Bankruptcy Dynamics in Japan: A System Estimation Approach

Abstract

The number of bankruptcies in Japan fluctuates over time, and the characteristics of these fluctuations have not been sufficiently investigated. This paper intends to examine the trends in bankruptcies in Japan, with a particular focus on identifying the dynamic features of this series. Using quarterly data from 1975 to 2005, we first classify bankruptcies into three industrial sectors and two levels of firm size. Then, for each of these six (three by two) categories, we estimate a vector autoregression (VAR) system comprised of four economic variables (the Tokyo Stock Price Index (TOPIX) by industry, the break-even point ratio, the debt-equity ratio and the quick assets ratio) and the bankruptcy rate. The VAR is a simple but widely used approach for identifying the path and depth of interrelationships among economic time-series variables. We also construct impulse response functions, which enable us to assess the dynamic features of the system. In particular, considerable differences exist in both the strength and lag-structure of dynamic responses to shocks across different industrial sectors and firm sizes. It indicates that the total number and movement of bankruptcies reflect the accumulated impact of various shocks that have distinct response structures in each sector and firm size.