A Spatial Analysis of the Voting Patterns in the South Korean General Elections of 2016

In this study we analyze the spatial patterns in the South Korean voting behavior in the context of the 2016 general election along with the socio-economic determinants of the South Korean voters' choice. To this end we applied spatial econometric analysis to a unique dataset on the outcomes of the 2016 general elections in South Korea at a highly disaggregate level of 229 provinces. Our empirical model accounts for three types of spatial dependence in the data that has to do with the fact that geographic proximity may imply similar voting behavior. Our empirical findings align well with the existing evidence on South Korean voting behavior, in particular regarding the influence produced by the voters' region of origin, and their age. Surprisingly, we do not find economic characteristics such as the regional income per capita or the rate of unemployment to produce a statistically significant effect on South Korean voters' choice. However, our results imply that a sound fiscal policy by the local government may act as a signaling device distinguishing between a conservative and a liberal political agenda. Our finding of the older voters leaning towards the conservative edge of the political spectrum suggests that the "silver democracy" now actively discussed in the South Korean media is increasingly assuming more conservative traits.