American Journal of Economics and Sociology, The: The trend of aggregate concentration in the United States

FindArticles > American Journal of Economics and Sociology, The > April, 1996 > Article > Print friendly

The trend of aggregate concentration in the United States

Patrick B. O'Neill

Problems of Scope and Measurement

Introduction

A sizable literature examining aggregate concentration and describing the potential adverse consequences of increasing aggregate concentration exists.(1) These adverse consequences involve the suspected relationship between increasing concentration and increasing corporate control over economic resources resulting in non-competitive behavior and/or enhanced political power.(2)

Caswell (1987) describes the potential for non-competitive behavior in terms of two linkages between aggregate concentration and economic performance. The first link concerns the relationship between aggregate concentration levels and concentration in individual markets. She states: "To the extent that aggregate concentration levels reflect concentration in individual markets, aggregate concentration measures act as an overall proxy for the level of competition in the entire economy."(3) The second link "is through the dynamic impact of the presence of very large, conglomerate firms on market structure." Activities such as reciprocity or mutual forbearance are dangers commonly thought to be associated with conglomerate firms.(4) The relationship between aggregate concentration and political power is that a danger exists that increases in aggregate concentration lead to the control of more resources that might be used to exert undue influence on the political process.

The data presented in the literature allows for a number of conclusions to be drawn concerning the trend in aggregate concentration.(5) First, during the 1970s, aggregate concentration in numerous sectors of the economy was either relatively constant (White, 1981) or declining (Shepherd, 1982). Second, Shepherd (1982) asserts that the main reason for declining aggregate concentration in the 1970s was the vigorous antitrust enforcement during this period. Enforcement of antitrust laws is a key element in controlling aggregate concentration since the lack of such enforcement is often cited as a primary underlying cause of increasing aggregate concentration. This is stated explicitly by Mueller (1990). He argues that there "exists a causal relationship between the growing merger-achieved centralization of control over American industry and the competitive structure and behavior found in particular markets."(6)

This leads to an obvious question: Given the lack of antitrust activity and the associated merger boom during the 1980s, has aggregate concentration risen? Several recent studies present mixed results. Greer (1988) shows that, for the early 1980s, there is little change in aggregate concentration in manufacturing but an increase in aggregate concentration in commercial banking. Caswell (1987) presents evidence indicating an increase in aggregate concentration in agribusiness between 1976-1986 due to mergers. O'Neill (1991) shows that aggregate concentration has declined slightly in the manufacturing sector during the 1980s.

Contributing to the mixed results of these studies is that there is no general agreement as to how aggregate concentration should be assessed. As Caswell (1987) points out, there are two interrelated problems. The first problem concerns the question: What level of the economy should be explored when measuring "aggregate" concentration? In other words, which data should be examined? Often the analysis is limited to the manufacturing sector (as is the case in Shepherd (1982), Attaran and Saghafi (1988), and O'Neill (1991)). Greer (1988) examines what he refers to as "macro aggregation," but his analysis is also limited to data from the manufacturing sector. The second problem concerns the question: What is the proper measure of aggregate concentration? These problems are addressed in more detail in the next section of the paper.

This paper expands upon the previous literature in two respects. First, the issue of the appropriate manner of assessing aggregate concentration is dealt with by presenting a broad assortment of data.(7) Second, the paper presents an updated look at aggregate concentration in the US economy by examining the period 1978 to 1987.(8) (Data from previous studies are also reported to provide a better picture of the long-run concentration trend.)

The remainder of the paper proceeds as follows. Two sections deal with aggregate concentration at the sector-wide level, the first covering manufacturing and the second examining various non-manufacturing sectors. Next, aggregate concentration on an economy-wide basis is explored. Finally, some concluding comments are presented.

A Plethora of Measures

As mentioned above, there does not exist any generally accepted "correct" way to assess aggregate concentration. Consequently, before examining the aggregate concentration data, some of the issues involved in choosing a measure are discussed.

