Estimating individual rates of discount: a meta-analysis

Abstract

In this article, we present the results from a meta-analysis conducted over 44 experimental and field studies, which report individual discount rate estimates. We find in our research that the experimental design of a study has a decisive impact on these estimates, and conclude that meta-analysis, in the case of individual discount rates, has a low transfer value.

Acknowledgements

The authors wish to thank participants in ‘Timescapes and Retroscapes’ (Palermo) and ERSA (Volos) conferences for useful comments on earlier drafts of this article.

Notes

¹ For the use of meta-analysis in economics, see, amongst others, Florax et al. (2002) and Nijkamp (2005).

² Notice that most of the studies present several discount rate estimates. The complete list of articles used in this article can be obtained from the authors upon request.

³Florax et al. (2002) provide an excellent review of statistical procedures to test for publication bias.

⁴ The source for these variables is the World Development Indicators, a data set maintained and distributed by the World Bank. GDP is expressed as thousands of dollars in 1995 Purchasing Power Parity.

⁵ Recall that meta-analysis was developed in the context of medical science to facilitate the anamnesis when there is limited information on a patient's condition.