An economic analysis of the returns to Canadian swine research: 1974-97

This paper reports a new set of estimates of the returns to swine research in Canada. These estimates are obtained using Agriculture and Agri-Food Can ada's Canadian Regional Agricultural Model (CRAM). Positive Mathematical Pr ogramming is incorporated into the model for use in this study. The CRAM al lows the effects of supply shifts from technological change in the hog indu stry to interact with product and factor market conditions in the rest of C anadian agriculture. Extensive sensitivity analysis is conducted to examine the robustness of the return estimates under variations in some of the key assumptions employed in the analysis. The costs of public and private sect or swine research are estimated. Public sector research costs are inclusive of the marginal excess burden of taxation. Overall, the estimated benefits from Canadian swine research are high relative to the estimated costs for the time period considered. Previous estimates of the returns to Canadian s wine research were obtained by Huot et al. (1989) with a partial equilibriu m model that did not allow for intra-sectoral resource use adjustments. The estimated returns obtained in the present study are generally higher than those obtained by Huot et al. For example, the estimates obtained from the direct application of the econometrically estimated supply, function in thi s study gave an internal rate of return of about 124% and a benefit-cost ra tio of 22.4 to 1. Huot et al reported comparable estimates of about 43% for the internal rate of return and 6-7 to 1 for the benefit-cost ratio. The d ifferences in returns are not solely attributable to the use of a multi-mar ket versus a single-market partial equilibrium approach. There are also dif ferences in the estimates of the marginal excess burden of taxation between the two studies.