The Clean Air Act, the most expensive and perhaps the most inefficient of the federal government's regulatory statutes, must be re-authorized before Sept. 30. Given the emphasis the Reagan administration is placing upon regulatory reform, it is surprising that it has done no more than endorse a handful of "principles" it would like to see followed if, as seems likely, the act is amended as part of its reauthorization.

It is even more surprising, and disappointing, that none of the administration's principles involves support for the simplification of air quality management through the use of economic incentives, instead of detailed, source-by-source controls.

This administration, more than any other, perhaps, might have been expected to look with favor on decentralized, market-like methods to meet environmental and other social goals. Instead, it seems content to endure the present cumbersome and legalistic approach, while turning more responsibility for enforcement over to the states.

The Clean Air Act can only be described as baroque. Its provisions are so numerous and complicated that they cannot easily be explained. The act directs the Environmental Protection Agency to set standards for thousands of different sources of pollution. Moreover, these standards are to differ depending upon the length of time a particular source has been operating, the air quality in the region surrounding it, the pollutants emanating from the source, and--in some cases--its financial condition. EPA simply has not had the time or the resources to set all these standards. For instance, the agency has yet to establish standards for new plants in certain important industries, even though 11 years have passed since the Clean Air Act first directed EPA to do so. Nor has it ever been made clear how "lowest achievable emission" controls (required of new plants in dirty areas) are to differ from "best available" controls (required in pristine areas) or garden-variety "new source performance standards," even though the act suggests they should be different.

Enforcing the standards and making them more efficient has also proved difficult. Very often compliance with a standard is ascertained solely from unaudited assurances set forth in letters submitted by polluters. Even when standards are enforced, EPA cannot be sure it is regulating efficiently. There is very little staff effort devoted to determining whether the costs of abatement across different sources are similar. In fact, most studies demonstrate that the individual source standards are set so inefficiently that more than one-half the current control costs could be saved if the burden could be shifted slightly from those finding pollution control expensive to those finding it relatively inexpensive.

Beginning with the Ford administration, EPA began to develop new approaches to air pollution control which could eventually save billions of dollars. Firms wishing to locate in dirty areas were permitted to do so provided they installed the most modern control equipment and purchased further reductions in pollution ("offsets") from other sources operating in the area. This policy was codified in the 1977 amendments to the Clean Air Act and was expanded and strongly supported by President Carter. Unfortunately, the policy still requires rigid, technology-based standards for new sources even when pollution can be reduced much more cheaply by other existing sources in the same area.

In addition, under Carter, EPA began to experiment with trade-offs across sources of pollution within the same firm. If a firm could meet EPA's objectives with a different and less expensive set of controls than those initially prescribed, it could--under certain circumstances --substitute the more efficient controls. This became known as the "bubble"policy since a firm could treat all of its emissions as if they occurred under a bubble or a plant-wide umbrella, worrying only about the final emissions from this bubble to the environment. The firm could choose its own mix of controls to get to this result. Armco Steel, Du Pont, 3M and other firms have discovered that this new approach can save them millions of dollars, but cumbersome procedural requirements have limited its applicability.

These two new initiatives, if pursued, could eventually reduce pollution control costs substantially at the same time air quality is being maintained or even improved. With some changes in the act and prudent administration, we could be on our way to a simpler, less burdensome and more effective clean air policy. Unfortunately, the Reagan administration appears to be ignoring these new mechanisms and the opportunities they create for relieving the burden of detailed government regulation. Rather, it is choosing to perpetuate the system of detailed standards for the myriad of individual smokestacks, as well as strict, technology-based standards for new facilities. This combination of tight technology-based standards for new plants, little trading of emissions-reductions among existing plants, and weak enforcement against these older plants will generate the worst possible outcome. The standards on existing sources will continue to be set inefficiently and enforced rather loosely. As a result, the maintenance or improvement of air quality will require further tightening of new-source standards. New plants, which are inherently cleaner, will not be built; and the dirtier, less productive old equipment will be kept in place longer. The government will continue to try to impose a cumbersome regulatory structure on the nation's basic industries, slowing economic growth and impeding progress toward a cleaner environment.

Why does the administration not support the transition toward decentralized trading of emissions reductions among firms as a substitute for detailed point- source regulation? Part of the reason is that there is very little constitutency for economic efficiency in Washington. Everyone could benefit from a more efficient air pollution policy, but few are aware of the savings they could reap. Politicians and lobbyists, on the other hand, are more interested in fighting over the distribution of the costs of regulatory programs than they are in minimizing the costs of meeting environmental or other goals.

There are other, more troubling reasons. Business firms are reluctant to support changes in the existing system of air pollution controls because this system clearly favors older plants. Many of these older plants are in the northeastern and north-central sections of the country, which are losing industry to the Sun Belt. Allowing new plants to be built without onerous new-source standards might hasten the demise of the older facilities in the Frost Belt. Firms which own these older plants are not eager to relieve the burden imposed on new facilities. Even those firms with good growth potential may be opposed to a system that allows them to buy pollution offsets from existing nearby plants. Not only may they fear the uncertainty a market in pollution would bring; they may also resent having to pay for environmental clean-up at another firm. They would prefer to be told how to curb pollution by the government rather than rely upon market forces to do so efficiently.

There can be little doubt that the Reagan administration is squandering an important opportunity. Rarely can one find an opportunity to produce more of any social good-- clean air, in this case--while lowering costs. A market in pollution abatement could surely generate this unusual combination of outcomes, but the administration seems unwilling to take advantage of this rare free lunch.

Mr. Crandall is a senior fellow of the Brookings Institution. Mr. Portney is a senior fellow of Resources for the Future.