This study verified the relative consequence of influential factors on social welfare expenditure as a key parameter in the development of social welfare policy and correlations between those factors along with multiple regression analysis. In other words, this study performed an empirical analysis about the relationships between a dependent variable(the amount of social welfare expenditure per capita as spent by Korean local government from 1995 to 2002) and two independent variables, i.e. socioeconomic variables(per capita GNP, local financial independence, the ratio of the aged over 65 and the ratio of the poor) and political-administrative variables(rate of political participation, inter- party political competition, local financial grant). This study developed a framework helpful to analyze determinants with related variables selected, while researching and designing a theoretical model about determinants in social welfare expenditure so as to come to final conclusions by empirical analysis(correlation analysis and multiple regression analysis). As the result of analysis, it was found that the results of multiple regression analysis between political -administrative variables(independent variable) and per capita social welfare expenditure(dependent variable) amounted to 87.7% in explanatory power, which was relatively lower than explanatory power(98.3%) of results obtained from multiple regression analysis of socio-economic variables. In the analytic framework of mixed model, it is difficult to explain a relationship like ‘political-administrative variables → socio- economic variables → production of social welfare expenditure’, but it is convincing to set up another relationship like ‘socio-economic variable → political-administrative variable → production of social welfare expenditure.’ In this regard, it is reasonable to set up a model that describes a modified version of sequential relationship as suggested herein. Based on results of analysis herein, this study could come to the following conclusions: First, the ratio of the aged over 65 has a considerable impact on the amount of social welfare expenditure per capita, which implies that such influential trend is so significantly associated with our contemporary aging-induced social problems in South Korea, because it will be possibly necessary to keep spending social welfare expenditures as a part of remedial solutions for ever-intensified social elder problem. Secondly, social welfare expenditure is still under the influence of per capital GNP, and the higher GNP yields the higher financial income and the more availability of governmental resources necessary to promote social welfare policies. Thirdly, another noteworthy point is the financial independence of local government. Only in view of the results hereof, it may be concluded that the higher level of financial independence had a tendency to the lower amount of per capita social welfare expenditure. This tendency points out that budget expenditure is largely focused on our sense of welfare and local development. Fourthly, it was found that interparty political competitions had impact on social welfare expenditure, which implies that local candidates from each party struggle to advocate public commitment for the benefit of low-income class, so the amount of social welfare expenditure becomes increasing inevitably. In view of the results hereof, it is somewhat controversial whether the extent of change in factors of social welfare expenditure can be enough identified only with 8-year data on local government\'s welfare policies after local government system was initiated. However, it is so significant that this study verified the welfare budget of Korean local government successfully based on research results. Especially, it is noteworthy that this study analyzed determinants in budget-making process after intermediate progress of local government system in Korea. Summing up, since determinants in the social welfare expenditure of local government are affected by various factors, it is recommended to explore countermeasures from an integrated angle by allowing for the appropriateness of determinants in social welfare expenditure, rather than focus on the explanatory power of a certain factor.