Social Networks and Factor Markets: Panel Data Evidence from Ethiopia

Abstract

We investigate the role of an indigenous social network in Ethiopia, the iddir, in facilitating factor market transactions among smallholder farmers. We use detailed longitudinal household survey data and employ fixed effects estimation approaches to identify the effect of iddir membership on factor market transactions among farmers. We find that joining an iddir network improves households’ access to land, labour and credit transactions. Our findings also hint that iddir networks may crowd-out borrowing from local moneylenders (locally referred as ‘Arata Abedari’), a relatively expensive credit source. These results suggest that non-market institutions can play crucial roles in facilitating market transactions.

Acknowledgements

Two anonymous referees provided useful comments that substantially improved the paper. This paper also benefited from comments by Henning Tarp Jensen, Erwin Bulte, Adriaan Soetevent, Jonas Nordström and seminar participants at the IFRO seminar at the University of Copenhagen. The data used for this study come from a large longitudinal survey which benefited from funding by the World Bank Group (Award 100025484/2010), the Department for International Development (DFID), the US Agency for International Development (USAID) and the International Food Policy Research Institute (IFPRI). All remaining errors are ours.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Previous studies point to the role of social networks in risk and consumption smoothing (Ali & Deininger, 2014; Ali, Deininger, & Duponchel, 2014; Fafchamps & Lund, 2003; Hoddinott et al., 2005; Kinnan & Townsend, 2012; Okten & Osili, 2004; Udry, 1994; Wydick, Hayes, & Kempf, 2011); credit, saving and transaction costs (Dercon et al., 2006; De Weerdt & Dercon, 2006); and technology adoption (Bandiera & Rasul, 2006; Conley & Udry, 2010; Fafchamps & Minten, 2002; Foster & Rosenzweig, 1995; Krishnan & Sciubba, 2009).

2. However, richer households could obtain better coverage against risk by joining multiple iddir networks, and perhaps by joining iddir networks established in rich neighbourhoods. As suggested by Hoddinott et al. (2005), the income and wealth status of households could affect the intensity of participation in iddirs, but not the extensive margin of participation in these egalitarian networks.

3. Sharecropping is a tenancy agreement between landowners and their tenants. Sharecropping depends on the premise that tenants share a portion of the harvested production with the landowner depending on their agreement, usually half or two-third of gross production (see, Pender & Fafchamps, 2005). In some cases, landowners contribute some production inputs, generally draft-animal (oxen) or labour. In contrast, in fixed land rentals, the tenant pays a fixed amount of money, commonly in advance and assumes ownership of the land and the harvested production for the agreed production season.

4. For example, if a household’s crops are not ready for harvest, the household continues to credit labour to other households who are in demand for it and gets the labour back when its crops are ready for harvest.

5. These four major regions are Tigray, Amhara, Oromia, and Southern Nations, Nationalities, and Peoples (SNNP).

6. We exclude households who are beneficiaries of the direct support part of the PSNP programme in Ethiopia, those households without adequate labour force and hence cannot contribute to public works of the PSNP programme.

7. Some previous studies that focus on specific regions where iddir networks are particularly more prevalent report higher iddir participation than are seen in our sample (Dercon et al., 2006; Hoddinott et al., 2005). The lower share in our data is due to the inclusion of regions that have a lower prevalence of iddirs.

8. An exception to this could be observed in those iddir networks which are organised along professional, ethnic and religious lines. These iddir networks, although not common, are expected to have some entry restrictions.

9. Equib is a form of ‘rotating credit and saving association’ (ROSCA) in Ethiopia. Although both equib and iddir are social networks that operate through powerful social pressures, equib mainly functions as a financial intermediary, rather than as an inclusive social network of broader purpose.

10. We categorise shocks into idiosyncratic and covariate shocks considering whether these events affect specific households or communities living in a similar area. Idiosyncratic shocks include death and illness of family members as well as other similar events that specifically affect a specific household. Covariate shocks are those spatially covariant, naturally bad events whose effects go beyond a specific household and include drought, flood, pests, crop diseases and others.

11. As discussed in Krishnan and Sciubba (2009), while there are different types of labour-sharing practices in Ethiopia such as ‘debo’ that involve larger-scale borrowing of labour from a large number of households which may easily lead to iddir formation, we focus on a specific type of labour-sharing practice that commonly involves symmetric reciprocation of labour among parties involved in the network, commonly two or three households reciprocating labour with each other.

12. Although some iddir networks provide soft loans to their members, this accounts for less than 1 per cent in our data. Thus, our focus is restricted to the indirect role of iddir networks in facilitating credit access from neighbours and friends.

13. A similar table in the Appendix (Table A2) compares households’ factor transactions between those households who recently joined iddir networks and those remaining non-members in both surveys.

14. Not surprisingly, the treatment effects from the linear regression models are very comparable with the implied marginal effects from the probit models. For this reason, the latter estimates are not reported but available from the authors upon request.

15. One possible question here is whether the three factor markets are interlinked so that one market (for example, credit market) is deriving the other market (for example, land market) as discussed in Ray (1998). While these markets may be interlinked, simple cross-correlations among the three factor markets do not seem to be significant in our data.

16. This is particularly appealing given the fact that formal credit markets are commonly thought to be ineffective in crowding-out informal moneylenders in rural areas (Hoff & Stiglitz, 1990; Udry, 1990).

17. For example, Krishnan and Sciubba (2009) emphasise that the impact of social networks on economic performance heavily depends on the size and type of the network.