This paper examines convergence in environmental/carbon performance by constructing a measure based on production theory, where production processes explicitly result in the production of two outputs; a good output (GDP) and a bad output (CO2). We use the derived measure to test the beta-convergence hypothesis for a panel of 94 countries. The results reveal evidence in support of beta-convergence in environmental, or carbon performance for the entire (global) sample and each of the sub-samples. The evidence points to a slower convergence rate for the high-income countries relative to low-income countries. Moreover, the rate of convergence does not vary with capital in the global sample, but does vary in the high-income sample, possibly reflecting differences in abatement cost induced by differences in the stringency of environmental regulation and enforcement. Additionally, we find evidence of a negative relation between environmental performance and fossil fuel share, both at the global level as well as at the middle and high sub-samples, which tend to vary with capital intensity. As such, the results conform to the results from studies on the dynamics of per capita emissions.