The Myth of the Explicit Reaction Function’s Necessity
Critics of Kevin Warsh’s new approach to monetary policy have fixed upon a strange idol: the reaction function.
The careful critics insist they are not demanding forward guidance. It’s just fine for the Fed to give up this heirloom of the financial crisis. They merely want the chairman to articulate the Fed’s “reaction function,” describing in detail how the central bank will adjust interest rates as inflation, unemployment, growth, and financial conditions change.
But this is largely a distinction without a difference. A promise tied to a date is calendar-based forward guidance. A promise tied to unemployment or inflation is also forward guidance, albeit state-contingent forward guidance. Calling it a reaction function makes the demand sound scientific and less like Wall Street asking the Fed to hold its hand.
Central Bank Actions Speak Louder Than Words
Investors are obviously going to form expectations about how the Fed will behave. Those expectations affect bond yields, stocks, the dollar, and credit spreads. A well-understood reaction function can help investors anticipate what the Fed will do before Fed officials tell us. They can respond to economic news before the Fed acts with more confidence that they know what the Fed will do when it does act. That’s nice if you are a bond trader, but it does not establish a requirement that the chairman provide the function in advance.
The most relevant recent research comes from Michael Bauer of the Federal Reserve Bank of San Francisco, Carolin Pflueger of the University of Chicago’s Harris School, and Adi Sunderam of Harvard Business School. In their 2024 Quarterly Journal of Economics paper, “Perceptions about Monetary Policy,” they use professional forecasts going back to 1985 to estimate the reaction function investors believed the Fed was following.
They found that these beliefs matter. As the perceived responsiveness to economic data on the part of the Fed rose, the sensitivity of interest rates to economic surprises doubled and sometimes tripled. Perceived Fed behavior also affected Treasury term premiums and stocks. The paper shows that investors develop beliefs about a reaction function. But this isn’t the same as showing that the Fed must talk about the reaction function.
The same authors make the point more clearly in “Changing Perceptions and Post-Pandemic Monetary Policy,” presented at the Kansas City Fed’s 2024 Jackson Hole conference. Before the Fed began raising rates in March 2022, a 10-basis-point surprise in core inflation produced essentially no response in the two-year Treasury yield. After liftoff, the same surprise raised the yield by 9.6 basis points.
Crucially, markets became convinced of the Fed’s willingness to fight inflation only after the Fed started hiking. The authors conclude that substantial rate increases were apparently necessary to change perceptions. Markets learned the reaction function from policy action after Fed communication failed to convince them.
In other words, the Fed really communicates its reaction function not by talking but by doing. The demand for more explicit talk about the reaction function is just not well-grounded in economic theory or research.
Volatility Suppression Is Not a Free Lunch
The case for guidance is strongest when rates are stuck at zero. If the Fed cannot cut the current rate, it can provide accommodation only by convincing markets that rates will remain lower in the future. Outside the zero bound, the Fed can act, observe the consequences, and act again. A surprise rate increase is still transmitted into the economy.
But isn’t a surprise bad? Won’t it cause unwanted volatility? The explicit reaction function advocates apparently assume that reducing market volatility is an obvious good. But that’s not what the financial-stability literature tells us.
In “Learning from History: Volatility and Financial Crises,” Jon Danielsson of the London School of Economics, Marcela Valenzuela of the University of Chile, and Ilknur Zer of the Federal Reserve Board examined more than 200 years of data covering 60 countries and 262 banking crises. Published in the Review of Financial Studies, the paper found that unusually low volatility was followed by credit buildups and increased banking-crisis risk.
So now we have the demand for a clearly stated reaction function being explicitly defended as a way of suppressing near-term interest-rate volatility and compressing risk premiums. But while that may be worth doing during a crisis, it’s not something you want to permanently incorporate into monetary policy. At the most basic level, the low volatility can convince investors that the Fed has insured them against changes in the policy environment. More broadly, the appearance of stability encourages ever-increasing leverage and inflates the values of financial assets (and the assets those support, such as real estate).
For a good discussion of this, we would direct you to this October 2020 episode of Blooomberg’s Odd Lots podcast. The title of the episode makes the argument clear: “Volatility Suppression Turned the Entire Economy Into One Big Carry Trade.” Odd Lots hosts Joe Weisenthal and Tracy Alloway interviewed Tim Lee, Jamie Lee, and Kevin Coldiron, who are the authors of The Rise of Carry, a book arguing that central-bank suppression of volatility had turned much of the financial system into a fragile leveraged carry trade. The Lees and Coldiron explained that when central banks hold down volatility, carry trades appear safer and more profitable than they really are. That draws in weaker balance sheets, encourages leverage, and produces long periods of apparent stability punctuated by sudden crashes. Weisenthal closed the discussion by describing volatility suppression as “bottling up a potential crisis,” which is as good as a summary as we can imagine.
Warsh Has Explained What the Markets Need to Know
Warsh has supplied a benchmark against which the Fed can be judged: two percent inflation. He has declined to supply a schedule explaining how every new piece of data will alter policy. This may create some uncertainty, some volatility, some (gasp) actual losses on trading books when markets bet the wrong way on Fed policy. At least at first, the new dispensation will take some getting used to, and there will be bruised investors along the way.
But the gods of central bank credibility do not require encantations about reaction functions. They just require markets to believe that persistent inflation will bring sufficiently forceful action in a timely manner. That was damaged greatly by the insistence by Jerome Powell and his Fed that the post-pandemic inflation would be transitory. Warsh’s Fed will repair the damage if inflation falls to the target or add to the injury if it does not. All the talk about the reaction function will not make a difference.
A reaction function is useful to investors trying to forecast the Fed—and helping investors forecast the Fed is appropriate during a crisis. But in normal times, it is not a requirement of sound monetary policy. The demand that Warsh articulate one is backdoor forward guidance, dressed up as a defense of central bank credibility.
Read the full article here