Monopoly leveraging concerns in a broadband world

The Computer Inquiries were designed to govern competition in a narrow-band world in which Internet access was provided almost exclusively by common carrier telephone companies. The economics of monopoly lever-aging within the Internet sphere has become significantly more complicat-ed now that broadband technology has begun supplanting dial-up as the Internet access mechanism of choice for consumers. The ascendancy of broadband is significant here for four reasons.

First, it dramatically increases the competitive stakes for the communi-cations industry as a whole. Widespread use of broadband connections makes industry-wide technological convergence less of a theoretical possi-bility and more of an imminent, transformative reality. Suddenly, the com-pany that efficiently controls the highest quality pipes to end user premises could theoretically dominate not just one communications service, but all communications services: voice, data, and video programming.

Second, broadband access threatens to marginalize the ISP intermedi-aries that, in the narrowband environment, act as competitive buffers between the monopoly provider of last mile access and the Internet at large. An ISP necessarily plays a less central role in a user’s broadband experience than it plays in the traditional dial-up setting. In the latter con-text, the end user scarcely notices the telephone company’s involvement in her relationship to the Internet because the call she places to her ISP appears much the same as any other “local” call, the telephone company charges nothing extra for it, and the ISP performs all protocol conversion functions. In contrast, and for technical reasons discussed below, the broadband consumer is inevitably aware that the existence, quality, and price of her Internet connection depends largely on her cable or telephone company or other platform provider. That remains the case whether or not her ISP is affiliated with that company—as it usually is.

These first two factors could be cited as reasons why the arrival of broadband technology warrants greater government intervention to

pre-vent monopoly leveraging by providers of last mile transmission services.

The third and fourth considerations, however, cut in the opposite direction.

The third relates to cross-platform competition. Whereas dial-up Inter-net access has traditionally required the use of a single company’s circuit-switched telephone network, today’s Internet access market is subject to competition among rival broadband platforms. Opponents of government intervention say that robust platform competition will safeguard the inter-ests of consumers better than regulation ever could, avoiding any of the investment disincentives and unintended consequences that regulation is often thought to produce. They thus argue that, so long as consumers can choose among a range of broadband providers, the resulting competition will force each provider to ensure nondiscriminatory use of the Internet;

otherwise, consumers will retaliate by moving to another platform.

Of course, the prospect of cross-platform competition can protect against anticompetitive practices only to the extent that consumers may easily drop one broadband service and subscribe to another. Several factors can complicate this competitive dynamic. Although consumers in many densely populated markets today have a choice of two or more broadband providers (cable, DSL, and perhaps wireless), consumers in less densely populated areas may have no realistic choice of providers at all; indeed, they may be lucky even to have one. Even if they have two, a duopoly is unlikely to produce optimal output and pricing decisions over the long term, although it is surely better than a monopoly.¹³Also, consumer choice can provide a check on a dominant provider’s market power only to the extent that these platforms are close market substitutes. For example, if consumers perceive that cable modem service is significantly faster for the money than DSL—as, in fact, it often is—they may tolerate some discrim-ination by their cable modem provider against certain applications or con-tent providers before canceling their accounts and calling the telephone company. Likewise, if consumers invest heavily in the platform they initial-ly adopt—say, by making extensive use of an e-mail address specific to that platform—they may hesitate before incurring the costs necessary to switch to an alternative platform. But the central point remains: over time, con-sumers can expect increasing choices in the broadband marketplace, and those choices will tend to weaken a firm’s incentives to engage in signifi-cant monopoly leveraging.

Monopoly Leveraging Concerns and the Internet 157

The final factor that may reduce the need for heavy preemptive safe-guards against such leveraging relates to a firm’s internalization of comple-mentary externalities. This economic phenomenon, which we will call the

“complementary externalities” principle for short, is closely related to the

“one monopoly profit” principle we addressed in chapter 1. Recall that the total profits a monopolist can earn if it seeks to leverage its monopoly in one market (here, the market for physical-layer broadband access) by dom-inating a complementary market (here, the applications and content mar-kets) are theoretically no greater than the extra profits it could earn in an unregulated environment simply by charging more for the monopoly prod-uct itself. This fact gives the monopoly platform provider a powerful incen-tive to enhance its platform’s attracincen-tiveness to consumers so that more of them will pay a higher price for it. Accordingly, the monopolist can nor-mally be expected to take whatever steps are necessary, including steps to promote competition in the applications market, to spur the creation of complementary products that will drive demand for its platform.

