Four Empirical Essays on the Market for General Practitioners' Services

This dissertation consists of empirical essays within the subject of health economics. There are four essays in applied micro-econometrics and, as data in Essays 2 and 4 have a panel format, econometric methods for panel data are applied. Tobit-type models for limited dependent variables are applied in Essays 3 and 4, Nested logit models for choice of general practitioner (GP) are applied in Essay 1 and in Essay 2 modeling and estimation procedures involving latent structural variables are applied. The market for GPs' services is the recurrent theme, and a common feature of the empirical modeling and estimation conducted in the essays of this dissertation is that latent variables play an important role. There are several motivations for studying the market for GPs. GPs play a key role and constitute the cornerstone of the health care sector in Norway and other countries with a national health service. The GP is often a patient's first encounter with the health care sector. In Norway the GP also acts as a “gatekeeper”, and a referral from the GP is necessary in order to receive specialized care. Further, decisions made by GPs have a large impact on public spending such as sick-leave benefits and drug reimbursements. Due to this pivotal role of general practice, any research providing policy guidance for the sector will potentially have noticeable welfare effects. Studying this market may also provide advances in terms of enhanced understanding of economic behavior in general, and within the discipline of health economics in particular. Arrow's (1963) article describing various imperfections in the market for medical care is often considered to mark the founding of health economics (Culyer and Newhouse, 2000). Many of the peculiarities Arrow describes are, indeed, present in the market for GPs' services. Asymmetric information limits consumer sovereignty and creates challenges for designing appropriate contracts and payment mechanisms. Quality of services is diffcult to observe and quantify, and optimal consumption of health care services is diffcult to achieve. While the general research questions in health economics are relevant in the specific context of general practice, the conditions for knowledge accumulation seem favorable as well. The availability of detailed and disaggregated panel data enables identification and quantification of the mechanisms in focus by applying a large variety of modeling and estimation methods. In addition, one may argue that economic theory is well suited to model individual behavior and that assumptions of rational decision makers are more realistic when describing individuals such as GPs than when describing institutions such as hospitals. Fuchs (2000) describes two related missions of health economists: providing valuable input into health policy and enhancing understanding of economic behavior. Both missions motivate research on the market for GPs. Essays 1 and 2 in this dissertation focus mainly on factors influencing the patients' or consumers' decision to seek the services of a particular GP, while Essays 3 and 4 focus on factors influencing the GP's supply decision. The rest of the text proceeds as follows: The next subsection provides a brief introduction to the topic of this dissertation. Summaries of the four essays are given in subsection 1.3. Policy implications of the findings are presented in subsection 1.4. Limitations and ideas for future research are discussed in subsection 1.5. Complete versions of Essays 1-4 follow in sections 2-5.