Before the
FEDERAL COMMUNICATIONS COMMISSION
Washington, D.C. 20554
In the Matter of )
)
Implementation of the ) CC Docket No. 96-128
Pay Telephone Reclassification )
and Compensation Provisions of the )
Telecommunications Act of 1996 )
)
SECOND REPORT AND ORDER
Adopted: October 9, 1997 Released: October 9, 1997
By the Commission: Commissioners Quello and Ness issuing separate statements.
Table of Contents
Topic Paragraph No.
I. Introduction 1
II. Background 6
III. Per-Call Compensation 16
A. The Standard for Determining Per-Call Compensation 16
B. Market-Based Compensation Analysis 29
C. Alternatives to Market-Based Compensation Rate 68
D. Per-Call Compensation Rate 111
E. Other 123IV. Procedural Matters 134
A. Paperwork Reduction Act Analysis 134
B. Final Regulatory Flexibility Act Analysis 135
V. Conclusion 165
VI. Ordering Clauses 166
Appendix A List of Parties Filing Comments
Appendix B List of Parties Filing Replies
Appendix C Rules Added
I. INTRODUCTION
1. In this order, we address the default per-call compensation rate(1) for subscriber 800
and access code calls(2) originated from payphones in light of the decision of the United States Court of
Appeals for the District of Columbia Circuit (the court) in Illinois Public Telecommunications Ass'n v.
FCC,(3) which vacated and remanded portions of the Payphone Orders.(4) In that decision, the court
concluded that the Commission did not justify adequately setting the per-call compensation rate for
subscriber 800 and access code calls at the deregulated local coin rate of $0.35,(5) because it did not justify
its conclusion that the costs of local coin calls are similar to those of subscriber 800 calls and access code
calls.(6) After seeking additional comment on this issue, we conclude in this order that the default rate for
per-call compensation of subscriber 800 and access code calls from payphones is the deregulated local coin
rate adjusted for cost differences. As discussed herein, based on our analysis of the record and the
statutory policy goals of Section 276 of the Communications Act,(7) we establish a rate of $0.284 per call
as the default per-call compensation rate for subscriber 800 and access code calls for the first two years
of per-call compensation.(8) This rate will continue to be the default rate for coinless payphones absent a
negotiated rate. Interexchange carriers (IXCs) must pay this per-call amount to payphone service providers
(PSPs) for access code and subscriber 800 calls beginning October 7, 1997, as required by the Payphone
Orders.(9) After the first two years of per-call compensation, the market-based local coin rate adjusted for
certain costs is the surrogate for the default per-call rate for subscriber 800 and access code calls.(10)
2. The compensation amount we adopt in this Second Report and Order is applicable,
as Section 276(d) provides, to "[t]he provision of public or semi-public pay telephones, the provision of
inmate telephone service in correctional institutions, and any ancillary services."(11) We previously have
declined to treat 0+ and calls from inmate payphones differently from other payphone calls,(12) and we
reaffirm that decision here. As of October 7, 1997, PSPs must be compensated for all payphone calls not
otherwise compensated pursuant to contract, including 0+ and inmate calls.
3. The immediate implementation of the rule provisions adopted herein is crucial to
the Commission's efforts to ensure fair compensation for PSPs, encourage the deployment of payphones,
and enhance competition among payphone providers, as mandated by Section 276 of the Act.(13) The
Commission's Payphone Orders require that per-call compensation for certain payphone calls begin by
October 7, 1997. To meet this obligation, we must revise those rules vacated by the court in Illinois Public
Telecomm. that relate to the implementation of a per-call compensation scheme and commence on October
7, 1997. The Report and Order, released September 20, 1996, informed parties that per-call compensation
would commence on October 7, 1997.(14) Therefore, parties affected by this rule change have had notice
since the release of that order that they would be subject to certain obligations beginning October 7, 1997.
Making this order effective immediately minimizes disruption within the payphone industry by eliminating
disputes about payment obligations and enhances the general availability of payphone services to the public.
4. This order does not address other issues vacated and remanded by the court or
otherwise alter the requirements of the Payphone Orders. Other requirements remanded in Illinois Public
Telecomm., including the compensation obligations applicable during the period from November 1996,
through October 6, 1997, will be addressed in a subsequent order in this proceeding. We tentatively
conclude in this regard that the $0.284 per-call rate we are adopting as a default rate on a going forward
basis should also govern compensation obligations during the period ending October 6, 1997, We also
tentatively conclude that PSPs are entitled to compensation for all of their access code and subscriber 800
calls during this period. We plan to address the manner in which the total payment obligation for that
period will be calculated and allocated among IXCs in a subsequent order.
5. We note that the Common Carrier Bureau (Bureau) has granted a limited waiver,
until March 9, 1998, for those payphones that cannot provide payphone-specific digits as required by the
Payphone Orders.(15) This limited waiver applies to the requirement that local exchange carriers (LECs)
provide payphone-specific coding digits to PSPs, and that PSPs provide coding digits from their payphones
before they can receive per-call compensation from IXCs for subscriber 800 and access code calls. This
limited waiver was granted by the Bureau to afford LECs, IXCs, and PSPs an extended transition period
for the provision of payphone-specific coding digits without further delaying the payment of per-call
compensation as required by Section 276 of the Act and this order. The Bureau made this limited waiver
effective immediately in order to ensure that PSPs receive per-call compensation beginning October 7,
1997.
II. BACKGROUND
6. In the Payphone Orders,(16) the Commission adopted new rules and policies
governing the payphone industry to implement Section 276 of the Act. Those rules and policies: (1)
establish a plan to ensure fair compensation for "each and every completed intrastate and interstate call
using [a] payphone[;]"(17) (2) discontinue intrastate and interstate carrier access charge service elements and
payments in effect on such date of enactment, and all intrastate and interstate payphone subsidies from
basic exchange services;(18) (3) prescribe nonstructural safeguards for Bell Operating Company ("BOC")
payphones;(19) (4) permit the BOCs to negotiate with payphone location providers on the interLATA carrier
presubscribed to their payphones;(20) (5) permit all payphone service providers to negotiate with location
providers on the intraLATA carriers that presubscribed to their payphones;(21) and (6) adopt guidelines for
use by the states in establishing public interest payphones to be located "where there would otherwise not
be a payphone[.]"(22)
7. In the Report and Order, the Commission noted that the 1996 Act erects a
"procompetitive deregulatory national framework designed to accelerate rapid private sector deployment
of advanced telecommunications and information technologies and services to all Americans by opening
all telecommunications markets to competition."(23) Thus, we sought to advance the twin goals of Section
276 of the Act of "promot[ing] competition among payphone service providers and promot[ing] the
widespread deployment of payphone services to the benefit of the general public . . . ,"(24) by eliminating
the effects of some long-standing barriers to full competition in the payphone market. To effectuate this
objective, we concluded that we would continue to regulate certain aspects of the payphone market, but
only until such time as the market evolves to erase these sources of market distortions.(25)
8. Section 276(b)(1)(A) of the Act directs the Commission to establish a plan to
ensure that all PSPs are fairly compensated for every completed call.(26) We defined "fair compensation"
as the amount to which a willing seller (i.e. PSP) and a willing buyer (i.e. customer, or IXC) would agree
for the completion of a payphone call. For certain calls, the PSP received no revenue for originating
certain calls (i.e., for subscriber 800 and other toll-free number calls) and could not block callers from
making such calls (access code calls). Based on evidence in the record, we noted in the Report and Order
that the number of these types of calls completed from payphones had proliferated in the past several
years,(27) and we concluded that PSPs must be compensated for access code, subscriber 800, and other toll-free number calls, whether they are jurisdictionally intrastate or interstate.(28)
9. In the Report and Order, we concluded that the payphone marketplace has low
entry and exit barriers and likely will become increasingly competitive,(29) and that the market generally is
best able to set the appropriate price for payphone calls, including local coin calls, in the long term.(30)
Therefore, because we have an obligation under Section 276 to ensure that the compensation for all local
coin calls is fair, we concluded that the local market should be allowed to set the price for all compensable
calls unless a state demonstrated that competition would not constrain prices; for example, payphones at
certain locations would be priced at monopoly rates. This approach is appropriate, because once PSPs are
free to enter the market, and once callers are free to choose payphones for their calls, the market ultimately
will determine whether a particular payphone is economically viable. Therefore, in the Payphone Orders,
we concluded that the appropriate per-call compensation amount, in the absence of a negotiated agreement,
ultimately is the amount the particular payphone charges for a local coin call, because the market will
determine the fair compensation rate for those calls. We further concluded that if a rate is compensatory
for local coin calls, then it is an appropriate compensation amount for other calls as well, because we found
the costs of originating various types of payphone calls such as access code and subscriber 800 calls to be
similar to the costs incurred when initiating a local coin call.(31)
10. Before we moved to a local coin call default rate, however, we found that it was
necessary to observe over time how the payphone marketplace would function in the absence of regulation.
