chevron-down Created with Sketch Beta.

Antitrust Magazine

Volume 36, Issue 2 | Spring 2022

Preserving Potential Entry is Not the Holy Grail in Vertical Merger Enforcement

Brianna L. Alderman and Roger D Blair

Summary

  • The rationale for example 4 of the withdrawn 2020 Vertical Merger Guidelines would have wrongly sacrificed an immediate and certain consumer benefit from eliminating double marginalization between the merging firms in a vertical relationship for an uncertain future benefit of preserving new entry into the downstream market -- an unwise policy trade-off.
  • The FTC's effort to block or unwind the Illumina/Grail vertical merger will likely involve costs of one sort or another, but without any likely benefit.
Preserving Potential Entry is Not the Holy Grail in Vertical Merger Enforcement
Tim E White via Getty Images

Jump to:

In 2020, the Department of Justice (DOJ) and the Federal Trade Commission (FTC) jointly issued the newest version of the Vertical Merger (VM) Guidelines. Before their release, the draft VM Guidelines received comments, criticism, and suggestions from 74 interested parties. The final version of the VM Guidelines were intended to inform the business and legal communities of how the DOJ and FTC evaluate proposed vertical mergers. No doubt, this guidance was welcome since it should have reduced uncertainty in ways that would facilitate decision making. In our view, there was much not to like about the VM Guidelines. After less than 15 months, however, the FTC rescinded the VM Guidelines and thereby created considerable uncertainty. Although the DOJ technically left the 2020 VM Guidelines in place, Assistant Attorney General Jonathan Kanter indicated that the DOJ would participate in revamping the Guidelines.

Now, with the announcement that a new version of the Guidelines is forthcoming, there is an opportunity to correct some of the most serious misadventures that were present in the VM Guidelines. One such problem is a misplaced concern with protecting potential entry. Depending on the pre-merger market structure, horizontal mergers that eliminate a potential entrant may be problematic. But vertical mergers do not pose the same risks. Our concern is that procompetitive vertical mergers may be blocked because the Agencies believe that vertical integration might chill entry. We will identify this misplaced concern in the 2020 VM Guidelines in the hopes that it is not duplicated in the forthcoming revisions, and explore the consequences in more detail with our analysis of the Illumina-Grail matter. In our analysis, we will show that unwinding Illumina’s acquisition of Grail’s stock is apt to reduce consumer welfare. More generally, we suggest that the Agencies should rely on sound economic analysis to avoid blocking vertical mergers that enhance consumer welfare.

Preserving Potential Entry

The importance of entry in disciplining the economic conduct of participants in a market has long been recognized. When competitively structured markets are in disequilibrium, entry helps restore order: quantity expands and price falls to marginal and average cost. Excess profits disappear and social welfare is maximized. Consequently, it is clear that protecting the possibility of entry is sound antitrust policy and the Agencies should be applauded for doing so. In some markets, potential entry may also be worth preserving as the threat of actual entry leads the incumbents to temper their pursuit of profit.

A simple numerical example will illustrate the economic benefits of entry. Suppose that demand is represented as:

P = 140 − Q

where P represents the price and Q represents the quantity, and the constant marginal and average cost of production is equal to $20.

If the market were competitive, price would equal marginal cost of $20 and output would be 120.

In contrast, if the market is monopolized, the price will be $80 and the quantity will be 60 units. The monopoly profit will be $3,600. These profits will attract entry. If the incumbent and the entrant behave as Cournot Duopolists, i.e., compete on quantity, the output will expand to 80 units and the price will fall to $60. Total profit in the market will fall to $3,200 and will be split between the two firms. These profits will attract further entry.

As long as the firms continue to behave as Cournot oligopolists, industry output will be times the competitive output, where n is the number of rivals in the market.

Table 1 summarizes the results of further entry. As one can see, entry causes further reduction in the price and expansion in output.

Table 1 Economic Impact of Entry

Number of firms (n)

Price (P)

Quantity (Q)

Total Profit (Π)

Profit per Firm (Π/n)

1

80

60

$3,600

$3,600

2

60

80

$3,200

$1,600

3

50

90

$2,700

$900

4

44

96

$2,304

$576

5

40

100

$2,000

$400

6

37

103

$1,751

$292

7

35

105

$1,575

$225

8

33

107

$1,391

$174

9

32

108

$1,296

$144

10

31

109

$1,199

$120

 

It is plain to see that entry leads to lower prices and expanded output, which benefits consumers. Although the profits of the incumbent fall as the number of entrants increases, the result is positive. Thus, protecting the possibility of entry boosts consumer welfare.

