generalized law of relationship between the average of all wholesale prices and the average prices of oils and fats. This line is algebraically represented by the simple equation : y=1.935308x-94.896.

This equation says that, on the average, if the index of all wholesale prices is multiplied by 1.935308 and from the product is subtracted 94.896 the result will be the wholesale price index of oils and fats. Applying this formula to the average of wholesale prices for the year 1947 (11 months), we obtain a theoretical figure for the wholesale price of oils and fats of 196.8. The actual index is 186.1, or nearly 11 points below what might be expected of it on the basis of relationships between the two series in the years from 1926 through 1941. Thus the index which recorded the largest gain of all appears to be somewhat below, rather than above, its normal course. Oils and fats throughout the period 1926-41 were subject to much wider fluctuations in price than was the general average and so would tend to move up more rapidly during the past year or two than general prices.

But this is not the whole story. The law of relationship between the two series as set forth by the equation and by the line in the chart has been determined upon the basis of only 16 years, and it is an average relationship. Suppose it had been possible to observe these relationships for a thousand years. Is the law as determined markedly different from what would be obtained in studying a much longer period? What, then, is the extent of our sampling error?

The answer to this question is given in columns 3 and 4 of table 1. The equation as written is more likely to be correct than any other, but there are many variations with degrees of likelihood. If we write the computed value of the 1947 index of oils and fats 196.8±22.6, as we do in column 3, we shall include the true figure 95 out of 100 times. That is to say, the odds are 19 to 1 that the true figure lies between 174.2 and 219.4.

The actual index for fats and oils, 186.1, falls well within this range, and must, on this basis, be regarded as not significantly deviating from normal. Even if we narrow the range so that the odds are only 7 to 3 that the true figure lies within it, we get 196.8±11.3, leaving the actual deviation of 10.7 of doubtful significance.

A similar study has been made of all the other price groups included in table 1. In a number of instances, charts are provided, presenting the results graphically. The only exceptions (and those only partial exceptions) are found in the cases of lumber, petroleum products, shoes, meats, and hosiery and underwear. In these instances, it became clear that there was a longer-term drift in the group prices which was not shared fully, if at all, by the trend of all wholesale prices. It therefore was necessary to take a third factor, time, into account. The procedure in these instances is somewhat more involved, but the net result is the same. The estimated "normal" result, the actual index, the deviation, and the "confidence limits" are shown for these as well as for the others in table 1. These data are to be interpreted precisely as are those for the other series of which oils and fats is a typical example. The charts show the actual index for these groups, and the computed index in the upper section, In the lower sections, one of the factors is eliminated while the relationship between the other two are shown as in the case of oils and fats.

Similar analysis of other groups of commodity prices, which preliminary study suggests are not in a very definite way related to changes in the general average of all wholesale prices, is under way. The results, so far as they appear helpful, will be published in the March issue of the Business Record, along with a study of price changes during and after World War I.

ANITA R. KOPELSON, FREDERICK W. JONES, Division of Business Economics.

Mr. GAINSBRUGH. If I may, I would like to perhaps recap what I have said in the price sector by giving you a contrast between prices at their zenith in 1920 and then perhaps their zenith in January 1948:

From 1914 to May 1920, the whole commodity wholesale price index went up by 145 percent. That is from pre-World War I to the turning point in May 1920.

From 1939 to January 1948 wholesale prices this time went up 115 percent. They have not gone up quite so much as post-World War I, but there is a decided difference in the items which contributed to the increase in average prices in the two periods.

Last time, post-World War I, hides and leathers rose far above the average; textile products rose 245 percent; fuel and light, 182 percent; building materials, 212 percent; house furnishings, 152 percent.

This time, only two groups are substantially above the increase in our wholesale prices, and those two groups are again farm products and foods, the series in which the major price correction has been taking place.

You can do the same thing for the cost of living index and see how much the cost of living index rose post-World War I and what pushed the items up then, and how much the cost of living index arose postWorld War II and what pushed the items up now.

The CHAIRMAN. Roughly speaking, would it be accurate to say the field of inflation is food and food products?

Mr. GAINSBRUGH. The primary area we have discovered in the price zone which is out of keeping with our present price structure are farm products and food. There are some others-lumber and building materials which are on the high side.

But the two dominant items, because of their weight, that affect both wholesale and cost of living index, which are out of line on the high side are farm products and food, and both of those are being corrected. There has also been quite a good deal of emphasis placed upon a correction which has been taking place in the retail trade.

We hear a good deal about the fact that jewelry sales have turned downward; that liquor purchases have fallen off in recent months; that women's apparel sales are not so high as they used to be.

