Economic discrimination in a two-good model

Conclusions The main conclusion of this study is that in the general case we cannot make any predictions regarding the effects of discrimination on the welfare of whites or blacks. To reach definite results special assumptions have to be made, and even then much depends on in terms of which good the effects are measured.Joseph E. Stiglitz has criticized the trade approach to discrimination on the grounds that once the step is taken from the simple Beckerian one-good model to the full-fledged two-by-two-by-two framework used in international trade theory this approach can no longer explainwage differences, since the factor equalization theorem will apply, at least as long as the two groups are not totally specialized in production⁸. This criticism can be extended in the light of the results of the present work. We have in the foregoing seen how even in a model that assumes total specialization in production we cannot reach any clear-cut conclusions regarding under which circumstances it pays for the majority group to discriminate against the minority. Maximization of majority welfare by means of economic discrimination simply requires too much information, at least when the trade approach is used. Other theories are likely to yield better explanations regarding the causes of discrimination, and future research is likely to become more fruitful if attention is concentrated more on alternative approaches.The research was financially supported by the Swedish Council for Social Science Research (Statens råd för samhällsforskning). Thanks are due to an anonymous referee for thorough and helpful comments on an earlier version of this paper.