Depository institutions — banks — are contractually obligated to hold a certain level of reserves at a Federal Reserve bank, more or less to demonstrate a baseline of solvency. On top of that, banks can leave excess reserves, if they wish. In October 2008, less than a month after Lehman Brothers’ chapter 11 bankruptcy, the Federal Reserve for the first time began paying interest on these excess reserves. According to the October 6, 2008 press release, the objective was “to address conditions in credit markets” (link).
Immediately upon the policy reform, banks feverishly moved cash to the Fed, from $200 billion within a month, to $800 billion by the end of 2008, to a peak of $1.6 trillion in July 2011. Today there remain $1.4 trillion in excess reserves sitting idle at the Federal Reserve (link).
My analytical view is these excess reserves could tell us a lot about what to expect from the US economy going forward. If banks have the incentive (and if the Federal Reserve/US government can afford and/or manage) to move this cash into the private sector, through investment or spending, the economy could enjoy a ready-built stimulus of significant proportions.
As such, a policy change that ought to be debated is whether the Federal Reserve should reduce or eliminate the interest it pays on excess reserves, and return to the pre-Lehman reality. This study by two at the NY Fed argues the Federal Reserve, not the banks, is behind the growth in excess reserves, a finding that suggests the Federal Reserve could engineer the reserves’ unwinding, as well.
My concern is this: it is hard to see imagine a real flourishing recovery in the US economy as long as banks continue to sit on this level of reserves, and I find it a puzzle why the Fed would continue to provide incentive for so much dormant money.
In favor of rethinking the policy is Bruce Bartlett, writing at the NY Times, with a column headlined “The Fed Should Stop Paying Banks Not to Lend” (link). An excerpt:
In a Wall Street Journal commentary on July 22, Alan S. Blinder, the Princeton economist and former vice chairman of the Federal Reserve Board, suggested that lowering the interest rate the Fed pays banks on their reserves would force them to lend more and thus stimulate growth.
To explain why this is potentially important, it is first necessary to explain something about the relationship between the Fed and the banking system.
Typically, when the Fed wishes to stimulate growth it increases the money supply. It buys Treasury securities from banks and credits their accounts at the Fed. All banks maintain accounts at the Fed, just as people have accounts at commercial banks. When the Fed credits their accounts, they have more money to lend.
Banks must maintain a certain level of required reserves in the form of vault cash or balances at the Fed as a percentage of their deposits, in order to provide adequate liquidity. Reserves over and above those required are called excess reserves.
Historically, banks held as little in the way of excess reserves as possible, because this was money that could be lent immediately, upon which no income was earned. One way income on excess reserves could be earned was by lending them to other banks overnight, through what is called the federal funds market.
The interest rate charged on such overnight loans is called the fed funds rate and it is essentially controlled by the Federal Reserve, which routinely adds or subtracts reserves so as to meet its target rate. Since Dec. 16, 2008, the target fed funds rate has been between zero and 0.25 percent.
Under normal conditions, such a low fed funds rate would be more than adequate to create a considerable amount of new lending. Since the rate is the basic cost of money to banks, they would make a profit even if they made loans at a 1 percent interest rate.
But rather than make loans, banks instead are simply sitting on the money, so to speak. According to the Federal Reserve, they have $1.5 trillion in excess reserves. This is extraordinary. It is as if individuals took $1.5 trillion of their savings out of stocks, bonds and every other income-producing financial asset and put it all into non-interest-bearing checking accounts back in 2009, and just left it there.
Economists have puzzled about this phenomenon for years. They note that historically the Fed never paid interest on reserves, but in October 2008 it began doing so. Moreover, the Fed pays interest on excess reserves as well as required reserves. Originally, the rate was 0.75 percent to 1 percent, but since Dec. 17, 2008, it has been fixed at 0.25 percent.
This may not sound like much, but keep in mind that interest rates on United States Treasury securities with maturities of less than two years are currently less than 0.25 percent. The effective fed funds rate is also lower than 0.25 percent. In recent weeks, it has been as low as 0.13 percent. Compared with these rates, a riskless return of 0.25 percent looks pretty good.
In the end, however, Bartlett proposes a pretty cautious policy response:
[In] a liquidity trap, monetary policy is impotent because the Fed cannot reduce nominal interest rates below zero.
However, the Fed can penalize banks for holding excess reserves by charging them interest rather than paying them interest. This has been done in other countries. From July 2009 to July 2010, the central bank of Sweden charged banks 0.25 percent on their reserves, and on July 5 the central bank of Denmark announced that it would begin charging an interest rate of 0.2 percent on reserves.
In effect, this reduces the real interest rate received by banks and thus, ironically, would ease monetary policy and encourage bank lending.
No one thinks the Fed is ready to follow Denmark and Sweden. However, economists speculate that the Fed may be ready to reduce the 0.25 percent rate paid on reserves and see what happens. It would be a move in the right direction.