agriculture and construction at the other end of the scale measured by the assets. In finance 2.6 percent have 78 percent of the assets; manufacturing, 1.5 percent having 66 percent of the assets; mining and quarrying, 2.5 percent having 65 percent of the assets; in trade, 0.5 percent with 35 percent of the assets, and service, agriculture, and construction falling below that.

Remember, this is a record of corporations. For example, in agriculture there are only seven or eight thousand corporations, so that while it is true that among those seven or eight thousand in agriculture 0.7 percent of them have 28 percent of the assets in existence, the figure means very little in terms of agriculture as a whole. However, for four groups, at any rate-about 90 percent or more of the activity is done by corporations, and therefore one can get a pretty definite indication from these figures of the degree of concentration.

EXHIBIT No. 60

ASSETS OF LARGE MANUFACTURING CORPORATIONS OF
$5,000,000 AND OVER PERCENT OF INDUSTRY TOTAL 1935

The CHAIRMAN. This chart to which you have just been referring deals only with corporate assets?

Dr. THORP. That is correct.

The CHAIRMAN. And you presented a chart earlier in the morning, the chart entitled "Importance of Corporate Activity," in which it was shown, in the second column, that, in agriculture, corporations control only 7 percent of the entire activity, and that, in mining, corporations control 96 percent. In finance, which is the second figure on exhibit No. 59, according to your chart this morning, 84 percent of the activity is controlled by corporations. Don't you think it might be helpful to tie this chart into the other one and, by indicating the percentage opposite each subdivision here, indicate the percentage of the entire activity in that field which is carried on by corporations?

1 See exhibit No. 55, supra, p. 96.

Dr. THORP. That would enhance the usefulness of the chart. The CHAIRMAN. Suppose you do that, beginning with transportation and other public utilities.

Dr. THORP. The grouping is not exactly the same. The industry classes in exhibit No. 591 are those used by the Bureau of Internal Revenue and are based on the financial reports submitted by taxpaying units, whereas the classification in exhibit No. 55,2 showing the importance of corporate activity, is based primarily on reports to the Bureau of the Census in which companies may be divided according to plants or establishments. That means that a corporation chiefly engaged in manufacturing which operates a retail store, would be wholly a manufacturing concern in the eyes of the Bureau of Internal Revenue, but would be subdivided into the two activities by the Census Bureau. With this limitation and some other minor differences in mind, the percentages of activity that are accounted for by corporations for the industries listed in exhibit No. 59 are:

You will note that the only class in exhibit No. 59 showing a percentage different from the "like named" class in exhibit No. 55 is Transportation and Other Public Utilities. This class, appearing in exhibit No. 59, is a summation of the third, sixth, and seventh branches of industry listed in exhibit No. 55. The similarity in the other items is due to the smallness of the differences in classification of the items going to make up the groups in the two charts.

These percentage figures, of course, refer to activity while the data used in exhibit No. 59 relates to corporate assets.

Senator KING. Before you leave this field, you spoke about the corporations engaged in agricultural activities. What particular field of agriculture would fall within that classification?

Dr. THORP. I think the most important area in which any large agricultural corporations appear is in dairying.

Senator KING. Would those factories that slaughter animals and dispose of the meat fall within the agricultural classification?

Dr. THORP. No; I think meat packing is regarded as manufacturing. Senator KING. Would the production of sugar from cane or sugar beets fall within your classification of manufacturing?

Dr. THORP. The refining of sugar would fall under manufacturing, but the actual raising of sugar beets or sugar cane would come down in the agricultural classification.

ASSETS DISTRIBUTION FOR MANUFACTURING SUBDIVISIONS

Dr. THORP. Now we turn to exhibit No. 603 which is presented to show how varied the picture is if one looks at different divisions within the manufacturing field.

This is the chart giving the assets of large manufacturing corporations, the same corporations as were included in the last chart

which we discussed, but broken down by subdivisions for manufacturing. There we find that corporations with $5,000,000 of assets and over are of varying importance in different areas within the manufacturing group, the order being indicated very clearly from the chart-tobacco products; chemicals and allied products; rubber products; metal and metal products; paper and pulp products; food and kindred products; stone, clay, and glass products; printing, publishing, and allied industries; textiles; forest products; and liquor and beverages.

In the extremes, in tobacco products less than 6 percent of the corporations have assets of $5,000,000 and over but they have over 90 percent of the assets; and on the other extreme, in the liquor and beverages industry, 1 percent of the corporations fall in this category, and they have 28 percent of the assets.

I indicated previously that much depended upon the measure one used. The fact is, that one needs a number of different measures in order to determine size. It is important to keep in mind that what is a big unit in one industry might be a small unit in another. A big millinery manufacturing enterprise would be much smaller than a small steel mill, probably when measured either in terms of assets or measured in terms of wage earners.

LIMITATIONS ON MEASURES OF SIZE

Dr. THORP. Now we have certain other things to consider if we are trying to do a complete job, and I hope in the course of time we can report to the Committee a much fuller analysis of this problem of size. For example, many of these enterprises are in fact larger than they appear. They may have foreign branches, for example; they may have such close working relationships with their sources of supply that in fact the size is much greater than appears in the formal organization. And then there are interlockings of various kinds, financial interlockings, interlocking directorates, holding companies and such, which also need to be taken into account if one is to get an accurate measure of our large business units.

