Telecommunications: FCC Needs to Improve Its Ability to Monitor
and Determine the Extent of Competition in Dedicated Access
Services (29-NOV-06, GAO-07-80).
Government agencies and businesses that require significant
capacity to meet voice and data needs depend on dedicated access
services. This segment of the telecommunications market generated
about $16 billion in revenues for the major incumbent
telecommunications firms in 2005. The Federal Communications
Commission (FCC) has historically regulated dedicated access
prices. With the Telecommunications Act of 1996, FCC reformed its
rules to rely on competition to bring about cost-based pricing.
Starting in 2001, FCC granted pricing flexibility on the basis of
a proxy measure of competition. GAO examined (1) the extent that
alternatives are available in areas where FCC granted pricing
flexibility, (2) how prices have changed since the granting of
pricing flexibility, and the effect on government agencies, and
(3) how FCC monitors competition. GAO's work included analyzing
data on competitive alternatives, list prices, and average
revenue, and interviewing FCC officials and industry
representatives.
-------------------------Indexing Terms-------------------------
REPORTNUM: GAO-07-80
ACCNO: A63770
TITLE: Telecommunications: FCC Needs to Improve Its Ability to
Monitor and Determine the Extent of Competition in Dedicated
Access Services
DATE: 11/29/2006
SUBJECT: Competition
Contracts
Cost analysis
Data collection
Price regulation
Prices and pricing
Regulatory agencies
Strategic planning
Telecommunications
Telecommunications industry
Price controls
******************************************************************
** This file contains an ASCII representation of the text of a **
** GAO Product. **
** **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced. Tables are included, but **
** may not resemble those in the printed version. **
** **
** Please see the PDF (Portable Document Format) file, when **
** available, for a complete electronic file of the printed **
** document's contents. **
** **
******************************************************************
GAO-07-80
* [1]Results in Brief
* [2]Background
* [3]Facilities-Based Competition to End Users Does Not Appear to
* [4]Competitive Alternatives Exist in a Relatively Small Subset
* [5]FCC's Metric Shows a Decline in the Extent of Colocation in
* [6]Limited Competition Could Be Caused by a Variety of Factors
* [7]Prices for Dedicated Access Services in MSAs with Phase II P
* [8]Price-Flex List Prices Are Higher on Average Than Price-Cap
* [9]Effects of Contracts on Prices Varied, but Are Generally the
* [10]Conditions and Terms May Inhibit Switching Circuits to Compe
* [11]Average Revenue Has Declined Over Time, but Is Generally Hig
* [12]Data from Two Departments Show That Their Spending on Dedica
* [13]FCC Plays an Important Role in Ensuring Competition, but Lac
* [14]Ensuring Competition Is a Central FCC Responsibility
* [15]FCC Uses Various Data to Assess Competition for Dedicated Ac
* [16]Data Available to FCC Are Not Current, Specific or Reliable
* [17]Conclusions
* [18]Recommendations for Executive Action
* [19]Agency Comments
* [20]Extent of Competitive Alternatives to Major Incumbent Firms
* [21]Analysis of Available Information on Dedicated Access Pricin
* [22]Change in List Prices
* [23]Customized Contracts
* [24]Change in Average Revenue
* [25]Federal Spending on Dedicated Access Services
* [26]FCC Oversight of Dedicated Access
* [27]GAO Contact
* [28]Staff Acknowledgments
* [29]GAO's Mission
* [30]Obtaining Copies of GAO Reports and Testimony
* [31]Order by Mail or Phone
* [32]To Report Fraud, Waste, and Abuse in Federal Programs
* [33]Congressional Relations
* [34]Public Affairs
Report to the Chairman, Committee on Government Reform, House of
Representatives
United States Government Accountability Office
GAO
November 2006
TELECOMMUNICATIONS
FCC Needs to Improve Its Ability to Monitor and Determine the Extent of
Competition in Dedicated Access Services
GAO-07-80
Contents
Letter 1
Results in Brief 12
Background 16
Facilities-Based Competition to End Users Does Not Appear to Be Extensive
19
Prices for Dedicated Access Services in MSAs with Phase II Pricing
Flexibility Are on Average Higher Than Prices Elsewhere 27
FCC Plays an Important Role in Ensuring Competition, but Lacks Sufficient
Information to Determine the Success of Its Deregulatory Policies 36
Conclusions 41
Recommendations for Executive Action 43
Agency Comments 44
Appendix I Objectives, Scope and Methodology 50
Appendix II Analysis of Average Revenue Data and List Prices 62
Appendix III Comments from the Federal Communications Commission 71
Appendix IV GAO Contact and Staff Acknowledgments 76
Tables
Table 1: Conditions for Granting Different Phases of Pricing Flexibility
in an MSA 16
Table 2: Percentage of Buildings with a Fiber-Based Competitive
Alternative by Demand (July 2006) 20
Table 3: Change in Competitive Colocation in "Price-flex" Wire Centers 24
Table 4: Examples of Contract Terms and Conditions 31
Table 5: MSAs in Analysis, by Price-Cap Incumbent and Applicable Pricing
Flexibility 50
Table 6: GeoResults' Dedicated Access Demand Model 54
Table 7: Summary Statistics of Average Revenue for Channel Terminations,
Unweighted Nominal Dollars 63
Table 8: Summary Statistics of Average Revenue for Channel Terminations,
Unweighted Adjusted Dollars Using Telecommunications Price Index 64
Table 9: Summary Statistics of Average Revenue for Channel Terminations,
Unweighted Adjusted Dollars Using General GDP Price Index 65
Table 10: Summary Statistics of Average Revenue for Channel Terminations,
Weighted Adjusted Dollars Using the Telecommunications Price Index 66
Table 11: Summary Statistics of List Price Comparisons for all DS-1
Combinations in Nominal Dollars 67
Table 12: Summary Statistics of List Price Comparisons for all DS-3
Combinations in Nominal Dollars 69
Figures
Figure 1: Simplified Components of Dedicated Access Circuits 5
Figure 2: Average Differences in Dedicated Access Prices Across All Terms
and Zones in Nominal Dollars 29
Figure 3: Change in Average Revenue for DS-1 and DS-3 Channel Terminations
in Phase I and Phase II Areas 33
Abbreviations
ARMIS Automated Reporting Management Information System
CALLS Coalition for Affordable Local and Long Distance Service
DOJ Department of Justice
FCC Federal Communications Commission GDP Gross Domestic Product
GSA General Services Administration
Mbps megabytes per second
MSA Metropolitan Statistical Area RBOC Regional Bell Operating Company
USDA Department of Agriculture
This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed in
its entirety without further permission from GAO. However, because this
work may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this material
separately.
United States Government Accountability Office
Washington, DC 20548
November 29, 2006
The Honorable Tom Davis Chairman Committee on Government Reform House of
Representatives
Dear Mr. Chairman:
Government agencies and businesses rely on "special access" services (also
known as "dedicated access") to meet their voice and data
telecommunications needs (i.e., large volumes of long-distance services,
secure point-to-point data transmissions, and reliable Internet access).^1
The federal government, with its extensive network of agency offices
spread throughout the nation, is a major consumer of these services. Due
to increasing data transmission needs, these dedicated access services are
a growing segment of the telecommunications market and represented about
$16 billion in revenues in 2005 for the major providers of those
services--the largest incumbent telecommunications firms (i.e., AT&T
Corporation [formerly SBC Communications], BellSouth Corporation, Qwest
Communications, and Verizon Communications). The incumbent firms have an
essentially ubiquitous local network that generally reaches all of the
business locations in their local areas. For long-distance or other
telecommunications companies (such as Sprint Nextel, Time Warner Telecom,
and Level 3 Communications) to provide their services to large business
customers, they often purchase dedicated access services on a wholesale
basis from the incumbents for local connectivity. The incumbent firms have
stated that the majority of the dedicated access services they sell are
sold wholesale to other carriers. Alternatively, competitors may build out
to reach customers using their own facilities, or purchase connections
from other competitive carriers that have built out to those businesses,
resulting in "facilities-based" competition. The Telecommunications Act of
1996 (the 1996 Act), allowed the major incumbent firms to compete in the
long-distance market;^2 therefore, incumbent firms are now competing to
provide businesses with long-distance services as well as acting as a
wholesale supplier of local connectivity to their competitors.
^1Because these services operate separately from the local "switched"
telecommunications network used to route telephone calls, they are
considered "dedicated." Customers do not consider switched access services
to be a viable substitute for dedicated access because they do not offer
the guaranteed bandwidth, high service levels, and security that dedicated
access provides.
The Federal Communications Commission (FCC), which is an independent
United States government agency, regulates interstate and international
communications by radio, television, wire, satellite, and cable. Because
the major incumbent firms initially controlled all dedicated access
connections, prices for these services have traditionally been regulated
by FCC. In 1991, FCC implemented a system of regulations that altered the
manner in which the incumbent firms established interstate dedicated
access prices. FCC "capped" the prices that could be charged by the large
incumbent firms. (Those firms are hereafter called "price-cap
incumbents.")^3 The 1996 Act, which Congress designed to foster a
procompetitive, deregulatory national policy framework for the United
States telecommunications industry, led FCC to reconsider its current
regulatory framework for access prices, including whether and how to
remove price-cap incumbents' access services from price caps and tariff
regulation once they are subject to substantial competition.
In 1999, FCC issued the Pricing Flexibility Order,^4 which, among other
things, permitted the deregulation of prices for dedicated access services
in metropolitan statistical areas (MSA)^5 where price-cap incumbents could
show that certain "competitive triggers" had been met. The competitive
trigger refers to the extent to which competitors have "colocated"
equipment in a price-cap incumbent's wire center (i.e., an aggregation
point on a local telecommunications' network).^6 FCC determined that once
a certain level of colocation in wire centers throughout a metropolitan
area had been achieved, it was a good predictor that competitors had made
significant, irreversible sunk investments in facilities, and indicated
the likelihood that a competitor could eventually extend its own network
to reach its customers. In FCC's view, sufficient sunk investments of this
sort would constrain monopoly behavior by price-cap incumbents.
Accordingly, FCC determined that colocation at the wire center level can
reasonably serve as a measure of competition in a given MSA, rather then
looking to more granular assessments of the level of competition at a
building level or at the level of individual customers. FCC also
determined that the colocation-based triggers would not be overly
burdensome on parties and on FCC's limited resources as would be more
granular assessments. The United States Court of Appeals for the District
of Columbia affirmed FCC's decision to grant additional pricing
flexibility to price-cap incumbents through a series of colocation-based
triggers.^7
2Upon a showing that local markets are open to competition, the Regional
Bell Operating Companies (RBOC) were granted authority to enter the market
for long distance services, pursuant to section 271 of the Communications
Act of 1934, as amended by the Telecommunications Act of 1996, Pub. L. No.
104-104, 110 Stat. 56 (1996). 47 U.S.C. S 271. Originally, seven RBOCs
formed after the break up of AT&T--Ameritech, Bell Atlantic, BellSouth,
NYNEX, Pacific Telesis, Southwestern Bell, and US West. Through various
mergers, these companies have combined into four--AT&T, BellSouth, Qwest,
and Verizon.
^3Sections 203 and 204 of the Communications Act of 1934, as amended,
establish tariff filing requirements applicable to common carriers. 47
U.S.C. SS 203, 204. FCC implemented the rules establishing regulations,
including the filing, form, content and notice, and the pricing rules and
related requirements that apply to incumbent carriers subject to price-cap
regulation. A tariff is the document filed by a carrier describing their
services and the payments to be charged for such services.
^4Access Charge Reform, CC Docket No. 96-262, Fifth Report and Order and
Further Notice of Proposed Rulemaking, 14 FCC Rcd 14221 (1999) (Pricing
Flexibility Order), aff'd, WorldCom, Inc. v. FCC, 238 F.3d 449 (D.C. Cir.
2001). Price-cap incumbents must file a petition seeking pricing
flexibility. 47 C.F.R. S 1.774.
Depending on the extent of competitive colocation that is achieved in an
MSA, FCC grants either partial or full pricing flexibility to the
price-cap incumbent carriers.^8
5The Office of Management and Budget defines an MSA as an area having at
least one urbanized area of 50,000 or more population, plus adjacent
territory that has a high degree of social and economic integration with
the core as measured by commuting ties.
^6FCC's Expanded Interconnection Orders required price-cap incumbent firms
to allow competing firms to install ("colocate") certain network equipment
in particular wire centers at reasonable terms and conditions. 47 CFR S
64.1401(a).
^7Specifically, the court found that FCC made a reasonable policy
determination that colocation was a sufficient proxy for market power, and
that the court had no basis upon which to require FCC to engage in a more
searching analysis of competition before granting pricing flexibility.
WorldCom, Inc. v. FCC, 238 F.3d 449 (D.C. Cir. 2001). See also Covad
Communications Co. v. FCC, 450 F.3d 528 (D.C. Cir. 2006). The court noted
that under the 1996 Act, colocation can reasonably serve as a measure of
competition in a given market, particularly where it is superior to the
various alternatives proposed by objecting petitioners.
^8Competitive, nonincumbent firms are not subject to rate regulation and,
therefore, do not apply for pricing flexibility.
o In MSAs where price-cap incumbents can demonstrate a certain
level of competitive colocation, they would satisfy the triggers
that would result in partial price deregulation (known as "phase
I" flexibility). With phase I flexibility, FCC allows price-cap
incumbents to offer customized contracts to customers that provide
discounts off the price-capped "list prices." This flexibility was
designed to allow price-cap incumbents to more adequately respond
to competition, where price-cap regulation may be too constricting
to allow the incumbent to lower its prices to respond to
competitive pressures. Price-cap incumbents must file their
contract terms and conditions---on a day's notice--with FCC and
make that same contract available to other customers that meet the
contract's specified terms and conditions. Alternatively, for
customers that do not sign up for contracts, incumbents are
required to offer dedicated access at price-capped prices. Those
prices may include term and volume discounts (e.g., lower list
prices may exist for 3-year or 5-year terms compared with
month-to-month list prices or for purchasing greater amounts of
dedicated access).
o In MSAs where price-cap incumbents can demonstrate a higher
level of competitive colocation, price-cap incumbents may meet
more stringent competitive triggers and qualify for greater price
deregulation (known as "phase II" flexibility). Because FCC deems
phase II areas to be sufficiently competitive to ensure that rates
for dedicated access are just and reasonable, phase II flexibility
frees the incumbent from price caps and allows it to raise or
lower its list prices. Price-cap incumbents must still file these
new "price-flex" list prices with FCC. As with phase I
flexibility, contracts can be offered that provide additional
discounts to respond to competitive pressures.
o Where neither trigger for competition is met, price-cap
incumbents' prices remain subject to FCC's price cap and customers
can only purchase dedicated access from the price-capped list
prices (which can include volume and term discounts).
FCC's pricing flexibility pertains to two separate components of
dedicated access services--the end user channel termination and
dedicated transport. In general, the end user channel termination
component (sometimes referred to as a "local loop") connects an
end user's location (e.g., the corporate headquarters or field
office) with the nearest incumbent's serving wire center. The
dedicated transport component connects one wire center to another
wire center or to another carrier's point of presence. Figure 1
illustrates these components in MSAs with different levels of
pricing flexibility for channel terminations and the pricing that
applies for each component. In the MSA on the left-hand side of
the figure, the price-cap incumbent has received phase I
flexibility for channel terminations. As figure 1 shows, with
phase I flexibility, the price-cap price is still available for
channel terminations. In the MSA on the right-hand side of the
figure, the incumbent has received phase II flexibility for
channel terminations, and the price-flex price is used. If a
competitor is colocated in the wire center as the figure
illustrates, FCC has noted that the potential exists for the
competitor to build out its own network to end user B.
Figure 1: Simplified Components of Dedicated Access Circuits
In 2000, prior to its granting any pricing flexibility, FCC further
reformed its price-cap rules. That reform was initiated by a group of
incumbent firms and long-distance companies, called the Coalition for
Affordable Local and Long Distance Service (CALLS).^9 The CALLS plan was
envisioned as a 5-year transitional regime to resolve, among other things,
price-cap issues.^10 Specifically, as FCC adopted, the CALLS plan provided
for yearly reductions in price caps for dedicated access services based on
agreed-upon percentages. The percentage decreases were 3 percent in 2000
and 6.5 percent each year from 2001 through 2003. Beginning in 2004,
price-cap rates have essentially been frozen, with no further decreases in
prices, with the exception of adjustments based on cost factors outside of
the incumbents' control (e.g., taxes and fees). The "CALLS Order" was
intended to run until June 30, 2005, but the order remains in place until
FCC adopts a subsequent plan.
^9CALLS consisted of four of the five largest incumbent firms and two of
the three largest long-distance carriers at the time (mergers since then
have reduced the number of incumbents and long-distance carriers). CALLS
consisted of the AT&T Corporation, Bell Atlantic Telephone Companies,
BellSouth Corporation, GTE Service Corporation, SBC Communications Inc.,
and Sprint Corporation.
In 2001, concurrently with the scheduled decreases in price caps resulting
from the CALLS Order, FCC began granting pricing flexibility to price-cap
incumbents. Some level of pricing flexibility has since been granted to
the four major price-cap incumbents in 215 of the 369 MSAs in the United
States and Puerto Rico. These four price-cap incumbents have received full
price deregulation (phase II for all circuit components) in 112 MSAs. Only
3 of the 100 largest MSAs in the United States and Puerto Rico are not
under any pricing flexibility.^11
In January 2005, in response to a petition that AT&T filed in 2002, FCC
initiated a rulemaking proceeding on dedicated access price regulation to
examine whether the pricing flexibility rules should remain intact or be
revised.^12 The basic economic theory underlying FCC's regulatory approach
postulates that greater competition should constrain incumbent pricing
power and drive prices toward the marginal cost of providing those
dedicated access services. However, competitors and business customers
have raised concerns that, in places where FCC has granted phase II
pricing flexibility, prices have incongruously risen. Concerns also have
been raised that the competitive triggers that FCC used were inadequate to
accurately judge the extent of competition in the market. Price-cap
incumbents, on the other hand, generally oppose the petition. They contend
that their dedicated access rates are reasonable, that there is robust
competition in the dedicated access market, and that the colocation-based
triggers are an accurate metric for competition. FCC's rulemaking is still
ongoing.
^10Access Charge Reform, CC Docket Nos. 96-262, 94-1, 99-249, 96-45, Sixth
Report and Order in CC Docket Nos. 96-262 and 94-1, Report and Order in CC
Docket No. 99-249, Eleventh Report and Order in CC Docket No. 96-45, 15
FCC Rcd 12962 (2000) (CALLS Order). See also Texas Office of Public Util.
Counsel v. FCC, 265 F.3d 313 (5th Cir. 2001).
^11Price-cap incumbents have also received some level of pricing
flexibility in the non-MSA areas of 14 states, and phase II flexibility
for all circuit components in the non-MSA area of 1 state. The 3 MSAs of
the top 100 in the United States and Puerto Rico without pricing
flexibility are San Juan-Bayamon, Puerto Rico; Youngstown-Warren, Ohio;
and Sarasota-Bradenton, Florida.
^12Special Access Rates for Price Cap Local Exchange Carriers, WC Docket
No. 05-25, Order and Notice of Proposed Rulemaking, 20 FCC Rcd 1994
(2005).
