Theory of the Growth of the Firm
Page 34
Although in a rapidly expanding economy, early acquisition and merger may quickly enable the larger firms to achieve near-monopoly positions in many industries, it is a game that two can play, and if the market is expanding rapidly enough, the merger of smaller firms or later arrivals may enable them quickly to overcome the handicaps of small size, and through merger to challenge the dominant position of the larger earlier-established firms. If merger is a ‘cause’ of early dominance, it is sometimes equally a ‘cause’ of a subsequent weakening of that dominance.236 But the mere fact that merger raises the maximum rate of growth of the individual firms means that some firms will be larger than they would have been in the absence of merger, and in all probability larger also in relation to the economy as a whole.
The occasion for profitable acquisition exists under numerous and varied circumstances and it therefore is not surprising that extensive acquisition should mark industrial history. When the rate of expansion is straining the productive services available to the large firms, while at the same time extensive opportunities for profitable expansion are to be seen on all sides by the enterprising entrepreneur, acquisition and merger will appear particularly attractive; and not only attractive but at times ‘necessary’ to a particular firm as a safeguard for its long-run position whenever other firms are observed to be taking advantage of opportunities so closely related to that firm’s own activities that they appear as a competitive threat. In addition, acquisition will be attractive if continued expansion appears likely to be hindered by bottlenecks in the flow of supplies, and integration by merger the most appropriate safeguard. Acquisition will also tend to be unusually prevalent when economic growth is stimulated by important innovations or discoveries that are not in their inception substantially within the preserve of the larger firms. Under these circumstances there may be a scramble of old and new firms to gain a footing in the new fields of activity. If entry is easy there will be many failures, the environment will be favourable to merger, and the few who survive the initial struggle will often gain their advantage in this way.
Merger in Relation to Indices of Business Activity
Economists have not been particularly satisfied with a heterogeneous list of ‘causes’ to explain the occurrence of merger and acquisition, and have tried to discover whether there are not more persuasive considerations giving rise to ‘waves’ of mergers. Interest has centred particularly on the business cycle and on the levels of security prices, but the results of attempts to correlate the number of mergers with other indices of business behaviour have not been very successful, for the differences between ‘merger movements’ seem to be more significant than the similarities.237 The same factors seem to have different effects under different conditions. For example, high security prices are said to have encouraged merger in the 1920s because they made outside capital relatively cheap, while low security prices are supposed to have done so in recent years because the assets of many existing firms were undervalued at ruling prices.238
The failure of the attempts to correlate the number of mergers with other indices of economic change is not surprising, partly because of the variety of more or less independent economic and institutional factors influencing merger, many of which were discussed in Chapter VIII, and partly because the usual indices of merger activity do not necessarily reflect changes in the strength of the inducement to expand through acquisition. Merger is one of two possible methods of expansion, and whether or not it is the method chosen will depend on its relative profitability. The number of mergers in a given period, or even the ratio of the number of mergers to the number of existing firms, will tell us nothing about the ratio of expansion through merger to expansion through new capital formation; yet this is the information we require if the purpose of comparing merger activity to other indices of business behaviour is to discover the conditions under which merger is promoted.
If the number of mergers rises in a period, for example, but expansion through new investment rises even faster, presumably the existing constellation of circumstances, while favourable for expansion of capital investment, is not particularly conducive to the choice of merger as the method of expansion. And whether high or low security prices are conducive to merger can only be tested by finding out whether the proportion of merger to total expansion varies with the level of security prices.
It has been objected that although merger and internal expansion may be alternative uses of funds from the point of view of the individual firm, merger does not, from the point of view of the economy as a whole, absorb real savings. Hence in prosperity real investment must be high by definition. Only in depression could acquisition be the dominant form of expansion in the economy as a whole.239 While it is true that the condition of prosperity is a large amount of real investment, and depression, by definition, is a deficiency of real investment, the relevant question is whether or not acquisition forms a larger percentage of total expansion at times when real investment is high than it does when investment is low. The answer to this is not a priori clear; yet to know whether prosperity or depression is relatively more favourable to expansion by acquisition than to internal expansion, we would have to examine this percentage in the two conditions. The difficulties of constructing an index of the importance of merger that takes account of its role in total capital expansion would be considerable and the results might not justify the effort. But in the absence of such a measure, it is hard to see what purpose is served by correlating merger activity with security prices, business activity, or other measures of cyclical or secular change.
