Chairman Bachus, Ranking Member Johnson, and Members of the Subcommittee, I am Scott Hemphill, a Professor of Law at Columbia Law School. I write and teach about antitrust law, innovation, and competition. My research has considered the application of antitrust law to Internet service providers (ISPs) and video distributors. n1 I recently served the New York Attorney General as Chief of the Antitrust Bureau, which strives to protect competitive markets on behalf of New York consumers and businesses. n2
I welcome the opportunity to testify today about the antitrust implications of the proposed merger of Comcast and Time Warner Cable (TWC). Antitrust law has a critical role to play in preserving competition. Competition benefits the economy through low prices, efficient production, and innovative new products and services. Antitrust law accomplishes this, in relevant part, by prohibiting mergers that may "substantially . . . lessen competition" or "tend to create monopoly." n3
Some of the concerns raised about this merger are best framed as antitrust objections. Critics have charged that the deal will have anticompetitive effects by raising cable or broadband prices to consumers; by harming video programmers; or by foreclosing competition from online video distributors such as Netflix. A closer evaluation of competition in these markets, however, demonstrates that the merger is unlikely to have such anticompetitive effects. n4 Let me take these points in turn.
Most mergers that receive antitrust scrutiny are combinations of rivals. Such mergers are troubling because they eliminate head-to-head competition between the firms or lessen competition among the remaining market participants. Usually, the primary focus is products and services sold by the parties, so-called "output markets," though mergers can also have an effect on the market for products and services purchased by the parties.
To evaluate output markets, antitrust analysis ordinarily adopts the perspective of a particular purchaser of the goods or services in question--for example, a purchaser of wireless service in New York City. Such buyers can choose among AT&T, T-Mobile, and other providers. A merger's removal of a significant competitive choice, for a particular set of buyers, can harm competition through higher price and lower quantity, among other effects.
Or consider an individual who wishes to fly from Washington, DC, to Chicago. Such buyers can choose among US Airways, United, and other airlines. Here, once again, the removal of a competitive option for those local consumers threatens substantial harm. This reduction in competition, considered from a particular purchaser's standpoint, was the central premise of the Justice Department's antitrust lawsuits challenging the AT&T/T-Mobile and US Airways/American mergers.
The Comcast/TWC merger is nothing like that. A consumer in a particular New York City apartment seeking traditional "multichannel" video service chooses among TWC and other providers. Comcast is not an option unless she moves to another city. Similarly, consumers in Philadelphia can pick among Comcast and other options, but TWC is out. Before and after this merger, consumers would have the exact same number of options to choose from. This is true not only of video service, but also Internet access, telephone service, and bundles of all three. Nor are Comcast and TWC plausibly potential competitors. Comcast has evinced no interest in building facilities to compete with TWC in TWC's present territory, or vice versa.
The lack of overlap means that the usual concern of antitrust enforcement is completely absent. This is not to say that cable prices won't rise anyway if the merger is approved. They might. ESPN, among other programmers, has achieved remarkable price rises in recent years in its negotiations with distributors, increases that distributors pass along to consumers. Comcast has a powerful incentive to bargain for lower prices, but programmers have considerable bargaining power. The key point is that the proposed merger has no tendency to affect consumer prices through any reduction in competition between the merging parties.
As noted above, mergers can have an effect not only in output markets, but also in markets for products and services purchased by the parties. For example, suppose the only two employers in a small, isolated town compete for labor. If they merge, the merged firm may thereby acquire increased "monopsony" power. A monopsonist may have an incentive to inefficiently reduce its hiring in order to drive down wages. Monopsony is thought to be a significant issue in labor and agricultural markets, because a cutback in the purchase of labor or agricultural commodities might plausibly reduce the price of those inputs.
Once again, the Comcast/TWC merger is nothing like that. Video programming is different from labor or grain. When a farmer sells grain to a buyer, that sale reduces the amount of grain left over for other buyers. By contrast, when ESPN sells programming to Comcast, nothing has been used up. ESPN remains free to make a similar deal with TWC or Cablevision. Comcast and TWC are not competing for the rights to a scarce resource. As a consequence, the merger provides no similar opportunity to economize on this input as a way to reduce its price.
