Re:       WC Docket No. 23-320, Safeguarding and Securing the Open Internet

Dear Ms. Dortch,

I[1] offer these replies to aid the Commission in reaching the proper conclusion in the matter of the regulatory classification and treatment of Broadband Internet Access Services (BIAS.) The FCC re-opened this issue in an NPRM released on Oct. 30, 2023.

Summary

Comments in support of the NPRM tend to repeat arguments made in support of the hyper-regulation of broadband Internet service providers (ISPs) over the last two decades. The theme of these arguments can be summarized:

1. The market for Internet services is dominated by monopolies with gatekeeper power.
2. Even in competitive markets, ISPs have a new kind of monopoly, a “terminating access monopoly,” that reinforces their power.
3. Gatekeepers want to extort edge service providers such as Google, Facebook, X, and Netflix for access to Internet users on favorable terms.
4. Privacy and free speech are uniquely threatened by gatekeepers.
5. The only way to make gatekeepers behave is to shackle them with (often vague) regulations enforced by an expert agency according to the pressures of the moment.
6. If ISPs behave according to our wishes, the rest of the Internet will be fine.

When these ideas emerged from the Stanford Law School in the late 1990s, the ISP market was highly concentrated. Professors Mark Lemley and Lawrence Lessig and their proteges Tim Wu and Barbara van Schewick wanted structural separation between monopoly broadband providers and over-the-top ISPs following the blueprint of laws passed in the European Union and, in some readings, by the 1996 Telecom Act.[2]

When the courts refused to uphold structural separation in the US, these advocates turned to behavioral rules to protect the openness and neutrality of the Internet. Even with structural separation, the EU adopted behavioral rules of its own because they judged the results of structural separation to be less than ideal, chiefly for its corrosive effects on innovation, investment, and free expression.

The only new arguments deal with issues cited by the DC Circuit Court in the Mozilla v. FCC case. These secondary concerns are described as pole attachment, Lifeline, public safety, national security, and privacy. While the issues are real, they can all be resolved without Title II reclassification, as we can see from the fact that the FCC’s Title II order was only in effect for two years while the FCC managed to keep the sky from falling all along.

The NPRM repeats this line of argument and the comments by pro-Title II organizations (T2O) such as Free Press, Public Knowledge, New America’s Open Technology Institute, Engine, and the Electronic Frontier Foundation reinforce it.

Fortunately, the ISP marketplace is wildly less concentrated and more competitive today than it was prior to the turn of the century. In today’s market, there is no longer a rationale for bright line rules banning conduct that has never taken place on a substantial scale. There is significant risk that the ban on for-fee traffic shaping (AKA “paid prioritization”,) the mandate for free interconnection, and a murky general conduct standard will be applied in ways that prevent more sophisticated network management and greater network efficiency.

It is also concerning that T2O advocates continue to blame virtually all the Internet’s shortcomings on ISPs. The disregard for privacy, threats to national security, the lack of consistent and reasonable speech moderation practices, weak network management by public safety, and lax cyber defenses overall will not be curbed by the order contemplated by this NPRM.

The RIF Order does need to be recast so that the FCC retains authority over the secondary issues as well as the authority to preempt state laws in conflict with federal policy. Reliance on ancillary authority has always been sufficient for this purpose in the past.[3]

The Actual State of the Broadband Market

As we pointed out in our original comment, the marketplace for broadband Internet service in the U. S. is healthier than it has ever been.

1. Performance metrics continue to exceed all-time highs.[4]
2. The number of nationwide ISPs has tripled since 2018.[5]
3. The number of markets served by more than three broadband providers is at a record level.[6]
4. Most Internet edge traffic is wireless.[7]
5. TV has largely switched from linear cable and satellite to Internet-based services.
6. Video conferencing is routine, built-in to current operating systems and useable by children.
7. The rate of fiber optic installation is only limited by available workforce.
8. The rate of FWA installation is limited only by available spectrum.
9. LEO satellite launches by SpaceX alone are expected to take place every 2.5 days in 2024.[8]
10. Newer LEO satellites connect directly to newer mobile phones.
11. Reports of ISP misbehavior with respect to traffic management are virtually non-existent.
12. Wireless Internet subscribers are more satisfied with their services than those who subscribe to completely wire-based offerings.[9]
13. While gatekeeping-as-a-service is lucrative for Cloudflare, malicious gatekeeping doesn’t exist in the U. S.
14. Prices of networking silicon have fallen so low that Apple no longer sees value in making its own Wi-Fi and 5G chips.[10]
15. Share prices of semiconductor manufacturers involved in general-purpose computing and AI are skyrocketing.[11]

