Kevin Warsh has created quite a stir in his first two months as Federal Reserve chair by saying too little, which breaks with more than a decade of practice.
If Warsh gets his way on scaling back Fed communication, it’s likely to lead to a bit more volatility in interest rates and bond prices. But this isn’t reason for investors to panic.
Why didn’t the Fed hike rates? That question was on the minds of many investors following the July 29 Federal Open Market Committee meeting, but an answer was never forthcoming.
That’s not to say the decision was a surprise. The market-implied probability of a rate hike had been just 30% the day before the meeting—but that’s still high enough to say the decision was up for grabs. Hence, the lack of a postmeeting explanation of the decision was a major departure from prior practice.
Kevin Warsh has been on the record for at least a decade criticizing the Fed’s expanded communication. Fed watchers had some notion that Warsh would dial back communication upon becoming chair in May 2026, but the extent to which he’s done so has still been a shock to the system.
Warsh in particular has long criticized forward guidance—the practice where the Fed signals the future path of interest rates. He appears to believe that he can’t meaningfully elaborate on why the Fed didn’t hike rates at the last meeting without signaling what might prompt the Fed to hike rates in the future.
He artfully dodged specifics in the press conference, and the whole 45 minutes elapsed without him divulging much of anything of substance. Warsh has floated the idea of holding fewer press conferences, but a formal change may not be necessary; a few more like the last one, and people may stop showing up.
Warsh believes that, at worst, forward guidance deprives the Fed of vital flexibility in the face of changing economic conditions.
Of course, Fed officials often emphasize that guidance about the federal-funds rate (including the “dot plot”) is generally just a forecast, not a commitment. This forecast is conditional on forecasts of other macroeconomic variables like inflation. If those play out differently than the Fed expects, it will adjust accordingly.
Despite this, Warsh is skeptical that Fed officials can manage to avoid becoming “prisoners of their own words” once they make their forecasts public.
Insofar as the Fed is able to preserve its flexibility, it does so by burying its statements in layers of qualification and nuance such that Fed communication becomes a “cacophony,” which adds more noise than signal to financial markets
At the same time, financial market participants become so fixated on the Fed’s forecasts that they abdicate their job of projecting economic fundamentals and incorporating them into interest rates and asset prices. As Warsh argued at the last press conference, in the old regime, financial markets were merely “echoing” the Fed.
Warsh’s criticism of forward guidance has some merit in light of the 2021-22 runup in inflation. The Fed said in late 2020 that it would keep the federal-funds rate fixed near zero until labor markets reached “maximum employment.” Throughout the second half of 2021, despite PCE inflation shooting up over 5%, the Fed adhered to this language. It did not ultimately hike rates until March 2022.
Without the forward guidance, it is likely the Fed would have been a little earlier in hiking rates—but maybe only by a few months. The real problem wasn’t forward guidance per se, but rather that the Fed was fighting the last war. It was so afraid of a long-lasting shortfall in economic output (such as during the 2010s after the Great Recession) that it neglected the inflation surge until it came back to bite.
The critique that forward guidance constrains the Fed’s flexibility is legitimate, albeit exaggerated. On the other hand, Warsh’s proposition that the absence of Fed communication will unleash a cornucopia of information gathering by financial markets has little basis.
The first flaw is that markets have to have some information about the Fed. Specifically, they need to know the Fed’s reaction function, which spells out how the Fed will adjust interest rates in response to deviations from its inflation target and maximum employment. Even if markets aren’t relying on the Fed to forecast the latter, they still need to know the reaction function in order to understand the implications for interest rates.
Asked about this in the recent press conference, Warsh implied that the problem was trivial, saying “any central banker, when he or she sees underlying inflation moving higher, he or she is more inclined to tighten policy” and vice versa. But this clearly only scratches the surface; it says nothing about the magnitude of reaction (which is likely to vary over time), much less other complications. Markets can’t parse this nuance without Fed guidance.
What’s the alternative to relying on the Fed for information about its reaction function, along with providing a baseline set of economic forecasts? If history is any guide, it’s not markets coming to a well-informed, independent view. Instead, it’s markets flying blind.