A related concern is that a post-merger Comcast might have greater bargaining power with programmers, thanks to its enlarged subscriber base. Enhanced bargaining power, to the extent that it simply shifts profits among firms, is not an anticompetitive harm. But in any event, it is hardly clear that Comcast would be able to strike a better bargain. On the one hand, the stakes would be higher for ESPN, compared to today, because ESPN could lose more revenue from lost viewers if its contract negotiations with an enlarged Comcast broke down. On the other hand, Comcast would have more to lose too, as more customers could plausibly complain or cancel their service in the event of a breakdown. To conclude that Comcast gains bargaining power on a per-subscriber basis, it is necessary to establish that there is some special, disproportionate consequence to the programmer in the case of bargaining failure.
One such disproportionate consequence might arise if the post-merger Comcast were so large that, without its business, a programmer would be unable to effectively function. But post-merger Comcast lacks the requisite scale. If we take as a starting point the Federal Communication Commission's previously expressed view on this subject, a video distributor must have more than 30 percent of traditional video subscribers to pose such a risk. n5 But the FCC's view was likely too cautious when it was reached in 2007; indeed, it was rejected by the D.C. Circuit for understating the degree of competition in video markets. n6 It is surely too conservative today. Among other developments, the rise of online video provides opportunities for programmers to reach viewers without selling their content to a traditional distributor. This transaction, which results in a share of traditional video distribution slightly less than 30 percent, is therefore unlikely to be of a sufficient size to make a meaningful difference in this respect.
Foreclosure of Online Video
A final possibility, raised by some mergers, is that a transaction might lessen competition by enabling foreclosure--that is, by undermining rivals' ability to compete, resulting in consumer harm. The exclusion of competitors is potentially even more worrying than the distortion from changed prices, because it can slow new entry and thereby harm innovation. Under certain circumstances, a merger can increase the risk of foreclosure, by strengthening the resulting firm's incentive and ability to exclude. A full antitrust analysis of the Comcast/TWC transaction therefore requires an evaluation of the prospects for foreclosure.
In any complex business environment, it is possible to raise a speculative concern about an enormous number of theoretical foreclosure effects. Discussing them all is beyond the scope of my testimony, but let me focus on one prominently stated foreclosure concern, that a traditional video distributor such as Comcast might have an incentive to inhibit the competitive prospects of online video distributors. Netflix is the best-known example, but there has been enormous growth in this sector in recent years. Such firms threaten the traditional model of multichannel video distribution by enabling "cord-cutting."
Cord-cutting is a misnomer, because consumers still rely on a broadband Internet connection to access online programming. The broadband connection is often supplied by traditional video distributors, using the same physical plant. That combination of businesses provides a potential opportunity to undermine online video, by choking off the Internet access on which it relies.
But it also furnishes a powerful reason not to do so. Online video is an important and increasing part of the value provided by broadband Internet. Harming a growing business to preserve a declining one is a costly and doubtful business strategy. That fact reduces the incentive to engage in foreclosure.
If Comcast nevertheless wished to foreclose competition, how would it do so? The most obvious routes are cut off by Comcast's existing regulatory commitments, made as a condition of its NBC-Universal acquisition. These commitments include acceptance of the FCC's Open Internet rule. Critics have focused on an interaction between online video distributors and Comcast not subject to the existing restrictions. ISPs are increasingly receiving payment for direct interconnection to the ISP network. This practice of "paid peering" has raised concerns that such payments might harm online video, resulting in a form of foreclosure.
Paid peering is an ineffective tool of foreclosure. Online video distributors are under no obligation to pay directly for interconnection. They are typically free to contract with middlemen, such as backbone providers and content delivery networks, that in turn deliver the content to the ISP. Those alternatives mean that an ISP is unable to degrade online video delivered in this fashion without also degrading other traffic delivered by the middleman. Such protection is particularly potent for smaller distributors that are more easily pooled with other traffic.
To be clear, the fact of payment is not in itself problematic. Online video distributors and other content providers have long paid for interconnection, and Comcast has long received payment for interconnection. Payment for interconnection has always been made using some mix of cash and reciprocal carriage of the other firm's traffic. If a content provider or its agent takes on no reciprocal carriage, then cash is a natural alternative. In this respect, paid peering is a new variant of an old business practice.