Potential Competition

If a firm is not an actual participant in an antitrust market, but is a potential entrant in that market, it may influence the conduct of existing participants. A potential entrant’s decision to enter or not is rationally driven by profit considerations. The more profitable actual entry appears to be, the more likely a potential entrant will become an actual entrant. Consequently, the conduct of the incumbent firms may be influenced by the threat of entry. If the incumbent curtails its pursuit of short-run profit and behaves more competitively, it will reduce the probability of entry and thereby avoid the enhanced competition that would accompany entry. In this way, preserving potential competition provides some procompetitive benefits.

Under the conditions described above, entry increases the number of firms, output expands, and price falls, which benefits consumers. Given these potential benefits, preserving the possibility of entry in horizontal merger cases should be a policy goal of antitrust enforcement.

However, the likelihood that potential competition will prove to be so beneficial that an otherwise unobjectionable merger should be barred should not be overstated. Whether those conditions are met frequently is an empirical matter, but they are demanding. Nonetheless, when the conditions are met, barring a horizontal merger can have procompetitive benefits.

These unambiguously positive results do not necessarily carry over to vertical mergers. We demonstrate this in our discussion of the Illumina-Grail merger, but first we examine entry in the 2020 VM Guidelines.

Misplaced Concern for Protecting Potential Entry

The Agencies exhibited a misplaced concern for protecting potential entry in the 2020 VM Guidelines. This can be seen in the discussion of an example from the Guidelines involving a proposed merger in a successive monopoly environment, where the Agencies expressed unwarranted concern for a potential downstream entrant.

The example and discussion is as follows:

“Situation: Company A is the sole supplier of an active ingredient required to make an off-patent pharmaceutical drug produced only by Company B. Company A’s supply of the active ingredient is the related product. It sets a constant unit price. Company C is considering entering the relevant market with its own version of the drug. Were it to enter, head-to-head competition with Company B would be significant and prices for the drug would likely fall significantly, leading to increased sales. Company B buys Company A. In the absence of the merger, Company A would benefit from Company C’s entry.

Discussion:
The merger may diminish competition in the relevant market by making entry by Company C less likely. In the absence of the merger, Company A would likely have an incentive to facilitate the entry of Company C and to supply Company C if it did enter. The merged firm, on the other hand, may have the ability to prevent Company C from successfully entering the relevant market by refusing to supply Company C with any active ingredient. In this case, Company C’s successful entry into the relevant market may require Company C to produce the active ingredient as well. This two-level entry may be more costly and riskier than entering the relevant market alone, and thus may deter Company C from entering. Moreover, the merged firm may have an incentive to refuse to supply Company C unless it is markedly more efficient or targeting additional customer groups or markets.”

If the FTC or the DOJ blocked this merger, the outcome would be no better than permitting the merger and could be much worse. We illustrate this point in the context of the Illumina-Grail merger. By doing so, we hope to help to assist the Agencies in their expected future revision of the VM Guidelines.

Illumina-Grail: Facts

Illumina

Illumina is a biotechnology company whose areas of interest include oncology, microbiology, and cellular microbiology. Illumina is known for its specialization in DNA sequencing technology. Specifically, its development of next-generation short-read DNA sequencing (NGS) technology. This technology is a key component in Multi-­Cancer Early Detection (MCED) tests, like those made by Grail.

In its Complaint, the FTC acknowledges that there are four other companies in the same general market for NGS technology: Thermo-Fisher Scientific, GenapSys, Pacific Biosciences of California (PacBio), and Oxford Nanopore Technologies, but minimizes their competitive significance.

According to the FTC, the latter two companies are actually providers of long-read NGS platforms, and the FTC notes that “MCED developers do not view the long-read NGS platforms of PacBio and Oxford Nanopore as viable alternatives to Illumina’s short-read NGS platform due to their lower read counts, lower accuracy, and higher costs.” They also point out that GenapSys has “significant performance limitations compared to Illumina’s that make it unsuitable for MCED testing,” and that Fisher Scientific’s platform is “difficult to use, its throughput is too low for MCED screening, . . . [and] Thermo Fisher’s error rate is far too high to be used for a commercially viable test.”

