FCC Hints at Including More Robust and Relevant Data in QRA Methodology

A March 9, 2012 Wireline Competition Bureau filing might give hope to the many Quantile Regression Analysis (QRA) critics that the FCC is actually paying attention to the extensive complaints that QRA is fatally flawed. The filing explains that the Wireline Competition Bureau “is currently considering the record received on this topic…In addition, consistent with Office of Management and Budget (OMB) guidelines, the Bureau recently sought peer review of the methodology proposed in Appendix H of the FNPRM.” The Bureau makes available the peer review submissions, and more importantly, a list of additional resources and data sets that it is “considering.”

One of the most pervasive complaints about QRA is that the methodology relies upon data that is inappropriate for depicting the costs of deploying broadband in rural areas. For example, QRA does not account for soil texture of frost index, even though soil texture and frozen ground have tremendous impacts on network deployment and maintenance costs. The FCC initially argued that it would rely on widely available data sets, but was criticized because “widely available” does not necessary mean “accurate or relevant.” In the Bureau’s filing, an extensive list of alternative data is now apparently being considered for the QRA methodology, including:

• USDA SSURGO Soil Survey Data and STATSGO2 Soil Map
• USDA Plant Hardiness Zone Map for climate information
• US Department of Commerce, Census Bureau TigerLine Shape Files for road information
• US Department of the Interior, Fish and Wildlife Service National Wetlands Inventory for climate information
• US Department of the Interior, US Geological Survey Digital Elevation Model Information for topology information and National Hydrography Dataset for water level information
• Study area information filed by various RLECs and PSCs

Is the FCC second-guessing the abhorrent methodology or just following protocol by seeking peer review and considering a more robust data set? A February 21 memorandum from Wireline Competition Bureau chief Sharon Gillett requesting peer review explains that the FCC is “currently considering the best means to implement these regressions”—nearly 4 months after the methodology was adopted in the USF/ICC Transformation Order. The memo continues, “Consistent with OMB peer review guidelines, you should evaluate whether the econometric and economic analyses are reasonable and technically correct, and consistent with accepted practices in the field. You should identify any uncertainties and explain the potential implications of the uncertainties for the technical conclusions drawn, and provide suggestions, if any, for ways to minimize key uncertainties and how the methodology might be improved.” The rural telecom industry, of course, has provided this information and then some in comments—but the FCC is right to seek unbiased analysis from expert economists in addition to the companies that QRA will likely devastate.

A peer review by Office of Strategic Planning and Policy Analysis economist Paroma Sanyal seems generally favorable of using QRA to identify rate-of-return carriers “that may be considered less efficient (higher cost) than their peers, and limiting their payments, to incentivize cost-minimization behavior and efficiency improvements.” However, Sanyal identifies 12 issues with QRA. Some notable critiques include:

• “One major concern with the proposed specification is the underlying assumptions behind the model…applying the quantile regression to the individual cost components may miss some high cost carriers, or mislabel other as high cost.” This phenomenon was explained in reply comments by the Rural Broadband Alliance, which illustrated that some companies invested significantly less than the ninetieth percentile caps in some areas, but were capped nonetheless—in other words, mislabeled as being too high cost.
• “The idea behind capping reimbursements is to incentivize carriers to reduce their costs. However, individual cost capping ignores any complementary or substitutability between the various components…A more flexible approach may be to estimate the ninetieth percentile over the total costs.” This speaks to concerns by RLECs that companies might just shift costs to other categories if they fear being capped in one or more categories—which doesn’t really amount to “efficient investment” by the FCC’s standards.
• Regarding the seemingly-arbitrary decisions on which costs are capped, Sanyal comments “There was little discussion in [Appendix H] why some costs were chosen to be capped, while others were not.”
• Sanyal seems to concur with RLECs that “the specifications my suffer from omitted variable bias, as several important factors that may explain loop costs have not been included in the regression.” Sanyal mentions percentage of bedrock, soil type, roads and streams in the construction area, annual rainfall, and frost-free days. Sanyal comments, “The Nebraska Rural Independent Companies’ Capital Expenditure Study makes a fairly compelling case for including these variables in the regression analysis.”
• Sanyal also questions the ninetieth percentile itself—“It would be interesting to compare the results with regression of other percentiles, and observe whether the effects of the explanatory variables are the same across percentiles.”

Another economist, Tracy Waldon from the Media Bureau, believes that “the method rests on a sound theoretical footing, though it is in need of additional analysis of specific implementation issues;” and “quantile regression is the appropriate tool for the estimation problem at hand.” One interesting issue Waldon brings up is that “in its current form, the Appendix does not make a convincing argument that the existing explanatory variables are sufficient to adequately determine similarly situated study areas. A more convincing presentation would being with a detailed discussion of each of the cost categories and the factors which are likely to drive those costs.” Waldon also provides some analysis about variables, statistical significance vs. economic significance, and mathematical aspects of QRA.

