Agricultural Dynamics and Structural Transformation: Can they Withstand Weather Extremes?
Kristin Muthui & Natalia Zugravu-Soilita
Abstract
Extreme weather increasingly influences the economies of low- and middle-income countries, where agriculture remains a major employer. Yet its role in structural transformation is poorly understood. Using within-country anomalies for droughts, heatwaves and excess rainfall, we estimate a simultaneous-equations system for agricultural labour productivity and employment on a 35-country balanced panel (1996–2017). Static models suggest that extreme heat consistently depresses both labour productivity and agricultural employment, while moderate droughts appear benign in isolation but become highly damaging when combined with heat. Irrigation buffers isolated droughts and moderates heat-induced labour exits, yet offers little protection against compound shocks. Nevertheless, these static estimates overstate immediate impacts and underestimate longer-run effects. Once we incorporate persistence and compute long-run effects, isolated heatwaves lose significance, whereas compound “hot droughts” emerge as the main source of agricultural labour productivity decline, with cross-equation feedbacks further amplifying these losses. Excess rainfall consistently raises labour productivity but, by increasing short-term labour demand, slows reallocation and shifts the economy toward a more labour-intensive transformation trajectory. These results challenge the conventional view that labour productivity gains automatically draws workers out of agriculture: under certain climate conditions, higher output per worker can coincide with sustained labour intensity. Policy must therefore pair rapid-response safety nets for heat stress with longer-term investments tailored to the specific profiles of climatic risk.
Evidence weight
Balanced mode · F 0.40 / M 0.15 / V 0.05 / R 0.40
| F · citation impact | 0.50 × 0.4 = 0.20 |
| M · momentum | 0.50 × 0.15 = 0.07 |
| V · venue signal | 0.50 × 0.05 = 0.03 |
| R · text relevance † | 0.50 × 0.4 = 0.20 |
† Text relevance is estimated at 0.50 on the detail page — for your query’s actual relevance score, open this paper from a search result.