WCB_Comprehensive Market Data Collection for ISAS_SS Part B rev 081514 final

3060-xxxx
December 2013 (rev. August 2014)
Comprehensive Market Data Collection for Interstate Special Access Services

SUPPORTING STATEMENT
B. Collections of Information Employing Statistical Methods
The collection is a one-time census survey, consisting of standardized questions. The goal of the
collection is to obtain quantitative and qualitative data and information for a competition analysis of the
special access market. One question in the survey, directed at Competitive Providers, uses sampling to
obtain additional information on the Locations served by the Competitive Provider.1
1.
The Commission plans to collect data from all Providers and a wide range of Purchasers of
special access services, including all facility-based mobile wireless service providers, as well as certain
entities that provide Best Efforts Business Broadband Internet Access Service, i.e., those entities with
15,000 or more customers or 1,500 or more business broadband customers. The estimated potential
respondent pool for the collection is 4,000.
We plan to collect data for all geographic areas subject to price cap regulation nationwide.2 The
potential respondent universe includes incumbent local exchange carriers (ILECs), competitive local
exchange carriers (CLECs), interexchange carriers (IXCs), cable system operators, fixed and mobile
wireless service providers (including wireless Internet service providers (WISPs) and wireless
telecommunications carriers), terrestrial and satellite mobile wireless service providers, electric utilities,
local government entities, third party network providers, certain private wireless licensees and certain
providers of information services.3
We expect a high response rate for the collection as a whole. To help ensure that we have the full
universe of facilities-based providers of special access services, all “Facilities-Based Providers of
Broadband Connections to End User Locations” that were required to file the FCC Form 477 “Local
Telephone Competition and Broadband Reporting” to report broadband connections to end user locations
in 2012 are required to respond to this data collection to affirmatively indicate whether or not they are
covered by the scope of the data collection, i.e., whether or not the entity is a Provider or Purchaser of
Dedicated Services or a covered entity providing Best Efforts Business Broadband Internet Access
Service. We expect that most, if not all, of the facilities-based Providers of special access and best efforts
services for which the Commission mandates the submission of data and information are likely to have
filed the Form 477 based on that form’s reporting criteria.4 Form 477 filers not covered by the scope of
the collection will only have to certify as much and are not required to submit data and information in
response to the standardized questions contained in the data collection.

1

Capitalized and italicized terms contained herein are defined in the collection. See Attachment A § I, Definitions.

2

See Special Access Data Collection Order, 27 FCC Rcd at 16327, 16328, paras. 20, 23.

3

Wireless Telecommunications Carriers is a Census Bureau Category that includes: beeper (i.e., radio pager)
communication carriers; cellular telephone communication carriers; cellular telephone services; mobile telephone
communication carriers, except satellite; paging services, except satellite; radio paging services communications
carriers; ship-to-shore broadcasting communication carriers, except satellite; telecommunications carriers, cellular
telephone; telephone communications carriers, wireless, except satellite; two-way paging communication carriers,
except satellite; wireless data communication carriers, except satellite; wireless Internet service providers, except
satellite; wireless telephone communications carriers, except satellite; and wireless video services, except satellite.
See Census Bureau, NAICS 517210, http://www.census.gov/econ/industry/def/d517210 htm (last visited July 15,
2013).
4

See Instructions for Local Telephone Competition and Broadband Reporting, FCC Form 477, at 2 (Sept. 2013),
available at http://www.fcc.gov/forms (last visited Sept. 24, 2013).

1

3060-xxxx
December 2013 (rev. August 2014)
Comprehensive Market Data Collection for Interstate Special Access Services
The estimated number of potential respondents that are Providers, covered entities providing Best
Efforts Business Broadband Internet Access Services, or facility-based mobile providers is approximately
1,700-1,800. We estimate that 1,662 “Facilities-Based Providers of Broadband Connections to End User
Locations” file FCC Form 477 (Local Telephone Competition and Broadband Reporting).5 This group
includes the largest known providers of facilities-based special access and “best efforts” service, i.e.,
ILECs, cable system operators, and CLECs, and mobile wireless service providers. There may also be a
small number of Providers and mobile wireless service providers that must respond to our data request,
which are not included in the identified 1,662 providers. For example, there may exist a small number of
facility-based suppliers of Dedicated Services (including firms that engage in self-provision, for example,
by use of fixed wireless) that do not report broadband connections to end user locations on the Form 477.
Leaning toward overstating the burden of the proposed collection, we estimate the total number of
Providers, covered entities providing Best Efforts Business Broadband Internet Access Services and
facility-based mobile wireless service providers that do, and do not, file the Form 477 but that will be
required to respond to the data collection is between 1,700-1,800.
The preceding estimate focuses on Providers, covered entities providing Best Efforts Business
Broadband Internet Access Services, and facility-based mobile wireless service providers. We expect
most of these entities are Purchasers as well. We also anticipate the collection will potentially require
responses from additional Purchasers. These Purchasers fall into two groups: (1) those that provide
communications services or are otherwise required to file the Form 499-A “Telecommunications
Reporting Worksheet;” and (2) other Purchasers.
All intrastate, interstate, and international providers of telecommunications in the United States
must file the Form 499-A.6 The large majority of the potential respondents to the collection are likely to
fall within the reporting requirements of the Form 499-A due to its filing criteria.7 Based on the number
of filers – counted at the holding company level – that submitted the Form 499-A for 2012, we estimate
that the number of potential respondent Purchasers that file the Form 499-A and may be required to
respond to the collection is about 5,200. This estimate includes the 1,700 to 1,800 Providers that are
likely to file both the Form 477 and the Form 499-A.8