Recent mergers of major telecommunications firms--SBC's acquisition of
AT&T (and subsequently assuming the AT&T name); Verizon's merger with MCI;
and, more recently, AT&T's proposed purchase of BellSouth--have further
complicated the issues surrounding dedicated access services. As
long-distance companies, the former AT&T and MCI were two of the largest
purchasers of dedicated access services from the incumbents and were major
competitors for providing large business customers with telecommunications
services. At the federal level, FCC and the Department of Justice (DOJ)
reviewed these mergers.^13 DOJ filed separate civil antitrust complaints
on October 27, 2005, seeking to enjoin the proposed acquisitions. DOJ
found the likely effect of these acquisitions would be to lessen
competition substantially for dedicated access in 19 metropolitan
areas.^14 FCC approved the proposed mergers on October 31, 2005, subject
to the parties' agreeing to certain commitments, including freezing the
prices for dedicated access for 30 months.^15 More recently, AT&T
announced plans to purchase BellSouth. Concerns have been raised that this
proposed merger also may lessen competition in the dedicated access
market. FCC's review of this merger is ongoing.
^13Because mergers involve a change in the ownership or control of
companies holding licenses or lines needed to offer telecommunications
services in the United States, merging firms must apply to FCC for
approval of the transfer of those licenses or lines. 47 U.S.C. SS 214(a);
310(d). The purpose of FCC's review is to determine that the license
transfers are in the "public interest." In making this determination, FCC
considers several factors such as the effects of a merger on competition
in the industry, the FCC's ability to enforce its obligations under the
1996 Act and the deployment of advanced telecommunications services. When
FCC finds a merger to be in the public interest, it will approve the
transfers of licenses and lines necessary to allow the merger to go
forward. If FCC finds the public interest harm outweighs the public
interest benefit of a transaction, it may enter into discussions with the
merging parties, and ultimately, adopt conditions--that is, specific
activities that the merged company would have to perform--that will change
the balance of the public interest effects and thus enable FCC to find the
license transfers to be in the public interest. 47 U.S.C. SS 214(c);
303(r). FCC may, after taking the necessary steps, determine that the
merger is not in the public interest and decline to approve the merger.
Whatever FCC actions are taken in a particular case, interested parties
(including, but not limited to, the merging companies) can file a lawsuit
challenging FCC's decision. Any party filing such a lawsuit against an FCC
decision bears the burden of proof in showing that the decision was
"arbitrary and capricious" or beyond FCC's authority.
DOJ reviews mergers under federal antitrust laws to assess whether a
merger may substantially lessen competition. If DOJ determines that a
merger will substantially harm competition and, therefore, violates
antitrust laws, it can bring a court action to stop the merger. DOJ also
can negotiate an agreement with the merging companies, where the merging
companies agree to undertake activities that would eliminate the
competitive harm of the merger, such as divesting certain properties. That
agreement, called a proposed Final Judgment, is filed with the court and
is legally enforceable upon compliance with the Antitrust Procedures and
Penalties Act, 15 U.S.C. S 16(b)-(h) ("Tunney Act"). Under a Tunney Act
review, the court may enter the judgment if it concludes that it is in the
public interest.
The availability of unbundled network elements (UNE) also complicates the
issues surrounding facilities-based competition in dedicated access
because they are functional equivalents to certain dedicated access
services, but, where available, are generally less expensive than
dedicated access services.^16 The 1996 Act gave the FCC broad power to
require incumbent firms to make UNEs available to competitive carriers to
provide them with local connectivity.^17 Recently, a federal appellate
court upheld FCC's fourth attempt to impose UNE rules.^18 Under the new
rules, FCC modified its unbundling framework^19 for high-capacity loops
and transport. The Commission adopted a wire-center-based analysis that
used the number of access lines and fiber colocations in a wire center as
proxies to determine impairment for high-capacity loops and dedicated
transport.^20 Where such triggers are not met, the incumbent must make
UNEs available at rates based on forward-looking economic costs.^21 FCC
hopes that this framework will lead to the right incentives for both
incumbents and competitors to invest rationally in the telecommunications
market.
^14At the same time that the complaints were filed in the mergers between
SBC and AT&T and MCI and Verizon, DOJ also filed stipulations and proposed
Final Judgments that are designed to eliminate the anticompetitive effects
of the acquisitions in the affected buildings. Under the proposed Final
Judgments, defendants are required to divest, in most situations,
indefeasible rights of use for lateral connections to certain buildings
located in a number of metropolitan areas. DOJ and defendants have
stipulated that the proposed Final Judgments may be entered into after
compliance with the Antitrust Procedures and Penalties Act, 15 U.S.C. S
16(b)-(h) ("Tunney Act"). These actions were consolidated. The Tunney Act
review is still ongoing.
^15Memorandum Opinion and Order, In the Matter of SBC Communications Inc.
and AT&T Corporation Applications for Approval of Transfer of Control, FCC
WC Docket No. 05-65 (rel. Nov. 17, 2005); and Memorandum Opinion and
Order, In the Matter of Verizon Communications Inc. and MCI, Inc.
Applications for Approval of Transfer of Control, FCC WC Docket No. 05-75
(rel. Nov. 17, 2005).
^16A network element is defined as "a facility or equipment used in the
provision of a telecommunication service." 47 U.S.C. S 153(29).
^17Congress left to the FCC the choice of elements to be "unbundled"
specifying that it must "consider, at a minimum, whether . . . the failure
to provide access to such network elements would impair the ability of the
telecommunications carrier seeking access to provide the services that it
would seek to offer." 47 U.S.C. S 251(d)(2) (emphasis added).
^18Covad Communications Company v. FCC, 450 F.3d 528 (D.C. Cir. 2006).
^19Unbundled Access to Network Elements, WC Docket No. 04-313, CC Docket
No. 01-338, Order on Remand, FCC 04-290 (rel. February 4, 2005) (Triennial
Review Remand Order).
In light of these issues, this report discusses (1) the extent to which
facilities-based competition to customer locations exists in areas where
FCC granted pricing flexibility; (2) how prices for dedicated access
services for businesses as well as federal government agencies have
changed since phase II pricing flexibility; and (3) what data FCC uses to
monitor competition in dedicated access services, along with the
limitations, if any, that exist in its monitoring effort. We are not
making a judgment on the legal sufficiency of competition in dedicated
access services, including whether recent mergers violate antitrust laws
or whether proposed remedies that DOJ identified would be sufficient to
eliminate the competitive harm of the mergers.
To determine the extent of facilities-based competition in areas where FCC
has granted pricing flexibility, we analyzed data on dedicated access
services in selected metropolitan areas from Telcordia(R) Technologies,
Inc., a leading global provider of telecommunications network software and
services, and GeoResults, which is a firm that the telecommunications
industry has used extensively to analyze Telcordia data.^22 We analyzed
data showing not only the extent that competitors continue to be colocated
in incumbent wire centers (FCC's measure), but also data showing the
extent to which competitors have equipment in commercial office buildings
that provides actual (or "lit") service to end users of dedicated access
services. We selected 16 metropolitan areas in which FCC has granted the
price-cap incumbents with varying phases of pricing flexibility.^23 We
selected 4 metropolitan areas in the geographic areas broadly served by
each of the four major price-cap incumbents (AT&T, BellSouth, Qwest, and
Verizon). We also interviewed officials with price-cap incumbents and
competitive firms, industry analysts, and representatives of major
telecommunications customers. To describe how prices have changed since
phase II pricing flexibility was granted, we used the following three
methods to analyze changes in prices in areas where phase II flexibility
was granted to areas where phase I flexibility was granted and areas still
under the price cap.
^20Competing carriers are impaired without access to DS-1 transport except
on routes connecting a pair of wire centers, where both wire centers
contain at least four fiber-based colocators or at least 38,000 business
access lines. Competing carriers are impaired without access to DS-3 or
dark fiber transport except on routes connecting a pair of wire centers,
each of which contains at least three fiber-based colocators or at least
24,000 business lines. Finally, competing carriers are not impaired
without access to entrance facilities connecting an incumbent's network
with a competitor's network in any instance. FCC adopted a 12-month plan
for competing carriers to transition away from use of DS-1 and DS-3
capacity dedicated transport where they are not impaired, and an 18-month
plan to govern transitions away from dark fiber transport. These
transition plans apply only to the embedded customer base, and do not
permit competitors to add new dedicated transport UNEs in the absence of
impairment. During the transition periods, competitive carriers will
retain access to unbundled dedicated transport at a rate equal to the
higher of (1) 115 percent of the rate the requesting carrier paid for the
transport element on June 15, 2004, or (2) 115 percent of the rate the
state commission has established or establishes, if any, between June 16,
2004 and the effective date of FCC's UNE Order.
^21The Commission has concluded that UNE prices must be based on each
element's Total Element Long-Run Incremental Cost (TELRIC). See 47 C.F.R.
S 51.505(b); Verizon Communs., Inc. v. FCC, 535 U.S. 467, 523 (2002).
TELRIC rates are akin to wholesale prices because competitors are supposed
to economically be able to rent UNEs and then use them to sell
telecommunication services to their retail customers. Covad Communications
Company v. FCC, supra. These rates are determined by states' public
utility commissions.
o We analyzed listed prices for channel terminations and dedicated
transport for month-to-month, 3-year, and 5-year terms across 3
density zones.^24 As previously noted, FCC requires price-cap
incumbent firms to file list prices in all areas that they serve.
Price-flex list prices are made generally available in areas with
phase II flexibility. Price-cap list prices are made generally
available to all customers in areas with phase I pricing
flexibility as well as all other areas that remain subject to full
price-cap regulation.
o Because many larger customers may purchase dedicated access
through various contracts with incumbents, we analyzed a
substantial number of these contracts, which each price-cap
incumbent firm also files with FCC.
o We could not obtain specific data on the prices paid by
individual customers purchasing dedicated access services at
various pricing levels (i.e., month-to-month or 3-year terms, or
different density zones) or under different contract options, or
the exact amount of dedicated access purchased. Therefore, as a
proxy for the average prices charged, we analyzed the average
revenue that price-cap incumbents received from selling dedicated
access in 56 selected MSAs under phase I flexibility or phase II
flexibility for channel terminations. We compared changes in
average revenue for channel terminations between the period prior
to pricing flexibility being granted and 2005, and also compared
average revenue in 2005 across areas under phase I flexibility,
phase II flexibility, and remaining under the price cap. We
obtained average revenue data for the 56 MSAs under pricing
flexibility from the four major price-cap incumbents. We obtained
average revenue data for price-cap areas from annual tariff review
plans submitted to the FCC by price-cap incumbents. Because only 1
of the MSAs in the data provided to us by the price-cap incumbents
was under phase I flexibility for dedicated transport, we were
unable to conduct a comparison of price trends for transport under
different phases of pricing flexibility. These averages mask
variation that exists across MSAs and across price-cap incumbents.
Because the data provided by the price-cap incumbents are
proprietary, we relied on these averages to examine overall trends
in markets under different phases of pricing flexibility.^25 We
were unable to independently verify the reliability of these data.
However, we performed logic tests that were based on listed prices
and available discounts to determine if there were any major
inaccuracies.
In each of the three methods, we limited our analysis to prices
for high-capacity dedicated access services at two speeds--1.544
megabytes per second (Mbps), which is known as a DS-1 circuit, and
45 Mbps, which is known as a DS-3 circuit--because they represent
the majority of dedicated access revenues that the price-cap
incumbent firms generate. We also were unable to collect data on
prices that competitors charged; therefore, those prices are
excluded from this analysis. According to competitors, they could
not provide data on prices because of nondisclosure agreements
they have in place. We interviewed representatives from these
firms, and they provided anecdotal information about their prices.
Furthermore, we were unable to measure the extent to which price
trends related to cost trends, because these data were also
unavailable.^26 In addition, we analyzed available data on prices
that two federal government departments paid under General
Services Administration (GSA) contracts as well as prices paid
under separate agency contracts. To determine what data FCC uses
to monitor competition and any limitations that may exist to their
monitoring efforts, we reviewed and analyzed FCC triggers for
predicting competition as well as FCC data collection processes
for determining and monitoring competition. We analyzed FCC's
strategic plan, performance budget, and measures used by the
agency to track their progress toward meeting their stated goals
of increasing competition and choice for businesses. We also
reviewed the rulemaking proceeding on dedicated access and public
comments filed in that proceeding. We conducted our work from
November 2005 through October 2006 in accordance with generally
accepted government auditing standards. See appendix 1 for a more
detailed discussion of our objectives, scope, and methodology.
Results in Brief
In the 16 major metropolitan areas we examined, facilities-based
competition for dedicated access services exists in a relatively
small subset of buildings. Our analysis of data on the presence of
competitors in commercial buildings suggests that competitors are
serving, on average, less than 6 percent of the buildings with at
least a DS-1 level of demand. Competition is more widespread where
buildings have a higher level of demand. For the subset of
buildings identified as likely having companies with a DS-3 level
of demand, competitors have a fiber-based presence in about 15
percent of buildings on average. For buildings identified in our
model with 2 DS-3s of demand, competitors have a fiber-based
presence in 24 percent of buildings on average. The data also show
that the theoretically more competitive phase II areas generally
have a lower percentage of lit buildings than phase I areas,
indicating that FCC's competitive triggers may not accurately
predict competition at the building level. The data also show that
there has been a decline in some MSAs in the level of competitive
colocation in the wire centers used by the price-cap incumbents to
obtain pricing flexibility. Limited competitive build out in these
MSAs could be caused by a variety of entry barriers, including
zoning restrictions, or difficulties in obtaining access to
buildings from building owners that discourage competitors from
extending their networks. In addition, where demand for dedicated
access is relatively small, such as buildings with less than three
or four DS-1s of demand, it is unlikely to be economically viable
for competitors to extend their networks to the end user.
Incumbent firms have noted that, where competitors can lease UNEs
from incumbent providers, there may be less incentive for
competitors to invest in their own facilities. However, even if
UNEs were not available, competitors still may find it
uneconomical to extend their own networks to end users if their
demand for dedicated access is relatively low.
Since FCC first began granting pricing flexibility in 2001,
average revenue from channel terminations and average revenue for
dedicated transport across the four major price-cap incumbents has
generally decreased. This suggests that average prices may have
fallen as well and is generally what would be expected with
automatic decreases to the price-cap list prices required under
FCC's existing CALLS Order. Additionally, the decrease appears to
be consistent with the prospect of competition that FCC predicted.
However, our analysis of data from the four major price-cap
incumbent firms and FCC, which was intended to determine how
prices have changed since the granting of phase II pricing
flexibility, generally shows that prices and average revenues are
higher, on average, in phase II MSAs--where competition is
theoretically more vigorous--than they are in phase I MSAs or in
areas where prices are still constrained by the price cap.
o Since phase II pricing flexibility was first granted, list
prices for dedicated access that apply under phase II, on average,
have increased. Conversely, price-cap list prices available in
phase I and price-cap areas were pushed downward over the same
period--largely by the CALLS order. As a result, average list
prices in areas with phase II flexibility are higher than average
list prices in phase I and price-cap areas.
o According to representatives of the price-cap incumbent firms,
many customers that represent a significant amount of revenue
purchase services under price-flex contracts that apply additional
discounts to circuit components in areas with phase I or II
pricing flexibility. However, most of these contracts provide
overall discounts off the list price, and, therefore, since
price-flex list prices are higher on average than price-cap list
prices, prices will remain higher in phase II areas. Over time,
however, our analysis shows that most contracts provide discounts
that, coupled with CALLS Order decreases in phase I areas, can
eliminate any increases in the list prices and result in an
overall decrease in price when compared with prices that existed
prior to pricing flexibility. For many contracts we were unable to
determine their effect on net prices because certain data were
unavailable. Competitors also argue that price-flex contracts
require customers to meet contractual terms and conditions that
may limit the ability of competing vendors to win that business.
For example, contracts may include termination penalties that
discourage customers from switching to competing firms during the
length of the contract.
o Not all customers are under price-flex contracts, and detailed
data on the number of customers and circuits that are purchased at
the price-flex list price were not available to us. Therefore, we
compared average revenue data for dedicated access services under
price-cap regulation (filed with the FCC), and under phase I or
phase II flexibility (on the basis of data from 56 MSAs--27 phase
II and 29 phase I--provided by the four major price-cap
incumbents) to examine the net effect of changes in list prices
and the application of contract discounts. Average revenue for
channel terminations and dedicated transport for DS-1 and DS-3 has
generally decreased over time, although the decline in average
revenue for channel terminations is larger in phase I areas
compared with phase II areas. Comparing average revenue across
price-cap areas, phase I areas, and phase II areas as of 2005--the
most recent period available--we found that average revenue in the
27 phase II areas is higher, on average, than it is in the 29
phase I areas and not statistically different that average revenue
in areas that are still under a price cap.
Although no total spending figures are available, the federal
government is also a large consumer of dedicated access services.
Our review of spending on dedicated access services by the
Department of Agriculture (USDA) and DOJ indicated that they
procured most of their services through the government-wide
telecommunications contract, FTS2001. With FTS2001 expiring and
GSA currently negotiating a new contract, it is unclear what
prices the government will pay for dedicated access services.
FCC uses various data to assess competition for dedicated access,
but most of these data have significant limitations in their
ability to describe the presence, extent, or change in competition
in any given area. For example, the data presented in a price
flexibility petition measure potential competition at one point in
time and FCC does not revisit or update them, even though
competitors may enter bankruptcy or be bought by another firm. FCC
also attempts to collect data from external parties through its
rulemaking proceedings, but those parties generally have no
obligation to provide data, and FCC has limited mechanisms to
verify the reliability or accuracy of any data submitted. For
example, as part of its rulemaking proceeding on dedicated access,
FCC requested data on price indices in price flexibility areas to
determine how prices have changed in areas with varying levels of
price deregulation; however, no incumbent firm provided these
data. FCC's strategic plan and various rulemakings have defined
FCC's obligation to assess and ensure competition for dedicated
access. FCC has stated that gathering and analyzing additional
data would be costly and burdensome. FCC has expressed concern
about its ability to gather data without disturbing the market and
also noted that any additional reporting requirements on incumbent
firms or any requirements on competitive firms to report
information on their networks would have to conform to the
Paperwork Reduction Act^27 and also proceed through a lengthy
administrative process. Certainly, FCC must balance the additional
costs of gathering more data with the potential benefit that might
result from additional data. Yet without more complete and
reliable measures of competition, FCC is unable to determine
whether its deregulatory policies are achieving their goals.
We are making recommendations to FCC to revisit the issues it
initiated in its rulemaking proceeding on dedicated access and to
develop measures and methods to monitor competition on an ongoing
basis that more accurately represents market developments and
customer choice. We provided copies of the draft report to FCC for
its formal comment. FCC did not disagree with the facts presented
in the report but contended that the report implied a need for
regulatory price controls and consequently disagreed with the
recommendations. Counter to FCC's interpretation, the report does
not call for the reregulation of dedicated access prices. Instead,
the report concludes that in order to better meet its regulatory
responsibilities, FCC needs a more accurate measure of effective
competition and needs to collect more meaningful data. We also
made copies of the draft report available to the major incumbent
carriers. They generally disagreed with the information presented,
stating that the data we used were incomplete and unreliable. We
recognize the limits of available data on the extent and effect of
competition in the market for dedicated access services, but
believe the data used provided a reasonable and sufficiently
reliable picture of the extent of facilities-based competition.
