How many experts does it take to figure out a better way to regulate?

In this case, four, who gathered recently at the AEI-Brookings Joint Center for Regulatory Studies to discuss "How Can the Government Improve Regulation?"

The program encouraged looking ahead as well as mulling over some of the missteps of the past. Among the questions that came up:

Why do some rules deliver what they promise while others create problems? Why do some not cost as much as expected while others far exceed expectations? How reliable is the science and economic analysis that agencies rely upon to come up with rulemakings? What should be required of agencies when they begin to write a rule? Does Congress hamstring agencies by writing legislation that is too prescriptive? Is the risk of not regulating properly considered? Does the public fixate on the most improbable of risks, such as getting shot in the sniper attacks that recently terrorized the Washington area?

Timely questions all, as the Bush administration puts its imprint on regulatory policy, with an emphasis on strict cost-benefit analysis, and the reversal of several Clinton-era rules. Much of this direction has been coming from Bush's regulatory czar, John D. Graham. He is head of the Office of Management and Budget's Office of Information and Regulatory Affairs.

At the panel discussion, C. Boyden Gray, counsel to President Bush and a proponent of market incentives to clean up pollution, said his principles for better regulation are so basic that they tend to be forgotten.

Gray suggested that regulators not pick favorites among all the groups interested in the outcome of a rule: "It's easy to play favorites . . . and the Congress is always willing to help out." He sees a need for a "rigid adherence" to cost-benefit analysis of rules, prescribing performance standards that let industries figure out how best to comply with rules, and for setting up market incentives such as trading pollution credits to get the best results. "The markets will sort out problems, I think, better than bureaucrats can."

The second panelist, Sally Katzen, who was head of the OMB regulatory office in the Clinton administration, had the advantage of very recent hindsight. She described ideas for motivating agencies to write "better" rules that take into account economic and scientific principles early in the writing process. "That's not to say the system is broken or people have been fools or incompetents," she said.

She suggested that some carrots that might encourage agencies to use better data and analysis in their rules. For example, if an agency gets its cost-benefit analysis critiqued by a peer review panel, it might warrant a less strenuous review at the OMB. Or, the OMB could assure agencies that well-thought-out rules would be introduced publicly with fanfare by the president (the old face-time incentive). Or, simply give the agencies the funding they need to do their jobs. "Good analysis costs money," she said. "How about $50 -- anything -- to show your heart is in the right place?"

Cass R. Sunstein, a law professor at the University of Chicago and author of several books on regulation, suggested having agencies fill out a scorecard that would accompany their rules. Sunstein called it a "two-pager that would be easily accessible" and tell anyone who cared to know the cost, benefit, assumptions used, and alternatives that the agency explored.

"From outside the Beltway, it looks like we're still stuck in 1970s environmentalism," he said, where rules are highly prescriptive and directed at a single problem. He suggested that regulators pay greater attention to the magnitude of problems, disclose the risks involved in making regulatory decisions and choose "the cheaper, more effective tools" to solve a regulatory problem.

Over the years, there has been debate over how to best regulate the regulators: Should it be the OMB, Congress or the courts that play the primary role in keeping tabs on the agencies?

Paul Noe, Graham's counselor, said at the forum that his office is best suited to the role because it offers perspective about what regulatory policies are "best for society" and offers consumers a kind of insurance that rules will be of high quality. He said that good review focuses on efficiency (get rules reviewed in a timely way), sound science (make sure there is credible scientific evidence underpinning a rule, and transparency (let the public see how the rule came together and was reviewed).

In the past few weeks, there have been chances to test Noe's ideas. The Bush administration has made some far-reaching regulatory decisions. After years of opposition by organized labor, Mexican trucks will soon be rolling over the U.S. border; the administration proposed thinning of national forests to reduce forest fires; a Clinton administration rule to allow states to use unemployment funds for paid family leave was rescinded; the Clinton administration ban on snowmobile use in Yellowstone National Park was set aside; power plants and refineries will no longer have to install new pollution controls when they expand.

"We need effective centralized regulatory review as a form of consumer protection because it's about maximizing social welfare," said Noe, explaining how the Bush administration OMB sees its role in approving -- or rejecting -- new rules.