Modelling the macroeconomic effects of population ageing in Japan and the international economy

Abstract

Most developed nations are experiencing a transition towards higher median ages and slower population growth. Many developing nations are likely to experience similar transitions in the coming decades. The implications of such demographic changes for economic growth and standards of living, public finance, and international capital flows have been the subject of much discussion and research. This study seeks to make a contribution to the literature on the modelling of macroeconomic effects of demographic transition. Specifically, it seeks to adapt, extend and enhance two well-known modelling frameworks: the empirical MSG3 model (McKibbin and Wilcoxen, 1999) and the theoretical Blanchard (1985) model.

The MSG3 model is an important tool in the multi-country, general equilibrium modelling literature and is well suited to the analysis of saving and investment, capital accumulation, economic growth, standards of living, international capital flows, and transition dynamics in a general equilibrium context. In this study, we report some of the efforts that have been made in using formal mathematical analysis, as well as empirical implementation and calibration, to adapt the MSG3 framework such that it becomes suitable for the analysis of population ageing. We also report some key findings from simulations based on the adapted MSG3 model.

Some of these findings are consistent with those reported in previous studies. For example, other things being equal, a fall in the birth rate is likely to result in slower growth in labour supply and real output, as well as in per capita real GDP and consumption. Other findings help shed new light on old questions, especially those involving transitional dynamics. For example, the contrast between simulations with and without the presence of children in the model helps to clarify the effects of demographic change on investment and saving: if the transition to slower population growth is anticipated well in advance, the short-term saving response (a rise in saving) may outweigh the staggered investment response (a rise in investment) so that the country tends to export capital (experience current account surpluses) for a number of years.

This study also makes contributions towards enhancing the Blanchard (1985) model, a seminal framework that has served as the conceptual basis of numerous analyses of policy changes and demographic shocks. The Blanchard model makes a simplifying assumption, namely that all individuals face a common mortality rate. To relax this assumption, we apply an overlapping generations approach to the Blanchard model. In the new resultant model, an individual's mortality rate rises with the person's age, and this age-mortality relationship is allowed to change over time. A version of this theoretical model is numerically implemented and simulated.

The new discrete-time, cohort-based theoretical model is readily amenable to direct calibration with the use of historical data and authoritative projections. By incorporating variable (increasing) mortality rates, it is better equipped to capture key demographic features such as the population age structure, in comparison with a corresponding model based on the constant-mortality assumption. Results obtained from simulations of the empirical model indicate that such differences in demographic modelling translate into material differences in projections for important macroeconomic variables, including per capita output. Show more