Spatial econometric analysis of the ımpact of the agricultural sector on economic development

Purpose: The main purpose of this study is to examine the contribution of the agricultural sector to the Turkish economy through spatial econometric neighborhood relations. The relationship between the variable of GDP per capita and non-agricultural employment, in-agricultural employment, production amount, agricultural area and agricultural exports was examined. The relationship between the dependent variable GDP per capita and the independent variables non-agricultural employment (number of people), in-agricultural employment (number of people), production amount (tons), agricultural area (decare) and agricultural exports was examined. Design / methodology / approach: In the study, spatial econometric models were examined under three subheadings and it was decided that the appropriate model was the fixed effect model with the help of the Hausman test. Necessary tests were applied to determine the most appropriate model among the SDM, SAR and SEM models, and it was concluded that SDM gave the best estimation results, supported by the results obtained from the SAC model. Findings: When the variables in the model were examined, it was determined that GDP per capita had a positive and significant relationship with non-agricultural employment, in-agricultural employment and production amount. A significant inverse relationship was found between the agricultural area variable and economic growth. When the variables representing the regional neighborhood were considered, it was concluded that the GDP per capita showed a significant inverse relationship only with the non-agricultural employment variable. Originality / value: The study investigated the effect of the agricultural sector on economic growth for a 12-year period by taking data at the provincial level. While the use of spatial econometrics as a method in the study constitutes the originality of the study, it differs from other studies because the spatial effects of the data and variables are also addressed in the model.