In particular, we concluded that consumers facing time constraints may not be able to find, in certain
locations, a reasonable substitute for a payphone located on the premises. We stated that in these cases
where the location provider has an exclusive contract with a PSP, the PSP may be able to charge supra-competitive prices. The location provider would share in the resulting "locational rents" through
commissions paid by PSPs. We concluded that to the extent that market forces cannot ensure competitive
prices at such locations, we may want to continue regulating, along with the states, the provision of
payphone services generally or in particular types of locations where the size of the location or the caller's
lack of time to identify potential substitute payphones could lead to locational monopolies. To allow us
to ascertain the status of competition in the payphone marketplace, we concluded that we should establish
the default per-call rate before leaving it to the market to set the rate, absent any changes in our rules.
11. We recognized that competitive conditions, which are a prerequisite to a
deregulatory market-based approach, did not exist yet, and would not be achieved instantaneously.
Therefore, we established an interim compensation plan to ease the transition to market-based local coin
rates and ensure fair compensation for coin and noncoin calls. In particular, we established a two phase
interim plan to address coin calls. During the first year (phase) the states would be responsible for
ensuring that PSPs were fairly compensated for local coin calls as well as for protecting consumers from
excessive rates. We concluded that states could continue to set the local coin rate during the year prior
to market-based per-call compensation. During the second phase, beginning October 7, 1997, we stated
that the market would set the price for the local coin call, absent particular state concerns, and the need
for modification.(32)
12. Additionally, in the Payphone Orders, the Commission established a two-year
interim plan for payphone compensation for subscriber 800 and access code calls based on a rate of $0.35
per call that began November 7, 1996. For the first year after the effective date of the rules adopted in
this proceeding, we required that IXCs pay flat-rate compensation to PSPs. More specifically, under the
first year of the interim plan, IXCs with annual toll revenues in excess of $100 million were required to
pay, collectively, a flat-rate compensation of $45.85 per payphone per month in shares proportionate to
their share of total market long distance revenues. During the second year of the interim plan, which is
the first year of per-call compensation, all IXCs were required to pay $0.35 per subscriber 800 call or
access code call unless they contracted with the PSP to pay a different amount.(33)
13. Numerous parties filed petitions in federal court seeking review of the Payphone
Orders. In Illinois Public Telecomm, the court affirmed important parts of the Commission's rules
implementing Section 276, but also vacated and remanded certain other aspects of those rules. The court
overturned our determination in the Payphone Orders regarding: (1) the interim and permanent
compensation rates established for access code and subscriber 800 calls; (2) the requirement that only those
IXCs with annual toll revenues over $100 million pay PSPs for these calls during the first year of the
interim period; (3) the failure to provide any interim compensation to BOC PSPs for "0+" calls and calls
made from inmate payphones; and (4) the use of fair market value for payphone assets transferred from
a BOC to a separate affiliate.(34)
14. By Public Notice released August 5, 1997, we sought comment on the issues
remanded by the court.(35) We sought comment on the differences in costs to the PSP of originating
subscriber 800 and access code calls as compared to local coin calls.(36) We sought comment on whether
these potential differences in costs should affect a market based compensation amount, and if so, how.(37)
We sought comment on whether the local coin rate--subject to an offset for expenses unique to those
calls--is an appropriate per-call compensation rate for calls that are not compensated pursuant to a contract
or other arrangement, such as subscriber 800 calls and access code calls.(38) We stated that parties should
respond specifically to concerns raised by the court in setting forth their views on the appropriate per-call
compensation amount.(39)
15. This order addresses only the amount of default per-call compensation. We decline to address in this order other issues related to the implementation of the per-call compensation structure.(40) Because the court vacated and remanded the per-call compensation rate for access code and subscriber 800 calls, we have sought to act expeditiously to reevaluate the default per-call rate. We conclude, because of the exigency of the situation wherein PSPs are not receiving per-call compensation as required by Congress in Section 276, that we must address quickly and efficiently the most urgent issue -- the per call compensation amount to be paid by IXCs to PSPs beginning on October 7, 1997, the beginning of per-call compensation.
III. PER-CALL COMPENSATION
A. The Standard for Determining Per-Call Compensation
16. In the Notice, we sought comment on whether the market-based local coin
rate--subject to an offset for expenses unique to those calls--is an appropriate per-call compensation rate
for calls that are not compensated pursuant to a contract or other arrangement, such as subscriber 800 and
access code calls.(41) In Illinois Public Telecomm., the court in particular concluded that the Commission
did not adequately justify "tying the default rate [for per-call compensation] to local coin rates."(42) The
court found evidence in the record that the costs of coin calls are higher than those for coinless calls
because: (1) additional costs are incurred for equipment and coin collection; and (2) the PSP pays for
originating and terminating local calls, while for coinless calls the PSP only pays for originating the calls.(43)
Therefore, the court stated that setting the per-call compensation for subscriber 800 calls and access code
calls at the deregulated local coin rate of $0.35 was not justified, and vacated and remanded the issue to
the Commission for further consideration.(44)
1. Comments(45)
17. APCC asserts that Illinois Public Telecomm. affirms the Commission's market-based approach to determine compensation and does not mandate an analysis of costs.(46) According to
APCC, the court also affirmed the Commission's finding that the payphone marketplace is competitive,
even if market forces do not yet operate freely for dial-around calling.(47) APCC further argues that the
court did not preclude the Commission from relying on market-based surrogates, such as the local coin
rate, or require the Commission to calculate an exact cost differential to be reflected in the per-call
compensation figure.(48) The Commission, APCC asserts, could exclude consideration of cost evidence
altogether and focus solely on market price indicators.(49) APCC contends that the court objected only to
the Commission's attempt to compare the costs of dial-around calls and local coin calls.(50) Only if the
Commission continues to rely on cost comparisons as a factor in the application of a market-based
approach, must the Commission adhere to the reasoning issues raised by the court, states APCC.(51) Parties
further contend that a market-based approach will fulfill the requirements of the statute, i.e., provide rates
that "fairly compensate" PSPs and "promote competition among payphone service providers and the
widespread deployment of payphone services."(52) APCC alleges that the IXCs do not provide any
arguments for rejecting a market-based approach, and challenges the arguments that there are local
payphone provider monopolies that prevent the payphone market from being competitive.(53) Peoples adds
that PSPs are not monopoly providers because Commission rules require PSPs to unblock access code
calls, giving every caller the option to dial around a PSP's presubscribed service provider or to use a debit
card to reach a carrier of their choice.(54)
18. The Coalition argues that the court did not question the Commission's decision to
rely on market-determined prices rather than regulatory accounting procedures.(55) The Coalition asserts
that the court did not require the Commission to abandon its market-based proxies, but instead required
the Commission to consider appropriate differences, such as originating costs, between coin and coinless
calls.(56)
19. AT&T asserts that the court found that the Commission acted unlawfully in
establishing an assumed market rate for coinless calls, because the Commission ignored record evidence
on the cost differences between coin and coinless calls.(57) Because of this error, AT&T states, the court
found that there was no rational basis for the Commission's conclusion that per-call compensation should
be set at the assumed deregulated market price, and therefore, that the Commission's compensation rate
could not stand.(58)
20. Frontier similarly argues that the court did not endorse the Commission's market-based approach,(59) and further, that the court found the Commission's conclusion that the local coin rate
represents the best surrogate of the costs of completing local calls unjustified.(60)
21. Sprint asserts that although the Commission used a market-based approach to
determine local coin rates, the Commission never purported to use a market-based approach for per-call
compensation for access code and subscriber 800 calls.(61) Instead, Sprint contends that the Commission
has viewed costs as the appropriate approach from the outset, and has sought surrogates for originating
costs while rejecting non cost-based market surrogates.(62)
22. PageMart and CPI argue that the great disparity in the record between the market
rates and costs demonstrates that the payphone market is not yet competitive,(63) because price in a truly
competitive market would have been driven closer to cost.(64) PageNet argues that market rates are
misleading, because, as consumers, IXCs cannot decline a sale, i.e., block incoming payphone calls, and
thus have a weakened market power.(65) WorldCom asserts that market-based rate would be more arbitrary
and artificial than rates based on objective and verifiable costs.(66)
2. Discussion
23. Despite a careful review, we find no statement in the court's decision that precludes us from relying on market-based surrogates, or requires us to determine a rate based on cost data submitted by incumbent LECs, independent PSPs, and other parties to determine the new per-call rate. The court did not reject the concept of linking the market-based local coin rate to the per-call rate for access code and subscriber 800 calls based on the similarity in costs, nor conclude that our approach was irrational. Rather, the court concluded that the Commission had not responded to information on the record regarding the cost disparities between the cost of providing coin calls and subscriber 800 and access code calls. Therefore, the court concluded that adoption of the default rate without further explanation was arbitrary and capricious.(67)
24. The 1996 Act does not prescribe a particular course to ensure that all PSPs are
fairly compensated for each and every call.(68) Nothing on the record in response to the Notice persuades
us to change the deregulatory scheme established in the Payphone Orders. Based on the record in this
proceeding, we affirm our decision in the Payphone Orders to use a market-based default rate for per-call
compensation for subscriber 800 and access code calls. We conclude for the reasons stated there that a
market-based rate best responds to the competitive marketplace for payphones consistent with the
deregulatory scheme we adopted in the Payphone Orders for the provision of payphone services pursuant
to Section 276, and also will effectively advance the statutory goals of encouraging competition and
promoting the deployment of payphones.
