Healthcare providers are motivated by a combination of financial and non-financial incentives. This thesis focuses on two specific forms of these incentives: pay-for-performance (P4P) and reputation. Despite increased use, there is limited evidence on how financial and reputational incentives interact, or on how financial incentives affect patients and professionals. We further our understanding with respect to P4P and make recommendations about the design of future schemes. We achieve the thesis aims by producing four empirical studies. Each empirical study uses data collected as part of a national P4P in the English National Health System, the Quality and Outcomes Framework. We begin with an investigation of the relevant importance of financial and reputational incentives in determining provider performance. We use administrative data for nine years of practice performance on a range of indicators totalling close to five million observations. This analysis covers a period during which the financial and reputational rewards were changing. We find that initially, financial incentives had a larger effect on performance. Over time, reputational incentives become more important. Our second study uses changes in the organisational structure of healthcare providers to explore whether the observed similarity in the performance of nearby practices can be explained by peer effects. We measure the performance of 8,000 individual practices and their peers for five years. When peer groups are merged, there is a reduction in peer effects for old peers and an increase in peer effects for new peers. Practices seem to be pulled down by the presence of poor peers in their group. In our third study we measure the impact of variations in the proportion of income at risk to P4P on the working lives of GPs. We combine administrative data with survey data from before and after the introduction of P4P. Our sample consists of approximately 2,000 GPs who provide detailed information about their working lives. We find that providers are unaffected by these variations despite income at risk being high. Finally, to observe the relationship between quality of care reported at the patient level and at the practice level we link practice performance with a detailed survey of the English population aged over 50 years. Correlations are generally smaller than expected and negative for some areas. Practices may have lacked an incentive to communicate their care adequately to patients and may have diverted attention away from areas of care without financial incentives. Non-financial incentives can be effective motivators when peer performance is observable. Professionals are also unlikely to associate negatively with income being related to their performance, along as incomes remain high. However, patients may suffer from a lack of communication about the type of care they are receiving.