Daily Rules, Proposed Rules, and Notices of the Federal Government
This is a summary of the Commission's Report and Order in WC Docket No. 05-25, RM-10593, FCC 12-92, adopted on August 15, 2012 and released on August 22, 2012. The summary is based on the public redacted version of the document, the full text of which is available electronically via the Electronic Comment Filing System at
1. In this Report and Order, we suspend, on an interim basis, our rules
2. Special access continues to play a critical role in our economy. Four of the largest incumbent LECs recently reported that their combined 2010 revenues from sales of DS1s and DS3s exceeded $12 billion. Competitive carriers rely heavily on special access to reach customers; a large competitive local exchange carrier (LEC) that offers enterprise services to businesses using special access services as a critical input has reported that it purchases [
3. We continue to strongly believe, consistent with the goals set forth in the
4. The approach we take is based on our evaluation of our 1999 rules, the predictive judgments upon which they were based, and market developments since their adoption. As discussed in greater detail below, the Commission decided in 1999 to use an administratively simple proxy for the presence of actual or potential competition in special access markets—the extent of collocation within broad geographic regions. The Commission predicted that certain levels of collocation within a Metropolitan Statistical Area (MSA) would serve as an accurate indicator of competitive pressure sufficient to constrain prices throughout that area.
5. Based on the evidence in the record and thirteen years of experience with this regime, we now conclude that the Commission's existing collocation triggers are a poor proxy for the presence of competition sufficient to constrain special access prices or deter anticompetitive practices throughout an MSA. We therefore suspend, on an interim basis, the operation of those rules pending adoption of a new framework that will allow us to ensure that special access prices are fair and competitive in all areas of the country.
6. Although we currently lack the necessary data to identify a permanent reliable replacement approach to measure the presence of competition for special access services, we emphasize that the forbearance process set forth by Congress in the 1996 Act provides an avenue for targeted relief based on a complete analysis of competitive conditions in a geographic area.
7. Going forward, in the absence at this time of clear evidence to establish reasonable and reliable proxies to determine where regulatory relief is appropriate, we will collect necessary data and undertake a robust competition analysis that may identify reliable proxies for competition in the market for special access services going forward. We will issue a comprehensive data collection order within 60 days to facilitate this market analysis. We anticipate that during the pendency of the data request, we will continue to analyze the information submitted in the record, and may issue further decisions as warranted by the evidence. Nonetheless, the record in this proceeding demonstrates that a comprehensive evaluation of competition in the market for special access services is necessary, and that further data to assist us in that evaluation is needed with respect to establishing a new framework for pricing flexibility.
8. Through the end of 1990, interstate access charges were governed by “rate-of-return” regulation, under which incumbent LECs calculated their access rates using projected costs and projected demand for access services. An incumbent LEC was limited to recovering its costs plus a prescribed return on investment. It also was potentially obligated to provide refunds if its interstate rate of return exceeded the authorized level. However, a rate of return regulatory structure bases a firm's allowable rates directly on the firm's reported costs and was thus subject to criticisms that it removed the incentive to reduce costs and improve productive efficiency.
9. Consequently, in 1991 the Commission implemented a system of price cap regulation that altered the manner in which the largest incumbent LECs (often referred to today as price cap LECs) established their interstate access charges. The Commission's price cap plan for LECs was intended to avoid the perverse incentives of rate-of-return regulation in part by divorcing the annual rate adjustments from the cost performance of each individual LEC, and provide for sharing efficiency gains with customers in part by adjusting the cap based on industry productivity experience.
10. In contrast to rate-of-return regulation, which focuses on an incumbent LEC's costs and fixes the profits an incumbent LEC may earn based on those costs, price cap regulation focuses primarily on the prices that an incumbent LEC may
11. Pursuant to the pro-competitive, deregulatory mandates of the 1996 Act, the Commission in 1996 began exploring whether and how to remove price cap LECs' access services from price cap and tariff regulation once they are subject to substantial competition. Three years later, in 1999, the Commission adopted the
12. In 2000, after a comprehensive examination of the interstate access charge and universal service regulatory regimes for price cap carriers, the Commission adopted the industry-proposed CALLS plan. This plan represented a five-year interim regime designed to phase down implicit subsidies and (as it pertained to switched and special access charges) to move towards a more market-based approach to rate setting. In adopting the CALLS plan, the Commission offered price cap carriers the choice of completing the forward-looking cost studies required by the
13. The CALLS plan separated special access services into their own basket and applied a separate X-factor to the special access basket. The X-factor under the CALLS plan, unlike under prior price cap regimes, is not a productivity factor. Rather, it represents “a transitional mechanism * * * to lower rates for a specified period of time for special access.” The special access X-factor was 3.0 percent in 2000 and 6.5 percent in 2001, 2002, and 2003. In addition to the X-factor, access charges under CALLS are adjusted for inflation as measured by the GDP-PI. For the final year of the CALLS plan (July 1, 2004-June 30, 2005), the special access X-factor was set equal to inflation, thereby freezing rate levels. Thus, in the absence of a new price cap regime post-CALLS, price cap LECs' special access rates have remained frozen at 2003 levels (excluding any necessary exogenous cost adjustments). The Commission hoped that, by the end of the five-year CALLS plan, competition would exist to such a degree that deregulation of access charges (switched and special) for price cap LECs would be the next logical step.
