incentives for R&D into diseases such as mnalaria. The proposed policy, she con-tends, would have the effect of lowering the prices in developing countries of phar-maceuticals for rich country diseases while also maintaining incentives for researchand development for therapies for poor country diseases.Anne Case analyzes the links between income and health. It is generally assumedthat there is a strong relationship between the two throughout the world. But thedirection of causality is an important question for those involved with health policy.Research on this issue has been limited by the lack of data on the health status ofindividuals, as most surveys tend to be based on the household. In addition, to studythe causal effect of income on health, one needs to identify a source of income thatis not determined by an individual's health status.Case resolves the causality issue by considering the role of the state old age pen-sion, a function only of age, in South Africa, using data from an integrated surveyof individual health and economic well-being. Her empirical analysis shows thatincome has a causal effect on health status through several channels, including nutri-tional status, sanitation and living standards, and reduction of psychosocial stress.Case concludes that governments interested in improving health status may find theprovision of cash benefits to be one of the rnost effective policy tools.Tomas Philipson and Rodrigo Soares note the common concern that while somecountries are getting richer, poor developing countries are falling behind. Per capitagrowth rates over the period 1962-97 seem to confirm that poor countries are notcatching up in income. But in other dimensions of welfare, such as life expectancy,different results emerge. Countries beginning the period with low life expectancytended to gain more in longevity than those starting with high life expectancy. Thuslooking only at income to draw conclusions about changes in overall economic wel-fare across nations may be misleading.The United Nations Development Programme has sought to account for nonma-terial dimensions of welfare through the Human Development Index, whichincludes education and longevity as measures of human welfare. Philipson andSoares argue that this index has limitations because of the "arbitrary way" in whichit is constructed. They propose an alternative, an income-equivalent measure of wel-fare, based on the value the population attaches to gains in different dimensions ofdevelopment. They argue that their proposed measure is less arbitrary than theHuman Development Index.
6 Boris Pleskovic, Nicholas Stern, and E Desmond McCarthyAs in previous years, the planning and organization of the 2001 conference was ajoint effort. Special thanks are due to Francois J. Bourguignon for very useful sug-gestions and advice. We also wish to thank several anonymous reviewers for theirassistance and conference coordinators Julee Allen, Mantejwinder K. Jandu, andJean Gray Ponchamni, whose outstanding