Background
FCC is an independent United States government agency, directly
responsible to Congress. Established by the Communications Act of
1934, FCC is charged with regulating interstate and international
communications by radio, television, wire, satellite, and cable.
The Telecommunications Act of 1996 established that FCC should
promote competition and reduce regulation to secure lower prices
and higher quality services for American telecommunications
consumers and encourage the rapid deployment of new
telecommunications technologies. FCC's strategic plan clarifies
its support for these principles by stating that competition in
the provision of communications services, both domestically and
overseas, supports the Nation's economy.
After passage of the 1996 Act, FCC started several actions to
encourage competition in the telecommunications market. As
previously discussed, FCC instituted pricing flexibility where
incumbents could meet specific competitive triggers. Table 1
summarizes the conditions under which incumbents could qualify for
phase I or phase II flexibility. According to FCC, all of the
applications for pricing flexibility have utilized the
revenue-based triggers, and not the percentage of total wire
centers in an MSA.
^22Telcordia and COMMON LANGUAGE are registered trademarks and CLCI, CLEI,
CLFI, CLLI, and NC/NCI are trademarks of Telcordia Technologies, Inc.
^23The 16 MSAs we included in our analysis were as follows: Atlanta,
Georgia; Chicago, Illinois; Detroit, Michigan; Greenville, South Carolina;
Los Angeles, California; Miami, Florida; Minneapolis, Minnesota; New
Orleans, Louisiana; New York, New York; Norfolk, Virginia; Phoenix,
Arizona; Pittsburgh, Pennsylvania; Portland, Oregon; San Jose, California;
Seattle, Washington; and Washington, D.C.
^24Typically, price-cap incumbents offer prices across different zones
that reflect the concentration of business demand for dedicated access
within a geographic area. Zones generally correspond with areas of
relatively high, medium, and low business demand density. Zone 1 is
generally considered as inclusive of the central business area, where a
large portion of businesses that would require DS-1 and DS-3 would reside.
Prices are generally lower in zone 1 than in zones 2 or 3--with zone 3
generally having the highest prices, because costs to provide services are
likely higher in less dense areas. Occasionally an incumbent will offer
prices across five zones. In cases where an incumbent provided pricing
across five zones, we analyzed prices associated with zones 1, 3, and 5.
^25Not all of the major incumbent firms were able to include every
discount that was based on price-flex contracts. One firm was unable to
include discounts that were based on revenue commitments; however, because
these discounts are available in both phase I and phase II areas, there is
little reason to believe that these discounts would affect the prices
available in phase II areas greater or less than it would affect prices in
phase I areas.
^26FCC discontinued cost studies several years ago. In addition, FCC gave
price-cap incumbents the option to accept the CALLS Order price decreases,
or to have prices reinitialized on the basis of detailed cost studies. No
price-cap incumbents provided a cost study.
^27The Paperwork Reduction Act sets standards for information collection.
These standards include avoiding unnecessary duplication; reducing burdens
on the public and small entities; ensuring that collection is developed so
that information is used in an efficient and effective manner; and using
information technology to the maximum extent practicable to reduce burden
and improve data quality, agency efficiency, and responsiveness to the
public. 44 U.S.C. SS 3501 et. seq. In addition, pursuant to the Small
Business Paperwork Relief Act of 2002, Pub. L. No. 107-198, [35]44 U.S.C.
S 3506 (c)(4), FCC generally seeks specific comment on how it might
"further reduce the information collection burden for small business
concerns with fewer than 25 employees."
Table 1: Conditions for Granting Different Phases of Pricing Flexibility
in an MSA
Requires installed colocation
Level of pricing Dedicated access equipment from at least one
flexibility components competitor in:
Partial ("phase I") Channel terminations Wire centers with or 50 percent
price deregulation to end users 65 percent of of total
revenues wire centers
Dedicated Wire centers with or 15 percent
transportation^a 30 percent of of total
revenues wire centers
Full ("phase II") Channel terminations Wire centers with or 65 percent
price deregulation to end users 85 percent of of total
revenues wire centers
Dedicated Wire centers with or 50 percent
transportation^a 65 percent of of total
revenues wire centers
Source: FCC Pricing Flexibility Order.
Note: In addition to the wire center requirements, price-cap incumbents
must also demonstrate in their petitions for price flexibility that, in
each wire center relied on in the applicant petition, at least one
competitor relied on dedicated transport facilities provided by a
nonincumbent carrier.
^aDedicated transport includes entrance facilities, direct-trunked
transport, and the flat-rated portion of tandem-switched transport as well
as dedicated access services other than channel terminations to end users.
Phase I flexibility is essentially downward pricing flexibility. Under
phase I flexibility, prices charged by price-cap incumbents are generally
not expected to increase (except for changes to the price cap resulting
from cost factors outside the price-cap incumbents' control, such as taxes
and fees, or from increases in certain prices under the price cap, which
would require the price-cap incumbent to lower other prices under the
cap). Phase II flexibility allows price-cap incumbents to raise or lower
their list prices. Although prices are permitted to increase, under phase
II flexibility competition is expected to constrain incumbent pricing
power. The Pricing Flexibility Order noted that it is possible past
regulation may have resulted in setting some prices below cost, and,
therefore, some price increases would be expected under phase II
flexibility,^28 although the order does not speculate on which prices may
increase or to what extent price increases were expected. Regulation could
have caused prices to be below costs in some areas because price-cap
incumbents are required to offer the same "average" price throughout a
geographic area, although costs may not be uniform throughout that area.
As a result, the price-cap incumbent may have had to price certain
services too high where costs were low, and too low where costs were high.
Therefore, if the triggers are correct, and sufficient competition does
exist in phase II areas, one might expect prices in the more dense areas
of metropolitan areas to decrease (perhaps as a result of contract
offerings) because costs are likely lower in these areas. Prices in more
sparsely populated parts of the MSA, with fewer businesses, may show some
increases due to the likely higher cost of providing service in such
areas.
If the price-cap incumbents raise their prices, and there is not
sufficient competition to constrain these prices, competitors' input costs
to serve business customers may rise. Generally speaking, economic theory
holds that as competitors' input costs increase, they would have greater
incentive to search for an alternative supply of that input, or produce it
themselves, leading to further entry by competitors into building their
own facilities to compete with incumbents. Representatives of several
competitive providers with whom we spoke stated that they generally prefer
to provide service over their own facilities or to purchase from other
competitive providers, wherever it is possible to do so, rather than use
incumbent facilities because of price considerations, concerns over the
reliability of the facilities, and the ability to quickly fix service
problems should they arise. Alternative supply for dedicated access can
also be provided by competitors in the form of alternative technologies,
such as point-to-point wireless connections. Some industry analysts when
we spoke were encouraged by the prospect of fixed wireless and WiMax
technology that could provide alternative dedicated access. However,
according to these analysts, this technology is still being developed and
has only been used in limited circumstances to replace high-capacity
dedicated access connections.^29
28Pricing Flexibility Order, 80.
An alternative theory suggests that an incumbent can discourage full
facilities-based entry into the market by allowing competitors to lease
the incumbent's network at a price just equal to the competitors' cost to
build out their own networks, thus making the competitors indifferent to
leasing or building. This theory would suggest that prices will adjust to
the point where it remains uneconomical for competitors to build their own
facilities. By extension, this theory argues that full facilities-based
competition is unlikely to ever occur, because the incumbent will
discourage entry by keeping competitors on its own network, rather than
risk splitting or losing the market. Representatives of incumbent firms
have pointed out that the vast majority of their dedicated access services
are sold to competitors on a wholesale basis.
The federal government has been somewhat shielded from market developments
through long-term contracts that GSA negotiated. Government agencies can
acquire telecommunications services and dedicated access services through
GSA's FTS2001 contracts. FTS2001 contracts will soon be replaced by GSA's
"Networx" contracts, which are planned for award in 2007. FTS2001 was
negotiated with set rates for dedicated access connections over the life
of the contract. However, government agencies are not required to use the
FTS2001 contract, and not all agencies purchase off of this contract.
^29WiMAX is defined as Worldwide Interoperability for Microwave Access by
the WiMAX Forum, and was formed in April 2001 to promote conformance and
interoperability of the IEEE 802.16 standard, officially known as
WirelessMAN. The forum describes WiMAX as "a standards-based technology
enabling the delivery of last mile wireless broadband access as an
alternative to cable and DSL."
Facilities-Based Competition to End Users Does Not Appear to Be Extensive
Based on the data available to us, facilities-based competition for
dedicated access services to end users at the building level (i.e.,
analogous to channel terminations to end users) does not appear to be
extensive in the MSAs we examined, although moderate levels of competition
appear where demand for dedicated access exceeds the DS-3 level. The data
further suggest that there have been some declines in competition in wire
centers used by incumbents to obtain pricing flexibility. These findings
suggest that FCC's competitive triggers--which look at competition at the
wire center level--may not adequately predict competition at the building
level throughout an MSA. The limited amount of facilities-based
competition could be due to a variety of factors, including the high cost
of constructing local telecommunications networks, government regulations,
and limited competitive access to buildings.
Competitive Alternatives Exist in a Relatively Small Subset of Buildings
According to data from July 2006, facilities-based competitors have
extended their networks to a relatively small subset of buildings in the
MSAs that we examined.^30 Of the buildings with a level of demand greater
than the DS-1 level in our model, we found that only about 6 percent of
buildings, on average, have a fiber-based competitor. Competition is more
widespread where buildings have a higher level of demand. For the subset
of buildings identified in our model as likely having companies with a
DS-3 worth of demand, competitors have a fiber-based presence in 15
percent of buildings, on average. For buildings identified in our model
with at least 2 DS-3s of demand, competitors have a fiber-based presence
in 25 percent of buildings, on average (see table 2). The data also show
that phase II areas--which are the theoretically more competitive
MSAs--generally have a lower percentage of lit buildings than phase I
areas.
^30These figures are the product of the following two sources: (1) the
number of buildings believed to be served by competing firms present in
the database and (2) estimates of the total number of commercial buildings
in each MSA that include commercial businesses likely to demand dedicated
access with at least a DS-1 level of service, at least one DS-3 of
service, or 2 DS-3s or greater of service. We used the Telcordia Location
Registry (formerly "CLONES") to make inferences about buildings lit by
competitors. The Location Registry is a hosted database of network
locations and related network functions for the telecommunications
industry. Estimates of commercial buildings with demand for dedicated
access are derived from a proprietary model owned by GeoResults. That
model estimates likely demand generally on the basis of the type of
business, number of employees per business location, and existence and
size of the corporate parent to that business location. GeoResults reports
that all major incumbent firms, as well as a number of competing firms,
also use the model to forecast demand.
We analyzed the extent to which competitors had extended their networks to
end users utilizing fiber and wireless based connections. While other
transmission mediums are available (such as copper wire) competitors have
generally built fiber networks that have greater capacity and lower costs.
See appendix I for additional information.
Table 2: Percentage of Buildings with a Fiber-Based Competitive
Alternative by Demand (July 2006)
Buildings
Buildings Number of Number of with Number of
with buildings Buildings buildings demand of buildings
demand of with a Percent with with a Percent 2 DS-3s with a Percent
DS-1 or "lit" with a demand of "lit" with a and "lit" with a
MSA^a greater competitor competitor DS-3 competitor competitor greater competitor competitor
Phase II MSAs
Atlanta 12,718 446 3.5% 278 25 9.0% 67 10 14.9%
Los Angeles 22,639 508 2.2% 650 26 4.0% 265 34 12.8%
Miami 14,300 363 2.5% 421 14 3.3% 136 21 15.4%
Norfolk 5,008 2,080^b 41.5% 56 35 62.5% 13 9 69.2%
Phoenix 7,981 297 3.7% 155 17 11.0% 51 5 9.8%
Pittsburgh 4,733 383 8.1% 78 15 19.2% 25 9 36.0%
Portland 3,683 126 3.4% 67 9 13.4% 26 3 11.5%
San Jose 4,653 287 6.2% 98 13 13.3% 20 2 10.0%
Total 75,715 4,490 5.9% 1,803 154 8.5% 603 93 15.4%
Phase I MSAs
Chicago 16,732 361 2.2% 325 37 11.4% 185 47 25.4%
Detroit 12,174 298 2.4% 168 13 7.7% 48 2 4.2%
Greenville 2,551 68 2.7% 28 5 17.9% 4 0 0.0%
Minneapolis 6,786 389 5.7% 147 31 21.1% 56 12 21.4%
New Orleans 3,540 207 5.8% 65 15 23.1% 37 9 24.3%
New York 34,650 2,354 6.8% 762 198 26.0% 380 158 41.6%
Seattle 4,951 188 3.8% 100 15 15.0% 51 14 27.5%
Washington,
D.C. 20,472 1,967 9.6% 518 131 25.3% 146 40 27.4%
Total 101,856 5,832 5.7% 2,113 445 21.1% 907 282 31.1%
Grand total 177,571 10,322 5.8% 3,916 599 15.3% 1,510 375 24.8%
Source: GAO analysis of Telcordia and GeoResults data.
^aThis table includes the portions of the MSAs that are served by the
incumbent firm. Therefore, we excluded portions of those MSAs where
another incumbent firm provides service, but may not have the same level
of pricing flexibility. For example, both AT&T and Verizon serve parts of
Los Angeles, but we only considered those areas that AT&T serves, because
AT&T has received phase II flexibility for channel terminations in Los
Angeles.
^bAccording to a local cable official, Norfolk's high competition numbers
are due to the local cable company's long-term financial commitment to
build a fiber optic network to provide business telecommunications
services.
The data in table 2 may overstate the availability of facilities-based
competition to some extent. Some equipment that does not provide service,
no longer provides service, or no longer exists may remain in the
database, falsely indicating a competitive presence. Several companies and
government agencies, such as mobile telephone companies and GSA, are
included in the number of competitors, even though they do not provide
dedicated access connectivity for businesses. Also, according to
GeoResults, cellular phone sites are significantly underrepresented in the
number of buildings with demand for dedicated access. However, cellular
sites with competitive fiber are included in the number of buildings with
a fiber-based competitor. Furthermore, these numbers include bankrupt
companies, such as Jato Communications and Ciera Network Systems, whose
equipment is still listed in the database. It is unclear whether these
assets are being used by another company or have been liquidated. These
data also include equipment owned by the former AT&T and MCI prior to the
recent mergers. We did not filter out these data because DOJ has required
divestiture of some of these assets and the courts have yet to finalize
that action. DOJ's analysis is discussed further later in this report.
In addition, the results from table 2 also may understate facilities-based
competition to some extent. Both incumbent and competitive firms
voluntarily populate their network locations and functions into the
database for the purposes of interconnection and network management.
According to Telcordia, data on competitive firms may be less
comprehensive than data on incumbent firms, but a precise estimate of
underreporting is not available from Telcordia. In order to gauge the
extent that the data are underreported, we compared entries in the
database with lists of "lit" buildings provided to us by two of the
largest competitive firms. One firm showed 465 lit buildings in the data
they provided to us in the 16 MSAs we examined, of which, 436 showed the
presence of a "lit" competitor, suggesting an underreporting error of a
little over 6 percent. However, the database also showed this same
competitor as being the sole competitive presence in 81 additional
buildings that were not on the firm's list of lit buildings, suggesting,
that, for this competitor, the database is overreporting the level of
competition by about 12 percent. However, the other firm from which we
obtained data provided us a list with 693 lit buildings in the MSAs we
examined, of which, 289 showed the presence of a "lit" competitor,
indicating underreporting of about 400 buildings across the MSAs for this
competitor. These two examples show that individual competitor's presence
may be underreported and overreported. One price-cap incumbent has
suggested that the database may be underreported by 30 percent, although
representatives of GeoResults disagreed that the data are underreported to
that extent. If the data were underreported by 30 percent, we would find a
competitive presence in 8 percent of buildings with demand greater than
DS-1; 20 percent of buildings with demand of DS-3, and about 32 percent of
buildings with demand greater than 2 DS-3s. These estimates still suggest
that competitive alternatives exist in a relatively small subset of
buildings, with more moderate levels of competition in buildings where
demand is higher.
Because there is no compulsory process through which telecommunications
companies report such data to FCC or private data sources, no single
public or private data source is universally recognized as comprehensive.
As we have indicated, the data may be understating or overstating
competition to varying degrees. It is not clear the extent to which
underreporting by competitors will be offset by the inclusion of bankrupt
and merged companies. Regardless, this database is the most comprehensive
available to us, and price-cap incumbent firms, such as AT&T and
BellSouth, have used the database for similar purposes. We discuss data
reliability in more detail in appendix I.
Our competition analysis, while not a complete representation of
competition, is a more granular view than that taken by FCC in its Pricing
Flexibility Order--which was to extrapolate the state of competition
throughout an MSA by the presence of competitors' equipment colocated in
incumbent firms' wire centers. We analyzed the extent of competitive entry
in a market at the level of individual buildings--that is, at individual
locations where business or government end users would choose from service
providers to purchase dedicated access. In its review of the SBC/AT&T and
Verizon/MCI mergers, DOJ's Antitrust Division also adopted the building
level as its basis of analysis. There is some disagreement among FCC,
incumbent firms, and competitors on the appropriate level of analysis to
judge the state of competition for dedicated access. For example, some
observers have stated that the proper level of granularity for any
competition analysis is not the presence of a competitor within a
building, but the presence of the competitor at the business location
within that building. Competitors have pointed out that while they may
have a connection to a building, they are unable to connect to businesses
on all floors within that building. In this case, our analysis would be
overstating the level of competition. However, that level of detailed data
is not available.
Another view taken by some observers is that, from a business'
perspective, demand for dedicated access will be determined by that
business' individual location and the other locations where the business
needs dedicated access, such as field offices or branches. These other
locations could be within the same MSA or could be spread out over several
MSAs, several states, or even nationwide. For example, a bank may have 30
or 40 locations in 12 states in one region of the country that require
dedicated access. To serve that customer wholly over its own facilities, a
competitor would need to extend its network to all of those locations.
Alternatively, a competitor could compete for that customer against the
incumbent, who likely has connections to all of the customer's locations
in that region, using some of its own facilities and some facilities
purchased from the incumbent or from another competitor. Our analysis does
not consider an individual customer's total demand--a level of data that
is unavailable--but rather their demand within a building in an MSA.
However, because the percentage of buildings in these MSAs with a
competitor appears to be relatively small, our analysis suggests that it
is unlikely that a single competitor would have very many of its own
facilities to serve such a customer.
FCC's Metric Shows a Decline in the Extent of Colocation in Some MSAs since the
Granting of Pricing Flexibility
Using FCC's competition metric--competitive colocation in incumbent wire
centers--the data suggest that, for some MSAs, fewer competitors exist in
the wire centers used by the incumbents to meet FCC's competitive triggers
than when the incumbents were granted pricing flexibility. In fact, in
many MSAs we examined, some wire centers that had competitive colocation
several years ago, appear to no longer have any competitive colocation.
Price-cap incumbents have noted in the rulemaking proceeding on dedicated
access that competitors will often bypass their wire centers, and that
FCC's trigger would not detect these competitors. This analysis cannot
test the extent to which formerly colocated competitors have removed
equipment to bypass incumbent facilities, or the extent to which bypass
occurs in general. Table 3 shows the change in the number of price-cap
incumbents' price-flex wire centers used to meet FCC's competitive
triggers in which competitors were colocated as of July 2006.^31
31Because Telcordia's database is used primarily for interconnection
purposes, it is likely that there is little underreporting of competitors'
presence in price-cap incumbent wire centers.
