실업률을 고려한 최적 통화정책 분석

This paper studies the role of unemployment as a feedback variable in monetary policy interest rate rules. While labor market variables are perceived to be important ingredients to explain business cycle fluctuations, many New Keynesian dynamic stochastic general equilibrium (DSGE) model literatures based on Korean economy consider output for the real sector variable in the feedback rule. We also investigate the role of growth variables in the monetary policy rules. Orphanides and Williams (2002) argued that a monetary policy rule responding to growth variables, such as unemployment growth, rather than gap variables can reduce policy mistakes from mis- perceptions of natural rates of output and unemployment. A monetary policy rule responding to output gap or unemployment gap could bring about unexpected or undesirable results, since natural rates of output and unemployment are not directly observed but measured with uncertainty or measurement errors, and policy actions based on such mistakenly perceived variables could cause the economy to move to the unintended directions. On the other hand, monetary policy rules can avoid such mistakes by responding to growth variables because such rules are not based on the uncertain measures. In order to address these issues, we build a DSGE model with sticky prices and wages and unemployment, and estimate the model using Korean data. The monetary rules considered are interest rate rules responding to past interest rate, inflation rate, and a real sector variable. We consider four different Taylor type rules with respect to a real sector variable. For a real sector variable in the feedback rule, output gap, unemployment gap, output growth, and unemployment growth are used in the analysis. Based on estimated model parameters, we search feedback rule coefficients in a way to minimize welfare losses and evaluate the performance of the rules. In order to analyze the optimal policy rules when output gap and/or un- employment gap are measured with errors, we simulate the case where the central bank adjusts interest rate responding to the output gap or un- employment gap which are measured with errors. We find that the rule responding to unemployment growth is more welfare improving than other rules. It is because unemployment growth contains information on not only output gap, but also price inflation and wage inflation which are components of welfare loss function. Therefore, responding to unemployment growth mitigates inefficiencies from sticky prices and wages at the same time. This result holds with and without uncertainty in output gap and unemployment gap. While, in case of cer- tainty, the rule responding to output gap is more welfare improving than the rule responding to output growth, the result is reversed in case of uncertainty because the rule responding to uncertain output gap measure misleadingly increases fluctuations of other variables. These results imply that, when central banks practice monetary policy, unemployment can be used as a useful feedback variable together with inflation and output vari- ables.