Spatial Regression Models on Factors Influencing Regional Minimum Wages

Abstract

Regional minimum wages might well represent the economic development of a region. The most likely spotlight province regarding the wage determination issue is East Java. This work is intended to obtain the best regression model on factors influencing East Java's regencies/cities' minimum wages in terms of spatial approach. The methods are Spatial Autoregressive (SAR) and Spatial Error Model (SEM). This study aims to obtain the best spatial model based on the factors influencing the regional minimum wage in districts/cities in East Java and the mapping. The data source is secondary data from Statistics Indonesia (BPS) of East Java. It consists of several variables, namely the Regional Minimum Wage, Total Working Population, Gross Regional Domestic Product, Total Population, and percentage of Population with a minimum education of senior high school. It shows that two significant factors are the number of working civilians and the percentage of high school-college graduates, affecting regional minimum wages. It proves that minimum wages among regions in East Java are spatially correlated with a closed area. Spatial regressions are the better ones than classic ones since they have higher R-sq and satisfy assumptions. Meanwhile, the selected model is SAR rather than SEM as it has a smaller AIC and explains variation better in minimum regional wages. It is indicated that some regions need more care due to small regional wages.