Some past deregulatory successes came in “big bang” changes, like the airline deregulation of the 1970s, which shut down the Civil Aeronautics Board. What we need today is the selective pruning of bad regulations. Cost-benefit studies are a good idea, but they tend to be done when we have the worst possible information about the effects of regulations — namely, before the regulations are passed. Furthermore, cost-benefit studies may look only at some of the largest regulations, and not the general problem of regulatory accretion over time.
Better bureaucratic incentives are needed. Agencies are now motivated to generate regulation after regulation, because those are the formal assignments set before them. One possible step forward would be to require agencies to submit plans for retiring some fraction of their regulations over the next few years, and to reward these agencies for seeing this process through.
For a while I toyed with the idea of automatically sunsetting some subset of regulations, but I couldn’t quite bring myself to endorse it. I like the basic idea, but I worry about the ongoing uncertainty imposed on businesses. It makes it harder for businesses to make “once and for all” adjustments and may impose an even greater cost on the allocation of attention from top management. Businesses like certain regulation, partly for good reasons (easier to deal with), and partly for bad (it may hurt smaller competitors even more and keep them out of the market). Still, to the extent you understand the burden of regulation as about dealing with the regulations, rather than the regulatory mandates per se, the case for sunsetting is not a slam dunk.
Addendum: Some readers have asked me about a reference on the discussion of asthma treatments and medications, try this Cass Sunstein column.