FOREIGN TAKEOVERS of America's entertainment giants are arguably benign, but you'd think that someone would draw the line at a company involved in our nuclear weapons program. Unfortunately, you'd be wrong. Just such a takeover is imminent, and the hands-off attitude being adopted in Washington illustrates graphically the nation's abysmal lack of a sensible policy on foreign buyouts of U.S. businesses.

The acquisition target is Moore Special Tool Company, of Bridgeport, Conn., the world's best maker of ultra-precise machine tools, and one of America's industrial crown jewels. The prospective buyer is Fanuc, a major Japanese player in the machine tool field. Moore is the only U.S. supplier of high-precision machine tools that can meet Department of Energy needs. The components fabricated with these tools directly affect critical variables such as the explosive force of a nuclear weapon and the risk of accidental detonation.

The United States will now be totally dependent on foreign machines (German, Swiss and Japanese) for the most sensitive operations in maintaining the arsenal that has anchored its defense for almost 50 years. A double irony is that the most precise tools (from Moore) would be provided by Fanuc, whose home country has forsworn nuclear weapons as a matter of principle.

How did we wind up in such a fix? In general, it's the fault of the business environment in which U.S. companies operate. Our cost of capital is too high, partly as a result of government borrowing; our tax laws penalize investment and reward consumption; our piecemeal regulatory structure is insensitive to effects on international competitiveness. Moore also has specific problems of its own. The company has worked for the U.S. government on projects that were neither profitable nor applicable elsewhere. As a niche-market producer, Moore must depend on exports (more than 50 percent of sales) to generate enough volume to stay profitable. But because of the strategic value of Moore's machines, the Departments of Defense and Energy have blocked or delayed approval of key overseas sales.

Even so, the existence of alternative foreign suppliers has frustrated U.S. export control. Foreign customers, aware of Moore's problems in getting export licenses, have often settled for second-best, but still satisfactory, products from other sources. The net result: Sales and profits dwindle for Moore but worrisome technology continues to spread as foreign competitors supported by their governments rush to fill the gap.

Concern that the machines might be misused or fall into the wrong foreign hands is valid, but the application of our export control laws often serves neither national security nor business interests. Ultimately, we can only influence the flows of technologies if we are a technology leader. If our companies are put out of business or sold to foreign concerns, then we have little leverage over the actions of others.

The lack of a coherent national economic strategy, where business decisions, government procurement, export control, anti-trust and tax policy are all linked, has produced this situation.

Washington's mechanism for reviewing sales of U.S. companies to foreign entities is CFIUS, the interagency Committee on Foreign Investment in the U.S. The committee's guidelines, known as the Exon-Florio provisions, say that the acquiring company must be shown to pose a national security risk for a sale to be denied. But how should national security be defined? Is it military security only or should it encompass economic security as well? To date CFIUS has chosen to define security only in narrow military terms. Of some 480 takeovers reviewed, the process has blocked only one sale -- to a company owned by the Chinese government. A bill to make economic security an explicit criterion in the decision-making process failed to pass in the last Congress. Indeed, Congress failed to renew the entire Defense Production Act, of which Exon-Florio is a part. As a result, the world's greatest economy currently has no guidelines whatsoever on foreign takeovers.

Operating in a legal vacuum, CFIUS last week voted informally to recommend that the president approve the sale of Moore to Fanuc. Although many CFIUS agencies were uneasy, none was courageous enough to recommend denial. But if ever there was a clear national security case, even according to the narrow definition, the Moore case is it. What could be more deeply involved in national security than the machines that produce the nation's nuclear arsenal?

The United States has a long history of controlling exports for political reasons. What if the Japanese, Germans or Swiss decide not to sell to or service the DOE's nuclear program? After all, Switzerland is a neutral, Germany does not possess nuclear weapons and has a large, politically active anti-nuclear movement, and sensitivity to things nuclear is integral to modern Japan. Can we rely on these governments never to respond to political considerations and to continue to sell to DOE? The grudging cooperation we have received from Germany and Japan in the Persian Gulf crisis, as well as the stalled GATT negotiations, demonstrates vividly that these countries often do not view their respective national interests as coincident with ours, even on critical issues.

Further, will these countries tightly control the technologies involved and keep them out of the hands of the Saddam Husseins of the world? German companies have been heavily involved in the Iraqi chemical, missile and nuclear-weapons programs. The Arab blacklist for years has influenced Japan's relations with Israel, and Japanese companies have sold restricted technologies -- the most famous case being Toshiba Machine's sales of nuclear submarine propeller-quieting technology to the Soviets.

Moore's plight is far from unique. Another current CFIUS case, for example, is also mired in policy confusion: the sale of SemiGas Systems to its Japanese competitor, Nippon Sanso. SemiGas is the key supplier of specialized gas cabinets to Sematech, the government-business consortium established to help our computer chip industry become competitive again. Although the Japanese would stand to gain valuable insight into the technology of their U.S. competitors by acquiring SemiGas, CFIUS approved the sale in August. But last week, the Department of Justice, which did not object in the CFIUS process, announced it would go into court to block the sale on anti-trust grounds.

With respect to the impending sale of Moore, ordinarily, the president would have 15 days to act on the case. But the normal rules have lapsed thanks to Congress's inaction. Moreover, the president has plenty on his plate currently. Yet critical cases like Moore cannot be allowed to slide through even when the White House is caught up in an immediate crisis. A way must be found to keep Moore and companies like it in U.S. hands and to help them to prosper.

The problem is that such measures would look like "industrial policy" -- a taboo to some key White House economic policy makers. Yet export controls, which they accept, are also a form of industrial policy. Such controls certainly hurt Moore financially. The real problem is not that we might wander into industrial policy but that we already have too many industrial policies, too many uncoordinated laws and regulations that sandbag our companies in their competition with foreign rivals.

We desperately need a more strategic approach to our national economy. Decisions to block the Moore sale and fully support the Justice Department's suit against the SemiGas acquisition are the places to start.

Kevin Kearns, a former State Department official, is a fellow at the Economic Strategy Institute in Washington.