THE RELATIONSHIP between economic growth and environmental quality has been a source of great controversy for a very long time. At one extreme has been the view that greater economic activity inevitably leads to environmental degradation and ultimately to possible economic and ecological collapse. At the other extreme is the view that those environmental problems worth solving will be addressed more or less automatically as a consequence of economic growth. The longevity and passion of this debate has, in part, been a reflection of the lack of substantial empirical evidence on how environmental quality changes at different income levels. Compilation of such evidence has been constrained by the absence of data for a large number of countries. While data remain a problem, the situation is much improved and this paper takes a first step at systematic analysis of what data are available (see Appendix I for details). A number of caveats are in order. The data on environmental quality are patchy at best, but are likely to improve over time with better monitoring. Comparability across countries is affected by definitional differences and by inaccuracies and unrepresentative measurement sites. At this stage of knowledge, this paper has the modest objective of opening up the empirical debate using a relatively simple modeling technique applied on a consistent basis to a large number of environmental quality indicators and countries. The relationship between income and the costs and benefits associated with any given level of environmental quality is complex because it operates through a number of different channels, such as preferences, technology, and economic structure. The types of environmental degradation that occur depend on the composition of output, which changes with income. Some income levels are often associated with increases in certain polluting activities such as the development of heavy industry whereas economies with large service sectors may generate less pollution. There is a view that rising incomes imply that the cost of environmental degradation is greater because the wages used to value the opportunity cost of illness or work days lost are higher. This would imply increases in marginal benefits as incomes rise. But the poor are often the most exposed and vulnerable to the health and productivity losses associated with a degraded environment. There are some environmental problems where thresholds like survival are at stake. Here, the willingness to pay to avert damage is close to infinity and the level of per capita income only affects the capacity, not the willingness, to pay. With other environmental issues, most of