The literature on the GATS and trade in health services is limited, largely speculative, and polarised between ‘trade’ and ‘health systems’ perspectives. The distinctive features of the market for health services, particularly in developing countries, means that the ‘benefits of trade’ argument is not applicable: the potential economic benefits of trade in health services are limited, while its potential effects on health systems, particularly through the supply of health professionals to public services, are substantially negative in almost all developing countries. GATS commitments have not encouraged trade, and there is little reason to anticipate this, except possibly in Mode 3; but they seriously constrain health policy. The effectiveness of GATS safeguard provisions is at best doubtful; and the stronger domestic regulation needed in open health service markets is beyond the capacity of most developing countries. Proposed GATS-consistent regulations to offset adverse effects of trade in health services are untested, unrealistic, impracticable, ineffectual, and/or have serious limitations or adverse side-effects. GATS commitments are binding, backed by sanctions, and of indefinite duration; but key provisions are ambiguous, and the future extent, nature and effects of trade in health services very uncertain. Analysis and data are very limited and analytical frameworks undeveloped. There is thus no basis for informed decision-making on GATS commitments; and decision-making processes mean that health considerations play a very limited role. Therefore, the GATS framework is inappropriate as a basis for international rules on trade in health services; and it is undesirable for developing countries to make GATS commitments in the health sector.