Inequality Decomposition and Human Development

Abstract

The human development index (HDI), which takes into account achievements in health, education, and income, is considered a good measure of the social attainments of a country. The global cross-country distribution of human development is imbalanced and the degree of convergence is low. This inequality has varied during recent years. In this paper we present evidence that improvements in the convergence of human development across countries are mostly attributed to education, whereas health and income have made poor contributions. To do this we exploit the multiplicative structure of the HDI and several decompositions of the Theil inequality index.

Keywords:

• World
• Human development
• Inequality
• Inequality decomposition
• Theil index

Jel Classification:

• I30
• D63

Acknowledgements

I thank the two anonymous reviewers whose comments have improved the quality of the paper. I also thank Lorena Fernández for her assistance during the revision of this manuscript. This work was supported by Ministerio de Ciencia under Grants ECO2011-29355 and ECO2014-5376-P; Network MOMA under Grant ECO2014-57673-REDT, Junta de Andalucía under Grants SEJ5980 and SEJ4941; and Grupo PAIDI under Grant SEJ426.

Disclosure Statement

No potential conflict of interest was reported by the author.

About the Author

Ricardo Martínez holds a Ph.D. in Economics (University of Alicante) and is Associate Professor at the University of Leon. His research focuses on welfare economics, equity, social choice and decision-making processes.

Notes

1. Despite improvements made in the new version of the HDI, some authors have shown that a few drawbacks still persist (Herrero, Martínez, and Villar 2010, 2012; Chakravarty 2011; Klugman, Rodríguez, and Choi 2011; Ravallion 2011, 2012).

2. These indicators are normalized to fit in the interval [0,1] for comparability.

3. Alcalde-Unzu, Ezcurra, and Pascual (2009) use a similar technique to study cross-country disparities in health-care expenditure.

4. Readers are referred to UNDP (2013) for a more detailed discussion on the construction of the HDI.

5. We thank one of the anonymous referees for suggesting this alternative version of the decomposition.

6. In our sample, countries are assigned to development levels as in the Human Development Report 2013.

7. The UN Development Program provides retrospective data for the study period according to the new criteria, variables, and methodologies for defining the revised version of the HDI. We also use the same lower and upper bounds in computing the partial indices for health (), education (), and income () so that they are comparable over time.

8. We keep the group composition constant for the whole study period of 1980–2012.