Cutting the Gordian Knot: Providing the American Public with
Advanced Universal Access in a Fully Competitive Marketplace at the
Lowest Possible Cost
The Advanced Universal Access Dilemma
A common vision is shared among both the business community and the federal political establishment as to what the broad characteristics of the national information infrastructure (NII) should be. Both the Democratic Administration1 and the Republican Congress2 share the view with the private sector that the NII must be built by private business in a competitive marketplace, rather than directed and regulated by government. Moreover, it should be a "network of networks," consisting of interconnected but competing systems, not just constructed but also operated in accordance with the competitive market model.
Yet neither the Congress nor the Administration is content with laissez-faire, purely market-driven deployment. A sense of national urgency, not only to ensure our ability to compete with other nations in the global economy, but also to help address major domestic concerns, has prompted other, not easily reconcilable goals. The first is to have the advanced infrastructure deployed as widely and rapidly as possible, with particular emphasis on early access by education and health care providers and users.3 There is also the second, broader social concern about preventing the potential disparity between "information haves" and "information have-nots" for critical advanced access to the NII at reasonable and affordable rates. The current Senate proposal is to provide subsidies to designated carriers of last resorta continuation of the existing narrowband mechanisms and an extension of those mechanisms into the realm of new and as yet unknown services.4
This public vision of the NII is not just internally inconsistent. It relies on a foreknowledge of the events that will define future technology deployment. Yet the actual composition of the anticipated advanced physical architecture and technology that will link schools, hospitals, government, businesses, and residences and will ultimately constitute the NII is unknown today. Although it is a given that advances in technology should benefit certain broad areas of our society, such as education and health care, there is no consensus as to what specific applications will be most valuable to the economy and the social fabric of the country.
The Magnitude of Current Basic Service Subsidies and Proposed Broadband Service Subsidies Under Current Operating Arrangements
The dimensions of the problem are well documented. The nationwide local exchange carrier (LEC) collection of revenues from customers and interconnecting carriers in 1992 exceeded $91.5 billion. Subsidy level estimates from various sources range from $4 billion to $20 billion.5 In a study of data from Tier 1 LECs which discussed only the averaging effects within individual LECs (excluding the complex subsidy flows among large and small LECs and interexchange carriers), costs for providing service to rural customers exceeded rural revenues by 35 percent$5 billionor more than $19.00 per line per month.6
A national policy to create a fully capable broadband network universally available in rural areas and provided by regulated carriers using traditional costing methods would require far greater subsidies: total loop and non-loop costs per line in rural areas would range from $154 to $168 per month per line, depending upon whether deployment completion was targeted for a 10- or 20-year period. Since rural telephony per line revenues average approximately $54 per month, customer revenues or subsidies would have to bridge a $100 per month gap. These estimates exclude the incremental cost of any customer premises equipment that would be necessary.7 The regional bell companies and several of the largest cable multiple system operators are seeking or have entered into agreements with set-top box manufacturers, such as Digital and Scientific-Atlanta, to develop proprietary consumer premises equipment with an eye toward bringing the cost down to well below $1,000 per unit.8 Even if the nation chose to afford it, providing subsidies under a business-as-usual model would effectively prevent competitive entry and keep costs high. Since there is currently no two-way universal broadband network in place, there is no rationale for selecting and subsidizing one potential provider over another.
This paper does not propose to chart a hybrid middle ground between regulation and competition. It also neither evaluates existing and proposed narrowband transition plans nor offers any new transition mechanisms to end the current subsidy scheme for those universally available services. What this paper offers instead is a mechanism to provide universal two-way broadband service to high-cost areas in an industry- and technology- neutral way, through the use of a fully competitive marketplace, and at the lowest possible cost.
