Would It Help To Eliminate Interest on Reserves?
Although we can’t be certain of the size of the effect, the ECB’s recent experience suggests that eliminating interest paid on reserves held with the Federal Reserve would not substantially increase bank lending and money growth.
Since 2008, the Federal Reserve has purchased several trillion dollars’ worth of securities in an effort to push down long-term interest rates and stimulate the economy. As a result of these quantitative easing (QE) transactions, the public now holds more short-term claims on the Federal Reserve (currency and bank deposits with the Federal Reserve) and fewer long-term interest-bearing assets, which has reduced long-term interest rates while providing safe, liquid, short-term assets with a low or no nominal yield. The intent of QE was to encourage interest-sensitive spending and investment through its effect on long-term interest rates and other asset prices. QE did not, however, dramatically increase bank loans and the growth of broader monetary aggregates. This essay reviews the recent experience of the European Central Bank (ECB) to consider if reducing interest paid on reserves could stimulate such lending and aggregate growth. But first some background on central banks’ monetary policy is needed.
An important point to understand about monetary policy is that central banks determine the monetary base (total reserves plus currency), but commercial banks decide the extent to which an increase in the monetary base changes the quantity of bank loans and the broader monetary aggregates. When the Fed bought securities from primary dealers, it credited the reserve accounts of the dealers’ banks, which in turn credited the dealers’ accounts. The banks chose to hold the deposits as “excess reserves” in the form of deposits with the Federal Reserve instead of lending them, which would have increased the broader monetary aggregates. 1 Because of these decisions by individual commercial banks, the Federal Reserve’s asset purchases supported the growth of broader monetary aggregates at normal, but not extraordinary, rates. Is the small amount of interest that banks earn on their Federal Reserve deposits acting as a disincentive for them to lend in the current environment?2
Some observers have argued that QE would be more effective if bank loans and monetary aggregates grew at faster rates. Some have suggested lowering the interest paid on reserves to encourage banks to make additional loans and thereby increase the broader monetary aggregates. The minutes from the September 2011 Federal Open Market Committee meeting indicate that the Committee considered this possibility.
How much would eliminating interest paid on reserves increase bank loans and thus monetary aggregates? The experience of the ECB provides a case study of such an effect. On July 11, 2012, the ECB reduced the interest it pays on excess reserves held at its deposit facility from 25 basis points to zero. The chart shows that the European monetary aggregates did not grow unusually fast in the months following this action. Of course, other factors also influenced monetary growth; the data cannot rule out a small positive effect from the small reduction of interest paid on reserves. As the economy improves, short-term rates rise, and banks begin to lend their excess reserves, we might get a much better picture of the relation between bank lending and interest paid on reserves.
1 Keister and McAndrews (2009) and Wheelock (2010) discuss in some detail why banks choose to hold large quantities of excess reserves.
2 In 2008, Congress granted the Federal Reserve the authority to begin paying interest on banks’ excess reserves to improve the New York Fed’s ability to manage the federal funds rate target and remove an implicit tax on holding reserves.