The Effect on the Interstices of ‘Natural’ Limitations on Acquisition
In general the inducements to merger are so numerous and so pervasive that the appropriate question perhaps is not, why so much merger? but rather, why not more merger? Except where owners are stubbornly independent in the face of attractive offers for their firms, why should not small firms with bright prospects always have opportunities to sell out on especially favourable terms? Why should they ever grow large in the interstices of the economy?
We have already touched on one type of explanation—the limits on the ability of any one firm to absorb others in a given period of time. In periods of very rapid growth of the economy, therefore, many smaller firms may find no other firm actively seeking to buy them. But there is another, perhaps even more relevant, explanation that is also implied by our previous analysis. It is obvious that for acquisition to be the preferred method of expansion there must be ‘suitable’ firms available, that is to say, firms that fit the expansion plans of the acquiring firms. Yet there are many types of expansion for which suitable firms are unlikely to exist; for example, the type which is stimulated by the prospects of introducing new products or of using new technology developed by the expanding firm. Or when the internal inducement to expand arises partly from special skill in the organization and operation of particular processes in production, internal development may have overwhelming advantages over acquisition. Large firms maintaining their own research and development laboratories may be expected to develop new processes and products the best use of which can often be made only if the firm itself undertakes both the creation of new factories and the promotion of its products. So far as the large firms are themselves the source of much of the new technology introduced into the economy, this alone will decrease their desire to expand through acquisition. Similarly the introduction of new industries into new geographical areas may require the building of new plants.
For these and allied reasons, internal expansion may often be more profitable than acquisition for individual firms, and this ‘natural’ restriction on the advantages of expanding through acquisition, together with the general limit on the rate of expansion of the firm may, if the large firms are very large, reduce the extent to which the small firms growing into the interstices are absorbed by larger firms. It is true that in the United States economy at the present time, large firms make more acquisitions than do smaller f
irms,240but at the same time acquisition is a less important source of growth for large firms than for small firms.241 This means that acquisition promotes the growth of those small firms that make acquisitions more than it does the growth of large firms making acquisitions, and thus promotes an increase in the number of medium-size and larger firms, which may improve the competitive position of these firms vis-á-vis the older large firms. On the other hand, there can be no question that acquisition facilitates the diversification of the larger firms and in this respect it may partly offset the tendency of an expanding, changing economy to create opportunities for smaller firms.
Interstices and the Business Cycle
Before pulling together the analysis of this and the two previous chapters into a general theory of the process of industrial concentration, we should briefly examine some of the more important effects of cyclical fluctuations in industrial activity on the relative positions of small and large firms. Severe downswings in industrial activity obviously reduce the interstices in the economy chiefly because they reduce the opportunities for expansion of all firms; the question is whether small firms are relatively worse off in comparison with large firms.
One of the significant characteristics of cyclical fluctuations is the interaction between actual and expected changes in economic activity: a change in one direction tends to cause anticipations of further change in the same direction which may themselves actually induce the change anticipated. Thus, in spite of idle resources during a depression, and in spite of the fact that widespread efforts to use these resources would, within limits, create the income which itself would justify the effort, firms in general do not make it because of unfavourable expectations on the part of each individual firm. The cumulative effect is to reduce both anticipated and actual opportunities for profitable investment.
If the larger firms have competitive advantages over smaller firms, one would expect that the impact of depression on them would be lighter even if their competitive advantage showed up only in relatively fewer failures. As a group they have greater financial strength; individual large firms may have more control over prices, and if the income elasticity of demand for their products is low enough, they may be able to maintain earnings at a level higher than would have been possible under more competitive conditions. In addition, large firms often plan further ahead than do smaller firms, partly because their greater financial strength enables them to afford it, and partly because the nature of their operations more or less forces them to. For this reason, depression is sometimes looked on as a good time to expand: costs are low, plant can be constructed and equipment bought cheaply. All that is necessary to justify such action is the ability to finance the project and an anticipation that demand will pick up in the not-too-distant future. Needed improvements in organization and productive processes may be undertaken which were not profitable at peak levels of activity. Depressed economic conditions unquestionably favour large firms as a group in comparison with small firms, and often permit individual large firms to consolidate their position and further improve their competitive strength.
However, all of the advantages of the larger firms are equally available to medium-sized firms with adequate finances, and although there are probably more outright failures among the smaller firms, individual medium-sized and small firms may well come out of the depression in equally good shaped. 242 To the extent that these smaller firms have lower overheads and are more able to reduce costs as output is reduced, they may even incur proportionately lower losses than many of the larger firms. Consequently, although the depression may permit larger firms as a group to improve their position, and may eliminate many small firms, there will, when the upswing comes, still be numerous smaller and medium-sized firms in a position to expand as the interstices widen.
