انعطاف ناپذیری اسمی و سیاست های پولی در کانادا

This paper develops and estimates a dynamic, stochastic, general-equilibrium model with price and wage rigidities to analyze monetary policy in Canada. A monetary policy rule allows the Bank of Canada to systematically change the short-term nominal interest rate in response to deviations of inflation, output, and money growth. The structural parameters of the model are estimated econometrically using a maximum-likelihood procedure with a Kalman filter. The estimates reveal that either price or wage rigidities are key nominal frictions that generate real monetary effects. Furthermore, the simulation results show that the Bank has, since 1981, increased the short-term nominal interest rate in response to exogenous positive demand-side disturbances, and used modest but persistent reductions in the nominal interest rate to accommodate positive technology shocks.

In recent years, an extensive literature has emerged on the role of nominal rigidities in shaping key features of the business cycles and in evaluating short-run dynamic monetary policy effects on aggregate variables. Researchers have generally used dynamic, stochastic, general-equilibrium (DSGE) models in which prices and/or nominal wages are sticky. These models act on the assumptions that private agents have rational expectations and that their optimizing behavior determines the time paths of nominal and real variables, such as output and inflation. Furthermore, in contrast to the previous generation of models, these new models predict that the real effects of monetary policy shocks would differ sharply under sticky prices and sticky wages.1 Chari et al. (2000) show that staggered price-setting alone does not generate endogenous persistence in an economy of imperfectly competitive price-setters. Christiano, Eichenbaum and Evans (CEE) (2005) find that a version of a DSGE model that has only nominal-wage rigidity does almost as well as a model with price and wage rigidities, while the version of their model that has only price rigidity gives very poor results. Erceg et al., 2000, Huang et al., 2000, Kim, 2000 and Huang and Liu, 2002 find that combining staggered wages and imperfectly competitive households generates more output persistence in response to monetary policy shocks. Dib (2003) uses a closed economy, estimated DSGE model with both nominal and real rigidities for the Canadian economy to show that the degree of nominal rigidity is substantially increasing in the presence of some form of real rigidity. The Bank of Canada manages short-term nominal interest rates to control inflation. Thus, the inflation rate has fallen significantly from its peak in 1981 and has remained low and stable since then. Moreover, the long-run inflation decline has been accompanied by long-run declines in the short-term nominal interest rate and money growth. The dynamic relationships between the nominal interest rate, money growth, and inflation may reflect the ways in which the monetary authority and private agents respond to economic disturbances. Following Ireland, 2001, Ireland, 2003, Kim, 2000 and Dib, 2003, we develop and estimate an optimization-based model for the Canadian economy to examine how price and nominal wage rigidities allow the Bank’s policy actions to affect economic activities.2 Though Canada is a small open economy, using a closed economy framework is still a useful exercise to estimate and simulate DSGE models if we do not address issues that require an open economy framework. Clarida, Galí and Gertler (CGG) (2001) argue that the optimal policy problem of a small open economy is isomorphic to that of a closed economy. Similarly, Khan and Zhu (2002) estimate a sticky-information model using both closed and open economy frameworks for Canada and the United Kingdom. They find that open and closed economies lead to similar parameter estimates for both countries.3 Our model features monopolistic competition between firms and between households, nominal rigidities in the form of price- and wage-adjustment costs, and convex costs of adjusting capital. The model includes four sources of disturbance: monetary policy, money-demand, technology, and preference shocks. Temporary rigidities in prices and nominal wages allow the Bank of Canada to affect the behavior of real variables in the short term. Furthermore, under these nominal rigidities, exogenous money-demand shocks become a significant source of aggregate fluctuations. Empirical work shows that such shocks are large and highly persistent.4 Taylor (1993) describes Federal Reserve behavior with a monetary policy rule that adjusts the short-term nominal interest rate in response to inflation and output deviations. However, this paper follows Ireland, 2001 and Ireland, 2003, who generalizes Taylor’s specification, by allowing the Bank of Canada to respond to deviations of money growth as well. Such a policy still implies an endogenous money supply. An increase in inflation allows the interest rate to rise which, in turn, reduces nominal asset demand and restrains money supply growth. Similarly, if money growth increases, reserve demand will rise, and the Bank will increase the nominal interest rate, which should automatically reduce aggregate money demand.5 To evaluate empirically Canadian monetary policy, the structural parameters of four versions of the model are estimated using a maximum-likelihood procedure with a Kalman filter: three alternative versions with nominal rigidities (sticky-price and/or sticky wage models) and a version with perfectly flexible prices and wages. Quarterly data on consumption, a short-term nominal interest rate, inflation, and real balances are used. Since the Bank of Canada effectively abandoned M1 growth targeting by the middle of 1981, the data used cover the period 1981Q3–2000Q4. The maximum-likelihood estimates reveal that price- and wage-adjustment cost parameters are of reasonable magnitudes. They imply that, on average, prices remain unadjusted for more than two-quarters in the model with only price rigidity, but that they are almost completely flexible in the model combining price and nominal wage rigidities. The average estimated duration of unadjusted nominal wages, however, is about three-quarters in both models that include nominal wage rigidity. Thus, the nominal wage rigidity appears to be more important than price rigidity in explaining real effects of monetary policy shocks. On the other hand, the likelihood ratio test suggests that the model with no nominal rigidities is strongly rejected against the models considering some form of price and wage stickiness. The estimated values of the monetary policy rule coefficients are positive and highly significant. They indicate that the Bank has actively responded to the deviations of inflation, output, and money growth by increasing the nominal interest rate. Thus, monetary policy during this period can be better described as following a modified Taylor (1993) rule that adjusts the short-term nominal interest rate in response to deviations of inflation, output, and money growth from their steady-state levels. This result is similar to that found by Ireland, 2001 and Ireland, 2003 for the US economy. Alternatively, monetary policy can be described as influencing a linear combination of the nominal interest rate and the money-growth rate to control inflation and to reduce output fluctuations. The simulation results show that, the Bank has successfully reduced the effects of money-demand and preference shocks on output and inflation by modestly, but persistently, increasing the short-term nominal interest rate. More importantly, the Bank has also reduced the short-term nominal rate to accommodate technology shocks. This finding is consistent with the results, found by Ireland (1997b) and CGG (2001), showing that an optimal monetary policy should actively respond to technology shocks. The rest of the paper is organized as follows. Section 2 presents the model. Section 3 describes the data, the calibration procedure, and the econometric method used to estimate the structural parameters. Section 4 discusses the empirical results. Section 5 offers some conclusions.

