Long-Run Effects of Macro Shocks on the Korean Stock Market: A Structural VAR Approach

Abstract

This paper examines the long-run effect of major macroeconomic shocks, using the structural vector autoregression (SVAR) approach, on stock prices for the Korean stock market, one of the leading emerging markets. Novel features of the modeling is to expand the dimension of the bivariate SVAR model of Blanchard and Quah (1989) and of Hess and Lee (1999) by including stock market shocks in addition to the pre-identified supply and demand shocks. Moreover, our three-variable SVAR model is further generalized to a five-variable SVAR, augmenting interest rate shocks and foreign exchange rate shocks. The model is estimated using monthly Korean market data from January 2003 to September 2015, and further estimated with two sub-samples using the recent global financial crisis period and non-crisis period. The estimation results using a three-variable model show that stock returns have a negative relation with demand shocks and a permanent positive relation with supply shocks. Risk premium shocks show a negative relation with inflation and a positive relation with real output growth. Interestingly, the subsample results indicate that the market fluctuations during the global financial crisis period have relatively little effect on the Korean stock market and its macro variables. With the five-variable model estimation, a ‘price puzzle’ is documented with evidence of a positive correlation between inflation and the interest rate shock. A positive correlation is also observed between inflation and the supply shocks in the long run. Lastly the forecast error variance decomposition indicates that the monetary policy change is the leading determinant of the long-run level of stock prices in the Korean stock market. These findings suggest that the recent expansionary monetary policy should be cautiously implemented in accordance with inflation movement.