Reforms giving users of public services choice of provider aim to improve quality. But such reforms will work only if quality affects choice of provider. We test this crucial pre-requisite in the English healthcare market by examining the choice of 3.4 million individuals of family doctor. Family doctor practices provide primary care and control access to non-emergency hospital care, the quality of their clinical care is measured and published, and care is free. In this setting clinical quality should affect choice. We find that a one standard deviation increase in clinical quality would increase practice size by around 17%.