demand due initially to a variety of factors but always accompanied by higher growth of the money stock.

Second, as the inflation unfolds and subsides, it follows a complex and changing pattern of price movements spread out over a long period of time. It is not the case that all prices rise immediately by the same percentage. The pressure of excess demand initially raises production and the demand for factors of production. This puts pressure on products in the earlier stages of production, so that prices first begin to rise in raw materials markets and the more competitive sectors of the economy.

As materials prices rise, costs of production are higher and are passed through as higher prices from stage to stage of the productive process. As labor markets tighten, the growth of wages begins to accelerate, though this usually occurs with some delay. As the cost of living rises, wages rise again to compensate for the decline in purchasing power of wages, though this occurs with a longer delay. The wage acceleration raises unit costs of production and further pushes up on prices.

The excess demand therefore results in a wave of pressures on prices spread out over a long period. Each series of increases feeds back on costs to produce more increases in a process which continues long after the pressure of excess demand has finally subsided. Considerable evidence has accumulated from research over the past decade which supports this view of the inflationary process.1

From the evidence we know that raw materials prices tend to react earliest to inflationary pressures, then intermediate goods respond somewhat later, and at the tail end final goods and services respond. There are many exceptions, but this is the general pattern.

We know that changes in the demand for products are reflected in differential effects on output and prices. In raw materials markets and highly competitive industries such as lumber, prices respond sharply to changes in demand.

In most manufacturing industries and particularly those not characterized by a large number of small firms, prices initially respond little to changes in demand, and the change is absorbed by inventories, queuing of orders, and changes in output. Only later when costs rise do prices respond. The reasons for these differences in price behavior are still not well understood. But there is no doubt of the general nature of the process, and an important effect is that price increases in manufacturing industries tend largely to follow cost increases.

It is incorrect, however, to view these increases as cost push, in the mistaken view that costs are causing prices to rise. It is the excessive demand which has put pressure on prices throughout the economy; the process of inflation is such that the sequence of price changes tends to travel forward from earlier to later stages of production, while the pressure of excess demand had initially traveled backward from the final to the earlier stages of production.

One of the exasperating and misunderstood consequences of this process is that prices subside slowly as inflationary pressures wane and, indeed, continue rising for a while even after demand slackens

1 This evidence is discussed at length in two recent publications of mine. One is "The Hydra-Headed Monster. the Problem of Inflation in the United States," published by the American Enterprise Institute of Washington, D.C., late in 1974. The other is "Inflation and Market Structure 1967-73." which appeared in "Explorations in Economic Research," spring 1975, a publication of the National Bureau of Economic Research.

and markets develop excess supply. The reason is simply that price increases are still moving along the pipeline of costs from the earlier to the later stages of production. Under demand restraints, however, the process subsides in time.

This occurred in the 1966-67 minirecession and the 1970 regular recession, and is occurring again today in the severe recession of 197475. In a recent publication 2 I described this process as follows:

A deceleration of inflation due to restraints on aggregate demand [shows up first in output and only later in prices.] Output is reduced, factor demands decline, and unemployment rises. Prices nevertheless continue upward, because cost increases are still working forward through the price system. The first prices to weaken are generally those of raw materials and of other goods sold in highly competitive markets.

The upward pressure on costs in subsequent stages of production thus eases, weakening prices first for intermediate goods and finally for finished goods at the retail level. Wages resist still longer, and service prices that are tied closely to wages are usually the last to reflect a change in aggregate demand.

Since wage rates are slow to reflect changes in the growth of aggregate demand and are a major component of manufacturing costs, it might seem that manufacturing prices cannot decelerate much before wages do. In fact, these prices do decelerate ahead of wage rates, because demand restraints retard the rise in unit labor costs. A slackening of business activity leads firms to trim costs, chiefly by using workers more efficiently.

This process is at first concealed by the retention of skilled workers who are not fully employed but who are held in the work force against the day when output again expands. In the ensuing recovery, unit labor costs decelerate rapidly, relieving a significant pressure on prices. As the rise in prices eases, the slower increases in the cost of living feed on wage rates, allowing the process of deceleration to continue after the benefits of cost cutting have run out.

The 1970 recession generally followed this sequence. Output per manhour (an overall indicator of efficiency) had actually declined during the peak of the cyclical expansion in 1969, and unit labor costs (private nonfarm sector) rose 8.2 percent in that year while compensation per manhour rose 6.9 percent. Output per manhour began to increase in the second and third quarters of the 1970 recession, but the drop in output resulting from the General Motors strike in the fourth quarter, as well as the continuing recession, held the increase to 1.7 percent for the year.