14. On October 15, 2002, AT&T Corp. filed a petition for rulemaking requesting that the Commission revoke the pricing flexibility rules and revisit the CALLS plan as it pertains to the rates that price cap LECs, and the BOCs in particular, charge for special access services. AT&T claimed that the competitive showings required to obtain pricing flexibility failed to predict price-constraining competitive entry and, rather, that significant competitive entry had not occurred. It further contended that, based on Automated Reporting Management Information System (ARMIS) data, the BOCs' interstate special access revenues had more than tripled, from $3.4 billion to $12.0 billion, between 1996 and 2001 and that the BOCs' returns on special access services were between 21 and 49 percent in 2001. Further, AT&T stated that, in every MSA for which pricing flexibility was granted, BOC special access rates either remained flat or increased. Thus, AT&T contended both that the predictive judgment at the core of the
15. Price cap LECs generally opposed the
16. On January 31, 2005, the Commission released the
18. In July 2007, the Commission invited interested parties to update the record in the special access rulemaking in light of a number of recent developments in the industry, including several “significant mergers and other industry consolidation,” “the continued expansion of intermodal competition in the market for telecommunications services,” and “the release by GAO [the Government Accountability Office] of a report summarizing its review of certain aspects of the market for special access services.” While the special access rulemaking was pending, the Commission also addressed special access regulation for price cap carriers in several other proceedings. A petition for forbearance from dominant carrier regulation of enterprise broadband special access services (i.e., packet-based switched, high-speed telecommunications services for businesses) filed by Verizon was deemed granted in 2006. In orders issued in October 2007 and August 2008, the agency granted petitions filed by AT&T, Embarq, Frontier, and Qwest under 47 U.S.C. § 160 seeking similar forbearance relief, and, in August 2008, granted Qwest's petition for similar relief from regulation of enterprise broadband special access.
19. In November 2009, the Commission sought comment on the appropriate analytical framework for examining the issues that the
20. In October 2010, the Bureau issued a public notice inviting the public to submit data on the presence of competitive special access facilities to assist the Commission in evaluating the issues that the
21. On September 19, 2011, the Bureau issued a second public notice requesting the submission of special access data. In this request, the Bureau sought detailed data on special access prices, revenues, and expenditures, as well as the nature of terms and conditions for special access services. The Bureau requested that the data be submitted to the Commission by December 5, 2011.
22. In the
23. In the
24. The Commission found that different levels of collocation in an area would justify different levels of relief. Specifically, the Commission held that Phase I deregulatory relief would be appropriate in areas where the price cap LEC was able to show that competitors had made irreversible, sunk investment sufficient to “discourage[e] incumbent LECs from successfully pursuing exclusionary strategies,” such as “ `locking up' large customers by offering them volume and term discounts.”
25. The Commission held that Phase II deregulatory relief would be appropriate only in areas where a price cap LEC could show there was a higher level of collocation—specifically, that “competitors have established a significant market presence,
27. In the
28. MSAs can be geographically extensive and, in many cases, may encompass areas with vastly different business density within their borders. Some illustrative examples include the Pensacola, Florida MSA and the Atlanta, Georgia MSA.
30. The Commission established bright line “triggers” based on the extent of collocation within an MSA that it expected would allow a price cap LEC to demonstrate that market conditions in a given MSA would warrant relief. Specifically, the Commission held that price cap LECs would need to demonstrate
either that (1) competitors unaffiliated with the incumbent LEC have established operational collocation arrangements in a certain percentage of the incumbent LEC's wire centers in an MSA, or (2) unaffiliated competitors have established operational collocation arrangements in wire centers accounting for a certain percentage of the incumbent LEC's revenues from the services in question in that MSA. In both cases, the incumbent also must show, with respect to each wire center, that at least one collocator is relying on transport facilities provided by a transport provider other than the incumbent LEC.