5

At the holding company level, there were, for the June 2012 filing of the Form 477, 1,662 providers of Internet
access services at bandwidths of at least 200 kbps in at least one direction. See Industry Analysis and Technology
Div., Wireline Comp. Bur., Internet Access Services: Status as of June 30, 2012, Table 12 (May 2013), available at
http://hraunfoss fcc.gov/edocs_public/attachmatch/DOC-321076A1.pdf (last visited Sept. 24, 2013). We count
providers at the holding company level, i.e., a parent company and its affiliates are counted as one filer, though
providers can respond to the data request at lower levels of their operations. There were 1,906 total providers.
Thus, 244 providers (the difference between 1,662 and 1,906) did not provide broadband connections (Internet
access). The 1,662 providers included satellite telecommunications providers, non-interconnected VoIP providers,
interconnected VoIP service providers, and international service providers. Data are based on filings as of June
2012. Id.
6

See 2013 Telecommunications Reporting Worksheet Instructions, FCC Form 499-A, at 2-3 (Feb. 2013), available
at http://www fcc.gov/forms (last visited at Sept. 24, 2013).
7

As stated in the instructions to the Form 499-A, “[t]he term “interstate telecommunications” includes, but is not
limited to, the following types of services: wireless telephony, including cellular and personal communications
services (PCS); paging and messaging services; dispatch and operator services; mobile radio services; access to
interexchange service; special access; wide area telecommunications services (WATS); subscriber toll-free and 900
services; message telephone services (MTS); private line; telex; telegraph; video services; satellite services; resale
services; Frame Relay services; asynchronous transfer mode (ATM) services; Multi-Protocol Label Switching
(MPLS) services; audio bridging services; and interconnected VoIP services.” Id. at 3.

8

The count is again made at the holding company level.

2

3060-xxxx
December 2013 (rev. August 2014)
Comprehensive Market Data Collection for Interstate Special Access Services
In addition to the potential respondents that are required to file the Form 499-A, we estimate that
approximately 1,200 respondents are private wireless licensees that purchase Dedicated Service in price
cap areas and are therefore covered by the scope of the collection and will respond. This class of
licensees generally consists of state and local entities and institutions and businesses that operate wireless
facilities for internal, non-commercial use. While our licensing records show several thousands of such
licensed entities, we are unable to quantify the exact number of private wireless licensees that meet the
reporting requirements of the collection. We thus reach the estimate of 1,200 recognizing that there is
limited information available to estimate this class of potential respondents with any degree of certainty.
We then decreased the total estimate to account for the exclusion of Purchasers that purchased
less than $5 million in Dedicated Services in 2013 as a result of the recent changes. Working from the
Pareto principle, roughly 80 percent of purchaser revenue should come from 20 percent of the purchasers,
which means a large percentage of purchasers should qualify for the exclusion.9 However, in
acknowledging our lack of Dedicated Service revenue and expenditure information and attempting not to
underestimate the burden for the data collection, we estimate that a de minimis threshold of less than $5
million in expenditures would exempt 50 percent of Purchasers from responding. That said, we expect
more than 50 percent of Purchasers have less than $5 million in 2013 Dedicated Services expenditures.
Factoring in all the potential types of entities subject to the collection, the total estimated pool of
respondents is about 4,000. As detailed below in Table B.1, the vast majority of the respondents will be
businesses or other for-profit entities. While we are unable to quantify with any degree of certainty, we
expect that no more than ten percent of the covered respondents are state, local, or tribal government
entities. An even smaller number of respondents are expected to be not-for-profit institutions.
Table B.1 – Estimated Respondent Pool
Respondent Type
Businesses or other for-profit
entities
State, local or tribal entities
Not-for-profit institutions
Total

