Empirical studies of the structures of national political influence are rare due to the lack of direct measures of influence. This study analyzes personal relationships with influential people as an indicator of political influence. Using male data from the 1975 and 1995 SSM surveys (the national surveys of Social Stratification and social Mobility), I found that (1) there was a definite regional difference in the possession of relational resources in 1975. (2) In 1995, a regional difference still remained but the association between possession of relational resources and the scale of urbanization was not linear ; the residents of small cities (under 10, 000 residents) had more relations with assemblymen than those of rural areas and of large cities. (3) Determinants of relational resources were age, property, and self-employed occupational status. (4) In 1995, variables relating to achievement such as higher education and holding a managerial post had strong effects on the possession of relations with influential people, and the effect of region had decreased since 1975. This indicates that the structure of regional difference in Japanese political influence has changed in recent years.