Thereafter output per manhour began to advance sharply, so that from the first quarter of 1970 to the second quarter of 1971 it rose at an annual rate of 3.8 percent, and it accelerated to 5 percent per year in the first half of 1972. Although compensation per manhour rose about the same in 1971 and 1972 as in 1970, the advance in unit labor costs dropped from 8.2 percent in 1969 to 5.0 percent in 1970 to 2.2 percent per year overall in 1971 and 1972.

Because profit margins began rising in 1971 from their depressed levels, the deceleration in unit labor costs had little immediate effect on most prices. But the rate of increase in the private nonfarm deflator peaked in the fourth quarter of 1970, and inflation began to subside slowly; the deceleration continued well into 1972, though the imposition of price controls in August 1971 obscured the normal pattern of events. As we now also know, what might have been a happy ending to the restraint imposed on the inflation by the 1970 recession was shattered by the explosion of prices in 1973.

In another recent study of mine, I found that the response of prices to declines in aggregate demand has been diminishing since the recession of 1949. In each successive business recession, the rates of change of prices declined less, and this was not entirely due to the decreasing severity of the postwar recessions. The full explanation of declining responsiveness is not clear, but one reason appears to be the continuing experience with inflation. The economy begins to anticipate the continuation of price increases after prices have been rising for a while,

2 "The Hydra-Headed Monster." ibid., pp. 25-26.

3 "Changes in the Recession Behavior of Wholesale Prices in the 1920's and Post-World War II Explorations in Economic Research, National Bureau of Economic Research, winter 1975.

and prices tend to be adjusted upward in anticipation of increases in demand even though markets may at the moment be slack. The result is that inflation, once it gains a foothold, takes longer to subdue.

Now what role is there for price controls in this process? Regulations can be disruptive of production and not always easily enforced, but for purposes of this discussion let us waive those objectives and assume that regulations effectively restrain actual transactions prices. Excess control probably can do that, though I doubt that they can do it effectively for long.

In the inflationary process, excess demand puts pressure on costs. If these cannot be passed along into higher prices because of controls, profit margins will be squeezed, and output will eventually be curtailed. Bottlenecks and supply shortages will appear. That is not a solution to the problem of inflation. Moreover, we cannot control prices forever. Eventually we must imagine their being set free. Then profit margins will return to normal, and the total rise in prices will be the same as without controls, except that part of the inflationary movement will occur later after controls are removed. I see no benefit in bottling up price increases and stretching them out over a longer period in this way.

The usual rationale for price controls is that they will permanently prevent price increases that would otherwise occur. For the same degree of excess aggregate demand, however, prices will eventually reach the same general level. Because of the importance of aggregate demand, the proponents of controls assure that inflation originates in part in cost push of strong unions and of concentrated industries, and that the monetary authorities accommodate aggregate demand to the push.

There is evidence that unions, though they are laggard in responding to the initial outbursts of inflation, later catch up and then keep wages rising too rapidly as the inflation subsides. There is also evidence that more concentrated industries continue to raise prices above concurrent cost increases after demand has slackened. But the evidence indicates that this is a catching up and that profit margins in these industries do not rise higher than they would otherwise be in a noninflationary economy.

To be sure, those industries which have some market power due to oligopoly may have higher profit margins than other industries, but generally the concentrated industries respond more slowly to changes in demand and so catch up last to inflationary movements. This has been confirmed by a variety of studies.*

To impose price controls to hold down the increases of the business giants, therefore, will have the effect of depressing the profit margins of those who generally lag, not the sectors which spearhead inflationary movements. The public and the Congress may derive some satisfaction from shackling the corporate giants, but it will have virtually no effect in curbing inflation.

To restrain the sectors which spearhead inflation you will have to control commodity markets, and I believe you will find that task impossible to accomplish and, if tried, disastrous for the economy.

In particular. I discussed the evidence and presented new data in "Inflation and Market Structure 1967- 73," op. cit.

There is, to be sure, a different rationale for controls which has nothing to do with cost push or administered pricing or concentrated industries. I commented on it in my earlier study as follows:

[A rationale for controls] is that they prevent wage and price increases due to anticipation of inflation. According to this argument, the anticipated rate of inflation enters into wage negotiations and to some extent into the settling of prices. Anticipations appear to adjust slowly to the actual rate of inflation and in an upswing induce lagging behavior.

