LAUSR

Purpose This study aims to examine the relation between population composition and financial market variables in post-war Japan. Design/methodology/approach Cointegration and Granger causality tests are applied to annual data for the period 1948-2015. Findings Accounting for nonstationarity, this study finds long-run equilibrium relations between real financial price (stock and house) indices and the proportion of population in the prime earning (45-64) or retirement (65+) age. Granger causality tests that account for possibly nonstationary variables find some evidence of dynamic causation running from the 45-64 cohort to the real financial price indices. No such evidence is found for the 65+ cohort. Originality/value This study complements the existing literature primarily based on US data with analysis of Japanese data that has some unique population composition features.