To date the analysis on the wage penalty faced by over‐educated doctorate holders is surprisingly scarce. Being focused on college graduates, the existing literature has primarily tended to adopt a standard ordinary least squares regression technique that allows measuring the impact of over‐education on the conditional mean of the wage distribution. This paper aims to fill this gap, not only by assessing the extent to which the prevalence of over‐educated doctorate holders in science, technology, engineering, and mathematics (STEM) fields in South Korea is distributed across the wage distribution, but also by examining the extent to which their wage penalty relative to what they could have earned in a job corresponding to their level of education may have varied across the wage distribution. The main findings reveal that there is no evidence to support the notion that over‐education and ability are inversely related among STEM doctorate holders in Korea. Based on the empirical evidence, this article suggests that it is an oversimplification to characterise both the incidence and the wage effects of over‐education as merely reflecting lower‐ability levels, at least in the Korean context.