In this paper, we develop a new way to test hypotheses about policy-makers' targets, and we implement that test for Canadian monetary policy. If, for example, the Bank of Canada is using interest rates to target an inflation rate of 2 per cent and there is an 8-quarter lag in the effect of the interest rate on inflation, then deviations of inflation from 2 per cent should be unforecastable and uncorrelated with any information in the Bank of Canada's information set lagged by 8 quarters. This would imply that empirical causality tests of monetary policy on inflation could be very misleading. Our test indicates that there was indeed a major change in the Bank of Canada's objectives about the time when formal inflation targets were announced.

Also published as:

Canadian Journal of Economics (0165-1765)
May 2002. Vol. 35, Iss. 2, pp. 239-56