concern over the Fed's seeming inability to control
the situation, and financial markets were reflecting a need for a change in the methods by
which monetary policy was being conducted.

Chairman Volcker's pronouncement on

October 6, 1979, that henceforth the Federal Reserve would emphasize control of money
and credit, offered such a change.

Initially the new thrust of policy showed real promise. Growth of M1B (the money
stock measure consisting of currency and transactions balances) slowed abruptly in the
fourth quarter of 1979 to an annual rate of 3.2% and, for the first time in three years, the
Federal Reserve found itself able to report that money growth had remained within its
announced annual target ranges. As we entered 1980, hopes were high that the Fed's
new procedures were working.

Then something went wrong. Instead of more stable monetary growth, fluctuations
in M1B became more pronounced than at any other time in the past two decades. Interest
rate movements became equally erratic. Financial markets were dazed by interest rates
that rocketed to all-time highs by the end of March, plunged precipitously in the second
quarter of the year, and by year-end shot up again to record high levels.

The general pattern of money growth and interest rate movements throughout
1980, although perhaps somewhat more exaggerated in its ups and downs, was not unlike
what typically had occurred in the past.

As the economy slowed, interest rates and

money growth declined; when the economy accelerated, interest rates and money growth
increased. It is no wonder that, by year-end, "Fed watchers" were beginning to question
whether anything had really changed.

This afternoon, I should like to examine how such wildly erratic growth in money
could have been the product of a policy, the primary purpose of which was to concentrate

on controlling money growth.

I believe that we would all agree that the fundamental mission of monetary policy
must be to reduce inflation in a manner that does not have a destabilizing effect on the
economy. It is important to consider just how well the Federal Reserve's new program
met this test. Did it indeed contribute to a reduction in inflation? Did it promote greater
economic and financial stability?

First, consider its impact on inflation. To have any meaningful effect on reducing
inflation, money growth must be reduced and held below its long-run trend rate of
growth. This did not occur in 1980. The trend rate of growth of M1B in the five-year
period from the end of 1974 to the end of 1979 was 7%. From the fourth quarter of
1979 to the fourth quarter of 1980 . . . the first year of the Fed's new procedures . . .
M1B grew 7.3%. This was above the growth rate of money over the past five years and
certainly does not reflect the reduction necessary to achieve significant progress against

Furthermore, although M1B growth was targeted to grow at between 4% and 6-1/2%
in 1980, its growth last year actually exceeded the announced target range by nearly one
percentage point. And that overshoot would have been considerably greater had money
growth in December not dropped to a —8% annual rate.

Based on that record, it is

difficult to conclude that monetary policy in 1980 contributed in any meaningful way to
the reduction of inflation.

In judging the new policy against the objective of achieving stability in financial
markets, we again see something less than satisfactory performance.

The incredibly

erratic gyrations in the growth of money last year reflect anything but stability. Money
growth staggered and stumbled like a drunk on Saturday night, falling at —13% in April,
spurting to +24% in August, and then dropping at —8% in December. These excessive
gyrations, and the associated erratic behavior of interest rates, led to increased uncertainty
as to the intended course of monetary policy and resulted in severe instability in financial


Thus, from the standpoint both of reducing inflation as well as achieving financial
stability, the first year of the Fed's new program was disappointing.

There are two critical questions that must be answered: "What went wrong?"and,
"What can be done to avoid similar problems in the future?" First of all, I want to make
clear that I am in no way suggesting any lack of good intent on the part of policymakers.
The problem was one of procedure, not purpose. And it was not a matter of having set
incorrect annual targets for money growth. The experience of 1980 provides a perfect
example of what happens when policy changes are not accompanied by technical and procedural changes necessary to make them work. It is the same as if a football team, in the
middle of a losing season, changes its offensive strategy, but neglects to replace its old

Changing monetary policy from stabilizing interest rates to focusing more

directly on controlling the aggregates called for certain changes in practices and procedures. The Fed's reluctance to change its operating techniques to a sufficient extent
was, in my opinion, the primary cause of the poor results of 1980.

To avoid a repetition of the problems experienced in 1980, and to enhance the
Fed's ability to achieve its 1981 monetary growth targets, I propose the following changes
in the conduct of monetary policy.

