Towards sustainable agricultural development: empirical insights from Uganda on challenges and the way forward

Agriculture plays a vital role in Sub-Saharan Africa (SSA), engaging over 75% of the population and providing food and employment for more than 60% of the rural population. Despite its critical importance to regional economies and livelihoods, the sector receives insufficient recognition and support while encountering numerous challenges that restrict its benefits and jeopardise long-term sustainability. The impact of climate change is one of the major obstacles facing SSA agriculture because the agricultural sector in SSA is dominated by smallholders with a limited capacity to adapt to changing environmental conditions and mainly rely on rain-fed agriculture. Additionally, other economic and socio-demographic factors, including unstable food prices and gender inequalities, play a significant role in hindering agricultural development and food security in SSA. The recent COVID-19 pandemic has added further complexities by severely disrupting supply chains at all levels, locally and globally.
To overcome the challenges in agriculture sector in SSA, there is a growing global call to recognise and support smallholder farming in the region. Recently, sustainable agricultural technologies, such as drought- and heat-tolerant (DHT) crop varieties, have been developed and promoted to smallholder farmers, aiming to enhance their resilience in farming. However, the progress in supporting smallholder farmers has been generally slow, and empirical evidence to guide policy development for better support is limited. This thesis aims to empirically address the key challenges and prospects for smallholder farmers in SSA using Uganda as a case study.
The agriculture sector in Uganda, a developing country dominated by small-scale agricultural operations, presents a relevant empirical case for this thesis. The thesis comprises four papers, each written as a journal article with its own unique objective. The first paper aims to characterise trends and seasonal variations of observed temperature in Uganda over the past 120 years. By assessing national and regional temperature changes, the study provides valuable information for adaptation and climate risk mitigation efforts. The paper contributes to the literature by decomposing temperature data into seasonal and trend components using an innovative unobserved component (UC) model. The paper also employs a statistical testing procedure to test for a presence of a deterministic trend in temperature.
The second paper evaluates the effects on household food security of adopting DHT varieties for any crop (and specifically maize). The study utilises national-level data collected by the World Bank in 2009–10 and 2010–11 throughout Uganda. The Household Diet Diversity Score (HDDS) and Food Consumption Score (FCS) are used as proxies for food security. To isolate the effects of adopting DHT seeds on household food security, the study uses a treatment effect estimation procedure while exploiting regional temperature variations as an exogenous instrument.
In the third paper, a machine-learning approach is used to identify predictors of women’s participation in agriculture. The study uses data from 585 smallholder farmers in Nwoya District in Northern Uganda to build a random forests machine-learning model, which separates and ranks crucial predictors of women’s participation in agriculture from a set of 55 potential predictors. A two-step procedure of model training and validation, along with the recursive feature elimination (RFE) algorithm, is employed to identify the most important predictors of women’s participation in agriculture.
The fourth paper assesses the volatility of prices for major food commodities in Uganda, using monthly time series food price data collected by the World Food Programme (WFP). The study estimates and discusses differences in price volatility for various commodities at national and regional levels. The study also assesses the change in price volatility before and after the COVID-19 pandemic to develop an understanding of the impact of the global shock on the price volatility of major commodities traded in local food markets in Uganda.
The findings of this thesis provide valuable insights for enhancing the resilience and adaptive capacity of smallholder households and achieving the Sustainable Development Goals (SDGs) in Uganda and other SSA countries. The first paper offers an understanding of temperature trends and seasonal variations at national and regional levels, which is crucial for building resilience in smallholder systems in SSA against climate-related uncertainties. These findings are foundational for achieving SDG 13, which aims to take urgent action to combat climate change and its effects. The second paper highlights that adopting DHT crop varieties enables smallholder households to improve food security even amid climate change, aligning with SDG 1 (no poverty), SDG 2 (zero hunger) and SDG 13 (climate action). The third paper shows that economic factors are the most significant predictors of women’s participation in agriculture. This provides vital information for policymakers considering investing in projects that improve the economic situations of smallholders and increase women’s participation in agriculture. Additionally, these findings represent an essential step towards achieving SDG 5, which focuses on achieving gender equality and empowering all women and girls in SSA. The fourth paper highlights that the price volatility of major commodities in Uganda cannot be generalised because it varies regionally. This provides policymakers with essential information for promoting regional food security by identifying regions particularly vulnerable to price volatility in major staple food commodities.