To improve Fed policy, improve communications

“Price stability is the responsibility of the Federal Reserve and serves as the bedrock of our economy. Without price stability, the economy does not work for anyone. In particular, without price stability, we will not achieve a sustained period of strong labor market conditions that benefit all.” Federal Reserve Chairman Jerome Powell, Press Conference following FOMC meeting, September 21, 2022.

“When I was at the Federal Reserve, I occasionally observed that monetary policy is 98 percent talk and only two percent action. The ability to shape expectations of future policy through public statements is one of the most powerful tools the Fed has.” Former Federal Reserve Chairman Ben Bernanke, “Inaugurating a New Blog,” March 30, 2015.

Since May 2021, we have criticized the Federal Reserve’s lagging response to surging inflation. In our view, both policy and communications were inadequate to address the looming challenge. Early this year, we argued that the Fed created a policy crisis by refusing to acknowledge the rise of trend inflation, maintaining a hyper-expansionary policy well after trend inflation reached levels far above their 2% target, and failing to articulate a credible low-inflation policy.

Fed policymakers still face major challenges, both in what they say and what they do. We trace much of the current difficulties to the FOMC’s revised strategy. Implemented in August 2020, the current policy framework incorporates inconsistencies and asymmetries that weaken the credibility of the price stability commitment and complicate communication. First, policymakers say they are implementing a strategy of flexible average inflation targeting, but they do not specify a horizon—either looking back or looking forward—over which they are doing the averaging. Second, while the FOMC introduced a “make-up strategy” for inflation shortfalls (see our earlier post), participants have suggested that it does not apply to inflation overshoots, creating an asymmetry that is incompatible with an average inflation targeting regime (see, for example, the Chair’s January 2022 press conference transcript, page 26). Combined, these flaws make it more difficult to define a policy reaction function and implicitly expand policymaker discretion.

Implementation of the new framework also has made matters worse. While the revised strategy explicitly rejects efforts to preempt surges in inflation, presumably because inflation is difficult to forecast, in 2021, FOMC policymakers counted on projections that inflation would recede on its own. Finally, with regard to communications, the Fed failed to implement simple reforms (proposed in 2019 and 2020) that would clarify their policy reaction function and improve credibility.

Against this background, we suspect that, if the FOMC wishes to reach its inflation target within two to three years, the policy rate will need to rise as a much as a percentage point beyond the highest projection of any FOMC participant (4.9%) reported in the September Summary of Economic Projections (SEP Table 1).

Nevertheless, we commend the FOMC for its recent efforts. Not only is policy moving quickly in the right direction—with the policy rate rising 300 basis points in a mere six months—but communication improved markedly. In particular, despite the increasing likelihood of a near-term recession, Chair Powell made clear that price stability is necessary for achieving the second part of the Fed’s dual mandate (see the opening citation). That implicit hierarchy of goals—making price stability a prerequisite for achieving maximum sustainable employment—generally characterized FOMC behavior for many years prior to the pandemic. While it would be difficult to prove, we suspect that the combination of the Fed’s recent promise to make policy restrictive, along with its improved communications, is playing a key role in anchoring longer-term inflation expectations.  

In this post, we focus on central bank communication and its link to policy setting. By far the most important goal of communication is to clarify the authorities’ reaction function: the systematic response of central bank policy to prospective changes in key economy-wide fundamentals—usually inflation and the unemployment rate. Following the COVID shock, both policy and communication focused on ensuring the credibility of a commitment to “low-for-longer” interest rates. With hindsight, most observers can now see that the resulting mix of forward guidance and large-scale asset purchases tied policymakers to a hyper-stimulative mast for too long.

Even when the FOMC finally began to tighten in March of this year, communication focused on matters that were at best secondary—for example, on the path of the balance sheet or the scale of a policy rate change at a specific FOMC meeting. While this kind of information is of value to financial speculators, it is not a substitute for explanations of the central bank’s reaction function. The extreme short-run focus may even be counterproductive if it temporarily limits market volatility, thereby easing financial conditions. Put differently, so long as policymakers are confident that the financial system is stable, it makes no sense to provide investors temporary comfort if the goal is to tighten financial conditions.