1- The Federal Reserve should use either total reserves or the adjusted monetary base, instead of nonborrowed reserves, to control money growth. The
most direct and effective means of controlling money growth would be to
control the growth of reserves that constrain monetary expansion.


Fed's present procedure of choosing a nonborrowed reserve growth path,
and assuming a certain level of future commercial bank borrowing from the
Fed's discount window, unnecessarily complicates its ability to hit its
targets. This is because of the difficulty inherent in accurately predicting
the future level of bank borrowing. Not only is the practice of targeting on
nonborrowed reserves cumbersome, it is also based on the erroneous assumption that banks somehow respond differently to changes in borrowed reserves than to changes in nonborrowed reserves. This assumption is clearly


incorrect; it is the quantity of total reserves, regardless of their source, that
determines money growth. Consequently, a total reserve or monetary base
target would be distinctly preferable to the present targeting procedure.

2. The Federal Reserve should revert to contemporaneous reserve accounting.
The Federal Reserve currently uses lagged reserve accounting to determine
the required reserves that financial institutions must hold against their
deposit liabilities.

Any current week's required reserves are based on de-

posits held two weeks previously.

As a result, banks tend to create new

loans and deposits without knowing what reserves will be available to them
at settlement time two weeks hence. Under the present practice of lagged
reserve accounting, the monetary authorities have tended to accommodate
these loans and deposits by subsequently providing necessary required

Effective monetary control requires that reserves serve as a

constraint on future money growth, rather than an accommodation of
lending actions already taken by banks.

A return to contemporaneous

reserve accounting would strengthen the impact of monetary policy.

3. The Federal Reserve should "float" the discount rate so that it will conform more closely to movements in market interest rates. This change is
especially important under a nonborrowed reserve targeting procedure.
The extent to which financial institutions choose to borrow from the
Federal Reserve to meet reserve requirements is directly related to the
relationship between the discount rate and interest rates in financial markets.
Failure to move the discount rate often enough and by large enough
amounts increases the difficulty for policymakers to accurately estimate
and thereby control the extent of commercial bank use of the discount

One objection to floating the discount rate has been a belief that the Federal Reserve would lose the "announcement effect" on financial markets

that discount rate changes are alleged to have generated in the past. This
objection is irrelevant. Experience has shown that financial markets respond
primarily to published money numbers and the extent of their deviations
from announced targets, rather than the so-called "announcement effect"
of changes in the discount rate.

4. The Federal Reserve should target on a single monetary aggregate, preferably on M1B.

Currently the Federal Reserve announces target ranges

for a variety of aggregates:

M1A, M1B, M2, M3, and commercial bank

This multiplicity of targets is confusing to policymakers and the

public alike.

This is especially true when some aggregates are within,

while others are outside of their announced target ranges. It is simply not
true that all of the aggregates are "equal" with respect to their impact on
the economy. Research conducted at the Federal Reserve Bank of St. Louis
shows that M1B, when compared to the other monetary aggregates, is more
closely controllable by the Federal Reserve and is more closely related to
aggregate demand and inflation than other aggregates.

5. The Federal Open Market Committee's short-term money growth targets

must be consistently related to its long-term target. At each meeting of
the FOMC, short-run targets for money growth over subsequent months are

Too often in the past, these short-term growth targets have not

been consistent with the Federal Reserve's announced long-term target
ranges. Consequently, as was the case last year, money growth from month
to month may be within the short-term growth ranges, but at variance with
the long-term targets.

For example, the short-term target established at

the May meeting of the FOMC sought an M1B level of $393.6 billion in
June, while the midpoint of the long-term range called for $396.5 billion.
Similarly, at the October meeting, the short-term goal was $414.7 billion
for December, and the level consistent with long-term ranges would have
been $406.7 billion.

Such deviations from long-term path necessitate

severe adjustments late in the year, as actually occurred in 1979 and 1980.
By explicitly linking the short-term money growth targets to the long-term
ones, the FOMC would eliminate unnecessarily drastic swings in money
growth and would assure more stable financial market conditions during
the year.