Since the FTC did not find these alternatives to be reasonable substitutes in the NGS market, for all intents and purposes, it is appropriate to view Illumina as a monopolist in the market for NGS equipment and supplies. The FTC even goes so far as to say in its Complaint that “Illumina is the only NGS platform capable of meeting the technological demands required by MCED test developers” and that “Illumina’s NGS technology serves as a critical input to MCED tests, and there are no alternatives to it,” essentially conceding that Illumina is a monopolist in the NGS market. Therefore, if the merger between Illumina and Grail is permitted, Illumina would be the only vertically integrated supplier of MCED tests.

Grail

In 2016, Illumina founded Grail, Inc., a biomedical technology company that is now known for its development of Galleri Multi-Cancer Early Detection (MCED) tests. These are early cancer screening blood tests that can easily be administered by a general health practitioner. Before the development of these MCED tests, doctors were only able to detect four different cancers before they are symptomatic – breast, cervical, colorectal, and lung. With MCED tests, doctors are able to detect 50 different types of cancer before they are symptomatic.

Grail is not the only producer of MCED tests. Exact Sciences Corp. and Thrive Earlier Detection Corp. are just two of Grail’s rivals in the market for MCED tests. Despite the existence of competitors, there are no MCED tests sold commercially besides those sold by Grail. These competitors are developing their technology with the intent of entering the market for cancer screening technology. Illumina claims that these tests will not come with other modalities of cancer screening, and as such will not be in the exact same market despite being substitutes.

Illumina’s Acquisition of Grail

Illumina sold 88 percent of the equity in Grail in 2016, the same year it founded the company, and in doing so may have avoided the cost and risk that Grail faced to develop MCED tests.

Illumina has now re-acquired almost sole ownership of Grail. Alongside the FTC, the EU is also investigating the legality of the merger and asked Illumina to treat Grail as a separate company until further investigations are completed. Illumina has complied with this request, so as of now Illumina and Grail are not technically vertically integrated.

Regardless of the merger, Galleri MCED tests are available at a price of $950 per test. Unfortunately, at this time, health insurers do not cover the cost of these MCED tests since they have not yet been approved by the FDA. Illumina claims that it will drive down the price of each test by more than 40% by the year 2025 if the merger is permitted.

Illumina has previous experience in expanding biomedical testing for prenatal applications, and believes it can do the same with MCED testing. More specifically, Grail will need assistance in areas such as developing more effective ways to produce and distribute MCED tests on a larger scale, communicating with third-parties such as insurance companies and governments (both foreign and domestic), and getting pre-market approval in countries outside the US. These are all areas in which Illumina has had previous experience. By merging with Grail, Illumina believes that it will be able to produce more MCED tests sooner, in turn saving more lives.

The FTC has also issued an administrative Complaint to block or unwind the Illumina-Grail merger. The Complaint asserts that the vertical merger between Illumina and Grail would be anticompetitive by hindering potential entry in the market for MCED tests, and thereby harm consumers. In the next section, we show why this concern is unwarranted.

Economic Analysis of Illumina-Grail Integration

The Illumina-Grail integration is vertical in nature, and consideration of various alternative market structures will aid in evaluating competitive concerns regarding potential entry of Grail rivals and show why the FTC has failed to recognize that “a bird in the hand is worth two in the bush.”

Illumina-Grail Integration

The FTC alleges that Illumina is a monopoly supplier of Next Generation DNA Sequencing equipment and supplies that are adequate for MCED tests. For purposes of our analysis, we will assume that that this is correct. As for MCED tests, Grail’s Galleri seems to be ahead of others. Assuming that Grail finishes first in the technology race, it will have a monopoly in the market for MCED tests, which is expected to be both large and lucrative. For over 70 years, we have known that vertical integration in the presence of successive monopoly is procompetitive: the downstream price falls and quantity expands following vertical integration. The impact on consumer welfare is necessarily positive. These positive results flow from the elimination of double marginalization (EDM), which the FTC recognizes as a possible consequence of a vertical merger. We argue that this positive outcome would be both certain and immediate, and should be viewed as “a bird in the hand.”

