Published by EH.Net (December 2013)

Michael D. Bordo and Athanasios Orphanides, editors, The Great Inflation: The Rebirth of Modern Central Banking. Chicago: University of Chicago Press, 2013. xii + 532 pp. $120 (hardcover), ISBN: 978-0-226-06695-0.

Reviewed for EH.Net by Robert L. Hetzel, Federal Reserve Bank of Richmond.

The editor requested a one thousand word review of The Great Inflation, starting with a summary of the fourteen major essays, eight pithy commentaries, and accompanying wide-ranging discussion by eminent economists.  Given the richness of the material, that request is like asking someone to catch a bucket full of water from Niagara Falls and then pour it out in order to capture the falls for someone in a different location.  The first point to make is that the preface understates the value of the book.  The preface promotes the book with the observation that high inflation is costly and policy makers need to learn not to repeat the experience.  True enough, the incentive to adopt price controls when faced with high inflation came close to setting the United States on a path leading to state control of the economy and away from free enterprise.  However, the book is far more than a morality tale for central bankers.  The long, stumbling process of learning engaged in by central banks about operating in a regime of fiat money provides the kind of experiments that economists require in order to identify shocks.  When it comes to these experiments, the period known as stop-go monetary policy and as the Great Inflation is “as good as it gets.”

Economists form qualitative assessments of the economic outcomes of different monetary regimes and develop priors over the nature of the shocks in those regimes, real or monetary.  They then construct models that embody frictions that translate those shocks into real and nominal instability.  Ideally, estimation of the resulting models quantifies the shocks.  The methodological North Star is the formulation of hypotheses about what classes of policy rules provide for economic stability by eliminating monetary shocks and by mitigating real aggregate-demand shocks.  All such hypotheses are provisional.  They remain useful only to the extent that they predict outcomes in monetary regimes not observed in constructing the model.  Future outcomes are the obvious candidates, but experiences in other times and countries offer a largely unexcavated goldmine.

There is no way to short-circuit this process.  Different economists read the historical record of monetary policy experiments differently and build different models, which embed different frictions.  As pointed out by Chari, Kehoe, and McGrattan (2009), these different models when estimated produce very different shocks, and yet they all fit the data equally well.  One can use microeconomic evidence to distinguish among models.  However, no matter how sophisticated, models never will yield accurate measures of “natural” values of variables like the real interest rate, the unemployment rate, and the output gap.  Furthermore, modern models are not close to a satisfactory modeling of a phenomenon like monetary nonneutrality.

This state of affairs means that in narrowing the class of models economists explore the returns are very high to a better understanding of the historical policy “experiments” conducted by central banks.  What experiment did policy makers conduct in the Great Inflation?  Based on the essays in The Great Inflation, I take the following to be a least-common-denominator answer to this question (see also Hetzel 2008 and 2012).

Over a fifteen year period starting in the mid-1960s, the Federal Reserve operated within and largely accepted a political and intellectual consensus that aggregate-demand management should concentrate on achievement of a level of the unemployment rate consistent with full employment, widely taken to be four percent.  By the 1970s, the Keynesian consensus had come to accept that monetary policy constituted a significant influence on aggregate demand.  At the same time, that consensus assumed that inflation was a real phenomenon.  It follows that inflation is a phenomenon with multiple causes, which divide into the classifications of demand pull, cost push, and wage-price spiral.  From the perspective of inflation as a real phenomenon, the Fed had to make a difficult decision about balancing the benefits of low inflation against the real costs of the high unemployment required to control inflation.  Incomes policies (price and wage controls in the extreme) were considered to be an important means of alleviating the high cost of controlling inflation and relaxed the need for monetary policy to aim for low inflation rather than low unemployment.

Although the Fed avoids the language of trade-offs, implicitly, the Phillips curve, which makes lower inflation depend upon higher unemployment, organized the debate over the cost-benefit calculus of the degree of control the Fed should exercise over inflation.  Given the assumed chronic lack of a level of aggregate demand sufficient to assure full employment, policy makers believed that observed inflation was of the cost-push variety.  Price stability would, they believed, require socially unacceptable levels of unemployment (an unemployment rate sufficiently high in order to persuade labor unions to relinquish inflationary wage demands).  During the stop-go era, on an ongoing discretionary basis, the Fed implicitly weighed the benefits of monetary stimulus directed at moving unemployment toward its full-employment level against the costs of removing unemployment as a deterrent to cost-push inflation.  The contemporaneous behavior of unemployment and inflation caused the Fed to shift between monetary stimulus (the go phases) and monetary restriction (the stop phases).

The central characteristic of stop-go monetary policy was the absence of a nominal anchor.  Michael Bordo and Barry Eichengreen (“Bretton Woods and the Great Inflation”) document how the constraint on monetary policy exercised by gold outflows disappeared starting with the Johnson administration when capital controls replaced monetary policy as the instrument for rectifying international payments imbalances.  Andrew Levin and John B. Taylor construct an important measure of expected inflation, which places the start of the unanchoring of inflationary expectations in the mid-1960s.  After slowly drifting up, inflationary expectations rose dramatically toward the end of the 1970s.

