Skip to main content

Improving Our Monetary Policy Strategy

(PDF PDF icon)

I thank John Taylor for inviting me to participate in this conference. It’s a real pleasure to be on a panel moderated by Charles Plosser, whom I worked with at the Philly Fed. Although I learned a lot from Charles, the views I’ll present are my own and are not necessarily shared by my fellow panelists, other colleagues on the Federal Open Market Committee, or the Federal Reserve System.

The FOMC currently uses what has been called a flexible inflation-targeting framework to set monetary policy. It is briefly described in the FOMC’s statement on longer-run goals and monetary policy strategy.1 In my view, this framework has served the FOMC well in effectively promoting our policy goals. A milestone was reached in January 2012 when the U.S. adopted an explicit numerical inflation goal. I am certain that Charles remembers very well the careful analysis and discussions that helped the FOMC reach a consensus on the explicit 2 percent goal and the statement that describes the FOMC’s approach to setting policy to promote its congressionally mandated goals of price stability and maximum employment.

The FOMC is currently reviewing its policy framework. I am very supportive of this initiative. As a matter of good governance, a central bank should periodically review its assumptions, methods, and models, and to inform its evaluation, it should seek a wide range of perspectives, including those from experts in academia, the private sector, and other central banks. Another motivation to undertake the review now is that the post-crisis economic environment is expected to differ in some important ways from the pre-crisis world. Based on the aging of the population and the expected slowdown in population growth, higher demand for safe assets, and other factors, many economists anticipate that the longer-term equilibrium real interest rate will remain lower than in past decades.2 In fact, empirical estimates of the equilibrium real fed funds rate, so-called r-star, while highly uncertain, are generally lower than in the past.3 This means there is a higher chance that the policy rate will be constrained by the zero lower bound, and that nontraditional monetary policy tools will need to be used more often. To the extent that these tools are less effective than the traditional interest rate tool or are otherwise constrained, the potential is for longer recessions and longer bouts of inflation well below target.4 In addition, fiscal policy’s ability to buffer against macroeconomic shocks is also likely to be constrained, given projected large fiscal deficits and high government debt-to-GDP ratios.5 This raises the question of whether changes to our monetary policy framework would be helpful in maintaining macroeconomic stability in this environment.

A number of suggestions have been made for alternative monetary policy frameworks that potentially offer some benefits in a low-interest-rate environment. These include setting an inflation target that is higher than 2 percent (an option not being considered by the FOMC in its framework review), using price-level targeting or nominal GDP targeting instead of inflation targeting, targeting average inflation over the business cycle or some other time frame, or using what former Chair Ben Bernanke has called temporary price-level targeting (which is essentially doing inflation targeting in normal times and price-level targeting once the policy rate is constrained by the zero lower bound). An idea that has received somewhat less attention is defining the inflation goal in terms of a range centered on 2 percent rather than a point target.6 Although these alternative frameworks have theoretical appeal, none of them is without implementation challenges. For example, many of them work well in models of perfect credibility and commitment, where the public understands the framework and believes future Committees will follow through, and the Committee actually does follow through, implying that the Committee has control of inflation expectations. Whether these assumptions would hold in practice is an open question. One needs to ask whether it is credible for policymakers to commit to keep interest rates low to make up for past shortfalls of inflation from target even when demand is growing strongly or to act to bring inflation down in the face of a supply shock by tightening policy even in the face of weak demand. It is not clear what actually would happen to inflation expectations in these scenarios despite what is assumed in the models. So the FOMC is going to have to evaluate the assumptions that drive the theoretical appeal of each framework and determine whether in practice the net benefits of any of the alternatives will outweigh those of the flexible inflation-targeting framework, and if not, what, if any, enhancements should be made to our current framework.

Regardless of the framework the FOMC ultimately decides on, the public’s expectations about future monetary policy are an important part of the transmission mechanism of policy to the economy. This means effective communication will be an essential component of the framework. I believe there are ways we can enhance our communications about our policy approach that would make any framework more effective. Let me touch on three.

(1) Clarify how monetary policy affects the economy and which aspects of the economy can be influenced by monetary policy and which aspects cannot.

Monetary policy is more effective when the public’s and market participants’ policy expectations are aligned with our policy decisions. Before this alignment can occur, the public needs to have a basic understanding of our monetary policy goals and what monetary policy can achieve and what it cannot. My concern is that this understanding has diminished since the Great Recession. Regardless of the framework, the FOMC’s strategy document should articulate the relationship between monetary policy and our two policy goals of price stability and maximum employment. We should clarify that over the longer run, monetary policy can affect only inflation and not the underlying real structural aspects of the economy such as the long-run natural rate of unemployment or maximum employment. Although this concept is touched on in our current monetary policy strategy document, I do not think that the public fully understands. Indeed, former Chair Janet Yellen had to explain in one of her post-FOMC meeting press conferences that in an earlier speech, she did not mean to imply that she favored running a high-pressure economy as an experiment to affect longer-run growth and unemployment.7

