This paper, using the PIAAC (The Programme for the International Assessment of Adult Competencies）data, empirically analyzes the size of the public sector wage premium, which cannot be explained by the human capital. In this study, we estimated the wage premium by adding cognitive skills to covariates controlled by the conventional Mincer-type wage function. The main findings are as follows. The public sector wage premium, which cannot be explained by human capital including cognitive skills measured by the PIAAC, is around 4 to 5 percentage points for both males and females in the full sample (contains full-time workers and part-time workers). Such cognitive skills, which were not observed in conventional data, are important factors in explaining the wage gap between public and private sector workers. Any estimation without controlling for such cognitive skills may generate an upward bias in the size of the public sector wage premium.