Building on What Works: An Analysis of U.S. Broadband Policy.

• Author: Nuechterlein, Jonathan E.

TABLE OF CONTENTS I. INTRODUCTION AND SUMMARY 220 II. THE COMPETITIVE DYNAMICS OF THE U.S. BROADBAND MARKETPLACE 224 III. ASSESSING THE COSTS AND BENEFITS OF CURRENT PROPOSALS FOR BROADBAND REGULATION 233 A. A Brief History of the U.S. Approach to Broadband Regulation 233 B. The Costs and Benefits of Proposals for New Broadband Regulation 235 1. Facilities-Sharing Obligations 235 2. Rate Regulation 239 3. Interconnection Obligations 241 4. Open-Ended ISP Conduct Rules 245 C. State-Level Economic Regulation 252 IV. RECONCILING COMPETITION POLICY WITH SOCIAL EQUITY 254 I. INTRODUCTION AND SUMMARY

This year marks three milestones in telecom policy. Each conveys an important lesson for policymakers as they contemplate broadband regulation for the 2020s and beyond.

First, it has been ten years since the Obama FCC released the National Broadband Plan, which surveyed the U.S. broadband landscape in 2010 and offered policy recommendations for boosting deployment and adoption. In many respects, the Broadband Plan was a case study in regulatory humility. It recognized that broadband progress was "[f]ueled primarily by private sector investment and innovation"; that "government cannot predict the future"; that "the role of government is and should remain limited"; and that policymakers should thus focus not on imposing price controls or behavioral restrictions, but on "encourag[ing] more private innovation and investment." (1) This advice, which the FCC has generally followed, has fared well under the test of time. Ten years and hundreds of billions of investment dollars later, the broadband marketplace now offers consumers more choices and exponentially faster speeds than it did then. The Plan was also eerily prescient. In one passage, it anticipated "surge[s] in residential broadband network use during a pandemic" and the need for "high standards of reliability, resiliency and security." (2) Those are standards that U.S. networks have more than met during the COVID-19 pandemic, as broadband usage has surged. (3)

Second, it has been twenty years since the FCC issued the 2000 Notice of Inquiry seeking comment for the first time on "the appropriate legal classification of cable modem service" and "what regulatory treatment, if any, should be accorded" to it. (4) As the NOI noted, the FCC had consistently "taken a 'hands-off policy" for cable broadband services, (5) then the dominant form of broadband Internet access. The basis for that policy, which the FCC reaffirmed in later orders, was best summed up by then-FCC Chairman Bill Kennard, for whom we both worked at the end of the Clinton Administration. As he explained:

We sometimes get so caught up in the policy debates about broadband...that we forget what we need to do to serve the American public....We have to get these pipes built. But how do we do it? We let the marketplace do it....[T]he best decision government ever made with respect to the Internet was the decision that the FCC made...NOT to impose regulation on it. This was not a dodge; it was a decision NOT to act. It was intentional restraint born of humility. Humility that we can't predict where this market is going. (6) Twenty years later, private enterprise has invested more than a trillion dollars to "build the pipes." As a result, the typical American can now choose among multiple competing broadband services--both fixed-line and mobile--at speeds nearly unimaginable in 2000. And broadband ISPs made these investments against the backdrop of a light-touch regime that, with rare exceptions, declined to apply significant economic regulation for two decades. It is difficult to prove causal links between regulatory choices and specific investment decisions, but as a matter of economic logic, more intrusive forms of regulatory intervention would likely have reduced, not increased, incentives to commit private risk capital to broadband infrastructure and innovation.

Third, it has been 25 years since the House and Senate issued the bills that became the Telecommunications Act of 1996. (7) Much has been written about that legislation, both positive and negative. Among its undeniable achievements, the 1996 Act eliminated anticompetitive exclusive franchises, began rationalizing universal service mechanisms, and consolidated competition policy at the federal level at a time when technology had begun blurring the traditional distinctions between "intrastate" and "interstate" services. (8) But the 1996 Act is also notorious for launching years of unproductive regulatory churn, mainly surrounding the interventionist "unbundling" rules the FCC designed to mimic but not necessarily produce genuine facilities-based competition in landline telephone markets. (9) Those rules show how even the smartest regulators can do more harm than good if they underestimate prospects for facilities-based entry--in that case, the looming ascendance of mobile networks and VoIP technologies over the landline telephone network--and overestimate the efficacy and administrability of complex regulatory obligations.

