View all Articles

Why It Matters Whether the Fed Targets Inflation or Unemployment

Economics Finance

Today, more than four full years since analysts at the National Bureau of Economic Research declared the last recession officially

over, unemployment, at 7.3 percent, remains elevated. The jobless rate still exceeds the 2001 recession peak and stands not far below the higher peak from the 1990-91 downturn. Incomes, whether measured in the aggregate by Gross Domestic Product or on the individual level for those who are employed, have fallen far below levels that were reasonably expected prior to the financial crisis. In the meantime, with its federal funds target still up against its zero lower bound, the Federal Reserve seems incapable of providing much further monetary stimulus through additional interest rate changes.

How should Federal Reserve policymakers respond to this continuing crisis? And what new procedures might they put in place to ensure that nothing like it ever happens again? These questions are addressed in two studies released by top economists from the Federal Reserve Board, presented at a high-profile conference sponsored by the International Monetary Fund, and discussed widely in the popular press, including a recent article from the Wall Street Journal.  William English, David Lopez-Salido, and Robert J. Tetlow argue that even when the current short-term interest rate is constrained by the zero lower bound, monetary policy can still provide additional support for a weak economic recovery if policymakers promise to keep that interest rate low for a prolonged period of time through a form of so-called “state-dependent forward guidance.” Their analysis suggests that in the United States today, it might be appropriate for interest rates to remain low until unemployment crosses a “threshold” as low as 5.5 percent.  Dave Reifschneider, William Wascher, and David Wilcox propose a more permanent change to monetary policy. Going forward, they argue, the Fed should respond more vigorously to rising unemployment in the early stages of an economic downturn to prevent anything like a replay of what we experienced during 2008 and 2009. 

Both sets of authors are careful to point out that these views are their own, and not necessarily those of the Federal Open Market Committee. Furthermore, the authors of both studies write in a thoughtful, academic style that we very much admire. Circulating the results of internal research in the form of discussion papers such as these is essential for maintaining transparency about the thinking of Fed advisors, and so we applaud these authors’ willingness to share their views publically.

Transparency is especially important because these studies, written by leading economists at the Board, are widely viewed as providing much of the intellectual basis for the FOMC’s recent “forward guidance” statements, as well as the public statements of many top Fed policymakers regarding the benefits of setting thresholds for unemployment to guide the timing of future interest rate changes.

Does macroeconomic theory, or the empirical record of the relationship between monetary policy and unemployment, support the conclusions drawn in these studies and the policy actions that they recommend? With respect to one point in particular, we think not.

Specifically, while we share the authors’ concern that unemployment remains too high and agree that the US economy as a whole continues to operate at a level that is well below its long-run potential, we object to the notion that monetary policy decisions should be tied so closely to movements in the unemployment rate. A principal lesson of the 1960s and 1970s was precisely that monetary policy cannot successfully target the unemployment rate and that attempts to do so can have catastrophic consequences for both inflation and unemployment.

Both of the Fed studies rely heavily on the notion that the Phillips curve, a statistical relationship that associates higher inflation with lower unemployment, offers monetary policymakers an exploitable trade-off between these two variables. Unfortunately, in practice, this frequently turns out not to be the case: Policymakers’ efforts to “buy” lower rates of unemployment have often produced both higher inflation and higher unemployment – just the opposite of what the Phillips curve theory would suggest. That outcome reflects both the inability of monetary policy to reliably influence unemployment in the short term and the adverse influence of higher inflation on growth and employment in the intermediate and long term.

Some macroeconomists today would respond to this concern by saying that monetary policy will not produce inflation so long as unemployment is below its “natural rate.” But the natural rate is defined as the threshold rate of unemployment above which inflation will not accelerate in response to expansionary monetary policy, and therefore, this statement is tautological. In practice, the problem is whether one can ever say, reliably, whether the current rate of unemployment is above or below the natural rate. Many economists, but most notably Athanasios Orphanides, attribute the mistakes that the Federal Reserve made during the high-inflation years of the 1970s to these fundamental difficulties in measuring the natural rate of unemployment and the closely related concept of the “output gap” between actual GDP and its long-run potential. 

