The Washington Post

Putting Regulations to a Test

One sure consequence of the budget deal is that the president and Congress will now be strongly tempted to use regulatory mandates as a cheap way to deliver the political goods. That needn't be a bad thing; where the benefits of a regulation exceed its costs, clearly it ought to be done. And in fact, many of the regulations that deal with the environment, health, safety and employment—where the demands are likely to be the loudest—meet this common-sense test. But many don't. By our estimate, more than half the social regulations issued between 1982 and mid-1996 flunked a cost-benefit test. Getting rid of those regulations would have increased the size of the economic pie by almost $300 billion.

With the nation already spending more than $200 billion a year for social regulatory mandates, there is plenty of room to design regulations that get better results for less money. A reallocation of regulatory "expenditures" to those areas with the highest payoff could save tens of thousands of lives and, for consumers, billions of dollars annually.

Fortunately, help is on the way in a bill just introduced by Sens. Fred Thompson and Carl Levin. Their Regulatory Improvement Act would put teeth into executive orders requiring agencies to conduct a cost-benefit analysis of every rule with an annual economic impact over $100 million. Mindful of the critics who argue that cost-benefit focuses only on things that can be measured in dollars, the act allows the analyses to include intangible factors.

But the act goes further by requiring peer review of major rules by independent experts, so "junk science" can be trashed before political appointees recite it as gospel on Sunday-morning talk shows. It also requires agencies to justify any regulation that flunks a cost-benefit test, and describe alternatives (including doing nothing) that would bring it up to snuff. Those requirements will provide greater accountability by showing that much of the blame for bad regulation lies with Congress itself.

The bill would also require that the agency appoint a "balanced" advisory committee to recommend existing rules that need to be reviewed to increase net benefits for consumers. If the agency failed to complete a scheduled review within five years, rules not reviewed would be automatically repealed.

Finally, the bill recognizes that not all risks are created equal—that many of the risks regulators go after are no more a hazard than getting hit by lightning on a golf course. The act would require an "accredited scientific institution" to compare relative risks to safety, health and the environment and offer recommendations on setting priorities across agencies—something that is desperately needed if consumers are to get maximum bang for their buck.

We believe this is just a beginning. Federal government regulation has long had a major "forest and trees" problem. The regulatory trees are the many individual rules the government generates every year. Commendably, the act would help to ensure that those regulations are developed more sensibly.

Policymakers fail, however, to offer a process for tending to the regulatory forest—to ensure that Washington sets regulatory priorities on a regular basis, just as the budget act now does for direct government expenditures. As a result, there are wide differences among regulations in the costs of achieving a given benefit (such as lives saved).

A new American Enterprise Institute-Brookings report* that we coauthored with four other regulatory experts goes after the problem. We recommend that Congress rewrite key regulatory statutes to achieve better social outcomes at lower cost, give responsibility to the states for problems that can be better handled there and require an annual report on the costs and benefits of regulation.

We also suggest experimenting with a regulatory budget for some new regulations by which Congress would set annual allowable limits for the regulatory costs imposed on society. The budget would apply only to those rules for which the expected costs exceed the measurable benefits. That would increase accountability and reduce wasteful mandates but would not stop government from implementing rules that are clearly expected to improve the well-being of the average citizen.

In the meantime, enactment of the Regulatory Improvement Act would be a start. It would help transform the regulatory debate by demonstrating that reform is in the interest of consumers and voters, not "big business." That such legislation has already attracted bipartisan cosponsorship is a good omen that true regulatory reform is on the horizon.