The upswing, especially if sustained for a considerable period, is particularly favourable for the medium-sized and smaller firms; in some respects they may be in a better position to expand rapidly than the largest firms, for, as we have suggested above, the moderately large firms are likely to have some advantages in expansion over the largest ones. The interstices may increase appreciably and a long period of sustained activity may then see an improvement in the relative positions of smaller and medium-sized firms. When monetary policy becomes exceedingly stringent, however, severe credit rationing may reduce the expansion of small firms.243
The Process of Industrial Concentration
An analysis of the process of industrial concentration is an analysis of the forces determining changes in the relative importance of firms of different sizes in an economy or an industry. Firms may be looked on as employers of labour, owners of assets, or producers of goods, and a measure of their relative importance in each of these respects can be used as an index of the state of concentration at any given time. Plainly, the significance of any index of concentration depends upon the measure used, and a brief discussion of this problem is in order.
Measurement of Concentration
There are several ways of going about the measurement of concentation, and there is no general agreement on the best measure, or even the most useful concept. One may be concerned with inequality in the size distribution of firms (sometimes called ‘relative concentration’), or with the extent to which the quantity being measured is concentrated in the hands of a few firms (sometimes called ‘absolute’ concentration). Though changes in one may affect the other, inequality and absolute concentration are distinct concepts with a different significance and different uses in analysis.
Absolute concentration may be high, for example, and inequality very low, for an index of absolute concentration indicates nothing about the size distribution of firms within the group of largest firms nor within the remaining group of smaller firms. If an entire industry contained two firms only, the industry would be highly concentrated in the former sense, but if the two firms were of equal size an index of relative concentration (e.g., a Lorenz curve) would show perfect equality. Conversely inequality is not indicative of absolute concentration. Inequality may decrease while absolute concentration increases when the smaller firms leave an industry; extensive merger may reduce inequality but cannot reduce absolute concentration.244 1 am not here concerned with inequality, but only with ‘absolute’ concentration, which I shall call simply ‘concentration’, for our analysis of the growth of firms is most directly relevant to the question how and to what extent economic activity becomes and will remain concentrated in the hands of a few large firms.
The usual indexes of concentration relate either to some measure of assets or output, or to employment, and are most generally expressed in terms of an index of the percentage of total activity in the hands of a stated (few) number of firms or of the number of firms required to account for some stated (large) percentage of total activity.245 The basic purpose of attempts to estimate concentration is to discover whether a ‘few’ firms control a ‘large’ percentage of output, employment, or assets, presumably because this information is considered relevant for an appraisal of economic performance in some significant sense.
Each of the various economic quantities with respect to which concentration can be measured will yield results with a different significance, and the type of measure adopted will depend partly on the kind of problem one is interested in and partly on the type of data available. In any attempt to measure concentration a number of judgments have to be made, but we shall not be concerned with these here, for an analysis of the process of concentration does not require that we decide what level of concentration is significant for different purposes or even what economic quantities are relevant for the measurement of concentration. Nor is it practicable to attempt to take account of all of the qualifications required to adapt the analysis to each of the possible ways of measuring concentration. These questions have been the subject of much debate; 246 I mention them only in order to indicate some of the simplifications of the following brief discussion, and I shall not deal further wit
h any of the statistical or conceptual problems involved in measuring concentration or interpreting the results. Fortunately, over broad areas the various measures seem to be reasonably consistent, at least with regard to the direction of change, though not always with regard to the ‘level’ of concentration—concentration measured with respect to assets, for example, runs significantly higher than concentration measured with respect to employment or output, 247 which, as we shall see, is not surprising.
Concentration and Growing Firms in a Growing Economy
The emergence, extent, and persistence of industrial concentration, whether in the economy as a whole or in individual industries, is clearly a function of the relative rates of growth of the small and large firms, for the amount of growth that takes place in an economy (or industry) in a given period of time must be equal to the growth of all firms (including new firms) in the economy (or industry) in the period. With any given rate of growth of the economy (no matter how measured), a point will be reached at which a higher rate of growth of some firms will be inconsistent with the continued growth of some other firms, and finally with their continued existence. The total possible amount of expansion of an economy is not unlimited at any one time, and when the absolute growth of a few firms becomes greater than the absolute growth of the economy as a whole there must be a decline and eventual elimination of some other firms.