If adjusting monetary policy to changes in unemployment were the only way that the Fed could provide further support for the economic recovery, we, like the authors of the Board studies, might be willing to accept these risks. But, to the contrary, the Federal Reserve already has at its immediate disposal another, more reliable tool, in the form of the two percent long-run inflation target that the FOMC adopted in January 2012. Actual inflation has come in below two percent since the beginning of the 2008 recession. Extended periods during which actual inflation runs below target almost always indicate that monetary policy has been excessively tight. The inflation target alone, therefore, justifies the caution that many FOMC members have expressed against raising interest rates by too much, too soon. Besides, unlike policies that set ambitious thresholds for unemployment, the Fed’s existing inflation-targeting framework requires no elaborate and potentially confusing communication strategy: FOMC officials can easily explain that their two percent inflation target is a perfectly symmetric one, justifying monetary accommodation when inflation and inflation expectations fall below two percent but, likewise, requiring decisive moves to tighten policy when inflation rises above two percent persistently.

But that does not mean that all types of expansionary policy are justified by low inflation. During most periods, when the zero bound is not a constraint, lowering the federal funds rate target will do the job. When the zero bound is binding, however, the Fed should target an increased rate of growth of its balance sheet through open market purchases of U.S. Treasury bills only, so as to provide for growth in the monetary base that can then translate effectively into growth in broader measures of the money supply and, ultimately, economy-wide prices as well. By expanding its purchases to include long-term, mortgage-backed, and privately-issued securities, the Fed has confused the public and overstepped its bounds by attempting to allocate credit within the private sector instead of more effectively controlling inflation. It has, in other words, attempted to implement fiscal, as opposed to purely monetary, policies. Furthermore, these purchases can all too easily create adverse ramifications later on: If the sale of long-term securities would risk Fed insolvency, a necessary exit from the current, expansionary regime might be delayed, leading to much higher inflation down the road. 

Of course, despite the economic arguments against targeting unemployment, there are obvious political reasons that drive the Fed to do so. Tellingly, both of the two Board staff studies make reference to the Federal Reserve’s statutory “dual mandate” to support their arguments in favor of a more aggressive policy response to unemployment as well as inflation. This points to a more fundamental problem. The 1977 amendment to the Federal Reserve Act instructing the Fed to simultaneously achieve stable prices and maximum employment mistakenly views these goals as competing ones, to be managed by exploiting a stable Phillips curve. An enormous body of economic research compiled since the late 1970s has used the Fed’s unsuccessful experience in lowering unemployment by raising the rate of inflation to show instead that by stabilizing inflation first, the Fed best provides conditions under which unemployment can decline. This modern view has been laid out in detail by none other than current Fed Chairman Ben Bernanke himself, in a co-authored volume titled Inflation Targeting.  Most central banks – including the European Central Bank – follow a single mandate of price stability precisely because it is generally understood that this objective results in both lower inflation and lower unemployment. And many economists – including former Fed Chairman Paul Volcker – advocate the repeal of the Fed’s dual mandate and the adoption of a single mandate of price stability. 

President Obama, together with members of Congress and the Senate, would do well to set aside their partisan squabbles, if only for a moment, to update the Federal Reserve Act to reflect all that has been learned about what monetary policy can and cannot do, and thereby support the FOMC in its efforts to bring inflation back to target and improve the robustness of the ongoing economic recovery. By seizing this low-hanging fruit, our nation’s leaders would do much to help all Americans.


Charles W. Calomiris is the Henry Kaufman Professor of Financial Institutions at Columbia University and Peter Ireland is a professor of economics at Boston College. Both are members of the Shadow Open Market Committee.