White (1981) suggests that the commonly used measures showing the shares of sales or of assets attributed to the 50, 100 or 200 largest firms in any given year are necessarily flawed due to double counting problems. Aggregate sales data involve "obvious" double counting whereas aggregate asset data suffer from double counting problems "since the assets of the financial sectors are liabilities of the other sectors, which have been used to finance the assets of those sectors." White therefore suggests the use of value added figures as more appropriate. However Brozen (1982) maintains that the use of value added data "tends to mislead as well as enlighten" since increases in aggregate concentration measures (constructed using value added data) may represent an improvement, rather than a worsening, of competitive conditions. He argues that, when using value added data, any observed increase in aggregate concentration may simply result from firms that are vertically integrated entering the ranks of the largest (50, 100, 200) firms replacing non-integrated firms. His point is that an observed increase in concentration may be nothing to worry about as it is quite possibly a reflection of increased efficiency.

The issue of mobility among the largest firms is addressed by Clarke (1985). He suggests that, instead of looking at changing concentration ratios within a sector of the economy, a more general approach would be to "look at trends in concentration by comparing average concentration ratios in constant samples of industries over time." This method is intuitively appealing since it captures some of the entry and exit behavior that occurs over time (increasing concentration denoting exit and vice versa). However, this approach has one major drawback: the time path of aggregate concentration necessarily depends upon the (arbitrary) choice of the base year set of firms. This drawback has been illustrated by Bock (1979) who examined the share of total assets attributable to the 200 largest manufacturing firms between 1947 and 1972. When using 1947 as the base year, Bock found that the 200 largest manufacturing firms' share of total assets increased by 3 percentage points between 1947 and 1972. However, when 1972 was used as the base year, the 200 largest firms' share increased by 20 percentage points during this same time period.

Attaran and Saghafi (1988) argue for the use of an entropy index of concentration. The entropy index belongs to the same general family of indices as the four and eight firm concentration ratios and the Herfindahl index; all are weighted market share indices. In the entropy index, the weights are the negative of the log of the firm's market share. However, unlike the concentration and Herfindahl measures, increases in the entropy index signal a decrease in concentration rather than an increase in concentration. Thus the entropy index is commonly thought of as an index of diversification (i.e., an inverse index of concentration).(9) Unfortunately the entropy index suffers from the rather severe limitation that data for all firms in the industry (or sector of the economy) are required for its proper computation. Constructing the entropy index without a complete data set can result in incorrect index values.(10)

Given the existence of this disagreement in the literature, how should aggregate concentration be examined? Caswell (1987) furnishes a pragmatic guiding principle: "Since the alternative measures and bases of economic activity all have drawbacks, a broad set of measures using various data sources is likely to provide the most reliable indication of the level and trends of aggregate concentration." Following Caswell's principle, examined below are a variety of aggregate concentration measures for the manufacturing sector, various non-manufacturing sectors, and the economy as a whole. These measures, taken together, provide a broad overview of the trend of aggregate concentration in the United States. An explanation for the choice of the specific measures follows.

In examining the manufacturing sector, both White (1981) and Caswell (1987) make the claim that the Census of Manufactures data are the "best." The reason, as explained by White, is that these data include domestic operations only (and thus avoid any bias that might occur with the inclusion of foreign data).(11) The various non-manufacturing sectors of the economy examined were chosen so as to be able to update the work presented in White (1981).(12) To examine aggregate concentration at the economy-wide level, a set of measures that are typically reported for aggregate concentration in the entire economy are included.(13)

III

Aggregate Concentration in Manufacturing

The concentration data for the manufacturing sector appear in Table 1 below.

When examining aggregate concentration in manufacturing (using Bureau of the Census data), Greer (1988) states "Although the data are not yet available [TABULAR DATA FOR TABLE 1 OMITTED] for 1984-1987, if seems safe to suppose that the merger movement of the mid-1980s has substantially fostered further aggregate concentration." These data are now available.

The value added data do not tell a consistent story. They indicate a slight rise in aggregate concentration for the largest 50 firms between 1977 and 1987, but no change in concentration for the largest 100 and largest 200 firm size groups. The employment data (for all size categories) reveal a definite trend: a decrease in concentration during the 1977 to 1987 period.

