Previous studies of the association between pollution and income have tended to analyse cross-sectional or panel data for a sample of developing and developed countries. This paper presents an analysis for a single country, Malaysia. This south-east Asian country has more, and probably better, data on environmental quality than perhaps any other developing country. I find that pollution–income relationships from the cross-country studies fail to predict accurately trends in air and water pollution in Malaysia. In particular, none of six pollution-income relationships estimated using a panel data set for Malaysian states has the hypothesized 'environmental Kuznets curve' form. Although these results are inconsistent with the predictions of the cross-country relationships, they make sense in the Malaysian context. Perhaps most important, their interpretation confirms the importance of policy decisions in determining environmental outcomes.