None of the comments from T2O acknowledge these realities, let alone attempt to consider their impact on the need for strict, bright line rules prohibiting departure from the Commission’s understanding of the Internet’s history and essential architecture. For example:

• Free Press does better than the rest of the T2O crowd by mentioning fixed wireless broadband, even if it initially does so to downplay its importance:

“And though there are some green shoots of potential competition from fixed wireless providers, the home broadband market remains a rigid, cable-dominated duopoly. Furthermore, for the foreseeable future, tens of millions of people living in high-cost rural areas will continue to face a fixed terrestrial broadband monopoly.”[12]

Grudgingly, Free Press then admits that FWA providers are bullish:

“Fixed wireless services have existed in the marketplace for decades, typically offering a workable solution in high-cost areas where wired deployment is unprofitable. While these services were historically offered by small firms using unlicensed spectrum, the Commission has adopted policies that further enhance these providers’ capacity and access to the public airwaves. Recently, CMRS providers have entered the fixed wireless market, led by T-Mobile, then Verizon, and more recently, AT&T. These fixed wireless services appear to be popular with certain business customers, but are also increasingly an affordable option for residential users who may not need the higher capacities of wired services. See, e.g. , Verizon Dec. 5, 2023 Comments (“So we have said by ‘25, we should have 4 million to 5 million subscribers on fixed wireless access. Of course, we have built the network for more. The team has way more capacity.”)[13]

Finally, the leader of T2O becomes ever-so-slightly-optimistic about FWA in order to highlight the Lifeline problem:

“Fixed wireless is now the fastest-growing consumer broadband technology, and offers an easy-to-install, affordable option compared to traditional wired access services.”[14]

What we don’t see in these comments is much appreciation for the likely effects of three or four new residential broadband entrants in a market that was once highly concentrated. Wouldn’t it be nice if we had six or seven viable search engines or social media platforms? In reality, the ISP marketplace has moved from highly concentrated to massively competitive.

For all but the most arrogant of regulators, the transformation of this marketplace would at least prompt musing about the future of ISP regulation. Granted, FWA is still an emerging market with a number of limitations in terms of spectrum resources. But Congress paved the way for FWA by preempting local zoning authority over small cells in its 2012 Spectrum Act (embedded in Section 6409 of the Middle Class Tax Relief and Job Creation Act).

Following the Spectrum Act, the FCC picked up the ball and scored with its 2018 Small Cell Order preempting state and local zoning authority over 5G small cells. [15] The FCC’s authority to promote progress was upheld by the Ninth Circuit on August 12, 2020.[16]

Lawyers may well see this case as an affirmation of the FCC’s powers of preemption, but our interest is with the constructive role the FCC can play in nurturing competition. Contrast the Small Cell order with the obstacles the current FCC has erected against the progress of Low Earth Orbit (LEO) satellite constellations as yet another font of competitive choice.[17] None of the T2O comments mentions LEO constellations.