This much is clear thanks to a fascinating paper by Sebastian Hillenbrand. Hillenbrand looks at daily changes in the 10-year yield and sorts them into two buckets: those that occurred in three-day windows surrounding each FOMC decision, and those occurring outside those windows. These are then accumulated to create counterfactual scenarios for the 10-year yield starting in June 1989.
The results of the exercise are astonishing. From June 1989 to June 2021, the 10-year yield fell from about 8.5% to 1.5%. Almost all of that seven-percentage-point yield drop came from changes occurring within the three-day FOMC meeting windows, comprising a mere 10% of total trading days. The other 90% of trading days, occurring between meetings, didn’t matter for the long-run trend in yields. The odds this occurred by chance are nil.
The 10-year yield can be decomposed into the expected federal-fund rate over the next 10 years plus a term premium. The obvious interpretation, which Hillenbrand endorses, is that markets are heavily dependent on the Fed for forming their expectations of interest rates.
What about the shift to much larger changes occurring outside the window, starting in 2022? Arguably, that was the result of more Fed communication, not markets becoming less reliant on the Fed. This includes more frequent and better speeches from Fed officials between meetings. And, there was more forward guidance (which, even if issued at an FOMC meeting, market participants might chew on for more than the three-day window).
The Fed once communicated very little. FOMC decisions were explained in a terse handful of sentences, and that was only introduced in 1994. The flow of communication grew a trickle more in the 2000s but really gushed forth after the 2008 financial crisis. The postmeeting press conference was introduced in 2011, and the dot plot was introduced in 2012. Warsh’s seeming desire would be to return to something much closer to the 1990s and early 2000s regime.
What were markets doing before the Fed started communicating effusively? As the data shows, it was still the case that sustained yield shifts came solely from within the three-day FOMC windows. Markets seemed to have just been updating for the latest FOMC decision and making some sort of crude extrapolation, with also some assumption of long-run mean reversion.
Certainly, markets weren’t any less obsessed with the Fed in the era of lesser communication. Some of us can remember the silly attempts to predict rate decisions by the thickness of Alan Greenspan’s briefcase.
The idea that markets were ever churning through information independently of the Fed and projecting the course of interest rates is a fantasy. If that were the case, we would have seen much more sustained yield movement outside the three-day FOMC meeting window before the era of heightened communication. In this sense, Warsh is chasing a historical mirage.
Some commentators, though, were a bit premature in their reaction to the latest FOMC meeting. The spread between the 30-year and 2-year Treasury yields did widen by around 15 basis points, but it’s still narrower than it was back in January. The higher long-term yields didn’t come from higher long-run inflation expectations. Thus, calling this a credibility shock is a bit of an exaggeration. Although Warsh’s tactics look suspect, he has a track record as a monetary policy hawk, and we haven’t seen enough evidence yet to worry about his commitment to bringing down inflation.
What’s more accurate, probably, is that the market is bracing for more bond market volatility. That in itself can call for slightly higher long-term bond yields as risk premia widen a bit.
To the extent that Warsh curbs Fed communication, interest rates are likely to gyrate more wildly as markets are more surprised by future changes in the federal-funds rate and other monetary policy shifts. The market is continuing to price in rate hikes in coming months (one in 2026, a second in the first half of 2027). If inflation remains concerning but these rate hikes don’t play out, then we’re likely to see a jump in bond market volatility as the market struggles to understand the Fed’s strategy.
For now, though, we’re still in a world where Fed communication is light years ahead of where it was in the 1990s. Excluding Warsh, the other eleven voting participants of the FOMC are still issuing forecasts for the federal-funds rate and other economic variables in the Summary of Economic Projections. They’re still issuing speeches shedding light on their thinking.
As long as Warsh doesn’t impose his taciturn philosophy on the rest of the FOMC, then markets still have a lot of forward-looking information to work with. Even if the Fed did return to the low communication regime of the 1990s, it would be a recipe for higher bond market volatility, but not to a catastrophic degree, as that earlier era itself showed. There’s no reason for investors to panic.
The author or authors do not own shares in any securities mentioned in this article.
Find out about Morningstar’s editorial policies.
