SIMULTANEOUS ESTIMATION OF TECHNOLOGY ADOPTION AND LAND ALLOCATION

The paper considers the econometric modeling of technology adoption when crop choice is simultaneous. Bivariate probit is used to estimate a model of irrigation technology choice and land allocation using a unique field-level data set from California's Central Valley. Special attention is paid to the proper calculation of marginal effects in the bivariate probit model, which are often useful for policy purposes. Estimation results confirm that the choices of irrigation technology and land allocation are simultaneous. With regard to the influence of price incentives on agricultural water use, estimation results from the bivariate probit model indicate that the influence of water price on the adoption of precision irrigation technology is much larger than previously realized. A univariate model of technology choice that treats land allocation as exogenous underestimates the effect of water price on the adoption of precision technology by over 40 percent.