• Public Knowledge only mentions FWA once, in the context of arguing that broadband is a telecommunication service because market players emphasize transmission properties:“A brief survey of television and online advertising for both mobile and fixed broadband shows that ISPs compete with each other on the basis of speed, price, ease of use, reliability and availability. For example, cable operators claim that their networks have superior regular performance to fixed wireless competitors. Fixed wireless providers respond with advertisements emphasizing that they provide the same broadband capacity and speeds as cable broadband, but cheaper and easier.”[18]In fact, FWA players emphasize ease of installation among other benefits. FWA is generally bundled with a home router that provides 5G and Wi-Fi along with a cached DNS.
• New America’s Open Technology Institute fails to mention FWA and only considers wireless at all in connection with the regulatory treatment of mobile cell service. This is particularly odd, given OTI’s fascination with Wi-Fi.
• Engine fails to mention FWA and only mentions wireless in connection with denunciations of zero-rating. To be fair, the bulk of its comment is a regurgitation of a comment it made in 2017 on the RIF Order. FWA wasn’t much of a thing in 2017, but RIF’s deregulatory posture toward new technologies certainly didn’t hurt its deployment.
• EFF mentions fixed wireless once, merely in passing:“But many Americans have little or no choice as to who will provide that [broadband] service. For example, Comcast and Charter have an absolute monopoly over at least 47 million people, and another 33 million people’s only “competitive” alternative was slower and less reliable DSL. Further, the broadband market has also only further consolidated since the Commission last considered net neutrality regulations. In 2023, just 11 providers, some wireline and some fixed wireless, represented 96% of the high-speed internet access market.”[19]This is a very odd remark, given that FWA is the fastest-growing part of the ISP market today, and that customers are more satisfied with FWA than they are with cable. Once the technology leader of the T2O, EFF would do well to absorb Free Press’s comment: “Fixed wireless is now the fastest-growing consumer broadband technology, and offers an easy-to-install, affordable option compared to traditional wired access services.”

  This clearly contradicts EFF’s claim that the ISP market is increasingly concentrated.

The Presumed State of the Broadband Market

While the Title II lobby doesn’t like to acknowledge or discuss the realities of ISP competition brought about by the development of and investment in FWA and LEO, they’re fairly unrestrained in their praise for competition and denunciation of monopoly. Many of them argue that Title II will bring about increased competition among ISPs.

Free Press: “Title II also gives the FCC the power to protect and promote competition in communications markets.”[20]

Public Knowledge: “Title II can promote competition among ISPs as well as among edge providers. First, if the Commission grants the request by INCOMPAS to assert authority over interconnection, it will have the authority to prevent abuses that may arise in the interconnection context.”[21]

EFF: “EFF knows that unfair data discrimination by internet service providers is a serious problem that, given the general lack of competition in the market for broadband Internet access, must be addressed by effective consumer protection rules.”[22]

In principle and by mission, the FCC certainly can encourage and promote competition among ISPs. The agency has many tools for achieving this goal, some more effective than others. I would challenge the proponents of Title II to demonstrate a regulatory regime that has promoted more nation-wide ISPs than the status quo. If the FCC goes forward with its Title II plan, analysts will certainly look at the numbers of new ISPs in the five years following the transition.

As long as the competition dialog was limited to edge services, it was possible to make a case for the pro-competitive effects of Title II. But Title II has always failed to create new networks. In fact, that was never its mission. Title II was created by a group of lawmakers seeking to create competition in the market for residential voice service.

It relied on the massively complex scheme of wire sharing pioneered in nations with government-owned and -operated voice networks in the 1980s.[23]

Indeed, we have competition for voice services in the US today, but not because of Title II. Mobile phone technology held the key then, just as wireless technologies FWA and LEO do today.

Conclusion

The FCC can make the transition from wired to wireless ISPs faster or slower by controlling the supply of licensed spectrum, but there is very little the agency can do to stop it altogether. Just as consumers reaped benefits from bundling linear TV with broadband at the turn of the century, today’s consumers benefit from bundling residential broadband with mobile broadband. Rather than expending its resources trying to reanimate defunct regulatory paradigms, the FCC would be wise to promote the new normal.

Brightline rules against blocking and throttling aren’t terribly problematic because such practices don’t benefit players in a competitive environment anyway. But banning something that no sensible business has ever wanted to do is superfluous. In today’s world, gatekeeping is a valuable service that has made Cloudflare a billion-dollar business. If blocking denial of service attacks is good when Cloudflare does it, it’s certainly worthwhile when and if ISPs do it.

Similarly, so-called “paid prioritization” can be beneficial in some circumstances. The Magnificent Seven have built moats against competition by building their own networks, interconnecting to traditional ISPs at points of their choosing after massaging their own data streams internally. If ISPs offered a service that produced similar effects for a fee, one more entry barrier would be leveled.

Can they do that with a well-meaning but poorly informed regulator looking over their shoulders? Perhaps, but it will take a lot of jawboning. Wouldn’t it be more beneficial to tolerate experimentation, at least in competitive markets?