The FTC contends that one or more of Grail’s rivals may be foreclosed from the MCED test market if the Illumina-Grail integration is permitted. On this basis, it wants to forbid Illumina’s complete re-acquisition of Grail. It argues that blocking the merger will have desirable economic effects that presumably will go unrealized if the merger were permitted. The FTC’s reasoning seems faulty. Even if their reasoning were sound, those positive effects would be both uncertain and deferred. In our view, this outcome should be viewed as “two in the bush.”

If Illumina’s re-acquisition of Grail is prohibited, there are several possible outcomes. None of these outcomes matches the benefit of EDM, which would be certain and immediate.

First, suppose Grail is successful with Galleri and no one else can compete. The would-be competitors either fail to deliver or their MCED tests are decidedly inferior to Galleri. In that event, we do not realize the benefits of EDM, i.e., lower price and higher output. Since higher output means more MCED tests and more lives saved, there is obviously an important loss. Therefore, consumers would be worse off if the FTC (or EU) successfully compels Illumina to divest Grail.

Second, suppose that one of Grail’s rivals successfully entered with a viable alternative to Galleri. The economic consequences are unclear. The market structure would be monopoly upstream and duopoly downstream. If Grail accommodates its rival’s entry, tacit collusion may result and nothing changes. In effect, the market remains one of successive monopoly, and the benefits of EDM are lost, at least in part.

Third, suppose that Grail and its rival behave as Cournot duopolists. Since some double marginalization will remain, the economic results will still be inferior to those flowing from vertical integration. Consumers will face higher prices than would have prevailed if the acquisition had been permitted. The outcome will be better than with successive monopoly, but still inferior to that of vertical integration.

Finally, suppose that Grail and its rival compete on price and, therefore, behave as Bertrand duopolists. First, suppose that MCED tests supplied by Grail and those supplied by the entrant are perfect substitutes. In that event, consumer price and quantity will be precisely the same as those that would accompany a vertical merger. But there is still a cost associated with blocking the merger. For one thing, successful entry of a rival is not certain. As a result, benefits of entry must be discounted by the probability that entry may not occur. Moreover, the entry is not immediate. Consequently, the probabilistic benefits must be discounted to their present value.

Second, suppose that the MCED tests are somewhat differentiated. If the incumbent and the entrant produce identical products and compete on price, preventing vertical integration does no harm. If, however, they produce somewhat different products while employing the same NGS platform and consumables, blocking the merger of Illumina and Grail will yield deleterious results. Some double marginalization will remain. Consequently, Illumina’s profits will be lower, prices to consumers will be higher, and both consumer welfare and total welfare will be lower.

In sum, preventing vertical integration by merger provides no obvious benefits.

Numerical Example

A numerical example may be useful in gaining an appreciation for the quantitative significance of the qualitative results that we have presented. The demand and cost conditions are purely hypothetical and are intended to be illustrative. Suppose that the demand is

P = 150 − 0.1Q

while the constant marginal (and average) cost of production (MCP ) is $50 and the constant marginal (and average) cost of distribution (MCD ) is $10. The price, quantity, and social welfare losses under various market structures are as follows.

Successive Monopoly

With monopoly upstream and an independent monopoly downstream, the price to consumers will be $127.50 and the quantity consumed will be 225 units. In this case, the social welfare loss will be $22,781.25.

Vertically Integrated Monopoly

There is a profit incentive for successive monopolists to merge. Such a merger increases total profits, but also eliminates double marginalization and thereby decreases the price to the consumer, increases the quantity consumed, and reduces the social welfare loss. In our example, following the vertical merger, price falls to $105, the quantity consumed rises to 450 units, and the social welfare loss falls from $22,781.25 to $10,125. If the upstream monopoly is lawful, these economic results are as good as they are going to be. As we will see, the results of entry are never better and often worse.

Upstream Monopoly and Downstream Duopoly

If there is entry at the downstream stage, the market structure downstream becomes a duopoly. The economic results of that entry are unclear. If the duopolists collude—either tacitly or overtly—the economic results are precisely the same as those with successive monopoly. In that event, there is no gain from preserving entry, but there are continuing, avoidable losses in consumer welfare.

If the duopolists compete on quantity in Cournot fashion, the price to consumers will be $120 and the quantity will be 300 units. The social welfare loss will be reduced to $18,000, which is an improvement over successive monopoly, but not over vertical integration.