31. On February 2, 2001, the U.S. Court of Appeals for the DC Circuit upheld the
32. The Commission's 1999
33. The Commission rejected concerns from some parties that “competition may exist in only a small part of an MSA,” finding that “[t]he triggers we establish * * * are sufficient to ensure that competitors have made sufficient sunk investment within an MSA.” The Commission therefore rejected smaller geographies, such as wire centers, concluding that “the record does not suggest that this level of detail justifies the increased expenses and administrative burdens associated with these proposals.”
34. The Commission received little guidance from commenters on how to establish an appropriate geographic area for grants of pricing flexibility in areas that fall outside of MSAs. In the absence of such guidance, the Commission allowed price cap LECs to make a competitive showing for the entirety of the non-MSA portions of a study area for which they sought relief. It decided against requiring competitive showings at a more granular level—such as on a rural service area (RSA) basis, stating that
* * * we expect competitors to enter MSA markets first and then to extend their networks into less densely populated areas. Because rural areas by definition do not have large concentrations of population comparable to urban areas, we expect that competitive entry into rural areas will be less concentrated than in urban areas. Therefore, we do not expect that pricing flexibility will enable an incumbent to engage successfully in exclusionary pricing behavior with respect to one RSA because competitive entry is limited to another RSA.
35. The record in this proceeding suggests that, contrary to the Commission's prediction in 1999, MSAs have generally failed to reflect the scope of competitive entry. Rather, in many instances, the scope of competitive entry has apparently been far smaller than predicted.
36. In the sections that follow, we evaluate whether record evidence supports the Commission's prediction that MSAs and non-MSA sections of incumbent LEC study areas best reflect the scope of competitive entry. Entry is one of the many elements the Commission and antitrust agencies analyze when evaluating competition. As a general principle, firms are likely to enter a geographic area to compete “if the entrant generates sufficient revenue to cover all costs apart from the sunk costs of entry. Such entry succeeds in the sense that the entrant becomes and remains a viable competitor in the market.” In order to gauge whether entry would be profitable, firms are more likely to focus on areas with high demand for their services, relative to the cost of providing those services. Our review of the evidence suggests that demand varies significantly within any MSA, with highly concentrated demand in areas far smaller than the MSA. This leads us to conclude that competitive entry is considerably less likely to be profitable and hence is unlikely to occur in areas of low demand throughout an MSA, regardless of whether the MSA also contains areas with demand at sufficient levels to warrant competitive entry. This conclusion is confirmed by the available data, including the record of pricing flexibility grants since the Commission's 1999 Order, and data on subsequent competitive developments in these areas.
37. The Commission sought to define the geographic areas for which pricing flexibility requests would be considered “narrowly enough so that the competitive conditions within each area are reasonably similar, yet broadly enough to be administratively workable.” Our analysis of business establishment density indicates that business demand can vary significantly across an MSA. This suggests that competitive conditions within an MSA are also likely to vary significantly, since areas with higher demand tend to be more capable of supporting competition and are more attractive to potential entrants than low demand areas. These data provide context for our analysis of evidence about grants of pricing flexibility petitions and how competitive entry has occurred since adoption of the
38. The plots in Figures 1 and 2 below illustrate that business demand varies significantly within MSAs. They show the distribution of business establishment density by ZIP code in 12 of the sample of 24 MSAs for which we sought data in our voluntary data requests. Figure 1 shows the six MSAs with the least variance in business establishment density across ZIP codes—Fayetteville, North Carolina; Johnstown, Pennsylvania; Phoenix, Arizona; Ocala, Florida; Greenville-Spartanburg, South Carolina; and Lima, Ohio. The distributions show that, even within these relatively homogeneous MSAs, dense pockets of business establishments exist, as well as areas in which business establishments are few and far between. Johnstown, Pennsylvania is an extremely concentrated example. In Johnstown, seventy-five percent of the ZIP codes (from the minimum observation, represented by an upside-down “T” shape, to the top of the box) are clustered near the bottom of the scale with densities close to zero, while the remaining twenty-five percent (from the top of the box to the maximum observation, represented by a “T” shape) are scattered along the vertical axis between about five establishments per square mile and 230 establishments per square mile. The most dense ZIP code (15901), which covers the central business district of Johnstown, is 23 times more dense than the average zip code in the area. Phoenix is much larger and somewhat more uniform than Johnstown, but is nonetheless characterized by a few very dense ZIP codes amid a majority of less dense ZIP codes: while the Phoenix MSA has three ZIP codes with over 300 establishments per square mile, over half of the ZIP codes in the MSA have fewer than 40 establishments per square mile. Overall, these MSAs are similar in that a small number of ZIP codes are far more dense than the rest.
