Economics of land fragmentation : effects on productivity, technical efficiency and crop diversity in Tigray, Northern Ethiopia

Abstract

In this study I investigated the effect of land fragmentation (LF) in Northern Ethiopia on 1) farm productivity, 2) efficiency and 3) crop diversity using stochastic production frontier (SPF) analysis and farm household model (FHHM) with factor market imperfections. The analysis is carried out at plot level mainly in a plot-panel framework using a cross-sectional sample data of 421 households and their corresponding 1918 plots. Along with "plot size" and "farm size", I used three other land fragmentation indicators that are widely used in the literature: "number of operated plots", "distance to plots" and "SI-index" (where larger index means highly fragmented). Applying different econometric specifications, on 1) the dominant-crop model, 2) two-main-crops model and 3) aggregated-output models, I found no evidence to the conventional claim that land fragmentation could be detrimental to productivity or efficiency; in fact, the results indicate to the opposite. A non-trivial positive and significant association is observed between 1) number of plots and productivity, technical efficiency and crop diversity and 2) between SI-Index and crop diversity. A negative association is observed between farms size and productivity, technical efficiency and crop diversity. These signs of positive implications of fragmentation on productivity and efficiency can be explained by its indirect effect through diversification than by diseconomies of scale. Signs of negative implication of land fragmentation are observed only in the dominant crop model that ignores diversification and farm integration. However, such analysis is methodologically inconsistent when farm production involves diversification, since higher fragmentation can counterbalance the negative impact through diversification.