Taken together, these data suggest that the increase in concentration feared by Greer (1988) has not occurred, despite the merger wave of the mid-1980s. A plausible explanation for this non-increase in concentration is that the merger wave of the mid-1980s was counteracted by a divestiture wave as large conglomerates unloaded unprofitable acquisitions. Weston (1989) reports that "In the 1980s, some 35 to 40 percent of the mergers and acquisitions reported annually by W. T. Grimm were divestitures by other firms."

Aggregate Concentration in Non-Manufacturing

The concentration data for various non-manufacturing sectors appear in Table 2. An analysis of each sector follows.

The Banking Sector.(14) The concentration data for the banking sector appear in columns 2 and 3 of Table 2. The asset data indicate a significant increase in concentration between 1978 and 1987. This is likely due to the merger activity within the banking sector throughout the 1980s. Amel and Jacowski (1989) indicate that "Over the period from 1977 to 1987 . . . while the share of banking assets controlled by the largest banking organizations has increased sharply . . . the number of bank mergers and acquisitions has soared."(15)

In contrast to the trend in the asset data, the deposit share data show a decline in concentration between 1978 and 1982. This decline in concentration is an indication of the increasing importance of nonbank financial intermediaries. Amel and Jacowski (1989) state that "since the mid-1970s, assets held by domestic banking organizations have declined relative to those of other financial institutions."(16)

The trend of these two data series imply different conclusions concerning aggregate concentration in the banking sector. However, Amel and Jacowski (1989) appear unconcerned as they suggest competition from outside the banking sector may be sufficient to mitigate concern over the rise in concentration within the banking sector.

The Life Insurance Sector. The concentration data for the life insurance sector appear in columns 4 and 5 of Table 2. There is a clear decline in concentration in this sector between 1978 and 1987. This decline is evident in both the assets and insurance in force data, although it is more pronounced in the insurance in force data. Gart (1994) describes two changes within the life insurance industry that offer some explanation for this decline in concentration. First, life insurance companies became more diversified in the 1970s, with many companies entering the brokerage business. Secondly, an increase in competition from outside the industry has occurred as banks and securities firms have begun selling insurance.

The Electric and Gas Utilities Sector. The concentration data for the electric and gas utilities sector appear in columns 6 and 7 of Table 2. During the 1978 to 1987 period, both the assets and net income after tax data series indicate the same pattern, increasing concentration. This increase in concentration is probably being driven by the tremendous amount of merger activity that occurred among natural gas pipeline companies following deregulation of the industry with the passage of the Natural Gas Policy Act of 1978. Cramer (1989), commenting on [TABULAR DATA FOR TABLE 2 OMITTED] the merger activity in the deregulatory era, states that between October of 1982 and July of 1986 "there have been 23 major mergers or acquisitions involving major natural gas pipelines."(17)

The Retail Sector. The concentration data for the retail sector appear in columns 8 and 9 of Table 2. Both the revenues and employment data series indicate increasing concentration between 1978 and 1987. Increasing concentration among food retailers provides a (partial) explanation for this trend.(18) Cotterill (1993) provides details of a merger wave among food retailers that began in 1979 and continued throughout the 1980s.(19)

The Transportation Sector. The concentration data for the transportation sector appear in columns 10 and 11 of Table 2. The revenue share data indicate substantial variability within the 1978 to 1987 period, decreasing concentration in the earlier years, and increasing concentration since 1983. The employment share data also have substantial variability, but a trend toward a modest increase of concentration is discernable. Increases in concentration in the airline, trucking and railroad industries that occurred following deregulation are the likely underlying cause of the trend in these data.(20)

Morrison (1993) states that, following deregulation of in the airline industry "until early 1985, the industry became less concentrated. But from 1986 until mid-1988, the industry became much more concentrated because of the mergers that took place during that period."(21)

Boyer (1993) describes the post-deregulation activity in the trucking industry:

Unlike the airline industry, the for-hire trucking industry has proceeded to concentrate not through mergers but rather through the expansion of territory of individual carriers; expansion has been accomplished through the construction of new terminals and freight sales offices or the purchase of facilities of bankrupt carriers."(22)

Braeutigan (1993), in discussing the impact of deregulation on the railroad industry, states that the number of Class I railroads declined from 73 in 1975 to 16 in 1988.(23)

To sum up, the data for 1978 to 1987 from the non-manufacturing sector contain both good and bad news. The good news is that there is a trend toward increased competition between the financial sectors (banking and life insurance). The bad news is that the data indicate increasing concentration in the other three sectors: electric and gas utilities, retail trade and transportation.