6. The Federal Reserve should eliminate all constraints on the federal funds
rate. This change is, perhaps, the most crucial of all. Although the present
federal funds rate ranges are considerably broader than the ones that were
used prior to October 1979, they still, on occasion, frustrate the achievement of money growth targets. This is especially true whenever the federal
funds rate approaches the upper or lower bounds of its targeted range.

Unfortunately, a reluctance to totally eliminate federal funds rate constraint stems,
in large part, from a widespread belief that the Federal Reserve can, and therefore should,
control interest rates. This belief is totally false and, as far as I know, unsupported by
any serious evidence whatsoever.

Federal Reserve actions affecting the growth of the

money stock do influence interest rate movements, but usually in the exact opposite
direction to that espoused by common mythology.

The market interest rate is the price of credit and is determined by supply and
demand for credit. Federal Reserve actions influence both the supply and demand sides
of the credit markets. "Looser" monetary policy may temporarily increase the supply
of credit. However, at the same time it permanently increases the demand for credit as
the inflationary consequences of the Federal Reserve's expansionary actions are quickly
recognized by the public, particularly under present conditions when inflationary expectations are so important. "Loose" or "easy" monetary policy increases money growth,
enhances inflationary expectations, and thereby raises interest rates.

"Tight" policy

reduces money growth, reduces inflationary expectations, and reduces interest rates.
This association between money supply changes and interest rates is well documented.
In 1980, even over periods as short as one week, published changes in money stock

figures were directly correlated with changes in interest rates. Yet, the public is stillinundated with the economic and political nonsense that the Fed's tight policies are
responsible for high interest rates and that only monetary ease will bring rates down.

There is another widely-held argument put forth by the advocates of interest rate

It is based on a belief that stability in money growth necessarily means

instability in interest rates.

This questionable reasoning supports sentiment for the

Federal Reserve to revert to smoothing the movement of interest rates, even if this
means that money growth would be destabilized.

I find this argument to be without

Last year's erratic interest rate behavior was a reflection of the erratic pattern

of money growth and the market's increasing uncertainty as to the direction of policy.
More stable growth of money and the achievement of the Fed's publicly-announced
monetary growth targets would stabilize rather than destabilize interest rates and would
remove much of the uncertainty affecting financial market conditions.

I believe that U. S. monetary policymakers have learned a considerable amount in
the past two years.

In 1979, we learned that a fundamental change in implementing

monetary policy was called for. The rising rate of inflation and the rising level of interest
rates over the past fifteen years demonstrated the fallacy inherent in conducting monetary
policy through the control of interest rates.

In 1980, we learned that good intentions alone will not suffice and that basic
changes in Federal Reserve operating procedures are necessary to achieve monetary

As we gather here today, we have an unprecedented opportunity for economic
progress. We have a foundation of support for economic reform such as has not existed
for some time. The American public has presented policymakers with a mandate to overhaul the Nation's economic machinery from top to bottom. We have a national Administration that is dedicated to urgently-needed fiscal and regulatory reforms. The Federal
Reserve System through its Chairman has expressed its intention to supplement fiscal

reforms with a monetary policy directed toward reducing inflation in a stable and orderly

The necessary elements are in place for effecting major changes in the course of
economic events which, if successful, can have a profound beneficial effect for many
years to come. But the task will not be a simple one. To succeed, all economic systems,
fiscal and monetary alike, must be set on "Go." A failure in any single aspect of the
effort can doom the prospect for success of the entire mission.

Monetary policy is but one part . . . albeit an important one . . . of the process of
restoring stability and growth to our economy. The ability of monetary policy to achieve
its objectives, especially in the critical year 1981, is of special importance, not only for
the achievement of our Nation's economic revitalization, but also for the maintenance
of the Federal Reserve's credibility.

Today, I have suggested certain changes in the way the Federal Reserve conducts
monetary policy.

Those changes are essential for the success of the Fed's effort to

control monetary growth. Fortunately, they would be relatively simple to implement.

Whether they are put into effect will require the support of many diverse groups.
Participants in financial markets must be prepared to accept short-term movements in
interest rates.

Economists must be willing to test new techniques of monetary policy-

And the Federal Reserve must be innovative and ingenious in choosing those

operating techniques most likely to assure the attainment of the System's monetary

I hope that you will agree that the climate for economic progress has never been
better. Let us make certain that we do not let this important opportunity slip through
our fingers.