The Need for a Separate Broadband Paradigm
As noted above, the current Senate bill, S. 652, calls for an evolving definition of universal service. The law's effect would be to pull new broadband services into the subsidized and regulated carrier-of-last-resort mandatory provisioning scheme as such technology gained popular acceptance.9 As has already been discovered in narrowband communications markets supplied by multiple competing providers, the efficiencies of competition cannot be sustained if one or more parties' market behavior and obligations are being mandated by government. In an environment where there is a "network of networks" that both interconnect and compete with each other, intra-industry subsidies undermine and distort the marketplace while providing no direct end user benefit.
Although complaints about pricing and customer service led to the reregulation of cable television, cable companies are not monopoly suppliers of broadband services. A residential customer in search of broadband services has the technical capability to receive broadcast television, cable, multichannel multipoint distribution service, low-power satellite, and direct broadcast satellite signals. Portions of the entertainment and information that are received via such media are also available from newspapers, broadcast radio, on-line computer services, video cassettes, laser disks, computer diskettes, and CD-ROMs. The first principles of a new broadband paradigm rest on a free market foundation. Although robust narrowband local exchange competition will eventually arrive and enable the deregulation of that market, advanced infrastructure investment must go forward now in a "clean state" environment, in which government dictates neither the financial nor technical terms for the offering of new services. A distinct broadband deployment-friendly and regulation-free environment, possessing the following characteristics, is necessary:
A great deal of capital is currently being invested with the hope of creating the technologies, products, and services that will ultimately prove victorious in the marketplace and constitute the NII. Any participation by government for the purpose of ensuring universal access to those advanced broadband services by the general public must, therefore, be technology and service neutral and must have the smallest possible impact on the marketplace.11
The Development of the Competitive Market for Broadband Services
Telephone, cable, computer, software, and telecommunications equipment vendors have made massive commitments with each other to upgrade their products and services for a two-way broadband communications world.12 At the same time, legal barriers have been struck down in several federal court cases,13 and many state legislatures have either granted or are considering granting basic service price regulation instead of rate-of-return regulation, in return for lower prices, interconnection with competitors, and advanced infrastructure investments.14
As with other industries, it will be the combined effect of individual companies' strategies, and the public's reaction to the products and services offered pursuant to those strategies, that will determine how the total market for access to advanced two-way broadband services will actually develop.15 Obviously, in an unregulated marketplace, deployment will take place at differing speeds throughout the country. Different suppliers, only some of whom have ever been under Federal Communications Commission (FCC) jurisdiction, may be supplying access to advanced services. Many businesses in major markets today already have multiple suppliers of two-way, high-speed narrowband access. However, there is one certainty: variations in public acceptance of particular means of access will cause higher or lower rates of market penetration and differing market shares among competing access suppliers; winners and losers will emerge, and along with them, certain technologies will become more ubiquitous and others will fall by the wayside.
It will be the marketplace winners in this initial deployment phase that will help define, de facto, the standard for which it means to have advanced access to the national information infrastructure and thus become "advanced access interfaces" (AAIs). The role of government in this environment should be to augment the forces of the marketplace, rather than interfere with them, so as to bring about the availability of advanced access on a universal basis. In order for such access to be universal, it will require legislation enabling the FCC, in partnership with the states, via a Joint Board, to administer a plan that provides the incentives for universal deployment at reasonable rates within a competitive market environment. The mechanisms of that plan, their operation, and their impact are described below.
Advanced Universal Access Deployment Plan
Phase 1: FCC-State Joint Board "Advanced Access Interface (AAI) Recognition Rulemaking Proceeding"
The marketplace success of various types of advanced network access interfaces will, as noted above, create a de facto definition of what it means to be an "information have" in American society. This definition would become de jure when penetration of potential AAIs reached a specific level, either set by Congress or delegated to the FCC or the Joint Board. The plan's purpose is to prevent the creation of a long-term "have not" community of sparsely populated or untenably high-cost service areas, while preserving the operation of the competitive marketplace.
