Spatial Autocorrelation Panel Regression: Agricultural Production and Transport Connectivity

Using various new spatial data, the paper measures different types of transport accessibility and estimates their impacts in Ethiopia

Abstract

Transport infrastructure is an important determinant of agricultural productivity. Using various new spatial data, the paper measures different types of transport accessibility and estimates their impacts in Ethiopia. The paper takes advantage of a historical event that Ethiopia, a landlocked country, ceased freight rail operations connecting its capital and the main seaport in the late 2000s. Using the substantial changes in transport accessibility, the spatial autocorrelation panel regression is applied to show that the proximity to close markets and the access to the port are of particular importance for agricultural production. The elasticity is estimated at about −0.05 to −0.13, depending on type of accessibility. It is also found that there are considerable spillover effects that come from the spatial autocorrelation errors, meaning that crop production at one place is affected by its neighborhood environment, possibly including land fertility and weather conditions.

This is an output of the World Bank’s Strategic Research Program

Citation

Iimi, A., You, L. & Wood-Sichra, U. Spatial Autocorrelation Panel Regression: Agricultural Production and Transport Connectivity. Netw Spat Econ 20, 529–547 (2020). https://doi.org/10.1007/s11067-019-09489-y

Spatial Autocorrelation Panel Regression: Agricultural Production and Transport Connectivity

Published 17 January 2020