The male suicide mortality in Japan reached the highest in 1998, since statistics were first recorded in 1899. We investigated whether the recent high suicide rates were associated with economic and demographic factors, using the Pearson's correlation and multivariate regression analyses. The annual numbers of suicide deaths, economic variables, and the proportion of elderly people between 1995 and 2000 were obtained for all the 47 prefectures in Japan. The correlation analyses showed that the male SMR of suicide was mostly associated with the economic variables and the proportion of the elderly men. The female SMR of suicide was associated with the job application rate, and the proportion of elderly women in several years. The multivariate regression analyses showed that the male SMR of suicide was mostly associated positively with the job application rate and the proportion of elderly men and negatively with the amount of household savings and public assistance rate. The largest proportion of the variance of the male SMR of suicide was explained by the set of independent variables in 1998. The goodness-of-fit of the model concerning the female SMR of suicide was poor.