Page images

Doctor, if there is anyone with you, would you just identify them for the record. I should indicate that we had anticipated that the Assistant Attorney General in Charge of the Antitrust Division, Mr. Orrick, would be here this morning. Unfortunately he is unable to come.


Dr. Mueller. Thank you, Mr. Chairman, and Senator Fong. It is a pleasure to appear again before this subcommittee and to participate in this important inquiry into economic concentration.

Accompanying me this morning are Mr. Harrison Houghton, Assistant to the Director of the Bureau of Economics, and Dr. Stanley Boyle, the Chief, Division of Industry Analysis, Bureau of Economics.

Last year the Bureau of Economics of the Commission made two presentations dealing with various facets of the concentration question. My own presentation dealt with various measures of industrial concentration and related them to trends in the overall picture during the 1950's and early 1960's. Mr. Houghton presented a study of the recent growth of conglomerate corporations in which he traced the overall changes in the multiproduct activities of the large corporations, as well as the experience of five leading conglomerate corporations during the past decade and a half.

Senator Hart. I am sure you prefer that we not interrupt. I couldn't let pass at that point the expression of thanks from the subcommittee to you and to the Commission for developing those two studies.

Dr. MUELLER. Thank you very much. I was authorized to say in behalf of the Chairman of the Federal Trade Commission, that he feels a special obligation to the Congress to develop whatever facts we can which may be of assistance to you. This is one of the basic functions of the Federal Trade Commission as he sees it.

Today we are extending the analysis and shall (1) place the current merger movement in its historical perspective; (2) discuss the relationship between merger cycles and overall business conditions in order to see whether we can, on the basis of past experience, anticipate in broad compass the prospects for the contínuance of the current merger movement; and (3) analyze and describe the profile of so-called big mergers, that is, mergers involving the companies within the ranks of the about 2,000 manufacturing companies which have assets of $10 million or more and which together control over 80 percent of all manufacturing assets.

I shall first attempt to place the current merger movement in historical perspective and discuss certain general economic forces which have been common to all American merger movements.


Students of industrial organization generally agree that there have been three rather distinct merger movements in American industrial history. The first merger movement occurred around the turn of the century, and left a permanent imprint on the structure of American manufacturing. For the 7-year period, 1897–1903, Dr. Ralph L. Nelson recorded 2,864 mergers. Measured against the economy of 1900—the 1899 census put the value of shipments of all manufacturing companies at $11 billion, or less than the combined sales of the three largest industrial companies of 1964—the great merger movement was indeed massive in its effects. Indeed, according to one study, by 1909, when the merger movement was about spent, and prior to the breakup of the petroleum, tobacco, and explosives trusts, the 100 largest industrial corporations controlled about 17.7 percent of the assets of all industrial corporations.3

The second merger movement commenced after World War I and reached what Dr. Willard L. Thorp described as the "hysteria stage" during the late 1920's. In the 6-year period, 1925–30, 4,682 mergers were recorded. At the peak of the movement, in 1929, Dr. Thorp recorded 1,245 mergers.

The third merger movement has been underway since the closing years of World War II. The trends in the current merger movement are depicted in chart 1, based on the Federal Trade Commission's own merger compilations. We have divided the chart into three grids. The upper grid shows the trend based on “total” mergers in manufacturing and mining as reported in two financial reports. The middle and lower grids show trends in “large” mergers for the period 1948–64. By "large" we mean acquisitions of companies whose assets at the time they were acquired amounted to $10 million or more. This “arge” merger series is shown both in terms of the number of com" panies acquired (the middle grid) and the aggregate assets of the acquired firms (the lower grid).

Referring to the top grid on the chart, we note that the current merger movement has occurred in two rather distinct phases or cycles.

1 See particularly George W. Stocking,_"Comment on Mergers," Business Concentration and Price Policy, National Bureau of Economic Research, Princeton University Press, 1955, pp. 199 ff.

2 Ralph L. Nelson, "Merger Movements in American Industry," 1895-1956, Princeton University Press, 1959, p. 37. Nelson's merger series covering the years 1895–1920 is shown herein in appendix table 1.

* Collins and Preston, "The_Size Structure of the Largest Industrial Firms, 1909-58," American Economic Review, December 1961, 989. "Industrials” in that study included manufacturing, mining, and distribution.

. This series consists of mergers reported in Standard Corporation Records and Moody's Industrial Manual. Many additional mergers are reported in other sources ; however, chart 1 includes only those mergers reported in these two sources in order to maintain comparability for purposes of determining the merger "trend."


1940 - 1964 Number 1,000


t 600







[blocks in formation]

0. 1940





*Large Firms represent companies with assets of $10 million and over.

The first took place during the period 1943-47 and had an especially pronounced impact on the structure of particular industries, that is, textiles and distilling.5

Following 1947 merger activity subsided, reaching a post-war low in 1949. Beginning in 1950, however, the second major cycle commenced and the upward surge in merger activity is still continuing; Whereas at the peak of the first cycle of mergers the FTC's total manufacturing and mining series indicated a total of slightly more than 400 mergers per year, in each of the past 10 years, 1955–64, more than 600 mergers have been recorded—the lates figure being 854. (See appendix table 1.)

