Analyzing the environmental impact of agricultural production in Africa through a comprehensive spatial econometric framework

Abstract

This study investigates the environmental impacts of agricultural production in Africa, focusing on the continent's challenge of sustainably feeding a rapidly growing, food-insecure population. The study presents new evidence on three disaggregated sectors-crop, livestock, and fishery, using a fully modified agricultural-induced Environmental Kuznets Curve (AEKC) framework anchored on the Stochastic Impact by Regression on Population, Affluence, and Technology (STIRPAT) model. Using data from 50 countries (1990–2021), environmental quality is assessed through the agricultural total ecological footprint (ATEFP), measured in global hectares. The Spatial Durbin Model (SDM) is employed to assess spatial dependence and spillover effects. Exploratory Spatial Data Analysis (ESDA) reveals strong ecological spillovers between neighboring countries, highlighting proximity's influence on ecological outcomes. A 1 % increase in a country's agricultural output raises its ecological footprint by 0.543 % and its neighbors' by 0.0469 %. While the traditional Environmental Kuznets Curve (EKC) hypothesis is rejected, the AEKC is confirmed, suggesting that environmental improvements can eventually occur with continued agricultural growth. The sectoral analysis reveals varying environmental impacts: the standard EKC is observed exclusively in the fishery sector, while the AEKC is validated solely in the crop sector. The animal sector is U-shaped, becoming significant in the long run. Given spatial dependence, the study recommends deepening of agricultural regional integration and increased incorporation of agriculture in Paris Agreement related national development frameworks. Furthermore, policymakers should adopt sector-specific, spatially sensitive strategies that support a transition toward sustainable and resilient agricultural systems across the continent.