Table 3: Change in Competitive Colocation in "Price-flex" Wire Centers
Number of "price-flex" wire centers
with competitive colocation
In pricing flexibility
MSA application July 2006 Percentage change
Phase I channel termination markets
Chicago 58 42 (28)
Detroit 27 22 (19)
Greenville 5 5 0
Minneapolis 20 18 (10)
New Orleans 6 6 0
New York 80 75 (6)
Seattle 11 11 0
Washington, D.C. 46 39 (15)
Total 214 193 (10)
Phase II channel termination markets
Atlanta 16 16 0
Los Angeles 64 64 0
Miami 38 38 0
Norfolk 15 15 0
Phoenix 19 18 (5)
Pittsburgh 34 22 (33)
Portland 10 10 0
San Jose 11 11 0
Total 256 227 (11)
Source: GAO analysis of Pricing Flexibility applications, and Telcordia
and GeoResults data.
Note: Percentages have been rounded.
FCC does not monitor ongoing colocation to continually affirm that the
triggers are met, even in situations where major mergers or bankruptcies
may change the competitive landscape. According to FCC, continually
deregulating and reregulating prices on the basis of such changes would
not produce a desirable outcome and the costs of such actions would likely
outweigh the potential benefits.
The recent telecommunications mergers between AT&T and SBC, and between
Verizon and MCI--although still undergoing an Antitrust Procedures and
Penalties Act (Tunney Act)^32 review by the federal courts as of the date
of this report--are also likely to decrease the number of buildings with a
competitor for dedicated access in the MSAs we examined. DOJ's Antitrust
Division conducted a review of the effects these mergers would have on
competition. DOJ concluded that, viewed as a whole, the transactions were
likely to create substantial efficiencies that could benefit consumers.
However, DOJ found that, for the vast majority of buildings in the MSAs it
reviewed, no competitive providers of dedicated access facilities existed,
which is consistent with the data in table 2. For the purposes of its
merger review, however, DOJ did not review the state of competition in the
dedicated access market as a whole, but rather focused on the hundreds of
buildings where the transactions would combine the only two firms that
owned or controlled a direct fiber-optic connection to the building.^33
For those buildings where the competitors were reduced from two to one,
DOJ used the level of dedicated access demand in a building, coupled with
the distance of the building from the nearest competitor's network in the
MSA, to determine whether competitors could be induced to enter. For a
subset of the two to one buildings, potential entry was not sufficiently
likely to offset the potential anticompetitive effect. For this subset of
buildings, DOJ proposed a remedy designed to eliminate those
anticompetitive effects, under which the companies would be required to
divest "indefeasible rights of use" for connections to those buildings,
along with transport facilities sufficient to enable purchasers to provide
competing telecommunication services. The proposed divestiture involved
hundreds of buildings in 8 metropolitan areas in Verizon's franchised
territory and 11 metropolitan areas in SBC's franchised territory. Despite
the divestiture, for the other set of two to one buildings where DOJ
deemed entry to be likely, there will at least be an initial reduction in
competition before any entry occurs.
^32The Tunney Act requires that proposed consent judgments in antitrust
cases brought by the United States be subject to a 60-day comment period,
after which the courts shall determine whether entry of the proposed Final
Judgment "is in the public interest." 15 U.S.C. S 16(e)(1).
^33Dedicated access, as previously mentioned, can be provided over other
mediums, such as copper wire, or using wireless technology; however,
competitive firms have generally used fiber-optic cable to build their
networks.
Concerns have arisen that the proposed merger between AT&T and BellSouth
may cause a further decline in competitive alternatives available in
BellSouth's territory. DOJ has also reviewed this merger, and, using the
same criteria it used in the mergers between AT&T and SBC, and between
Verizon and MCI, DOJ determined that AT&T could provide service over its
own facilities to only a small minority of buildings in BellSouth's
territory. The potential of other competitors' extending their networks to
serve that minority of buildings was substantial enough to make
divestitures, such as those ordered in the other mergers, unnecessary.
FCC's review of this proposed merger is still ongoing.
Limited Competition Could Be Caused by a Variety of Factors
The apparent limited competition at the building level could be caused by
a variety of factors, including the high sunk costs--that is, costs that
once incurred cannot be readily recovered--of constructing local networks,
the cost of local government regulations, and limited access to buildings.
All of these factors can increase competitors' cost to deploy facilities
and provide dedicated access services to locations within an MSA.
Constructing a local telecommunications network can be extremely capital
intensive. Most communications equipment has no other use and therefore
can not be reused for alternative purposes. Because these investments
would have virtually no alternative value if the business fails,
competitors must have a certain level of expected revenue to extend their
networks. The level of demand required for a competitor to build out its
own facilities varied across the firms we interviewed depending on the
extent to which the firm had already invested in the market, and the
distance of the potential customer from the competitor's network. Based
purely on the expected returns on their capital investment and ignoring
other potential barriers, representative from one firm estimated that they
would need three to four DS-1s of demand, while representatives from two
other firms estimated demand of greater than 2 DS-3s was required.
However, one incumbent firm and one cable company noted that the necessary
revenue to extend a nearby network into a building is relatively low.
Incumbent firms also have noted that the availability of UNEs may provide
disincentives for competitive firms to extend their networks, because
these rates are generally below the prices charged for dedicated access
services. Therefore, competitive firms may have an incentive to continue
to lease UNEs as opposed to incurring the costs of extending their own
networks. However, FCC has recently limited the availability of UNEs.
Others have argued that, as UNEs become less available to competitors,
competitors may still find it uneconomical to extend their own networks to
those end users, and will instead have to purchase those connections as
dedicated access services from the incumbent. Moving from UNE rates to
dedicated access pricing effectively raised the cost of the competitor to
serve the end user customer, and may create a disadvantage for the
competitor in trying to win or retain that customer.
In addition, local government regulations also can increase the cost of
deploying these networks. The local government building and zoning
permitting process often includes extensive inquiries into the planned
construction activities of a competitor. This process can delay the
deployment of competitive networks and raises competitors' costs.
Representatives of some competitive firms we interviewed also stated that
some cities have moratoriums on construction that prevent them from
providing service to certain buildings.
Lastly, competitors also noted that it may be difficult to provide service
into buildings because some building owners may seek to charge competitors
for extending their network into their buildings or may refuse access to
additional carriers. This additional cost can be prohibitively expensive
because the building owner may demand a percentage of the revenue that
competitors earn in that building as a condition of granting access. In
such cases, competitors are forced to lease dedicated access lines from
the incumbent to serve customers in those buildings.
Prices for Dedicated Access Services in MSAs with Phase II Pricing Flexibility
Are on Average Higher Than Prices Elsewhere
Since FCC first began granting pricing flexibility in 2001, our comparison
of prices and revenue across phase I flexibility and phase II flexibility
suggests that list prices and revenue are higher on average for circuit
components in areas under phase II flexibility (areas where competitive
forces are presumed to be greatest) than in areas under phase I
flexibility or under price caps. First, our comparison of 1,152 list
prices for channel terminations and dedicated transport for both monthly
and multiyear terms found that price-flex list prices were almost always
higher than price-cap list prices. This is a result of the following two
effects: (1) price-flex prices have increased over time on average and (2)
the CALLS Order has pushed price-cap prices downward on average. However,
according to representatives of the incumbent firms, many of the largest
customers in pricing flexibility markets are under price-flex contracts.
Many of these contracts provide discounts off of the applicable price-cap
or price-flex list price. Because of the differences in the underlying
list prices, contract prices for dedicated access in phase II areas will
still be higher than phase I areas. Some contracts also contain terms and
conditions that, competitors argue, may limit a customer's ability to
choose other vendors. Third, average revenue for channel terminations and
dedicated transport for DS-1 and DS-3 in 2005 are generally lower than
average revenue in 2001 and 2002, although the decline in average revenue
for channel terminations is larger in phase I areas compared with phase II
areas. Furthermore, as of 2005, average revenue for channel terminations
is higher, on average, in phase II areas than in phase I areas or
price-cap areas.
Although no total spending figures are available, the federal government
is also a large consumer of dedicated access services, most of which are
procured through the FTS2001 contracts. Our analysis of federal agencies'
spending found that while most dedicated access services were acquired
through FTS2001--which is acknowledged to have low prices--there were
limited instances where agencies could be paying more, either by
purchasing independently of FTS2001 or not using the lowest cost FTS2001
service provider. With FTS2001 expiring and GSA currently negotiating a
new contract, it is unclear what prices the government will pay.
Price-Flex List Prices Are Higher on Average Than Price-Cap List Prices
Our comparison of 1,152 prices found that, as of June 2006, the price-flex
list price was on average higher than the price-cap price, regardless of
whether the price was for channel terminations, interoffice mileage, DS-1
or DS-3 service, different term arrangements, or different density zones.
This is due to two effects. First, price-flex prices as of June 2006 are
higher on average than list prices in effect just prior to FCC granting
pricing flexibility. As previously discussed, FCC expected price increases
in some areas, and these increases would likely be in areas where costs
were higher and, therefore, regulation had pushed prices below costs. Our
analysis showed that prices increased on average, regardless of density
zone or any other parameters--although prices did increase more on average
in lower density areas than higher density zones, and increased more for
shorter term lengths than longer term lengths. Second, list prices
available in areas where price caps remain (MSAs with phase I flexibility
and MSAs under full price-cap regulation) decreased on average over the
same period, mainly as a result of the CALLS Order. (See app. II for more
information on this analysis and for results using data adjusted for
inflation into 2005 dollars using the Bureau of Labor Statistics Producer
Price Index for Wired Telecommunications Carriers. Adjusting prices using
this price index does not change the result that prices are higher in
phase II areas on average, or that prices have increased over time in
phase II areas. However, adjusting prices using this price index resulted
in price-cap prices increasing in constant terms over the period.) Average
differences in dedicated access prices across all terms and zones are
shown in figure 2.
Figure 2: Average Differences in Dedicated Access Prices Across All Terms
and Zones in Nominal Dollars
Circuits may cross MSA boundaries, and customers may purchase circuits in
several different MSAs with different levels of pricing flexibility.
Therefore, looked at more broadly, a full circuit, or a customer's entire
purchase may cost less overall than it did prior to pricing flexibility.
In other words, decreases resulting from the CALLS Order, coupled with
contract discounts may offset any increases in price-flex list prices.
However, examining prices in this way would not show the effects of
differing levels of pricing flexibility, and, in particular, how phase II
flexibility has changed prices, which was the objective of our analysis.
Effects of Contracts on Prices Varied, but Are Generally the Same Under Phase I
Flexibility and Phase II Flexibility
In general, because many contracts provide for discounts off the list
price, effective prices for dedicated access under these contracts in
phase II areas will generally be higher than phase I areas because
price-flex list prices are, on average, higher than price-cap list prices.
The exceptions to this generality are those contracts with set prices for
circuit components, rather than discounts. Representatives of price-cap
incumbents, however, state that discounts from these contracts in phase I
areas will compensate for the increases in the price-flex list price, when
considering a customer's entire purchase of dedicated access or when
considering circuits that may cross MSA boundaries or are located in both
phase I and phase II MSAs. Our analysis confirmed that many contracts with
major incumbent carriers provide discounts that, along with CALLS Order
decreases to the price-cap list price, can eliminate any increases in the
price-flex list price that may have occurred as a result of phase II
pricing flexibility. However, for other contracts we could not determine
the effect on net prices, because key data (e.g., the length of dedicated
interoffice mileage) were unavailable.
Conditions and Terms May Inhibit Switching Circuits to Competitors
Customers who sign contracts may need to meet various conditions, which
competitors argue limits customers' ability to choose another provider.
These conditions include such things as revenue guarantees, requirements
for shifting business away from competitors, and severe termination
penalties. Table 4 shows examples of contracts with such conditions and
terms. In revenue guarantee contracts, the customer guarantees that it
will spend a certain amount with the incumbent (e.g., $301 million per
year), and, in some contracts, that amount will increase over the course
of the contract. These types of contracts may inhibit choosing competitive
alternatives because the customer does not receive the applicable
discount, credit, or incentive if the revenue targets are not met and
additional penalties may also apply. Unless a competitor can meet the
customer's entire demand, the customer has an incentive to stay with the
incumbent and to purchase additional circuits from the incumbent, rather
than switch to a competitor or purchase a portion of their demand from a
competitor--even if the competitor is less expensive. FCC has indicated
that if any party believes that these contract offerings are
discriminatory or unlawful in any way, they may file a complaint under
section 208 of the 1996 Act.^34 According to FCC, no such complaints have
been filed.
^3447 U.S.C. S 208.
Table 4: Examples of Contract Terms and Conditions
Type of terms and
conditions Specific contracts
Revenue guarantees Verizon contract number 1 (no longer available) was
a 3-year-term contract that required an annual
revenue commitment of $301 million in year 1, $346
million in year 2, and $386 million in year 3, with
discounts on the amount spent above these targets
but below a maximum. If the customer did not
achieve the revenue targets, no discount was
applied. If the customer spent more than the
maximum, the amount above the maximum was not
eligible for a discount.
BellSouth contract number 10 is a 2-year-term
contract with minimum revenue commitments of
$8,800,000 in year 1, and $10,100,000 in year 2,
with discounts on the amount spent above these
targets but below a maximum. If the customer does
not achieve revenue minimums, no discount is
applied, and the customer will not be allowed to
subscribe to another contract with revenue minimums
for a period extending 6 months beyond the term of
the contract. If the customer spends over the
maximum, the amount above the maximum is not
eligible for a discount.
Shifting business away Several AT&T contracts (Southwestern Bell contract
from competitors number 15, Ameritech & Pacific Bell contract number
20, and Southern New England Telephone contract
number 1) require that at least 4 percent of
services ordered from AT&T must be switched over
from a nonincumbent provider.
Termination penalties A variety of AT&T contracts contain severe
termination penalties. For example, if (SWBT)
contract number 3 is terminated at the end of year
3 of a 5-year contract, termination penalties are
50 percent of the remaining 2 years of recurring
charges, or approximately 100 percent of annual
billings. To provide comparable rates, a competitor
(1) would have to provide a 50 percent discount
over the next 2 years, just to match the
incumbent's offer, and (2) would need to provide a
higher discount to provide a lower rate.
Qwest contract number 06-009 is a 3-year-term
contract that requires a monthly commitment of at
least $16,935,000, but no more than $20,161,000. If
the contract is terminated, the customer is liable
for 50 percent of the minimum revenue commitment
for each remaining month of the contract term.
Verizon contract number 25 is a 2-year-term
contract that requires an annual commitment of at
least $162,500,000. If the contract is terminated,
the customer is liable for 50 percent of the
difference between the amount of revenue billed
when terminated and the minimum annual commitment
of $162,500,000. If the contract is terminated in
year 1, there is no penalty for year 2 of the
contract.
Source: GAO analysis of contracts filed with FCC.
Average Revenue Has Declined Over Time, but Is Generally Higher for Channel
Terminations under Phase II Flexibility Than under Phase I Flexibility
Not all customers use price-flex contracts, and detailed data on the
number of customers and circuits that are purchased at price-flex list
prices were not available to us. Therefore, to examine the net effect of
changes in list prices and the application of contract discounts, we
compared average revenue data for dedicated access services under phase I
and phase II pricing flexibility. Our analysis of average revenue data for
channel terminations and transport for both DS-1 and DS-3 shows that, in
general, average revenue has declined in nominal dollars.^35 However, the
decline in average revenue for channel terminations is larger in the 29
phase I areas for which we had data, compared with the 27 phase II areas.
Average revenue decreased about 17 percent for DS-1 and DS-3 channel
terminations in the phase I areas, and 6 percent for DS-1 and 5 percent
for DS-3 in the phase II areas. As of 2005, the data show that average
revenue in the phase II areas is about 4 percent higher for DS-1 channel
terminations, and 24 percent higher for DS-3 channel terminations,
compared with average revenue in the phase I areas. (See appendix II for
more information on this analysis and for results using (1) data adjusted
for inflation into 2005 dollars, using both the Bureau of Labor Statistics
Producer Price Index for Wired Telecommunications Carriers and the general
GDP price index, and (2) results weighted by the number of businesses in
the MSA. Regardless of weighting or the price index used, phase II average
revenue is consistently higher than phase I average revenue in 2005.)
Figure 3 shows the change in average revenue generated in 2005 for both
DS-1 and DS-3 channel terminations compared with the average revenue
generated from sales in the year just prior to when FCC granted pricing
flexibility. (We do not specify that year because it varied among
price-cap incumbents--i.e., 2000, 2001, or 2002.)
^35These averages mask the variation in average revenue generated by each
price-cap incumbent carrier and the variation across specific MSAs. In a
limited number of cases there was almost no difference between phase I and
phase II areas. Due to confidentiality concerns, we do not report data for
specific incumbents or MSAs and, therefore, rely on these aggregated
figures.
Figure 3: Change in Average Revenue for DS-1 and DS-3 Channel Terminations
in Phase I and Phase II Areas
Note: Data are in nominal dollars. The differences between phase I and
phase II revenue shown in 2005 are statistically significant at the 1
percent level or lower.
We also compared average revenue for channel terminations under phase II
flexibility with average revenue for channel terminations in areas
remaining under the price cap. We calculated the average price-cap revenue
from data submitted by price-cap incumbents in their annual tariff review
plans, where price-cap incumbents provide the FCC with detailed data on
the number of channel terminations sold under the various zone and term
prices available in their tariffs. Not all discounts available under price
caps were included in our calculation because we were not able to
determine how such discounts would be applied to only channel
terminations. Therefore, the average price-cap revenue is biased upward.
Comparing average revenue in the 27 phase II MSAs with average revenue in
price-cap areas governed by the same tariff, we find no statistical
difference on average. Therefore, on average, phase II flexibility does
not appear to have resulted in prices lower than are available under
price-cap regulation. These averages mask variation across price-cap
incumbents and across MSAs. For example, comparing average revenue in
price-cap areas with average revenue in phase II areas for DS-3 channel
terminations, two of the price-cap incumbents showed higher average
revenue in phase II areas and two showed lower.
Data on average revenue for dedicated transport that the incumbent firms
provided shows that average revenue per unit for dedicated transport has
declined over the same period. However, we were unable to compare revenue
across differing levels of deregulation for transport, because nearly all
MSAs with pricing flexibility are under phase II flexibility for dedicated
transport. In fact, only 13 of the 215 MSAs with pricing flexibility have
only phase I flexibility for transport, and all of the data provided to us
were for MSAs where the price-cap incumbent had received phase II pricing
flexibility for dedicated transport. Price-cap incumbent firms have argued
that the market for dedicated transport is a more competitive segment of
the dedicated access market. FCC's colocation triggers capture the extent
to which competitors have their own dedicated transport from incumbent
wire centers.
