22 pages, Commercialisation of smallholder agriculture is important for rural economic growth. While previous studies have analysed effects of commercialisation on productivity and income, implications for farm household nutrition have received much less attention. We evaluate the effects of commercialisation on household food security and dietary quality with a special focus on calorie and micronutrient consumption. We also examine transmission channels by looking at the role of income, gender, and possible substitution effects between the consumption of own-produced and purchased foods. The analysis uses survey data from farm households in Kenya and a control function approach. Generalised propensity scores are employed to estimate continuous treatment effects. Commercialisation significantly improves food security and dietary quality in terms of calorie, zinc and iron consumption. For vitamin A, effects are insignificant. Commercialisation contributes to higher incomes and increased nutrients from purchased foods, but it does not reduce the consumption of nutrients from own-produced foods. Enhancing market access is important not only for rural economic growth, but also for making smallholder agriculture more nutrition-sensitive.
15 pages, Generalized trust, which refers to trust towards people that are not well known (Yamagishi and Yamagishi, 1994; Stolle, 2002; Uslaner, 2002; Freitag and Traunmüller, 2009), is needed for many situations of economic interaction such as daily market activities. Considering that trust acts as a lubricant for social interaction, a lack of generalized trust can severely restrict a persons reach of efficient economic exchange. Increases of generalized trust within a society thus have the potential to create large efficiency gains (Fafchamps and Minten, 2002; Fafchamps, 2006).
Given the importance of trust for social interaction and various welfare dimensions, a growing body of economic, sociological, and psychological research has been devoted to examining the circumstances under which trust can thrive. A reoccurring notion in all three disciplines is that communication represents a key factor in the formation of trust (Lewicki et al., 2006; Glanville and Paxton, 2007). Ostrom et al. (1992), for example, find that communication and sanctioning in a common pool resource experiment lead to substantially more efficient outcomes. In a laboratory setting, personal communication has shown to enhance trust (Buchan et al., 2006), and is even more powerful in creating mutually benefitting exchanges than the possibility to engage in non-binding contracts (Ben-Ner and Putterman, 2009).
In this paper, we analyze whether mobile phones which constitute a fundamental component of modern information and communications technologies (ICT) can help build social trust among pastoral communities in Northern Kenya.1 In most African countries, trust levels are remarkably low; out of all regions in the world, people living in sub-Sahara Africa exhibit the lowest levels of generalized trust (Mattes and Moreno, 2018). In the study region of Northern Kenya, it is particularly relevant to increase trust for several reasons. First, the relatively weak legal system jeopardizes contract enforcement, which means that any economic interaction requires substantial amounts of trust between the contract partners. This has caused a strong reliance on trust-based relationships in Northern Kenyas livestock sector (Mahmoud, 2008; Pavanello, 2010; Roba et al., 2018). Furthermore, low trust levels between ethnic tribes have also reinforced longstanding intertribal conflicts in the region, and impede solutions to share resources peacefully and effectively (Schilling et al., 2012). Lastly, the low population density and long physical distances between settlements in the region make communication over long distances difficult and therefore induce high monitoring costs. Potential benefits of enhancing trust are therefore particularly high in the context of Northern Kenya.
To compensate for physical remoteness, rural communities have a high need for digital connectivity but have oftentimes suffered from poor connection and inclusion in existing networks in the past (Salemink et al., 2017). Over the last decade, however, mobile phones have become available to most pastoralists in Northern Kenya (Butt, 2015; Asaka and Smucker, 2016; Parlasca et al., 2020). A large and growing body of research has pointed out that mobile phones can help increase several paramount welfare dimensions of rural populations in sub-Saharan Africa, such as income and income equality, financial development, gender equality, or institutional quality (Aker and Mbiti, 2010; Asongu, 2015; Asongu and Nwachukwu, 2016; Rotondi et al., 2020). However, the research on the implications of mobile phones on social capital formation is much less extensive. To the best of our knowledge, the potential of mobile phones to affect trust is so far solely based on qualitative or anecdotal evidence (Molony, 2006, 2009; Overå, 2006) and lacks quantitative assessments. This paper aims to close this gap.
In this study, we elicit trust levels with an incentivized experiment, namely the canonical trust game by Berg et al. (1995). Experimental sessions were conducted from July 2018 to August 2018 in 17 different villages in Turkana County, Northern Kenya, and included a total of 402 respondents. We differentiate with regard to the object of trust by measuring trust towards fellow villagers, trust towards people from a neighboring village, and trust towards city dwellers from the county capital. Past research in rural sub-Saharan Africa indicates that smallholder farmers exhibit less trust towards people from different villages (Etang, 2010; Etang et al., 2011) or people from the next larger city (Parlasca et al., 2019). The differentiation of the object of trust therefore allows investigating heterogeneous effects of mobile phone use on trust depending on the physical distance between trustor and trustee.
This research adds add to the existing literature in several ways: to the extent of our knowledge, this is the first study to explore the direct link between mobile phone use and trust using quantitative household data from a low-income country. Thus far, no study has analyzed the role of geographical distance in the relationship between mobile phone use and trust. Lastly, this analysis contributes to the extremely sparse literature on trust in the context of pastoralist communities in sub-Saharan Africa.
The remainder of the study is organized as follows. Section 2 lays out the conceptual framework that guides the analysis. The data are explained in section 3 and the empirical framework is presented in section 4. The results are discussed section 5, followed by concluding remarks in section 6.