Aggregate Concentration in the Entire Economy

Data for aggregate concentration in the entire economy appear in Tables 3-5. The data in Table 3 show concentration measured in terms of employment.

[TABULAR DATA FOR TABLE 3 OMITTED]

For the largest 1300 firms, the data show a very slight drop in concentration between 1978 and 1981. For the largest 1000 firms, the data indicate a decrease in concentration between 1982 and 1987. Together, these data suggest an overall decline in aggregate concentration between 1978 and 1987.

Table 4 shows concentration measured by asset shares of the largest nonfinancial corporations. The pattern of concentration among each size category examined (i.e., the largest 50, 100, 150 and 200 nonfinancial firms) is essentially the same - from 1978 to 1987 there is a slight increase in concentration.

Table 5 shows concentration measured in terms of sales of the largest 500 firms. The data show only minor variability between 1978 to 1984, but indicate a decline in concentration between 1984 and 1987.

Taken together, the data from Tables 3-5 do not present a unified view of the economy-wide trend in aggregate concentration between 1978 and 1987 as two of the three measures suggest declining concentration while the third measure indicates the opposite trend. However, if one follows White (1981) and puts more faith in the employment share data than in any other economy-wide measure, it is possible to conclude that a decline in overall aggregate concentration has occurred between 1978 and 1987.

[TABULAR DATA FOR TABLE 4 OMITTED]

[TABULAR DATA FOR TABLE 5 OMITTED]

Summary and Concluding Remarks

This paper has presented data on aggregate concentration between 1978 and 1987 in order to explore how the lack of antitrust activity and the associated merger boom during the 1980s has affected the trend of aggregate concentration. The central question addressed is: Has a major increase in aggregate concentration, (as implicitly warned of by Shepherd (1982) and explicitly predicted by Greet (1988)), actually occurred? The answer depends upon which sector of the economy is examined.

When the focus is the typical one (the manufacturing sector) the answer is no.(24) However, when the focus of the analysis is extended to include other sectors of the economy, a simple answer to this question does not exist. The data for the non-manufacturing sectors indicate that some sectors have indeed experienced rising concentration (the electric and gas utilities, retail trade, and transportation sectors), one sector has experienced a decline in concentration (the life insurance sector) and another sector (the banking sector) has conflicting data. In examining the data on economy-wide aggregate concentration, the majority of the evidence suggests that overall aggregate concentration has declined. Thus the merger boom and lax antitrust enforcement during the 1980s appears to have resulted in rising concentration in some, but not all, sectors of the economy.

Can any policy conclusions be drawn from the preceding analysis? In looking at those sectors in which aggregate concentration has been rising, much of the increase in concentration has resulted from increased merger activity following deregulation. This suggests that industries wherein there is currently a move toward deregulation are likely candidates for increased concentration in the future. To the extent that aggregate concentration is a problem, the potential for rising concentration following deregulation should be kept in mind during the coming months (years) as the details concerning deregulation of the electric utility, telecommunications, cable television, and banking industries are worked out.

Notes

1. The seminal work in this area is Berle and Means (1939). More recent investigations include White (1981), Shepherd (1982), Weiss (1983), Attaran and Saghafi (1988), and Greet (1988).

2. For treatments of the issue of economics and power (broadly defined) see Rhoades (1983) and Peterson (1988).

3. The theoretical underpinnings of this link have been explored by Clarke and Davies (1983). They derive an index which shows that aggregate concentration is proportional to a weighted sum of concentration in individual markets, where the factor of proportionality is an index of diversification proposed by Berry. The details of this link follow.

To start, define Herfindahl indices of aggregate concentration ([H.sub.A]) and market concentration within industry j ([H.sub.j]):

[H.sub.A] = [summation of] [[S.sub.i].sup.2]/[S.sup.2] where i=1 to N [H.sub.j] = [summation of] [[S.sub.ij].sup.2]/[[S.sub.j].sup.2] where i=1 to N

where:

N is the number of independent firms in the economy,

[S.sub.i] is the total size of firm i,

[Mathematical Expression Omitted], is the aggregate size of the economy,

[S.sub.ij] is the size of firm i in industry j,

[S.sub.j] = [summation of] [S.sub.ij] where i=1 to N, is the size of industry j.

