There is debate about whether the rate of return to education in developing countries declines with the level of schooling. This paper reports evidence from urban Papua New Guinea which shows that the average private rate of return to an additional year of education rises with the level of education considered. This pattern is robust to the specification of the extended earnings function and is strengthened when the measure of employee compensation includes fringe benefits. The results are not sensitive to controls for sample selectivity bias and the estimated selectivity effects are more consistent with the principle of comparative advantage than is existing evidence from developing countries. The returns to on-the-job training are also higher than indicated by previous evidence in developing countries.