Paper short abstract:

This paper studies large landlords’ political advantage in rural Pakistan, its consequences for public goods provision, and response to shifts in land productivity.

Paper long abstract:

This paper studies vote buying and public good provision in a rural polity with unequal distribution of land and interlinked tenancy (sharecropping) contracts. I model a monopsonist landowner who can buy cheap votes through transferring private utility to his tenants, which leads to political advantage for the landlord over a non-landed competitor, and consequently low incentives to provide broad public goods. The political advantage of a landowner is considered in response to exogenous shifts in land productivity; particularly, I show that the benefit of running for office for landlords diminishes with exogenous technical change. Household data from Pakistan is used to test the model; I show that during a period with general elections, politician landlords transfer benefits to tenants by offering more favorable tenancy contracts, while landlords with no political incentives do not change their contracts. To extend the empirical analysis, I study the colonial roots of the ‘feudal effect’, by looking at prevalence of large estates or `jagirs’ during the British and pre-British rules. A higher incidence of these estates is correlated with low public goods and large landowning politicians in the post-colonial era. I call this the ‘feudal effect’, whereby the inter-linkage of land and power leads to an environment with weak political competition and low public good provision. Using the green revolution as a shock to agricultural productivity, I study the degree to which this ‘feudal effect’ persists across areas with varying response to technical change. The hypothesis I test is whether high productivity land makes it less likely that large landowners will run and win in elections and provide low public goods.