39. The distributions shown in Figure 2 demonstrate more extreme examples of intra-MSA variance of competitive conditions. Figure 2 depicts business establishment density variation for the six MSAs with the most business establishment density variation across ZIP codes: Chicago, Illinois; New Orleans, Louisiana; New York, New York; Seattle-Everett, Washington; Washington, DC; and Los Angeles, California. Except for New York, half of the ZIP codes in each MSA contain fewer than 100 establishments per square mile, whereas other areas within each MSA have upwards of 1,000 establishments per square mile.
40. This variance of competitive conditions within an MSA is an artifact of the way MSAs are defined. The resulting statistical entity can be large, including the entirety of distant counties if those counties contain exurban areas linked to the core by commuting behavior. The Atlanta, Georgia MSA, for example, includes Butts County, Georgia (see Figure 3 below). Of the three ZIP codes within that county, the densest (Jackson, Georgia 30233) has on average about 2.3 business establishments per square mile. This contrasts to the density level of the central business district of Atlanta's MSA, which contains thousands of business establishments per square mile. This kind of variation is common across the 12 MSAs we have examined for these purposes.
41. Given the foregoing evidence that MSAs do not have “reasonably similar” competitive conditions across their geographic areas, and as discussed fully below, when such competitive conditions are considered together with the evidence of how relief has been granted and how some competitive entry has occurred, we can no longer conclude that MSAs “best reflect the scope of competitive entry” by LECs.
42. Though the Commission acknowledged that demand for special access services might be concentrated in certain areas, it designed the competitive showings with the intent of ensuring that price cap LECs could not obtain pricing flexibility throughout an MSA in instances of extremely concentrated demand. While recognizing that “a few wire centers may account for a disproportionate share of revenues for a particular service,” the Commission attempted to set its revenue based collocation triggers at levels designed to “ensure that competitors have extended their networks beyond a few revenue-intensive wire centers.” Our analysis indicates that the 1999 rules have not effectively fulfilled this intent. This provides further evidence that MSAs likely do not reflect the actual scope of competitive entry.
43. As noted above, the Commission adopted two types of rules by which price cap LECs could make the competitive showings required to obtain relief. The first type of rule permitted price cap LECs to obtain relief by showing the presence of collocators in a certain percentage of its wire centers within an MSA. The second type, the revenue-based rule described above, reflected the Commission's concession that demand for special access services is often concentrated. Despite this concession, however, the Commission cautioned that the revenue-based threshold for dedicated transport services would need to be set high enough “to ensure that competitors have extended their networks beyond a few revenue-intensive wire centers.” With respect to channel terminations to end users, which the Commission noted were less competitive than dedicated transport, it doubled the revenue requirement for limited pricing flexibility and increased by almost a third the requirement for full relief. In short, the Commission made the revenue-based rule more difficult to meet specifically to protect against grants of pricing flexibility based on extremely concentrated demand.
44. We have analyzed the 217 incumbent LEC areas for which pricing flexibility relief for channel terminations to end users was granted by order of the Bureau, representing all such grants associated with pricing flexibility petitions available in the Commission's Electronic Tariff Filing System. These grants cover 199 MSAs and five non-MSAs. The majority of those grants were based exclusively on the revenue-based rule. Because the revenue-based rule has different revenue thresholds for each type of special access service, the Commission restricted its analysis to one type,
45. This analysis shows that our rules permitted MSA-wide relief on the basis of extremely concentrated demand in many instances. For example, as detailed in the chart below, 72 of the 212 grants for MSAs were based on revenues of no more than a quarter of the relevant wire centers within the MSA. For example, AT&T obtained Phase II pricing flexibility in the Pensacola MSA based on the revenues of three out of 12 wire centers. Further, 30 of those 72 grants were based on the revenues of only one wire center, 12 were based on the revenues of only two, and 5 were based on the revenues of only three.
46. In sum, more than a third of the cases in which pricing flexibility was granted were premised on the existence of collocations where 65 percent or more of the special access revenue generated within the MSA came from 25 percent or fewer of the wire centers in the MSA. This is consistent with extreme variations in business density. Qualitatively, this suggests that MSA-wide grants of pricing flexibility have encompassed areas in which little or no competitive entry would be expected.
47. Even with more relaxed standards for what constitutes extremely concentrated demand, the data shows that 97 grants were based on revenues from less than a third of the wire centers, and 144 were based on revenues from less than half of the wire centers. Conversely, only 28 grants were based on revenues of two-thirds or more of the wire centers within the applied-for MSA.