Pending legislation uses an imprecise standard for the circumstances under which subsidies should be provided for new services.16 A specific, easily recognized benchmark is far more desirable, because investment is augmented by reducing the risk that comes from regulatory uncertainty. The benchmark could be, for example, that 80 percent of institutions and businesses and 50 percent of residences in the top 25 metropolitan service areas have advanced access interfaces that have, as a minimum, the technical capability sufficient to simultaneously receive five channels and send one channel of National Television System Committee quality video (assuming that
this reflects the abilities of the most popular residential interfaces). It is important that access be viewed in terms of capabilities and not technologies, in order to ensure that any subsidies that assist deployment are technology neutral. For example, requiring a dedicated residential 440-MHz downlink and 6-MHz uplink is not meaningful if advanced access can also be in the form of a set-top box that couples digital cellular radio linked to massive regional servers with a half-meter dish of a co-venturing direct broadcast satellite system. What is most important is that a particular advanced access interface has succeeded in the market.
Once this plateau has been reachedi.e., when the level of market penetration has become sufficient to find that a core group of "information haves" has evolvedthe implementation of the plan can begin. The first phase of the plan is an "Advanced Access Interface Recognition Rulemaking Proceeding." In such a proceeding, the AAIs with a defined minimum market share whose specifications have by this time been well documented, will be described with particularity and declared by the FCC-State Joint Board to be critical to the national interest. Companies, standards groups, associations, and others may submit their AAIs for recognition, because it will be these interfaces, and only these, whose deployment will be eligible for subsidies.17
In order to ensure the availability of access throughout the country via equipment that individual users may purchase, all of the accepted interfaces for which subsidies will be made available must, in return, be open and nonproprietary. Any proprietary rights in the AAI identified by the rulemaking proceeding become subject to strict public policy limitations; in that way, no provider of technology can hold hostage to its licensing fee or product purchase demands those customers for whom a highly competitive marketplace is unavailable.18
Phase 2: Joint Board "Inverse Auction" Proceeding
Once the AAIs have been officially recognized, the Joint Board undertakes to implement universal access through a proceeding with the following elements:
As was widely noted at the recent auction of PCS spectrum, it is highly unlikely that strategy and game theory will play a major role in the bidding process. Some of these strategies may lead to later problems if not recognized early on. For example, an applicant who already has a significant market presence and believes that economies of scale are critical to long-term profitability may "lowball" a bid to discourage potential competitors from entering the market altogether. An applicant who has historically had protection from competition may also be tempted to lowball if it believes that it can bluff the government into providing a bailout or relaxing or
changing its rules with respect to cross-subsidies, pricing, or other requirements, as an unmet deadline approaches. As a consequence, it will be necessary to have penalties for failure to perform and, perhaps, performance bonds to protect the public from parties who underbid recklessly and then seek to abandon market obligations.21
The Post-Award Competitive Marketplace
The award of UAS status would have little immediate impact on the marketplace. Unlike radio licenses or exclusive service territories, no scarce franchises are granted. All parties, UAS and non-UAS alike, would still be free to compete in deploying the infrastructure, selling to any customer on a competitive basis. Only the UAS would be required, upon request, to provide access to the requesting customer at no more than the Top Ratean umbrella price for both its own and competitors' comparable access offeringsand annually report on its progress toward meeting the deployment target date. Neither federal nor state agencies would oversee the day-to-day construction or offering of AAIs, either by the UAS or other suppliers, until the UAS filed to receive its Performance Payment. This competitive marketplace coupled with the UAS Performance Payment incentives would encourage the most rapid deployment and meet the national deployment schedule targets.