I would like at this time to emphasize the fact that what we are talking about here is a series confined to just two sources of mergers, and to a series confined to mergers involving manufacturing and mining concerns. Last year the Federal Trade Commission recorded something like 1,700 mergers from all sources that we consulted, which include several leading financial papers plus the 2 that are shown in appendix table 1. Of course, our merger statistics do not include banking mergers, over which the Federal Trade Commission has no authority.

The Commission's “arge” merger series for the period 1948–64 parallels the general upward movement displayed by the "total" merger series—both in terms of numbers and aggregate assets acquired.

It should be pointed out that the series for so-called total mergers in manufacturing and mining has a number of serious shortcomings. While the series may reflect fairly accurately the overall trend in mergers, it is not a precise gage of the absolute level of merger activity, in fact far from it. This is the case because the series is based solely on manufacturing and mining mergers mentioned in two financial reports. If additional sources were consulted, for example, trade journals in each manufacturing industry, the number of mergers recorded would increase with each source consulted. For example, we have data on dairy mergers based on a nearly complete record of acquisitions by the top eight dairy companies, as well as most of the significant mergers among smaller dairies. In some years, that is, the late 1930's, we find that the total number of recorded dairy mergers actually exceeded the "total" number of recorded manufacturing and mining mergers, although the dairy industry accounts for less than 3 percent of total sales in manufacturing. In 1941 the 8 largest, alone, acquired 112 dairies, whereas the FTC series recorded 11i manufacturing and mining mergers in that year (appendix tables 1 and 2). This simply reflects the fact that many mergers by both large and small dairy companies, and this is true in other industries as well, of course, are not reported in the leading financial manuals and journals, and many are never reported in any source.

Again, I would like to emphasize that in a good many cases the Federal Trade Commission in the course of investigations of particular acquisitions discovers that companies have made additional acquisitions which had never come to our attention from any other source.

Despite the admitted shortcoming in our overall series, however, it does picture quite well the waves and tides of merger activity, if not

5 Report of the Federal Trade Commission on the Merger Movement, 1949. & See footnote 3, appendix table 1.

its precise depth. As such, our series indicates that once more the merger tide is running high.

Before elaborating on some of the key characteristics of the current merger movement, let us review certain cyclical characteristics of merger activity in American industry.


The rate of merger activity appears to be rather closely associated with overall business activity, or so-called business cycles. This is to say, merger activity is positively correlated with good times in the business community. As a result, there are year-to-year variations in the rate of merger activity even within particular merger movements.?

For example, the merger movement in the 1920's was choked off by the depression following 1929, and the first cycle of the current movement ended with the recession of 1948-49. On the other hand, the 13-percent increase in industrial production and 33-percent & increase in industrial stock prices (appendix table 3) between 1954 and 1955 were associated with a rapid rise in merger activity.

I should emphasize at this point that the increase in stock prices between 1954 and 1955, following as they did the substantial increase bet ween 1953 and 1954, represented one of the most dramatic increases in American history.

Mergers continued at a high rate into 1956 as industrial production rose another 4 percent. But with the slackening of business activity in 1957 and 1958, “total” merger activity fell off slightly. However, the Commission's "large” merger series showed a sharp decline. Total merger activity revived once again with the expansion of 1959 and has since continued at a high plateau.

The association between merger activity and business cycles appears to cut across all segments of industry. This is true, for example, in the dairy industry for the entire period, 1919–64 (appendix table 2). The dairy merger series is closely correlated with the Thorp-FTC merger series for the entire period since 1919.

The so-called Nelson, Thorp, and FTC series appear in appendix 1. I would like to turn briefly to the dairy industry to make a few comments on this relationship between business activity and conditions in the stock market, and relative merger activity. The dairy industry: A case study in merger cycles

To my knowledge, all prior statistical analyses of merger cycles have been confined to overall merger activity. Analysis of merger cycles in a particular industry usually is not possible because there are not enough companies to participate fully in more than a single merger movement. An extreme example is steel. After United States Steel was formed in 1901 as a "combination of combinations," it manufactured well over half of the country's steel products. This vast consolidation limited severely the extent to which the steel industry could participate in subsequent merger cycles. Similarly, after acquiring

* Nelson has found that merger cycles are quite regularly related to business cycles. Eleren of the twelve identifiable merger "cycles" occurring during the period 1895–1954 showed a definite time relationship to the turning points of reference cycle (Nelson, op. cit., pp. 110-111). His statistical analysis of merger activity and industrial stock prices for the period yields a simple correlation coefficient of 0.469, which is statistically significant at the 5 percent level. Ibid., p. 118.

* Based on Dow Jones industrials. Standard and Poor's industrial average rose by 40 percent between 1954 and 1955.

« PreviousContinue »