On the other hand, I also want to point out that for many purposes these large units are not so large from a business standpoint, in that for many purposes they operate on a decentralized basis. For example, a large enterprise may have 10 plants scattered around the country, and for certain purposes those 10 plants may operate with a relative degree of independence. They may have independence with regard to their labor policies, or independence with regard to the way in which they will distribute their products in that area, so that in some cases one may have to modify his analysis of size, taking into consideration the fact of decentralized organization.

Senator KING. Moreover, there are many gradations in the activities which culminate in the finished product, and each gradation might act in a decentralized form.

Dг. THORP. That is correct. One of the difficulties, of course, about measuring size is that if you try to measure in terms of volume of sales, the enterprise that is nearest the consumer will seem to be the biggest because it will have piled up all the work done by all the enterprises before it, and therefore, it will seem to be a very large enterprise, although it may have contributed only a small part of the total.

Any one of the measures that is available has its weaknesses, and I think we have to follow them all to understand the problem.

Senator KING. Is it not a fact that in many of the plants, the industrial plants, that the finished product is the result of many different activities, perhaps as you state, in various communities, indeed in various States?

Dr. THORP. Oh, yes; we have now a great many highly fabricated ized where many products are brought together.

GROWTH OF LARGE ENTERPRISES

Dr. THORP. I should like to make a few comments with regard to the historical development of this situation. Of course much of this growth is a matter of internal growth in which the corporation over a period of time has reinvested and has grown by that method, but there were two periods of active consolidation and mergers in the country. The first was from 1898 to 1902. John Moody, who wrote a book called The Truth About Trusts, listed 318 industrial corporations which he was able to locate in 1904 which he said had about two-fifths of all manufacturing capital at that time. About 30 of them were capitalized at more than $50,000,000.

The CHAIRMAN. You mean all manufacturing capital?

Dr. THORP. I think he meant all manufacturing capital at that time. Of course there is some question as to what capitalization meant in 1898 and 1902, particularly as these corporations were incorporated with a good deal of water included, and it would be difficult to place value on them comparable with whatever our current accounting practices may be.

That period of consolidation ended rather suddenly, and from then until the twenties, there was no further appreciable development of consolidation and merger, but in the twenties there was a revival. I made an elaborate study of that several years ago for the National Bureau of Economic Research. They were reporting on recent economic changes to a committee of which Mr. Hoover was chairman. At that time I found that in 1922 a little over 300 enterprises had been absorbed by other enterprises in the year 1922; that by 1929 the number of manufacturing and mining enterprises which disappeared through the process of merger and consolidation was 1,245. There was an extraordinary expansion in the merger movement during the twenties. Those figures are for manufacturing and mining. You would find the same thing in hotels and hospitals and motion-picture theaters. It seems to have been a very general phenomenon at that time.

Mr. OLIPHANT. What were the years covered?

Dr. THORP. The years 1922 to 1929, and the lowest was in 1922 when 309 companies were absorbed; the highest was in 1929 when 1,245 were absorbed.

Senator KING. Some were liquidated, were they not, voluntarily, and some through the courts? What I mean is some of the corporations that had existed prior to that time were liquidated.

Dr. THORP. Yes; there is a continual process going on. Many of the enterprises which were formed in the early period, 1898 to 1902 period, did not survive. Some of them made three efforts, were organized three times before they finally collapsed, and there were quite a number of cases which failed to survive.

Mr. ARNOLD. These figures were not intended to include bankruptcies?

Dr. THORP. No; these figures do not include bankruptcies. Representative SUMNERS. Did these consolidations result from schemes developed by somebody who wanted to sell a lot of stock and make money out of it, or was it an economic development from a business operating standpoint?

REASONS FOR CONSOLIDATIONS

Dr. THORP. It is hard to be certain about motives, but I should like to spend a little time on the reasons that the students have found to be present, and we can start with that one. I think there is no doubt but that both these periods of mergers were periods in which the desire to have new securities made available for flotation has been one of the most important existing motives. The promoters of these consolidations have frequently been people in the investment banking business, and the result of the consolidation has been new securities for flotation. That was very evident through the latter part of the twenties.

Representative SUMNERS. That was for the purpose of getting a commodity, to wit, a security, to sell the people.

Dr. THORP. Yes. What would happen would be this: It is not very good arithmetic, but one would take two corporations. Let's say corporation A had a value of 2, and corporation B had a value of 2. You would put those two together and give the new corporation a value of 5 on the assumption that it was bigger and more valuable and there were economies which I will discuss in a moment. You gave 2 to the owners of the first corporation, you gave 2 to the owners of the second corporation, and the remaining 1 you sold to the public. That was the process.

Of course that leads us into the second reason as to why these things take place, and that is the expectation of production and marketing economies. It is perfectly possible that when you put 2 and 2 together, you may get 5, for one reason or another. It may be that the larger enterprise will be able to gain in certain technological ways. My belief is that during the twenties they were particularly eager to expand size, because it was so helpful in marketing.

You can't afford a national radio broadcast if you are a little producer; you can't advertise in the Saturday Evening Post if you are a little producer. Size has certain definite advantages from the standpoint of using modern marketing methods, and, therefore, that was one of the definite considerations. There are many others. I think over the long run the increase in size is definitely to be attributed to certain changes in the economic system, the kind of products that we are producing. You can't produce automobiles with one or two employees; it has to be a fairly good-sized enterprise. The shift from household production over to producing in factories for many of our common necessities has been characterized by the growth of largescale enterprises.

The CHAIRMAN. Would it not be warranted to say that practically all of the so-called durable-goods industry, which was discussed yesterday by Dr. Lubin, is of necessity carried on by corporations?