April 10, 1998

Re:	Report to Congress, Federal-State Joint Board on Universal Service (CC Docket No. 96-45)

I welcome the opportunity provided by this Report to Congress, both to underscore my strong support for the goals of the universal service provisions of the 1996 Act and to share my current views regarding some of the principles I believe should guide the Commission's implementation of these provisions. I write separately (1) to highlight the challenges the Commission and Congress face as technological convergence erodes the foundations of our balkanized regulatory framework, and (2) to express my growing concern that we need to modify our universal service programs, particularly the Schools and Libraries (S&L) program, in order to more accurately meet demand and to limit their distorting effect on competition and consumer prices.

I. Internet Protocols and the Strain on Our Statutory and Regulatory Framework.

A. The Problem of Convergence

Congress has asked that we re-examine our interpretation of the critical terms "telecommunications," "telecommunications carriers," and "information service providers" contained within the universal service section of the 1996 Act.⁽¹⁾ At bottom, the question is whether these terms are mutually exclusive. That is, can a single service provider be offering both "telecommunications services" and "information services"? Or, as the Commission has previously held with respect to its treatment of "basic" and "enhanced" services, is a provider either one or the other.

I believe the Report rightly concludes that Congress intended to maintain the dichotomy between telecommunications services and information services that originated in the Commission's Computer decisions and was utilized in large measure by the court in the context of the divestiture of AT&T. Though I concede that there is merit in the views advocated by some that Congress intended to redefine telecommunications so as to capture information service providers within the snare of Title II and other regulation, I believe the express language of the Act and its legislative history prove otherwise, as the Report demonstrates.

Clearly, how the Commission chooses to categorize a provider under the statute significantly affects its regulatory treatment. A telecommunications carrier is subject to the full panoply of common carrier regulation and must contribute to universal service. A provider of telecommunications (one not offering telecommunications on a common carrier basis) may have to contribute to universal service if the Commission finds that it is in the public interest to do so. And, finally, an information service provider is neither subject to regulation nor, in my current view, should it be required to contribute to universal service.

Even though we are convinced that Congress in fact adopted a categorical "either/or" scheme in 1996, we are left with the difficult task of categorizing service offerings that do not fit neatly in either category, unless and until Congress makes a change in this scheme. This difficulty arises because the definitional dichotomy does not fully account for the explosion in innovation brought about by the Internet and its underlying network architecture. That dichotomy is premised on a network in which the nature of the service and the underlying infrastructure are highly integrated. Basic voice service was (and largely still is) distributed on networks constructed and optimized for that service, and the brains of that network resides within the central switches, signalling systems, and databases of the intelligent network owned by the operator, usually a Bell Operating Company (BOC). The Internet and the Internet protocol (IP) represent a dramatic paradigm shift in network architecture. A variety of services can be overlaid on an IP network and the intelligence of those services rests not centrally, but at the edges of the network -- in the case of the Internet, in the computers owned by the millions of users and information service providers throughout the world. Anyone with the right computer and software can offer and distribute new and innovative goods and services.

The advent of IP networks has placed great strain on the categorical definitions first set out by the Commission and adopted by Congress. Previously, we assumed that one could categorize a service by the degree to which the transmission was "processed" or altered by computers on the network. Basic service (assumed to be voice) traveled without manipulation or change in form to its destination. Thus, it was said to flow on a "transparent transmission path." This definition is being strained today even for voice networks. For example, in modern digital voice networks, such as the new cellular and PCS networks, advanced digital signal processing means the voice signal is being processed and manipulated between origin and destination. More importantly, information service (assumed to be data) involved the additional manipulation by computer of the transmitted information.

The infinite flexibility of IP switched-packet networks, has blurred these distinctions, making them difficult, if not impossible, to maintain. As we are seeing, one now can transmit voice, in addition to data, using a protocol that allows for a significant degree of computer processing and other advanced capabilities. Yet, from the perspective of a user who does not use those capabilities, the service may look nearly identical to traditional basic service. Were such a service to be classified permanently as an information service it should not, in my current view, be required to contribute to universal service. If innovative new IP services were all thrown into the bucket of telecommunications carriers, we would drop a mountain of regulations, and their attendant costs, on these services and perhaps stifle innovation and competition in direct contravention of the Act.