This complementary externalities principle is subject to several excep-tions, the best known of which, as mentioned in chapter 1, is called “Bax-ter’s law” in honor of the Justice Department antitrust official who invoked it in breaking up AT&T in the early 1980s. As applied here, Baxter’s law holds that a platform monopolist does have strong incentives to monopo-lize a largely unregulated applications market if the platform market itself is subject to rate regulation.¹⁴ This exception had great significance for Internet access in the pre-broadband world because regulators have long capped the price that telephone companies may charge consumers for use of the circuit-switched telephone network that served as the monopoly platform for Internet access for many years. Because these companies were barred from extracting monopoly profits from the use of that platform itself, they had an incentive to extract such profits instead by charging supracompetitive rates for Internet access and other information services—

an objective that would necessarily require discriminating against rival providers that could offer subscribers lower, cost-based rates for the same services. The regulatory response, as discussed above, was the set of non-discrimination requirements adopted in the Computer Inquiries.

As a practical matter, however, Baxter’s law does not apply—and the complementary externalities principle arguably does apply—in the broad-band world because regulators have not capped the rates that providers

may charge their customers for broadband Internet access. This fact removes a key incentive for even a dominant broadband provider to dis-criminate against unaffiliated applications and content providers, for such discrimination cannot increase the provider’s overall profits and could pos-sibly lower them if it degraded consumers’ perceptions of the platform as a whole.

We do not, however, wish to overemphasize this last point, for the com-plementary externalities principle is subject to several additional excep-tions beyond Baxter’s law. To take a few examples, a dominant broadband firm might have incentives to discriminate against unaffiliated content or applications providers if it views them as potential rivals in the platform market itself; or if the applications market is to some extent independent of the platform market and is itself subject to scale economies or network effects; or if the dominant firm is simply irrational and misperceives how its interests are affected by the “one monopoly profit” rule.¹⁵Our essential point is this: if they enjoy significant market power at the physical layer, non-price-regulated broadband providers might, but do not inevitably, have incentives to discriminate against unaffiliated content and applica-tions providers. The complex economics of that point often get lost in the rhetoric about whether to force last mile broadband providers to open up their networks to unaffiliated ISPs or to provide non-discriminatory access to content and applications providers.

II. Three Proposals for Addressing Monopoly Leveraging Concerns

Some policymakers view the steady mass migration to broadband Internet access with mixed emotions. On the one hand, they recognize that the cable and telephone providers of these new broadband services are adding enor-mous value to the economy, both by making electronic communications much more versatile and efficient and by exponentially expanding the range of communications-related services available to ordinary consumers.

They also acknowledge that, to an extent, these broadband companies need adequate incentives—i.e., the prospect of significant profits—to con-tinue investing billions of dollars in the construction of broadband facili-ties to ever greater numbers of consumers, including those who live in

“high cost,” less densely populated regions.

Monopoly Leveraging Concerns and the Internet 159

On the other hand, some observers worry that these companies, if inad-equately regulated, will carve up the country into broadband monopolies or duopolies and, after getting consumers hooked on this new technology, will discriminate against unaffiliated firms at the applications and content layers. For example, they speculate, once Time Warner’s cable system dom-inates the market for broadband access in particular areas, it will have incentives to discriminate subtly in favor of Warner Brothers’ content and against the content of corporate rivals like Disney. Advancing this type of argument, Stanford professor Lawrence Lessig claims that broadband providers, disdainful of the Internet’s traditionally non-proprietary open-ness, can and will “use the architecture [of their networks] to regain strate-gic control” of the Internet as a medium for communication. In particular, Lessig suggests: “‘Everyone knows that the broadband era will breed a new generation of online services, but this is only half the story. Like any inno-vation, broadband will inflict major changes on its environment.It will destroy, once and for all, the egalitarian vision of the Internet.’”¹⁶

The monopoly leveraging concerns raised by Lessig and others have spawned three general categories of proposed policy responses, each of which is distinguished from the others by the layers of the Internet on which it focuses. The first would entitle an end user to gain nondiscrimina-tory access to the ISP of her choice, no matter who her broadband trans-mission provider might be. This approach, sometimes known as open access—or, as the FCC calls it,multiple ISP access—seeks to preserve in the broadband environment the role of independent ISPs as competition-pro-tecting intermediaries operating at the logical layer. The second proposal, often accompanied by the catchphrase “Net neutrality,” would limit the ability of any broadband provider to deny end users access to particular applications or content over its system. The third aims to stimulate greater competition at the physicallayer, and thus limit monopoly leveraging con-cerns at the higher layers, through the creation or elimination of leasing obligations for the broadband-specific facilities of wireline telephone com-panies. We address each of these policy responses in turn.