Data from Two Departments Show That Their Spending on Dedicated Access Services
Is Generally Made Through FTS2001
The federal government spends millions of dollars annually on dedicated
access services. Our analysis of spending by USDA and DOJ found that, of
the estimated $9 million they spent annually on dedicated access services
in fiscal year 2006, most of these services were acquired through the
FTS2001 program.^36 Agencies purchasing under FTS2001 can obtain services
from one or more contractors, who may offer different prices for similar
services. Prices for services under the FTS2001 contracts decreased
annually over the life of those contracts, and GSA and the telecom
managers we interviewed recognized these prices to be generally below
rates with similar conditions and terms in price-cap, phase I, and phase
II markets.^37 We found that the data on prices provided by the federal
agencies regarding purchases through FTS2001 generally matched dedicated
access price estimates found in FTS2001. However, government agencies are
not required to use the FTS2001 program, and agencies may procure
dedicated access connections for high-capacity telecommunications services
directly from telecommunications carriers. Agencies may use their own
telecommunications contracts to procure dedicated access connections.
^36The FTS2001 program is the successor to the FTS2000 program. The former
program represented an improvement over its predecessor in terms of
available services and technology. FTS2001 provides voice, data, video,
Internet Protocol, and managed network services to federal agencies
nationally and internationally. Under FTS2001, GSA awarded an FTS2001 long
distance services contract to Sprint in December 1998 and another to MCI
WorldCom in January 1999. Under the terms of those contracts, each
contractor is guaranteed minimum revenue of $750 million over the life of
the contracts, which run for 4 base years and have four 1-year options.
Another GSA-managed telecommunications contract program is the
Metropolitan Area Acquisition, which provides local telecommunications
services in selected metropolitan areas. Under this program, certain
identified contractors as well as FTS2001 contractors are allowed to offer
services in both local and long-distance markets, a process termed
"crossover." For additional information, see GAO, FTS2001: Transition
Challenges Jeopardize Program Goals, [36]GAO-01-289 (Washington, D.C.:
Mar. 30, 2001).
USDA purchased all of its dedicated access services through FTS2001 to
connect the offices of its various agencies, such as the Forest Service
and the Agricultural Research Service. Our review of spending on dedicated
access in DOJ was limited to that contracted by DOJ's Justice Management
Division. It thus excluded any spending done by the Federal Bureau of
Investigation. See appendix I for additional information on the federal
agencies included in this review.
^37A direct comparison between FTS prices and commercial tariff and
contract prices cannot be made due to the difficulty in acquiring mileage
data as well as identifying applicable contracts. The government requires
carriers to provide it with the lowest applicable rates.
Our review of government expenditures on dedicated access services
indicated instances in which the government was not paying the lowest
price for these services in two ways.
o First, even if the procurement is made through FTS2001, an
agency may not necessarily use the lowest cost service provider
for a particular point-to-point circuit. Given a particular
dedicated circuit, FTS contractors each have prices that may
differ substantially. Agencies may not use the lowest cost service
provider because it may mean that they would have to switch
providers, and there may be significant costs involved in
switching.
o Second, because rates available through FTS2001 were usually
less than the rates government agencies received when purchasing
directly through a contract other than FTS, spending on dedicated
access services may be higher. However, these other contracts may
provide the agency with additional services, and thus we were
unable to fairly compare the different rates.
With the FTS2001 contracts expiring and a new contract vehicle,
Networx, currently being defined and negotiated, it is unclear
what prices the government will pay for dedicated access services.
In many instances list prices, as previously discussed in this
report, have increased since FTS2001 was initially negotiated.
However, because of newer technologies (e.g., metro Ethernet and
MPLS) and bundling of services, such increases in one service may
be offset by overall gains in efficiency and lower prices for
other services.
FCC Plays an Important Role in Ensuring Competition, but Lacks
Sufficient Information to Determine the Success of Its
Deregulatory Policies
FCC uses various data to assess competition for dedicated access
services, but most of these data have significant limitations in
their ability to describe the presence, extent, or change in
competition. FCC's strategic plan and various rulemakings have
defined FCC's obligation to ensure and assess competition for
dedicated access services. The agency attempts to collect data
from external parties through its rulemaking proceedings, but
those parties have no obligation to provide data, and the agency
has limited mechanisms to verify the reliability or accuracy of
any data submitted. FCC has stated that (1) gathering and
analyzing additional data would be costly and burdensome and (2)
that it is reluctant to impose additional reporting requirements
on incumbent firms or to require competitive firms to report
information on their networks. The agency must balance the
additional costs of gathering more data with the potential benefit
that might result from these additional data. Yet without more
complete and reliable data, FCC is unable to determine whether its
deregulatory policies are achieving their goals. FCC contends that
its open proceeding on dedicated access will address what steps
the agency should take to ensure that rates for dedicated access
services remain just and reasonable.
Ensuring Competition Is a Central FCC Responsibility
The promotion of competition is one of the two policy objectives
of the 1996 Act. The stated outcomes of this policy objective are
to lower prices and increase the quality of telecommunications
services available to American telecommunications consumers as
well as promote the rapid deployment of new telecommunications
technologies. FCC is the federal agency charged with executing and
enforcing the provisions of the 1996 Act.
In support of the goals of the 1996 Act, FCC implemented its
2006-2011 Strategic Plan, which defines the promotion of
competition as one of the agency's goals.^38 To support this goal,
FCC's current strategic plan says that it will collect and
evaluate information on competition in the domestic and
international communications markets. Additionally, FCC stated
that it will continually review FCC rules to determine what rules
need to be implemented, revised, or eliminated to achieve its
competition objectives effectively and efficiently. As part of its
review of the progress the agency has made toward meeting its
strategic plan, FCC determined that it was making adequate
progress toward ensuring that American consumers can choose among
multiple reliable and affordable means of communications.
In January 2005, based on a petition filed in 2002, FCC opened a
rulemaking proceeding in which it stated its commitment to
periodically examine its deregulatory judgments regarding
competition for dedicated access services.^39 FCC further affirmed
that its review of deregulation for dedicated access services is
consistent with its ongoing commitment to ensure that its
predictive judgments about competition for dedicated access are
consistent with actual marketplace developments.
FCC Uses Various Data to Assess Competition for Dedicated Access
Services
FCC uses data provided in incumbents' pricing flexibility
applications to determine the extent of competition in dedicated
access. As part of these applications, incumbents provide FCC with
data on competitors colocated in their wire centers. FCC has
determined that the collection of these data does not pose a high
administrative burden on applicants and allows FCC to assess the
state of competition in dedicated access. To minimize disputes
about these data, FCC requires the incumbent to notify colocated
competitors that the incumbent has listed their presence in their
application for pricing flexibility. Incumbents provide these data
to FCC in a confidential format that protects the exact names and
locations of these competitors.
A second major source of data that FCC uses to gauge competition
in the markets for dedicated access services comes from its
Automated Reporting Management Information System (ARMIS). FCC
initiated ARMIS in 1987 to collect and report on incumbent
financial and operational data. Incumbents annually report this
information to FCC. ARMIS data include general rates of return as
well as specific revenue figures and line counts for last mile
connections per incumbent. ARMIS data are publicly available
through FCC's Web site.
In other markets, FCC has identified some data needed to assess
competition. For example, in its 2005 Performance and
Accountability Report, FCC identified the following measures to
assess competition for various telecommunications services:
o Number of consumers having a choice among wireless and wireline
service providers.
o Percentage of households with competing providers for
multichannel video programming and information services.
o Relative prices for wireless and wireline services.
o Price for international calls.
However, these measures relate to competition for residential
customers, not business customers.
Finally, in various rulemaking proceedings, FCC has requested that
outside parties provide data regarding the state of competition in
dedicated access services. The information filed by these parties
is generally available for inspection and comment by the public.
FCC takes into account the data provided by outside parties as
part of its rulemaking proceedings. In its rulemaking proceeding
on competition in dedicated access services, FCC received a
variety of comments and data from incumbents, competitors,
cellular telephone companies, and large business users.
Data Available to FCC Are Not Current, Specific or Reliable
The data that FCC uses to assess competition for dedicated access
services have several limitations that prevent the agency from
describing the current state of competition--that is, these data
are not current, specific, or reliable. FCC has stated that
gathering and analyzing additional data would be costly and
burdensome. Furthermore, FCC has not traditionally required
competitive firms to report data, and the agency is reluctant to
impose further reporting requirements on incumbent firms, which
would be subject to OMB approval and the Paperwork Reduction
Act.^40
First, the data are not current. The data that FCC receives from
incumbents when they apply for pricing flexibility represent a
one-time assessment of the state of competition for dedicated
access services. Once it grants pricing flexibility, FCC does not
review the state of competition in dedicated access for those
incumbent markets. Because many pricing flexibility applications
were granted in 2001 and 2002, FCC has not reviewed the state of
competition in 4 to 5 years in markets, such as Atlanta, Los
Angeles, Phoenix, and Pittsburgh, where pricing flexibility has
been granted. Additionally, FCC has no mechanisms in place in its
rules to review competition. As previously shown, the amount of
competition has changed between 2001 and 2006 in some markets,
according to our analysis that used FCC's means of measuring
competition--colocation of incumbents and competitors in wire
centers. FCC also collects marketplace data from rulemaking
proceedings on an irregular basis, and the agency generally does
not establish a fixed timeline to resolve rulemaking
proceedings.^41
Second, data provided to FCC in rulemaking proceedings are often
not specific enough to be useful. Outside parties are under no
obligation to provide data FCC requests in rulemaking proceedings,
and these parties often do not provide requested data. For
example, in its rulemaking on competition for dedicated access
services, FCC requested data to create an average price index for
MSAs with pricing flexibility and MSAs under price-cap regulation
to determine how prices have changed. However, no companies filed
such indices. Instead, the companies provided aggregate figures of
average revenue. These average revenue figures were not
disaggregated to enable comparisons of price trends under
different levels of deregulation.
As with the major incumbent providers, FCC also has limited data
on competitors' provision of dedicated access services. For
example, ARMIS only requires certain price-cap incumbents to file
information. Competitors are not required to file any financial or
operational data through this system. In addition, competitors may
file tariffs for their dedicated access service offerings, but
they are not obligated to do so. As a result, FCC has no specific
or current data on competitors' prices for dedicated access
services or on the extent to which competitors have extended their
networks. FCC has noted that requiring competitors to disclose
that information could disturb the market by providing information
that would not otherwise be publicly available. However, some
competitors we interviewed stated that they have information on
where other competitors are because they use other competitors
wherever possible as an alternative to using price-cap incumbents.
Third, FCC's data also have limited reliability for assessing
competition in dedicated access. Outside parties are often
economically interested parties that have an incentive to provide
incomplete or biased data. For example, most of the outside
parties in the rulemaking proceeding on competition in dedicated
access are parties that would directly profit from further
regulation or deregulation. FCC is limited in its ability to
assess the reliability of these parties' information. Instead, it
relies on other parties to challenge or affirm these data's
reliability. Additionally, parties involved in the rulemaking on
dedicated access have raised concerns over the reliability of
ARMIS data to make assessments of competition in dedicated access
services. While competitors have used ARMIS data to show that
price-cap incumbents are earning large rates of return on
dedicated access, price-cap incumbents state that ARMIS cannot be
used to make competitive assessments due to outdated accounting
rules, including such things as arbitrary cost allocations and the
inclusion of certain revenues but not the corresponding costs. The
question of ARMIS data utility is part of the open proceeding on
dedicated access.
As previously noted, FCC's Performance and Accountability Report
recognizes the need for data to assess competition, but the
available data lack metrics for competition in dedicated access
for businesses. The Government Performance and Results Act of 1993
outlines criteria for agencies to define and measure their
progress in relation to the agencies' performance goals.^42 These
criteria state that an agency should identify performance measures
that adequately indicate progress toward its performance goals. In
our review of FCC's Performance and Accountability Report, we
found no data or measures of competition and choice that were
relevant to the business market or to dedicated access services,
although increasing consumers' and businesses' choice in
telecommunications is a stated goal of the agency. FCC's data
focus instead on consumers' access to residential wireless
providers or video programming (such as cable or satellite TV), or
on the cost of an international telephone call. However, FCC
stated that there is no easily identifiable and understandable
measure for competition in the business market or in dedicated
access. Furthermore, more competition has traditionally existed in
the business sector than in the residential sector, and,
therefore, FCC has focused on metrics in the residential sector.
Conclusions
The market for dedicated access services is complex and
multidimensional, including a wholesale market for dedicated
access facilities as well as a retail market that relies on
dedicated access facilities to provide high-capacity
telecommunications services for large business customers with
multiple locations. The wholesale market has historically been
controlled by the incumbent firms, who have virtually ubiquitous
networks within their regions. Competitors have entered segments
of this market with their own networks, encouraged by FCC's
actions dating back many years, and are active participants in the
retail market, reselling incumbent dedicated access services to
provide business services, or relying on their own or other
competitors' local connections, where they exist.
FCC has initiated several deregulatory actions and access charge
reforms in an effort to fulfill the intent of the 1996 Act and
allow market forces and competition to govern prices for dedicated
access. At the heart of FCC's actions was a vision of
facilities-based competition, where competitors would compete with
the incumbents mainly using their own networks and facilities.
Under facilities-based competition, incumbents would be
constrained from pursuing predatory and exclusionary pricing
practices, and prices would be driven toward marginal costs. FCC's
deregulatory actions were predicated on proxy measures that FCC
predicted would indicate whether sufficient facilities-based
competition existed for dedicated access services in order for
market forces to function in this way. However, our analysis of
facilities-based competition suggests that FCC's predictive
judgment -- that MSAs with pricing flexibility have sufficient
competition -- may not have been borne out, particularly for
channel terminations to the end users of dedicated access. Even
more troublesome is the fact that some of our analysis, which is
based on FCC's competition metrics, suggests that competitive
alternatives for dedicated access have declined in some MSAs in
the past few years. The effect that such changes may be having on
consumers of all sizes, including the federal government, could be
significant.
Taking a broader view of the competitive landscape, our analysis
suggests that wireline facilities-based competition itself may not
be a realistic goal for some segments of the market for dedicated
access. Long-standing entry barriers continue to exist and are not
likely to be alleviated. Where demand for dedicated access is less
than 3 or 4 DS-1's, it would appear unlikely that any competitor
would extend its network for that business. While competitors may
be able to serve such lower demand customers using UNEs in the
hopes that demand might increase to such a level that makes build
out a real possibility, FCC has recognized that the availability
of lower-priced UNEs has discouraged investment by competitors (as
well as incumbents). Furthermore, the FCC has recently limited the
availability of UNEs. New technologies, such as WiMax, also have
the potential to bring more competition. However, it is unclear
the extent to which this technology can provide a widespread
alternative to wireline dedicated access, how long that transition
will take to become an effective alternative, or who will be in
the best position to provide that alternative. Concurrently,
price-cap incumbents have received a significant amount of price
deregulation allowing them to negotiate price-flex contracts and
to raise their list prices. However, competitors argue that
price-flex contracts, in addition to other entry barriers,
discourage the use of competitive networks, and thus discourage
investment by competitive firms.
In its ongoing rulemaking proceeding on dedicated access, FCC
recognized its responsibility to revisit its proxy measures to
determine whether its predictive judgments comport with actual
market developments and to fulfill its mission to encourage
competition and to ensure lower prices, higher quality services,
and adequate choices for consumers. Much of the specific data and
information that FCC collects and has requested from incumbents,
competitors, and dedicated access customers that would enable the
agency to effectively analyze trends in competition and the
effects of deregulation were not provided by these parties, and
the information that has been provided is of limited reliability,
and has come from parties that would directly profit from further
deregulation or regulation. Even with the data that has been
provided, FCC's rulemaking proceeding, which began with a petition
filed in 2002, is still unresolved.
Regardless of where competition may come from in the future, it is
clear that FCC does not regularly monitor and measure the
development of competition, which will affect how FCC responds to
emerging trends, and the actions it takes to encourage and foster
such competition. We have consistently noted the need for better
data at FCC to track competition and deployment of
telecommunications services to a variety of consumers.^43 Without
data that are reliable, relevant, and current, FCC is limited in
its ability to adequately monitor the state of competition for
dedicated access, and thus is limited in its ability to determine
whether its predictive judgments were correct, and whether its
deregulatory actions are achieving their goals.
Recommendations for Executive Action
To more effectively monitor and determine whether its deregulatory
actions are achieving their goals of encouraging competition, and
ensuring lower prices and adequate consumer choice, FCC should
take the following two actions:
o Develop a meaningful and workable definition of effective
competition, or true customer choice, using an approach that
evaluates the competitive nature of a market by accounting for the
number of effective competitive choices available to customers.
o Consider collecting additional data and developing additional
measures to monitor competition on an ongoing basis that more
accurately represents market developments and individual customer
choice (e.g., price indices and the extent of competitors'
networks).
If, through this monitoring, FCC finds that competition is not
developing as it expected, it should determine what actions are
necessary to accelerate competition for dedicated access.
Agency Comments
We obtained comments on a draft of this report from FCC officials,
which are presented in appendix III. In summary, FCC stated that
our report "appears to imply the need for a return to price
control policies" and thus generally disagreed with the report's
recommendations. Consistent with that interpretation of the
report's overall message, FCC notes that "the cost of price
regulation to carriers and the public is still greater than the
benefits." FCC's comments assert that it "takes seriously its
obligation to foster competition ... and will use all available
data" to do so, but also indicates that gathering reliable data
and analyzing actual competition in communications markets would
be difficult. FCC adds that in 2001, a federal court agreed with
its theoretical approach to gauging competition based on proxy
measures.^44 Regarding our recommendation to develop a meaningful
and workable definition of effective competition, FCC states that
there is no universally accepted, bright-line definition of
"effective competition," and that any definition that suggests a
geographic market more granular than its existing measures would
be administratively infeasible to implement and may not be
consistent with the deregulatory goals of the 1996 Act. Regarding
our recommendation to consider collecting additional data and
developing additional measures to monitor competition, FCC states
that they continue to monitor competition in dedicated access, and
suggested that the detailed data FCC requested in its ongoing
rulemaking proceeding on dedicated access will allow it to
evaluate competitive entry in these markets.
Contrary to FCC's interpretation, the report does not call for the
reregulation of dedicated access prices, nor do we intend to imply
that broad reregulation of prices is either necessary or the
appropriate response to the evidence of less competition and
higher prices in deregulated markets. The market for dedicated
access services is estimated to be worth $16 billion annually. The
data developed in this report indicates that there are fewer
competitive alternatives and that prices for dedicated access in
the theoretically more competitive phase II markets are higher on
average than prices in phase I markets, and not statistically
different than prices in price-cap markets. The report also
demonstrates that FCC does not have the type of meaningful data
that would allow it to effectively oversee the extent of
competition in the market. Thus, the report calls for FCC, serving
in its capacity as the federal regulator of interstate
communications services, to better define effective competition
and then collect meaningful data on the state of competition in
the marketplace. Only by doing so can FCC measure its progress
toward its stated goals to encourage competition and secure lower
prices and higher quality services for telecommunications
consumers, or adjust its approach toward that goal, as needed.
We recognize that the United States Court of Appeals for the D.C.
Circuit held that FCC made a reasonable policy determination
regarding its proxy measures of competition, and that regulation
could impose costs that outweigh benefits.^45 However, given the
changes in the market and the questions raised by our analysis, we
believe FCC should not be static and should seek more discrete
measures that are necessary in an evolving environment. Indeed,
although FCC's existing rules on pricing flexibility were adopted
based on predictive judgment and its ongoing rulemaking on
dedicated access is intended to examine whether available data
support maintaining, modifying, or repealing those rules, FCC's
comments now suggest a preference for economic theory rather than
empirical data. The data used in our analysis was obtained in a
manner that prohibited our sharing it; FCC could obtain those data
and analyses contractually in the same way that we did. The data
developed in this report, at a minimum, raise questions about
FCC's assertion that higher prices will induce competitive entry.
