View original document

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

Safeguarding Good Policy Practice
“Reflections on Monetary Policy 25 Years After October 1979”
Federal Reserve Bank of St. Louis
St. Louis, Missouri
October 8, 2004
Published in the Federal Reserve Bank of St. Louis Review, March/April 2005, 87(2, Part 2), pp. 303-06


can’t help recognizing an emotional note
in my reaction to this conference. Yesterday,
we enjoyed three superb scholarly papers.
Allan Meltzer’s paper left me depressed,
and the Lindsey, Orphanides, Rasche paper left
me elated. Marvin Goodfriend’s paper left me
with hope for the future.
But now I’ll try to be a dispassionate social
scientist. This panel inevitably overlaps somewhat
the previous one on what we have learned since
October 1979. In no small part, what we have
learned since October 1979 starts with what we
learned from the Great Inflation and how it was
brought to an end. Going forward, we need to
incorporate in policy practice both sound theory
and lessons from history.
I will make five major points, none of which
is new but all of which deserve attention in the
context of this conference. First, good science is
extraordinarily important. Second, market confidence in the central bank is essential for good
monetary policy. Third, stability of the real
economy requires price stability. Fourth, central
bankers have an obligation to communicate
clearly with the general public. Fifth, we should
not underestimate the role of leadership.

The problems of the 1960s and 1970s were
partly—not totally, but partly—the consequence
of bad economics. Allan Meltzer has discussed
those issues, and I do not need to repeat his argument here.
I note especially that Chairman Martin’s dismissal of economics and economists does not

make for happy reading today. I hope that we
never again see Federal Open Market Committee
(FOMC) members with that attitude. Policymakers
need not be professional economists, but they
must be able to understand what economists bring
to the table.
How do we safeguard a high level of expertise
in the FOMC of the future? There is no way to
ensure that the appointment process will always
put the right people on the FOMC. But I think we
can help to guard against appointment errors by
working with Reserve Bank directors, who choose
Reserve Bank presidents, with Congress, and
with opinion leaders in general. Those of us in
leadership positions today, and everyone else
with monetary policy expertise, need to spend
time in helping to instill in the general public a
deeper understanding of monetary policy responsibilities. We need to discuss what characteristics
are necessary for policymakers to be successful. I
hope that we never again have appointments
yielding the results of the 1930s, 1960s, and 1970s.
The largest gap in macroeconomics is the
weak understanding of the relationship between
real and nominal variables. In our models, we
employ a Phillips-curve type of relationship to
model inflation, or changes in the rate of inflation.
In our models, a departure of the actual rate of
inflation from the expected rate depends on a
current and expected future real gap measure of
some sort. I simply distrust this model, on both
theoretical and empirical grounds. Empirically, I
don’t think it works very well, and theoretically
it ought not to work very well.
I’d love to hear Chairman Greenspan offer a
systematic exposition of his enormous success


in forecasting inflation pressures. My sense of
what I do, which I think is not dissimilar to what
most FOMC members do, is attempt to intuit
future inflation pressures from current observed
pressures as they show up in both price changes
and resource pressures, or real gaps, in individual
markets. The approach is not totally without
theory; for example, wage changes are evaluated
in light of expected productivity trends. I attempt
to sort out temporary from more lasting wage and
price changes and attempt informally to construct
an appropriately weighted average of disparate
experience in various sectors. I look closely at
data on inflation expectations, but treat such data
carefully because longer-run expectations are
really a vote of confidence on the Fed and not an
independent reading on inflation.
I am extremely uncomfortable with this
approach and believe that it is an invitation to
future mistakes. I don’t know what better to do.

A standard feature of monetary analysis in
recent years is that market confidence in the
central bank is tremendously important. Retaining confidence requires, above all, successful outcomes. There is no adequate substitute for good
results. The market does not require perfection—
people do understand in broad outline what the
central bank can and cannot do. People understand that some small mistakes are inevitable.
Still, the market will surely lose confidence from
mistakes occurring year after year after year.
Once confidence is gone, restoring it is incredibly costly. That is one of the prime lessons of U.S.
experience in the late 1970s and early 1980s and
experience around the world. To restore confidence, it is necessary to achieve, or at least make
progress on, policy objectives of price stability
and full employment.
Making progress on policy objectives is far
more important than hitting an intermediate target
such as a steady, moderate rate of money growth.
To the noneconomist, intermediate targets are
highly technical in character. I am not an engineer,
for example, and really don’t care what engineers

say about the strength of steel when bridges fall
down. Similarly, the noneconomist really doesn’t
care about the rate of money growth. If it works,
fine, but stable money growth is not a substitute
for price stability. Of course, an intermediate target
may be of transitional importance in restoring
confidence, as the 1979-82 experience shows.
Restoring confidence may require—indeed, I
suspect in most cases does require—bearing a lot
of pain to demonstrate that a central bank is serious about meeting its responsibilities. The recession of 1981-82 is such an event. The market wants
to see the central bank is able to bear pain, and
the reasoning is simple. If you cannot withstand
a lot of pain, why should anyone believe you are
serious given all the pressures to change course?
Technical explanations can always be offered to
explain a change in policy direction, but if it
appears that technical mumbo jumbo is an excuse
for not completing the job, then confidence will
not be restored. Thus, a change in policy direction
will require a fairly understandable explanation
once a fair amount of pain has been endured.
The logic of pain seems inescapable. Inflation
cannot fall permanently unless inflation expectations come down. Expectations will not come
down in the absence of confidence that the central
bank will keep inflation down in the future. Confidence in the central bank will not be obtained
unless the market becomes convinced that the
central bank, and the political system more generally, has the institutional strength to maintain
low inflation. The real test of institutional strength
is capacity to bear pain.
The rational expectations argument of costless
disinflation through restoration of credibility never
appealed to me. In 1979, given what the Fed had
said and done over the preceding 15 years, it
would have been irrational to have granted the
Fed instant credibility.