I think we could do a better job of explaining how monetary policy promotes the economy’s growing at potential and operating at maximum employment. In particular, we tend to move our policy rate up when resource utilization tightens and down when resource utilization eases in order to bring our policy rate into alignment with the economy’s natural rate of interest, which changes over the business cycle as the economy adjusts to shocks. There doesn’t need to be an exploitable Phillips curve tradeoff between the unemployment rate and the inflation rate in order for policymakers to want to respond to changes in the unemployment rate, an indicator of resource utilization.8 The response is not an attempt to actively use monetary policy to affect the longer-run growth rate of the economy or the longer-run unemployment rate. A benefit of explaining things in this way makes it clear that the FOMC is not trying to rob the economy of jobs when it raises interest rates. Another benefit is that it should allay concerns that because the empirical Phillips curve has flattened, monetary policy has become anemic.

Improving the public’s understanding of how monetary policy works and what it can achieve would help not only in normal times but also in bad times. The Great Recession was an enormous negative shock, some part of which was likely permanent or very persistent rather than transitory. Monetary policy should not have been expected to make up for that permanent loss. Fiscal policy should have taken on a larger part of the burden.

(2) Clarify how uncertainty is accounted for in monetary policymaking and incorporate this uncertainty into monetary policy strategy to avoid giving a false sense of precision.

According to Voltaire, “Uncertainty is an uncomfortable position, but certainty is an absurd one.” In our context, this means it is important to convey that monetary policymakers have to deal with uncertainty in several forms. Monetary policy has to be forward looking because it affects the economy with a lag, but the economy is buffeted by shocks that can lead economic conditions to evolve differently than anticipated. Moreover, our view of economic conditions in real time can be cloudy because the data come in with a lag and many economic data are revised over time. In addition, there is model uncertainty.

The public needs to understand that given the lags and revisions in the data, incoming information can alter not only the policymaker’s view of the expected future evolution of the economy but also his or her understanding of current and past economic conditions. New information could alter the expected future path of policy and might even result in ex post regret of a recent action. Robert Hetzel says that policymaking has a flavor of “guess and correct.”9 It is a normal part of monetary policymaking that policymakers will always be learning about whether their policy settings are the appropriate ones to promote their goals.

The public has to hold the FOMC accountable for its performance, but it should not hold monetary policymakers to an unrealistic standard. The FOMC took an important step in communicating uncertainty when it began showing 70 percent uncertainty bands around the median projections of FOMC participants, but these are not emphasized. I think they deserve more attention and should be released at the time of the post-FOMC press conference. They are a good illustration of the reasonable amount of deviation to expect between the projections and outcomes. Some have argued that the FOMC’s projections of appropriate monetary policy, the so-called dot plot, should be dropped because actual policy can differ from the projections. I think that would be a mistake. The dots can change over time because of economic developments, but that’s a design feature, not a flaw. Omitting the dot plot would not eliminate the uncertainty around the projections, the divergence in views across FOMC participants, or the fact that policymaking always entails learning and recalibration, but it would be a significant step back in transparency.

We need to recognize uncertainty in our broader monetary policy strategy as well. Consider the FOMC’s inflation target. After much deliberation, the Committee chose a point target instead of a range and a total inflation measure rather than a core measure. While there were arguments on both sides, the Committee was persuaded that a point target would better anchor inflation expectations. Implicit in the choice was that the Committee would tolerate small deviations from target given the precision with which we can measure inflation, the precision with which we can guide the economy, and the typical revisions to the PCE inflation measures, which tend to be revised up over time.10 It is interesting to think through whether our policy choices or communications since 2012 might have differed had the Committee opted for a range rather than a point target, as some other central banks do, and for a core measure rather than a total measure of inflation. These data revisions and measurement issues, as well as potential difficulties in maintaining anchored inflation expectations during the periods of higher inflation meant to make up for periods of lower inflation, and vice versa, would seem to be amplified in price-level targeting and nominal GDP targeting frameworks.

(3) Clarify our monetary policy strategy by taking a more systematic approach to our policy decisions and in how we communicate those decisions.

Households, businesses, and investors make economic and financial decisions based on their expectations of the future, including the future course of monetary policy, and the FOMC strives to avoid surprising the public with its policy decisions. The communications challenge for the FOMC is to give the public a good sense of how policy is likely to respond conditional on how the economy evolves without implying that policy is pre-committed to a particular policy path regardless of how the economy evolves. Essentially, the FOMC needs to convey the strategy it uses to determine its policy actions over time to promote achievement of its policy goals, i.e., its reaction function. And this will be true regardless of which monetary policy framework the FOMC ultimately adopts. Ironically, the FOMC’s strategy document does not offer much in the way of strategy, and this can lead to a misunderstanding that our policy decisions are discretionary. The term “data-dependent” has been used to explain the FOMC’s policymaking strategy, but this term could be potentially misinterpreted as suggesting that policy will react to every short-run change in the data rather than the accumulation of changes that affect the medium-run outlook.