The peer review submissions indicate that RLECs are not just making up the pervasive problems in QRA—apparently even high-level government economists have independently reached similar conclusions. Although these economists generally believe that QRA is an appropriate methodology, as usual, the devil is in the details...However, in this case, RLEC salvation from the current version of QRA may be in the details if the details are widely disputed by economists. Based on the economists' peer review submissions, it is fairly clear that the FCC has a lot of work to do on QRA before it is an acceptable methodology for limiting HCLS.  

Do you think the peer review submissions, combined with the overwhelming outcry by RLECs, will be enough to get the FCC to course-correct QRA before significant damage is done to RLEC investment? Will the additional data under consideration at the FCC improve the methodology?

Big Players Stress Hands-off Approach to Fledgling IP-to-IP Market

In the still-developing IP world, the mere possibility of heavy-handed IP-to-IP interconnection regulation is enough for some companies to jump to doomsday conclusions and insist that such regulations would derail innovation and unnecessarily intervene in a market that isn’t far enough along to have experienced any notable failure. In response to the FCC’s USF/ICC Reform FNPRM, powerful telecom industry players like Verizon, Comcast and Google commented that IP-to-IP interconnection should be achieved through individual commercial negotiations, not regulatory directive. Although these commenters bring up some valid objections to regulating IP-to-IP interconnection, the Rural Associations (NECA, NTCA, OPASTCO and WTA) express concern that the large players could abuse their market power when negotiating IP-to-IP interconnection with RLECs. When the time finally comes for the FCC to embark on an IP-to-IP interconnection rulemaking (and that time will come), the agency will definitely face the daunting task of balancing new market momentum with market power protection for small carriers.

In the FNPRM, the FCC asks various questions about IP-to-IP interconnection with the hopes of developing a more complete record—which is highly desired before any drastic regulations are considered. The USF/ICC Transformation Order establishes “the duty to negotiate in good faith has been a longstanding element of interconnection requirements under the Communications Act and does not depend upon the network technology underlying the interconnection, whether TDM, IP, or otherwise.” The FCC seeks comment “on the implementation of good faith negotiations, and…on any additional actions the Commission should ‘take to encourage transitions to IP-to-IP interconnection.’”

Although the FCC has required interconnection in situations where “network owners have incentives to refuse reasonable interconnection to other network operators,” the FCC has also taken a hands-off approach with newer markets like Internet backbone service, where “regulation might impose structural impediments to the natural evolution and growth process which has made the Internet so successful.” Somewhere in between these two extremes, IP-to-IP interconnection questions are lingering unanswered. According to the FCC, “There are conflicting views regarding what role interconnection requirements should play in an increasingly IP-centric voice communications market.” Therefore, the FCC is seeking input from the industry. The rubble was cleared from the road in the Order, but now the FCC wants directions to the final, all-IP destination. Specifically, the FCC wants to know if IP-to-IP interconnection should fall under existing policy frameworks (like Section 251), if a new policy framework should be developed, or if IP-to-IP interconnection should be negotiated purely through unregulated commercial agreements.

Verizon strongly asserts that IP-to-IP interconnection should not be regulated, and voluntary commercial agreements are completely sufficient. According to Verizon, “New regulations—even ones that may at first seem innocuous—by contrast would disrupt the transition [to IP] and harm consumers.” Verizon reasons, “Regulatory history amply demonstrates that, especially in industries marked by rapid technological change, rules based on static assumptions about technology and markets quickly become obsolete—and worse, can lead to unintended negative consequences, including stifling investment and innovation.”

Verizon points to the rapid emergence of cable companies as competitors in the VoIP market as proof that investment and innovation occur without regulatory constraint. In fact, cable VoIP has completely changed the market and turned upside-down traditional incumbent/competitor distinctions: “No longer does one technology or platform dominate over another. Neither the business model or the technologies underlying the Communications Act of 1934 and the Telecommunications Act of 1996 exist as they did when those laws and the resulting regulations came into being. In this innovative new world of IP networks, there are no incumbents. Everyone is a new entrant, and it makes no sense to regulate these new networks and technologies based on a regulatory model that developed around a very different set of circumstances.” Although Verizon’s “everyone is a new entrant” comment might sound prolific and egalitarian, it doesn’t change the fact that there is still a distinction between powerful and powerless when it comes to commercial negotiations between companies of vastly different sizes.

