Scott Sumner  

The letter and the spirit of the law

Letter From a Former Student... Henderson Speech at Brescia Un...

David Beckworth has a new podcast, interviewing George Selgin on the subject of interest on reserves. I found the final part of the interview to be particularly interesting. Beginning about the 42:30 mark, they discuss the legal issues surrounding the payment of interest on reserves.

During 2006 and 2008, Congress gave the Fed the right to pay interest on reserves, at a rate not to exceed the prevailing level of short-term interest rates. The actual interest rate paid by the Fed does exceed most measures of short-term interest rates; for instance, it has often exceeded the fed funds rate, or the rate on 3-month T-bills. But George points out that the Fed cleverly crafted the language so that in practice they have almost unlimited ability to pay as much interest on reserves as they like. Thus while the actual program is a subsidy to banks that clearly violates Congressional intent, it's not technically illegal.

The payment of such a high interest rate has created a "floor system", whereby the Fed can inject massive quantities of reserves into the banking system without substantially depressing interest rates. Instead of using open market operations as the prime policy tool, the Fed now uses interest on bank reserves. In other words, monetary policy used to be all about the Fed controlling the supply of money; now it's mostly about the Fed controlling the demand for money.

And this leads to something else that Congress did not intend. Decisions on open market operations are made by the FOMC, which includes 5 regional bank presidents. In contrast, the 7-member Federal Reserve Board determines the interest rate on excess reserves. Thus the switch to a floor system where IOR is used to set policy could be viewed as a sort of "coup", where the Board seizes monetary policy control away from the regional bank presidents.

However that's not what actually occurred, at least thus far. The Board was presumably uncomfortable with the idea that it would be using IOR to do an end around on Congressional intent regarding Fed governance. Clearly the idea was that the FOMC would determine monetary policy, and the Board didn't feel comfortable unilaterally changing that system. Whatever you think about Bernanke, he is not a Machiavellian figure.

On the other hand, you could argue that Greenspan was a Machiavellian figure, and that future Fed chairs might also be willing to exploit this loophole in Fed governance rules. David mentioned an episode in the past where Greenspan used various stratagems to prevent the regional bank presidents from interfering with his policy preferences.

How should we feel about all of this? A few thoughts:

1. I believe the Fed felt a need to adhere to the spirit of the law when it involved personnel issues that are easily understandable to Congress and the public, but not when it involved highly technical issues that are confusing to most non-economists. It's about expediency, not ethics.

2. I believe that the precedent set by Bernanke and Yellen is likely to endure. I don't see Powell going back to Greenspan's more devious practices. Just as after a number of years you can have a common law wife or an easement because you frequently walk across someone's land, the fact that the regional presidents have taken part in IOR decisions for many years makes it almost impossible to take away that privilege.

3. In some ways it might be better if the Board did seize power away from the regional bank presidents. For years, people viewed as "monetary cranks" have argued that the semi-private nature of the Fed is unconstitutional. I didn't pay much attention, assuming they were crackpots. But after reading Peter Conti-Brown's excellent history of the Fed, I'm convinced that the FOMC actually is unconstitutional. Any person making public policy is supposed to report to the President, or one of the President's staff. Regional bank presidents are independent of the federal government, and yet they make public policy.

Thus while taking power away from the regional presidents on the FOMC might violate the spirit of the Banking Act of 1935 (which created the current system of Fed governance), the law itself was probably unconstitutional.

For people like me who believe in the rule of law, this creates a quandary. Whatever the Fed decides, it's likely to seem a bit lawless from one perspective or another. What a mess!

Screen Shot 2018-03-10 at 5.43.58 PM.png

Comments and Sharing

COMMENTS (9 to date)
Vaidas Urba writes:

If IOR is a subsidy, why isn't it competed away? An alternative explanation is that rents are dissipated, and IOR - 3m bill spread is a compensation for lower liquidity of reserves compared to treasury bills.

Matthew Waters writes:

I'm sure Conti-Brown made the same argument in his book, but I'll link to this argument I found about the unconstitutionality of Fed presidents.


I was pretty skeptical when I read this post, since Congress has very broad powers to regulate interstate commerce and to spend money. But requiring cause for the Fed governors to remove regional directors does look, on its surface, unconstitutional under the Appointments Clause.

The constitutionality is really beside the point though. It's unconscionable how the Fed is organized and should be changed by Congress to operate in the public's, not banks', benefit.

Scott Sumner writes:

Vaidas, Whether something is a subsidy and whether it is competed away are two very different questions.

Vaidas Urba writes:


In any case I think there is no subsidy, just a compensation for lower liquidity of reserves compared to T-bills. The right way to resolve this would be to let the Fed issue securities which are as useful for market participants as T-bills are.

Brian Donohue writes:


Given the flattening of the yield curve, isn't it time for the Fed to end the 0.25% subsidy on IOER and sell some of its longer-dated assets rather than continuing to increase the FFR?

Scott Sumner writes:

Vaidas, I'm not convinced that reserves are less liquid that T-bills. Liquidity is usually defined as the ease with which an asset can be turned into base money.

Brian, I'd recommend they do both. BTW, the IOR rate is now far higher than 0.25%

Vaidas Urba writes:


That is a circular definition. More useful definition is that assets which facilitate transactions are more liquid.

In any case, we deserve an explanation why treasury is able to borrow at below-IOR rates. This is a puzzle and saying that IOR is a subsidy is just a hand-waving.

Philo writes:

Life is a mess!

Matthew Waters writes:


Treasuries can be owned by anybody while Fed reserves can only be owned by Fed members. If Treasuries earn less than IOR, then the arbitrage by Fed members is limited by the cost of borrowing Treasuries for shorting them.


Return to top