Remarks as prepared for delivery.
Good evening, everyone. Thank you, Patricia,1 for the warm introduction. It's a pleasure to be with all of you.
As always, the views I express today are my own and do not necessarily reflect those of my FOMC colleagues or anyone else at the Federal Reserve.
I want to start with a question that sounds almost too simple to deserve serious attention: Should a central bank tell you what it plans to do with interest rates?
I suspect that for most people in this room, the answer is, "Yes, obviously. More information is better than less." I'd like to suggest a more useful question: Do we understand each other?
By "we," I mean the Fed and all of you: markets, businesses, and households. Do you understand how we read the economy and how we're likely to respond as it changes? And do I understand key market signals?
I think that understanding matters more than the specific ways we communicate. When we understand each other, we don’t need to say as much when new data arrive. You can anticipate how we'll respond, and we can act in ways that are steady and, for the most part, predictable.
That matters for the dual mandate. The expected path of interest rates shapes decisions about hiring, investing, and borrowing. When people understand how we are likely to respond to economic developments, three things follow. Rates can more effectively guide spending in the right direction. We can look through temporary shocks without losing credibility. And rates move in largely predictable ways, without unnecessary jolts that force households, businesses, and financial institutions to adjust in a hurry. All of that helps us deliver price stability and maximum employment.
To understand where communication stands today, it helps to look at how it has evolved. For much of the Fed's history, it did not even tell you what it had done. Until 1994, the FOMC didn’t even announce its policy decisions. Since then, the Fed has said steadily more: statements describing the balance of risks, the Summary of Economic Projections, press conferences, the dot plot, and, especially in the wake of the Great Financial Crisis, forward guidance as an active policy tool.2 3 The trend is clear: more information, shared more often, in more forms. But volume is not the same as clarity. The measure is not how much we say. It is whether we are understood.
I’m not the only one thinking about this. The Fed has a task force dedicated to communications, and I’m looking forward to their insights.
Here is how I will proceed. I will look at research on expectations and credibility and how markets coordinate around what we say. Then I will offer a test I use to evaluate any form of communication.
Expectations
So, what makes communications work? I find it helpful to start with expectations.
Expectations matter. The rate the Fed controls is an overnight rate, but almost no household or business borrows overnight. Families borrow for 30 years to buy a house. Businesses borrow for five years to buy equipment. What they pay depends on where markets expect short-term rates to go over that whole period. So, their investment decisions depend on the whole path of rates, not just today's rate.
Markets know this better than anyone. A 10-year Treasury yield reflects, in part, expectations for the next 10 years of Fed decisions, so markets naturally move when they learn something new about how policymakers think.
That means influencing the expected path is part of how monetary policy works. The Nobel Committee recognized this insight in awarding Michael Woodford this year's prize in economics, together with Bob Hall.4 5 One way to shape that path is through forward guidance: communicating where rates are likely to go. Over time, the term has broadened. Today it often means describing where the economy is expected to go and what policy would fit that outlook. Think of it as a conditional forecast: if the economy evolves this way, policy will likely evolve that way.
A conditional forecast can help illuminate something else I want to talk about tonight: the reaction function, or how policy adjusts to changes in the economy. But the two are not the same thing. In the early 1990s, John Taylor showed that a simple rule, now known by his name, did a remarkably good job of describing monetary policy, at a time when the Fed did not even announce its rate target.6 People could understand how the Fed responded to the economy simply by watching what it did. Forward guidance is one way to build understanding, not a prerequisite for it.
Guidance about the rate path is most valuable, I think, when the policy rate is near the zero lower bound, and the Committee's usual tool to fight a recession, lower rates, isn't available. There, guidance can be a promise to depart from the normal reaction function and keep rates lower for longer than economic conditions would otherwise call for.7 This should lower longer-term rates and give the Committee a way to provide stimulus when it can't cut the overnight rate.
What about in normal times? I think this is a more open question. I lean toward the view that guidance about the rate path is less valuable in normal times, because people can usually work out the path themselves if they understand the reaction function. When markets, households, and businesses understand how the FOMC is likely to respond to economic news, they can update their expectations as the data come in. What matters in normal times is that understanding.
Let me give you an example. The Committee raised rates in September and held them steady in October. As I said in a speech earlier this month, I supported both decisions, and I see holding rates steady in October as a matter of pace, not direction.8 A pause like that works best when the direction of policy is well understood. In October, markets expected further tightening, so the pause did not leave policy behind where it needed to be. A shared understanding of the reaction function allowed policy to move deliberately and gave people time to adjust, without the risk that a pause was mistaken for a stop.
A reaction function is also conditional: it tells you not just where policy is headed, but what would change it. In that same speech, I said I would want to move more quickly if inflation expectations drift up, if wage pressures build, or if underlying inflation moves higher.
