Among my ambitious New Year’s resolutions, the most difficult is my commitment to kill the oft-repeated claim that the Federal Reserve sets interest rates.
Wait, what? It is a basic truism of financial journalism that the Fed determines rates from on high. From the Wall Street Journal to the New York Times, central-bank control over interest rates is so thoroughly accepted that it’s hard even to think about things any other way.
Alas, it just isn’t true. Things have admittedly gotten more complicated since the 2007-8 financial crisis. With the switch from a corridor system to a floor system, the federal funds rate waned and the rate paid on excess reserves waxed in importance. Yet even now, the claim that the Fed sets interest rates obscures more than it illuminates. Let’s explore this, step by step.
Up until the financial crisis, the key policy interest rate used by the Fed was the federal funds rate. This is the rate banks charge each other for short-term (overnight, unsecured) loans. The Fed does not set this rate. Supply and demand in the market for bank reserves does. The Fed can increase or decrease the amount of reserves in the banking system, thereby affecting the fed funds rate. However, it isn’t setting anything. The Fed influences a market rate by acting within the market. It buys or sells assets, which changes liquidity conditions in that market.
Furthermore, Fed policy does not have long-term effects on interest rates. Say the Fed expands the supply of bank reserves by purchasing assets. Immediately after, interest rates should fall. This is the textbook result. But many commenters close the textbook too early. Turn the page and you’ll see we need to account for secondary effects. Because expansionary Fed policy eventually creates inflation, interest rates will go up. The effects on real (inflation-adjusted) interest rates is small and transitory.
Okay, fine. The Fed technically doesn’t set interest rates. Isn’t this just linguistic nit-picking?
No, it isn’t. In this case, precision matters quite a bit.
By exaggerating the Fed’s control over interest rates, the public conversation around Fed policy tacitly assumes that monetary policy is about messing with interest rates. It’s not. Monetary policy is about money. Good monetary policy adjusts the quantity of money supplied to meet changes in the demand to hold it. Effects on interest rates take a back seat. In econ-speak, we’d say interest rates aren’t the relevant transmission mechanism for monetary policy. In plain language, focusing on interest rates puts the cart before the horse.
The key error in the “monetary policy is about interest rates” paradigm is that it confuses instruments and targets. An instrument is something the Fed controls. A target is what it’s trying to achieve. The financial press writes about interest rates as if they’re an instrument, when in reality they’re a target. The Fed uses anticipated responses to interest rates to gauge whether they’ve injected or withdrawn the right amount of liquidity from the economy. This is why official Fed publications discuss changes in the target for the fed funds rate.
Econ 101, if taught well, hammers into students the importance of prices in guiding resource allocation. Interest rates are a price: the price of time. Allocating resources across time is no less important than allocating them across space. There’s no good reason to mess with the capital-allocation process, so there’s no good reason for anybody, including the central bank, to go around tinkering with interest rates willy-nilly. Fortunately, that’s not what monetary policy is about, although monetary economists are apparently quite good at keeping that a secret.
Now let’s look at a more complicated case. Unlike the fed funds rate, the rate paid on excess reserves is an administered rate — an instrument, in other words. Interest on excess reserves isn’t determined in a market. It’s determined by the Fed itself, similarly to the discount rate, which is the rate for borrowing directly from the Fed. So: Does the Fed set the excess reserves rate? Yes and no. Administratively, the Fed can hypothetically make this rate whatever it wants. Economically, the Fed is constrained here, too, by a combination of political and economic forces.
Set the rate on excess reserves too low, and banks that keep their money at the Fed will find it more lucrative to put those funds to work, financing real economic activity. Set the rate on excess reserves too high, and you can certainly give banks an incentive to tie up more of their liquidity in Fed accounts. But eventually that eats away at Fed profits, which are supposed to be remitted to the Treasury.
Fed remittances aren’t huge. Last year they totaled $88.5 billion. While this was a big increase over 2019, that number is still a very small percentage of the federal budget. Still, the fiscal authority would probably get cranky if the monetary authority siphoned off anticipated funds by paying exorbitant interest on excess reserves.
So even with interest rates that the Fed does “set,” there are countervailing forces that restrict its scope for administrative fiat. This doesn’t mean we should be cavalier about the Fed switching to a system where interest on excess reserves matters more than the fed funds rate. In fact, there is a big problem: Under the new (post-2008) system, the volume of bank reserves can become enormous, without creating inflation. This means the Fed has a lot more wiggle room to use its balance sheet for fiscal purposes.
As we saw with the plethora of credit policies during the COVID-19 crisis, the Fed is only too happy to engage in fiscal policy masquerading as monetary policy. I’d much prefer we get back to basics and return to the fed-funds system. But those basics don’t include setting interest rates, which are a barometer for monetary policy, not the content of monetary policy.