25. As discussed above, because of market imperfections such as the inability of PSPs
to block access code and subscriber 800 calls, we concluded in the Payphone Orders that a default rate was
necessary to ensure that PSPs received fair compensation during the transition to a deregulated market.
We also concluded in those orders, as we conclude here, that the default rate should be market-based. The
method we use in this order to estimate a reasonable default per-call compensation rate addresses the
court's concerns as well as those raised on the record in response to the Notice by LECs, IXCs, and PSPs.
Specifically, our approach continues to rely on a market-based rate (the local coin rate).
26. We, however, adjust the market-based local coin rate for differences in the costs
of coin and coinless operation, reducing the market-based local coin rate for coin-related costs and
increasing the market-based local coin rate to reflect costs that are related to access code and subscriber
800 calls. In addition, in response to the arguments of parties in this proceeding that a market-based rate
would be unreasonable and that we must establish a rate based on cost data submitted by the parties, we
also have performed an analysis of those cost data to test the reasonableness of the selected per-call market-based rate. As discussed below, we find based on this analysis that the adjusted market-based rate is
reasonable. Accordingly, we conclude that the deregulated local coin rate, adjusted for cost
considerations, is a reasonable market-based surrogate for determining the default per-call compensation
rate and specifically responds to the court's concerns that cost differences between coin calls and coinless
access and subscriber 800 calls be explained. Furthermore, we conclude that the per-call rate established
in this order will further the goals of Section 276 and is in the public interest.
27. The record on remand supports our prior conclusion that per-call compensation
should be set by the marketplace and that full and unfettered competition is the best mechanism to achieve
Congress' dual policy objectives.(69) Competition over time will lead to the more efficient placement of
payphones, improved payphone service, and lower prices for consumers. To encourage competition in
the payphone marketplace, we ensure in this Second Report and Order that PSPs are fairly compensated
for "each and every completed intrastate and interstate call."
28. We conclude that because we make the per-call amount subject to negotiations, the
marketplace will make the appropriate adjustments in the per-call rate. We established the per-call default
rate to be applied only if the PSP and the IXC are unable to negotiate some other rate of compensation for
compensable calls. Negotiations may lead to rates other than the default rate for several reasons. First,
because virtually all of the costs are fixed costs and are not incurred on a per-call basis, an IXC and a PSP
might agree to a flat-rated charge rather than a usage-based compensation rate. Second, there may be
locations where a payphone would not be viable financially if compensated at only the default rate per
compensable call, but would be viable at a higher compensation rate. If an IXC found it profitable to carry
calls at this higher rate, it would be in the mutual interest of the two parties to agree on a higher rate.
Third, IXCs may choose to pass on the per-call compensation rate to their customers. In the case of 800
subscriber calls, the IXC could pass on the cost to the called party. If the called party refused to accept
calls for which it was charged the default rate, but was willing to accept calls with a lower charge, the IXC
and the PSP may find it in their mutual interest to negotiate a per-call rate lower than the default rate.
Fourth, in locations where a competing payphone could be placed without the permission of the location
provider, a PSP may be willing to negotiate a lower rate than the default rate, rather than give an IXC the
incentive to place a competing payphone.
B. Market-Based Analysis
29. As discussed above, we conclude that the appropriate rate of per-call compensation for access code and subscriber 800 calls is the market-based local coin rate adjusted for costs. In setting the per-call compensation rate for the first year of per-call compensation, we begin with the $0.35 market-based local coin rate established in the Payphone Orders and adjust that rate to remove coin-related costs and add costs specific to subscriber 800 and access code calls.
1. Comments
30. Market Rate. APCC, the Coalition, Peoples, and CCI request that the Commission adopt a market-based per-call compensation rate, and furthermore, assert that the underlying costs attributable to both coin and noncoin calls are similar.(70) APCC contends that any market-based rate-setting mistakes are self-corrective, because the market will demonstrate the mistake.(71) APCC further contends that contrary to the IXCs position, the market will prevent PSPs from gaining any long term windfall, and would force any such "windfall," to be passed on to consumers.(72) APCC contends that market-based rates are more objective than the subjective components of cost-based rates.(73)
31. The Coalition further maintains that the market will reflect variations from region
to region and payphone to payphone.(74) The Coalition urges that the market rate be the local coin rate
adjusted to reflect the relative elasticities of demand of the various types of calls.(75) The Coalition contends
that under market conditions sellers will tend to load costs onto services for which prices are less likely
to fluctuate, i.e., that have a lower elasticity of demand, than onto services that have a higher price
sensitivity. The Coalition further argues that the elasticity of demand for local coin calls is higher than for
long distance calls. In other words, the Coalition argues, customers of local calls will respond more
quickly to price changes than customers of 0+, subscriber 800 and dial-around calls.(76) Thus, the Coalition
contends, the price of long distance calls should be the local call rate adjusted upward to reflect the lower
elasticity of demand and the greater proportion of costs, relative to local calls, that such calls will carry
under true market conditions.(77)
32. CCI, an independent payphone provider, argues that the Commission should adopt
a market-based surrogate, and contends that there are few differences between the costs of a local coin call
and a subscriber 800 or access code call.(78) CCI argues, however, that even under a cost-based approach,
the cost of a local coin call and a dial around call is approximately $0.35.(79)
33. Several of the IXCs assert that the retail price for local coin calls is not an
appropriate surrogate for the costs of a noncoin call, because there are substantial cost differences between
these two types of calls.(80) AT&T and MCI assert that if the Commission develops a rate based on an offset
from the local coin rate, the offset should be at least fifty percent,(81) or based on the rate negotiated between
AT&T and APCC in 1994 for dial-around access code calls.(82) MCI asserts that a market-based rate, being
higher than a cost-based rate, would lead to increased blocking by 800 subscribers, as those subscribers
try to avoid having to pay IXCs for unduly high payphone charges.(83) MCI also asserts that market-based
rates are artificially driven up by location owners holding out for the highest bidding PSP.(84) These higher,
market-based rates will lead to an unwarranted income transfer from consumers to payphone providers,
MCI contends, because excessively high rates will encourage PSPs to place payphones in increasingly
marginal locations.(85) The Coalition disputes MCI's assertion that a market-based rate would lead to
increased blocking arguing that PSPs have an interest in seeing calls completed, which call blocking would
defeat, and an acceptable market rate would result in more completed calls.(86)
34. Local Coin Rate as Surrogate. Several of the PSPs argue that if the local coin calling rate is used, no significant adjustment for cost differences between the coin rate and dial-around calls is required, because any cost differences are minimal.(87)
35. Peoples argues that a single, flat default rate would simplify procedures, much as
a first-class postage stamp covers mail that goes various distances.(88) Peoples further argues that the local
coin rate is such a flat rate, because it is used to originate all types of calls from a payphone.(89) Moreover,
Peoples argues, coinless calls alone do not justify installing a payphone; payphones are installed for coin
calls, thus, the local coin rate is a good market measure for all of the calls that originate from it.(90)
36. Several of the IXCs oppose the use of the local coin rate as a surrogate, but state that if the Commission uses the local coin rate, then the Commission should reduce the local coin rate so that it reflect only expenses unique to access code and subscriber 800 calls.(91) CPI objects to the use of the local coin rate as a starting point because the coin rate does not represent the result of a competitive market.(92) TRA says that using the local coin rate will lead to a grossly inflated default rate.(93) Frontier states that the coin rate bears little relationship to the costs of completing a coin call, much less a coinless call.(94)
37. Other Surrogates. APCC requests that the Commission consider other surrogates