Saying more matters most when policy departs from what a standard reaction function would suggest. Take a shock that pushes prices up but whose effect is likely to fade. The standard Taylor rule would call for higher rates. If long-run inflation expectations are well anchored, it can be appropriate, though, to respond less, or not at all. Someone who understands only the standard rule might read that as going soft. That is when it helps to explain why: what I think is temporary, what I am watching to make sure it stays that way, and what would change my mind. Inflation expectations should stay anchored if people understand that I will act if the effect doesn’t fade.
Of course, when something big changes, understanding can't be assumed, and saying more can be worth it. But offering a precise path then would claim more than I know. What I can do is explain how I'm thinking.
Credibility
Everything I've said so far depends on one thing: credibility. When I describe how I will respond, you have to believe that I will, in fact, respond that way. The same is true for the Committee as a whole.
The research that explains why credibility matters was not written with communication in mind, but it is very relevant. In the late 1970s, Finn Kydland and Edward Prescott identified a problem called time inconsistency, which helped them earn the 2004 Nobel Prize.9
Here is the idea. A central bank can promise to act in a certain way in the future to influence how people act today. But when the future arrives, the bank may have an incentive to break its word, and in that moment, breaking it can look like the best choice. If the public sees that coming, its expectations and behavior won't respond to the promise in the first place.
With Thanksgiving around the corner, let me bring this to life. Parents everywhere are telling their kids, "Eat your peas and carrots, or no pumpkin pie." When dessert arrives, the vegetables are still on the plate. The parents can follow through, but then the kids will cry and Thanksgiving is spoiled for everyone. Just giving them the pie seems easier. Of course, the kids saw this coming. They knew they'd get the pie, so they skipped the vegetables.
The research points to three ways to solve this problem.
The first is to set a rule and make it hard to break. The parents announce "no clean plate, no dessert" before dinner and hand the pie to a grumpy uncle to enforce it. Countries have spent decades building their own versions of this. Legislative mandates are the rules and independent, but accountable, policymakers are the grumpy uncles.
The second is reputation.10 Thanksgiving comes once a year, but parents and children eat together every day. Parents who hold firm at every meal end up with promises that are believable. A central bank builds its reputation the same way, by acting consistently over time. A reaction function that is followed year after year is credible.
The third way to solve this problem is to better match the central bank's promises with its incentives. After the Great Financial Crisis, the Committee's early forward guidance was tied to the calendar.11 But if the economy changed suddenly, a calendar-based commitment might have to be broken. So, guidance shifted to being tied to unemployment and projected inflation, the outcomes the Committee is charged with delivering.12 In my view, that made it more believable.
But I don't take credibility for granted. Parents don't get to coast on the reputation they've built. Neither can a central bank.
When Everyone Is Listening
So far, I've argued that communication works best when it builds a shared understanding of the reaction function, and that credibility is essential, because people act on messages they trust. Now I want to turn to a risk that comes from how closely Fed communications are watched.
The caution comes from work by Stephen Morris and Hyun Song Shin on how markets coordinate around public signals.13 Their insight is that when payoffs depend on what others do, as they often do in markets, people care about being right, and they also care about being aligned with everyone else.
Think about what that means for central bank communication. Statements, press conferences, and the dot plot are all public. Everyone sees them, and everyone knows that everyone else sees them. So, they do two things. They inform, and they become a reference point, something people can coordinate around precisely because they are shared.
That creates a risk. People might give a public signal more weight than their own information. That is not irrational. If you know others are watching the same signal, leaning on it makes sense. But if the signal is less precise than people believe, or is misread, everyone can end up moving in the same direction at the same time.
The 2013 episode known as the taper tantrum shows how this can play out. When the Fed discussed slowing asset purchases, markets interpreted it as a signal that rate increases would come sooner. Given what had been said, this was a reasonable conclusion. And because everyone was reaching the same conclusion at the same time, yields rose sharply as investors adjusted together. Yet the Committee saw decisions about rates and about asset purchases as separate. What stands out for me from this episode is how much is riding on clear communications.
A less dramatic example involves the Summary of Economic Projections. There is some evidence that the public's forecasts of the policy rate do not fully update as new information arrives.14 Instead, they stay anchored to the latest dot plot, even after it has gone stale.
To be fair, some of this risk is theoretical, and there is real debate about how often it matters. Some researchers have argued that the problem arises only when the public signal is much noisier than what markets already know.15 I read this research as a caution, not a prediction.
I take two lessons from this.
The first lesson is care. Fed communications carry a lot of weight, so I want what I say to complement what people know rather than crowd it out. That means being clear about the conditions behind my views, honest about my uncertainty, and focused on explaining how I'm thinking, rather than offering a precise path. It’s also why I see the goal as mutual understanding. I work hard to understand what markets are telling me, just like many of you work hard to understand policymakers. When understanding runs both ways, a public signal is less likely to be given too much weight.
The second lesson is about the value of many views. Monetary policy in the United States is made by a committee, and I think that is a real strength. Members have varied backgrounds, speak to a wide range of people, and communicate in different ways. That diversity improves the information available to the FOMC and to the public. It also guards against any single communication becoming the only one that matters.
