Zero nominal interest rates and deflation
By Stephen Cecchetti
March 24, 2003
Overnight interest rates have fallen to their lowest levels in decades. With the US Federal Reserve's federal funds rate at 1.25 per cent, the European Central Bank's main refinancing rate at 2.5 per cent and the Bank of England's repo rate at 3.75 per cent, we are approaching a very clear limit. And since nominal interest rates cannot go below zero, what will policymakers do if they reach that limit? Importantly, what will they do if nominal interest rates are zero and the economy is experiencing deflation?
These questions have been on the minds of central bankers since before the Bank of Japan, in response to a mild but persistent deflation, moved its policy rate to zero in early 1999. Policymakers have studied a set of unconventional options thoroughly and are convinced that they will work. What are these unorthodox policies? How do they work and what do they mean for policy both now and in the future?
The mechanics of unconventional monetary policy is straightforward and based on the fact that the central bank controls the size of its balance sheet. It is through changes in its assets and liabilities that it influences the economy. During normal times, policymakers operate by controlling the supply of their liabilities to meet an interest rate target. The details vary with each central bank, but the thrust is always the same. All monetary policy, conventional or not, is the result of balance sheet manipulation.
In recent public addresses, both Alan Greenspan, chairman of the Federal Reserve, and Ben Bernanke, his fellow governor, have suggested a series of options that they could pursue should they wish to ease policy further when the federal funds rate is at zero. Some of these concern technicalities about what the Fed would buy and how they would buy it. But their most important suggestion, and the one that has universal relevance, is that, once the target for the overnight interest rate hits zero, they would begin targeting longer-term interest rates. That is, they suggest that the Fed would put a ceiling on US Treasury securities interest rates at maturities of two, three or even 10 years. For example, the Fed might decide to put a ceiling of 3 per cent on the 10-year Treasury, while holding the federal funds rate at zero.
Operationally, this involves decisions about which assets they will hold. Everyone agrees that this is feasible. The Fed can simply stand ready to purchase these securities at a price consistent with the interest rate target. If the market price exceeds the cap, the interest rate is lower than the target interest rate. Otherwise, the Fed will simply start to accumulate the bonds in question.
These policies are bound to be effective, driving up prices and eliminating any deflation. However, a problem arises from the fact that we have no experience with them. While we have rules-of-thumb for how much a cut of 25 or 50 basis points in the federal funds rate affects output and inflation over the following two years, we do not know what the quantitative impact of these alternative measures would be. What would happen if the Fed reduced the 10-year Treasury rate by 25 basis points, while holding the funds rate at zero? While we can be sure of the direction things will go, we do not know how far or how fast.
Nevertheless, overall we should rest assured that central bankers have both the knowledge and the tools to continue to act even when the short-term interest rate that serves as their primary target hits zero. To think that there is somehow a limited amount of ammunition is simply wrong.
It is important to remember that these unconventional policies are part of the extensive contingency planning every central bank does as part of its daily routine. But they are like the fire insurance on your house: you are happy to have it, but you certainly do not want to use it. And the large uncertainties surrounding the implementation of these unorthodox measures make everyone wary of using them. This means that the mere fact that interest rates cannot go below zero has an important impact on policy at all times.
This line of reasoning leads to two important conclusions for policy today. First, the mere prospect of hitting zero means acting earlier and faster. What looks like caution in moving slowly, can turn out to be very risky. This was one element that led the Fed to lower interest rates 4.75 percentage points over a period of 11 months in 2001. While there is every reason to believe that tomorrow's meeting will not produce another cut, Open Market Committee members are surely ready to pull the trigger should conditions worsen perceptibly between now and their next meeting in early May. And, with the main refinancing rate at 2.5 per cent the ECB should be giving serious consideration to the same sort of pre-emptive action.
The second conclusion is that, in order to avoid having to use unconventional policies of uncertain effect, policymakers are willing to tolerate a bit more inflation. A year ago, the Federal Reserve would probably have been comfortable with consumer price inflation of 1-2 per cent. Today I am not so sure. My guess is that Fed officials are nervous enough that they are going to aim for inflation in the 2-3 per cent range in the hope that unconventional policies remain unnecessary. At least, I hope so.