Therefore, the challenge we face is how to categorize the growing number of hybrid services, in light of the Act's twin objectives of promoting competition and advancing universal service. Hybrid services are those that have components of both a basic service and an information service (as traditionally defined). On one end of the spectrum, for example, may be some forms of IP telephony that resemble traditional basic service in nearly every way, except that they have at their core a network and protocol that makes it possible to substantially enhance the basic voice service (for example, speak French and have it come out in English or have some users choose higher fidelity sound than others). On the other end of the spectrum are service providers whose fundamental function is to offer access to information to its customers using computer intelligence, but whose service rests on a transmission network fully capable of providing a clear path for basic transmission (same thing in, same thing out). Sorting hybrid services into their appropriate regulatory bin is difficult, yet something we will be forced to do more and more as new and innovative services explode from the fuel of IP networks. This reflects the growing challenge of adapting a balkanized regulatory structure to a world of technological convergence.

Congress is right to be both buoyed and fearful of this development. On the up-side, the arrival of digital technology and IP networks mean infinite possibilities for new and innovative services. The arrival of broadband digital networks promise to connect the world to an endless sea of information, significantly advancing how we buy and sell, communicate, create, and educate. Moreover, this technology will produce more competitive choice for consumers by eroding the traditional barriers between services and lowering the costs of entry. This hopeful state of affairs was born and has flourished in a deregulated and competitive atmosphere, the very type we are striving to achieve in communications services generally. On the down-side, these alternative network systems may present a threat to our universal service goals. Carriers that presently pay to support universal service may migrate more of their traffic to IP networks in a rational effort to avoid the cost of government regulation. Congress, will have to watch these developments vigilantly and take action if it becomes necessary, but until that day, what is an agency to do?

B. A Call for Case-By-Case Evaluation

As long as our universal service obligations require us to consider whether difficult-to-classify entities are "telecommunications carriers," provide "telecommunications," or are "information service providers" I think we should resolve those questions on a case-by-case basis. If the Commission is to preserve the dichotomy between information services and telecommunications, as the statute and legislative history appear to require, we will need to make highly fact-specific inquiries about whether particular entities provide transmission of information "without a change in the form or content" or whether the entities offer the capability of manipulating that information.

Attempts by this agency to set down its own prophylactic rules for categorizing classes of IP-based service will be, in my current view, futile if not dangerous for a number of reasons: First, the English language is no match for the infinite flexibility and innovation potential of an IP network. I am confident that any attempt to craft a rule to cover a class of IP-based service will be almost immediately frustrated by innovative changes to the service and technology that these advanced networks allow. I fear we would find ourselves in a never-ending chase for regulatory clarity. Second, adopting a rule that invades on a broad front the Internet field and its underlying technology is likely to chill, if not freeze, innovation in broadband digital services, and constrain the flow of capital investment in these growth industries, out of fear that the regulator and the tax-man cometh. Third, any rule will likely be over-inclusive and mire the Commission in waiver proceedings. What will undoubtedly result is a perforated rule -- one besieged by exceptions. Fourth, if we adopt broad rules, we wrong-headedly move in the direction of expanding the onerous body of regulations, rather than staying on the path of deregulation the Act commands us to take. Fifth, we risk serious loss of credibility internationally, having fought hard to win world trade concessions (e.g., international settlements, classifying the Internet as an information service).

These considerations favor resolving these matters (which I believe encompass the classification of Internet and IP telephony) on a case-by-case basis, rather than through rulemaking. It is for these reasons, I question the wisdom of even suggesting conclusions in the Report. Such conclusions, no matter how gingerly or narrowly drawn, signal a move toward developing a body of rules for classifying Internet-based services. I urge the Commission to seek comment on the appropriate method of designating new contributors to universal funds in the context of a future proceeding.

C. The Pitfalls of Competitive Neutrality Analysis

Some will argue that it is competitively unfair to subject competing communications services to regulation and not Internet companies. I constantly hear the mantra that we must "level the playing field," because exempting Internet service providers constitutes a massive subsidy. While this argument has some superficial appeal, I strongly caution against extending regulations solely on this basis.

I simply disagree with those who argue we are massively subsidizing the Internet by letting it operate in a free market while other companies labor under the yolk of government regulation. That seems to be an ironic characterization in light of the Act's stated goal of fostering a pro-competitive, deregulatory environment. The way to level this disparity, if at all necessary, is not to extend government imposed costs and regulations to the Internet, it is to take further actions to liberate those subject to regulation.

Moreover, competition is not a game of equally matched players. Competitors have different mixes of competitive advantages and burdens. It is too simple to focus on a single competitive inequity and then declare the game unfair, without examining the totality of advantages and disadvantages among competitors. Let me be clear: I believe it is entirely appropriate for the Commission to consider the extent to which its decisions will have a distorting effect on the market. We should not lose sight, however, of the fact that we can never avoid all such distortions so long as we regulate the market.

