This study examines the dynamic and asymmetric interrelationships between climate factors, agricultural performance, and economic growth in this country over the period 1974-2023. We use an advanced econometric approach combining ARDL (Autoregressive Distributed Lag), NARDL (Nonlinear Autoregressive Distributed Lag), and QARDL (Quantile Autoregressive Distributed Lag) models to analyze the nonlinear effects of climate variables (temperature and precipitation) on agricultural productivity and GDP per capita. The results highlight significant and heterogeneous climate effects, with notable asymmetry: a 1°C increase reduces agricultural GDP by 1.2% in the long term, while an equivalent decrease has no significant effect. Rainfall deficits have a greater impact on agricultural production than surpluses, with an amplified effect during periods of recession (2.3 times greater). Quantile analysis highlights structural disparities: small producers depend on imports to adapt, while large farms, although more productive, are vulnerable to heat and water stress. Robustness tests confirmed the validity of the models, with stable residuals and proven cointegration. These results highlight the need for differentiated policies, including: (1) progressive water pricing to limit overexploitation of groundwater; (2) targeted subsidies to encourage the adoption of water-efficient irrigation technologies; (3) training programs for smallholders to promote resilient practices. The study makes a significant contribution to the existing literature by proposing an innovative methodological framework for analyzing asymmetric climate effects in vulnerable agricultural economies, with direct implications for national resilience strategies, including.
M. Chérif (Fri,) studied this question.