Land misallocation and productivity

What is this research about and why did you do it?

A fundamental question in the field of economic growth and development is why some countries are rich and others poor. Existing literature has reached a consensus that cross-country differences in labour productivity are substantially larger in agriculture than in the non-agricultural sectors. Moreover, factor misallocation among heterogeneous producers is important in accounting for cross-country productivity differences. This paper builds on these two perspectives. Specifically, we use micro-level panel data from Malawi to assess the extent of factor misallocation in agriculture, aiming to shed new light on how resource misallocation contributes to the low agricultural and aggregate productivity observed in low-income countries.

‍

How did you answer this question?

We start by estimating farm-level productivity using panel data, controlling for transitory variation, and minimizing potential measurement error. We then construct a framework with the only imposed structure being the span of control of heterogeneous farmers and hence the equilibrium farm-size distribution is non-degenerate. This framework serves as a lens through which we evaluate the data, drawing a comparison between the actual allocation of resources and the model's implied efficient allocation. We also quantify the aggregate productivity gain achievable by reallocating resources to their efficient allocation and compare the impact on inequality among farmers between the actual and the efficient allocations.

‍

What did you find?

Capital and land inputs are substantially misallocated. Specifically, the efficient allocation implies that more productive farmers should operate larger farms, while actual capital and land inputs in the data are largely uncorrelated with farm productivity. Redirecting capital and land from less productive farms to more productive farms to equate their marginal products increases aggregate output by 1.7—2.0 folds, holding quantities of aggregate inputs unchanged. This gain is substantial even if we restrict the reallocation to be within narrowly defined geographic areas. Moreover, evidence suggests that land rental markets alleviate resource misallocation.

Figure notes: Each (blue) dot represents a household farm in the data whereas the (red) dashed line represents the efficient allocation. All variables have been logged. The efficient allocation implies that more productive farmers should operate larger farms, while actual capital and land inputs in the data are largely uncorrelated with farm productivity.

‍

What implications does this have for the study (research and teaching) of wealth concentration or economic inequality?

Inequality among farmers is always a concern of land reallocation. We find that the model-implied efficient allocation, despite concentrated land usage, reduces inequality among farmers and poverty compared to the actual allocation. The key is that low-productivity farmers, which are also low income, benefit more from land rental income than the farming revenue associated with operating their own land. This result highlights the possibility of improving productivity without exacerbating inequality by allowing for rentals (usership) while maintaining land ownership.

‍

What are the next steps in your agenda?

In companion projects, we investigate how land rentals causally reduce resource misallocation, exploring exogenous variations across space and time arising from a decentralized land reform in Ethiopia. We also assess how land insecurity in China affects rural-urban migration and inequality.

‍

Citation

Chaoran, C., Restuccia, D., and Santaeulalia-Llopis, R. 2023. “Land Misallocation and Productivity.” American Economic Journal: Macroeconomics, 15(2), pp 441-65.

‍