Paid peering is best seen not as an instrument of exclusion, but as a means to put a price on the additional capacity demands resulting from the increased popularity of online video. It is efficient for the distributor and its end-users, considered collectively, to pay for that capacity, rather than spreading the expense among all ISP customers. Doing so better aligns use with cost and incentivizes both investment and economical use. Paid peering is not the only possible solution to that problem, of course. Surcharging heavy users, provided that the surcharge is not itself an instrument of foreclosure, is a viable alternative.
Comcast's recent interconnection agreement with Netflix, far from suggesting an antitrust concern, is a sign that the market is working well. The proposed merger does not change that. In considering whether to impose a prophylactic restriction on Comcast's ability to engage in paid peering, the current moment of experimentation seems a particularly inapt time. We should be particularly cautious about intervening in the absence of a demonstrated problem.
A thriving online video distributor requires, in addition to a broadband connection to users, access to programming. That fact suggests a second potential strategy for foreclosure, which is to inhibit access to programming. In particular, the traditional distributor, as part of its contract with a programmer, might insist upon restrictions in the programmer's dealings with online video distributors. For example, Comcast might insist that Disney not make certain types of online content deals, or insert contractual clauses that have a similar but more indirect effect.
The Justice Department has reportedly investigated these contracts. n7 I am unaware of public information about the prevalence of such contracts or their practical effect. A contract that disadvantaged online video might well be resisted by the programmer, who would prefer to make profitable sales to such firms, and therefore such a contract would be more costly to the distributor to secure. Overall, whether this is a profitable or likely foreclosure strategy is currently unclear.
The key question for present purposes, in any event, is whether the proposed merger worsens whatever foreclosure problem might exist. Comcast is prohibited from enforcing any such anticompetitive contracts by its NBC-Universal conditions. n8 That prohibition would be extended to TWC if the merger is approved. That extension has the effect of strengthening existing protections against this potential form of foreclosure.
In conclusion, a merger between Comcast and TWC is unlikely to have several effects posited by critics. It is unlikely to affect prices for consumers; to have anticompetitive effects on programmers; or to incrementally foreclose competition from online video by impeding connectivity or access to programming. Thank you for the opportunity to discuss these issues with the Subcommittee.
n1 See, e.g., C. Scott Hemphill & Tim Wu, Parallel Exclusion, 122 Yale Law Journal 1182 (2013); C. Scott Hemphill, Network Neutrality and the False Promise of Zero-Price Regulation, 25 Yale Journal on Regulation 135 (2008).
n2 See, e.g., Proposed Final Judgment, United States and New York v. Verizon Communications Inc., No. 12-cv-1354 (D.D.C. Aug. 16, 2012).
n3 Section 7 of the Clayton Act, 15 U.S.C. [Sec.] 18, provides: "[N]o person . . . shall acquire . . . the whole or any part of the stock or other share capital . . . [or the] assets of any other person . . . where in any line of commerce or in any activity affecting commerce in any section of the country, the effect of such acquisition may be substantially to lessen competition, or to tend to create monopoly."
n4 For purposes of my testimony, I have not attempted an exhaustive analysis of the transaction, which would include an assessment of additional issues, such as video advertising, the limited programming assets contributed by TWC, and the efficiencies that the parties expect from the merger.
n5 FCC Fourth Report and Order, 23 F.C.C.R. 2134 (adopted Dec. 18, 2007; released Feb. 11, 2008).
n6 Comcast Corp. v. FCC, 579 F.3d 1 (2009).
n7 Shalini Ramachandran, "Favored Nations" Fight for Online Digital Rights, Wall St. J., June 14, 2012, at B3.
n8 Modified Final Judgment [Sec.] V.C, United States v. Comcast Corp., No. 11-cv-106 (D.D.C. Aug. 21, 2013), available at http://www.justice.gov/atr/cases/f300100/300146.pdf.
Read this original document at: http://judiciary.house.gov/?a=Files.Serve&File_id=14DA5814-6EF9-4313-8CE7-CE81440A7198