Further, we should recognize that competitive advantages are not limited to whether or not companies must contribute to universal service support. For example, while telecommunications carriers must contribute to universal service, they also enjoy significant benefits under sections 251 and 252 of the Act. Likewise, incumbents often have significant advantages over new entrants, such as capital, business and marketing savvy and technical expertise, such that declining to require new entrants to contribute to universal service may not give the new entrants an appreciable advantage over the incumbent. Leveling the playing field, when the players do not start from the same place, only institutionalizes the advantage of the stronger, better equipped, experienced players. We and the Congress regularly and consciously provide incentives for innovation and market entry.

Again, we should recognize that competitive advantage is not always an evil to be stamped out; the market is designed to reward companies for finding and exploiting their competitive advantages over other companies. Our goal should be to ensure that our policies do not have a significant impact on the overall competition between entities.

For these reasons, I am troubled by the discussions in which the Report suggests that new types of entities (e.g., ISPs that use their own transmission facilities, phone-to-phone Internet telephony providers) should be required to contribute to universal service based, at least in part, on "competitive neutrality." I am concerned that these discussions, among other things, may overstate the potential effect of declining to designate new contributors to universal service support and, in any event, oversimplify the competitive context. The danger here is that, as new technological and marketing innovations bring new entrants to the market, we will continue to expand the pool of contributors, whether or not we need additional contributors to keep the fund sufficient. Even worse, by continuously expanding the pool of contributors to encompass new entrants, we may discourage such entry.

I want to emphasize that we do have a duty to maintain a sufficient base of funds to support universal service. We can and will do so. As I explain below, however, we also must not throw our net out farther than is necessary, because unnecessarily expanding the scope or size of the contribution base for universal service would distort and inhibit competition and would put significant pressure on consumer prices for services subject to the government levy. In short, we should address these issues with a surgeon's scalpel and not a butcher's cleaver.

II. Tailoring Universal Service Programs to Sufficiency

One of the hallmarks of the Commission's implementation of the Act's universal service mandate should be that the funds must become and remain "sufficient" within the meaning of section 254(e). Section 254(b) requires that the Commission preserve and advance universal service and the Commission must establish mechanisms to meet that goal. At the same time, it would defy common sense, as well as the pro-competitive, deregulatory thrust of the Act, for the Commission to expand either the size or the scope of universal service programs unless necessary to fully-fund these programs or to otherwise satisfy the requirements of the Act. Thus, "sufficiency" under the statute is in essence a question of balance: our universal service funds must be sufficient to preserve and advance universal service, but these funds must not become larger than is necessary to achieve those goals. If we do not balance these contending objectives, we will unduly distort competition and add to the cost of service, which will likely result in higher rates to consumers.

I am concerned that the Report does not focus directly enough on the importance of maintaining the sufficiency of universal service funding. As the Report indicates, the Commission does not yet know how much any of the programs will cost; we will not know the demand for the Schools and Libraries program until the window for filing applications closes later this month, and we will not even decide the method for determining high cost support until later this year. With respect to Schools and Libraries, in particular, I am troubled that we continue to operate and collect based on initial assumptions about the demand and scope of the program. I fear the eagerness to keep this program moving, which I fully understand, may lead us inadvertently to overcollect or unnecessarily expand the program.

I am concerned that, by not squarely re-evaluating the question of sufficiency, we exacerbate a perception among many critics of our universal service implementation that somehow these programs are out of balance. This perception, paradoxically, is perhaps the most significant threat to the sufficiency of universal service funding. If we cannot find a way to make critics in Congress and elsewhere believe that we are working to preserve and advance universal service in a prudent and responsible manner -- that the funds will be sufficient but not too large -- I fear that support for these beneficial programs will erode in the minds of both legislators and consumers.

With respect to the Report's review of Commission actions taken in and subsequent to the Universal Service Order,⁽²⁾ questions of sufficiency and, conversely, overcollection are central. Congress has expressed its concern regarding the sufficiency of universal service programs in the Appropriations Amendment, which required the Commission to review its determinations regarding who is required to contribute to universal service, as well as the revenue base from which universal service support is derived. More generally, concern for sufficiency and overcollection are evidenced in several critiques of current universal service programs.