Google likewise favors a more hands-off approach to IP-to-IP interconnection and believes that “The obligation adopted in the Order requiring carriers to negotiate agreements in good faith for IP-to-IP interconnection will help to unlock the full potential of IP networks and will encourage their continued deployment.” Google believes that the FCC should avoid an “overly prescriptive approach,” and let the industry develop effective standards for IP-to-IP interconnection. However, Google does mention that the FCC could develop a “backstop” for cases where voluntary commercial agreements are unviable. Google also believes that “it is in the best interest for all for an industry-led body to take a leading role, at least initially” (translation: it is in the best interest for Google if a Google-led industry body takes a leading role). Industry-led bodies, or multi-stakeholder groups, have become popular for grappling with broad Internet topics; however, there is some concern that the most powerful multi-stakeholder groups are too exclusive and self-serving. John Staurulakis Inc. vice president Valerie Wimer commented to JSICA that she disagrees with Google’s recommendation that industry-led groups like the Technical Advisory Committee should set rules. According to Wimer, “There are not a lot of network owners on the current committee, only Google, programmers, and equipment vendors. The equipment vendors want a fast move to IP-to-IP so they can sell more equipment.”

Comments from Comcast further explain why IP-to-IP interconnection should not be regulated, from a big VoIP player perspective. Comcast argues that regulating IP-to-IP interconnection would be “counterproductive with far-reaching consequences,” particularly the threat that the FCC “could easily slip into broader regulation of the Internet backbone,” causing “serious distortions of that vibrant ecosystem.” Comcast also supports voluntary commercial agreements, insisting that “The Commission should allow market forces to continue to determine the pace and evolution of IP voice interconnection before rushing to impose regulations.” Comcast further argues that there is no evidence of market failure; therefore the FCC has no basis to regulate this area. Comcast does note that some companies are moving faster than others with IP-to-IP interconnection, but this is only because the negotiations and technical decisions are difficult, costly, and at this stage, still highly experimental. Comcast explains, “Experimentation is beginning to produce a variety of answers to these questions, but the questions themselves illustrate why this process is not happening overnight or as quickly as some marketplace participants might wish.” Regulatory intervention, however, is not the way to speed up market experimentation—“The Internet has thrived in the absence of regulation because this marketplace supplies its own checks and balances.”

The Rural Associations do not completely disagree with their large competitors regarding the dangers of premature regulation. The Rural Associations recommend that the FCC develop a more complete record before making a decision, for “detailed rules enacted at too early a stage are more likely to lead to unforeseen twists and consequences that may delay IP evolution or divert it down less than optimal paths.” Until the FCC has had an opportunity to establish a complete record, the Rural Associations recommend a few “simple rules and policies:”

1. Reliance primarily upon the existing and well-tested interconnection procedures of section 251 and 252 of the Act;
2. Monitoring of interconnection negotiations and the rules governing them, particularly to make certain that carriers with substantial bargaining power (e.g. large carriers negotiating with small carriers) are not misusing or abusing it;
3. Clarification that any ‘additional costs’ of IP-to-IP interconnection must be borne by those seeking to obtain such interconnection where it cannot be accommodated today.

The Rural Associations primarily seem concerned about the potential for large carriers abusing their power in commercial negotiations with RLECs. JSI’s Wimer concurred, arguing that “Rural carriers will get squashed by the large players if there is no basic set of rules. Currently, RLECs can barely enforce their rights under 251 let alone have real peer-to-peer negotiations. Verizon and AT&T just dictate what they want.” The Rural Associations further explain the problem: “To date, larger carriers have shown minimal interest in negotiating IP interconnection agreements with RLECs or other small carriers. Essentially, the perceived attitude is ‘you need us more than we need you.’” Given these market realities, the Rural Associations are rightly worried that commercial agreements alone might lead to unfavorable outcomes for small carriers.

RLECs are still in the early stages of transitioning to all-IP networks, and it certainly seems very premature for the FCC to swoop in with overly-stringent regulations. Until the market has had more time to develop, there are other options available to ensure economic and technical parity among players. JSI’s Wimer explains one option, “IP-to-IP is technically feasible but there is still much negotiation to make it happen and work in a manner that traffic does not get blocked. The FCC should not set rules on how this is ultimately done, but should depend on standards committees.”

IP-to-IP interconnection is one of the biggest emerging issues facing the RLEC industry right now, as the FCC continually pressures telecom providers to transition from PSTN to all-IP. What kinds of interconnection challenges and opportunities do you think await RLECs just around the corner along the road to all-IP?

Large ILECs Lobby for a Way Out of Unprofitable Service Areas

Proposed legislation in Kentucky, Senate Bill 135, may hasten the demise of landline service in the state. Although opponents of the bill argue that rural and low-income consumers could be stranded without any reliable and affordable telecommunications options, the “bigger picture” of this bill is that it appears to be aligned with the push by some telecom industry players to phase out the PSTN in the next decade. However, absent any federal rules mandating such a phase-out or implementing a timeline, this bill could be premature and detrimental for rural and low-income consumers in Kentucky.

