هدفگذاری نرخ تورم و سیاست پولی در کانادا : تاثیر بر عدم قطعیت تورم چیست؟

Inflation uncertainty has been demonstrated both theoretically and empirically to lower real output. This paper examines the impact of inflation targeting in Canada on inflation uncertainty, as well as persistence. Our results indicate that inflation targeting lowered inflation persistence, but actually increased uncertainty. Such an effect may be due to the failure of the previous formal target, the M1 money supply, to successfully combat inflation

Canada adopted a formal target for inflation in 1991. Adopting official targets for inflation is the latest of formal goals for monetary policy (the others being monetary aggregates and exchange rates). Advocates claim that a formal target for inflation will increase credibility, and thus decrease inflation persistence by anchoring expectations in the face of inflation shocks. Moreover, by anchoring expectations, inflation targeting (hereafter IT) should decrease inflation uncertainty. If true, this is an important feature of IT; while a moderate level of inflation may be neutral for output in the long run, inflation uncertainty can potentially lower investment and hence output growth (Cabellero, 1991; Craine, 1989; Friedman, 1977; Grier, Henry, Olekalns, & Shields, 2004; Grier & Grier, 2006). Canada announced an IT in steps between 1991 and 1993. Honda (2000) finds that the level of inflation was not affected by targeting, while Johnson (1997) finds that expected inflation experienced no change from IT. However, IT may still have an important effect, if it can lower inflation uncertainty, or the persistence of inflation shocks. Using data from 1970 through the second quarter of 2006, we employ the GARCH (Generalized Autoregressive Conditional Heteroskedasticity) methodology to estimate the impact of IT on inflation persistence and uncertainty. We find that while IT does appear to shorten the life of inflation shocks, it has actually increased uncertainty for any given level of inflation. We speculate that this increase in uncertainty in response to a formal target may be due to the failure of the previous formal target, that for M1. Given that uncertainty has been shown to impose real costs, the failure of IT to lower uncertainty in Canada gives a cautionary lesson for other nations contemplating adoption of a formal target for price level changes. This paper proceeds as follows. The next section describes the history of IT in Canada and the literature on the effect of inflation uncertainty. The third section explains the GARCH methodology we will utilize to investigate the impact of IT. The fourth section explains the results, and the fifth section concludes.

Several recent studies such as Honda (2000) and Ball and Sheridan (2005) have found results indicating that IT exerts no significant impact on the level or expectations of inflation. Our results suggest that, for the case of Canada, IT has actually raised inflation uncertainty, given the past level of inflation, although it has helped lower the persistence of inflation shocks, and was followed by a lower level of inflation. As there is a large literature on the negative impact of inflation uncertainty on real economic activity, the failure of IT to lower uncertainty is disappointing. Given that IT entails both costs and benefits (Kohn, 2005), these results suggest caution for other nations considering an inflation target. It is important to note, however, that the results should not be taken to suggest that IT can never be beneficial under any conditions. Country circumstances differ; some nations may have such a difficult inflationary history that a formal, visible target may be the only way to control price level increases. Moreover, even in the case of Canada, IT does appear to lower the persistence of inflation shocks. Benefits for other countries may be greater.