For example, although FCC's comments note that high prices will
induce competitive entry, the data developed in this report
suggest otherwise. There appear to be fewer competitors in areas
where prices are higher. Moreover, economic theory generally holds
that competitive entry would occur if markets are "contestable."
FCC itself recognizes that the substantial sunk costs required to
compete in these markets may serve as a barrier to entry.
We agree with FCC that there is no universally accepted,
bright-line definition of "effective competition." However, we
believe that FCC is in the best position to develop meaningful and
workable definitions despite any difficulties associated with such
a task. Further, we maintain that this is a relevant and important
task for the requisite federal regulatory body. Furthermore, we
maintain that FCC would be significantly hindered in its ability
to fulfill its regulatory responsibilities and statutory goals of
promoting competition if it cannot define competition, does not
have measurable goals, and does not collect and analyze reliable
data on the state of competition for dedicated access.
Regarding our second recommendation that FCC consider additional
data and measures, we disagree with FCC's assertion that it
continues to monitor competition and that requesting detailed data
in the rulemaking proceeding on dedicated access is sufficient to
allow FCC to evaluate pricing behavior. FCC's strategic plan and
performance budget contain no measures by which the Congress or
the American public can ascertain the extent to which its
deregulatory polices are encouraging competition in the business
market or in the provision of dedicated access services.
Furthermore, while FCC requested information in its rulemaking
proceeding--such as price indices pertaining to services sold in
phase I and phase II areas and cost studies--it is our
understanding, based on review of the submissions of major
carriers, that such detailed data were not supplied by the parties
to the proceeding. Instead, FCC received data that were either
incomplete or in a more aggregated form that can obscure the
effect of phase II pricing flexibility. Thus, we disagree with
FCC's position that the data gathered from the rulemaking is
adequate to monitor competition and that additional data
collection is not needed. We support the FCC's rulemaking and
believe that data collection is a critical factor in the
proceeding.
We also provided the major incumbent carriers an opportunity to
review a draft of the report, as well as representatives of the
trade association representing competing carriers and a group
representing a coalition of major users of telecom services. Three
of the four major incumbent firms provided comments on a draft of
the report. Generally, these firms (AT&T, BellSouth, and Verizon)
took issue with the report's underlying data and the conclusions
we drew from those data, as follows:
o Concerning the extent of facilities-based competition in the
market for dedicated access services, the incumbent firms asserted
that data we used were incomplete. They asserted that competing
firms often did not supply information on the buildings served as
a competitive alternative. Further, the incumbents asserted that
even if individual buildings were not served by a competitive
alternative, the proximity of those firms' fiber networks (or the
presence of wireless alternatives and cable providers) could
provide a competitive check. Additionally, the incumbents
disagreed with the draft report's characterization of the
dedicated access product market as being defined by demand for a
DS-1 or greater level of service, commenting that we should have
examined competition at DS-3 and greater levels of demand.
Incumbent firms also asserted that DOJ and FCC have found the
markets for dedicated access to be competitive.
o Concerning our analysis of changes in prices, the incumbents
argued that the analysis was unreliable because the data we used
were incomplete. More specifically, they insisted that prices paid
by customers could not be analyzed by focusing on channel
terminations in particular areas because prices need to be
examined on the basis of the total circuit (i.e., including the
mileage portion), which may cross multiple geographic boundaries,
and that the data show that prices have decreased over time.
We recognize the limits of available data on the extent and effect
of competition in the market for dedicated access services. It is
highly unlikely that any data set on telecommunications networks
would be perfect. DOJ has noted that even with the ability to
obtain data through subpoena power, its analysis also likely
experienced some errors due to underreporting and overreporting.
Nevertheless, we believe the data used provide a reasonably and
sufficiently reliable picture of the extent of facilities-based
competition at the building level. The report acknowledges the
potential for underreporting and overreporting of competitors'
equipment and notes the extent of data errors using data supplied
by two large competitors. As noted, the database we relied upon
has been used by at least two incumbents in petitions before state
public utility commissions. We disagree that we are not accounting
for the competitive presence of cable and wireless providers. The
database we used shows the presence of these competitive
alternatives, and our analysis does not exclude such competitive
alternatives. And although a competing firm may have fiber
relatively nearby, that does not mean that competitive entry into
a location is necessarily likely. DOJ has recognized in its
Competitive Impact Statements that "such entry is a difficult,
time-consuming, and expensive process." We took note of the
incumbents' objections to our definition of the product market and
incorporated additional analyses of the extent of competition in
the market for higher levels of demand (e.g., DS-3 and higher). We
disagree that DOJ believes these markets to be competitive.
Although DOJ cleared the proposed mergers, it required Verizon and
SBC to divest portions of certain local fiber-optic network
facilities to proceed with their respective acquisitions.
Moreover, DOJ clearly noted in its Competitive Impact Statements
on the mergers that "[f]or the vast majority of commercial
buildings in [their] territory" the incumbents are the only
provider of dedicated access. We note also that FCC approved the
mergers with conditions.
On the concerns raised by the incumbents on the draft's analysis
of changes in pricing, it is true that we did not have complete
information on pricing; neither incumbents nor competitors were
able to provide that information, which is usually restricted
contractually by non-disclosure agreements. It is also true that
we focused our analysis on channel terminations and not on the
price of a total circuit, including transport, or on a customer's
entire purchase. Our objective was to examine the effect of phase
II pricing flexibility, which is the only circumstance under which
price increases can occur. By disaggregating the data on the basis
of how pricing flexibility was granted, we were able to examine
price trends under different levels of pricing flexibility. We
were unable to examine trends in transport prices under different
levels of pricing flexibility because the vast majority of MSAs
have received phase II pricing flexibility for dedicated transport
(and therefore very few data points with which to compare) and
because other key data elements associated with transport (e.g.,
varying mileage) was unavailable. Analysis based on more
aggregated data, such as suggested by the incumbents, obscures the
effect of phase II pricing flexibility by including prices that
are based on base rates resulting from the CALLS Order, which were
automatically decreasing until 2003. We took note of the
incumbents' issue that we did not compare average revenue under
pricing flexibility with average revenue under price caps, and
incorporated additional analyses comparing price-cap average
revenue to price-flex average revenue. The draft makes clear that
average revenue for both channel terminations and dedicated
transport have declined over time. At the same time, however, it
is also clear from the data provided by the incumbents themselves
that they generate higher levels of average revenue from sales of
channel terminations in the theoretically more competitive phase
II areas--a finding that is incongruous with greater levels of
competition.
Finally, we also provided GSA, USDA, and DOJ the opportunity to
comment on segments of the report that pertain to the data and
information they provided. GSA, USDA, and DOJ verified the key
facts we obtained from them, and provided technical clarifications
which we incorporated where appropriate.
We are sending copies of this report to the appropriate
congressional committees and to appropriate officials of the FCC,
GSA, USDA, and DOJ. We will also make copies available to others
upon request. In addition, the report will be available at no
charge on the GAO Web site at http://www.gao.gov.
If you or your staff have any questions about this report, please
contact me at (202) 512-2834 or at [email protected]. Contact points
for our Offices of Congressional Relations and Public Affairs may
be found on the last page of this report. GAO staff who made major
contributions to this report are listed in appendix IV.
Sincerely yours,
JayEtta Z. Hecker
Director, Physical Infrastructure
^38Federal Communications Commission, Strategic Plan 2006-2011 (2005), 8.
^39Special Access Rates for Price Cap Local Exchange Carriers, WC Docket
No. 05-25, Order and Notice of Proposed Rulemaking, 20 FCC Rcd 1994
(2005).
^40See footnote 20.
^41Generally, comments are due 60 days after a Notice of Proposed
Rulemaking is published in the Federal Register and reply comments are due
90 days after publication in the Federal Register. FCC extended the reply
comment period to on or before July 29, 2005.
^42Pub. L. No. 103-62.
^43GAO, Telecommunications: Broadband Deployment Is Extensive Throughout
the United States, but It Is Difficult to Assess the Extent of Deployment
Gaps in Rural Areas, [37]GAO-06-426 (Washington, D.C.: Mar. 2006); and
Challenges to Assessing and Improving Telecommunications For Native
Americans on Tribal Lands, [38]GAO-06-189 (Washington, D.C.: Jan. 2006).
^44Worldcom, Inc. v. FCC, 238 F.3d 449 (D.C. Cir. 2001).
^45Worldcom, Inc. v. FCC, 238 F.3d 449 (D.C. Cir. 2001).
Appendix I: Objectives, Scope and Methodology
This report examines the state of competition within the markets
for dedicated access services by addressing three issues: (1) the
extent to which competitive alternatives to the major incumbent
telecommunication carriers are available; (2) pricing for
dedicated access services in fully deregulated markets versus
regulated markets, as well as prices the government is paying for
dedicated access; and (3) the data the Federal Communications
Commission (FCC) uses to measure competition in dedicated access
and the limitations, if any, that may exist in such efforts.
Extent of Competitive Alternatives to Major Incumbent Firms
To determine the extent that competitive alternatives exist for
dedicated access, we analyzed the extent to which competitive
telecommunications providers provide dedicated access service to
end user buildings in 16 metropolitan statistical areas (MSA). We
selected those 16 MSAs, divided evenly between phase I and phase
II deregulated markets (for channel terminations) and evenly among
the major incumbent firms (AT&T Corporation, BellSouth
Corporation, Qwest Communications, and Verizon Communications).
Table 5 summarizes the selected MSAs. Our sample of MSAs is
intended to illustrate the extent that competition has entered the
market for dedicated access services only; the results are not
generalizable to all MSAs in the United States. We are not making
a judgment on the legal sufficiency of competition in dedicated
access services including, whether recent mergers violate
antitrust laws or whether proposed remedies that the Department of
Justice (DOJ) identified would be sufficient to eliminate the
competitive harm of the mergers.
Table 5: MSAs in Analysis, by Price-Cap Incumbent and Applicable
Pricing Flexibility
Phase II for channel
Price cap Phase I for channel terminations, terminations and dedicated
incumbent phase II for dedicated transport transport
AT&T Detroit, Chicago Los Angeles, San Jose
BellSouth Greenville, New Orleans Atlanta, Miami
Qwest Seattle, Minneapolis Phoenix, Portland
Verizon New York, Washington, D.C. Norfolk, Pittsburgh
Source: GAO analysis of FCC pricing flexibility report and orders.
Data on the presence of relevant telecommunications equipment in
businesses throughout the United States is not independently
available from public sources. To conduct our analysis, we
contracted with two firms: Telcordia Technologies, Inc., and
GeoResults.
We contracted with Telcordia, a leading global provider of
telecommunications network software and services, to obtain an
extract from the Location Registry (formerly, CLONES), which is a
hosted database of network locations and related network functions
for the telecommunications industry. A given location and network
function for an incumbent or competitive firm can be identified
uniquely using a CLLI(TM) Code. The CLLI Code is an alphanumeric
code and key into the Location Registry, providing additional
information such as physical address and coordinates. The CLLI
Code can be used to make inferences about network equipment. It is
not an equipment identifier. Both incumbent and competitive firms
can subscribe to the COMMON LANGUAGE(R) Location Information
Service from Telcordia to gain access to the Location Registry,
enabling the voluntary entry of their information into the
registry. Incumbent and competitive firms are responsible for
maintaining the integrity of their records and providing any data
reconciliation regarding records that may be incorrect or
incomplete. The Location Registry provides subscribing firms with
a standardized method for identifying network locations and
related network functions. The Location Information Service, in
conjunction with the Connection Information Service, provides a
method for standardizing orders for network interconnection and
network transport between the different firms.
We assessed the reliability of the Telcordia Location Registry and
determined that the information was sufficiently reliable for our
purposes. According to Telcordia, the information in the registry
may be less comprehensive for competitive firms than for incumbent
firms because some smaller competitive firms do not subscribe to
the service, and there may be some underreporting of competitors'
locations due to competitive concerns. However, Telcordia is
unable to estimate the extent to which competitors' data are
underreported. To gauge the extent that the data are understated,
we compared entries in the database with lists of "lit" buildings
provided to us by two of the largest competitive firms. One firm
showed 465 lit buildings in the data they provided to us in the 16
MSAs we examined, of which 436 showed the presence of a "lit"
competitor, suggesting an underreporting error of a little over 6
percent. However, the database also showed this same competitor as
being the sole competitive presence in 81 additional buildings
that were not on the firm's list of lit buildings, suggesting,
that, for this competitor, the database is overreporting the level
of competition. However, the other firm from which we obtained
data provided us a list with 693 lit buildings in the MSAs we
examined, of which 289 showed the presence of a "lit" competitor,
indicating underreporting of about 400 buildings across the MSAs
for this competitor. These two examples show that individual
competitor's presence may be underreported and overreported.
The data in the location registry may also overstate the presence
of competitors for other reasons. Bankrupt and merged companies,
such as the former AT&T and MCI, still have entries in the
registry, although that equipment may not be in use or may now be
equipment owned by the incumbent firm. Furthermore, the registry
may also incorporate some equipment from active competitors that
is not currently providing service or is no longer in service, as
shown above. For instance, there may be equipment in a vacant
building where a competitor used to provide service, but the
competitor had not removed the equipment from the registry.
For the purposes of analyzing the presence of competitors at
incumbent wire centers, the registry is likely more accurate.
Because the registry is primarily used for interconnection
purposes, and because wire centers are locations where aggregation
of traffic and interconnections take place, underreporting of
competitors' presence in wire centers is unlikely. There is no
single public or private data source universally recognized as
comprehensive. This is because there is no compulsory process
through which telecommunications companies report such data to FCC
or private data sources. This database is the most comprehensive
available to us, and price-cap incumbent firms, such as BellSouth
and AT&T, have used the database for similar purposes. While we
recognize that there is both underreporting and overreporting in
the database, it would appear that these two errors offset one
another to some degree. Therefore, while our analysis does not
provide pinpoint accuracy regarding the state of facilities-based
competition, we determined that the database was sufficiently
reliable to illustrate the general level of competitive build out
to end-user locations.
To analyze the extract of the Location Registry, we contracted
with GeoResults, which is a firm that the telecommunications
industry has used extensively to analyze Telcordia data.
GeoResults analyzes the CLLI Codes within the Location Registry to
make inferences about the presence of fiber-optic equipment within
a given building. GeoResults' analysis (known in the
telecommunications industry as its "GeoLit" report), provides us
with the necessary filtered data to indicate which end user
buildings have a "lit" presence from a competitor. Telcordia has
not validated methods or assumptions of any analysis performed by
GeoResults for accuracy or completeness. GeoResults' GeoLit
analysis is based on July 2006 extract from the Location Registry.
The firm provided GAO with analysis on the extent to which
competitors provide dedicated access using fiber and wireless
facilities at commercial buildings in those 16 MSAs. While copper,
fiber, and wireless facilities may be used for access, industry
participants told us that for practical purposes, competitive
local access providers extend their facilities primarily via fiber
and to a lesser extent via wireless, due to the higher revenue
capacity and the lower maintenance costs of fiber or wireless.
We also analyzed the presence of any type of telecommunications
equipment owned by competitors located in those buildings. This
analysis thus included data on telecommunications equipment that
GeoResults identified as non-fiber optic or could not be
positively identified as fiber optic. It also included data on
other equipment not used to provide service, such as testing
equipment. In general, this analysis found presence of any type of
equipment in buildings with greater than a DS-1 of demand ranging
from about 17 percent to 30 percent, excluding Norfolk. We believe
this analysis is not a valid measure of facilities-based
competition because the equipment included in the analysis
includes non-fiber optic equipment attached to non-fiber optic
dedicated access from price-cap incumbents--which may represent
leased lines--and testing equipment, both of which falsely
indicate a facilities-based competitive presence.
To analyze the extent to which business users are likely to
purchase dedicated access from incumbent firms or competitors, we
also contracted with GeoResults. GeoResults provided GAO with
their standard demand model that estimated the number of buildings
that might require dedicated access at three levels of demand, as
shown in table 6.
Table 6: GeoResults' Dedicated Access Demand Model
Level of demand Definition
At least DS-1 A commercial building with one or more business tenants
that have a dedicated access demand for one or more DS-1
circuits. An individual business that has a data
bandwidth demand of 512 Kb to 8 Mb will be defined as a
business that has a dedicated access demand for one or
more DS-1 circuits.
At least DS-3 A commercial building with a business tenant that has a
dedicated access demand for one DS-3 circuit. An
individual business that has a data bandwidth demand of
8 Mb to 16 Mb will be defined as a business that has a
dedicated access demand for one DS-3 circuit.
At least 2 DS-3's A commercial building with one or more business tenants
that have a dedicated access demand for two or more DS-3
circuits. An individual business that has a data
bandwidth demand of 16 Mb or more will be defined as a
business that has a dedicated access demand for two or
more DS-3 circuits.
GeoResults' model focuses on the number of employees per business;
the type of business (e.g., ones that are telecommunications
intensive versus ones that are not, such as bakeries); and the
"family size" of the business (i.e., the extent to which a
business was a branch office of a larger corporate parent). For
our analysis of facilities-based competition for buildings with at
least one DS-1 of demand, we included cellular phone sites, mobile
switching offices, "carrier hotels"--locations where several
competitors locate for interconnection purposes--and any other
locations where competitors had placed fiber-based equipment,
regardless of whether the model indicated any demand for dedicated
access. For our analysis of competition in locations with a
greater level of demand, we only examined those locations
GeoResults identified as having a DS-3 level of demand or a level
of demand of 2 DS-3s or higher.
GeoResults obtained data on businesses from Experian's National
Business Database. Experian is a national company that provides,
among other products, information regarding businesses in the
United States. We assessed the reliability of Experian's National
Business Database and reviewed Experian's quality procedures that
it uses to verify the information contained within its National
Business Database and found it sufficiently reliable for our
purposes. No available database is 100 percent inclusive of all
commercial buildings. GeoResults uses records for some 15 million
commercial buildings in its demand model. GeoResults estimates
that this data covers about 70-75 percent of the total number of
commercial buildings. GeoResults officials told us that this
demand model is widely used by a variety of incumbent and
competitive firms as well. According to GeoResults, these firms
use their demand model to identify target dedicated access
customers within various buildings throughout MSAs in the United
States.
Analysis of Available Information on Dedicated Access Pricing
To describe how deregulation has affected available prices for
dedicated access services, we analyzed changes in list prices,
prices available under customized contracts, and average prices in
MSAs under phase I flexibility and phase II flexibility from the
period prior to the granting of pricing flexibility (generally,
2001 or 2002) to the present, or to the latest period for which
data were available. We limited our analysis to prices for
high-capacity dedicated access services at two speeds--1.544
megabytes per second (Mbps), which is known as a DS-1 circuit, and
45 Mbps, which is known as a DS-3 circuit--because they represent
the majority of dedicated access revenues. Where possible, we
compared prices for the two major components of dedicated access
services--channel terminations and dedicated transport. Although
dedicated access services can be ordered with multiple options and
configurations, such as value-added services geared toward
providing added network reliability, we focused our analysis on
DS-1 and DS-3 monthly recurring charges only, without any such
features or options.
