Effects Of Social Networks On Household Food Consumption Among Smallholder Farmers In Kisii And Nyamira Counties, Kenya

Abstract

Malnutrition is a major burden worldwide. One strategy for curbing malnutrition is through food and nutrition policies that promote both sufficient calorie intake and dietary diversity. However, effects of shocks and lack of nutrition knowledge have led to consumption of insufficient quantity and poor quality diets respectively. In absence of formal institutions to offer nutrition education and insurance against shocks, social networks act as an informal insurance and a channel for individuals to obtain new nutrition information. Literature has mixed findings on the insurance role of social networks, while information on the role of social network in improving dietary diversity is scarce. This study evaluated effects of idiosyncratic shocks on social network formation; the effects of social networks in insuring household food consumption against idiosyncratic shocks; factors influencing formation of nutrition information network and effects of such networks on household dietary diversity. The study was undertaken in Kisii and Nyamira counties of Kenya and data was collected using semi-structured questionnaires. A two-stage sampling procedure was used to identify the sample, farmer groups were first sampled and then households were sampled from the selected farmer groups. Two rounds of data were collected, the first round between October and December 2015, and the second was round between October and December 2016. A total of 824 and 745 farmers were interviewed in the first and second rounds respectively. A dyadic linear probability model was used to evaluate the effects of shocks on the formation of financial and non-financial networks as informal risk sharing strategies. Findings showed that kinship, geographical proximity, education and age are important determinants of both financial and non-financial links, while health shocks influences formation of financial links. xii The effects of health shocks on food consumption and also the effects of social networks in insuring food consumption against health shocks were estimated using a fixed effects model. The finding showed that health shocks have a negative and significant effect on purchased food but not on total food consumption. The results further indicate that financial credit networks have positive and significant effect in insuring food consumption against health shocks. Lastly, factors influencing nutrition information link formation, and the effects of such networks on household dietary diversity in Kenya were estimated using Probit and Poisson regression models respectively. The results indicate that education level and the number of neighbours positively affect the probability of having a nutrition information network. Further, the average household dietary diversity of an individual’s network members has a positive effect on the dietary diversity of the individual. Furthermore, education level of an individual’s network members, household size, wealth status and farm size also have a positive effect on household dietary diversity. The study concluded that social networks play a risk sharing role and insure food consumption against idiosyncratic shocks, though the social networks don’t provide full risk sharing against health shocks. Therefore, policy makers would make policies, for the public sector to create safety nets that complement the ones formed by the smallholders, to enhance resilience to shocks sharing. Additionally, there is need to implement nutrition education programs targeting to improve the quality of diets. Such programs can use social networks as pathways through which nutrition information flows and would also widen nutrition information networks among smallholder farmers. Most importantly, the programs can take advantage of the social multiplier effect generated by the endogenous network effects.