The danger of prematurely raising interest rates

October 12, 2009

Central bankers like Mark Carney of Canada are musing aloud about the possibility of raising rates despite very high unemployment, a highly valued Canadian dollar heading toward parity with the U.S. dollar and generally suppressed aggregate demand. The situation is even worse in the U.S. where unemployment now is 1.5 % points higher than in Canada, standing at 9.8 %. Whenever central banks raise rates to cut off a recovery or slow down a boom it usually takes over a year for the impact to be felt. If one examines the data from the 1990-92 recession in the U.S. we can see this quite clearly. I have examined the monthly change in the federal funds rate, the rate of change in the monthly Consumer Price Index and the monthly result for the unemployment rate from 1988 to 1995 for the U.S. The results   clearly show that it takes a number of months for the interest rate rises to have an impact but once they do they raise the rate of unemployment for many months to come despite reductions in the interest rate once the bank has clearly unleashed a recession. There are very big risks then for the central bank to prematurely raise rates particularly when globally there is still deflation and high unemployment, even if there is a lag time of several quarters before the   rate rises begin to bite.

Paul Krugman had a good piece on this in the New York Times today and he is right to worry that monetarist leaning central bankers will jump the gun on rate rises unless they are pressured not to do so. The problem is they base their decision far too much on a NAIRU mentality and take far too seriously expectations among bond dealers and traders rather than among the general public as to which direction prices will be heading.

Just to make the argument clearer look at the evidence from the time series I have identified above.(These series are all available for the U.S.Bureau of Labour Statistics and the Federal Reserve.) The federal funds rate was 6.58 % in February 1988. At that time unemployment was 5.7 % and month to month inflation 0.3 %. By March 1989 the federal funds rate had peaked at 9.85 % when unemployment was 5.0 % and inflation was running at 0.5 % month to month. The unemployment rate began to rise slowly at first, but then by December 1990 it hit 6.3 % rising to 7.3 % by December 1991. It peaked at 7.8 % in June of 1992. It stayed above 7% for the next 12 months even as the federal funds rate dropped to 2.96 %. It stayed above 6.0 % for the next 14 months after that despite federal funds rates below 4% for most of that period.

Overall unemployment was above 6 % for a total of 46 months imposing considerable unnecessary hardship on Americans. The only positive accomplishment was lowering the inflation rate from over 4.5 %-5.2 % down to 2.5 %. A better employment performance would have been possible by accepting a 3-4 % target for the inflation rate. Unemployment above 6 % for close to five years was a high price to pay for this extra 1.5 % point gain in the inflation reduction result.

Central bankers ought to think long and hard before they prematurely pull the trigger on interest rate rises.