Are There Gender Differences in Demand for Index-Based Livestock Insurance?

ABSTRACT

Risk management plays a role in avoiding and escaping chronic poverty throughout the world, particularly for women, who are disproportionately negatively affected by shocks. Using three years of household survey data, administrative records and qualitative interviews, this paper examines the relationship between gender and demand for index-based livestock insurance (IBLI) among pastoralists in southern Ethiopia. IBLI appears to be equitably accessed by men and women alike and we find limited evidence of gender-differentiated demand for IBLI. We also find only modest differences associated with age and share of income from livestock.

Acknowledgements

This research uses data collected by a collaborative project of the International Livestock Research Institute, Cornell University, Syracuse University and the BASIS Research Program at the University of California at Davis. We wish to specifically thank Munenobu Ikegami, Birhanu Taddesse, Wako Gobu, Nasibo Bonaya and Anne Gesare for their efforts to collect useful and accurate data. We also thank Andrew Mude, Mark Constas, Nathan Jensen, Richard Palmer-Jones, Megan Sheahan, two anonymous reviewers and Cornell seminar participants for valuable feedback throughout the process.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplemental Material

An Online Appendix is available for this article which can be accessed through the online version of this journal available at http://dx.doi.org/10.1080/00220388.2016.1214717

Notes

1. NDVI images used in the construction of the IBLI contract has resolution of 8 km² and is taken every 10 days from a United States National Oceanic and Atmospheric Association satellite used largely for weather forecasting (Chantarat et al., 2012).

2. Five relevant variables cannot be imputed for the periods between survey rounds: Expected rangeland conditions, assets, savings, IBLI knowledge and financial literacy. The outcome variables as well as key determinants of demand (price, herd size, previous losses, income and so forth) vary across the interim periods. Estimation using only the rounds for which we can measure these five variables is underpowered and infeasible. Rainfall data for Figure 1 from Lasage, Seifu, Hoogland, and De Vries (2010).

3. See Online Appendix A for further details on panel construction. See Online Appendix B for complete attrition analysis.

4. TLU, or tropical livestock units, are calculated based on metabolic weight. 1 TLU = 1 cattle = 0.7 camel = 10 sheep/goats. Cattle are the dominant species in Borana, 77 per cent of the animals in our sample. Women are more likely to own goats than men, but once herd size, income, savings and marital status are accounted for there is no statistically significant gender difference in herd composition.

5. The only exception was for 20 participants in a separate, annual herd migration survey, who had a 50 per cent chance of receiving a 100 per cent voucher for IBLI purchase up to 15 TLU in a given sales period.

6. Complete qualitative methodology available upon request.

7. Results using reported purchase only and the 87 per cent of cases where reported purchase matched OIC records can be found in the Online Appendix.

8. Four respondents had no household members participating in groups.

9. According to the survey data, the most important information sources for both male- and female-headed households were community meetings and NGOs, followed by the insurance company and informal conversations with friends and family. Qualitative data contradict this. All respondents who attended community meetings where IBLI was discussed reported not effectively learning about the IBLI product at community meetings. No one reported community meetings as a preferred channel, though for many people they were the only product-focused channel, which may explain why this was chosen as ‘most important’ in the survey data. No one indicated that they learned about the IBLI product from informal conversations with friends and family. The category ‘NGO’ meant different things to different people, including ILRI, OIC or anyone who comes to the community in a car.

10. Estimates using alternative formulations of the IBLI information source variable yield the same (lack of) gender differentiation along this dimension of demand.

11. The effective price of IBLI per TLU of coverage, accounts for discount coupons received in addition to spatial and temporal price variations. However, IBLI is priced by species, not TLU. Therefore, the price facing each individual depends on the animals they choose to insure. For simplicity, we have calculated the effective TLU price as the price of insuring one cow rather than using the actual prices paid for the diverse combinations of animals individuals chose to insure. The latter method makes it difficult to define a price for those who chose not to purchase IBLI.

12. Non-livestock assets are compiled using principal components analysis using a host of durable goods (Online Appendix A). The choice of assets is challenging in a context where people own little other than livestock. Our results, however, are robust to inclusion or exclusion of the asset index, underscoring their relative unimportance in this context.

13. First-stage regressions of IBLI knowledge on assigned cartoon and assigned tape for the time period used in this study yield p-values ranging from 0.05–0.88, depending on the specification. In their study of IBLI demand using the same data, but only the first two sales periods, Takahashi et al. (2016) instrument for IBLI knowledge using the educational treatments. Their first-stage regressions suggest that the instrument is weak in the first purchase period, but acceptable in the second. Given that these interventions were only used in the first year of IBLI implementation, it makes sense that their influence on knowledge wanes over time and is not visible in the third and fourth purchase periods.

14. The first indemnity payments under the IBLI contracts were made in November 2014, outside the time span of the data analysed here.

15. These results are consistent across all specifications, including those using reported IBLI purchase rather than OIC record of purchase. Those results can be found in Online Appendix C.

16. Empirical tests of intra-household bargaining following Thomas (1990) reveal non-identical preferences across several household expenditure classes (sugar, tobacco, cooking fuel, and education) using lactating herd and HS animals. These results are available upon request.

Additional information

Funding

This project was made possible by the generous funding of the UK Department for International Development through FSD Trust Grant SWD/Weather/43/2009, the United States Agency for International Development grant No: EDH-A-00-06-0003-00 and the World Bank’s Trust Fund for Environmentally and Socially Sustainable Development Grant No: 7156906. This paper represents the views of the authors alone and not the positions of any supporting organisations. Any remaining errors are the authors’ sole responsibility. Data and code for replication are available upon request.