Abstract
This article investigates different modes of public policy in health care and their impact on health care financing and health service provision. In order to investigate the relationship between health expenditure and health service provision, we construct an “index of health care providers”. The empirical analysis of expenditure and this index demonstrates that there is only a weak correspondence between the level of total health expenditure and the number of health service providers in OECD countries. Different modes of health policy can help to explain why some countries are more successful in translating monetary inputs into health care personnel than other countries. Our results indicate that policies which favor self-regulation by non-governmental actors (as in Germany) lead in general to high levels of health care providers at above OECD average health expenditure. Policies which favor direct state control (as in the United Kingdom), on the other hand, are characterized by lower levels of health care providers and below average health expenditure. Policies which favor market elements are more difficult to categorize. However, it is noteworthy that especially countries that give market mechanisms higher priority than other countries (as the United States) offer below average numbers of health care providers at comparatively high total health care costs.
Acknowledgements
The authors gratefully acknowledge the helpful comments and criticism by Ted Marmor and the other members of the Yale workshop on comparative health policy analysis in 2007 as well as by Carolyn Tuohy, David Wilsford, Bernhard Ebbinghaus, Bruno Palier, and Olli Kangas. The research reported here has received financial support from the German Research Foundation (DFG). Claus Wendt would also like to thank Harvard's Minda de Gunzburg Center for European Studies for the time and intellectual community provided to him when working on the final draft of this article as a John F. Kennedy Memorial Fellow.
Notes
1. Due to lack of data, Japan, Korea, Mexico, and Turkey are not included in the analysis. Iceland with a population of about 289,000 has also been excluded. Hence, our study covers 25 of the currently 30 OECD countries (OECD 25). However, reliable data are not provided by the OECD for all time points, thus the analysis is partly based on less than 25 countries.
2. For the debate on the perception of health care systems see Mossialos Citation1997, Gelissen Citation2002, Kohl and Wendt Citation2004, Marmor et al. Citation2006.
3. Due to double counting the figures on coverage by private insurance, Medicaid, Medicare, and on those without coverage do not add up to 100%.
4. “Public health financing” here comprises financing by general tax revenues as well as by social security contributions.
5. Alternatively one could analyze the effect of different modes of public policy and/or of health care financing on the numbers of each health care provider type available in OECD Health Data. In this case, however, it would be difficult to estimate the relation between health expenditure and the overall level of health employment.