Separately, we analyzed spending on dedicated access services by
selected federal departments and agencies. We analyzed available
data on prices that selected federal government departments under
General Services Administration (GSA) contracts paid, as well as
prices paid under separate agency contracts for services purchased
directly in the marketplace.
Change in List Prices
We analyzed listed prices for channel terminations and dedicated
transport for month-to-month, 3-year, and 5-year terms across
three density zones.^1 FCC requires incumbent firms to file list
prices in all areas that they serve. "Price-flex" list prices are
made generally available in areas with phase II flexibility.
"Price-cap" list prices are made generally available to all
customers in areas with phase I pricing flexibility as well as all
other areas in which FCC has granted neither phase I nor phase II
flexibility. We analyzed 1,152 elements of dedicated access
service across the four major price-cap incumbents' filings. We
compared both current price-flex prices with current price-cap
prices as well as current price-flex prices to prices in effect in
2001. We left these prices in nominal dollars and did not adjust
for inflation. These comparisons were made using like components
and parameters. For example, we compared the price for a 3-year
term, zone 1, channel termination in a phase II MSA in 2006
against the price for a 3-year term, zone 1, channel termination
in that same MSA in 2001, as well as against the price for a
3-year term, zone 1, channel termination in a phase I MSA in 2006.
Some price-cap incumbents also offered 1-year, 2-year, 4-year, or
7-year terms. We did not include these prices in our analysis,
because not all price-cap incumbents offered such terms. Including
these additional term prices would not change the overall results
of the analysis because term prices are generally determined by a
percentage discount off of the month-to-month list prices.
Customized Contracts
Because larger customers may purchase dedicated access through
various contracts with incumbents made allowable under phase I and
phase II deregulation, we analyzed a substantial number of these
contracts, which FCC requires incumbents to file. In reviewing
contracts, we generally compared net prices (i.e., after discount
and credits) under the contracts to the initial price prior to the
granting of pricing flexibility and examined the effect of
contract discounts on the price-flex list price and the price-cap
list price. However, because we do not know the details of how the
circuits purchased under these contracts are configured (i.e.,
some circuit components can be in phase I areas, others in phase
II, some circuits may traverse both phase I and phase II areas or
even price-cap areas), we could not determine the overall effect
of the contracts on customers' entire purchases.
In some cases, prices on some contracts did not vary on a per-mile
basis. Because list prices have a mileage component, we were not
able to compare many of the flat contract prices with prices
available prior to pricing flexibility. Additionally, some
contracts cover multiple regions. For example, several AT&T
contracts require subscription of concurrent dedicated access
services in specific MSAs in four regions, each of which has its
own list prices. Since base prices are not identical across AT&T's
regions, it is possible that under a multiregion contract, the
contract discounts result in lower prices in one region, but not
another. The data required to analyze the various factors that
contribute to an overall contractual price were not available. For
example, mileage data for individual contracts or specific
detailed data on the number of DS-1 and DS-3 circuits purchased
under each contract were not available.
Change in Average Revenue
Because we could not obtain specific data on the number of
customers purchasing dedicated access services at various pricing
levels (i.e., month-to-month, term, various zones, and various
contract options) and the exact amount purchased, we could not
test the effect of phase II pricing flexibility over time through
an analysis of list prices and contract discounts. Therefore, we
requested that incumbents provide us with data on their average
revenue per unit for channel terminations and dedicated transport
from the period just prior to the granting of pricing flexibility
and the most current period for which data were available across
MSAs with phase I flexibility for channel terminations and phase
II flexibility for channel terminations. We requested that the
incumbents provide us with data representing total average monthly
recurring charges, which would include any discounts or
termination penalties from price-flex contracts,^2 and exclude any
non-recurring charges associated with the initial purchase of the
services. The average revenue per unit effectively suggests the
average (arithmetic mean) price that customers paid for the
specific dedicated access components. As an average, the data
reflect the net effect of circuits purchased in different density
zones, and across different term lengths or volume arrangements.
The mean price is susceptible to effect from a few large customers
with heavily discounted prices. One major incumbent carrier told
us that 5 percent of its customers had contracts with customized
discounts, and those customers represented about 50 percent of the
firm's dedicated access business. To compensate for such an
effect, data on the median prices paid would also have been
useful, but were not available to us.
We analyzed the average amount of revenue in nominal dollars that
the major incumbent carriers reported from the sales of DS-1 and
DS-3 dedicated access channel terminations in 56 MSAs--27 MSAs
with phase II flexibility for channel terminations, and 29 MSAs
with phase I. We received data for 20 MSAs from AT&T and Verizon,
and data for 10 MSAs from BellSouth and Qwest, for a total of 60
MSAs. We excluded two MSAs from the data that Verizon provided and
two from AT&T's data because those MSAs were not under phase I or
phase II flexibility for channel terminations. We were unable to
independently verify the reliability of the data provided by the
price-cap incumbents. However, we performed some logic tests based
on listed prices and available discounts to determine if there
were any major inaccuracies. Due to confidentiality concerns, we
aggregated these averages across MSAs and across all four major
price-cap incumbents, which masks some variation across the firms,
as well as variation across MSAs, but still allows us to examine
overall trends in markets under different phases of deregulation.
We also calculated the average revenue in areas that remain under
full price-cap regulation from data submitted by price-cap
incumbents in their annual tariff review plans, where price-cap
incumbents provide the FCC with detailed data on the number of
specific circuit components sold under the various zone and term
prices available in their tariffs. We calculated an average
price-cap price for DS-1 and DS-3 channel terminations for tariffs
corresponding to the phase II MSAs for which we had average
revenue data. For example, the phase II MSAs in Verizon's
territory for which we received data corresponded to the area
covered by Verizon's FCC No. 1 tariff filing. The average
price-cap revenue is likely to be biased upward. Because areas
still under price-cap regulation have not qualified for phase I or
phase II flexibility, these areas are likely to have lower
business density. Therefore, a higher percentage of circuits are
likely to be sold under zone 3 pricing, which is generally priced
higher than circuits under zone 1 pricing. For example, in Qwest's
annual tariff review plan, 51 percent of DS-1 channel terminations
were sold under zone 3 pricing. Because we do not have detailed
data on the number of channel terminations sold under different
zones in phase II areas, we were unable to correct for this bias.
Furthermore, not all discounts available under price-cap
regulation (such as AT&T's Managed Value Plan) were included in
our calculation because we were not able to determine how such
discounts would be applied to only channel terminations. However,
despite these biases, we find that phase II average revenue for
the 27 MSAs, on average, is not statistically different than
price-cap average revenue.
We compared how the average revenue for channel terminations in
phase I and phase II areas had changed over time, from prior to
deregulation through 2005 (the latest full year for which data
were available), and we compared average revenue figures for 2005
across phase I areas, phase II areas, and areas remaining under
the price cap. We also performed our analysis using two different
price indexes and after weighting the data based on the relative
size of the MSAs to determine how sensitive our results were to
such effects. We used both the general GDP price index, as well as
the Bureau of Labor Statistics' Producer Price Index for Wired
Telecommunications Carriers to adjust the data to 2005 constant
dollars. To account for the relative difference in the size of the
MSAs, we weighted the data on the basis of the number of
businesses with 20 or more employees in each MSA. While an
imperfect weight, this was used as a rough estimate of the level
of demand in these MSAs. Regardless of the deflator used or
weighting the data, phase II average revenue was higher than phase
I average revenue. See appendix II for the detailed results of
this sensitivity analysis.
We also analyzed the changes in average revenue for dedicated
transport. Because all but one of the MSAs in the data provided to
us were areas where the price-cap incumbents had received phase II
flexibility for transport, we were unable to compare changes in
average revenue for transport under different levels of pricing
flexibility. In fact, of the 215 MSAs where pricing flexibility
has been granted, the four major price-cap incumbents have
received phase II flexibility for dedicated transport in 202 MSAs,
and phase I flexibility in only 13 MSAs.
We were unable to collect data on prices that competitive firms
charged; therefore, those prices are excluded from this analysis.
We asked competitive firms to supply prices, however, they did
not. We interviewed representatives from these firms who provided
anecdotal information about their prices. Furthermore, we did not
include the costs of providing dedicated access services in our
analysis to measure the extent to which prices approach costs,
because these data also were unavailable. FCC, which previously
collected data on costs, discontinued cost studies several years
ago. In addition, FCC gave price-cap incumbents the option to
accept price decreases resulting from the CALLS Order, or have
prices reinitialized on the basis of detailed cost studies that
the incumbent could provide. However, no price-cap incumbent
provided a cost study.
Federal Spending on Dedicated Access Services
To examine prices that the government has paid for dedicated
access, we interviewed and obtained dedicated access monthly
prices from the Department of Agriculture (USDA) and DOJ. We
selected those departments because they have many offices that
require high-speed dedicated access. USDA officials indicated they
provided data for all dedicated access purchased by the entire
department. At DOJ, we obtained data from the Justice Management
Division, which both uses and orders a substantial portion of
telecommunication services for agencies within the department. The
Justice Management Division provided data on dedicated access
services. However, these data did not include information from the
Federal Bureau of Investigation. The results from these
departments and agencies may not be representative of the federal
government as a whole.
We interviewed officials with GSA, which awarded and administers
the governmentwide FTS2001 telecommunications contracts with
Sprint and MCI. We also obtained access to GSA's automated pricing
tool, the "SDP Pricer," which provides pricing estimates for
dedicated access and other services and is available to federal
departments and agencies. We compared the prices provided by USDA
and DOJ with prices obtained from the pricing tool. GSA officials
indicated that the SDP Pricer provides estimates only, and that
actual prices paid may be different. For example, GSA officials
indicated that they are aware of instances where service
initiation charges are negotiated and waived between the FTS2001
contractor and the federal government entity purchasing the
services. GSA also indicated that federal entities are not
required to purchase services using FTS2001, nor are they required
to use the lowest cost contractor.
USDA and DOJ provided aggregate and individual circuit pricing
data for 1 month of data and indicated that the month provided was
representative of average spending. These entites also indicated
whether the circuit was purchased under the FTS2001 contracts and
also the identity of the service provider. Each entity provided
individual circuit data that included circuit endpoints, monthly
recurring charges, and speed of the service, among other
information.
We compared the channel termination and local interoffice costs of
dedicated access services, and not the total or long-distance
costs. We were not able to directly compare total entity or GSA
prices to list or contract prices offered by incumbents, as data
on interoffice mileage and the closest wire center were not
available. For circuits purchased under FTS2001, we compared
prices paid with the estimates that the SDP Pricer produced.
FCC Oversight of Dedicated Access
To determine what data FCC utilizes to monitor competition and any
limitations that may exist to its monitoring efforts, we analyzed
FCC triggers for predicting competition as well as FCC data
collection processes for determining and monitoring competition.
We analyzed FCC's strategic plan, performance budget, and measures
that the agency uses to track its progress toward meeting its
stated goals of increasing competition and choice for business. We
then compared these plans, budgets, and measures against criteria
developed from the Government Performance and Results Act of
1993.^3 We discussed all of those elements with FCC senior staff.
In addition, for our three objectives we analyzed and summarized
comments in the rulemaking proceeding on dedicated access, and
interviewed the major incumbent telecommunications carriers,
competitive local exchange providers, Wall Street analysts
covering the dedicated access markets, and representatives of
large telecommunication users.
We conducted our work from November 2005 through October 2006 in
accordance with generally accepted government auditing standards.
^1Typically, price-cap incumbents offer prices across different zones that
reflect the concentration of business demand for dedicated access within a
geographic area. Zones generally correspond with areas of relatively high,
medium, and low business demand density. Zone 1 is generally considered as
inclusive of the central business area, where a large portion of
businesses that would require DS-1 and DS-3 would reside. Prices are
generally lower in zone 1 than in zones 2 or 3--with zone 3 generally
having the highest prices, because costs to provide services are likely
higher in less dense areas. Occasionally an incumbent will offer prices
across five zones. In cases where an incumbent provided pricing across
five zones, we analyzed prices associated with zones 1, 3, and 5.
^2Not all of the major incumbent firms were able to include every discount
that was based on price-flex contracts. One firm was unable to include
discounts that were based on revenue commitments; however, because these
discounts are available in both phase I and phase II areas, there is
little reason to believe that these discounts would affect the prices
available in phase II areas greater or less than it would affect prices in
phase I areas.
^3Pub. L. No. 103-62.
Appendix II: Analysis of Average Revenue Data and List Prices
We analyzed the average amount of revenue that the major incumbent
carriers reported from the sales of DS-1 and DS-3 dedicated access
channel termination in 56 MSAs--27 MSAs with phase II flexibility
for channel terminations and 29 with phase I. We excluded two MSAs
from the data that Verizon provided and two MSAs from AT&T's data
because those MSAs were not under phase I or phase II flexibility
for channel terminations. We performed our analysis in nominal
dollars. We also performed our analysis using the Bureau of Labor
Statistics' Producer Price Index for Wired Telecommunications
Carriers ("telecommunications price index"), as well as using the
general GDP price index to adjust these data to 2005 constant
dollars. Regardless of the price index used, phase II average
revenue in 2005 was higher than phase I average revenue in 2005.
However, using nominal dollars or dollars adjusted using the
general GDP price index did not result in any increases in average
revenue in phase II MSAs over time, whereas adjusting these data
using the telecommunications price index did result in increases
in phase II MSAs.
Because MSAs in these data varied greatly in their size (e.g., Los
Angeles and Greenville), we also performed the analysis and
weighted these data on the basis of the number of businesses with
20 or more employees in each MSA. While an imperfect weight, this
was used as a rough estimate of the level of demand in these MSAs.
We obtained data on the number of businesses from the U.S. Census
Bureau's 2002 Statistics of U.S. Businesses. These were the most
recent data available. Weighting these data did not change our
finding that phase II average revenue was higher than phase I
average revenue.
Tables 7 through 10 show the results of our analysis using
unweighted nominal dollars, unweighted adjusted dollars using the
telecommunications price index, unweighted adjusted dollars using
the general GDP price index, and weighted adjusted dollars using
the telecommunications price index.
Table 7: Summary Statistics of Average Revenue for Channel Terminations,
Unweighted Nominal Dollars
Time frame and level of
current pricing Lower Number of
Speed flexibility Mean limit^a Upperlimit^a MSAs
DS1 Pre-flex--all MSAs $146.53 $139.10 $153.96 56
Pre-flex--MSAs that 152.35 140.40 164.29 29
became phase I
Pre-flex--MSAs that 140.28 131.56 149.01 27
became phase II
2005--all MSAs 128.88 123.49 134.27 56
2005--Phase I MSAs 126.20 120.28 132.12 29
2005--Phase II MSAs 131.77 122.24 141.29 27
2005 less pre-flex--all (17.65)^b (22.63) (12.66) 56
MSAs
2005 less pre-flex--Phase (26.15)^b (33.95) (18.35) 29
I MSAs
2005 less pre-flex--Phase (8.52)^b (12.72) (4.32) 27
II MSAs
2005 less pre-flex--Phase 17.63^b 8.92 26.35 56
II less phase I
DS3 Pre-flex--all MSAs 1,341.46 1,264.28 1,418.65 56
Pre-flex--MSAs that 1,287.32 1,183.23 1,391.41 29
became phase I
Pre-flex--MSAs that 1,399.62 1,282.08 1,517.15 27
became phase II
2005--all MSAs 1,194.97 1,125.05 1,264.90 56
2005--Phase I MSAs 1,069.58 997.75 1,141.41 29
2005--Phase II MSAs 1,329.65 1,225.40 1,433.91 27
2005 less pre-flex--all (146.49)^b (202.75) (90.23) 56
MSAs
2005 less pre-flex--Phase (217.74)^b (296.17) (139.31) 29
I MSAs
2005 less pre-flex--Phase (69.96) (144.84) 4.91 27
II MSAs
2005 less pre-flex--Phase 147.78^b 41.49 254.06 56
II less phase I
Source: GAO analysis of data from AT&T, BellSouth, Qwest and Verizon.
Note: The average revenue data for pre-flex are from 2000, 2001, or 2002.
Verizon provided data from 2000, AT&T and BellSouth provided data from
2001, and Qwest provided data from 2002.
^aThe values are based on 95 percent confidence intervals.
^bThe difference is statistically significant at the 1 percent level or
lower (two-tailed), using mean-difference tests.
Table 8: Summary Statistics of Average Revenue for Channel Terminations,
Unweighted Adjusted Dollars Using Telecommunications Price Index
Timeframe and level of
current pricing Lower Number of
Speed flexibility Mean limit^a Upperlimit^a MSAs
DS1 Pre-flex--all MSAs $133.50 $127.49 $139.51 56
Pre-flex--MSAs that 138.31 128.57 148.05 29
became phase I
Pre-flex--MSAs that 128.33 121.42 135.25 27
became phase II
2005--all MSAs 128.88 123.49 134.27 56
2005--Phase I MSAs 126.20 120.28 132.12 29
2005--Phase II MSAs 131.77 122.24 141.29 27
2005 less pre-flex--all (4.62)^c (8.82) (0.42) 56
MSAs
2005 less pre-flex--Phase (12.11)^b (18.08) (6.14) 29
I MSAs
2005 less pre-flex--Phase 3.43 (0.99) 7.86 27
II MSAs
2005 less pre-flex--Phase 15.54^b 8.26 22.82 56
II less phase I
DS3 Pre-flex--all MSAs 1,226.36 1,156.10 1,296.61 56
Pre-flex--MSAs that 1,173.73 1,078.19 1,269.27 29
became phase I
Pre-flex--MSAs that 1,282.88 1,177.30 1,388.47 27
became phase II
2005--all MSAs 1,194.97 1,125.05 1,264.90 56
2005--Phase I MSAs 1,069.58 997.75 1,141.41 29
2005--Phase II MSAs 1,329.65 1,225.40 1,433.91 27
2005 less pre-flex--all (31.38) (81.71) 18.95 56
MSAs
2005 less pre-flex--Phase (104.15)^b (173.45) (34.84) 29
I MSAs
2005 less pre-flex--Phase 46.77 (17.89) 111.43 27
II MSAs
2005 less pre-flex--Phase 150.91^b 58.29 243.54 56
II less phase I
Source: GAO analysis of data from AT&T, BellSouth, Qwest and Verizon.
Note: Average revenue figures are in 2005 dollars using the Bureau of
Labor Statistics Producer Price Index for Wired Telecommunications
Carriers.
^aThe values are based on 95 percent confidence intervals.
^bThe difference is statistically significant at the 1 percent level or
lower (two-tailed), using mean-difference tests.
^cThe difference is statistically significant at the 5 percent level or
lower (two-tailed), using mean-difference tests.
