Input/Output Measures and Implications for Productivity Estimation: The Case of US Dairy
Department of Agricultural & Applied Economics
University of Wisconsin - Madison
Wednesday, September 27, 2017
Taylor-Hibbard Seminar Room (Rm103)
12:00 pm-1:30 pm
Lack of detailed quantity measure of input and output, and ignoring output quality differences are common issues encountered in production function estimation and associated productivity assessment. Using home-grown feed, family labor, and milk quality as example, we quantify the contributions of above mentioned factors and consequent estimation biases in the evaluation of total factor productivity. Production functions are estimated using alternative measures of inputs and outputs. The Agricultural Resource Management Surveys (ARMS) data of the US dairy farms over the period of 2000-20010 are employed. The panel structure of the data allows us to investigate changes in total factor productivity under different input/output measures while controlling for input endogeneity. The results indicate that including unpaid labor and homegrown feed has minor impact on the ranking of productivity assessment. Total factor productivities estimated from deflated revenue are significantly different from those using direct quantity measure. We find that dairy operations with large scale of production and relatively large share of revenue from dairy tend to be more overestimated using revenue measure. Accounting for milk quality results in a significant change in the distribution of productivities of the sampled farms.