Genetic Testing to Signal Quality in Beef Cattle: Bayesian Methods for Optimal Sample Size

Genetic testing is one way that feeder cattle producers can credibly signal quality to buyers. However, quality signaling in the presence of asymmetric information typically requires paying measurement costs. Given that previous research has indicated that the value of genetic information is generally not enough to offset the current cost of testing, we evaluate random sampling as a strategy to reduce the overall cost of testing. An economic approach to sample size determination is introduced utilizing a Bayesian decision theoretic framework to balance the expected costs and benefits of sampling. Data from 101 pens (2,796 animals) of commercially-fed cattle are used to empirically evaluate optimal sampling. Assuming profit is linear (nonlinear) in genetic information, results indicate that at the baseline parameter values an optimal sample size of nine (five) out of 100 animals generates returns from sampling of $7.87/head ($5.96/head). Sensitivity analyses indicate that the degree of asymmetric information (absolute difference between seller and buyer prior expectations of quality) is the major driver of the overall results. The results provide strong evidence that random sampling generates benefits that far exceed the costs.