1 Introduction
Governments facing scal pressure have increasingly turned to proposals to create or enhance consumer choice for public services (see, e.g., Besley and Ghatak 2003, Blochliger 2008, Hoxby 2003, Le Grand
2003). In health care, choice is a popular reform model adopted by administrations of dierent political orientations in many countries, including the US, the UK, Denmark, Italy (Lombardy), the Netherlands,
Germany and Sweden. The belief is that by increasing choice for patients, providers of care or insurers will become more responsive to patient demand, which in turn will drive greater eciency in the delivery and funding of health care. However, whether enhanced patient choice will make hospital choice more responsive to quality is not well established, although the consequences of poor quality in health care can be dire. Patients' health can be severely compromised by poor quality care, including, as we show below, an increased risk of death. Thus there is a need to understand the responses of health care consumers when they are oered more choice. This is exactly the issue we address here.
To do this we exploit a reform which introduced patient choice and tie this to the estimation of a structural demand model that explicitly incorporates the institutional features of the reform. This enables us to identify the eect of increasing choice on patient behavior. We use the model to quantify the gains from the reform in terms of patient welfare and survival and to analyze how the changes in patients' choices translate into changes in the competitive environment faced by hospitals.
The reform we exploit is from the English National Health Service (NHS). In 2006, the UK government mandated that patients in the English NHS had to be oered a choice of 5 hospitals when referred by their physician to a hospital for treatment. Prior to