On the eve of World War II, the Italian dictator Benito Mussolini told Adolf Hitler that if the League of Nations had succeeded in imposing an oil embargo in response to his invasion of Ethiopia in 1935, he would have had to withdraw "within a week." He added, "That would have been an incalculable disaster for me." It would have also made an incalculable difference to the rest of the 1930s and 1940s.

Matters have now turned right around with the imposition by the United Nations of an embargo against oil exports from Iraq and Kuwait. But there is a big risk that should be faced. Under some circumstances, oil prices could surge again and to such a level as to lead to panic and then -- in the face of inflation, recession, disruption and domestic outrage -- to the collapse of the embargo, which would be a disastrous foreign policy defeat.

The essential problem for the United States is this: to maintain the embargo without inadvertently doing intolerable harm to its economy and that of other Western nations. We have learned from the oil crises of the 1970s that the psychology of shortage and panicky bidding for supplies can drive prices to heights far above what is justified by the fundamentals of supply and demand.

In the present crisis, prices have fallen back from their initial peaks primarily because of the expectation that a few countries -- mainly Saudi Arabia, United Arab Emirates and Venezuela -- will produce to the limits of their capacity. If all goes right, that extra production will just about compensate for the 4 million barrels per day of exports lost from Kuwait and Iraq.

But it is dangerous to depend upon such surge production alone. Simply put, there is no security margin if something further goes wrong. (Note that U.S. oil production has fallen by 2 million barrels per day since 1986, an amount greater than Kuwait's entire output.)

It is not at all difficult to conceive of new difficulties. The surge could be reduced either by the expansion of conflict in the Gulf or because of physical accidents somewhere along the supply chain. Installations and facilities suddenly pushed to 100 percent operating rates are more vulnerable to operational difficulties and accidents than those working at lower levels. Or production that today is part of the equation could be disrupted by anything from industrial strikes (the North Sea) to weather (hurricanes in the Gulf of Mexico) to ethnic and social conflict (the Soviet Union). Moreover, by October the approach of winter will be putting more pressure on the overall supply system.

A critical lesson from previous crises is the great value of "early response" to salve disrupted logistics, as we have here, and to counter the psychology of shortage, which would be reignited by the appearance of any shortfall. It is better to take prudent steps now before new developments push prices out of sight or motorists are fuming in gas lines -- and the administration suddenly finds itself pushed on the defensive.

Since the first Senate investigation of high gasoline prices by Sen. "Fighting Bob" LaFollete in 1923, sudden increases at the pump have been an inevitable subject for congressional hearings. The risk here is that a grave international crisis, requiring national cohesion, could be transmuted into a rancorous domestic debate about gasoline prices that would divert time and attention and miss the main point. Such a spectacle would also undercut the willingness and ability of other nations to stick with the embargo.

Much better to head off such difficulties at the pass and move now to smooth out present and possible disruptions, moderate demand and alter psychology. With Iraq in Kuwait and with tens of thousands of U.S. military personnel now in Saudi Arabia, and more on the way, the crisis is already here. Some aspects of the long-wrought energy emergency programs should be activated now.

A reasonable objective might be to, in a nonpunitive way, reduce U.S. oil demand by 5 to 6 percent, or about 1 million barrels a day, which is just about one-quarter of the oil lost from Kuwait and Iraq. This can be achieved through fuel switching, particularly to natural gas, and by modest conservation measures.

Even a few months ago, some were questioning the value and need of the Strategic Petroleum Reserve. No longer. Still, there is doubt as to whether it will be used and even whether it will work. It makes great sense to bring some SPR oil into the market in the next few weeks, even if only very modest amounts, to demonstrate the SPR's effectiveness and that it is definitely part of the response. Such a demonstration would make buyers think twice before acquiring oil at panic prices on the expectation that prices will surely still go higher.

"Early response" in the United States needs to be matched by coordinated international action, as suggested by the recent International Energy Agency meetings in Paris. The other industrial nations could, in total, match a million barrels per day reduction in the United States. That would give a much greater security margin for the uncertain days ahead. In this way, panic reactions, which can drive the oil market, can be preempted. And that is all to the good, for another lesson of the recent past is that democratic nations cannot resist such emotions for all that long.

The writer is president of Cambridge Energy Research Associates.