Skip to Main Content

Honoring Marvin Goodfriend

The 2007 Monetary Policy Consensus in Retrospect

Goodfriend, Marvin. 2007. “How the World Achieved Consensus on Monetary Policy.” Journal of Economic Perspectives 21, no. 4 (Fall): 47-68.


Keynes famously ends the General Theory with a description of the "academic scribbler" whose ideas from "a few years back" eventually find their way into policymaking. When Marvin Goodfriend wrote "How the World Achieved Consensus on Monetary Policy" in 2007, that time lag had largely disappeared, at least in central banking. For example, in the Federal Reserve System it had become the norm for a substantial fraction of those in important decision-making positions to hold PhDs in economics. Not only did these officials have advanced degrees, they often earned the stature that led them to be chosen for their position by doing cutting edge research. The most prominent example was the Federal Reserve chairman at the time, Ben Bernanke. It is also now the case that the research done by staff at the Fed and other central banks is as sophisticated as any that occurs in academia. As a result, ideas flow freely and instantly between the halls of academia and central banks. The time lag is gone.

An important theme of Marvin's 2007 paper is how the development of this symbiotic relationship between academic research and central bank policymaking led to a consensus on a new framework for monetary policymaking, a framework that has proved highly useful and remains with us today. What Marvin leaves out is the significant role that he played in this development. To my knowledge, Marvin was the first economist to simultaneously contribute to the modern literature on monetary policy and hold a nontrivial policymaking job in the Federal Reserve System. A happy coincidence is that at Brown he was classmates with his future coauthor Bob King. The two would play an important role in developing the New Keynesian framework and making it operational for analyzing monetary policy. Being able to work with Bob kept Marvin in close proximity to academia. At the same time, Marvin's experience with the policymaking process provided him with important insights into how to make their work, as well as his other research, most useful in practice.

The early years: disarray and revolution in macroeconomics

When Marvin joined the Richmond Fed in 1978, communication between central banking and academia may have been at an all-time low. The failure of the large econometric models developed more than a decade earlier to anticipate the stagflation of the late 1960s and '70s left central banks without a framework to provide guidance for monetary policy, at least one in which they could have confidence. In academia, the rational expectations revolution was heating up. Popular at the time was the Phelps/Friedman natural rate theory, which related output to unanticipated movement in inflation. It was standard to assume that expectations were formed adaptively, so that a monetary expansion in the short run would increase real output temporarily but then produce a subsequent increase in inflation. Rational expectations turned things upside down: Within the context of the Phelps/Friedman framework, predictable movements in the money supply (which produced predictable movements in inflation) had no effect on real output, as Robert Lucas (1972) famously showed. To put it mildly, central bankers were not particularly hospitable to the idea that only unpredictable movements in the money supply could affect real activity.

A more extreme development from the vantage of central bankers was the advent of real business cycle (RBC) theory, which involved the use of the stochastic competitive equilibrium growth model to explain business cycles. The virtue of the approach is the explicit use of microfoundations to build a macroeconomic framework. A striking implication, however, is the total irrelevance of monetary and financial factors. Another dramatic implication was that business cycles, while unfortunate, represented efficient responses of the economy to exogenous disturbances. Needless to say, this development did not exactly enhance academic interaction with monetary policymakers.

The Volcker disinflation: consequences for research and policymaking

As Marvin emphasizes, a critical turning point was the shift to tight monetary policy in late 1979, engineered by Paul Volcker. The aim was to bring the era of high inflation to an end. As Marvin describes, the sudden and unexpected tightening can be thought of as a kind of natural experiment to study the impact of monetary policy on output and inflation. The tightening succeeded in reducing inflation, though with a lag. But in the process it induced the largest recession of the postwar period up to that point. As Marvin notes, the episode sent a clear message to central bankers: they did have the ability to control inflation. At the same time, disinflations were not costless, even if the factors that determined these costs were not clearly understood.

I would add that the Volcker disinflation also had a profound effect on the course of academic research. It was clear that neither the Phelps/Friedman model with rational expectations nor RBC theory could easily account for the effect of the Volcker tightening on output and inflation dynamics. The need for a new framework was obvious. But it was also clear that the field could not retreat from the methodological advances ushered in by the rational expectations/RBC revolution. These considerations led to an effort to rebuild Keynesian economics using microfoundations. Out of this effort would emerge a framework that could be used — and eventually would be used — in the policymaking process. No, the framework has not come anywhere close to the point where it can be used to put monetary policy on automatic pilot. But it has reached the point where it does play a significant role in helping organize thinking about policy implementation. As a result, the relationship between academic research and central bank policymaking has become highly symbiotic. Economic events influence the development of the model. The model in turn informs policymaking.

Marvin's work with Bob King (Goodfriend and King, 1997) played a significant role in the development of what is now widely known as the New Keynesian (NK) model.2 Marvin and Bob perhaps more aptly refer to this paradigm as the New Neoclassical Synthesis, as it begins with an RBC model and then adds three crucial ingredients. First, money is introduced so that the model can account for nominal variables. Second, monopolistic competition is incorporated so that it is possible to characterize price setting by firms. Third, nominal rigidities are added, which gives rise to the nonneutraliy of money and inefficient fluctuations in output.3 Absent nominal price rigidity, the framework behaves essentially as an RBC model. With nominal price rigidity, the Keynesian features emerge.

The interest rate as the policy instrument

Another important component of the consensus that Marvin emphasizes is the use of the short-term interest rate as the instrument of monetary policy, in keeping with actual practice at central banks. As late as the 1980s, it was still commonplace in academic work to model the money supply as the policy instrument. However, central banks have learned through practical experience that trying to directly regulate monetary aggregates was problematic. Broad monetary aggregates were difficult to control due to the endogeneity of inside money. Controlling narrow aggregates like reserves generated wild gyrations in interest rates due to fluctuations in reserve demand. These wild fluctuations in interest rates, in turn, wreaked havoc on the economy.

Given his proximity to policymaking, Marvin quickly saw that to get the attention of central bankers, academic work needed to treat the interest rate as the policy instrument. Indeed, Marvin was among the earliest researchers to interpret monetary policy actions through the lens of interest rate decisions, not only about current rates settings, but also about communication of the paths of future rates.4

Click here to continue reading this essay.


Cite as: Gertler, Mark. 2022. "The 2007 Monetary Policy Consensus in Retrospect." In Essays in Honor of Marvin Goodfriend: Economist and Central Banker, edited by Robert G. King and Alexander L. Wolman. Richmond, Va.: Federal Reserve Bank of Richmond.

 
1

Thanks to Bob King and Alex Wolman for helpful comments.

2

Related work includes Rotemberg and Woodford (1997), Clarida, Gali, and Gertler (1999), Woodford (2003), Christiano, Eichenbaum, and Evans (2005), and Gali (2015). For a collection of the early contributions to the New Keynesian framework, see Mankiw and Romer (1991). For a critique, see Chari, Kehoe, and McGrattan (2009).

3

The most common way to incorporate nominal rigidities is via the staggered contracting approach of Calvo (1983), which is a more tractable version of Taylor's (1980) overlapping contracts framework.

4

See Marvin’s paper “Interest Rates and the Conduct of Monetary Policy” (1991), which is also discussed in this volume, in an essay by John Taylor.

Phone Icon Contact Us

Lisa Davis (804) 697-8179