In order to clarify the problems of rising and persistently high unemployment rate in the recent Japanese economy, the paper attempted to examine th regional dimension of unemployment. At the beginning of the paper, this authors presented a brief review of the long-run movement of regional unemployment rates covering the just thirty years. It is clear that the differentials of unemployment rates between regions were smaller compared to the differentials in 1965 or 1970. On the other hand, the paper tried to define the structural determinants of regional unemployment rate, based on the following model, u=a_0+a_1A+a_2S+a_3V where u is unemployment rate of a region, A is the age-structure of labour force, S is the variable representing industrial structure, and V is an index of balance between supply of and demand for labour. Although the directly estimated results of the above model seem to confirm some important effects of age-structure of labour force on regional unemployment rate, they were not completely convincing and satisfactory. To extract the age-structure effect on regional unemployment independently, standardised unemployment rates were calculated and then compared with actual unemployment rates by region. The result shows the clear but minor impact of age-structure upon regional unemployment rates. When the standardised rates of unemployment were regressed on the variables concerning the industrial structure and labour market condition, these variables could explain about 65 per cent and 83 per cent of the regional variation of unemployment rates for males and for females respectively. One of the most interesting findings is the pronounced effect of industrial structure on regional unemployment rates, in term of the ratio of occupants engaged in tertiary industry to those in secondary industry.