To anticipate our conclusions, we argue for two changes to the FOMC’s quarterly SEP. First, we encourage publication of more detail on individual participants’ responses. The goal is to illuminate the reaction function by making it possible to link individual projections of inflation, economic growth, and unemployment to the path of the policy rate. Second, we see a role for scenario analysis in which FOMC participants provide their anticipated policy path contingent on one or more adverse supply shocks that present unappealing policy tradeoffs (for example, between the speed of returning inflation to its target and the pace at which the unemployment rate returns to its sustainable level).

To set the stage, we start with a simple scatter plot that highlights the current challenge: namely, trend inflation that is well above target and a labor market that is very tight. The vertical axis shows the excess of the Fed’s preferred measure of trend inflation (measured by the percent increase from a year ago of the price index of personal consumption expenditures excluding food and energy) above their two-percent target. Along the horizontal axis, we plot the gap between the current unemployment rate and the Congressional Budget Office’s noncyclical rate of unemployment (u-u*). In the top left quadrant, which we label overheated, inflation is high (>2%) and unemployment is below u*. Continuing clockwise, in the top right, inflation is high and unemployment above u*, so there is stagflation. We label the lower right sector, where inflation is below target (<2%) and unemployment above u*, as undesirable disinflation. Finally, in the lower left, inflation is below target and unemployment below (but usually close to) u*, so we label this quadrant as bliss.

Trend Inflation and Unemployment Rate Gaps (scatter chart), 1960-August 2022

Notes: Trend inflation is measured by the annual percent change of the price index of personal consumption expenditure excluding food and energy. The deviation from target subtracts 2 percentage points from this measure. The unemployment gap shows the actual minus the noncyclical rate of unemployment (NROU). Source: FRED.

During the COVID recovery, the unemployment gap plunged from a peak of +10 percentage points (masked by the truncation of the horizontal axis at 6 percentage points) to its September 2022 level of nearly -1 percentage point. Over the same period, the deviation of trend inflation from target rose from a trough of -1 percentage point to beyond +3 percentage points. The red squares denote the points along this 2020-22 trajectory, with the red diamond (enclosed in black) denoting the latest (August 2022) observation. Importantly, the dashed line (with the arrow) connecting the two red circles shows the extraordinarily rapid shift up and to the left from March 2021 to January 2022. Indeed, this 10-month rise brought trend inflation to above 5 percent, its highest level since 1983.

Of course, the relationship between inflation and unemployment depends on inflation expectations. Consequently, it does not present a stable, structural link that policymakers can exploit. Nevertheless, the chart makes clear that, for the Fed to achieve its inflation target, it should aim to set policy rates at a level sufficient to restrain the growth of aggregate demand below that of aggregate supply. This is the lesson embedded in monetary policy rules, including those tracked by the Fed in its semi-annual Monetary Policy Report to Congress. (The four rules published there point to the need for policy rates to be in the 3 to 7 percent range since at least the final quarter of 2021.)

So, do we think that policy has improved markedly in the past few months? The answer is clearly yes. As is widely noted, the FOMC is in the midst of the most rapid increase of policy rates in 40 years. But this shift merely constitutes a belated catch-up (see our earlier post that highlights the risks of acting too little, too late). Far more important is that the FOMC’s guidance now focuses clearly on the need for a restrictive stance to restore price stability, combined with humility regarding the level of restraint that ultimately will be needed to bring inflation back to 2%.

We can see this clearly in the history of the FOMC’s quarterly SEP. For each SEP, every FOMC participant submits their projection of the policy rate (roughly) two years ahead. In the following chart, we show the median (red line), the interquartile range (pink-shaded area), and the full range (gray area) of the projections. So, for example, the median December 2019 federal funds rate projection for end-2021 was 1.9% (with a range of 1.6% to 2.4%).  Importantly, as recently as March 2021, the two-year-ahead median projection (for end-2023) still had the federal funds rate at 0.1%, albeit with a range to 1.1%. The change over the past 16 months is striking. More important, investors responded to the Fed’s evolving quantitative guidance, as market expectations for the federal funds rate at the end of both 2022 and 2023 now exceed 4.5 percent (shaded black circles).