for the market rate, such as 0+ commissions, 0- transfer rates and sent-paid toll call surcharges.(95)
According to APCC, the 0+ call commissions are the only known instance where carriers and PSPs meet
in the marketplace to negotiate a price for routing a call from the payphone to the carrier, and therefore,
the Commission should reconsider 0+ commissions.(96) APCC further contends that sent-paid tolls are
another reasonable indicator of the market price.(97) Additionally, APCC contends that the 0- transfer rates
are a reasonable surrogate, because these rates indicate the minimum price IXCs are willing to pay to
obtain telephone traffic.(98) APCC concludes that the most appropriate market-based surrogates are local
coin calls, operator-assisted call commissions and sent-paid toll surcharges, because these three surrogates
are based on prices actually charged in the marketplace for origination of payphone calls. APCC states
that a weighted average price for these three charges is $0.45 per call.(99)
38. Several of the IXCs argue that 0+ commissions cannot be used as a market guide
because these commissions include factors unrelated to the use of payphones for the use of access code and
subscribers 800 calls.(100) Furthermore, carriers argue, sent-paid calls are not a reliable surrogate, because
these charges cover such services as a payphone's capability to track time and amount, and recognize types
of coins, services not needed for 800 subscriber calls.(101) MCI argues that these surrogates are not
representative because they are narrowly tailored to specific types of calls.(102) Moreover, MCI contends,
some of so-called surrogates apply to calls from telephones that are not even payphones.(103) Sprint argues
that the only truly reliable indicator of the market for subscriber 800 and access code calls is what the
market provided to PSPs for such calls prior to the imposition of the Commission's orders in CC Docket
No. 91-35.(104) At that time there was no compensation to PSPs for these calls, and therefore, Sprint
contends, the market price was zero.(105)
39. Excel argues that the Commission should start with a local coin rate at $0.25,(106) then subtract those costs unique to the local coin service coin equipment and collection, coin rating, originating and terminating access from the local coin rate.(107) AT&T, CompTel, and CWI argue that the Commission should not rely on avoided costs in establishing the default compensation rate, because this method inappropriately compares the price of coin calls with the costs of coinless calls and may overcompensate PSPs. Nonetheless, if the Commission adopts this method, AT&T argues, the Commission must set the local coin rate at $0.25 and determine the actual avoided costs related to coinless calls,(108) and CompTel and CWI argue that the Commission should subtract the costs of tracking and billing compensation.(109) MCI argues that if the Commission adopts a top-down approach, it should calculate the default rate by subtracting the coin specific costs from the cost of a coin call, not from the market rate.(110) RCN argues that the Commission should determine a nationwide default rate and then subtract those costs that are unique to coin calls.(111)
40. The Coalition argues that the avoided cost methodology will not produce a per-call compensation rate lower than the deregulated coin rate, and in fact, will increase the amount of compensation owed to the PSPs.(112) Furthermore, the Coalition argues, avoided cost methodology will not produce competitive outcomes, because joint and common costs are a significant portion of the total costs, and the market does not price goods or services on costs alone.(113)
2. Discussion
41. In the Payphone Orders, we found that the market rate for a local coin call is $0.35
and we stated that this is also the rate for access code and subscriber 800 calls for the first two years of
per-call compensation. In response to the court's concern that there may be differences in cost in between
providing local coin calls and subscriber 800 and access code calls, we have evaluated the evidence on the
record to develop a default rate for access code and subscriber 800 calls that reflect those cost differences.
On the record, parties discuss several cost factors suggesting that compensation for access code and
subscriber 800 calls should be either above or below the market price for coin calls.(114) In section (a) we
conclude that based on differences in costs, a market rate for access code and subscriber 800 access calls
likely would be between 5.9 and 7.3 cents lower than the market rate for a local coin call, resulting in a
rate of $0.284. In section (b) we conclude that the parties failed to provide sufficient information to adjust
the default dial access and subscriber 800 rate to reflect differences in the elasticities of access code and
subscriber 800 calls compared with local coin service. Thus, we do not make any adjustment for elasticity
differences.
a. Adjustments to the local coin market rate based on cost differences
i. General approach
42. Our general approach is to start with the market rate for local coin service ($0.35),
and subtract costs directly attributable to coin calls and add costs specific to access code and subscriber
800 calls. The majority of the costs associated with a payphone are joint and common costs that are shared
by the different types of calls made by means of the payphone. These costs do not increase or decrease
as the number or composition of calls changes at a particular location. By making no adjustment to the
coin rate for these costs, we conclude that each call placed at a payphone should bear an equal share of
joint and common costs.
43. The long distance and paging companies argue that we should limit the costs
attributed to access code and subscriber 800 calls to the costs that would be incurred from providing access
at a coinless payphone; coin-related costs should not be included. Under this theory, all other costs that
are incurred to support a payphone coin call would be attributed to coin calls and either removed from any
market-based rate or excluded from any other type of cost estimate.(115) PSPs, however, maintain that few
locations could support a coinless instrument.(116) Instead, they explain that most payphones are installed
to handle both coin and coinless calls.(117)
44. We agree with the IXCs, and paging companies, that costs directly associated with
the coin mechanism should be borne by coin calls. Under their general approach, however, compensation
for subscriber 800 and access code calls would not fairly contribute to the recovery of joint and common
costs of payphone service that would occur, even if the payphone is used solely to place such calls. In our
view, such joint and common costs are not "additional" costs occurred to provide local coin calls. Hence,
compensation for subscriber 800 and access code calls should contribute to the recovery of such costs.
Our calculation assumes that each call will contribute to a multi-use payphone's joint and common costs.
45. We reject AT&T's contention that using a coinless payphone results in a per-call compensation rate of 11 cents per call and that this rate should be the basis for selecting a per-call compensation rate. We note that AT&T divided its monthly costs to install, operate, and maintain a coinless payphone ($76.85) by the number of calls at a coin payphone estimated by APCC.(118) The APCC study showed that the average payphone carried 713 calls per month, and that 511 of these calls were coin calls and 202 of these calls were coin-less calls.(119) It is more reasonable to assume that you would divide AT&T's estimated monthly costs for a coinless payphone ($76.85) by 202, the number of coinless calls. This calculation results in a cost of 38 cents per call, rather than the 11 cents estimated by AT&T. If the number of calls at coinless payphone were adjusted for a marginal location as we do in our analysis above, the per-call cost would be even greater. Thus, we conclude that the 11 cent rate obtained by AT&T in its analysis would not be an appropriate per-call compensation rate for subscriber 800 and access code calls.(120)
46. Selecting the number of calls to represent a low traffic location. Any analysis of
the costs incurred for a call from a payphone must be based on a particular number of calls. Most of the
parties presented cost information based on coin payphones serving locations with an average amount of
calling. We believe, however, that it is appropriate to analyze cost for a location with less than average
calling. Prices in competitive markets tend to be set at the marginal cost of production. For payphone
service, the marginal unit of production is the installation of a payphone at a low traffic location. If prices
for payphone calls increased, providers would be willing to install more payphones; however, customers
would likely place fewer calls. At the equilibrium price for payphone calls, newly installed payphones
would be expected to generate just sufficient calls to earn only a normal return on investment. Thus, we
believe that setting a default compensation rate to achieve fair and reasonable compensation requires that
a payphone operator be able to cover costs at a low traffic location. A single instrument would be required
to provide both coin and coinless calls at such a location, with neither class of calls, by itself, sufficient
to justify installation of a payphone.