The Bank of Canada follows a monetary policy that actively manages the short-term nominal interest rate to control inflation. Instead of Taylor’s (1993) rule, we assume that the Bank responds not only to output and inflation deviations from their steady-state levels, but also to those of money growth. To evaluate this policy, a DSGE model with price and nominal-wage rigidities has been developed and estimated for the Canadian economy. To compare the implications of nominal rigidities three versions of the DSGE model were estimated using a maximum-likelihood procedure with a Kalman filter: sticky-price, sticky-wage, and combined sticky-price and sticky-wage models, as well as a model with flexible prices and wages. The estimates reveal that either price or nominal wage rigidities are key nominal frictions to account for real monetary policy effects, and the hypothesis of price and wage flexibility is strongly rejected. Furthermore, the estimates of the monetary policy rule coefficients indicate that, since 1981, Canadian monetary policy has responded actively to changes in inflation and money growth, but modestly to output deviations. Alternatively, one can consider that the Bank has used a policy that influences the short-term nominal interest rate and money growth to control inflation. The decomposition of the forecast-error variance reveals that both technology and preference shocks are the most important sources of output fluctuations in both the short and long terms followed by monetary policy and money-demand shocks. The simulation results show that the responses of output, inflation, and money growth to monetary policy shocks under the modified Taylor (1993) rule are very similar to those generated under a rule with a smoothing terms. More importantly, by a small and persistent increase in the short-term nominal interest rate, monetary policy has been able to reduce the negative effects of money-demand and preference shocks on real economic activities. Monetary policy has also accommodated positive technology shocks by persistently reducing the short-term nominal interest rate. Thus, as a positive technology shock produces temporary deflationary pressures, the Bank responds to these pressures with a modest but persistent reduction in the short-term nominal interest rate. In contrast, to reduce the inflationary pressures implied by positive demand-side (preference) shocks, the Bank modestly, but persistently, increases the short-term nominal interest rate and temporarily decreases money growth.