Fairness and Efficiency in Smallholder Farming: The Relation with Intrahousehold Decision-Making

Abstract

Agricultural households face collective action dilemmas when making decisions about investments in their common household farm and the allocation of resources and benefits derived from it. We relate intrahousehold decisions, as measured in a lab-in-the-field experiment conducted with spouses in agricultural households in western Uganda, with actual investments and intrahousehold resource allocation. Intrahousehold decision-making that supports cooperation and equitable sharing is associated with greater investment in the intensification of cash and food crop production, and more equitable access and control over income. Freeriding behaviour by husbands is associated with the intensification of cash crop production, but not with equitable sharing.

Acknowledgements

We are grateful to Cheryl Doss for her comments on an earlier version and to Nathalie Holvoet, Piet van Asten, Jennifer Twyman, Gyde Feddersen, Ghislaine Bongers, Marijke Verpoorten, Kristof Titeca, Bjorn Van Campenhout as well as participants of the 24th International Association for Feminist Economics Annual Conference, the CGIAR-Uganda Research Seminar at IITA/IFPRI Kampala, the DEV Seminar of the School of International Development at the University of East Anglia and the Institute of Development Policy and Management research seminar for useful comments on earlier drafts and presentations. We would like to thank the anonymous reviewers of the article for valuable suggestions. We acknowledge the CGIAR Fund Council, Australia (ACIAR), Irish Aid, European Union, International Fund for Agricultural Development (IFAD), Netherlands, New Zealand, Switzerland, UK, USAID and Thailand for funding to the CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS) which facilitated funding this research through the Uganda office of the International Institute of Tropical Agriculture (IITA). We would like to thank the Hanns R. Neumann Stiftung for contributions to funding the implementation of this research. We are thankful for an excellent and fruitful collaboration with IITA and with the Hanns R. Neumann Stiftung, especially Ghislaine Bongers, Fortunate Paska and Doreen Mbabazi. We would like to thank the field researchers for their dedication to excellent data collection and the farmers in Kasese district who enthusiastically participated in this research. The research tools and data are available on request. The experiment script is available in the Supplementary Materials 4.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Sustainable is understood as not compromising future use and with the potential to adapt to climate changes.

2. Pareto or allocative efficiency occurs when a distribution of resources cannot be improved for anyone without making anyone else worse off.

3. The experiment script is available in the Supplementary Materials 4. Data and questionnaire are available on request.

4. We did not reveal the exact amount but informed participants that each box contained between 0 and 4000 UGX. The exchange rate at that time was 3250 UGX to the Euro.

5. Illiterate participants were individually assisted.

6. Given that participants are informed that remainders become savings, the proportion of the total household income consumed by husband and wife in the second stage may depend on their ambiguity aversion, risk and time preferences.

7. We use the teffects ipw command in STATA13 including the types of couples as a multivalued treatment (see chapter 21.6.2 in Wooldridge, 2010).

8. The POM for ‘Other’ couples is calculated based on the estimates for the other types of couples.

9. We use logistic regression for binary outcome indicators, Poisson regression for counts and Ordinary Least Squares for continuous outcome indicators.

10. We refrained from using Augmented IPW (AIPW). First, Glynn and Quinn (2010) showed that in the case of non-linearity of the outcome mean function, possible confounding and relatively small sample sizes, which our analysis is confronted with, IPW may produce less biased and more precise results than AIPW (and regression analysis). Second, the benefits of IPW (versus regression analysis) of saving degrees of freedom by controlling for multiple covariates captured in one score would be lost with the inclusion of covariates in the outcome model in AIPW.

11. See the  Figures 1-10 in Supplementary Materials 3 for an assessment of overlap and balance of the matching procedures.

12. The (non-confirmed) perception that shade trees create favourable conditions for pests and fungi in coffee plantations has caused many farmers to uproot them.

13. We limited to couples in which spouses agreed on the most important cash crop, which is coffee in 98 per cent of the cases.

14. We defined control over income as making the (majority of) decisions on who can spend the money and on what the income will be spent on.

15. The presidential elections, a commonly remembered event about four years prior to data collection, were used as a reference point.