THE PRESIDENT must decide by tomorrow whether to sign the Orphan Drug Act, and some of the advice he's getting is bad.

There are dozens of rare diseases that could be relieved by drugs that are uneconomical for pharmaceutical companies to produce because of the small number of people who need them. Or there may be a problem in finding companies to do the research necessary to develop drugs to treat rare diseases--an unprofitable exercise unless there are likely to be benefits for other areas of the companies' business. Some companies will test or even produce an uneconomical drug just out of the goodness of their hearts or for public relations purposes. But they can hardly be counted on to do this as a routine matter. Is there a role for government here?

By unanimous votes, the House and Senate have said yes. The Orphan Drug Act would give tax credits to subsidize production of these so-called orphan drugs, at a projected annual cost of $15 million, and would create a small discretionary grant program totaling $12 million to help finance preliminary testing. It would also streamline Food and Drug Administration procedures and grant limited patent-like rights when a drug is orphaned because it would be unpatentable. The legislation is carefully constructed to make maximum use of market incentives, rather than to create an elaborate regulatory scheme or a system of government controls over research and production decisions by drug companies. But there is strong opposition within the administration.

Put aside the usual objections based on cost, because the price tag is too small to treat it as a budget threat. And put aside the objection to a grant program for private research, because there is certainly a public purpose here and there are plenty of other examples of research subsidies for profit- making enterprises. And how seriously can one take Treasury's objection to using the tax code to provide economic incentives? The incentive approach employed by the legislation is quite consistent with the administration's own general approach to government intervention.

In fact, the problem seems to be that Sen. Orrin Hatch attached a rider to the bill instructing the secretary of health and human services to conduct a study on the relationship between exposure to radiation and the incidence of certain cancers. This revives the old argument over the exposure of residents of Utah and Nevada to nuclear testing radiation years ago. The government has argued that any health effects were minimal and has tried to avoid liability, so a study might complicate matters. And some scientists believe the kind of study the Hatch amendment would force them to produce would be scientifically spurious.

This amendment, however, is not a good reason to veto an important bill. The proposed study would have no legal or regulatory effect. Nothing would be banned, no one would receive a court judgment, no workplaces would be regulated as a result of it. Sen. Hatch just wants to force the government to state as best it can the current scientific consensus on radiation hazards--or at least document the extent of the ignorance.

Medical science could give hundreds of thousands of people productive and relatively comfortable lives if the economics of the pharmaceutical industry did not work against them. Congress has given the president a way to help, and he should sign the legislation.