Estimated Percentage
of Total
≥91%

Estimated Number of
Potential Respondents
3,640

≤8%
≤1%
100%

320
40
4,000

We do not expect that all of these potential respondents will actually have to submit data and
information. Many will fall outside the scope of the collection because they do not actually provide or
purchase Dedicated Service in price cap areas or provide covered Best Efforts Business Broadband
Internet Access Services. The respondent pool is therefore an overestimate. For additional details on our
methods to maximize responses rates see the response to Item B.3.
The Commission considered and rejected sampling, except in one narrow instance, because the
Commission “believe[s] that the process of identifying and collecting a representative sample would be
unlikely to substantially reduce provider burdens, and could significantly lengthen the data collection
process.”10 For an additional explanation of the Commission’s decision not to use sampling, see the
response to Item B.2 below.
The Commission has not previously conducted this collection.
9

See Y.S. Chen, P.P. Chong, & M.Y. Tong, Mathematical and Computer Modeling of the Pareto Principle, 19
Mathematical and Computer Modeling 61-80 (May 1994).

10

Special Access Data Collection Order, 27 FCC Rcd at 16328-29, paras. 24-25 (explaining decision not to use a
sampling approach).

3

3060-xxxx
December 2013 (rev. August 2014)
Comprehensive Market Data Collection for Interstate Special Access Services

2.
This information collection applies to Providers and Purchasers of Dedicated Service and nonexempt entities providing Best Efforts Business Broadband Internet Access Service as explained in the
answer to Item B.1.
Almost the entire information collection relies on a census and not a sampling methodology.
The Commission considered whether it could reduce the burden of this information collection by
collecting all of the data from a sample of locations and or/geographies. The Commission did not adopt a
sampling approach, however, because due to the process of identifying and collecting a representative
sample, the Commission believed a sampling methodology would not reduce the burden on respondents
sufficiently, if at all.11 Further, the Commission determined that a sampling methodology could lengthen
the collection process.
As the Commission noted in the Special Access Data Collection Order, it would be very difficult
to design a representative sample without coming close to covering the entire country.12 The Commission
designed the collection to obtain data at a very granular level because it determined that competition may
occur at the building/tower.13 There are a wide range of factors that could influence the effectiveness of
competition in any location: state and local rules and regulations; regional input prices, weather
conditions, soil types and gradients; the nature of localized demand, including the number and density of
potential purchasers, and the volume and density of their demands; the identity and number of nearby
competitors; and a host of other factors. Consequently, identifying a representative sample of geographic
regions likely to provide appropriate controls for such factors represents a challenge. On the one hand, a
representative sample of regions used in the pricing flexibility rules, i.e., MSAs and rural equivalents,
would nearly encompass the entire country and would be unlikely to result in cost savings as compared to
a census. On the other hand, using much smaller geographies would also create substantial uncertainty
for respondents. For example, respondents would face substantial costs, if provided with a list of sample
census blocks, in determining which of their serviced buildings lay within those census blocks. Indeed, in
many cases, this would be more difficult than simply providing all their data.
Alternatively, we could require all respondents to identify all relevant locations so that a sample
could be drawn from that census in a scientific way. However, that methodology would likely lengthen
the data collection process, because it would require two collections to be conducted sequentially:
initially requiring carriers to conduct a census of their served locations from which a sample could be
drawn, and then having them answer a subsequent set of questions about locations in the sample. Second,
respondents likely would have to do the same or greater amount of coding to “pull” a sample of records as
it would if it pulled all records. Third, while the costs in burden saved through sampling are likely to be
relatively small, the statistical error of any conclusions based on a sample could be significantly higher
than conclusions based on a census.
We do employ sampling, however, for the limited purpose of analyzing the evolution of buildout
to Locations with Connections by Competitive Providers. In Question II.A.6, we plan to provide
Competitive Providers with a sample from the total Locations they report elsewhere in the collection. For
each sampled Location, the Competitive Provider must state the month and year that it first provided a
Connection that is owned, leased pursuant to an Indefeasible Right of Use (IRU) agreement, or obtained
as an Unbundled Network Element (UNE) to the Location. If the Connection was originally supplied to
the Location over a UNE, the Competitive Provider will indicate when (if at all) it switched to using a
Connection that it owns or leases as an IRU. The sampling question requires the respondent to research
11

Id.

12

Id.

13

Id. at 16327-28, para. 22.