Terminating access monopolies are no longer a cause for concern given that the typical family is now served by both mobile and fixed ISPs. If my wireline ISP misbehaves or simply fails, I can always switch to the Wi-Fi hotspot until the crisis is remedied.

Twenty-odd years of debate over net neutrality/open Internet regulations has created a turgid historical record where it has become nearly impossible to distinguish ancient grievances from current realities.

Rather than going forward with backward-looking Title II regulations it would be wise for the FCC to issue a Further NPRM seeking comment on the state of competition in the Broadband ISP market. The NPRM barely touches this topic, but it’s actually at the center of the current issue set. There is much the Agency can do to accelerate the transition from a wire-dominant broadband regime to a wireless future in which the Internet is fully pervasive.

The last FCC took the first step with the RIF and Small Cell Orders. The current FCC can take the next step on that path.

Best regards,

Richard Bennett
Founder, High Tech Forum
Wi-Fi Pioneer

References

[1] I am an independent network engineering consultant and policy analyst, presently working at High Tech Forum as editor and founder. These remarks are offered in my personal capacity and do not necessarily represent the opinions of any client or sponsor. I have previously offered comment in the “Preserving the Open Internet”, “Broadband Industry Practices”, and “Restoring Internet Freedom dockets,” GN 09-191, WC 07-52, and WC Docket 17-108 respectively, and offered testimony at the FCC En Banc Public Hearing on Broadband Network Management Practices in Cambridge on February 25, 2008 as an invited technical expert. My amicus brief was cited by the DC Circuit in its opinion on the Mozilla challenge to the RIF Order. My CV is available at http://www.bennett.com/resume.pdf.

[2] Richard Bennett, “Designed for Change: End-to-End Arguments, Internet Innovation, and the Net Neutrality Debate” (Washington, DC: Information Technology and Innovation Foundation, September 2009), http://www.itif.org/index.php?id=294.

[3] American Library Ass’n v. FCC, 406 F.3d 689, 693 (D.C. Circuit 2005).

[4] In 2018, the fastest ISP in the US was Xfinity with a Speedtest score of 104.67; the current leader is Cox with a median download speed of 260.09 Mbps.

[5] Adding 4 FWA networks and 2 LEO.

[6] Follows from point 2.

[7] Cisco et al.

[8] Stephen Clark, “Next Year, SpaceX Aims to Average One Launch Every 2.5 Days,” Ars Technica, October 20, 2023, https://arstechnica.com/space/2023/10/next-year-spacex-aims-to-average-one-launch-every-2-5-days/.

[9] “2023 U.S. Residential Internet Service Provider Satisfaction Study,” J.D. Power, October 12, 2023, https://www.jdpower.com/business/press-releases/2023-us-residential-internet-service-provider-satisfaction-study.

[10] Dan Robinson, “Apple Still Struggling with Its Own 5G Modem Silicon,” accessed December 14, 2023, https://www.theregister.com/2023/11/17/apple_qualcomm_modem_independence/.

[11] See NVDA and AVGO.

[12] Free Press, “Comments,” December 14, 2023, p. 43.

[13] Free Press.

[14] Free Press.

[15] FCC, “Accelerating Wireless Broadband Deployment by Removing Barriers to Infrastructure Investment, Declaratory Ruling and Third Report and Order, WT Docket No. 17-79,” 2018.

[16] City of Portland v. FCC, No. 18-72689 (9th Circ. August 12, 2020).

[17] FCC, “In the Matter of Application for Review of Starlink Services, LLC; File No. 0009395128,” December 12, 2023, https://docs.fcc.gov/public/attachments/FCC-23-105A1.pdf.

[18] Public Knowledge, “Comments,” December 14, 2023, p. 23.

[19] Electronic Frontier Foundation, “Comments”, December 14, 2023, p. 6.

[20] Free Press, “Comments,” p. 3.

[21] Public Knowledge, “Comments,” p. 60.

[22] Electronic Frontier Foundation, “Comments,” December 14, 2023. p. 2

[23] Jeffery J. Wheatley, “Competition, Privatization and Change at British Telecom,” Technovation 5, no. 1–3 (October 1986): 115–24, https://doi.org/10.1016/0166-4972(86)90047-7.