If anything can go wrong, it will.

Murphy knew what he was talking about. His justifiably famous law, which points to the general perversity of nature and man, applies nowhere better than to economic policy.

Everyone complains about economic policy. And everything does seem to go wrong. But I have yet to hear anyone comment on a remarkably systematic and quite unnecessary perversity in the way economic advice is used in policy- making. I call it Murphy's Law of Economic Policy:

Economic advice has the most influence where economists know the least and the least influence where economists know the most.

This law, which has a profound effect on our nation's economic policy, seems to have escaped the attention of both social scientists and the media. Let me explain the law and offer some examples.

The first part of the law predicts that economic advice will be taken most seriously on issues about which economists disagree most vehemently and in areas where economic science is in greatest disarray.

Take forecasting the state of the economy. Economists have some, albeit imperfect, understanding of the fundamental causes of business fluctuations. But, frankly, we have never been very good at predicting the precise timing of recessions and recoveries. Yet here the public dotes on every word we say.

In recent years, theoretical and doctrinal controversies have made a shambles of macroeconomic theory. But this chaos did not stop the Federal Reserve from embracing monetarist policies in 1979, precisely when the theoretical foundations of monetarism were crumbling and the doctrine was coming under increasing attack in scientific circles. The uncertainties that divided economists did not deter policy-makers from striking out boldly in the wrong direction.

Reaganomics provides an even more extreme example. What is popularly known as supply-side economics was a creation of polemicists and journalists that conservative economist Herbert Stein once characterized as "punk economics." The doctrine had no scientific basis, was riddled with internal contradictions and was rejected as wishful thinking by the vast majority of professional economists.

The idea that you could actually increase tax revenues by cutting income tax rates, for example, was never a contentious issue among tax experts: no serious economist thought it would happen. Yet this notion was one of the cornerstones of the 1981 tax cuts.

The other half of Murphy's Law of Economic Policy predicts that when economists offer advice on issues about which they agree, and think they know what they are talking about, that advice typically will be ignored.

Government policy toward tariffs, quotas and other restrictions on imports is probably the most outstanding case. Almost all economists, whether liberal or conservative, steadfastly support free trade, and have done so consistently for years. In this, they have usually stood alone, with support from neither workers nor capitalists nor politicians (though many have paid lip service to free trade).

Another important example is environmental policy. Most economists believe that a system of financial penalties on polluters is a far more efficient, workable and effective way to protect the environment than the web of direct regulations that we have used. Though this idea has been accepted by economists for more than 50 years, it was considered outlandish by policy-makers until quite recently. And it still has had precious little influence on policy.

A third example is the tax system. Virtually all economists agree that tax laws normally should be designed so as to interfere as little as possible with individual decision-making. For example, a good income tax would tax different types of income at the same rate. Otherwise, lightly taxed activities will be encouraged and heavily-taxed activities will be discouraged.

But our actual tax system is a travesty of this principle. Countless loopholes and special provisions lead to huge differences in the tax rates levied on different types of income. And each successive revision of the tax code seems to make things worse rather than better.

Murphy's Law is bad news for those who seek to improve national economic policy. Let me draw a parallel. Imagine that patients scrupulously followed their doctor's advice on problems about which there is little scientific agreement (like how to cure chronic backache), but systematically ignored their doctors on problems about which medical knowledge is most secure (like how to prevent polio). People would surely be far less healthy than they are.

Yet this is roughly what happens in economic policy. Like doctors, economists know a good deal about the treatment of certain ills, but rather little about others. So economic policy, like medical care, will never be perfect. But it is more than a little disheartening to see sound economic advice routinely ignored while quack remedies are adopted.

The surest way to improve economic policy would be to repeal Murphy's Law of Economic Policy. But, mindful of what the law says, I will give no such advice.