These three milestones should inform today's debates about broadband policy. The National Broadband Plan reflects the best traditions of the FCC'S professional staff: focusing on the facts, recognizing the complexity of markets, and promoting policies that enhance rather than undermine private incentives for investment and innovation. Chairman Kennard's turn-of-the-milennium policy of "intentional restraint born of humility" reflects the same commitment to data-driven broadband policy. But the network-sharing regime adopted under the 1996 Act offers a more cautionary tale. It illustrates the costly detours that telecom policy can take when well-intentioned regulators pursue novel schemes of regulatory intervention on the mistaken assumption that markets would stagnate without them.

The broadband industry today is more technologically dynamic and competitive than the landline telephone industry of 1996, but the two share one similarity. Much like the turn-of-the-millennium telephony market, the broadband industry is in new period of technological transition. Fixed-line networks are deploying technologies that support increasingly mobile functionality, while mobile networks--first with LTE and now with 5G--are increasingly capable of cost-efficiently supporting high-bandwidth services that were once the unique province of fixed-line networks. (10) If experience with the 1996 Act taught us nothing else, it is that policymakers must be careful neither to exaggerate the need for major intervention in such transitional markets nor to overlook the costs of doing so.

Unfortunately, current proposals for market intervention are often long on rhetoric and short on real analysis of likely tradeoffs and actual consequences. This paper thus analyzes the asserted need for, and likely consequences of, four types of proposals in recent circulation: (1) facilities-sharing obligations, (2) retail price controls, (3) Internet interconnection obligations, and (4) amorphous and open-ended ISP conduct rules like those the FCC imposed on consumer broadband services in 2015. (11) For the most part, we see little merit to any of these proposals under current market conditions. None of them addresses any identifiable market failure and each would impose significant costs, including the investment-chilling prospect of regulatory creep. That said, we support re-imposition of bright-line prohibitions on blocking or throttling to guard against any risks to the Internet's status as an open, positive-externalities-generating platform for communication and innovation. Although those risks appear remote, such bright-line rules would reduce them to zero and impose minimal costs because such rules would simply codify what have become industry norms in any event.

All this said, government retains a critical role to play in the broadband marketplace. Market forces are unmatched in their power to bring the greatest benefit to the greatest number. But market forces by themselves will not help America close two stubborn and unacceptable digital divides: between rich and poor, and between urban and rural. (12) As the COVID-19 pandemic underscores, broadband is critical to equal opportunity and to full participation in civic and economic life, but underemployment has made it unaffordable for many Americans. At the same time, many Americans in rural areas cannot buy the connectivity they need at any price. The great broadband challenge of the next decade is to close both divides by boosting adoption in low-income communities and deployment in high-cost areas.

These are real, universally acknowledged problems that call for real solutions. In particular, they call for expanded subsidy mechanisms--one directed to low-income subscribers and the other to broadband providers that commit to new infrastructure deployment in rural and other high-cost areas. But the challenge of closing these digital divides does not even logically support a call for more intrusive regulation of the broadband industry. To the contrary, such regulation would, if anything, make the underlying problems worse by placing a thumb on the scale against additional broadband investment.

This paper is divided into three main sections. Section II addresses the types of market conditions that do--and do not--call for economic regulation, the focus of this paper. By "economic regulation," we mean rules intended to constrain the exercise of market power (e.g., retail rate caps) or force firms to cooperate with other firms, including their rivals (e.g., asset-sharing, interconnection, and "neutrality" obligations). (13) As we discuss, a rigorous analysis of tradeoffs and consequences generally disfavors economic regulation in industries that, like broadband, are technologically dynamic and subject to competition. Section III then summarizes the history of light-touch broadband regulation in the U.S...