Nudging the Fed toward a rules-based policy regime.

• Author: Sumner, Scott

There is a great deal of academic research suggesting that monetary policy should use a rules-based approach (e.g., Kydland and Prescott 1977, McCallum 1985, Plosser 2014). However, Fed officials have generally been opposed to any sort of rigid policy rule.

There are two types of policy rules, both of which the Fed finds problematic. One involves a commitment to target a macroeconomic variable such as inflation, or nominal GDP, at a specified rate of growth. Today many central banks aim for approximately 2 percent inflation, although such rules are generally regarded as being flexible--with some weight also being given to output and/or employment stability. Even the European Central Bank, which has a simple inflation mandate, must also ensure that the eurozone monetary regime remains stable and viable.

The Fed has a dual mandate for stable prices and high employment, which it interprets as 2 percent inflation and unemployment close to the natural rate. However, there is no clear indication of the weights assigned to each variable, and hence current policy cannot be viewed as a fully rules-based monetary regime. If both inflation and unemployment are above target, the Fed has discretion as to which problem deserves more attention.

In other cases, the term "policy rule" refers to an instrument rule, such as the famous Taylor rule, which would require that the Fed target the nominal fed funds rate (see Taylor 1993). Key Fed officials also oppose instrument rules, which they suggest do not provide adequate flexibility. They worry that if the natural rate of interest and/or the natural rate of unemployment change, then the Taylor rule could lead to a suboptimal policy. In principle, the rule can adapt to changes in these parameters, but it may be very difficult to estimate the natural rate of either unemployment or the real interest rate.

Elsewhere, I have argued that the Fed's discretionary approach did very poorly during the Great Recession and that the Fed should adopt level targeting of nominal GDP (Sumner 2012). I have also suggested that policymakers should target the market forecast of future nominal GDP, or at least the Fed's internal forecast, if a market forecast is not available (Sumner 2015). (1) In this article, I will simply assume that a nominal GDP-level target is the best option; however, all of the arguments presented here could equally be applied to a different policy target, such as one for 2 percent inflation.

Given the Fed's opposition to a rigid policy rule, it's worth asking whether the Fed can be "nudged" in the direction of a policy rule, through some more modest and less controversial policy reforms. Here I'll suggest three such reforms: first, asking the Fed to more clearly define the stance of monetary policy; second, asking the Fed to more clearly evaluate past policy decisions; and third, asking the Fed to define the outer limits of acceptable deviation in aggregate demand from the target path. I will also argue that if the Fed starts down this road, it will likely lead to the eventual adoption of nominal GDP-level targeting.

What Do We Mean by the "Stance" of Monetary Policy, and Why Does It Matter?

Economists frequently refer to monetary policy using terms such as "expansionary" or "contractionary," "easy" or "tight," and "accommodative" or "restrictive." Those terms are said to refer to the "stance" of monetary policy. Given their frequent use, one might assume that they have a clear meaning, at least to professional economists. Unfortunately, that is not the case. References to the stance of monetary policy are often vague and misleading, and frequently hinder clear thinking about the role of monetary policy in the business cycle. Given that the Fed discusses its policy stance while communicating with the public, it is essential that policymakers clearly define what these terms mean. We need some metric for evaluating the stance of monetary policy.

I am not the first economist to express frustration with the way pundits characterize real-world monetary policy stances. Milton Friedman made a similar complaint in 1997:

Low interest rates are generally a sign that money has been tight, as in Japan; high interest rates, that money has been easy.... After the U.S. experience during the Great Depression, and after inflation and rising interest rates in the 1970s and disinflation and falling interest rates in the 1980s, I thought the fallacy of identifying tight money with high interest rates and easy money with low interest rates was dead. Apparently, old fallacies never die [Friedman 1997].