4

3060-xxxx
December 2013 (rev. August 2014)
Comprehensive Market Data Collection for Interstate Special Access Services
the deployment history of each sampled Location in a given area. This information will then show how
Competitive Providers expand their facilities to nearby Locations over time, allowing the Commission to
test whether providers can easily, and do in fact, extend service to buildings near served Locations as
claimed by commenters in the proceeding.
Obtaining this history for all served Locations is highly burdensome (respondents often do not
keep data on past deployment decisions in readily accessible formats), and so realistically data of this
nature can only be obtained by sampling. Moreover, we can develop a true random sample of the data,
because we are collecting the addresses of all relevant Locations, from the full list of Locations reported
by each Competitive Provider, and we can do so without significantly slowing down our data collection
process (it will take significant time to process the data received from the main data request, which
creates a window in which the respondents can provide their facilities history).
We will derive the sample list of Locations as follows. A random sample will be drawn from the
Locations reported in response to Question II.A.4 equal to the maximum of: (i) p percent of the
Locations to which the Competitive Provider had a Connection in 2013 rounded up to the nearest integer
or (ii) the number two.14 Competitive Providers with one such Location will not need to answer this
question. The percentage p will be selected to ensure that about 1,600 sampling units are drawn; based on
a current estimate that Competitive Providers serve on their own facilities 200,000 Locations, the value of
p is expected to be on the order of 0.8 percent. This sample size is expected to provide ranges for
estimates of proportions of the national population that are with 95% confidence expected to include the
true value of the estimated proportion within ±2.5% of the estimate. For example, if the sample produces
an estimate of 50% of the national population, the true value of the proportion would be in the range
(47.5%, 52.5%) with 95% confidence. The four served Locations nearest each of the sampled served
Locations will be added to the random sample of served Locations, and the resulting list, with duplicates
removed, will be given to the Competitive Provider.
Detailed data on the evolution of Competitive Provider networks are necessary for the
Commission to understand how competitive facilities are deployed over time and whether the presence of
competitive facilities provides a threat of competitive entry in nearby or adjacent areas. We believe that
the sampling methodology described above is appropriate for this limited purpose.
3.
The Commission plans to use the collected data for a one-time, market analysis. The analysis
will include “econometrically sound panel regressions . . . of the prices for special access on
characteristics such as 1) the number of facilities-based competitors (both actual and potential); 2) the
availability of, pricing of, and demand for best efforts business broadband Internet access services; 3) the
characteristics of the purchased service; and 4) other factors that influence the pricing decisions of special
access providers, including cost determinants (e.g., density of sales) and factors that deliver economies of
scale and scope (e.g., level of sales).”15 The Commission also plans to assess the reasonableness of terms
and conditions offered by ILECs for special access service.16 Once the data are obtained and analyzed,
the Commission will evaluate whether it is appropriate to make changes to its existing pricing flexibility
rules to better target regulatory relief in competitive areas and evaluate whether remedies are appropriate
to address any potentially unreasonable terms and conditions.
We expect the proposed collection is adequate for our intended uses. The Commission’s
regression analysis will rely heavily on the quantitative data obtained from Providers on their facilities
14

Forcing a minimum of two sampling units per Competitive Provider allows for unambiguous estimates of the
stratified sample’s standard deviation.
15

Special Access Data Collection Order, 27 FCC Rcd at 16346, para. 68.

16

Id. at 16354-56, paras. 91-93.