As inflationary pressures subside, anticipations may cause some wages and prices to perpetuate rates of increase whose justification has disappeared with the onset of slack market conditions. If controls prevent those increases or actually help to reduce the anticipated rate of inflation, they bring actual wage and price increases into line with slower growth in aggregate expenditures. If the controls go no further, they may prevent the prolongation of inflationary movements due to incorrect anticipations. By preventing prices from overshooting their equilibrium level, the controls achieve a higher level of employment. To be effective and beneficial, the controls must distinguish between those wage and price increases that represent catching up with past inflation and those that are anticipating future inflation. To prevent the former increases is futile; and such an attempt, if directed against union wages, creates strife. But at any moment the two kinds of increases are for all practical purposes indistinguishable.

Furthermore, the anticipatory increases should not be repressed unless they are indeed mistaken, which implies that policy will in fact be successful in its intention to subdue inflation. In that event, controls may be marginally effective when they are least needed and demanded by the public.

They cannot, moreover, rescue an unsuccessful policy... When controls are most wanted by the public-in periods of rising demand-pull inflation-they collapse. This happened to the guideposts in the mid-1960's and to the freeze in June, 1973. . . In view of their limited usefulness, their dislocations to the economy, and the danger they will mislead policy makers into believing that suppressed inflationary pressures are in fact subdued and thus into overstimulating the economy, as did happen in 1972, their value as a placebo is not worth the costs.

Thank you.

Mr. BLANCHARD. Thank you very much, Dr. Cagan.

Our next witness is Ralph W. Borsodi, a retired economist, representing the National Retired Teachers Association and the American Association of Retired Persons. I would like to point out to my colleagues that it is the intent of the Chair that testimony by these witnesses and all colloquy be completed by 11:30.

At that time, the subcommittee will take testimony from the Honorable John McFall who is the sponsor of H.R. 4214, known as the Concentrated Industries Anti-Inflation Act.

Mr. Borsodi?

STATEMENT OF RALPH W. BORSODI, ECONOMIST, ON BEHALF OF THE AMERICAN ASSOCIATION OF RETIRED PERSONS

Mr. BORSODI. I am Ralph W. Borsodi, a retired economist and a member of the American Association of Retired Persons. This association and its affiliate, the National Retired Teachers Association, represent a combined membership of over 8,300,000 older Americans. With me this morning is James M. Hacking, a member of the legislative staff.

5 "The Hydra-Headed Monster," op. cit., pp. 56–57.

We appreciate this opportunity to testify before your subcommittee on S. 409, the Council on Wage and Price Stability Act Amendments, and H.R. 4214, the Concentrated Industries Anti-Inflation Act.

Despite the slow and perhaps fragile upturn in the economy that seems to be occurring, the United States is still experiencing the effects of its worse economic recession since the late 1930's and, simultaneously, unacceptably high rates of inflation.

Our associations cannot ignore what is currently happening in the economy. Let it be understood at the outset-inflation and recession are the most serious threats to the income security of our constituency at the present time and are seriously jeopardizing the financial integrity of the income maintenance structure on which they are dependent. It is our belief that administered pricing practices in noncompetitive, concentrated markets, both at home and abroad, are responsible for a good deal of this current situation.

Our associations believe it is essential to develop an effective micropolicy to deal with administered pricing and deny to the large and concentrated industries the excessive and continuing profits they have been able to exact when these are wholly unjustified on the basis of production costs and fair rates of return.

Only by developing an instrument effective to deal with this type of inflation, will we be able to use the highly stimulative monetary and fiscal policy which is essential to ending the recession. No economic program which attempts to deal with inflation while ignoring recession or with recession while ignoring inflation will be successful.

For the poor and the fixed-income aged, the combination of inflation and recession-with its attendant unemployment-during 1974 was catastrophic. With less purchasing power to begin with, it was these groups who suffered the most from inflation. While the magnitude of their dollar income decline may not have been as great as that of other groups, the decline was from a level that was, at best, marginally adequate.

We are not suggesting that all the aged are poor. We point out, however, that 45.3 percent of the aged have total money income of under $4,000 per year and older family units tend to be concentrated more in the lower and less in the upper extreme of the national income distribution.

Although recent increases in social security benefits have reduced the number of aged in the poverty class to under 3.7 million, the incidence of poverty and low income is still substantial. Because of the higher rates of inflation with respect to necessities such as food and housing on which the poor and fixed-income aged tend to spend a far higher portion of their total income, they have, over the last 3 years, suffered a relatively greater loss of purchasing power than other

groups.

The impact of the recession and its present 9.2 percent unemployment rate has rendered even more difficult, if not impossible, any relief of the impact of inflation through increased income from active employment. Even in the best of times, the aged encounter a formidable combination of barriers to employment. With an increasing number of workers competing for a diminished number of jobs, the employ