Friedman thought monetary policy in Japan was quite contractionary during the 1990s, despite near-zero interest rates and quantitative easing (QE). Of course, in 1963, Friedman and Schwartz famously argued that contractionary monetary policy had caused the Great Depression. And, as with Japan, this occurred despite near-zero interest rates and rapid growth in the monetary base. (2)

This raises an interesting question: If Friedman were alive today, would he have regarded Fed policy during 2008 and 2009 as expansionary or contractionary? Indeed, is it possible that the recession of 2008 was caused by tight money? I won't definitively answer that question here. Rather, I will show that this hypothesis should not be summarily rejected merely because most economists saw monetary policy during 2008 and 2009 as being "obviously" highly expansionary.

The Problem of Identifying the Stance of Monetary Policy

Joan Robinson (1938) argued that easy money could not have caused the German hyperinflation, because interest rates were not low. Modern economists might be inclined to smile at this example of "old Keynesian" thinking, perhaps recalling the more than billion-fold increase in the German monetary base between 1920 and 1923, when currency was being printed at a furious pace. On the other hand, modern economists are not supposed to rely on changes in the monetary base as an indicator of the stance of monetary policy. So is that really a good reason to dismiss Robinson's claim (which applied to the first part of the hyperinflation)?

For instance, in the United States, the monetary base had been growing at about 5 percent a year in the period leading up to August 2007. Then, over the next nine months, growth in the base came to a sudden halt. And yet you would be hard pressed to find many economists who regarded this sudden change in the growth rate of the monetary base as a contractionary move by the Fed. The reason is obvious: interest rates were cut repeatedly during this nine-month stretch, from 5.25 percent all the way down to 2 percent. Contemporaneous discussion of monetary policy during 2007-08 almost invariably referred to the Fed's actions as expansionary or "easy money." This characterization implicitly rejected the monetary base as a useful indicator, and (presumably) relied instead upon changes in interest rates.

A more sophisticated argument against Joan Robinson's claim would be that nominal interest rates don't matter, and that real interest rates are the proper measure of the stance of monetary policy. Certainly, real interest rates would be a superior policy indicator during a period of hyperinflation. But once again it seems highly unlikely that this is the variable that economists actually focus on--or should focus on. Between early July and early December 2008, the real interest rate on five-year inflation indexed Treasury bonds rose from less than 0.6 percent to more than 4 percent, one of the sharpest increases ever recorded in such a short period of time. If economists regarded real interest rates as the proper indicator of the stance of monetary policy, then one might have expected almost universal outrage about the Fed's "highly contractionary" policy shift during a period of financial turmoil and deepening recession. Yet it is difficult to find any criticism of this sort during the second half of 2008. On the contrary, most commentators claimed that monetary policy was expansionary. (3)

We have already seen that nominal interest rates might be highly misleading due to the effects of inflation. But, in fact, the same sort of criticism can be lodged against real interest rates, which reflect other macroeconomic variables, such as expected real GDP growth. A real interest rate of 2 percent during a period of rapid economic growth certainly represents a different monetary policy stance from a 2 percent real interest rate during a deep depression. So it is not at all clear that real interest rates are actually a good indicator of policy.

Milton Friedman is not the only economist to criticize the way members of his profession describe the stance of monetary policy. Other highly regarded economists, in many cases those closer to the (new Keynesian) mainstream, have expressed similar concerns. Take Frederic Mishkin, who served on the Federal Reserve Board, and wrote the number one monetary economics textbook in the United States (Mishkin 2007). Toward the end of the book he listed several important points about monetary policy. Here are the first three as they appeared back in 2008:

1. It is dangerous always to associate the easing or the tightening of monetary policy with a fall or a rise in short-term nominal interest rates.

2. Other asset prices besides those on short-term debt instruments contain important information about the stance of monetary policy because they are important elements in various monetary policy transmission mechanisms.

3. Monetary policy can be highly effective in reviving a weak economy even if short-term rates are already near zero [Mishkin 2007: 606-7].

One of the most striking features of these three key lessons for monetary policy is how incompatible they seem with the consensus view of events circa 2008 and 2009. Policy was almost universally viewed as being expansionary precisely because the Fed cut...