Many competition authorities have provisions for transactions of “minor importance” where the parties involved can avoid mandatory pre-merger notification. Several industries have seen merger waves whereby consolidator firms engage in a strategy of “strategic rollups” – the serial acquisition of many small firms within a highly fragmented industry (“Sashimi M&As”). In this article the literature on sequential mergers is reviewed and three industry cases are examined. It is found that there is a path dependency in merger sequences. M&A series which start with the larger entities merged first should make competition regulators wary. Furthermore, how the acquired entities are organized post-merger – that they continue to operate independently can be an indication of a consolidator intending to engage in future M&As within the relevant market. While the individual transactions may not raise competition concerns, ultimately, the merger sequences may warrant the scrutiny of regulators.
By Mihkel Tombak[1]
Many competition authorities have provisions for transactions of “minor importance” being able to avoid mandatory pre-merger notification. Such thresholds are understandable given the under budgeting of those antitrust regulatory bodies and the daunting caseloads of many regulatory bodies. Under the Hart Scott Rodino Antitrust Improvements Act of 1976 there is a size of transaction test, which the U.S. Federal Trade Commission announced in January of 2022 would be $101 million over which there is a requirement for the parties to file premerger notification reports. In Canada the pre-merger notification- transaction size threshold is $93 million ($ Cdn) for 2022 and when the combined assets in Canada exceed $400 million. The EU bases such a threshold on the turnover of the firms involved, i.e. (i) a combined worldwide turnover of all the merging firms over €5 000 million, and (ii) an EU-wide turnover for each of at least two of the firms over €250 million. There are, however, instances where firms have engaged in merger and acquisition (“M&As”) strategies in highly fragmented industries, where the transactions are below the above thresholds and which should ultimately raise competition concerns. There are many examples of industries undergoing “merger waves.” One salient issue is whether and at what stage should such merger waves be stopped and what indicators, observable to regulators, are reliable signs of anticompetitive merger waves.
Kocourek, Chung & McKenna (2000)[2] describe merger waves at the industry level as “strategic rollups” whereby specific firms (which we will label as “consolidators”) have a strategy to buy up many small firms within a highly fragmented industry (Sashimi M&As). One can also observe a number of instances where the consolidator buys target firms in close proximity to one another, in part, in order to capture some efficiencies. In industries where consumers do not travel far for the good or service, this local preference for targets can also entail a localized increase in market power. This can occur where the price of the good is low relative to the transportation cost or when the demand for a service is more immediate such as for certain health care services.
The subject of this article is that there may be mergers which because of their small size may escape the scrutiny of competition authorities and yet the accumulation of those and many other acquisitions may result in establishments that ultimately do raise antitrust concerns. What follows is a survey of the literature on dynamic mergers and the regulation of the process which indicate conditions and behaviors which may indicate possible antitrust concerns. A few cases – in the tobacco, funeral homes and cardio-vascular surgery industries – where such concerns arose (and are arising) are then also discussed. We then summarize the messages for competition regulation.
I. A SURVEY OF THE LITERATURE ON SEQUENTIAL MERGERS AND THE INTUITION IT PROVIDES
There is a steadily growing literature in the theory of sequential mergers (Kamien & Zang, 1991, 1993[3]; Nilssen & Søgard, 1998[4]; Faulli-Oller, 2000[5]; Horn & Persson, 2001[6]; Tombak, 2002[7]; Nocke & Whinston, 2010, 2022,[8] among others). In much of the earlier analyses monopolization was uncommon. That would suggest that antitrust concerns could be muted. One basis for this was that many models feature “business stealing” whereby the firms in the industry but outside the merger transaction gain sales as the merged entity consolidates output (first noted by Stigler, 1950 and then examined in a single merger context by Salant, Switzer, & Reynolds, 1983)[9].
Kamien & Zang (1990, 1993) model the sequential merger process as a series of interdependent auctions and show that a free riding problem – where firms outside the merger may engage in “business stealing” may inhibit mergers that are profitable. Nilssen & Sorgard (1998) focus on the interrelatedness of sequential mergers and relate it to the business strategy taxonomy of Fudenberg & Tirole (1984).[10] They show that the Farrell & Shapiro (1990)[11] criterion for welfare increasing mergers (that the combined effect on consumers and non-merging firms is positive) is not sufficient for sequences of mergers as a particular merger may have the motive of affecting future mergers, i.e. motivate future consumer price increasing mergers or discouraging future price decreasing mergers. The above generally point to the importance of history of the industry – that there is a path dependency of sequences of mergers.