5

3060-xxxx
December 2013 (rev. August 2014)
Comprehensive Market Data Collection for Interstate Special Access Services
used to serve Locations, billing information at the circuit level, network maps, etc. The information from
covered entities that provide a Best Efforts Business Broadband Internet Access Service will help to
assess the extent to which it is a substitute for Dedicated Service. The information obtained from
Purchasers will help us identify harmful, anticompetitive conduct in the sale of Dedicated Service. The
Bureau has also directed respondents to report data according to specified record formats to facilitate
analysis.
Entities that provide or purchase Dedicated Service or provide Best Efforts Business Broadband
Internet Access Service in price cap areas are required to respond to the collection unless otherwise
exempt. Covered entities failing to respond are subject to monetary forfeitures of up to $160,000 for each
violation or each day of a continuing violation, up to a maximum of $1,575,000 for any single act or
failure to act that is a continuing violation.17 We plan to conduct outreach, e.g., through webinars, to
provide additional notice and an opportunity for the public to ask questions about the requirements of the
collection.
Providers and covered entities that provide Best Efforts Business Broadband Internet Access
Service that will provide the bulk of the data for the Commission’s panel regressions are likely required to
file the Form 477 and/or the Form 499-A. Accordingly, the Commission can look to previous filers of
those forms to help determine instances of non-compliance. In addition, as part of the collection, Form
477 filers that are required to report broadband connections to end user locations must affirmatively state
whether or not they fall within the scope of the collection. These mechanisms will maximize the
responses from facilities-based Providers for the analysis that will include panel regressions.
As discussed above, the Commission will use the information obtained from Purchasers to help
identify harmful, anticompetitive conduct, i.e., terms and conditions, in the sale of Dedicated Service.
The Commission recognizes that the information collected from Purchasers need not, and indeed cannot,
be comprehensive to serve this purpose.18
For Purchasers that provide communications services, we can look to previous filers of the Form
477 and Form 499-A to help identify potential instances of non-compliance and maximize the response
rate from these respondents. As discussed above, we believe that facilities-based Providers of Dedicated
Service largely fall within the universe of entities that provide broadband connections to end users and are
required to file the Form 477. Many of these entities that are required to file the Form 477 are not only
Providers but are also likely Purchasers of Dedicated Service in areas outside of their service areas to
provide seamless service to their customers. The Form 477 therefore provides us with a mechanism to
identify this subcategory of Purchasers for instances of potential non-compliance. In addition, entities
filing the Form 499-A report revenues for the provision of special access service, even when on a resale
basis, providing the Commission with another tool for identifying Purchasers. We estimate that this
group of Purchasers represents the largest Purchasers of Dedicated Service in the market.
We do not have a mechanism, however, for identifying instances of non-compliance for
Purchasers that are not required to file the Form 477 or Form 499-A, e.g., entities that merely hold
private wireless licenses for internal, non-commercial operations. Moreover, for these “other”
Purchasers, we recognize that many of them do not engage with the Commission regularly and may be
largely unaware of the collection requirement despite our planned outreach efforts. Our ability to
maximize the response rate from this group is therefore limited. Even having a limited response to the
17

47 U.S.C. § 503(b)(2)(B); see also 47 C.F.R. § 1.80(b)(2). Part 1.80(b) of the Commission's rules was recently
amended to increase penalty amounts to account for inflation. See Amendment of Section 1.80(B) of the
Commission's Rules, Adjustment of Civil Monetary Penalties to Reflect Inflation, Order, DA 13-1615 (Enf. Bureau
rel. Aug. 2, 2013); see also 78 Fed. Reg. 49370 (Aug. 14, 2013).
18

See Data Collection Implementation Order at para. 12.

6

3060-xxxx
December 2013 (rev. August 2014)
Comprehensive Market Data Collection for Interstate Special Access Services
data collection is worthwhile because it provides useful information that is not otherwise available, and
the fact that the response may be limited does not undermine the usefulness of the data collected because
it is going to be non-comprehensive under any circumstances given the size of the market and the limits
of our jurisdiction.
The Commission will create a secure Special Access Web Portal for the electronic submission of
responses. Filers will login using an FCC Registration Number (FRN) and password and download a data
container that will include record specifications for compiling data responses and software tools to verify
that data is submitted in the appropriate format. Filers will subsequently log in using its FRN and
password and electronically submit responses to the data collection. The Commission will provide
additional details on the electronic filing process in advance of the submission deadline.
4.
The standardized questions contained in the collection are based on prior questions tested by the
Commission on a voluntary basis.19 In addition, the questions include revisions and the instructions
provide potential respondents with additional clarifications based on feedback received from the public
following the release of the Special Access Data Collection Order, which contained an earlier version of
the questions.20
5.
The following individuals were consulted on statistical aspects of the design and may be
contacted regarding the methodology of the information collection:





Eric Ralph, Chief Economist, Wireline Competition Bureau, (202) 418-0771, [email protected].
Kenneth Lynch, Industry Economist, Industry Analysis and Technology Division, Wireline
Competition Bureau, (202) 418-7356, [email protected].
Jack Erb, Industry Economist, Office of Strategic Planning and Policy Analysis, (202) 418-1025,
[email protected].
William Layton, Attorney Advisor, Pricing Policy Division, Wireline Competition Bureau, (202)
418-0868, [email protected].

The survey data will be collected and analyzed by these same individuals and other Commission
staff. We do not anticipate seeking assistance from outside the agency unit for collecting and/or
analyzing the information.

19

See, e.g., Competition Data Requested in Special Access NPRM, WC Docket No. 05-25, RM-10593, Public
Notice, 26 FCC Rcd 14000 (2011) (requesting billing and mapping information). The data solicited from the public
on a voluntary basis were not subject to the Paperwork Reduction Act. See id at fn.8; 5 C.F.R. § 1320.3(h)(4).

20

See Attachment A (Revised Data Collection) and B (Instructions).

7