The phenomenon of regression to the mean might seriously bias estimates of the impact of hospital rate setting laws. We examine this problem and find that, while estimates of the average effectiveness of rate setting laws are not strongly affected, estimates for individual states may be skewed.