Although the competitive marketplace debate has centered on visions of "last mile" competition among local exchange carriers, cable systems, and interexchange carriers, advances in technology, auctioning of spectrum, and the lessening of regulation will open the door to many potential participants, especially consortia using combinations of diverse technologies. Among possible providers:
Distribution of the Performance Payment and Market Growth Beyond the Initial Deployment Period
At the end of the deployment target period (or prior thereto if the UAS has met the goal ahead of schedule), the UAS would petition for its Performance Payment under a "UAS Completion Proceeding." The distribution mechanism could be administered at either the state or federal level:
In either case, the subsidy would be simple to administer and should provide incentives for the desired behavior. Drawing rights could be made available to each state PSC or to the FCC by the Treasury for the amount of subsidies maturing in markets within their jurisdiction during each year. Since the state PSC or the
FCC could withhold certification and thereby delay the Performance Payment, the UAS would have an incentive to make the strongest case and be responsive to customer complaints and government inquiries. However, in order to be sure that the subsidies are directed to the purposes for which they were originally bid, neither the state PSC nor the FCC should be permitted to use the proceeding as an opportunity to negotiate quid pro quo forgiveness of missed targets in return for lower prices or regulatory oversight of operations, and so on.
The resulting high market penetration of advanced access would probably also reflect a competitive marketplace. Ideally, it should render unnecessary both continuing subsidy of remaining and growth infrastructure investment and subsidies for reasonably priced access. Nevertheless, although these mechanisms will have enabled the initial deployment of advanced access, there may be some markets where there is a continuing need to assure the availability of service out into the future or to allow for circumstances where competition may not be adequate to ensure affordable rates. If such circumstances were to occur, the state PSC could petition the FCC to provide for subsequent rounds of UAS bidding for a designated market in time blocks, i.e., bidding to be the provider of AAI access at no more than the Top Rate to all requesting parties on a last-resort basis for a designated number of years.
Funding the Performance Payment
There is already an abundance of proposals on how and from whom subsidies for narrowband universal service should be collected.23 For those required to participate in providing the subsidy, the issue is not only the participation itself, but also the annual amount and open-ended duration of the subsidy. This plan addresses those issues and limits their ability to disrupt both the marketplace and the participants' efforts at business planning.
The first principle with respect to imposing any subsidy should be to spread the burden as broadly and fairly as possible across market participants. This is especially true in this instance because there will be a variety of communications technologies and varying levels of distribution of computing power (i.e., some systems of access may rely on large regional servers connecting to comparatively simple consumer premises equipment, while others may have set-top boxes whose capabilities rival those of today's workstations). Excise taxes on the value of shipments of computers and peripherals, telecommunications equipment, and satellite ground and mobile equipment, along with taxes on LEC, cellular, IXC, and CATV service revenues, will provide that base.
Ordinarily, subsidies distort markets because transfer payments flow from the pockets of companies designated efficient and profitable into the pockets of the designated inefficient and needy. This plan neither designates nor offers preferences to any potential provider.24 Consequently, these proposed subsidies would flow only to parties providing access via interfaces that have already proven themselves successful in the marketplace. As a result, the distortion of transfer payments is minimized; the funds will flow back to the same industry participants who contributed to the fund and, potentially, in roughly similar proportion, preserving market efficiency as well as equity.
The second principle with respect to imposing a support burden is that it be predictably measured and finite. Under this plan, the national liability and maturity date for each and every market will be known well in advance of the time that any payment comes due, because the bidding and selection will occur years before the Performance Payment will be distributed. The process can be further fine-tuned by holding auctions on a staggered basis, using an accurate cross section of markets rather than the largest markets first. Such a procedure would enable the establishment of an approximate market price for each type of service area as the process went forward. The industry tax could then be set in a manner that would ensure proper levels of funding.
The collection of new taxes is always politically unpopular, but the fact that these funds would be collected from a discreet group and segregated and ultimately paid out to members of that same class of taxpayers, while simultaneously benefiting the general public in every congressional district, should alleviate much of the concern about government involvement. Since there is no day-to-day administration of a high-cost
service fund or other regulatory oversight of the potential recipients, there is no need for an independent entity to handle the funds in order to avoid a debate on the application of federalist principles. Opposition would also likely come from independent telephone companies, who have been both recipients of subsidies and protected from competition within their own service territories in the provision of narrowband services. Yet they will be positioned better than almost any other competitor to take on the role of UAS and will likely be the low bidders in their local markets; they are also likely to form partnership or consortia with other independent companies and bid successfully for the regional market.