Actual end-year federal funds rate, SEP quarterly projections from two years earlier, and market expected federal funds rates for end-2022 and end-2023, 2012-2024P

Notes: The dashed line shows the actual federal funds rates for the end of each year. The red line shows the two-year earlier median SEP projection of the end-year federal funds rate. For example, the projections shown in 2014 are from the 2012 SEPs. The pink-shaded area is the interquartile range of the two-year earlier SEP projections, while the gray-shaded area displays the full range of the projections. The black shaded circles are the probability-weighted averages of the CME FedWatch tool projections (based on Fed Fund futures prices as of October 14, 2022) for December 2022 and 2023, respectively. Note that “two years earlier” is only an approximation: the SEP projection is for the end of a calendar year, so the horizon is about two years for each December projection, but somewhat longer for each March, June and September projection. The gap reflects that there was no SEP at the start of the pandemic shutdown in March 2020.

So, what can the FOMC do to ensure its credibility? The most important thing is to match future policy deeds to the Chair’s recent words: that means hiking rates sufficiently to restore price stability even if, as we expect, the appropriate level of policy restraint will be accompanied by a recession. Aside from standing behind its revitalized price stability commitment, perhaps the most important thing would be to announce an early review of its flawed policy strategy.

Short of a full review, we see two meaningful reforms that would improve both communication and policy by facilitating a broader understanding of the Fed’s policy reaction function. The first is to publish in each SEP a table (or matrix) that links the projections of each FOMC participant for interest rates, economic growth, unemployment, and inflation both on a given date and over time. Currently, this information is revealed only after a lag of five years (see, for example, Table I.A. in the March 2016 SEP). As we previously argued, timely disclosure would focus attention on the FOMC consensus that underlies the reaction function, helping observers to anticipate when and why that consensus might shift.

Second, the Fed should introduce a procedure for scenario analysis that would help reveal the relative importance that the FOMC assigns to inflation and unemployment developments. If we have the SEP, why do we also need this? The problem is that the SEP provides information about participants’ expected outcomes, together with non-quantitative indications of uncertainty and risk relative to that baseline case. These data mask how policymakers would respond to possible, or even quite probable, alternative scenarios that involve sensitive policy trade-offs.

Scenario analysis offers a natural antidote to these weaknesses. For example, in the second quarter of 2020, when both unemployment and uncertainty were very high, Bordo, Levin and Levy (BLL) proposed the use of illustrative scenarios as a supplement to the SEP. At the time, they identified three alternatives that would inform observers about the FOMC reaction function: namely, an incomplete recovery, a benign outcome supported by effective vaccines, and a severely adverse scenario involving high unemployment, financial strains and undesirable disinflation. As BLL make clear, by using scenario analysis, the FOMC can develop and articulate plans that are contingent on key developments without making commitments. Just as Chair Powell emphasizes that SEP projections are not Committee plans, we share the BLL view that communicating contingent thinking in a period of heightened uncertainty can help build confidence in the FOMC’s preparedness and in its commitment to meet its mandated objectives.

In the current context, consistent with FOMC participants’ views that unemployment and inflation risks remain upward-biased (see SEP Figures 4.B. and 4.C.), in our view the most useful alternative scenario for the FOMC to explore would be a lingering stagflationary one in which inflation remains stubbornly above target for longer and with higher unemployment than SEP projections currently anticipate. Eliciting and publishing FOMC participants’ quantitative policy responses to such a stagflationary scenario would allow observers to assess the FOMC’s willingness to “stay the course” needed to bring inflation back to 2%. That kind of credibility can go a long way to keeping long-term inflation expectations anchored and limiting the cost of lowering trend inflation.

To conclude, over the past six months, the path of Fed policy shifted dramatically. We applaud the changes in both what policymakers are doing and saying. While we still expect that bringing inflation down to 2% will require higher interest rates than either policymakers or markets currently believe, things are clearly moving in the right direction. Furthermore, the tone of Fed communication now displays the sort of resolve needed for policymakers to achieve price stability.

That said, there is room for improvement on at least two fronts. First, to correct the inflationary bias in its current policy strategy, the FOMC should initiate an early review. Second, to improve the communication of its reaction function, they should further enhance the SEP, publishing more detailed information on participants’ projections as well as adding participants’ quantitative responses to scenarios that illuminate contingent policy thinking regarding sensitive policy trade-offs.