47. We select the number of calls to represent a low traffic location by estimating the
number of calls that could cover all of the costs of operating a payphone with the exception of commissions
paid to location owners. This number represents the lowest number of calls at which a payphone could
be operated without requiring a subsidy. Most of the costs associated with a payphone do not vary with
the number of calls made at an individual payphone. Thus an individual call must cover its own marginal
costs as well as a share of the non-varying costs. The contribution made by an individual call is the price
of the call less the marginal costs of the call. If the price of calls remains constant, each additional call
adds a fixed amount of contribution. If the number of calls is high enough, the total of this contribution
will exceed the total of non-varying costs, including a normal return on investment. The amount by which
total revenue exceeds total cost is referred to as economic rent. In the long run, premises owners will be
able to extract any economic rent from payphone owners through commissions.(121) If a location generates
only enough traffic to support the installation and upkeep of a payphone, however, there will not be any
commission payments. Some PSPs may choose to pay standardized commission amounts.(122) These
companies will not serve as wide a mix of locations. All things being equal, the owner of a high traffic
location would seek out the potential profits by choosing the PSP that is willing to pay the highest
commissions. On the other hand, if the owner of a low traffic location insisted on a commission, no PSP
would be willing to install a new payphone at that location because no PSP could pay the commission and
generate a sufficient return on its new investment.(123) Accordingly, a marginal location is a location where
traffic just covers costs other than premises owner commissions.
48. Based on the data provided by the commenters, it is necessary to complete several steps to determine the appropriate number of calls needed to sustain a payphone at a marginal location. As explained more thoroughly below, we rely on APCC cost data, because these data are representative of the payphone industry as a whole. However, APCC did not provide a breakdown of the 689 calls that it reported as the average per payphone when it collected the cost data. Therefore, we first used APCC data from the call type study--which provided data based on an average of 713 calls--to determine the proportion of access code and subscriber 800, coin and other calls for the 689 calls reported in the cost study. Second, using these derived call numbers, we estimated the amount of coin and other calls necessary to generate commission payments, and subtract those calls to yield the number of calls needed to sustain the marginal payphone.
49. We use APCC data to estimate the number of calls per month that an average PSP
would need at a location to cover costs other than commissions.(124) APCC reported $240 monthly cost per
payphone, including $45 in commissions, based on an average of 689 calls of all types.(125) Until October
1996, $6 of the monthly cost per payphone was met from dial around compensation and the balance of the
monthly cost per payphone had to be met with coin revenues and revenues from 0+, 0-, and 00- calls.(126)
To determine the amount of revenue that the average coin, 0+, 0-, and 00- call had to produce so that the
average number of calls would cover total costs, we had to determine the total number of each such call
type. Therefore, we used the data in the APCC call distribution study, which produced a total of 713 calls
of all call types--154 access code and subscriber 800 calls and 561 coin and other calls--and applied this
breakdown to the 689 calls in the cost study to develop a call distribution. Applying the representative
percentages of the call types resulted in the following distribution: 147 access code and subscriber 800
calls, 494 coin calls, and 48 other calls.(127) Thus, to recover the $240 in monthly costs at an average
location, the PSPs surveyed by APCC had to collect an average of 43.5 cents per call in revenue from coin
and other calls.(128)
50. The APCC data illustrate that PSPs pay an average of $45 per month in commissions. For the purposes of this analysis, we impute the number of calls at a low traffic location by taking the number of calls at an average location, and subtract the number of coin and other calls that would produce marginal revenue of $45. As explained above, to break even at an average location, PSPs must have generated 43.5 cents per call from an average number of coin and other calls. This revenue per call, however, is offset by about 4.7 cents of marginal cost per call,(129) meaning that payphone providers must realize about 38.8 cents in average net revenue per call. Dividing $45, the average compensation to premises owners, by 38.8 cents, which is the marginal revenue per call, results in 116 coin and other calls. In other words, if the number of coin and other calls is decreased by 116, all other things being equal, the PSP's net revenue would be reduced by $45 (116 calls times 38.8 cents per call). Assuming a proportionate reduction in all calls, a break even or low traffic location would have 116 fewer coin and other calls and 31 fewer access code and subscriber 800 calls.(130) Using the total number of all calls from the cost study (689), we subtracted 116--the number of coin and other calls that would generate $45 in commissions. This resulted in 573 calls. We also expect that the number of access code and subscriber 800 calls at a marginal payphone location would be less. As noted above, we determined that 147 of the 689 calls at an average location would be subscriber 800 and access code calls. To reduce that amount (147) by the decrease in access code and subscriber 800 calls that would be originated at a marginal location, we then determined how many of the remaining calls were subscriber 800 and access code calls. Comparing the numbers from the APCC call volume study, we determined that the number of coin and other calls (excluding subscriber 800 and access code calls) was approximately 20.68% less in the cost study.(131) Assuming that the subscriber 800 and access code calls also would decrease proportionately, we determined that there would be 31 fewer subscriber 800 and access code calls.(132) Thus, we subtracted 31 from 573, which results in 542 calls. Accordingly, we use this number, 542, as the total number of calls that would be made from a low traffic location.(133)
ii. Estimate of avoided and added costs.
51. The parties submitted data on avoided and added costs of dial access and subscriber 800 calls compared with local coin calls. Different parties have different costs by category due to differences in the type of location served and differences in accounting treatments. Line charges, for example, vary from state to state. One party may treat a specific cost as overhead while another party might include the same sort of cost a direct cost of maintenance. It is not possible to fully reconcile differences in cost estimates by analyzing the data filed on the record. Accordingly, we have used the information submitted by the parties along with information from Securities and Exchange Commission 10K filings to develop ranges within which cost for an average PSP might reasonably be expected to fall.(134)
52. Coin Mechanism Capital Costs. While a single payphone may be installed to
handle both coin and coinless traffic, the direct costs of the coin mechanism should be recovered by coin
calls. After installation, the capital costs of a payphone become fixed. Because we are looking at the long
run, where all costs are avoidable, we consider the decision made by the PSP at the time the phone is
installed. When a payphone provider considers installing a telephone at a new location, it must consider
whether the additional coin traffic at that location would justify the additional cost of installing a coin
telephone. The PSP would not install a coin payphone instead of a coinless payphone unless the additional
coin traffic would at least cover the additional costs of a coin mechanism. Therefore we conclude that
costs directly associated with the coin mechanism should be attributed to coin traffic. We assume that the
market rate for local coin calls recovers these costs and therefore conclude these costs should be removed
from the adjusted market rate.
53. David Robinson, in a study submitted by AT&T, provided the most detailed
information on the costs of purchasing and installing different types of telephones. Independent PSPs
typically use smart payphones. Robinson estimated that new smart coin payphones cost about $900 to
$1200 per unit compared with $200 to $250 per unit for coinless units.(135) The differences in cost are
primarily due to equipment used to accept, count, and hold coins.(136) Some cost differences, however, may
be due to quality features that allow the payphone to be used in harsher environments. We selected the
$900 figure for smart coin telephones as an amount that would be suitable for general locations instead of
the $1200 figure, because the latter figure, likely included additional features that go beyond the standard
smart coin telephone that would not be necessary at the general location. We determine that $250 is an
appropriate amount for the coinless phone operated in a general location, to reflect some quality features,
and further, because there is not a significant difference in the capabilities among the coinless phones and
the difference between the estimates ($200 to $250) is not significant. The difference in price, from $900
to $250, $650 per telephone, would be due to added costs associated with coin traffic. Robinson also
estimates that a smart coin telephone requires $60 more for installation than does a coinless telephone due
to additional testing and programming for the coin rating and collection functions.(137) Thus, we estimate
a total investment cost of $710 per payphone that is related to coin functions.(138) This equates to $12.36
in investment costs per month for a coin telephone.(139) Thus, we impute that the market rate for local coin
service includes 3.2 cents per coin call at a low usage location and that this amount represents an avoided
cost for dial around and subscriber 800 calls.(140)
54. Line Savings. In some areas, all payphones are charged per-message or per minute
charges for all local calls. In other areas, all payphones use unmeasured lines. In still other areas,
payphone providers can choose between using some form of measured service and unlimited calling. PSPs
taking measured service pay message charges for local coin calls, but not for access code or 800 subscriber
calls. This represents a marginal cost difference of coin versus coinless service. Based on the record, we
conclude that the average cost savings for line charges is about 2.5 to 3.0 cents per call.(141)
55. Collection and Maintenance Savings. The parties concur that coin collection costs
are related to coin calls, that coin telephones have higher maintenance costs than coinless telephones and
that maintenance costs increase as the number of coin calls increases.(142) It is difficult to separate
maintenance from coin collection costs, however, because some coin collection and routine maintenance
may occur at the same time.(143) Not all maintenance is related to coin calls.(144) For example, key pads and
handsets are used for both coin and non-coin calls and vandalism may be directed against the phone or the
enclosure as well as targeted against the coin box. Based on the record, we conclude that the average
savings from coin collection and maintenance is 2.1 to 3.0 cents per call.(145)
56. Bad Debt / Collection Charges. Peoples identifies some collection and bad debt expenses that it attributes solely to compensation for access code and subscriber 800 calls. Under the interim compensation plan, Peoples was unable to collect from IXCs approximately $4.50 per payphone per month, which translates to $0.03 per access code and subscriber 800 call.(146) Conversely, CompTel alleges that Peoples' bad debt expenses arose primarily from operator service operations.(147) CWI opposes including any allowance for increased collection costs of access calls, arguing this is not a cost of access and that the IXCs also bear such costs.(148) Furthermore, AT&T notes that collection costs should decrease steadily with the implementation of ANI and other Commission requirements.(149) CWI and CompTel contend that per-call compensation should not include billing or bad debt costs.(150) Neither the Coalition nor the other PSPs included specific estimates of increased collection and bad debts. As such, we do not have sufficient information to attribute an amount to bad debt and/or collection charges.
