This paper proposes a simple decomposition of the widely used DRC measure of international competitiveness into three distinct elements: (1) changes in international prices, (2) changes in production techniques and (3) total factor productivity change. The decomposition provides a clear analytical link between two largely separate methodologies for assessing economic performance, cost-benefit indicators based on the world price rule and total factor productivity analysis. The methodology is applied to evaluate changes in measured comparative advantage of industries in Thailand for the period 1963-76. The results broadly indicate that changes in price competitiveness and in total factor productivity are the major sources of changes in DRC ratios for Thai industries.

THE X-inefficiency costs of protection and industrial concentration have been topics of considerable interest in the literature on economic development. Balassa (1975) and Bergsman (1974), for example, have argued that protection, by increasing X-inefficiency, generates a major welfare cost which is not captured by traditional costs of protection calculations, and White (1976) has examined the relationship between inappropriate factor intensities which he associates with Xinefficiency and industrial concentration. The common theme which emerges from each of these studies is that public policy can have a major impact on economic performance not only by influencing factor proportions, but also by influencing the intensity with which a non-measurable input, X-efficiency, is employed. Despite the intuitive appeal of the X-efficiency concept there have been very few empirical tests of this hypothesis. This partly reflects the imprecise nature of the basic concept but also the lack of adequate micro economic data. X-efficiency clearly relates to efficiency within firms so that the most appropriate tests would rely on data on firm-level efficiency. In this paper we propose a direct test of the X-efficiency hypothesis using firm-level data. We build on previous work by Corden (1970, 1974) and Martin (1978) who showed how to model X-efficiency effects using the concept of managerial leisure. We extend these simple models of managerial behavior, emphasizing the role of the external managerial labor market. The model is then used to explore the relationship between changes in public policy, managerial effort and X-inefficiency. Certain predictions of the model are tested using cross-section data from a survey of firms in two subsidized industries in Ghana. Efficiency indices are computed on the basis of a translog frontier production function. Variations in relative efficiency are then correlated with several explanatory variables including the presence or absence of subsidy payments to the firm. Subsidized firms in both industries are found to exhibit higher relative levels of X-inefficiency.