You can name quite a few retail trade groups in which undoubtedly received report of declining dollar value.

you have There, again, if we go through the same line of analysis, we will find that those were the very areas in which you would expect declines to begin to appear as supplies of other goods became more abundant.

This is no great surprise-that women's apparel sales have fallen off. They were one of the items that were furthermost out of line on the upside relative to levels of income that were prevailing.

Some of our apparel manufacturers were caught short by that swing, but they should not have been.

Millions of women entered the labor force during the war, stayed in the labor force in part during the initial postwar period, and exercised a far more independent option as to how their income would be spent than when they became the member of the group in which the head of the household determined how that income should be spent. Liquor sales: The figures were again on the high side relative to our national or disposable income. But perhaps life in the war period and postwar period called for that type of relaxation. It was in the vulnerable area and we anticipated that particular downturn which comes before downturns developed in other areas.

The soft-goods field was on the high side and durable goods were on the low side. As more and more durable goods became available, the proportion of durable goods would rise and the proportion of nondurable to total sales would decline. That was a correction that was inevitable if we were going to get a restoration of balance in our consumption patterns. These patterns do not change violently for major classes of consumption. That is what has been going on.

There, again, I can make the statement I did with respect to price change. Such changes as have been taking place in the retail area have been in the reduction and restoration of a normal balance between durable and nondurable, between one form of consumption and another, in the same way the corrections have been taking place in the price area-have been primarily confined to the items on the high side. It is for that reason it is possible to state each of those changes has been salutary and each change, whether in retailing or whether in price, does not necessarily signify we have begun the downturn.

There is one further line of development that I would like to present to you, and then perhaps it would be best that I stop and, if you have any questions, try my hand at answering them for you.

With regard to this field of retailing, I have already said that we have moved from the easy catch-up stage of the initial postwar period into the next unknown postwar period.

The next stage will be characterized by more violent competition for the consumer's dollar as backlogs are exhausted and supply approaches levels of demand.

The composition of production and consumption in this next stage needs far more attention in business and market research than it has thus.far received.

We know as little about the shifts in production and employment and consumption under a stabilized price level as we knew about the upswing of national income from less than $100,000,000,000 prewar to well over twice that amount currently.

The low proportion of the consumer's dollar spent for optional items of consumption, such as automobiles, is a particular element of strength in the current business picture.

A high rate of optional spending is rightly regarded as a vulnerable element, a vulnerable element in the business picture.

The reason for that is that such spending rises more rapidly on the upswing of income and contracts more violently on the downswing.

Unfortunately, it declines more violently than the income.

Conversely, where little signs of weakening of saturation are found in optional consumption, a strong basis exists for the continuation of both high-consumption expenditures and sustained high production. Optional or sensitive consumption is today not as high, relatively, as in previous periods of high employment. As best we can estimate it for 1947, optional expenditures accounted for only 13 percent of total consumption expenditures as compared with 18 percent in 1929 and 14 percent in 1941.

The bulge in 1946-47 came primarily in semioptional outlays rather than in the volatile optional items.

There is little evidence here of extreme vulnerability in this annalysis of consumption, particularly in light of the backlog for new cars and other durable optionals.

It has been my favorite habit to, at a forum we hold each year on the business outlook to which I turn next, raise one question with our forum at the end of the evening's discussion.

That question is on the business outlook for next year. That question has always been this: Can you cite any past period of depression in which the durable-goods industry has been going full blast and has had months of effective orders on their books?

I have not yet had chapter and verse of such example cited. The bulk of our consumption is still in the forms of consumption which will not depart too violently from changes in the rate of income contraction. Our total output of passenger cars last year was less than 3.6 million, and the production figure that is most likely for 1948 is 4,000,000. The 1929 production exceeded 4.6 million. The average age of the car on the road today is some 9 years.

This back drop of unparalled strength in optional, consumption, which is far from saturated, for at least the immediate short term,

may tower above the foreboding back drop of the historic pattern of postwar price collapse after previous wars.

One final exhibit for the benefit of the record, because it bears directly upon a question that is of concern to you.

What I gave you has been largely qualitative in character rather than of specific-figure content, and I know a great deal of interest attaches to a specific figure rather than to a general discussion, particularly a figure on national income.

We have at the conference board for the past 2 years assembled a group of nationally known economists and asked them to discuss the business outlook for us-such men as Lionel D. Edie, Jules Backman, Louis H. Beam, Solomon Fabricant, Edwin B. George, Everett E. Hagen, A. D. H. Kaplan, members of the economic fraternity in Government; Robert Nathan, Clyde Rogers, Glenn Saxon, Helen Slade, Bradford B. Smith, and Rufus Tucker, and others.

We incorporate that discussion in this study. (The matter referred to is as follows:)