57. ANI ii. The Commission's rules require that LECs provide certain automatic number identification information (ANI ii) to the IXC with each call. These digits provide IXC's with automated information that enables them to bill, block, and track calls. On the record, the parties disagree about the costs associated with the provision of ANI ii digits, and further, who should bear those costs.(151) USTA estimated the cost of providing ANI ii digits through hardcoding and through FLEX ANI. The estimated total capital cost for hard coding the digits was about $1.035 billion of which $559 million was for upgrading all non-equal access switches and $476 million was for hard coding switches.(152) Sprint notes that the USTA figure assumes equipment upgrades for every non-equal access switch, while many of these switches do not support any payphones.(153) Given that not all non-equal access switches would be upgraded, and that the upgrade would benefit all users of the switches, it seems unlikely that all the upgrade expense would be attributed to payphone service. For the purpose of translating the USTA cost estimates into additional pay telephone costs, we assume that $600 million of additional LEC investment would be recovered from increased payphone line rates. $600 million in increased investment recovered over 10 years would require increased monthly line charges of $5.50.(154) Divided by the low traffic location number of calls, 542, would equal approximately 1 cent per call.
58. AT&T notes that less expensive alternatives to the plan advanced by USTA exist.(155)
The Coalition indicates that if LECs are allowed to use a combination of FLEX ANI or original line
screening technology, payphone digit identification costs may be as low as $0.01 per call.(156) As discussed
above, we have evaluated the data supplied by the USTA, the Coalition, AT&T, and Sprint, and we
estimate a cost of $0.01 per call.
59. Interest. Several payphone providers note that they have the use of coin receipts
almost immediately while they must wait to collect compensation on access calls.(157) Peoples, for example,
collected payphone compensation for access calls completed between October 8 and December 31, 1996
in April 1997.(158) Accordingly, we conclude that the delay in receipt of compensation for access calls
represents an additional cost of providing access code and subscriber 800 service calls that would not be
included in the market rate for local coin calls.
60. AT&T uses 11.25% as the interest rate and the return requirement for payphone
investment.(159) APCC claims that the appropriate interest rate for many payphone providers would exceed
that rate significantly.(160) Peoples used a 10% interest rate in its calculations.(161) Most payphones, however,
are owned by large local exchange carriers, whose authorized interstate rate of return has been 11.25%
representing a weighted average of debt and equity costs.(162) Accordingly, we conclude that 11.25% is the
appropriate cost of capital for payphone providers in this context. Thus, the delayed receipt of
compensation for access code and subscriber 800 calls justifies an upward adjustment of .8 cents (11.25%
for 3 months times the market rate adjusted for other costs).
61. Opportunity Costs. Teleport contends that the Commission should recognize the opportunity costs associated with use of a payphone for non-coin calls.(163) This cost theoretically arises because the payphone provider does not have the opportunity to realize coin or 0+ commission revenue whenever its payphone is being utilized for an access code or subscriber 800 call. Sprint, however, notes that the payphone will be available for 0+ and coin calls 98.2% of the time based on average amounts of access code and subscriber 800 calling. Sprint also states that when a given phone is not available, another phone from the same company may be available, so the call is not necessarily lost.(164) Therefore, we make no adjustments to the local coin rate based on opportunity costs.
62. Commissions. Several IXCs argue that commissions paid to location owners on
0+ and 1+ calls should not be attributed to per-call compensation rate.(165) CompTel argues that these
commissions have been paid on 0+, 1+, and local calls, and recovered through these revenues. CompTel
and RCN argue that there is no assurance that these commissions are just and reasonable.(166) WorldCom
argues that 0+ commissions should not be included as a cost in computing per-call compensation because
these commissions reflect the value of being selected as the default 0+ provider and as such are unrelated
to the costs of providing subscriber 800 and access code calls. The Coalition and the independent PSPs
propose that per-call compensation default be set on the basis of the average commission received by
independent payphone providers on 0+ calls to set the rate for access code and subscriber 800 calls.(167)
CompTel and RCN argue that there is no assurance that these commissions are just and reasonable.(168)
Accordingly, we do not need to make any adjustments to reflect commission costs.
63. Total Adjustments to Market-Based Rate. The preceding analysis suggests that
costs associated with coin equipment, line, coin collection and maintenance are not directly attributable to
provision of access code or subscriber 800 call. We estimate that in total, between 7.7 cents and 9.1 cents
per call are directly attributable to local coin calls, and thus should be subtracted from the market rate.
There are uncertainties with the estimates but we found no evidence to suggest a preponderance of either
high or low biases. On the other hand, we adjust the local coin market rate upward by 1 cent to account
for additional costs to PSPs resulting from ANI ii implementation to identify payphone originated calls for
the benefit of IXCs, and 0.8 cents for interest attributable to the delay in compensation for access code and
subscriber 800 calls. These additions and subtractions produce an adjusted market-based range of $0.277
to $0.291. The midpoint of that range is $0.284. Thus, we conclude that the surrogate or adjusted market
default price is $0.284 per access code and subscriber 800 call.
b. Adjustments to the Local Coin Market-Rate Based on Demand Differences
64. The Coalition filed a study by Dr. Hausman that adjusts the local coin market rate
for differences in demand. Dr. Hausman explains that in an industry with a significant amount of joint and
common costs, competitive firms take into account demand conditions and competitive conditions as well
as costs when setting price.(169) A competitive firm recovers joint and common costs through markups over
marginal costs. Dr. Hausman states that the markups are set so that the firms recover total costs. Dr.
Hausman then asserts that services, where the demand is relatively price elastic, compared to other services
provided over the joint facility, would receive lower markups.(170) Dr. Hausman uses several methods to
translate relative elasticities into relative prices for coin calls versus access code and subscriber 800 calls.(171)
Dr. Hausman uses derived elasticities to show that access code and subscriber 800 services are less elastic
than local coin calling.(172) His analysis concludes that the Commission should set the default compensation
rate at the local coin rate plus approximately $0.07 to $0.08 per call.(173)
65. AT&T replies with a study by Dr. Warren-Boulton, who contends that the derived
elasticities presented by Dr. Hausman significantly underestimate true elasticities. Dr. Warren-Boulten
notes that customers faced with a $0.35 increase in toll rates at payphones likely would substitute toll
services that did not increase in price, rather than simply deciding not to make the calls.(174) This view is
supported by MCI's comment that many 800 customers are interested in blocking subscriber 800 calls from
payphones to avoid paying the compensation charge.(175) MCI, however, suggests that the demand for coin
calls is significantly less elastic than Dr. Hausman suggests.(176) These customers may anticipate that at least
some potential callers subsequently would make a subscriber 800 call from another location.
66. Dr. Hausman's derived elasticities are sensitive to several of his underlying
assumptions. He based the average price of an access code call on historic AT&T data. These data
probably overstate the current average price for an access code call because many firms exclusively operate
by providing prepaid calling cards, which do not include a surcharge,(177) and because there have been
significant decreases in some interstate and international toll rates. Furthermore, Dr. Hausman uses the
overall toll elasticity as the elasticity for dial around access calls. Customers placing access code calls,
as opposed to 0+, 0-, and 00- calls, have already made choices based on perceived price differences.(178)
These customers therefore may be much more price sensitive than average toll customers, and may be far
more willing to forego or delay calls than indicated by Hausman's derived elasticity. We conclude that
the demand for access code and subscriber 800 calls are significantly more responsive to price than Dr.
Hausman suggests.
