Environmental offsets have been proposed as a technique for managing the environmental impacts of new developments in regions that are not in compliance with environmental standards. By requiring developers to 'offset' any impacts by purchasing 'environmental credits', environmental quality can be maintained or even improved. Environmental offsets have a lot of intuitive appeal, and are being used widely in the USA, Australia and other countries. However there is at present no robust theoretical framework for analyzing the use of offsets, which has led to some of the weaknesses of existing programs and criticisms against the use of offsets. We present an economic model for designing offset programs that is based on identifying and valuing environmental service flows. We also discuss a number of factors that influence the effectiveness efficiency of offset programs including fungibility, effects on incentives of landholders and uncertainty and make recommendations about how to respond to these factors based on our model. The distributional effects of offsets are also explored and it is noted that offsets are not distributionally neutral. We argue on the basis of distributional effects that it is not appropriate to use offsets alone to seek to improve environmental quality. Furthermore, we recommend the combining of offsets with other market-based instruments such as Pigouvian taxes or cap and trade systems in order to reduce the negative distributional effects of offset programs, provide greater scope for achieving environmental improvements and to increase the probability of achieving first best (socially optimal) solutions.