Specifically, in my current view, much of the concern regarding the size and scope of the S&L program centers around the perception that, intentionally or not, the Commission has cast its net broadly to capture substantial funding and taken more and quicker steps to ensure that that program will be sufficient than it has taken with respect to the high cost and other programs. Indeed, one could argue that the S&L program may be poised to overcollect: First, the Commission concluded that, despite the fact that States may establish their own S&L programs, the federal S&L program would be funded out of both interstate and intrastate revenues, whereas high cost and low income support would be based only on the former.⁽³⁾ Second, the Commission set the cap for the S&L program at $2.25 billion, even though projected first-year demand for the program is substantially less than that amount.⁽⁴⁾ Third, the Commission directed the establishment of a separate corporation to administer aspects of the S&L program, rather than allowing the fund administrator to run the entire program, as is the case for the high cost and low income funds.⁽⁵⁾ Fourth, as a general matter, critics point out that the S&L program is already collecting funds, whereas the high cost program will not be fully fleshed out until later this year and will not begin collecting money until next January. Indeed, many states clamor, and the Report concedes, that the percentage of federal funding adopted by the Commission in the Universal Service Order,⁽⁶⁾ if not modified on reconsideration, would ensure that some high cost states will receive substantially less federal support than they do currently.

While these decisions, taken individually, constitute a reasonable exercise of the Commission's discretion, the overall picture sketched by these decisions suggests to some critics that the Commission has taken more pains to ensure that the S&L program is sufficient than it has taken with respect to the high cost and low income programs. It is not lost on these critics that section 254 provides no basis for the Commission to favor certain classes of recipients over others with respect to the level or timing of universal service support flows.⁽⁷⁾ Thus, I have become increasingly concerned that, from the standpoint of sufficiency, the S&L program is, or at least appears to be, out of balance.

I support a vigilant approach to ensuring that the sufficiency of universal service be kept in balance, as a general matter and with respect to particular programs, such as the S&L program. As general matter, the Commission should establish an "early warning system," whereby we regularly assess whether the funds and the pool of available contributors are sufficient to satisfy statutory requirements. These efforts could build on the work already begun by the Common Carrier Bureau pursuant to the monitoring authority delegated to it in the Universal Service Order.⁽⁸⁾ In establishing such a monitoring system, we could request comment on whether specific new technologies or types of providers are contributing indirectly to the funds by, for example, generating new revenues for carriers that are required to pay into the funds.

With respect to Schools and Libraries, I believe that the Commission must quickly commit itself to a modest modification of the program if it wishes to maintain crucial political support for it. To be blunt, I fear that if we do not move quickly to modify our approach to implementation of this and other programs, we will fail to carry out adequately our universal service duties, endanger the pro-competitive and deregulatory goals of the Act and make it necessary for Congress to step in to show us what the public interest requires. With respect to the S&L program, some have raised substantial questions regarding whether the Commission has jurisdiction to assess contributions based on intrastate revenues, as well as important policy concerns. On this basis, and in light of the perception that the Commission has somehow favored the S&L program, I would support limiting that program to assessment of contributions based on interstate revenues only. Likewise, I would favor reducing the cap on the S&L program so as to bring it down to current demand levels. Further, I would support moving the functions performed by the S&L Corporation to the fund administrator, both to remove the appearance that the Commission is favoring the S&L program and to resolve serious questions raised by the General Accounting Office regarding whether the Commission was authorized to direct the establishment of the S&L Corporation.⁽⁹⁾

III. Conclusion

In closing, I remind my colleagues that the Commission is not alone in the effort to promote universal service. I, for one, believe that if we reach a point where there is a potential shortfall in universal service funds, Congress could craft additional legislation narrowly tailored to assess universal service contributions without the accompanying risk of heavy regulation (perhaps to include express preemption of State efforts to regulate the Internet). In the meantime, I urge the Commission to be mindful of the threats to competition and innovation if we do not keep the sufficiency of universal service funds "in balance" both as to their size and scope. By following some of the approaches I have described here, I believe we would greatly benefit the development of competition and innovation by refraining from imposing contribution obligations on entities that do not fit neatly into the traditional categories. Just as importantly, by taking swift action to ensure that the funds are sufficient to preserve and advance universal service, but are no larger than is necessary, we will do much to quell the criticisms that threaten to undermine support for these beneficial programs.

1. See 47 U.S.C. § 254.

2. See In re Federal-State Joint Board on Universal Service, Report & Order, CC Docket No. 96-45, FCC 97-157 (rel. May 8, 1997) (hereinafter "Universal Service Order").

3. See Universal Service Order, ¶ 823; compare id., ¶ 831.

4. See id., ¶ 425.

5. See NECA Report & Order, ¶ 57.

6. See Universal Service Order, ¶ 269.

7. See generally 47 U.S.C. § 254.

8. See Universal Service Order, ¶ 869.

9. See Feb. 10, 1998 Letter from Robert P. Murphy, General Counsel, General Accounting Office to Senator Ted Stevens.