Fauli-Oller (2000) examined takeover waves in a Cournot setting with cost asymmetries and found a positive feedback in that a merger raises the profitability of future acquisitions. This he finds holds for both a consolidator and its rivals. Tombak (2002) analyzes an asymmetric Cournot model with efficiency gains from acquisitions, consolidation being endogenous, and firm acquisitions taking place as sequential (but interrelated) bargaining games. The model, having asymmetric firms, allows for efficiencies from a merger. Another essential feature in this analysis is that the output decision(s) may remain decentralized post-merger. The results of such a study indicate that those firms which are intent on continuing their sequence of mergers (i.e. have a strategy of a “rollup” of the industry) would tend to be those firms which: are bigger and more efficient; are those which post-merger do not centralize their output decisions (at the beginning of the sequence of mergers), and; are those which buy up their more efficient rivals first. The intuition for the consolidator buying up the more efficient rivals first is to eliminate an acquisitions competitor for subsequent acquisitions thereby saving on the purchase prices of ensuing procurements. The insight for leaving post-merger output decisions decentralized is to prevent the “business stealing” phenomenon which would enlarge those subsequent acquisition targets, again with the intent of keeping the price of successive purchases lower. That those consolidators would be the more efficient (and larger) firms means that ultimately there would be an efficiency gain from the series of mergers and that there is an incentive for the firms to endogenously engage in mergers. Furthermore, on this last efficiency point, it would suggest that antitrust concerns might be subdued. Ultimately, if allowed to proceed to the end, such a “strategic rollup” would entail a loss of welfare as the market power effects would in the end dominate the efficiency effect. This last argument is suggestive of an optimal point in the sequence of mergers at which antitrust authorities should intervene as the expression of market power would come with a consolidation of output decision making which would take place when the sequence of acquisitions is halted.
In more recent studies focusing on the regulation of such merger sequences Nocke & Whinston (2010) find conditions under which a myopic regulator would make reasonably good consumer surplus maintaining or increasing decisions. It should be noted that among those conditions are the complete information of all firms of all other firms’ merger possibilities and efficiencies. Nocke & Whinston also acknowledge that their conclusion may be weaker when: firms face price competition; when products are differentiated, and; where demand may increase over time. Also on the regulation of merger sequences, Nocke & Whinston (2022) find both theoretical and empirical justification for thresholds based on concentration changes regardless of concentration levels to test whether the merger would have unilateral increase in price. They also find that present U.S. threshold levels are not stringent enough unless the efficiency effects are greater than 5 percent.
In we now turn to examine some cases in light of the theory above. In what follows we scan the cases of: American Tobacco, U.S. Funeral Homes, and the U.S. market for cardiovascular procedures. American Tobacco is interesting because it is well studied and there were a sufficient number of M&A transactions to warrant an econometric analysis (Burns, 1986).[12] The U.S. funeral homes industry is of interest as it involved multiple consolidators and regulatory agencies partially intervened with sanctions on M&As for certain parties in certain geographical areas. The cardiovascular procedure market is chosen as it is an M&A process that is ongoing and shows some features of how sashimi M&As have evolved.
II. THE AMERICAN TOBACCO MARKET AT THE TURN OF THE PREVIOUS CENTURY
Burns (1986) examined the process of purchasing of 43 competitors acquired by American Tobacco Company and two affiliated companies in order to establish dominant positions in the tobacco industry. Ultimately, American Tobacco bought approximately 250 firms in the industry and achieved average market shares during the period 1900-1910 of 68.8 percent, and 73.9 percent in the smoking and fine cut markets, respectively.
While Burns focuses on the predatory pricing effect on purchase prices, there are two other important observations of the behavior of the Company relevant for this commentary. The first observation is that rivals were acquired sequentially and that, American Tobacco tried to buy its largest and most expensive rivals first and then it acquired smaller firms after it was well established in each segment of the industry. Secondly, often when companies were acquired their brands continued to be sold and Burns describes some of those brands as “fighting brands.” This indicated the retention of some degree of competition between brands in the product market particularly when American Tobacco was in the process of acquiring competing firms.
III. THE U.S. FUNERAL HOMES MARKET IN THE 1990S
Service Corporation International (SCI), the largest funeral home company in North America, made a takeover offer for the second largest firm in that industry, The Loewen Group, for US$ 4.1 billion on September 24, 1996. Such a takeover would have “remove(d) what has become SCI’s impediment to eternal growth: Loewen’s emergence as a rival bidder for small prey” (Economist, 1996).[13] The industry generated at the time annual revenue of $16 billion in North America and was highly fragmented (89 percent of the 23,500 funeral homes in the U.S. and Canada were owned by independent operators, Martin, 1996). High barriers to entry exist due to location, zoning and other regulations, the high fixed costs of building new funeral homes, and due to the pre-need sales backlog of existing competitors. The industry had been going through a period of rapid consolidation with the three largest capitalization companies (The Loewen Group, Service Corporation International (SCI), and Stewart Enterprises) making acquisition expenditures totaling approximately $1.2 billion and $1.8 billion in 1994 and 1995, respectively. Loewen has focused on the acquisition of rural homes while SCI and Stewart concentrate on urban properties.