My third point is that maintaining price stability is extraordinarily important for stability of
the real economy. The idea that we can trade off

Safeguarding Good Policy Practice

employment stability against inflation stability is
flawed. I do not want to deny that there may be
some trade-off around the edges, but the key regularity is that instability of inflation and real growth
are positively correlated. Tolerance of higher inflation is not a recipe for creating higher employment or improved employment stability, but just
the reverse. The reason is that inflation instability creates more instability in inflation expectations and wider dispersion in the expected rate
of inflation.
Greater variability and dispersion of inflation
expectations increases the magnitude of expectational errors and therefore increases misallocations
in the real economy. Moreover, an increase in
inflation tends to reduce the market’s confidence
in the central bank, which, in turn, makes it more
difficult for the central bank to adjust its policy
to help stabilize the real economy. This point was
demonstrated dramatically in the 1980-82 period.
Given weak market confidence in the Federal
Reserve’s willingness to control inflation, the Fed
was not able to switch gears toward a less restrictive policy as employment weakened in the 1981
recession. The central bank could not switch gears
because doing so ran the risk of undoing tentative
progress in restoring the market’s confidence in
the central bank.
The arguments I have just offered flow from
sound economics—the observed positive correlation between inflation instability and employment instability is what we ought to expect.

Allan Meltzer discussed the intellectual
environment that made the Great Inflation possible. By the 1960s, traditional central bank concern
over inflation had come to be regarded as old
fashioned and the new economics promise of an
optimal trade-off of modest inflation to buy lower
unemployment had won many converts. Although
the Federal Reserve, especially the Board of
Governors, included converts, the Fed also
included leaders who shared traditional concerns
about inflation. My memory of this period, which

I have not tried to research for accuracy, is that
traditional concerns were not stated forcefully
by articulate defenders of price stability within
the Fed.
Central bankers can influence public debates,
if they try. One of the lessons I draw from the
Great Inflation is that those of us in leadership
positions in the Federal Reserve have an obligation to communicate actively. If we do not, by
default we leave the debate to others. I think that
academics are important to public debates primarily through the longer-run force of their scholarly
contributions. These are all that really matter in
the long run; in the short run, some academics
command public attention, but not very many
and not much attention in the scheme of things.
Press attention is concentrated on politicians,
office-holders in general, and business leaders
who control large resources. Federal Reserve
office-holders immediately attract press attention,
by nature of their positions. As a Reserve Bank
president, I have an opportunity to reach an audience far larger than I ever had as a professor at
Brown University.
The communications environment is quite
different today from the early 1980s, when the
Fed released relatively little information. In the
interest of full disclosure, I was one of the skeptics
when the Fed abandoned reserve targeting in the
late summer of 1982. My fear was that the Fed
would embark once again on a policy that would
permit inflation to rise. As a monetary economist,
perhaps I knew too much; I found the Fed’s explanation for switching from nonborrowed-reserves
to borrowed-reserves control in 1982 an example
of the technical mumbo jumbo I referred to earlier.
But the market bought the argument, and the fact
that monetarists such as I were suspicious was
irrelevant. I was wrong, and I am certainly happy
that I was wrong.
Still, the current environment of much greater
Fed openness has probably raised the standard
of what will be required in debates in the future.
If the Fed makes major mistakes and must again
embark on a campaign to restore credibility, I
suspect that it will have to pursue a more open
dialog with the public.


In any event, safeguarding good policy practice from political pressures will require ongoing
communication with Congress, market professionals, leading citizens, and the general public.
Good monetary policy will be easier, and more
effective, with widespread understanding of what
constitutes good policy. That to me is one of the
clear lessons of the Great Inflation.

My last point concerns the role of leadership.
This conference is a celebration of Paul Volcker’s
Central bank leadership requires at times a
willingness to push hard enough to get the job
done—and recognition of how hard is too hard.
The central bank does not want to get itself fired
through changes in law or appointments that
undermine the bank’s authority. Pushing hard
enough but not too hard is obviously a dicey act
at times, requiring political judgment and acumen,
but it is nevertheless one that central bank leadership must be able to pull off successfully.
I appreciate, at a much deeper level today
than I did at the time, the extent of Paul Volcker’s


achievements in the 1979-82 period. Saying that
is not meant to imply a negative comment about
his achievements in later years. But certainly
1979-82 was a critical period in U.S. monetary
history. I know that Paul Volcker did not do the
job alone—support from President Reagan was
critical. That said, there was no guarantee that
President Carter would appoint Paul Volcker.
Volcker was a logical, but not inevitable, appointment. President Carter could instead have
appointed a Chairman who would have continued
the policy of drift. The inflation rate would have
continued to rise, and the pain of unwinding the
inflation would have been greater.
The Great Inflation is understandable, but
was not unavoidable. Stronger leadership by
Chairman Martin would have cut short the early
development of inflation. Chairman Burns could
have stopped it. The intellectual and political
environment of the 1960s and 1970s certainly had
a lot to do with making the Great Inflation possible. Still, the Great Inflation was not inevitable.
Leadership really does matter.