Comment on “Aging, Savings, and Public Pensions in Japan”

Abstract

A good policy paper is, in general, either written well before any policy action is required, or is written well after everything is settled and gives a fair evaluation of the entire historical experience. Horioka et al . (2007) is a unique combination of the two. It is concerned with Japan's savings behavior in a long historical perspective and an evaluation of public pension reform and its impact on Japan's savings rate in which a real challenge will come in the middle of the twenty-first century. This paper is written in a very simple manner, but its content is very deep and complex. Many issues are yet to be solved. Horioka et al . argue that (i) Japan's high household savings rate was a temporary phenomenon during the 1955 -1995 period, and that it was not unusually high during the prewar and early postwar periods or after 1995; and (ii) the demographic structure of Japan's population can explain the level as well as the trends of the household savings rate over time. Putting these two arguments together, the authors seem to suggest that the demographic phenomenon was unusual in the 1955 -1995 period, and it was normal before 1955 and after 1995. I cannot observe a steady-state demographic structure before 1955 and after 1995. If the actual demographic phenomenon are observed, the baby boom in the 1946 -1949 period and the dramatic drops in fertility rate after the 1990s are two of the biggest demographic changes in the post-World War II period. The authors need to pay more attention to the demographic history per se. The authors' point that the rapid aging of Japan's population is likely to cause Japan's household savings rates to continue declining is well taken. But I have difficulty in finding any evidence that the demographic structure peaked out at around 1973 when Japan's household savings rate was at its highest. In 1973, the inflation rate reached its highest level in the postwar period, and generous public pension reforms and civil servants' wage structure reforms were carried out by the Tanaka government. The demographic story of household savings behavior does not seem to explain the turning point in 1973 -1975. I might expect that the demographic structure would be one of the main explanatory variables, but the rapid inflation in 1973 -1975 and other elements would explain temporal increases/decreases in the savings rates from time to time. The authors argue that existence of a pay-as-you-go public pension system increased the savings rate of the cohort born after 1965. This argument is based on simulation results from the OSU II model, which suggest that cohorts born in 1965 or later will receive less than they contribute to the public pension system. However, the OSU II model does not