Senate Bill 135 essentially proposes to end Carrier of Last Resort obligations, particularly for AT&T, Windstream, and Cincinnati Bell. A February 16 Kentucky.com article explains that AT&T has lobbied hard for this bill to be passed, and the legislation has earned the nickname “The AT&T Bill.” Kentucky.com also notes that AT&T has a powerful lobbying machine in Kentucky, with 31 legislative lobbyists and state campaign donations of $91k since 2007. AT&T and proponents of the bill argue that it is “unfair to burden them with state regulations and service requirements that don’t apply to their cable and wireless competitors;” and there is basically no point in providing landline service to new customers if they are eventually going to migrate to wireless or VoIP anyway. AT&T argues that it “must follow where the market leads,” and being relieved of the obligation to provide landline service will open up new opportunities to invest in wireless and broadband.

What AT&T does not say is why they cannot invest in wireless and broadband anyway, without a rule that allows them to escape COLR obligations. Is providing landline service as a COLR actually hindering investment in other telecommunications services in Kentucky? There is no denying the fact that landline use is declining rapidly, but it is unclear if there is a direct correlation between providing landline service as a COLR and not investing in advanced networks—which AT&T seems to imply is the case. The overall impression is that AT&T wants a legislatively-backed excuse to not have to provide service to rural, remote and low-income consumers—especially if any other company is willing to do it.

The bill would allow an ILEC to refuse landline service upon request if there are at least two other competitive alternatives for voice service, which could be wireless or VoIP, or at least one other broadband provider. The bill also further deregulates landline service, building upon rules adopted in 2006. Senate Bill 135 would relieve the state Public Service Commission from the duty to “set, investigate or determine rates as to any electing telephone utility,” but the commission could still conduct investigations initiated by consumer complaints.

A February 28 Louisville Courier-Journal article clarifies that existing telephone subscribers would not lose service—the bill would only apply to customers requesting new landline service. Since AT&T thinks “Kentucky [has] rotarty-dial laws in an iPhone world,” it is possible that AT&T could go cherry-picking through new service applications, weeding out the unprofitable consumers who have a competitive alternative even if the service quality and reliability is not on par with landline service.  This possibility is raising concerns from public safety, rural consumer groups, and the AARP. A February 27 letter from the Rural Assembly’s Rural Broadband Policy Group to members of the Standing Committee on Economic Development, Tourism and Labor states, “Without basic telephone service, rural people will be further isolated from economic and civic participation, and disconnected from our nation’s vital emergency service network.” The group continues, “The real tragedy of this bill is to further disadvantage the most vulnerable people in Kentucky by cutting their ability to communicate with loved ones, elected officials, potential employers, medical providers, and the society at large.” Among their top concerns are carriers ultimately abandoning landline service completely and “’redlining’ of poor and remote communities where providing service is more costly and higher maintenance.”

It is unclear what the Kentucky RLEC industry thinks about this legislation, but there are surely some interesting interplays between a state bill to eliminate COLR for large ILECs and the overarching federal Connect America Fund reforms. There may be a silver lining for RLECs in AT&T’s desire to abandon unprofitable rural areas—by way of CAF Phase II. If the large ILECs are pushing to end COLR obligations, they might not be interested in accepting CAF subsidies for unprofitable rural areas either. CAF support for these unwanted areas could then be up for grabs via reverse auction—possibly opening up new opportunities for RLECs. Essentially, AT&T’s trash could become an RLEC’s treasure.

Senate Bill 135 on the whole doesn’t appear to be particularly good for rural consumers in Kentucky, especially rural consumers with few competitive alternatives. Also, many questions remain about how service decisions would play out. What metrics for “at least two competitive alternatives” do the ILECs plan to use when making the decision to accept or reject an application? If, for example, a rural consumer has access to 2G wireless and cable service—but cannot afford the cable service or reliably call 911 from the wireless serivce—would the ILEC be allowed to reject the application? Would the broadband alternative have to meet the FCC’s 4/1 Mbps standard? Are satellite and fixed wireless services included, even if they do not support a stand-alone voice service? Does this bill violate longstanding principles of Universal Service, or has that game changed with the passing of the USF/ICC Tranformation Order? One would hope that some of these questions will be resolved prior to a legislative decision.

If this bill passes, it would not be out of line to assume that AT&T will press hard for similar legislation in other states. A similar bill in Mississippi that would relieve AT&T of COLR obligations and downgrade the role of the state commission is also receiving strong opposition because of the perceived threats to rural service. Is the end of days coming for COLR obligations? How will this impact the RLEC industry, given that RLECs are strong advocates for COLR obligations?