A STOCHASTIC FORECAST MODEL FOR JAPAN'S POPULATION

Obtaining appropriate forecasts for the future population is a vital component of public policy analysis for issues ranging from government budgets to pension systems. Traditionally, demographic forecasters rely on a deterministic approach with various scenarios informed by expert opinion. This approach has been widely criticized, and we apply an alternative stochastic modeling framework that can provide a probability distribution for forecasts of the Japanese population. We find the potential for much greater variability in the future demographic situation for Japan than implied by existing deterministic forecasts. This demands greater flexibility from policy makers when confronting population aging issues.