Central Banks Should Stop Paying Interest on Reserves

In 2008, the Federal Reserve began paying interest on reserve balances held on deposit at the Fed. It took more than seven decades from the US leaving the gold standard — in 1933 — for the fiat regime to do this and thus revoke a cardinal element of the old gold-based monetary system: the non-payment of any interest on base money.

The academic catalyst to this change came from Milton Friedman’s essay “The Optimum Quantity of Money” where he argued that the opportunity cost of paper money (any foregoing of interest compared to on alternative money-like instruments such as savings deposits) should be equal to its virtually-zero marginal cost of production. Opportunity cost could indeed be brought down to zero if base money (bank reserves, currency) in large part paid interest at the market rate. Under the gold standard, the opportunity cost of holding base money largely in metallic form (gold coin) was indeed typically significant. All forms of base money paid no interest. And the stream of interest income foregone in terms of present value was equal in principle to the marginal cost of gold production (this was equal to the gold price).

Interest on Reserves are Important to Controlling Markets and Imposing Negative Rates

Friedman, however, did not identify the catch-22 of his proposal. If the officials of the fiat money regime indeed take steps to close the gap between the marginal production cost and opportunity cost of base money, with both at zero, then there can be no market mechanism free of official intervention and manipulation for determining interest rates.

That is what we are now finding out in the few years since central banks in the US, Europe, and Japan started paying interest on reserves. (The ECB was authorized to do this since its launch in 1999, while the Fed and BoJ began following the 2008 financial crisis.) Central banks can now bind the invisible hand operating in the interest rate market to an extent almost unprecedented in peacetime. In some cases, central banks have even deployed a negative interest rate “tool” which would have been impossible under the prior status quo where base money paid no interest.

How We Got Here

The signing into law of the Financial Services Regulatory Relief Act in 2006 authorized the Federal Reserve to begin paying interest on reserves held by depository institutions beginning October 1, 2011. On the insistence of then Fed Chief Bernanke, that date was brought forward to October 1, 2008 by the Emergency Economic Stabilization Act. He was in the process of dispensing huge loans to troubled financial institutions but wanted nonetheless to keep interest rates at a positive level (one purpose here was to protect the money market fund industry).

Accordingly, the Federal Reserve Board amended its regulation D so that the interest rate paid on required reserves and on excess reserves would be at levels tied (according to distinct formulas at the start) to market rates. An official communiqué explained that the new procedure would eliminate the opportunity cost of holding required reserves (and thereby “deregulate”) and help to establish a lower limit for the Federal Funds rate, becoming thereby a useful tool of monetary policy.

This was useful indeed from the viewpoint of rate manipulators: by setting the rate on excess reserves the Fed could now determine the path of short-term interest rates and strongly influence longer term rates regardless of how the supply of monetary base was growing relative to trend demand. By contrast, under the gold standard and the subsequent first seven decades of the fiat money regime, interest rates in the money market were determined by forces which brought demand for base money into balance with the path of supply as set by gold mining conditions or by central bank policy decision respectively. A rise in rates meant that the public and the banks would economize on their direct or indirect holdings of base money and conversely.

Back Before the Fed Paid Interest on Reserves

Yes, under the fiat money system the central bank could effectively peg a short-term rate and supply whatever amount of base money was needed to underwrite that — but the consequential growth of supply in base money was a variable which got wide attention and remained an ostensible policy concern. Right up until the Greenspan era, the FOMC implemented policy decisions by directing the New York Fed money desk to increase or reduce the pace of reserve growth and changes in the Fed funds rate occurred ostensibly to accomplish that purpose. This old method of determining money market interest rates under a fiat regime — in which banks’ need for reserves was minute given deposit insurance, a generous lender of last resort, and too-big-to-fail — depended on the banking industry enduring what was essentially a tax on its deposit business, which was then magnified by fairly high legal reserve requirements. Thus, it is not surprising that the original impetus to paying interest on reserves, whether in the US or Europe, came from the banking lobby. There was no such burden under the gold standard even though the yellow metal earned no interest. Banks in honoring their pledge to deposit clients that their funds were convertible into gold had to visibly hold large amounts of the metal in their vaults or at hand in a reserve center. Actual and potential demand for monetary base by the public is more limited under a fiat money regime than under the gold standard as bank notes are hardly such a distinct asset as gold coin from other financial instruments.

More Problems with Friedmanite “Solutions”

Friedman, when he advocated eliminating the opportunity cost of base money under a fiat regime, hypothesized that this could occur under a long-run declining trend of prices rather than by the payment of interest. The real rate of return on base money could then be in line with the equilibrium real interest rate. This proposal for perpetually declining prices would also have been problematic, though. The interest rate would fluctuate, and in boom times be well above the rate of price decline. In any case, the rate of price decline would surely vary (sometimes into positive territory) in a well-functioning economy even when the long-run trend was constant (downward). The equilibrium real interest rate would be below the rate of price decline sometimes (for example, during business downturns), meaning that market rates even at zero would be too high. That situation did not occur often under the gold standard where prices were expected to be on a flat trend from a very long-run perspective and move pro-cyclically (falling to a low-point in the recession from which they were expected to rise in the subsequent business expansion, meaning that real interest rates would then be negative).

What Can Be Done?

So what is to be done to escape the curse? A starting point in the US would be for Congress to ban the payment of interest on bank reserves. And the US should use its financial power with respect to the IMF to argue that Japan and Europe act similarly within a spirit of G-7 coordination such as to combat monetary instability. We have seen in recent years how rate manipulation and negative rates are made possible by the payment of interest on reserves, and are potent weapons of currency warfare. Yes, the ban in the immediate would force the Federal Reserve to slim down its balance sheet so that supply and demand for base money would balance at a low positive level of interest rates. The Fed might have to swap its holdings of long-maturity debt for T-bills at the Treasury window so as to avoid any dislocation of the long-term interest rate market in consequence. That, not the Yellen-Fischer “rate lift off day and beyond,” is the road back to monetary normalcy.

February 23, 2016•Brendan Brown