Titolo della tesi: Essays on Positioning in Value Chains, Proximity to Markets and Development
This thesis examines the relationship between value chain positioning, market proximity, and development. The research constituting this work focuses on the more recent debate in the development literature on market positioning and participation by farmers in rural developing areas. Although nowadays, crop commercialization is generally considered to have positive effects on development, the literature on the main determinants of this relationship still needs to reach definitive conclusions.
Market participation allows farmers to diversify their production, bringing significant advantages to small-scale producers, like access to new customers, lower costs, and business risk diversification. Nevertheless, this outcome could be very different in a risky environment like the one in developing countries. This scenario can be exacerbated (or eased) in agricultural product value chains. The literature has vastly debated how a market value chain is structured, but it has yet to discuss how it gets structured, especially at the farmer level.
From a “macro” point of view, the conceptual framework for (global) value chain participation and positioning stands on what was primarily outlined by Antràs & Chor in 2013 and 2018. At the “micro” level, particularly in development studies, the theoretical framework around positioning in VCs for smallholder farmers is still unsettled. At the same time, research on the links between market proximity and farmers’ vulnerability to food security is scarce.
Drawing from both the new trade literature on value chains and the development literature on household development and food security, this thesis addresses, at multiple scales of analysis, the research question of whether there is a connection between value chain positioning and welfare, as well as how proximity to markets affects the relationship between resilience and food security. More specifically, this work delivers a theoretical framework for farmers’ positioning in agricultural value chains; secondly, it explores the linkages across proximity to markets, resilience, and food security; finally, it tests whether better farmers' commercialization positioning both in terms of downstreamness in value chains and proximity to private companies, increases farmers' consumption as well as firms' productivity.
The contributions of this thesis to the current literature are conceptual and empirical. Essay 1 provides a first indicator for farmers’ positioning in value chains and tests its validity compared to its current alternatives. Based on a unique dataset, Essay 2 outlines the application of the Chaudhuri (2003) Vulnerability as Expected Poverty (VEP) framework to a subjective, non-monetary variable of food security and a new empirical approach using machine-learning techniques in the VEP calculation procedure. Lastly, Essay 3 goes beyond the boundaries of a sole-farmer value chain analysis with the integration in the analysis of a dataset focusing on firms, allowing to study the impacts of farming households' value chain positioning beyond the farm.
Although one cannot rely on ad hoc datasets for micro analyses on value chains, this work employs robust empirical methods and panel household data to test the validity of the proposed indicator. In addition, specifically in Essay 2, a mediation analysis estimates the indirect effect of resilience on food security volatility and vulnerability via proximity to markets (a standard proxy for market positioning in the development literature). Ultimately, the empirical assessment around the impact of farmers’ value chain positioning is improved by the inclusion of spatial panel models accounting for spatial spillovers in their calculations (see Essay 3).
Essay 1 proposes a micro-level measure for the positioning of farmers in value chains (in the absence of information on farmers’ inputs of production it turns out to measure positioning in “market chains”) inspired by the conceptual framework outlined by Antràs and Chor (2013, 2018), leveraging for farming households’ selling location and buyers. Using the World Bank's panel data for Ethiopia and Nigeria from "Living Standards Measurement Study: Integrated Surveys on Agriculture," this work also provides an empirical application of the proposed indicator showing how it performs better compared to the current alternatives at the micro-level. Secondly, by investigating the dynamics of farmers' food and total consumption over time and controlling for various household and production characteristics, as well as possible confounding factors, this Essay demonstrates that changes in farmers' welfare as proxied by food and total consumption, are positively affected by better positioning in the market chain.
Essay 2 advocates that proximity to final markets drives the link between resilience and food security. This work uses an exclusive dataset made available by the International Fund for Agricultural Development in 2017-2018 to contribute to the understanding of this impact. The paper applies a hybrid empirical approach combining machine learning algorithms with standard vulnerability approaches. Specifically, this work finds positive and statistically significant associations among proximity to markets, resilience, and food security. The work tests the plausibility of the exclusion restriction that market proximity does not affect food security fluctuations other than its impact on resilience capacity by implementing an instrumental variable approach. Moreover, using mediation analysis, this Essay reveals that market proximity accounts for a significant share of the positive correlation between household resilience and food security outcomes. The dampening role played by market proximity in decreasing welfare fluctuations is also confirmed when replacing food security outcomes with income ones. Overall, these findings suggest that policymakers should prioritize interventions to improve infrastructure and access to markets to boost household resilience and, in turn, decrease welfare fluctuations and vulnerability to food insecurity.
Essay 3 applies spatial panel regression models to a unique longitudinal dataset of firms and farmers’ surveys. This work stands on the availability of two datasets of surveys conducted by the World Bank Living Standards Measurement Study (LSMS-ISA) and by the Central Statistics Agency of Ethiopia in three data collection waves between 2010 and 2016 in Ethiopia. Specifically, this paper tests the relationship between farmers' positioning in markets (estimated both in terms of geographical distance and positioning in value chains) and firms that import and export abroad, as well as the relationship between firms' closeness to farmers and their productivity levels. The key results are i) better farmers' positioning, both geographically to firms in global markets and in value chains, boosts households' welfare; ii) firms' proximity to farmers operating in value chains affects their total sales as well as productivity. These findings highlight how better farmer-to-firm and firm-to-farmer relationships represent a crucial means to foster local development.