US president Donald Trump has, on numerous occasions, considered out loud replacing current Federal Reserve chair Jerome Powell. His main criticism of Powell has been centered on interest rates. But the choice of new Fed chair should consider a much wider array of issues.
With that in mind, I drafted a list of questions the media, Congress, and the president should be thinking about when evaluating any future nominee for the Fed’s top job, and that aspirants to the job should be prepared to answer. The questions originally appeared as part of an op-ed I wrote for the Wall Street Journal; here, I add my answers. The point is not to get the right answers, or if my answers are right. The point is that a Fed chair, and those who appoint him or her, should understand the questions and the pro-con arguments.
Controlling inflation
Should the Fed modify its 2 percent inflation interpretation of its “price stability” mandate? Should it aim for zero inflation or a steady price level?
After an inflation surge, should the Fed try only to bring subsequent inflation back to target, while accepting higher prices, as it does now? Or should it slowly bring back the level of prices, to achieve its inflation target as a long-term average?
I favor zero inflation, and a price-level target. Like former Fed chair Alan Greenspan, I think the Fed should get there slowly and opportunistically rather than all at once, because I distrust my own understanding of inflation dynamics and adjustment to new regimes.
A price-level target implies that if there is a spurt of inflation or deflation, the Fed should aim to gradually bring prices back to the original level. We don’t shorten the mile by 2 percent every year; why should we undo the standard of value by 2 percent every year?
I think this is what Congress had in mind when it mandated “price stability” as a Fed objective in 1977. We had something closer to price-level stability under the gold standard, and that experience was certainly in Congressional minds. But my argument does not rest on legalisms, and indeed one can take Congress’s acquiescence to the Fed’s target as an implicit endorsement.
The second part of the question asks a candidate to discern between a forward-looking target that forgets past mistakes and a target of overall average performance. I prefer the latter because it is more accountable and transparent. Following the former approach, the Fed could say that over the past five years it has acted perfectly consistently with its target because, after a mysterious surge, it did what it could to bring inflation back to target, and it always forecast a quick return to 2 percent inflation.
What strategy or rule should the Fed adopt in place of its ill-fated “flexible average inflation targeting,” under which it accepted short-term deviations from its target? Or should it abandon strategies and just raise and lower interest rates as it sees fit?
You might expect me to endorse the Taylor rule, the formula developed by Stanford’s John Taylor that ties interest rates to inflation and output. I do admire the rule a lot.
But I am torn between that and the view, expressed by Harvard’s Lawrence H. Summers at the 2024 Hoover Institution monetary policy conference, that each shock is different, nobody really knows how the economy works, and therefore a firm “I’ll do whatever it takes” commitment to the target without specifying the exact action is better.
The right answer lies somewhere in between. At least know the arguments for and against each approach.
I don’t think QE did much. If I give you two $5 bills and a $10 bill for each $20 bill, do you spend more?
Should the Fed address employment by simply focusing on price stability, avoiding the inflation and disinflation that upsets labor markets? Or should it sacrifice some price stability in pursuit of employment goals?
Price stability. The record on trading inflation for better employment is poor. Just keep the price level stable and employment will take care of itself. Avoid surges of inflation and panicked rate rises that cause recessions. Don’t repeat 1979, when inflation was allowed to climb, and you won’t repeat 1980, when the United States went into recession.
The Fed bought trillions of dollars of new Treasury debt, thereby financing the huge 2020–22 deficits with new money. It also kept interest rates at zero for a year after inflation surged in 2021. Were these mistakes? And if so, how will you avoid repeating them?
These were mistakes. I would avoid a repeat by introducing the concept that not all shocks are the same, and therefore the Fed should respond differently to supply shocks than it does to demand shocks. (The Fed is in danger right now of responding to a tariff supply shock as if it were low demand: with stimulus.)
I would also introduce the concept of fiscal inflation, which obviously was what was happening in 2021. And yes, I would pay some attention to the Taylor rule, and if inflation surges to 10 percent, I would certainly think about raising rates.
More deeply, the inflation episode needs a thorough review to determine what happened and how the Fed got it so wrong. “Supply shocks hit that weren’t our fault” is not good enough. That excuse implies that any future “shocks” will lead inevitably to more inflation.
Which economic theory do you think best describes monetary policy?
This is in part a trick question to see if a candidate is familiar with various schools of thought—Keynesian, New Keynesian, monetarist, modern monetary, and fiscal theories, all of which disagree fundamentally—evidence for and against them, and where their bodies are buried.
Just echoing, “I think high rates lower spending, which lowers output and, via the Phillips curve, lowers inflation” gets you a C-, and shows you don’t know what’s actually going on in the Fed’s own models.