Although this plan envisions a major role for state agencies, the overall operation of the proposal will preempt state jurisdiction over the day-to-day local activities, such as pricing, service standards, and so on, traditionally regulated at the state level. There is already, however, a recognition within many states, as noted above, that new services will be offered in a competitive, market-driven environment, where regulation does not need to be substituted for competition. With respect to the UAS itself, the state will be able, at its option, to participate in or control key aspects of the approval process. Finally, it should be noted that the plan, which will enable more rapid deployment of infrastructure, will by its operation not only improve the quality of life within the jurisdiction, but also increase commerce for both CPE and ancillary services, thereby increasing local and state sales and income tax receipts.
The ideal of relying solely on unfettered competition among multiple suppliers of broadband access to reach a goal of affordable, universally available advanced services carries with risk of failure that policymakers within the United States are unwilling to take. At the same time there is a realization that neither regulation nor managed competition are viable long-term options. The proposals set forth in this paper offer one possible avenue to achieve advanced universal access in a fully competitive environment with the least economic and social costs and minimal regulation.
1. Information Infrastructure Task Force (IITF). 1993. The National Information Infrastructure: Agenda for Action. IITF, U.S. Department of Commerce, Washington, D.C., September 15.
2. S. 652, Telecommunications Competition and Deregulation Act of 1995, 104th Congress, 1st Session, Report No. 104-23, Sec. 4–5, p. 3.
3. Ibid. at Sec. 103(d), pp. 304–310.
5. Weinhaus, Carol, and the Telecommunications Industries Analysis Project Work Group (Weinhaus et al.). 1994. Getting from Here to There: Transitions for Restructuring Subsidies, p. 16.
6. Weinhaus et al. 1994. Redefining Universal Service: The Cost of Mandating the Deployment of New Technology in Rural Areas, pp. 8–9.
7. Ibid. at p. 12.
8. Telecommunications Reports. 1995. "Ameritech to Deploy Scientific-Atlanta Technology," 61(8), February 27, p. 7; and "RFP Aims to Lower Set-Top Box Manufacturing Costs," 61(9), March 6, p. 27.
9. S. 652, Report No. 104-23, Sec. 103(d), pp. 39–40.
10. This is arguably not new regulation. Indeed, if there were no prior history of telecommunications regulation, it might well be argued that new legislation to ensure such interconnection would not be necessary. Application of Sec. 3 of the Clayton Act might reasonably be interpreted to prevent telecommunications service suppliers from denying other service providers access to its customers. See Datagate v. Hewlett-Packard, 914 F.2d 864 (9th Cir. 1991); XETA Inc. v. ATEX Inc., 852 F.2d 1280 (D.C. Cir. 1988); and Digidyne Corp. v. Data General Corp., 734 F.2d 1336 (9th Cir. 1984).
11. This paper does not discuss programs to fund or provide broadband services to those individuals or institutions determined to be in need of special subsidies. Rather, it is limited to the issue of how best to make access to advanced broadband services universally available to the general public. Nevertheless, certain principles that further the overall goal of advanced universal access do apply: in a competitively neutral marketplace, public support, whether targeted recipients are poor individuals or distressed education, health, and public institutions, must provide the recipients
with the flexibility to obtain services from competing vendors rather than require last-resort carriers to offer services at noncompensatory rates. Armed with the ability to select services from among several vendors, the needy will make choices that will reflect the efficiencies of the marketplace.
