Abstract: In the modern regulatory state, there is a serious tension between two indispensable ideas. The first is that it is important to measure, both in advance and on a continuing basis, the effects of regulation on social welfare, usually through cost-benefit analysis. The second idea, attributable above all to Friedrich Hayek, is that knowledge is widely dispersed in society. As Hayek and his followers emphasize, governments planners cannot possibly know what individuals know, simply because they lack that dispersed knowledge. When important information is missing, cost-benefit analysis can be exceptionally difficult to conduct. There are three ways to respond to that problem. The first involves notice-and-comment rulemaking, which has particular promise in the modern era, where regulators are in a far better position to collect the dispersed information of the public. The second involves retrospective analysis, accompanied by a process for obtaining public comment. In many cases, retrospective analysis has found that the ex ante estimates were wrong, thus pointing the way toward potential improvements both in rules and in future estimates. The third, and potentially the most valuable, involves experiments, above all randomized controlled trials, which can give a clear understanding of the likely effects of regulations.