Do We Understand Each Other?
Let me return to where I started. Should a central bank tell you what it plans to do with interest rates?
Sometimes, definitely. But I think the better question is whether we understand each other. The goal is always the same: a shared understanding of how policymakers are likely to respond to economic conditions. In normal times, when people understand the reaction function, specific guidance about the rate path may not add very much. When rates are near zero, when policy departs from what people would expect, or when something big changes, understanding can't be assumed, and saying more can be well worth it.
The research I've described suggests a test I use for evaluating any form of communication. Does it help people understand how policymakers respond to the economy, and why? Does it match what policymakers actually do? And does it help people combine what policymakers say with what they already know, rather than replace it?
How well a given tool does on these three questions will vary with the moment. For example, I view the dot plot as very helpful in some periods and less helpful in others. That is a reason to value having several ways of communicating, since at least one of them is likely to be doing useful work at any given time. Reasonable people can disagree about the right mix, and the answer will change with the economy. That is why I welcome the work of the Fed's Communications Task Force.
When communications are doing their job, when we understand each other, rates can better guide spending in the right direction, the Committee can look through a temporary shock without losing credibility, and the path of rates can adjust to incoming data without unnecessary jolts. That helps deliver price stability and maximum employment.
Over more than 30 years, Fed communication has gone from silence to statements to a steady stream of speeches, projections, and press conferences. The tools will keep changing. But the question will not: do we understand each other?
Thank you.
- The views expressed here are solely those of the author and do not necessarily reflect the views of the Federal Reserve Bank of Philadelphia or the Federal Reserve System.
I am grateful to Satyajit Chatterjee, Thorsten Drautzburg, Patrick Fischer, Igor Livshits, Makoto Nakajima, Keith Sill, and Skander Van den Heuvel for their assistance with these remarks. - Patricia Zobel, governor, Money Marketeers of New York University, Inc., and head of macroeconomic research and market strategy, Guggenheim Investments
- Blinder, Alan S., Michael Ehrmann, Marcel Fratzscher, Jakob De Haan, and David-Jan Jansen. 2008. "Central Bank Communication and Monetary Policy: A Survey of Theory and Evidence." Journal of Economic Literature 46(4): 910–945.
- Campbell, Jeffrey R., Charles L. Evans, Jonas D. M. Fisher, and Alejandro Justiniano. 2012. "Macroeconomic Effects of Federal Reserve Forward Guidance." Brookings Papers on Economic Activity 2012 (1): 1–80.
- Woodford, Michael. Interest and Prices: Foundations of a Theory of Monetary Policy. Princeton University Press, 2003.
- Hall, Robert E. 1978. “Stochastic Implications for the Life Cycle-Permanent Income Hypothesis: Theory and Evidence” Journal of Political Economy 86 (6): 971–87.
- Taylor, John B. 1993. "Discretion versus Policy Rules in Practice." Carnegie-Rochester Conference Series on Public Policy 39 (1): 195–214.
- Eggertsson, Gauti B., and Michael Woodford. 2003. "The Zero Bound on Interest Rates and Optimal Monetary Policy." Brookings Papers on Economic Activity 2003(1).
- Paulson, Anna (2026). "Economic Outlook," speech delivered at the Federal Reserve Bank of Philadelphia's 81st Annual Field Meeting Capstone, Conshohocken, Pa., November 10. https://www.philadelphiafed.org/the-economy/monetary-policy/261110-81st-annual-field-meeting-capstone-speech.
- Kydland, Finn E., and Edward C. Prescott. 1977. "Rules Rather than Discretion: The Inconsistency of Optimal Plans." Journal of Political Economy 85(3): 473–491.
- Barro, Robert J., and David B. Gordon. 1983. "Rules, Discretion and Reputation in a Model of Monetary Policy." Journal of Monetary Economics 12(1): 101–121.
- Board of Governors of the Federal Reserve System, “FOMC Statement,” press release, August 9, 2011, https://www.federalreserve.gov/newsevents/pressreleases/monetary20110809a.htm.
- Board of Governors of the Federal Reserve System, “Federal Reserve issues FOMC statement,” press release, December 12, 2012, https://www.federalreserve.gov/newsevents/pressreleases/monetary20121212a.htm.
- Morris, Stephen, and Hyun Song Shin. 2002. "Social Value of Public Information." American Economic Review 92(5): 1521–1534.
- Eric Engstrom, "Anchored to the Dot Plot: Central Bank Projections and Interest Rate Expectations," Finance and Economics Discussion Series, no. 2026-026 (May) Washington: Board of Governors of the Federal Reserve System, https://www.federalreserve.gov/econres/feds/anchored-to-the-dot-plot-central-bank-projections-and-interest-rate-expectations.htm.
- Svensson, Lars E. O. 2006. "Social Value of Public Information: Comment: Morris and Shin (2002) Is Actually Pro-Transparency, Not Con." American Economic Review 96(1): 448–452.