Analyzing the spatial autocorrelation between the growth of agricultural and non-agricultural sectors in Afghanistan's Economic-Agricultural Areas

The agricultural sector is one of the most important and main economic sectors of Afghanistan to improve the situation of poverty, unemployment and food insecurity in this country. On the other hand, due to the nature of production regions, the adoption of regional policies is necessary for the growth of the agricultural sector. Also, major studies show that in the field of investigating the factors affecting growth, ignoring the issue of spatial dependence of areas on each other, will lead to bias and inefficient estimates and incorrect results. Therefore, the aim of the present study is to determine and explain the spatial dependence by considering the effects of spatial overflow on the growth of the agricultural sector in the economic-agricultural areas of Afghanistan using static and dynamic panel spatial models. For this purpose, economic-agricultural areas of Afghanistan have been determined and after confirming the existence of positive spatial correlation using diagnostic statistics and first-order spatial autoregression model, Spatial correlation between agricultural sector growth in these areas evaluated in the form of spatial errors model (SEM) as well as static and dynamic spatial autoregressive model (SAR) as the final model. In the estimation models for the period, 2001-2019 to the role of other effective factors such as the amount of government investment and international support, price index, the role of human capital and specific ethnic, religious and spatial characteristics of each the area was also considered and the direct and indirect effects of each factor were extracted. According to the obtained results, the necessity of using spatial models in agricultural sector policy-making and also the importance of continuing to support the agricultural sector of Afghanistan are emphasized.