Files in this item

FilesDescriptionFormat

application/pdf

application/pdfAMADU-DISSERTATION-2018.pdf (6MB)
(no description provided)PDF

Description

Title:Three essays on the adoption, impact, and pathways of climate-smart agriculture: An ex-post impact evaluation of a natural resource management intervention in southern Malawi
Author(s):Amadu, Festus Onesimus
Director of Research:Miller, Daniel C.
Doctoral Committee Chair(s):McNamara, Paul E.
Doctoral Committee Member(s):Brazee, Richard J.; Kalipeni, Ezekiel
Department / Program:Natural Res & Env Sci
Discipline:Agricultural & Applied Econ
Degree Granting Institution:University of Illinois at Urbana-Champaign
Degree:Ph.D.
Genre:Dissertation
Subject(s):Adoption
Agroforestry
Africa
Climate Smart Agriculture
Developing country
Climate financing
Double hurdle
Endogenous switching regression
Environmental sustainability
Externally funded
Food security
Household
Income
Malawi
recursive bivariate probit
Poverty reduction
Propensity score matching
Yield.
Abstract:This dissertation consists of three (3) papers on climate-smart agriculture (CSA) – an increasingly important approach for achieving sustainable development objectives in the face of global climate change and extreme weather events. It advances theoretical and empirical knowledge at the intersection of agricultural development, environmental economics, and natural resource management through a set of analysis of a large USAID-funded CSA project in southern Malawi. Specifically, it contributes to narrowing the following gaps in the literature: a) lack of conceptual clarity on farm-level CSA practices with highest adoption potential, b) paucity of evidence on the effectiveness of externally supported CSA projects and c) dearth of empirical evidence on specific pathways through which CSA projects generate effects. The dissertation utilizes primary survey data collected from 808 households in five districts across southern Malawi. To obtain plausible estimates of a counterfactual for the CSA intervention, I used rigorous analytical techniques that control for endogeneity and selection bias due to non-random program placement and unobserved heterogeneity. In the first paper, I developed a typology of farm-level CSA practices, which helped to generate and test hypotheses on CSA adoption dynamics in the study area. I then used recursive bivariate probit regression to estimate CSA adoption by CSA practice type (or category). Results showed that the program increased adoption probability by at least 41% and that CSA adoption rates were highest for labor-intensive practices such as installation and maintenance of physical infrastructure like stone bunds and water absorption trenches. Paper 2 used endogenous switching regression to estimate food security impacts of CSA adoption in terms of agricultural yields and household income. I found that on average, CSA adopters obtained yield and household income increases of 90% and 41% respectively. The third paper utilized a double hurdle model with control function to estimate the impact of CSA program participation on agricultural yields, conditional on agroforestry adoption as a CSA impact pathway. The result indicates that CSA program participants that adopted agroforestry saw their yields increase by an average of 31%. In addition to the conceptual and empirical contributions, this dissertation has significant policy implications for sustainable rural development in Malawi and elsewhere in Africa and beyond. For instance, development policies that promote externally funded CSA programs could enhance the adoption of resource-intensive, but higher impacts CSA categories such as agroforestry and physical infrastructure like continuous contour and water absorption trenches, thereby improving environmental conservation and food security in the developing world.
Issue Date:2018-07-12
Type:Text
URI:http://hdl.handle.net/2142/101559
Rights Information:Copyright 2018 Festus Amadu
Date Available in IDEALS:2018-09-27
Date Deposited:2018-08


This item appears in the following Collection(s)

Item Statistics