The Bank of Canada's Reaction Function

Abstract

Introduction The Bank of Canada (BOC) is charged with influencing the price level, employment, and growth of the Canadian economy. In recent years the BOC has developed a reputation of being very concerned with a low rate of inflation, perhaps to the exclusion of other goals. Critics have charged that the Bank is too concerned with anti-inflation targets, and should implement policies that pay more attention to unemployment and economic growth (Lucas, 1989). Supporters of the BOC's policy contend that such actions would merely cause inflation and a devaluation of the dollar, and cause no increase of economic growth in the long run. Whatever policy the Bank of Canada pursues, it must consider a number of economic factors. Its actions may influence, and may be influenced by, such things as the growth rate of the economy, the unemployment rate, the rate of inflation, exchange rates, and the actions of other central banks. Its response to changes in these factors is undoubtedly complex. The factors are likely to change in conflicting ways, and the Bank may be torn as to how to respond. For instance, it is quite possible that the rate of inflation may increase in the same period that the value of the Canadian dollar increases. The BOC may want to increase interest rates to dampen inflation, but at the same time may want to cut interest rates to mitigate the adverse effects that a strong Canadian dollar could have on exports. Because it cannot do both, the Bank must decide which is more important and react accordingly; or if it values each effect equally, do nothing. The purpose of this paper is to discover which economic factors matter most to the BOC as it implements its monetary policy. Specifically, ordinary least squares (OLS) regressions are performed to capture the effects of changes in output, inflation, unemployment, exchange rates, and the U.S. federal funds rate on the changes that the Bank of Canada makes in the overnight interest rate. The structure of this paper is as follows: The next section briefly describes past literature regarding central banks' reactions to a wide variety of economic factors. A third section describes the data and variables selected for the analysis. The fourth section presents the OLS results, and a final section provides a discussion of the results. Literature Review Previous research on the topic of central banks' response to economic variables has for the most part been limited to the behavior of the U.S. Federal Reserve. (1) Much of this research was undertaken prior to the mid-1980s and did not attempt cross-country comparisons. (2) Most of this older research found that the Federal Reserve reacts to increased unemployment by loosening monetary policy, and reacts to increased inflation by tightening monetary policy. In those studies including international factors, little reaction was found to changes in the balance of payments surplus or changes in the exchange rate. Other factors such as measures of fiscal policy and government debt, measures of monetary aggregates, and output are not always examined in this older work, and where it is included there is little consensus. A more recent vein of reaction function research has focused on the Taylor Rule (Taylor, 1993), a mathematical relationship suggesting that central banks should manipulate their short term interest rates in response to the deviation of inflation from some target level and the deviation of output from its potential level. Recent articles that estimate this rule (Judd and Rudebusch, 1998; Taylor, 1998) find that these factors do quite well in explaining movements in the federal funds rate. Although most research on monetary policy reaction functions has been performed on U.S. data, there are some studies that shed light on the BOC responses. One of the first of these (Reuber, 1964) specifies a regression with the Canadian money supply as the dependent variable and unemployment, the consumer price index, and productivity as explanatory variables. …