The Effect of Land Tenure and Erosion Measures on Productivity and Investments: Plot and Household Level Evidence from Mali

The importance of land tenure and erosion structures as important soil conservation measures that increase productivity and make agricultural production more sustainable in developed and developing countries have received a lot of attention in the literature, especially during the last two decades. However, given the highly diverse climatic regions, especially in sub-Saharan Africa, evidence is mixed. This study examines these relationships in Mali and it provides evidence on the effect of land tenure and soil erosion on productivity using two different definitions of land tenure. Unlike other studies, we provide aggregate results, results by poverty quintile and for each of the five most important crops in Mali. We use a recent nationally representative household survey from Mali to, first, examine the land tenure affects erosion structures. Then we focus on the effects of erosion structures and land tenure on maize, rice, millet, peanut and sorghum productivity in Mali. We use several empirical methods to address endogeneity concerns. After controlling for a variety of farm and household attributes we find that a land title does not affect the probability of building an erosion structure but household size positively affects the probability of having one. Contrary to some of the literature on land titles, our findings indicate that having the land title for a plot does not affect agricultural productivity. However, an erosion structure on the other hand has a positive effect on three (rice, millet and peanut yields) out of the five main crops in this analysis. Finally, quintile regressions indicate that erosion structures are more common among poor households while households that are better off are more likely to plant trees and the effects of land tenure on erosion outcomes depend on the definition of land tenure used. These results indicate that here are heterogenous effects across households based on poverty level and also crop specific results. Thus, policies targeted to investments in erosion structures may be more effective if they are constructed based on household types. The effects of erosion structures and other soil conservation measures could shed more light if they are examined across a variety of crops.