Now define Berry's index of firm level diversification ([D.sub.i]):

[D.sub.i] = 1 - [summation of] [[S.sub.ij].sup.2]/[[S.sub.i].sup.2] where j=1 to K

where:

K is the number of industries (or markets) in the economy.

Combining all of the pieces above allows for [H.sub.A] to be written as a relationship between [H.sub.j] and [D.sub.i] as follows:

[Mathematical Expression Omitted]

where

[w.sub.j] = [[S.sub.j].sup.2]/[S.sup.2],

[Mathematical Expression Omitted], is an aggregate index of diversification,

[Mathematical Expression Omitted].

The meaning of the [H.sub.A] index is summarized by Clark and Davies as follows:

"Measuring concentration by the Herfindahl index and diversification at the firm level by Berry's [D.sub.i] index, the level of concentration in a given sector (when that sector might be an aggregate economy), depends positively on the weighted sum of concentration in the constituent industries of that sector, and ceteris paribus positively on an aggregate index of diversification. The latter is a weighted average of diversification at the firm level, with especially large weights attributable to the very large firms."

4. Reciprocity refers to a situation where firm A agrees to buy from firm B provided that firm B agrees to buy from firm A. Mutual forbearance concerns a situation in which firm A agrees not to compete with firm B in a particular market provided that firm B agrees not to compete with firm A in some other market. For a more detailed discussion, see Greer (1992).

5. The majority of the aggregate concentration literature deals with the link between concentration in individual markets and aggregate concentration. The evidence concerning the problems of reciprocity and mutual forbearance is largely anecdotal (see Adams and Brock 1990), although in an empirical test Alexander (1985) finds some (limited) support for this hypothesis. Concerning the potential for increased political power, the evidence is also largely anecdotal (see Petr, 1988). For empirical tests of this issue, see Salamon & Siegfried (1977) and Mann & McCormick (1980).

6. Greer (1988) and Caswell (1987) express similar views.

7. The data series examined here do not constitute an all-inclusive set of measures. Rather, they are a set of measures selected so as to facilitate comparability with past studies of aggregate concentration.

8. Unfortunately, not all of the data series discussed below are available for this entire period. To maximize comparability with the prior literature, data series are discussed even if the data have stopped being reported prior to 1987.

9. Attaran and Saghafi (1988) provide a definition of the entropy index D:

D([E.sub.1], [E.sub.2], . . ., [E.sub.N]) = - [summation of] [E.sub.i] [log.sub.2] ([E.sub.i]) where i=1 to N

where N is the number of firms and Ei is the market share (of some relevant variable, typically sales or assets) of the ith firm. The entropy index is bounded between 0 (in the case of monopoly) and log N (in the case of perfect competition). A main benefit of the entropy index is that it can be disaggregated to provide an indication of the degree of diversification among and between the firms in an industry. See Attaran and Saghafi (1988) for details of the disaggregation procedure.

10. See O'Neill (1991) for more on the limitations of using the entropy index without data on all firms in the industry.

11. Unfortunately, these "best" data are not entirely flawless These data suffer from two practical limitations: they are not available annually, and there is an extreme time lag involved in their being reported.

12. These data appear in Table 2, which is primarily an update of Table II in White (1981), who reports concentration data for these five sectors between 1955 and 1977. As White points out, only the data for the transportation sector include any significant amount of foreign values.

13. These data are presented in Tables 3-5. The data in Table 3 represent a partial update of Table III in White (1981). Table 4 is an update of Table 2 in Schwartzman (1979). (Which is duplicated as Table IV in White (1981).) Table 5 is an update of Table 5 in Shenefield (1979). None of these data qualify as "best" in the sense described by White. In each case the reported values overestimate the true domestic concentration levels as the numerators contain both domestic and foreign figures while the denominators contain only domestic data. These data do, however, disclose some information about the trend in economy-wide aggregate concentration between 1978 and 1987 and, therefore, are worthy of examination.

