Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Doctorat → États-Unis → 2006 → Agrarian Households in Semi-Arid Tropics Evaluating Policy Options

Pardee RAND Graduate School (2006)

Agrarian Households in Semi-Arid Tropics Evaluating Policy Options

Mukherji Arnab

Titre : Agrarian Households in Semi-Arid Tropics Evaluating Policy Options

Auteur : Arnab Mukherji

Université de soutenance : Pardee RAND Graduate School

Grade : Doctoral degree in public policy analysis 2006

Approximately 165 million people, about a fifth of the world’s poor, are known to be resource scarce agrarian households, or "smallholders’’, living in rain fed semi-arid tropical areas (SATs). Uncertainty in livelihoods in SATs is higher than in other areas of intensive farming with short burst of intense rainfall, high soil erosion and cycles of drought. Chapter 2 describes economic condition of the smallholders, their significance, and the state of knowledge on some micro-interventions that have been used to reduce the uncertainty in their incomes. Chapter 3, following Rosenzweig and Wolpin (1993), characterizes decision making in risky environments by formulating a structural dynamic model of agricultural investment (in bullocks) for smallholders. This incorporates key characteristics of decision making for smallholders e.g. income uncertainty, no rental market, credit constraints, and preferences over mean and variance of incomes. The model’s utility function primitives are calibrated using data from the ICRISAT village level studies. Once the calibrated parameters are obtained, counterfactual policy simulations identify which ones are most successful at preserving household wealth. The interventions I consider are (a) livestock intervention, (b) a soil and water conservation intervention and (c) an employment guarantee scheme (that allows households a fixed income during droughts). I find that livestock management and soil and water conservation have a minimal impact on asset holdings, while the employment guarantee scheme provides substantial asset protection throughout the lifetime of these households. However, the livestock intervention is the most cost-effective intervention. Chapter 4 evaluates an agricultural CDD that was designed and implemented by a village development committee in the semi-arid tropics of Madhya Pradesh, India. I use non-parametric propensity score adjusted difference in difference models and retrospective data from a field study to evaluate the effects of this CDD on a variety of outcomes. The findings suggest that the CDD led to : (1) a decrease in agricultural income but no change in total income. (2) a reduction in the incidence of skipped meals and collecting non-timber forest produce, and (3) reduction in sharecropping in favor of independent farming. This study is based on an original data set collected by the author. Chapter 5 concludes with policy implications and a discussion of areas of research that will further improve policy design and implementation.


Version intégrale

Page publiée le 13 décembre 2011, mise à jour le 10 octobre 2018