Assumptions regarding the magnitude and direction of energy-related technological change have long been recognized as critical determinants of the outputs and policy conclusions derived from integrated assessment models. Particularly in the case of developing countries, however, empirical analysis of technological change has lagged behind simulation modeling. This paper presents estimates of sectoral productivity trends and energy-augmenting technological change for several energy-intensive industries in India and South Korea, and, for comparison, the United States. The key findings are substantial heterogeneity among both industries and countries, and a number of cases of declining energy efficiency. The results are subject to certain technical qualifications both in regards to the methodology and to the direct comparison to integrated assessment parameterizations. Nevertheless, they highlight the importance of closer attention to the empirical basis for common modeling assumptions.