Assessing health efficiency across countries with a two-step and bootstrap analysis

Abstract

Regressing Data Envelopment Analysis (DEA) output efficiency scores on nondiscretionary variables, with a two-stage DEA/Tobit and bootstrap procedures, we show that health inefficiency in Organization for Economic Co-operation and Development (OECD) countries is related to Gross Domestic Product (GDP) per head, education level, obesity and smoking habits.

Acknowledgements

The authors are grateful to I. Joumard, K. Wendorff and the participants at workshops at the Zentrum für Europäische Wirtschaftsforschung (ZEW) and at the German Ministry of Finance for helpful comments. The opinions expressed herein are those of the authors and do not necessarily reflect those of the authors' employers.

Notes

¹See Simar and Wilson (2007).

²See Farrell (1957) and Coelli et al. (2005).

³See Ruggiero (2004).

⁴Afonso and St. Aubyn (2006) explain the algorithms (available on request), implemented in Matlab.

⁵Data sources are detailed in Afonso and St. Aubyn (2006).

⁶See Afonso and St. Aubyn (2006).

⁷Afonso and St. Aubyn (2005) addressed health efficiency for 2000 without PCA. Countries labelled as efficient were in line with our results.

⁸Educational level, percentage of population that achieved tertiary education in 2000–2003; GDP per capita, PPP USD in 2003; obesity, percentage of obese population in 2002; smoking, percentage of population that consumed tobacco in 2003.