Athanasios Orphanides and John Williams (“Monetary Policy Mistakes and the Evolution of Inflation Expectations”) use FOMC documents to demonstrate the activist character of policy directed toward achievement of a low unemployment rate.  In the spirit of the Phillips curve, Board staff at the Fed forecast that monetary stimulus would raise output (lower unemployment) with no impact on inflation in the presence of a negative output gap (“high” unemployment).  That is, the Fed could stimulate aggregate demand without fear of creating demand-pull inflation.  The inflation that did occur had to arise from cost-push forces.  Ricardo DiCecio and Edward Nelson (“The Great Inflation in the United States and the United Kingdom: Reconciling Policy Decisions and Data Outcomes”) show the prevalence of this view.

The discretionary character of monetary policy meant that the FOMC weighed off each period whether to move policy in a restrictive direction (raise the funds rate) or in a stimulative direction (lower the funds rate).  The emphasis placed on financial conditions conducive to the encouragement of investment in the housing and corporate sectors led the FOMC to limit the magnitude of increases in the funds rate.  Marvin Goodfriend and Robert G. King (“The Great Inflation Drift”) provide a model showing one way in which interest rate smoothing imparts random drift to the price level.

Alan Blinder and Jeremy Rudd (“The Supply-Shock Explanation of the Great Stagflation Revisited”) defend the traditional Keynesian view of aggregate-demand management aimed at balancing employment and inflation goals.  They argue that there is an underlying core rate of inflation determined by the difference between the growth rate of aggregate nominal demand and potential output with monetary policy only one of many influences on aggregate nominal demand.  The large fluctuations in inflation in the 1970s came from inflation shocks (increases in the relative price of energy and food) that pushed inflation above its core value.

As recounted by William Poole, Robert H. Rasche, and David C. Wheelock (“The Great Inflation: Did the Shadow Know Better?”), the Shadow Open Market Committee challenged the prevailing cost-push explanation of inflation and the inference that the control of inflation required an ongoing calculation of the costs of that control in terms of excess unemployment.  In contrast to the prevailing Keynesian sentiment, the Shadow made the monetarist assumption that monetary policy is the dominant force in the growth rate of aggregate nominal demand at cyclical and trend frequencies.  Given the stability of the trend rate of growth of M1velocity at the time, it follows that steady M1 growth would have produced steady growth in aggregate nominal demand at cyclical and trend frequencies.  Given steady growth in potential output, the policy of steady, moderate growth of M1 espoused by the Shadow would have provided a nominal anchor and steady trend inflation.

Given the Blinder-Rudd explanation of the Great Inflation that exonerates the Fed from any blame, the discussion of monetary policy in Japan and Germany is especially interesting.  Takatoshi Ito (“Great Inflation and Central Bank Independence in Japan”) contrasts monetary policy before and during the two inflation shocks of the 1970s, the first in 1973-1974 and the second in 1979-1980.  He argues that expansionary monetary policy in the early 1970s had already created high inflation before the first shock.  In the second episode, monetary restraint led to only a short-lived, moderate increase in inflation.  In a similar spirit, Andreas Beyer, Vitor Gaspar, Christina Gerberding, and Otmar Issing (“Opting Out of the Great Inflation: German Monetary Policy after the Breakdown of Bretton Woods”) credit a monetary policy, which started in the mid-1970s, with a firm nominal anchor for price stability as a source of nominal and real stability relative to other countries.  They especially emphasize the role of money targets as a commitment device for aligning inflationary expectations with the goal of price stability.

Macroeconomists cannot run controlled experiments, but they can do a much better job of identifying and elucidating the extraordinary range of experiments that central banks have delivered.  The Great Inflation is a terrific example.


V.V. Chari, Patrick J. Kehoe, and Ellen R. McGrattan. “New Keynesian Models: Not Yet Useful for Policy Analysis.” American Economic Journal: Macroeconomics 1 (January 2009), 242-66.

Robert L. Hetzel. The Monetary Policy of the Federal Reserve: A History. Cambridge: Cambridge University Press, 2008.

Robert L. Hetzel. The Great Recession: Market Failure or Policy Failure?  Cambridge: Cambridge University Press, 2012.

Robert L. Hetzel is an Economist and Senior Policy Advisor at the Federal Reserve Bank of Richmond. The views expressed in this review are those of the author not those of the Federal Reserve Bank of Richmond or the Federal Reserve System.

Copyright (c) 2013 by EH.Net. All rights reserved. This work may be copied for non-profit educational uses if proper credit is given to the author and the list. For other permission, please contact the EH.Net Administrator ( Published by EH.Net (December 2013). All EH.Net reviews are archived at