More cost-benefit? Better cost-benefit?

In this morning's WSJ, Edward Glaeser and Cass Sunstein write on "How to Deregulate Cities and States ... Cost-benefit analysis and 'lookbacks' could lift unnecessary burdens of occupational licensing." You bet. But who will do these studies?  I think that's the real challenge.

Some years ago, Robert Hahn and Paul Tetlock published. "Has Economic Analysis Improved Regulatory Decisions?" They found, "... there is not strong support for the view that economic analysis has had a significant general impact."  What to do?

Given these unimpressive results, where should we go from here? Perhaps what is needed is a more disciplined and formal commitment to benefit-cost balancing, led by the president and Congress, along with comparable officials abroad. As noted above, such a commitment could entail mandating benefit-cost analysis of important regulations in statutes. Congress could also codify a version of the current executive order requiring benefit-cost analysis. It may also want to consider subjecting some proposed laws to at least a crude benefit-cost analysis prior to voting on them. Already, Congress often asks for estimates of the budgetary impacts of laws and proposed laws.

But Congress' record on this is, to be polite, mixed. Pork first. Even when studies are done, they are all over the place. Here is an apparently serious cost-benefit analysis of California's high-speed rail plan.

Many years ago, Charles Lave suggested that consultants be required to bond their forecasts. If taxpayers are asked to accept huge risks into the indefinite future, why should those who grease the wheels accept none of the risk?

What would the specifics of such a consulting contract look like?  This discussion should occur before we simply demand more cost-benefit studies.