Table 9: Summary Statistics of Average Revenue for Channel Terminations,
Unweighted Adjusted Dollars Using General GDP Price Index
Timeframe and level of
current pricing Lower Number of
Speed flexibility Mean limit^a Upperlimit^a MSAs
DS1 Pre-flex--all MSAs $161.62 $152.82 $170.42 56
Pre-flex--MSAs that 168.47 154.38 182.56 29
became phase I
Pre-flex--MSAs that 154.26 143.81 164.71 27
became phase II
2005--all MSAs 128.88 123.49 134.27 56
2005--Phase I MSAs 126.20 120.28 132.12 29
2005--Phase II MSAs 131.77 122.24 141.29 27
2005 less pre-flex--all (32.73)^b (38.72) (26.75) 56
MSAs
2005 less pre-flex--Phase (42.26)^b (52.05) (32.49) 29
I MSAs
2005 less pre-flex--Phase (22.49)^b (27.09) (17.89) 27
II MSAs
2005 less pre-flex--Phase 19.77^b 9.12 30.43 56
II less phase I
DS3 Pre-flex--all MSAs 1,475.83 1,391.02 1,560.64 56
Pre-flex--MSAs that 1,419.26 1,305.38 1,533.15 29
became phase I
Pre-flex--MSAs that 1,536.59 1,406.46 1,666.72 27
became phase II
2005--all MSAs 1,194.97 1,125.05 1,264.90 56
2005--Phase I MSAs 1,069.58 997.75 1,141.41 29
2005--Phase II MSAs 1,329.65 1,225.40 1,433.91 27
2005 less pre-flex--all (280.86)^b (343.40) (218.32) 56
MSAs
2005 less pre-flex--Phase (349.68)^b (437.96) (261.40) 29
I MSAs
2005 less pre-flex--Phase (206.94)^b (292.16) (121.72) 27
II MSAs
2005 less pre-flex--Phase 142.74^c 22.85 262.63 56
II less phase I
Source: GAO analysis of data from AT&T, BellSouth, Qwest and Verizon.
Note: The average revenue figures are in 2005 dollars using the general
GDP Price Index.
^aThe values are based on 95 percent confidence intervals.
^bThe difference is statistically significant at the 1 percent level or
lower (two-tailed), using mean-difference tests.
^cThe difference is statistically significant at the 5 percent level or
lower (two-tailed), using mean-difference tests.
Table 10: Summary Statistics of Average Revenue for Channel Terminations,
Weighted Adjusted Dollars Using the Telecommunications Price Index
Timeframe and level of current Lower Upper Number of
Speed pricing flexibility Mean limit^a limit^a MSAs^c
DS1 Pre-flex--all MSAs $132.95 $132.88 $133.02 56
Pre-flex--MSAs that became 134.57 134.48 134.67 29
phase I
Pre-flex--MSAs that became 130.25 130.17 130.34 27
phase II
2005--all MSAs 126.80 126.73 126.86 56
2005--Phase I MSAs 123.93 123.86 124.00 29
2005--Phase II MSAs 131.56 131.44 131.67 27
2005 less pre-flex--all MSAs (6.15)^b (6.20) (6.11) 56
2005 less pre-flex--Phase I (10.64)^b (10.70) (10.59) 29
MSAs
2005 less pre-flex--Phase II 1.30^b 1.23 1.38 27
MSAs
2005 less pre-flex--Phase II 11.95^b 11.86 12.04 56
less phase I
DS3 Pre-flex--all MSAs 1,175.38 1,174.62 1,176.14 56
Pre-flex--MSAs that became 1,128.80 1,128.08 1,129.48 29
phase I
Pre-flex--MSAs that became 1,252.78 1,251.23 1,254.34 27
phase II
2005--all MSAs 1,163.53 1,162.74 1,164.32 56
2005--Phase I MSAs 1,081.89 1,081.28 1,082.50 29
2005--Phase II MSAs 1,299.13 1,297.54 1,300.71 27
2005 less pre-flex--all MSAs (11.85)^c (12.34) (11.37) 56
2005 less pre-flex--Phase I (46.89)^c (47.46) (46.32) 29
MSAs
2005 less pre-flex--Phase II 46.34^c 45.54 47.14 27
MSAs
2005 less pre-flex--Phase II 93.23^c 92.25 94.21 56
less phase I
Source: GAO analysis of data from AT&T, BellSouth, Qwest and Verizon.
Note: The average revenue figures are in 2005 dollars using the Bureau of
Labor Statistics' Producer Price Index for Wired Telecommunications
Carriers. Data are weighted on the basis of the number of businesses with
20 or more employees.
^aThe values are based on a 95 percent confidence intervals.
^bThe difference is statistically significant at the 1 percent level of
lower (two-tailed), using mean-difference tests.
^cThe number of observations for the frequency weights is 349,512 for the
56 MSAs; 218,164 for the 29 MSAs; and 131,348 for the 27 MSAs.
We also analyzed list prices in the published tariffs from the four major
incumbent firms to compare how phase II pricing flexibility and the CALLS
Order have changed these prices. We compiled data on prices for channel
terminations and dedicated transport (both fixed and variable charges) for
month-to-month, 3-year, and 5-year terms across three density zones from
the published tariffs as of June 1, 2006. We eliminated any comparisons
where the tariff contained price-flex prices for channel terminations, but
no MSAs covered by the tariff had phase II flexibility for channel
terminations (e.g., AT&T's tariff that covers Nevada has price-flex list
prices for channel terminations, yet AT&T has not received phase II
flexibility for channel terminations in any MSAs in Nevada).
We made the following three comparisons for all combinations of circuit
components, terms, and zones: (1) 2006 price-flex list prices compared
with initial prices (prior to pricing flexibility); (2) 2006 price-cap
prices compared with initial prices; and (3) 2006 price-flex prices
compared with 2006 price-cap prices. We also performed this analysis after
adjusting the data to constant dollars. Adjusting the dollars did not
change the basic findings of our analysis. Tables 11 and 12 show the
results in nominal dollars.
Table 11: Summary Statistics of List Price Comparisons for all DS-1
Combinations in Nominal Dollars
Mean price comparisons
Price-flex Price-cap
2006 less 2006 less Price-flex 2006
initial initial less price-cap Number of
Component Term Zone price price 2006 comparisons
Channel All All $7.73^a $(9.46)^a $17.20^a 144
terminations
Monthly All 20.56^a (3.45) 24.01^a 48
Zone 17.76^a (1.20) 18.96^a 16
1
Zone 21.03^a (4.25) 25.28^a 16
2
Zone 22.89^a (4.90) 27.79^a 16
3
3-yr All 2.74 (12.54)^a 15.28^a 48
Zone 0.87 (9.80)^a 10.67^a 16
1
Zone 3.17 (13.27)^a 16.45^a 16
2
Zone 4.17 (14.55)^a 18.73^a 16
3
5-yr All (0.10) (12.39)^a 12.30^a 48
Zone (1.12) (9.34)^a 8.22^a 16
1
Zone (0.21) (13.21)^a 13.00^a 16
2
Zone 1.05 (14.62)^a 15.67^a 16
3
Fixed All All 1.58^b (6.06)^a 7.64^a 216
transport
Monthly All 4.16^a (4.16)^a 8.32^a 72
Zone 3.60^a (4.11)^a 7.71^a 24
1
Zone 4.11^a (4.27)^b 8.37^a 24
2
Zone 4.78^a (4.09)^b 8.87^a 24
3
3-yr All 0.39 (6.79)^a 7.18^a 72
Zone 0.07 (6.11)^a 6.19^a 24
1
Zone 0.39 (6.73)^a 7.12^a 24
2
Zone 0.70 (7.52)^a 8.22^a 24
3
5-yr All 0.19 (7.24)^a 7.44^a 72
Zone (0.10) (6.28)^a 6.18^a 24
1
Zone 0.20 (6.77)^a 6.96^a 24
2
Zone 0.49 (8.68)^a 9.17^a 24
3
Variable All All 0.81^a (2.29)^a 3.10^a 216
transport
Monthly All 1.37^a (2.00)^a 3.36^a 72
Zone 1.28^a (1.91)^a 3.18^a 24
1
Zone 1.38^a (1.95)^a 3.33^a 24
2
Zone 1.44^a (2.13)^a 3.56^a 24
3
3-yr All 0.65^a (2.50)^a 3.14^a 72
Zone 0.51^c (2.39)^a 2.90^a 24
1
Zone 0.63^b (2.51)^a 3.14^a 24
2
Zone 0.80^a (2.59)^a 3.39^a 24
3
5-yr All 0.43^a (2.37)^a 2.80^a 72
Zone 0.40 (2.22)^a 2.63^a 24
1
Zone 0.45^c (2.39)^a 2.84^a 24
2
Zone 0.44 (2.50)^a 2.94^a 24
3
Source: GAO analysis of data from tariffs filed with the FCC in 2001 and
2006, including AT&T (Ameritech FCC No. 2, Nevada Bell FCC No. 1, Pacific
Bell FCC No. 1, Southern New England Bell FCC No. 39, and Southwestern
Bell FCC No. 73), BellSouth FCC No. 1, Qwest FCC No. 1, and Verizon FCC
Nos. 1, 11, and 14.
Note: Initial prices are from 2001 and 2006 prices are as of June 2006.
^aThe price difference is statistically significant at the 1 percent level
or lower, two-tailed.
^bThe price difference is statistically significant at the 5 percent level
or lower, two-tailed.
^cThe price difference is statistically significant at the 10 percent
level or lower, two-tailed.
Table 12: Summary Statistics of List Price Comparisons for all DS-3
Combinations in Nominal Dollars
Mean price comparisons
Price-flex Price-cap Price-flex
2006 less 2006 less 2006 less
initial initial price-cap Number of
Component Term Zone price price 2006 comparisons
Channel All All $98.12^a $(113.95)^a $212.08^a 144
terminations
Monthly All 137.37^a (118.78)^a 256.14^a 48
Zone 127.88^b (112.81)^a 240.69^a 16
1
Zone 137.37^b (121.90)^a 259.27^a 16
2
Zone 146.87^b (121.61)^a 268.48^a 16
3
3-yr All 90.59^a (115.75)^a 206.34^a 48
Zone 82.17^b (114.37)^a 196.54^a 16
1
Zone 90.07^b (113.81)^a 203.88^a 16
2
Zone 99.54^b (119.07)^a 218.61^a 16
3
5-yr All 66.41^a (107.34)^a 173.75^a 48
Zone 57.08^c (105.19)^a 162.28^a 16
1
Zone 66.92^b (106.71)^a 173.63^a 16
2
Zone 75.23^b (110.11)^a 185.33^a 16
3
Fixed All All 10.50^b (60.34)^a 70.84^a 216
transport
Monthly All 21.95^b (52.55)^a 74.50^a 72
Zone 21.72 (52.32)^a 74.03^a 24
1
Zone 21.96 (49.53)^a 71.49^a 24
2
Zone 22.17 (55.81)^a 77.98^a 24
3
3-yr All 3.46 (66.64)^a 70.09^a 72
Zone 3.12 (66.19)^a 69.31^a 24
1
Zone 2.98 (65.49)^a 68.47^a 24
2
Zone 4.28 (68.23)^a 72.50^a 24
3
5-yr All 6.09 (61.83)^a 67.92^a 72
Zone 5.78 (60.68)^a 66.47^a 24
1
Zone 5.65 (60.61)^a 66.25^a 24
2
Zone 6.86 (64.20)^a 71.06^a 24
3
Variable All All 2.64^a (12.57)^a 15.21^a 216
transport
Monthly All 3.84^a (13.46)^a 17.29^a 72
Zone 3.51^c (11.83)^a 15.34^a 24
1
Zone 4.04^c (13.11)^a 17.15^a 24
2
Zone 3.97^c (15.43)^a 19.39^a 24
3
3-yr All 2.18^c (14.01)^a 16.19^a 72
Zone 2.05 (12.30)^a 14.35^a 24
1
Zone 2.01 (13.78)^a 15.79^a 24
2
Zone 2.48 (15.96)^a 18.44^a 24
3
5-yr All 1.90 (10.24)^a 12.15^a 72
Zone 1.81 (8.81)^b 10.62^b 24
1
Zone 1.76 (10.12)^b 11.88^b 24
2
Zone 2.14 (11.80)^b 13.94^b 24
3
Source: GAO analysis of data from tariffs filed with the FCC in 2001 and
2006, including AT&T (Ameritech FCC No. 2, Nevada Bell FCC No. 1, Pacific
Bell FCC No. 1, Southern New England Bell FCC No. 39, and Southwestern
Bell FCC No. 73), BellSouth FCC No. 1, Qwest FCC No. 1, and Verizon FCC
Nos. 1, 11, and 14.
Note: Initial prices are from 2001 and 2006 prices are as of June 2006.
^aThe price difference is statistically significant at the 1 percent level
or lower, two-tailed.
^bThe price difference is statistically significant at the 5 percent level
or lower, two-tailed.
^cThe price difference is statistically significant at the 10 percent
level or lower, two-tailed.
Appendix III: Comments from the Federal Communications Commission
Appendix IV: GAO Contact and Staff Acknowledgments
GAO Contact
JayEtta Hecker (202) 512-2834 or [email protected]
Staff Acknowledgments
In addition, Steve Martin, Assistant Director; Eli Albagli; Edda
Emmanuelli-Perez; Colin Fallon; Brandon Haller; John Karikari; Logan
Kleier; Teague Lyons; Grant Mallie; Josh Ormond; Andrew Von Ah; and Mindi
Weisenbloom made key contributions to this report.
(544114)
GAO's Mission
The Government Accountability Office, the audit, evaluation and
investigative arm of Congress, exists to support Congress in meeting its
constitutional responsibilities and to help improve the performance and
accountability of the federal government for the American people. GAO
examines the use of public funds; evaluates federal programs and policies;
and provides analyses, recommendations, and other assistance to help
Congress make informed oversight, policy, and funding decisions. GAO's
commitment to good government is reflected in its core values of
accountability, integrity, and reliability.
Obtaining Copies of GAO Reports and Testimony
The fastest and easiest way to obtain copies of GAO documents at no cost
is through GAO's Web site ( www.gao.gov ). Each weekday, GAO posts
newly released reports, testimony, and correspondence on its Web site. To
have GAO e-mail you a list of newly posted products every afternoon, go to
www.gao.gov and select "Subscribe to Updates."
Order by Mail or Phone
The first copy of each printed report is free. Additional copies are $2
each. A check or money order should be made out to the Superintendent of
Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more
copies mailed to a single address are discounted 25 percent. Orders should
be sent to:
U.S. Government Accountability Office 441 G Street NW, Room LM Washington,
D.C. 20548
To order by Phone: Voice: (202) 512-6000 TDD: (202) 512-2537 Fax: (202)
512-6061
To Report Fraud, Waste, and Abuse in Federal Programs
Contact:
Web site: www.gao.gov/fraudnet/fraudnet.htm E-mail:
[email protected] Automated answering system: (800) 424-5454 or (202)
512-7470
Congressional Relations
Gloria Jarmon, Managing Director, [email protected] (202) 512-4400 U.S.
Government Accountability Office, 441 G Street NW, Room 7125 Washington,
D.C. 20548
Public Affairs
Paul Anderson, Managing Director, [email protected] (202) 512-4800
U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, D.C. 20548
www.gao.gov/cgi-bin/getrpt?GAO-07-80 .
To view the full product, including the scope
and methodology, click on the link above.
For more information, contact JayEtta Z. Hecker at 202-512-2834 or
[email protected].
Highlights of [47]GAO-07-80 , a report to the Chairman, Committee on
Government Reform, House of Representatives
November 2006
TELECOMMUNICATIONS
FCC Needs to Improve Its Ability to Monitor and Determine the Extent of
Competition in Dedicated Access Services
Government agencies and businesses that require significant capacity to
meet voice and data needs depend on dedicated access services. This
segment of the telecommunications market generated about $16 billion in
revenues for the major incumbent telecommunications firms in 2005. The
Federal Communications Commission (FCC) has historically regulated
dedicated access prices.
With the Telecommunications Act of 1996, FCC reformed its rules to rely on
competition to bring about cost-based pricing. Starting in 2001, FCC
granted pricing flexibility on the basis of a proxy measure of
competition. GAO examined (1) the extent that alternatives are available
in areas where FCC granted pricing flexibility, (2) how prices have
changed since the granting of pricing flexibility, and the effect on
government agencies, and (3) how FCC monitors competition. GAO's work
included analyzing data on competitive alternatives, list prices, and
average revenue, and interviewing FCC officials and industry
representatives.
[48]What GAO Recommends
GAO recommends that FCC better define effective competition, and consider
additional data to measure and monitor competition. FCC disagreed that
they need to better define competition and collect additional data. GAO
maintains that additional data collection is necessary for FCC to better
fulfill its regulatory responsibilities.
In the 16 major metropolitan areas we examined, available data suggest
that facilities-based competitive alternatives for dedicated access are
not widely available. Data on the presence of competitors in commercial
buildings suggest that competitors are serving, on average, less than 6
percent of the buildings with demand for dedicated access in these areas.
For buildings with higher levels of demand, facilities-based competition
is more moderate, with 15 to 25 percent of buildings showing competitive
alternatives, depending on the level of demand. Limited competitive build
out in these MSAs could be caused by a variety of entry barriers,
including government zoning restrictions and difficulty gaining access to
buildings from building owners. In addition, where demand for dedicated
access is relatively small, it is unlikely to be economically viable for
competitors to extend their networks to the end user. FCC has also noted
that, where competitors can lease unbundled network elements from
incumbent providers, there may be less incentive for competitors to invest
in their own facilities.
Available data suggest that incumbents' list prices and average revenues
for dedicated access services have decreased since 2001, resulting from
price decreases due to regulation and contract discounts. However, in
areas where FCC granted full pricing flexibility due to the presumed
presence of competitive alternatives, list prices and average revenues
tend to be higher than or the same as list prices and average revenues in
areas still under some FCC price regulation. According to the large
incumbent firms, many large customers needing service in areas with
pricing flexibility purchase dedicated access services under contracts
that provide additional discounts. However, GAO found that contracts do
not generally affect the differential cited previously, and that contracts
also contain various conditions or termination penalties competitors argue
inhibit customer choice. Government agencies, to the extent that they
purchase dedicated access off of General Services Administration
contracts, are generally shielded from price increases due to
prenegotiated rates. However, not all agencies purchase off of these
contracts.
FCC uses various data to assess competition in dedicated access, but these
data are limited in their ability to describe the state of competition
accurately. For example, these data measure potential competition at one
point in time and are not revisited or updated, even though competitors
may enter bankruptcy or be bought by the incumbent firm. FCC also collects
data from external parties through its rulemaking proceedings, but those
parties have no obligation to provide data, and FCC has limited mechanisms
to verify the reliability of any data submitted. FCC's strategic plan and
various rulemakings have defined FCC's obligation to assess and ensure
competition in dedicated access. FCC stated that gathering and analyzing
additional data would be costly and burdensome. Yet without more complete
and reliable data, FCC is unable to determine whether its deregulatory
policies are achieving their goals.
References
Visible links
35. http://www.lexis.com/research/buttonTFLink?_m=0a620515492fdec37dccb0ed40dbe58b&_xfercite=%3ccite%20cc%3d%22USA%22%3e%3c%21%5bCDATA%5b2006%20FCC%20LEXIS%204580%5d%5d%3e%3c%2fcite%3e&_butType=4&_butStat=0&_butNum=95&_butInline=1&_butinfo=44%20USC%203506&_fmtstr=VKWIC&docnum=8&_startdoc=1&wchp=dGLbVlz-zSkAb&_md5=ed621ef7aea5049eed3930d500f48cd6
36. http://www.gao.gov/cgi-bin/getrpt?GAO-01-289
37. http://www.gao.gov/cgi-bin/getrpt?GAO-06-426
38. http://www.gao.gov/cgi-bin/getrpt?GAO-06-189
47. http://www.gao.gov/cgi-bin/getrpt?GAO-07-80
*** End of document. ***