The best academic theory I know of is the fiscal theory of the price level, combined with an understanding of the role of long-term debt. But there is a lot I don’t know.
However, as an economist, my loss function gives a high reward for a new theory that turns out to be right, and only a mild penalty of comfortable obscurity if I am wrong. As a policymaker, my loss function would put much more weight on making sure I am not wrong and do not make disastrous mistakes. Given our uncertainty about how the economy works, that means putting both weight and distrust in lots of economic approaches.
Making asset purchases
Did quantitative easing significantly affect long-term rates, output, and inflation? Should QE continue as a routine part of the Fed’s tools?
If QE works, should the Fed more directly target long-term rates rather than buy some bonds and see what happens?
I don’t think QE did much. If I give you two $5 bills and a $10 bill for each $20 bill, do you spend more? The Treasury sold more long-term debt than the Fed bought. QE did move markets on announcement, but price-pressure effects don’t last forever.
At most, it had information effects—shock and awe that the Fed was so desperate—and it signaled rates would be low for a long time. On the other hand, that also means QE didn’t do much damage either.
I think a wide range of financial institutions should have access to something that looks like reserves.
The Treasury issues long-term debt, which pays a fixed rate. The Fed buys that debt and turns it into floating-rate debt (reserves). That action makes interest costs hit the budget faster.
How should the Fed and the Treasury agree on the proper maturity of the debt? Should the Fed buy only short-term debt?
To start, the Fed chair must admit that large QE operations that buy long-term Treasurys do shorten the maturity structure of the debt. Then admit that the current Fed losses are a cost to taxpayers because they remove what little interest rate protection the Treasury had bought with its long-term debt issues.
The Fed pretends to ignore the Treasury. The Treasury pretends to ignore that the Fed undoes whatever it does. We need a new accord about who is responsible for and accountable for the maturity structure of the debt and the exposure of taxpayers to interest rate risks. I think it should be the Treasury, and the Fed should hold only short-term securities, or swap out the interest rate risk with the Treasury.
Should the Fed continue to buy mortgage-backed securities?
In 2020, the Fed announced a corporate bond–buying program, to prop up their prices. Is it proper for the Fed to prop up asset prices? Should it worry about stock prices and exchange rates?
No, no, and no. The wider the mandate, the less the Fed can be independent.
There is a case for central banks to be forbidden from holding Treasury securities and restricted to holding private or foreign securities. Doing so helps isolate the central bank from pressure to monetize the debt. My judgment is that, at present, isolating the Fed from pressure to prop up a particular interest group’s favorite assets is a more pressing concern, so I favor an all-Treasury portfolio. But that may not last, and I get to change my mind.
Regulating the financial system
Should the Fed continue to pay interest on ample reserves? Or should it stop paying interest, return to a tiny quantity of reserves, and manage interest rates by changing that small quantity?
Should the Fed continue to pay banks more on reserves than it pays other institutions? Should it follow other central banks and elastically lend new reserves to banks as they demand, rather than strictly control the size of its balance sheet?
I favor ample reserves that pay a market interest rate. I favor a flat supply curve rather than trying to control the balance sheet as a separate policy lever, as with QE. I favor a narrow corridor—the Fed should lend at rates very close to the rates at which the Fed borrows—and equal treatment for all comers.
Ample interest-paying reserves have been a great innovation of monetary policy since 2008, with many benefits. The fear they would necessarily lead to inflation was proved wrong in a great worldwide experiment. Three cheers for the Fed, and for former chair Ben Bernanke, who inaugurated this policy.
Should the Fed maintain its prohibition of banks such as TNB that hold only reserves? Or should it encourage such narrow banks, as they are impossible-to-fail institutions that can provide low-cost transactions services and large, safe, uninsured deposits?
Should the Fed allow access to reserves and its payments system to a larger range of financial institutions that aren’t banks, such as money market funds or even stablecoins?
The Fed’s denial of a master account to TNB, short for the Narrow Bank, was a disgrace. The reasoning was beyond fanciful, showing that the Fed has really no idea what “financial stability” means.
The Fed likewise ruled against “segregated accounts” that funnel deposits directly to reserves, immune from bankruptcy, in the big banks. Large uninsured depositors could have used these vehicles. They ran in the Silicon Valley Bank fiasco, and then got an ex post bailout.
None of this would have happened had the Fed allowed them access to narrow banks or segregated accounts.
The Fed worried that people would run to narrow banks in a crisis. Never in history has the Fed’s response to a crisis been, “You have to sit and hold illiquid assets.” And people can run to bank deposits and short-term Treasurys directly now.