14. Data are not available for the entire period of 1978-1987 for the banking sector. Rather than excluding a discussion of this sector, those data that are available are reported here.

15. See Amel and Jacowski (1989) or Boyd and Graham (1991) for details on the number of mergers over this time period.

16. While deposit share data are unavailable after 1982, Amel and Jacowski (1989) suggest that the decline in concentration continues, ("assets held by domestic banking organizations fell from 58 percent in 1976 to 50 percent in 1987"). See Wheelock (1993) for a more recent study of this issue.

17. See Cramer (1991) Table 5-1 for a listing and description of these mergers.

18. Food retailers comprise a significant portion of the Top 50 retailing firms. For example, the share of total revenues of the Top 50 retailing firms belonging to food retailers in 1978 and 1987 was 38.5 % and 27.2 %, respectively.

19. See Table 3 of Cotterill (1993) for a list and description of merger activity in the food retailing industry between 1979 and 1989.

20. The airline industry was deregulated with the Airline Deregulation Act of 1978. The trucking and railroad industries were both deregulated in 1980 with the respective passage of the Staggers Act and the Motor Carrier Act. See chapter 17 of Viscusi, Vernon, and Harrington (1992) for a discussion of the economic impact of deregulation on these three industries.

21. See Table 2 of Morrison (1993) for an overview of airline mergers, acquisitions and failures between 1979 and 1991.

22. Boyer (1993) writes that "Deregulation has brought with it a jump in the four-firm concentration ratio in the LTL industry from 23% in 1978 to 42% in 1987" and furthermore that the "number of Class I motor carriers (those exceeding $5 million in revenue) providing inter-city LTL service in 1989 stood at 158, down from 494 prior to deregulation."

23. A Class I railroad was defined by the ICC in 1989 as one with a minimum annual operating revenue of $93.5 million, as reported by Braeutigan (1993).

24. With the exception of the value added data for the top 50 firms, all manufacturing data indicate a decline in concentration.

References

Adams, Walter, and James W. Brock, "Efficiency, Corporate Power, and the Bigness Complex," Journal of Economic Education, Winter 1990, 30-50.

Alexander, Donald L., "An Empirical Test of the Mutual Forebearance Hypothesis: The Case of Bank Holding Companies," Southern Economic Journal, July 1985, 122-40.

Amel, Dean F., and Michael J. Jacowski, "Trends in Banking Structure since the Mid-1970s," Federal Reserve Bulliten, Mar. 1989, 120-33.

Attaran, Mohsen, and Massoud M. Saghafi, "Concentration Trends and Profitability in the US Manufacturing Sector: 1970-84," Applied Economics, Nov. 1988, 1497-510.

Berle, Adolph A., Jr., and Gardiner C. Means, The Modern Corporation and Private Property, New York; Harcourt, Brace, & World, Inc., 1939.

Bock, Betty, "How Big Are the Biggest Companies? Aggregate Concentration: Concepts, Numbers, and Perceptions," in U. S. Senate, Subcommittee on Antitrust, Monopoly, and Business Rights, The Small and Independent Business Protection Act of 1979, 96th Cong. 1st Sess., 1979, 502-35.

Boyd, John H., and Stanley L. Graham, "Investigating the Banking Consolidation Trend," Federal Reserve Bank of Minneapolis Quarterly Review, Spring 1991, 3-15.

Boyer, Kenneth D., "Deregulation of the Trucking Sector: Specialization, Concentration, Entry, and Financial Distress," Southern Economic Journal, January 1993, 481-95.

Braeutigan, Ronald R., "Consequences of Regulatory Reform in the American Railroad Industry," Southern Economic Journal, Jan. 1993, 468-80.

Brozen, Yale, Concentration, Mergers, and Public Policy, New York; Macmillan, 1982.

Caswell, Julie A., "Aggregate Concentration: Significance, Trends, and Causes," in Wills, Robert L., Julie A. Caswell, and John D. Culbertson (eds.), Issues After a Century of Federal Competition Policy, Lexington MA: Lexington Books, 1987.

Clarke, Roger J., Industrial Economics, Oxford: Basil Blackwell, 1985.

-----, and S. W. Davies, "Aggregate Concentration, Market Concentration, and Diversification," The Economic Journal, Mar. 1983, 182-92.