The acquisitions strategy of these large consolidators has been to purchase smaller regional acquisition-oriented companies and small family-owned businesses. “The consolidators place a great deal of importance on the geographic proximity of the property in relation to other properties that they already own, and the ability to acquire a group of properties in a region, because it enables the implementation of clustering strategies. Clustering provides opportunities to capitalize on economies of scale, for example, by enabling a company’s various facilities to share vehicles, and employees, to reduce administrative expenses, …” (Martin, 1996, page 16).[14] Thus, there can be efficiencies due to acquisitions in this industry. Clustering, however, can also generate local market power as indicated by the four consent orders issued by the Federal Trade Commission restricting SCI from making acquisitions in ten areas of the U.S. over certain time periods.
The target firm, Loewen, had also a history of consolidations of family-owned businesses through partners which retain up to 10 percent interest in the future appreciation with the eventual purchase by Loewen of the retained interest. Thus, Loewen has ownership but the operations of the businesses are to some degree decentralized. The Loewen Group made the largest acquisition in the previous two years in North America by securing the right to purchasing Prime Succession for $295 million.[15] Prime Succession was a venture capital backed consolidator who started in 1992 with the objective of being a funeral home/cemetery acquisition vehicle, and rapidly became the fourth largest operator in North America. Hence an incentive for Loewen to make the offer was to “eliminate a strong acquisition competitor” (Martin, 1996, page 14). SCI was ultimately unsuccessful in the takeover bid of Loewen. These cases, both SCI’s attempt to acquire Loewen and Loewen’s purchase of Prime Succession, illustrate that a motivation for making acquisitions of large competitors could be to eliminate counteroffers for future acquisitions (as found in Tombak, 2002).
IV. THE U.S. CARDIO-VASCULAR MARKET
In the U.S., cardiovascular procedures continue to flow from hospital-based to physician-owned settings leading to increased fragmentation of the market. This has also led to increased incentives of consolidator activity. Despite two actions of the FTC during the 2010-2012 period – the Renown/Health Sierra Nevada Cardiology Associates case and the Providence Health & Services/Spokane Cardiology and Heart Clinics Northwest case[16] consolidations of cardiology offices have kept pace and are even accelerating. Renown had acquired Sierra Nevada with its 15 cardiologists in late 2010. It went on to purchase Reno Heart with its 16 cardiologists in March 2011 bringing Renown to 88 percent of the cardiologists in the Reno area. Furthermore, Renown used non-compete clauses in its contracts with those physicians. The FTC required Renown to suspend the non-compete provisions until at least six cardiologists accepted offers with competing practices in the Reno area. In the Providence case, the U.S. antitrust authorities expressed their concerns to the parties involved and thereafter Providence and Spokane abandoned the transaction.
As noted in a recent presentation by Provident Healthcare Partners (2022)[17] some of this new wave of consolidations have been instigated by private equity investors. It is claimed that private equity has an advantage in that cardiologist feel that it allows them to maintain clinical independence while gaining administrative efficiencies. The involvement of private equity can also make the work of regulators more onerous as there may be less publicly available information on the companies to monitor developments in the industry.
Recent examples include, Ares Management through its US Heart and Vascular group (USHV) completing an acquisition of Pima Heart and Vascular in 2021 and a purchase of Houston Cardiovascular Associates in 2022. Pima Heart was a group of five cardiovascular physicians and the merger brought the number of providers under the USHA umbrella in southeastern Arizona to 61. Houston Cardiovascular is a group of 16 cardiovascular physicians which combined with the partnership USHV announced on July 22, 2022 with Willowbrook Cardiovascular Associates’ 6 physicians brings USHV to a total of 22 physicians in the Houston area. Another private equity firm Deerfield Management through Novocardia obtained First Coast Heart & Vascular in 2021 and My Cardiologist in 2022. First Coast was the largest independent cardiovascular physician group in Northeastern Florida with 16 cardiologists. My Cardiologist has 27 cardiologists in the Miami-Dade and Palm Beach counties.
The above two examples in the cardiovascular procedures market (as well as the activities of Bain Capital in Texas and Arizona) are illustrative of the theory taking into consideration the regulatory context – the transactions are each below regulatory thresholds and localized enough to provide efficiencies and local market power. Also, the nature of the transaction and the mode of operation post-merger – that larger cardiovascular practices in a locale were acquired first and that they continue to operate with clinical independence are indicative of an intention to continue to acquire.
