| This thesis empirically measures the impact of formal credit on agricultural production in a developing country setting, using a structural model of farmers' acreage (net land leasing), groundwater use and labor use decisions. The dataset used in the thesis is from a World Bank power sector reform survey in Haryana, India, for the agricultural year of 1999–2000. The land leasing estimation incorporates the fact that there are transaction costs to entering the rental markets. Accordingly, the three sides of the market, leasing-in, leasing-out and non-participation, are estimated using a limited information maximum likelihood function with three regimes. The credit constraint is measured as a dummy and is treated as endogenous to the leasing decision. The thesis uses land owned as a proxy for the contractual choice under which the land is leased and finds that a credit constraint has a positive effect on leasing-in of land for farmers with smaller ownership of land (the sharecroppers), and the expected negative effect only on farmers with large landholdings (the renters). This is explained by the fact that landowners often provide cheap credit under sharecropping arrangements so that the credit-constrained farmers may lease in more land under sharecropping to overcome this constraint.; In the second part of the thesis, a system of demand equations is estimated for hired labor, groundwater and other inputs for each of the irrigation technology categories—the electric pump owners, diesel pump owners and water purchasers. In Haryana a large majority of electric pump users do not pay a marginal price for water. Therefore, to arrive at an effective price of groundwater, a production function of groundwater extraction is also estimated. The results of the second stage estimation indicate that controlling for acreage, credit constraint has a positive effect on groundwater use by electric pump owners who pay a very low marginal price for water, and a negative effect only farmers who purchase water at the full price. |