If the duopolists compete on price in Bertrand fashion, the price will be $105 and the quantity will be 450 units. The social welfare loss would then be $10,125, which is precisely the same as the social welfare loss with vertical integration.

As these numerical results show, there is no gain to prohibiting a vertical merger of successive monopolists, but there may be costs.

These results are summarized in Table 2, which is shown below.

Table 2 Comparison of Economic Results

Market Price (P) Quantity (Q) Welfare Loss
Successive Monopoly $128 225 $22,781
Downstream Collusion $128 225 $22,781
Cournot Duopoly $120 300 $18,000
Bertrand Duopoly $105 450 $10,125
Vertical Integration $105 450 $10,125
Competitive $60 900 -

Further Analysis of Illumina-Grail

There could be multiple entrants in the MCED test market. In this event, the economic results are a bit better, but still not superior to those of a vertically integrated monopolist. For one thing, tacit collusion is less likely the larger the number of rivals. For another thing, if the MCED test market is a Cournot oligopoly, the larger the number of rivals, the closer the outcome is to the competitive solution. This is still inferior to the price and output with vertical integration. The Bertrand oligopoly solution with homogeneous products is equivalent to that of vertical integration.

It is clear that blocking the vertical merger involves costs of one sort or another, but without any likely benefits.

From its Complaint and its statement, the FTC appears to misunderstand some fundamental economics of vertical integration and control. According to the FTC, Illumina is a monopolist in supplying NGS inputs for MCED tests. Neither Grail nor any other producers of MCED tests can function without Illumina’s inputs. As a result, Illumina has substantial bargaining power in negotiating prices and terms with all actual and potential MCED test producers. In principle, Illumina can extract all of the monopoly profit in the MCED test market because no one can supply MCED tests without Illumina’s NGS equipment and supplies.

If Grail is the only successful producer of MCED tests, there will be an element of bilateral monopoly. In that event, Illumina and Grail will cooperate to supply the socially optimal quantity of NGS, but negotiate on the price. In this case, the price serves only to divide the jointly maximized profit.

Since Illumina has other businesses, it will have the upper hand in the negotiation and thereby will garner a disproportionate share of the profit. Entry by other MCED test suppliers will increase Illumina’s bargaining power since it now has further options. In the limit, Illumina will end up with all of the profit, which is what will happen if the Illumina-Grail merger is permitted.

The Bottom Line

The FTC views Illumina as a monopolist that faces no effective competition from other NGS rivals. Although Grail is in the lead now for MCED tests, there are a few possible rivals in the MCED test market. A vertical merger between Illumina and Grail will yield immediate and certain benefits. If the merger is banned, the entry of MCED test rivals may be more likely. But will this entry be beneficial, i.e., will consumer welfare be enhanced?

The competitive concern with market foreclosure can easily be overstated with unfortunate economic results. Assume, for example, that Illumina is a monopolist in the supply of NGS equipment and supplies as the FTC alleges. Further, suppose that Grail wins the technology race and becomes a monopolist in the provision of MCED tests. The vertical market structure is one of successive monopoly. On this information alone, we know that there will be double marginalization, which a vertical merger would eliminate.

The competitive concern with the merger is that it may foreclose entry by other MCED test providers, on the basis that Illumina will refuse to sell to NGS technology to rivals of Grail, and Grail will refuse to buy NGS technology from rivals of Illumina. The prospects of entry are impaired both upstream and downstream… a double whammy. But should the Illumina-Grail merger be forbidden on these grounds?

Structured economic analysis can help answer this question, or at least it can highlight the issues. The merger would provide immediate and certain benefits: elimination of double marginalization leads to lower prices for MCED tests, a greater number of tests being performed, an increase in the number of lives saved, and an increase in social welfare. So far, vertical integration of Illumina and Grail confers significant benefits and imposes no costs. But what about foreclosure, either upstream or downstream?

One may object that an Illumina-Grail merger will require entry at both stages, which is more difficult, riskier, and more expensive. In response, we would point out that Grail was created by Illumina in the first place. It is true that Illumina sold off some 88 percent of Grail in 2016 for business reasons. This recognition, however, does not undermine the point that simultaneous entry at both stages is possible.

The FTC might speculate that entry at one stage may be deterred due to the integrated firm’s ability to make entry unprofitable. This sounds good in principle, but is there much empirical evidence supporting this concern?