48. Whereas our bright-line competitive showings suggested that some MSAs would soon be, or already were, competitive more than a decade ago, recent data indicates that competitors have a strong tendency to enter in concentrated areas of high business demand, and have not expanded beyond those areas despite the passage of more than a decade since the grant of Phase II relief. This provides further evidence that an MSA is probably a much larger area than a competitor would typically choose to enter.
49. For example, data about the Atlanta MSA, where BellSouth was granted Phase II relief in 2000, demonstrates the importance of geographic business establishment density as a driver of competitive entry. In 2011, staff collected data, on a voluntary basis, about the presence of competitive special access facilities for channel terminations to end users in 24 MSAs. The following providers submitted data indicating that they provide facilities-based competition in parts of the Atlanta MSA: [
50. The ZIP codes in which the reporting carriers in Atlanta offered facilities-based competition were those with the highest average business establishment densities. This is reflected in Table 5, which compares average business establishment density between ZIP code areas in which reporting carriers compete and ZIP codes areas in which they do not (and includes similar data for the Miami and Norfolk MSAs). Because the data submissions that serve as the basis for Table 5 were voluntary, the reporting competitors do not necessarily represent all competition in the three MSAs discussed above, and it is possible that competitors have higher market shares than our data show. However, Table 5 does not show market shares, but rather the geographic breadth of coverage by competitors within the MSA. Further analysis of these data indicates that the reporting carriers had a tendency to enter the same areas within the MSA. We have no reason to believe that the competitors' focus on high business establishment density indicated by these data would change if we were able to obtain data from any other competitive providers with access facilities in the Atlanta, Miami and Norfolk MSAs. Thus, despite the fact that our competitive showings rules were designed to predict competitive entry across an MSA, these data suggest a strong tendency for competitive LECs to deploy channel termination facilities to end users only in ZIP codes with the highest density of business establishments.
51. Chart 6 displays the distribution of establishment density for ZIP codes in the three MSAs of Table 5. The distribution at the top of Chart 6 is for ZIP codes in which no reporting carrier offered facilities-based competition for end-user channel terminations and the distribution at the bottom is for ZIP codes in which one or more reporting carriers did offer facilities-based competition for end-user channel terminations. The chart indicates that the reporting carriers had a greater tendency to offer competition in ZIP codes with business establishment density greater than 100 establishments per square mile than they did in ZIP codes with lower establishment densities. Based on an analysis of the individual ZIP code areas, the probability that the carriers' location decisions in these metropolitan areas were not tied to business establishment density is exceedingly small. The findings from this analysis are consistent with other evidence in the record.
52. The fact that there may be other competitors in these MSAs that are not reflected in our data, that more competitors may enter in the future, or that current competitors may build out to other parts of the MSA with high business density does not diminish our finding that competitors typically enter in areas of high business establishment density. Commenters rightly point out that we do not have comprehensive facilities data for the MSAs above. We recognize the limitations of our existing data set and, as described below, we intend to collect additional data in the coming months that will help inform our analysis. However, even this partial data provides insight into where competitors choose to enter within an MSA, and reinforces evidence we have received in this record.
53. Incumbent LECs generally concede that competitors have focused on areas in which demand for special access services is very concentrated. As SBC noted:
Demand for special access services is highly concentrated in a relatively small number of dense urban wire centers and ex-urban wire centers containing office parks and other campus environments. Indeed, more than
54. Some commenters also allege that extending new facilities is sufficiently easy that competitors could reach all parts of an MSA if warranted even if they only have facilities in part of an MSA today. SBC, for example, states that a large percentage of its demand for DS1 and DS3 services runs within 1,000 feet, or about three city blocks, of existing alternative fiber. Thus, incumbent LECs argue that potential competition exists throughout an MSA even if competitive facilities are only present in a small area. In contrast, competitive carriers assert that entry is far more difficult than incumbents describe in the record. Such commenters state that, as compared to incumbent providers who have achieved economies of scope and scale in the provision of telecommunications services, it is not economical for competitors to deploy their own facilities to serve all special access demand. Competitive carriers note that construction costs, the costs of fiber and electronics, backhaul costs, transaction costs involved in negotiating with
55. We need not resolve this controversy here, however, for data provided by incumbent LECs demonstrate that, even if competitors could easily deploy fiber to serve customer demand within 1,000 feet of incumbents' facilities, many parts of an MSA would still not be served by competitive fiber. For instance, a 2007 AT&T map depicting competitive fiber deployment in the Austin, Texas MSA appears to indicate that, out of the 24 AT&T wire centers in the MSA, competitive fiber does not extend to
56. As discussed above, the Commission selected the MSA because it decided the MSA best reflected the scope of