Thank you, Isabel, and thank you to the organizers for the invitation to be part of this discussion.1
As we get closer to the dinner hour, I will give you two thoughts to chew on about how monetary policy transmission has worked in the recent past and how that affects my view of appropriate actions when facing current challenges.
The first of these is that initial conditions are crucial. To decide where policy should go, you need a clear sense of where you are starting from. By “initial conditions,” I mean what is currently happening—not some average of experience in the past.
This lesson was brought home to me during the rapid escalation of inflation in the United States following the pandemic. Based on past experience, a considerable share of the economic profession believed that the rapid tightening of financial conditions needed to bring down inflation would unavoidably cause a sharp increase in unemployment. And while this expectation was a good summary of what had happened in the past, in 2022, it was a poor predictor of what would happen from tightening policy because initial conditions at that time were so different.
In particular, the combination of a negative labor supply shock along with a booming economy fueled by fiscal and monetary stimulus led to a situation in early 2022 where the ratio of job vacancies to unemployed workers was 2—a level never seen before. This initial condition for job vacancies was critical to understanding how tightening monetary policy would affect the labor market.
Historically, changes in labor demand had minor effects on vacancies. But that wasn’t the case from 2022 through 2023. Employers decreased labor demand by greatly reducing vacancies instead of shedding workers, and as I predicted in May 2022, that led to only a modest increase in unemployment.2
Along these lines, the initial size and suddenness of economic and monetary policy shocks need to be considered in order to assess the transmission of monetary policy. Economists typically use models that are linear and estimated based on historical data. These models tend to predict that, for example, a doubling of an interest rate shock means doubling the effect on the economy. But large shocks lead to behavioral changes that alter the transmission of monetary policy in nonlinear ways.3 The economists in the audience will recognize this proposition as an application of the Lucas critique. First, I argued in a 2023 speech that the unusual speed with which the Federal Open Market Committee (FOMC) had tightened monetary policy in 2022 and 2023 would cause more rapid changes in behavior, and the long lags typically associated with monetary policy actions might be a lot shorter. Second, research on “rational inattention” shows that people tend to ignore small changes in economic conditions but react to large ones. For example, if big shocks lead firms to increase the frequency of price changes, the Phillips curve steepens, which changes how monetary policy actions are transmitted to inflation and the real economy.
The 2021–22 experience shows how important it is for central bankers to consider initial conditions. For the academics here, think of this as a dynamic programming problem. To determine the optimal path forward, you need to know where you are starting from. That is why we list the current state variables as the arguments of the value function when determining the optimal policy path. What we don’t do is list the “historical average” of the state variable in the value function.
My second thought to chew on is about a subject that has been getting attention lately: providing forward guidance about the future path of monetary policy. I continue to believe that forward guidance can be a valuable tool that has, at times, significantly strengthened policymaking and will continue to be useful. But forward guidance is more art than science, and there have been times when it has hindered, rather than helped, policymaking. When it works, forward guidance can change economic conditions more quickly than adjusting the policy rate alone. For example, in September 2021, the Committee signaled that it would tighten policy in coming months. Even though we did not change the policy rate until March 2022, from September 2021 through mid-February, the two-year Treasury yield rose nearly 200 basis points. That rise effectively shaved off about 6 months from the usual 12- to 24-month lag that one might conjecture would be needed to see the 200 basis points of actual tightening affect the economy.
But forward guidance can impair policy transmission if the guidance is too strong or rigid.4 A case in point is the forward guidance the FOMC adopted in September 2020 when it said that liftoff from the effective lower bound would occur when conditions were consistent with maximum employment and when inflation had risen to 2 percent and was “on track to moderately exceed 2 percent for some time.”5 In addition, the FOMC’s guidance signaled that rate increases were on hold until tapering was well under way. This guidance didn’t change over the course of 2021 as inflation was rising quickly above 2 percent and unemployment was rapidly falling, leaving the public to wonder what “for some time” meant. In the end, this restrictive guidance tied the hands of the FOMC in 2021 and unnecessarily delayed rate increases.
Forward guidance is also problematic when there are potentially different economic scenarios confronting policymakers, each with a significant probability of occurring and requiring different policy paths. One cannot simply take a weighted average of these scenarios as the “base case” and use it to give forward guidance. Let me illustrate. I am driving and come to an intersection. The light turns yellow. I can stop before the intersection and sit through the light. Or I can drive through the intersection and keep going. But my base case is not to stop in the middle of the intersection. In these divergent situations, it’s much harder to give forward guidance, and, as a result, it is less useful.
So the takeaway is that forward guidance can help speed up policy transmission, but if it is not flexible enough, it can hinder policy transmission. And, in some cases, it’s best not to use it at all.
1. The views here are my own and are not necessarily those of my colleagues on the Federal Reserve Board or the Federal Open Market Committee. Return to text
2. See Christopher J. Waller (2022), “Responding to High Inflation, with Some Thoughts on a Soft Landing,” speech delivered at the Institute for Monetary and Financial Stability Distinguished Lecture, Goethe University Frankfurt, Germany, May 30; and Chris Waller and Andrew Figura (2022), “What Does the Beveridge Curve Tell Us about the Likelihood of a Soft Landing?” FEDS Notes (Washington: Board of Governors of the Federal Reserve System, July 29). Return to text
3. See Christopher J. Waller (2023), “Big Shocks Travel Fast: Why Policy Lags May Be Shorter Than You Think,” speech delivered at the Money Marketeers of New York University, New York, New York, July 13. Return to text
4. I addressed this problem in a June 2022 speech and expanded on the issue in a FEDS Note with Jane Ihrig; see Christopher J. Waller (2022), “Lessons Learned on Normalizing Monetary Policy,” speech delivered at “Monetary Policy at a Crossroads,” a panel discussion hosted by the Dallas Society for Computational Economics, Dallas, Texas, June 18; and Jane Ihrig and Chris Waller (2024), “The Federal Reserve’s Responses to the Post-Covid Period of High Inflation,” FEDS Notes (Washington: Board of Governors of the Federal Reserve System, February 14). Return to text
5. See Board of Governors of the Federal Reserve System (2020), “Federal Reserve Issues FOMC Statement,” press release, September 16, paragraph, 4. Return to text