12. A recent example is PacTel's $150 million head-end and set-top box purchase agreement as part of an interactive video network. See Wall Street Journal. 1994. ''Scientific-Atlanta Inc.Concern Is Awarded Contract to Equip PacTel's Network," April 24, p. B4. Yet even this appears modest by current standards; cf. Ameritech's $700 million in contract commitments to Scientific-Atlanta, note 8, supra. A great deal of publicity has surrounded announcements by regional Bell operating companies on the formation of programming consortia (Bell Atlanta, PacTel, and NYNEX; Ameritech, BellSouth, SBC Communications, and Walt Disney), as well as speculation with regard to US West's future intentions regarding its investment in Time Warner.
13. C&P Telephone Co. of Virginia et al. v. U.S. et al., 42 F.3d 181 (4th Cir. 1994); NYNEX v. U.S., 1994 WL 779761 (Me.); US West v. U.S., 48F.3d 1092 (9th Cir. 1994).
14. States where bills were introduced in 1995, and which currently appear to be under active consideration, include Colorado (HB 1335), Texas (HB 2128), and Tennessee (rival bills SB 891 and 827, HB 695 and 721). The Wyoming Telecommunications Act of 1995 (HB 176) was signed into law in March, and Georgia's Telecommunications and Competition Development Act of 1995 (SB 137) became law in April. In North Carolina, as of this writing, HB 161 awaits the governor's signature.
15. Telecommunications Reports. 1995. "Competitors and Allies: Cable TV Companies, Telcos, Broadcasters Jointly Will Create New Markets," 61(14), April 10, pp. 16–20.
16. S. 652, Report No. 104-23, Sec. 103(d), p. 40.
17. First in the market, especially in new technology, does not guarantee dominance, or even success (e.g., 8-track cassettes, Commodore computers, Atari and Magnavox video game systems, Bowmar calculators) even if the technology is superior (Betamax, digital audiotape). Commercial success is a reflection of market acceptance and manufacturing efficiency and yields a reduction in incremental risk per unit manufactured. If the public is required to subsidize the initial buildout of long-term infrastructure, there must be a high level of confidence that the access interfaces installed under subsidy will have a long useful life and that individuals who purchase equipment and later relocate or need to access nonlocal information sources will not find their investments rendered worthless.
18. Any proprietor is free not to participate in the proceeding so that there is no state taking of intellectual property.
19. It would be reasonable to expect that the rate would be in the top quartile of rates charged nationally, i.e., still affordable by definition, but not a windfall rate when compared to those paid by the majority of customers nationwide.
20. In calculating its bid, the applicant would, of course, factor in such items as the costs of meeting the service standards obligations, the Top Rate, current and projected levels of competitive deployment in the market, demographics, terrain, and so on. The development of competitive markets will provide vast historical data on costs, price elasticity of demand, and the like. Obviously, within that price-demand elasticity curve, the higher the Top Rate, the lower the Performance Payment necessary to attract the successful bidder.
21. The UAS could be required to post a performance bond at least 5 years prior to the target date. If it advises the FCC that it is abandoning its UAS status, qualified but previously unsuccessful bidders in the market may then rebid for the right to be the UAS. The Treasury would collect on the bond to the extent that the second bid series produced a result that was higher than the original UAS's bid plus a penalty and administrative premium.
22. In multistate markets, PSCs would designate members to sit on a panel that would serve as the sole intervenor.
23. See Weinhaus et al., supra, Getting from Here to There: Transitions for Restructuring Subsidies, 1994, for a review of the most commonly discussed alternatives. Although based on recent prices paid for PCS spectrum, an auction of unused UHF and VHF spectrum could well provide all of the funds necessary to finance advanced universal access; for the purposes of this paper, it is assumed that the funds must be raised from market participants.
24. This is particularly important if, by operation of the program, a UAS receives what could be considered a windfall profit: if the market develops more rapidly and efficiently than anticipated and the performance targets are met with little or no investment that is less than fully compensatory, the UAS would receive the Performance Payment merely for having been "on call."