67. We conclude that while differences in demand elasticities for access may prove
useful to some firms in setting prices, the information presented in the current record evidences wide
variations in assumed elasticities and the results are inadequate to determine whether access code and
subscriber 800 service or local coin service is the more price elastic service. Because we do not have
confidence in the elasticity analyses in the record given the variation in results, we decline to adjust the
market-based default per-call compensation rate for differences in demand.
C. Alternatives to a Market-Based Compensation Rate
68. As noted above, some commenters request that we establish the default per-call
compensation rate based on cost information filed by the parties in this proceeding. We decline to adopt
this approach, but we have assessed the record evidence on this matter and have calculated a cost-based
default rate below to validate that our market-based adjusted per-call rate is reasonable.(179)
1. Comments
a. Costing Methodologies
69. Several of the commenters argue that the Commission should derive a
compensation rate based on the costs that are incurred to originate coinless calls.(180) Several of the IXCs
request that the Commission adopt a bottom-up methodology to calculate per-call compensation.(181) AT&T
argues that a rate computed in this manner will be sufficient to provide for the widespread deployment of
payphones, and would not require the Commission to engage in lengthy cost proceedings.(182) AT&T argues
that its analysis is based on TELRIC, which, AT&T argues, is the most appropriate methodology in the
circumstances. Borden, Champion, and Sitel(183) argue that the fair compensation rate must be based on a
PSP's actual costs for handling 800 calling card calls. SDN supports a national rate based on verifiable
long range incremental costs for all PSPs. Excel argues that the Commission should adopt a rate that
reflects the actual costs incurred by an efficient PSP for delivering subscriber 800 and access code calls.(184)
70. CompTel and ITA argue that the Commission should base compensation for subscriber 800 and access code calls on the PSPs' incremental cost of originating these calls.(185) ITA contends that the Commission should use the cost of a payphone call as determined by Massachusetts Department of Public Utilities (Massachusetts DPU) and adjust that number downward.(186) Sprint and AT&T also argue that the Commission should use the coin rate filed by New England Telephone (NET) with the Massachusetts DPU indicating a per-call local coin rate of $0.167 as the point at which we should begin our analysis of a rate adjusted for costs related to coin calls.(187) The Coalition argues, however, that this cost study is not an appropriate basis for establishing per-call rate in this proceeding.(188) CWI, LCI, CompTel, and Sprint argue that the incremental costs to be included are the additional or marginal costs created by access code and subscriber 800 calls--additional maintenance and wear and tear for increased usage, and the per minute usage charges, if any, imposed by a LEC for originating access code or subscriber 800 calls.(189)
71. Alternatively, Sprint argues that if the Commission takes a fully allocated approach
to costs, then the rate should be based on the most efficient "bellwether" PSP's costs minus costs related
to coin functionality, local call completion and premises owner commissions from a local coin call.(190)
Sprint rejects Dr. Hausman's view that costs of the least efficient (or marginal) provider should be used
as the default rate to prevent the removal of payphones, arguing that this approach overlooks the
Commission's policy that inefficiency should not be rewarded in a multiprovider market and that rates
should be based on the costs of an efficient provider to promote competition.(191) The Coalition and APCC
contend that Sprint's "bellwether" approach is flawed, because large, fixed joint and common costs that
should be included as costs, were omitted;(192) relying on incremental costs only is inappropriate because
the PSP cannot recover the total costs of providing the service;(193) and cost estimates for a single state are
not representative.(194)
72. TRA and WorldCom argue that the Commission should apply total service long
term incremental costs (TSLRIC) principles to determine forward looking costs on efficient provider would
incur to provide access to noncoin calls.(195) CompTel, CWI, and LCI argue in the alternative that if the
Commission wants access code and subscriber 800 calls to bear some of the costs to ensure that PSPs are
fairly compensated, then the Commission should set the compensation rate based on forward looking direct
costs for access code and subscriber 800 calls.(196) Frontier and RCN argue that the Commission should
adopt a cost-based rate based on the costs of completing subscriber 800 and access code calls.(197) GCI
argues that PSPs should be compensated solely for the costs of subscriber 800 and access code calls.(198)
73. PageMart and PageNet argue that the Commission should adopt a caller- pays rate.
Alternatively, PageMart argues that it should remove the avoided costs of a coinless call from the
compensation rate.(199) Alternatively, PageNet requests that the Commission adopt a cost-based approach
that apportions only the additional costs that are incurred through the origination or subscriber 800 calls
on a per-call increment, not per-call basis.(200)
74. CCI argues that the Commission should not adopt a cost-based methodology
because a marginal cost rate does not fairly compensate all calls as required by Section 276 of the Act and
does not address fair compensation for other types of calls from payphones or whether additional costs
could be recovered through compensation available to PSPs.(201) CCI contends that if the Commission
adopts a marginal cost standard, then the rates would need to be sufficient such that revenues would
recover the total marginal costs of installing and operating payphones, which in the long run could increase
long distance rates and force some PSPs out of business.(202)
75. Peoples and the Coalition argue that the Commission should not adopt a cost-based
rate because the costs for local coin calls and dial around calls are similar, and further that access code and
subscriber 800 calls may be more costly than coin calls. Several of the PSPs and the Coalition further
argue that a cost-based rate would lead to the removal of payphones with low call volumes or above
average costs.(203) TEI argues that cost plus a fair rate of return is not appropriate, because the underlying
costs are similar and there is seldom agreement regarding costs or a fair rate of return.(204) APCC argues
that the Court did not require the Commission to adopt a cost-based methodology.
b. Cost Components(205)
76. Equipment. CWI contends that only forward-looking direct costs should be
considered, including the amortized cost of installing a coinless payphone and the cost of maintaining the
equipment, excluding the cost for coin equipment.(206) Several of the IXCs argue that coin equipment costs
should be excluded when determining per-call compensation.(207) PageNet argues that coin related costs
such as maintenance, repair and replacement for coin functions should not be included in determining per-call compensation.(208)
77. The Coalition contends that equipment costs are attributable to both coin and
noncoin calls. Teleport contends that the fixed costs associated with installing a coin operated payphone,
such as the cost of the payphone, the enclosure, the cable plant, and supporting network infrastructure,
are attributable to both coin and noncoin calls.(209) APCC states that most payphone costs, including
purchasing, installing, and maintaining equipment, are fixed and should be attributed to both coin and
noncoin calls.(210)
78. CCI contends that monthly direct costs such as the telephone bill (6 cents per call), location owner commissions ($0.05 per call), maintenance and collection ($0.05 per call), parts and supply are properly attributable to both coin and noncoin calls. CCI, however, discounts the telephone bill costs ($0.02 per call) and maintenance and collection costs ($0.01 per call) to deduct local measured usage charge and the costs associated with dial around collection.(211)
79. Payphone Lines. APCC states that local exchange line charges represent a small
differential between coin and noncoin calls--on average, about 3 cents per call.(212) AT&T argues that
tariffed screening and blocking service from the LECs as well as other reasonable expenses such as touch
tone and 911 charges should be included in the cost of a call when computing the appropriate amount of
per-call compensation.(213) CompTel argues that the line charge should be no more than $0.046 per call.(214)
CWI contends that basic phone line plus usage charges, if any, for subscriber 800 and access code calls
should be included in computing per-call compensation.(215)
80. Several of the IXCs contend that the costs associated with terminating local calls
should not be used to compute per-call compensation.(216) CompTel argues that per-minute usage charges,
if any, imposed by a LEC for originating access code or subscriber 800 calls are appropriate.(217) PageNet
argues that line charges should not be included because non-PSP carriers already pay the LEC for the use
of the payphone line through originating access charges.(218)
81. Peoples argues that line charges are attributable to coin and noncoin calls. Peoples
argues that there is a minimum fixed line charge, and that in some states, there is an additional usage
charge.(219) Peoples further argues, however, that as more states require fixed charges, there will be no
difference between line charges for coin and noncoin calls.(220)
82. The Coalition contends that the Commission should not impose an offset for the
local usage charge because in many cases payphone lines are flat-rated and PSPs do not recover
termination or local usage charges. The Coalition contends, however, that if there is an offset, it should
not be greater than $0.02 per call, which reflects the average local termination cost across all Coalition
members.(221) CCI does not include local usage charges in calculating per call compensation amount.(222)
83. Coin/Noncoin Collections. The Coalition contends that the cost of coin collection,
counting, and related equipment accounts for approximately $0.02 of the total cost of a local coin, but
argues that this rate may be inflated because it allocates coin collection costs among coin calls based on
coin volumes, not the number of coins deposited.(223) APCC argues that the differences between coin and
noncoin calls in the area of coin collection are limited because coin collection is generally combined with
general maintenance visits to the payphone, about $0.03. APCC further argues that coinless collection
costs are likely to increase and may actually be $0.05-$0.06, thus higher than coin calls.(224) Peoples
contends that coinless collection costs are greater than coin call collection costs, and further that in the past
six months, coin related maintenance accounted for only 38% of all maintenance visits.(225) Peoples
estimates that coin collection related costs are approximately $0.03 per call, and that coin collection costs
are slightly lower than the cost involved in collecting for noncoin compensation.(226) Peoples contends that
dial around collection costs are approximately $0.05-$0.06 per call.(227) CCI argues that it does not include
coin collection costs of dial around calls in computing the appropriate amount of per-call compensation,(228)
but argues, however, that the costs associated with noncoin calls may increase due to additional expenses
for collecting and auditing such compensation.(229)
84. CPI and CompTel contend that PSPs experience lower costs for subscriber 800
and access code calls than for coin calls because it is more costly to maintain a coin phone than a coinless
phone.(230) AT&T, CWI, Excel, Frontier, MCI, PageNet, RCN, and ITA state that coin collection costs
should not be included in the rate of per-call compensation.(231) TEI states that some service costs can be
deducted when determining the rate for a noncoin call.