A more systematic approach to setting monetary policy can better align the public’s policy expectations with policy decisions and help to reduce some of the uncertainty around how we conduct monetary policy. It can help insulate monetary policy from short-run political considerations, and it can also offer more policy continuity over time as Committee members change. In a time of rising public skepticism about “experts,” which can undermine public trust in institutions, being systematic will help the public understand how our decisions are actually made, which can enhance the Fed’s credibility.

The question is how to ensure that we are setting policy systematically and how to convey this to the public. I have three suggestions. First, while judgment will likely always be a part of policymaking, simple monetary policy rules can play a more prominent role in our policy deliberations and communications.11 The FOMC has been reluctant to relinquish policymaking to following a simple rule, because no one rule works well enough across a variety of economic models and circumstances. But the Board of Governors has begun to include a discussion of rules as benchmarks in the monetary policy report,12 and frameworks that try to build in some commitments and constraints on future policy actions, such as price-level targeting, average inflation targeting, and nominal GDP targeting, are being discussed. This suggests that systematic policymaking is garnering more support. As a first step, selecting a few benchmark rules that have been shown to yield good economic outcomes and using these as reference points to aid policy discussions and communicating why our policy may or may not differ from the rules’ policy descriptions could go some way in ensuring that our decisions are derived in a systematic way and could help us explain our own policy reaction function to the public.

A second suggestion is to enhance our own FOMC projections by asking the participants to provide a set of economic projections conditioned on a common policy path, in addition to the current projections, which are conditioned on each individual participant’s view of appropriate policy. This common path might come from a policy rule. This would be a step toward achieving a coherent consensus FOMC forecast, which has been a challenge but which could serve as the benchmark for understanding the FOMC’s policy actions and post-meeting statements, a recommendation I have made in the past.13

My third suggestion to help communicate systematic policymaking is to make our post-meeting FOMC statement consistent from meeting to meeting and less focused on short-term changes in the data released between FOMC meetings and more focused on the medium-run outlook and a consistent set of indicators on inflation, inflation expectations, the unemployment rate, employment growth, output growth, and financial conditions. Each statement could provide the rationale for the policy decision in terms of how accumulated changes in this consistent set of economic and financial conditions have or have not influenced the Committee’s assessment of the factors relevant for policy, i.e., the arguments in our reaction function. The statement would also consistently articulate the Committee’s assessment of risks to the outlook and other considerations that the Committee is taking into account in determining current and future policy. This assessment would be informed by the analysis of alternative forecast scenarios, which are discussed at each FOMC meeting. If we provided more consistency about the conditions we systematically assess in calibrating the stance of policy, the public and market participants would get a better sense of the FOMC’s reaction function over time and their policy expectations would better align with those of policymakers.

I note that all of the suggestions I have made today are relevant regardless of the framework the FOMC ultimately decides to use for setting monetary policy.


  1. See FOMC (2019). Return
  2. See Mester (2018a). Return
  3. For FOMC projections, see FOMC (2014) and FOMC (2019). For a review of the literature on the equilibrium interest rate, see Hamilton, et al. (2015). Return
  4. Other government policies might also be brought to bear to increase the long-term growth rate and equilibrium interest rate, which would give monetary policy more room to act. Such policies would focus on increasing productivity growth and labor force growth. Return
  5. See Peek, Rosengren, and Tootell (2018). Return
  6. For further discussion of these monetary policy frameworks, see Mester (2018b) and Mester (2018c). Return
  7. See Yellen (2016), p. 9. Return
  8. Brainard (2018) discusses the shorter-run neutral rate and longer-run equilibrium interest rate. Return
  9. See Hetzel (2019). Return
  10. Croushore (forthcoming) finds that the average revision from initial release to first annual benchmark revision to four-quarter PCE inflation over the period 1965Q3 to 2015Q4 was 0.10 percentage point and the average revision to four-quarter core PCE inflation over the period 1995Q3 to 2015Q4 was 0.14 percentage point. Return
  11. The Cleveland Fed provides updates for a set of monetary policy rules and their outcomes across several forecasts on the Cleveland Fed’s website, and the Federal Reserve Board’s Monetary Policy Report now includes a section on policy rules. See Federal Reserve Bank of Cleveland (March 22, 2019). Return
  12. See Board of Governors (2019), pp. 36-39. Return
  13. See Mester (2016). Hetzel (2019) also proposes a method to determine an FOMC consensus forecast that would entail the Committee’s agreeing to its preferred reaction function at the start of each year, and then using an iterative process among FOMC members based on that rule and the Board staff’s economic model. Return


Upcoming EventsSEE ALL