V. SUMMARY AND CONCLUSIONS
There are a number of lessons from the above commentary for antitrust regulators. The first is that history matters. There is often a path dependency in the sequences of mergers. Competition watchdogs need to examine the history of the industry and understand how a particular merger may affect the evolution of its market structure. There are indicators which can help with this projection – one is how the consolidator has organized and governed its past acquisitions. An acquirer which centrally controls its acquisitions, in particular, sales and output decisions would be less of a concern than one which allows the acquired enterprises to operate independently.
Another message to competition authorities is that acquisitions may fall below the maximum size of financial transaction for premerger notification and yet the mergers may have outsized effects on the evolution of the market structure. That is, while the transaction may fall outside the purview of the regulators, a small M&A can still have a significant effect on competition – at least on a local (and relevant) market level. Hence the Nocke & Whinston (2022) result that antitrust supervisors should monitor changes in concentration rather than levels may have to done at a more micro-level.
Finally, there are indications that such sashimi M&A strategies may be on the rise. In part, this may be facilitated by private equity consolidators. Such a development, in turn, will make the environment even more challenging for our competition regulators.
[1] University of Toronto. Mihkel.tombak@utoronto.ca.
[2] Kocourek, P., S. Chung, and M. McKenna, 2000, “Strategic Rollups: Overhauling the Multi-Merger Machine,” Strategy+Business, Booz, Allen & Hamilton, Issue 19, Second Quarter, pp.45-53.
[3] Kamien, M., & I. Zang, 1990, “The Limits of Monopolization through Acquisition,” Quarterly Journal of Economics, 105, pp. 465-499.
Kamien, M. & I. Zang, 1993, “Monopolization by Sequential Acquisition,” The Journal of Law, Economics, and Organization, Vol. 9, No. 2., pp.205-229.
[4] Nilssen T., & L. Søgard, 1998, “Sequential horizontal mergers,” European Economic Review, 42: 1683-1702.
[5] Fauli-Oller. R., 2000, “Takeover Waves,” Journal of Economics and Management Strategy, 9, (2) 189-210.
[6] Horn, H.,& L. Persson, 2001 “Endogenous mergers in concentrated markets,” International Journal of Industrial Organization, 19:1213-1244.
[7] Tombak, M, 2002, “Mergers to monopoly,” Journal of Economics and Management Strategy, 11, (3) 511-546.
[8] Nocke, V., & M. Whinston, 2010, “Dynamic merger review,” Journal of Political Economy, 118(6): 1201-1251.
Nocke, V. &, M. Whinston, 2022, “Concentration Thresholds for Horizontal Mergers,” American Economic Review, Vo. 112, No. 6, (June) pp. 1915-1948.
[9] Salant, S., S. Switzer, and R., Reynolds, 1983, “Losses from Horizontal Merger: The effects of an Exogenous Change in Industrial Structure on Cournot-Nash Equilibrium,” Quarterly Journal of Economics, 98, pp. 185-199.
Stigler, G., 1950, “Monopoly and Oligopoly by Merger,” American Economic Review Papers and Proceedings, 40, pp. 23-34.
[10] Fudenberg D. & J. Tirole, 1984, “The fat cat effect, the puppy dog ploy and the lean hungry look,” American Economic Review, 74, 361-368.
[11] Farrell, J., C. Shapiro, 1990, “Horizontal mergers: An equilibrium analysis,” American Economic Review, 80, 1007-126.
[12] Burns, M., 1986, “Predatory Pricing and the Acquisition Costs of Competitors,” Journal of Political Economy, 94, 266-296.
[13] Economist, 1996, “American takeovers – Barbarians at the pearly gates,” September, 28, pg. 87.
[14] Martin, D., 1996, “The Death Care Industry,” TD Securities Inc., Toronto, Canada.
[15] The author notes that Loewen group subsequently filed for bankruptcy which may, in part, have been due to the Prime Succession purchase price weighing on Loewen’s cash flows – it is tantalizing to imagine the possibility of a failing firm defense for the SCI bid.
[16] Renown/Health Sierra Nevada Cardiology Associates/Sierra Nevada Cardiology Associates, C-4366, FTC File No. 1110101 (final order issued November 30, 2012) (http://www.ftc.gov/enforcement/cases-proceedings/1110101/renown-health-matter) and, Providence Health & Services/Spokane Cardiology and Heart Clinics Northwest, (transactions abandoned on or about February 18, 2011) (https://www.ftc.gov/sites/default/files/documents/closing_letters/providence-healthservices/spokane-cardiology-and-hearts-clinic-northwest/110321providencestatement.pdf).
[17] Provident Healthcare Partners, “A New Wave of Investment and Consolidation in Cardiology,” A New Wave of Investment & Consolidation in Cardiology – Provident Healthcare Partners (providenthp.com), accessed January 4, 2023.