Cotterill, Ronald W., "Food Retailing: Mergers, Levereged Buyouts, and Performance," in Deutsch, Larry L., Industry Studies, Englewood Cliffs, NJ: Prentice Hall, 1993.

Cramer, Curtis, "Natural Gas Piplelines and Monopoly," in Nowotny, Kenneth, David B. Smith, and Harry M. Trebing, (eds), Public Utility Regulation: The Economic and Social Control of Industry, Boston: Kluwer Academic Publishers, 1991.

Gart, Alan S., Regulation, Deregulation, Reregulation: The Future of the Banking Insurance, and Securities Industries, New York: John Wiley & Sons, Inc., 1994.

Greer, Douglas F., "The Concentration of Economic Power," in Peterson, Wallace C., (ed), Market Power and the Economy, Boston: Kluwer Academic Publishers, 1988.

-----, Industrial Organization and Public Policy, New York: Macmillan, 1992.

Mann, H. M., and Karen McCormick, "Firm Attributes and the Propensity to Influence the Political System:" in Siegfried, John J., (ed.), The Economics of Firm Size, Market Structure, and Social Performance, Washington, D. C.: Federal Trade Commission, 1980.

Morrison, Steven C., "Airline Services: The Evolution of Competition Since Deregulation," in Deutsch, Larry L., Industry Studies, Englewood Cliffs, NJ: Prentice Hall, 1993.

Mueller, Willard F., "Conglomerates: a 'Nonindustry':" in Adams, Walter, (ed.), The Structure of American Industry, 8th ed., New York: Macmillan, 1990.

O'Neill, Patrick B., "Measuring Relative Concentration of Sales in U.S. Manufacturing," Southern Economic Journal, July 1991, 263-67.

Peterson, Wallace C., (ed.), Market Power and the Economy, Boston: Kluwer Academic Publishers, 1988.

Petr, Jerry L., "Economic Power and the Political Process," in Peterson, Wallace C., (ed), Market Power and the Economy, Boston: Kluwer Academic Publishers, 1988.

Rhoades, Stephen A., Power, Empire Building and Mergers, Lexington, MA: Lexington Books, 1983.

Salamon, Lester M., and John J. Siegfried, "Economic Power and Political Influence: the Impact of Industry Structure on Public Policy," American Political Science Review, September 1977, 1026-42.

Savage, Donald T., "Interstate Banking Developments," Federal Reserve Bulletin, February 1987, 79-92.

Schwartzman, David, "Prepared Statement:" in U. S. Senate, Subcommittee on Antitrust, Monopoly, and Business Rights, Mergers and Economic Concentration: Hearings on S.600, 96th Cong. 1st Sess., 1979, 581-614.

Shenefield, John H., "Prepared Statement:" in U. S. Senate, Subcommittee on Antitrust, Monopoly, and Business Rights, Mergers and Economic Concentration: Hearings on S.600, 96th Cong. 1st Sess., 1979, 61-95.

Shepherd, William G., "Causes of Increased Competition in the U.S. Economy, 1939-1980," Review of Economics and Statistics, November 1982, 613-26.

Viscusi, W. Kip, John M. Vernon, and Joseph E. Harrington, Jr., Economics of Regulation and Antitrust, D.C. Lexington, MA: Heath and Company, 1992.

Weiss, Leonard W., "The Extent and Effects of Aggregate Concentration," Journal of Law and Economics, June 1983, 429-55.

Weston, J. Fred, "Divestitures: Mistakes or Learing," Journal of Applied Corporate Finance, Summer 1989, 68-76.

Wheelock, David C., "Is the Banking Industry in Decline? Recent Trends and Future Prospects from a Historical Perspective," Federal Reserve Bank of St. Louis Review, Sept./Oct. 1993, 3-22.

White, Lawrence J., "What Has Been Happening to Aggregate Concentration in the United States?," Journal of Industrial Economics, Mar. 1981, 223-30.

[Patrick B. O'Neill, PhD., is associate professor in the department of economics, University of North Dakota, Grand Forks, ND 58202-8369.] The author thanks Frank M. Gollop, Carl A. Lundgren, and several anonymous referees for helpful comments.