The purpose of this paper is to examine how labour market changes impact the change in the aggregated household consumption, which is a topic that is under-researched in Japan.
The author uses a three-step approach. The first step is a descriptive overview of the trends for cohort, age and year. The second step is to test the variation attributable to age-period-cohort interactions using APC analysis. The third step is to check if identifiable linear trend exists between the consumption changes and the labour market changes.
The analysis shows that major labour market changes per se do not contribute to household consumption adjustment. Meanwhile, the labour market conditions at the time of joining the labour force may be more important in shaping consumption during working life period than labour market changes during employment.
The cohorts are created based on birth years, which is a limitation imposed by the data availability rather than characteristics of the population group. The main reason behind the limited data points is the survey being conducted every five years and 2014 being the most recent year for which the data are available.
Research about cohort consumption in Japan is limited to the consumption composition changes or to the growing population of unmarried singles. This analysis will examine how labour market changes impact the change in the aggregated household consumption, which is a topic that is under-researched in Japan. In the analysis, the author uses the Python APC model and Python statsmodel OLS regression, providing the notebooks with code and full results in Appendices.
