On the connection between energy prices and inflation, there are two extreme views:

1) That there is no connection; that inflation is entirely a manner of money supply and, secondarily, of federal deficits.

2) That increases in energy prices have been the entire "cause" of the inflation of the 1970s.

Neither proposition will withstand scrutiny.

Consider a concrete example. Between December 1978 and early 1980, the average or "composite" price of a barrel of oil to U.S. refiners increased from just under $13 to approximately $26. What has that increase done to inflation? The story comes in three parts:

The direct result has been to increase the price level -- the number of dollars it takes to "buy" the GNP -- by 3.2 percent, simply because the oil we burn producing the GNP costs $13 per barrel more. Phased over 1 1/2 years, to allow for the passing through of the higher oil price into the price of everything from gasoline to shoes to sealing wax, that 3.2 percent increase in the price level translates into a two-plus percentage point increase in the average annualized inflation rate (the rate of change of the price level). But that direct effect on the inflation rate is temporary; it will last only about 18 months, until the pass-through is complete.

Unfortunately, that is not the end of the story. Since the late 1960s, wages have become increasingly sensitive to the recent past rate of inflation. After 18 months of faster inflation -- two-plus points faster -- wage rates too are likely to be rising between one and two points faster than before.

If so, unit labor costs to business will be rising that much faster and so, therefore, will prices, even after the price of oil has stopped rising. And prices will keep pulling wages after them. A one-time increase in oil prices will have permanently ratcheted up the underlying wage-price "spiral," a circular process in which wages simultaneously chase and push up prices, like a dog chasing its own tail.

A good monetarist will quickly point out that the faster inflation cannot persist unless the Fed prints money fast enough to feed it. That is both true and deeply misleading. If the Fed fails to feed it -- if it allows money to become increasingly expensive and tight -- the proximate result will be to reduce aggregate spending for goods and services. (The only direct effect on prices, e.g., through mortgage rates, will be in the wrong direction.)

Confronted by a decline in spending, in orders and sales -- by the piling up of unwanted inventory -- U.S. business will typically cut back production and lay off people quickly, and shade price increases only very gradually. It's not a case of wickedness but of survival. As long as wage rates keep rising, so will costs of production; businessmen have to raise prices in order to stay in business.

The puzzle is posed by wages. For whatever reasons -- the story is understood only imperfectly -- wages tend to keep rising at their inherited rate in the face of increasing and substantial unemployment. Wage inflation in the United States exhibits extraordinary one-way inertia. If past regularities hold firm, it would take two extra points of unemployment, maintained for two years, to compress the core inflation rate by about two points, say from 10 percent to 8 percent, even in the absence of any further external price shocks (OPEC, a bad harvest).

The cumulative two-year cost would run to about $250 billion of unproduced output and real income ($4,000 per family), and that does not include the subsequent productivity slow-down due to the recession-caused reduction in capital formation. It seems a poor bargain.

We face a nasty dilemma. Even a one-time oil price increase will speed up the persistent, inertia-driven, wage-price inflation. Countering that by conventional fiscal and monetary measures alone, by compressing aggregate demand (as, for example, by President Carter's March 14 measures) will impose enormous costs. But not countering it, somehow, will guarantee that inflation will persist at the faster pace, until another oil price increase, or a bad harvest, or a bout of excess demand, as in 1965-68, ratchets it up again.

Is there a better way out? I believe there is, but it entails -- as part of a package -- "interfering" with prices, and especially with wage rates, by guidelines and tax penalties, and that is still anathema to most businessmen and many labor leaders.

They are right, of course, that by itself even a serious wage-policy won't work, but that is not the issue. And they are also right that, at best, and even as part of a comprehensive program, such an incomes policy would be both inefficient and inequitable. The question is -- compared with what?