The FTC may argue that an Illumina-Grail merger will reduce technological change in MCED tests. This, of course, would be undesirable. Once again, is there either empirical evidence or theoretical proof that this is likely? In fact, there appears to be considerable evidence that highly innovative firms produce technological breakthroughs that larger firms acquire and develop.

Some Cautionary Observations

It is noteworthy to question what would have occurred if the partial divestiture had not occurred in the first place. Would the FTC demand a break up of a vertically integrated firm? If it did, would that demand survive a challenge in court?

What if Illumina simply realized that divestiture was a mistake and began developing MCED tests with internal resources? If it were successful, would it be obligated to deal with an independent Grail and any other successful entrants in the design of MCED tests and/or supply of NGS technology? If so, on what theory?

Although we do not delve into it in this paper, it is also interesting to note that prior to Illumina’s reacquisition of Grail, Illumina charged Grail an ad valorem running royalty on its revenues. Such a royalty will lead to an increase in the price charged by Grail with a consequent reduction in output. Vertical integration would remove this undesirable effect on consumer welfare.

If vertical integration by internal means is permissible, then a vertical merger with the same economic results should also be permissible. Economic equivalents should receive the same antitrust treatment. To do otherwise is to choose form over substance.

Casting a skeptical eye on vertical mergers that arouse suspicions of monopoly can be a serious mistake. Nobel Laureate Ronald Coase once warned:

“One important result of [our] preoccupation with the monopoly problem is that if an economist finds something—a business practice of one sort or other—that he does not understand, he looks for a monopoly explanation. And as we are very ignorant in this field, the number of ununderstandable practices tends to be rather large, and the reliance on a monopoly explanation is frequent.”

Much the same may be said of antitrust policy makers. The unfortunate consequence of this tendency is to forbid some procompetitive mergers. Such errors reduce—rather than enhance—consumer welfare.

In their treatise, Areeda and Hovenkamp advise policymakers to be wary of unsupported suspicions regarding vertical mergers. They conclude that:

“In the absence of such clear proof, however, the merger should be considered lawful. This conclusion is supported by the uncertainty about the fact or scope of future entry, the utility of actual independent entry into many markets, and an underlying skepticism about the likelihood or severity of significant anticompetitive effects flowing from actual vertical relationships not accounting for the bulk of the relevant markets. We are suggesting, in short, that rather uncertain possibilities of future entry into supplying or consuming markets are insufficient for presumptive illegality.”

The DOJ and the FTC are responsible for enforcing Section 7 of the Clayton Act, which forbids mergers that may substantially lessen competition or tend to create a ­monopoly. The enforcement concerns with vertical ­mergers—raising rivals’ costs and market foreclosure—may be very real in some markets, but vertical mergers are often benign or procompetitive. Forbidding all such mergers would be a mistake. Separating the wheat from the chaff, however, requires an objective application of economic principles that apply to vertical integration and vertical mergers.

Concluding Remarks

Vertical integration, by contract or by merger, has long been a source of confusion for antitrust enforcers and policymakers. It would be a valuable contribution to get things straight in the new version of the Vertical Merger Guidelines. A positive step in this direction would be an improved understanding of the economic effects of entry in a vertical context.

The FTC has asserted that Illumina is a monopolist in the provision of NGS equipment and supplies, but has not acknowledged that the competitive problems in the MCED test market flow from this upstream monopoly. Its focus should be on preserving and promoting the possibility of entry in the NGS market. This is not to say that Illumina has engaged in monopolizing conduct. In fact, the FTC has attributed Illumina’s monopoly to its superior products.

Before blocking a procompetitive vertical merger to preserve potential entry, the Agencies should recognize that “a bird in the hand is worth two in the bush.” A vertical merger of successive monopolists will eliminate double marginalization and thereby improve consumer welfare immediately. The uncertain future benefits—if any—from blocking the merger should be viewed with the skepticism that they deserve.

Professor Sara Bensley provided some much needed advice and research assistance, and we also thank Greg Wrobel for his editorial suggestions. We are grateful to the University of Florida’s Ronald E. McNair Scholars Program, Social and Behavioral Sciences Graduate School Ph.D. Preparatory Program, and Department of Economics for financial support. The views expressed here are solely those of the authors.

    Authors