85. Teleport contends that costs associated with coin calls--collection, maintenance,
and cost of transporting a call--on a per call basis are de minimis, and further that the opportunity costs
associated with noncoin calls offset the de minimis difference in cost. TEI argues that the Commission
should include a cost for the time value of money used in collecting the compensation should the
Commission not prescribe collection tools for the PSP, and further, suggests that the Commission impose
a stated interest rate on late payers of per-call compensation.(232)
86. ANI ii. APCC contends that the Commission should not explicitly rule that such
charges incurred in restructuring the LEC networks to provide a unique screening digit for dumb payphone
lines may be assessed on PSPs. However, APCC contends, if LECs are allowed to assess such charges
on PSPs, then PSPs are entitled to recover those charges from IXCs dial-around compensation as part of
the cost of originating dial-around calls.(233) The Coalition contends that requiring PSPs to pay LEC tariffs
for ANI ii digits would add $0.05 to $0.08 to the per call rate, and Peoples supports attributing this cost
to subscriber 800 and access code calls.(234) AT&T, Excel, Sprint, and GCI argue that the PSPs are not
entitled to recover any costs for Flex ANI.(235) Excel and RCN state that IXCs should not be required to
pay for ANI information provided by the PSPs, because the PSPs are the beneficiary of the information.(236)
87. Depreciation/ Overhead. CWI, PageNet, and CompTel contend that per-call
compensation should not include depreciation costs or interest.(237) LCI, CompTel, and CWI argue that
administrative and overhead costs are not attributable to noncoin calls.(238)
88. CCI and TEI argue that overhead, depreciation, amortization, and interest are
attributable to coin and noncoin calls.(239) Peoples contends that overhead costs are attributable to all calls
made from payphones, and argues that the IXCs do not justify why such costs should not be included.(240)
89. Other. In its estimate, AT&T included an 11.25 percent interest on capital factor,
maintenance/warehouse/part costs and added averaged costs for the basic line and other related charges.(241)
AT&T admits that some costs such as overhead, general and administrative expenses and taxes are
appropriate in the computation of the cost of a noncoin call. According to AT&T, these costs are
approximately $0.012 per call.(242) CCI includes taxes and the return on invested capital in the calculation
of the costs of the per-call rate.(243)
90. CPI contends that subscriber 800 and access code calls are generally shorter in
duration than coin calls. Therefore, the longer duration of local calls could allow for opportunity costs
since few local calls displace shorter long distance calls.(244) TRA contends that per-call rates should not
include embedded or opportunity costs.(245) Excel argues that coin rating costs should not be included in
determining per-call compensation.
2. Discussion
91. As discussed above, we conclude in this order that an adjusted market-based local
coin rate is the appropriate surrogate for the default per-call rate for subscriber 800 and access code calls.
In this section, we explain our reasons for rejecting the proposals of various parties that we derive a default
per-call rate for such calls based on cost estimates submitted in the record of this proceeding.
a. Problems with the Proposed Methodologies for Deriving Payphone Compensation.
92. A number of commenters, notably the IXCs, argue that the Commission should
use the marginal cost of originating a payphone call as the basis for compensating PSPs.(246) Most of the
parties, however, estimate marginal costs based on the incremental cost of an individual coinless call.
Thus, as the Coalition explains, setting the rate at marginal or incremental costs means that joint and
common costs could not be recovered.(247) We conclude that the use of a purely incremental cost standard
for each type of call could leave PSPs without fair compensation for payphone calls, because such a
standard would not permit the PSP to recover a reasonable share of the joint and common costs associated
with those calls.(248) We also reject, for similar reasons, suggestions by commenters that we use local coin
rates currently in place as a surrogate for per-call compensation. As we stated in the NPRM, "local coin
rates in some jurisdictions may not cover the marginal [incremental] cost of the service."(249) Therefore,
basing the per-call compensation amount on current local coin rates, which are frequently subsidized by
state regulators, would not fairly compensate the PSPs. In the Payphone Orders, we rejected the use of
the $0.12 per-call compensation amount the Commission first discussed in its 1991 Notice of Proposed
Rulemaking in the access code call compensation proceeding. We noted that we never adopted the $0.12
per-call amount, and that rate was effectively rejected when the Commission adopted a $6 flat rate per
payphone per month based on a per-call rate for access code calls of $0.40.(250)
93. We determined in the Order on Reconsideration that reliance on cost studies, in
general, could reduce the revenue recovered by the PSPs, and therefore, might reduce the number of
payphones deployed.(251) We reaffirm that decision here. Adopting a per-call compensation scheme that
did not "promote the widespread deployment of payphone services" would be inconsistent with
Congressional intent.(252)
94. We also affirm our conclusion in the Report and Order that the cost-based TELRIC
standard that the Commission relied upon in the local competition proceeding is inapplicable here, because
the payphone industry is not a bottleneck facility that is subject to regulation at virtually all levels.(253) The
TELRIC pricing principles adopted in the local competition proceeding were designed to reflect the long
run cost of an element or physical facility. Since there are relatively few common costs between separate
facilities, TELRIC compensation will compensate a carrier for virtually all costs associated with providing
(the services of) that facility. With the addition of a share of the relatively small common costs, the firm
will be able to cover its total costs.(254)
95. Additionally, we conclude that Congress' use of the phrase "... payphone service
providers are fairly compensated for each and every completed interstate and intrastate call..."(255) is a
different standard than the cost-based standard articulated for the compensation for interconnection and
unbundled elements. We conclude that the PSP will be providing a competitive service (payphone use)
and should therefore receive compensation equal to the market-determined rate for providing this service.
In the Local Competition Order, we concluded that the cost-based interconnection standard, on the other
hand, compensates a carrier for the long run incremental cost of providing interconnection or the long run
incremental cost of providing an unbundled element plus a reasonable share of the common costs. Because
the local exchange is not yet competitive, we could not rely on the market to set competitive rates for
unbundled elements. In the case of payphones, the presence of multiple PSPs already operating in many
markets, and the structure of the industry that allows relatively easy entry and exit, leads us to conclude
that we can rely on market forces to provide for efficient pricing of these services in the near future.
96. In this proceeding commenters also argue that we should apply a TSLRIC cost
standard to only a subset of services (i.e., subscriber 800 and access code calls) provided by a facility
(payphone). In general, when several services are provided by the same facility, the incremental cost of
providing any one service is very small and the common cost among these services is very large. Thus,
a TSLRIC standard under which a carrier is compensated only for the incremental cost of each service
individually without a reasonable allocation of common costs, as suggested by commenters, would not
allow the carrier to recover the total costs of providing all of the services. A TSLRIC standard that yields
prices that recover a reasonable share of joint and common costs would require the difficult allocation of
those (large) costs among the different types of calls made from payphones.
97. We also reject suggestions that use of a market-based compensation standard, in lieu of one that is cost-based, will overcompensate PSPs. The marketplace will ensure, over time, that PSPs are not overcompensated. Carriers have significant leverage within the marketplace to negotiate for lower per-call compensation amounts, regardless of the local coin rate at particular payphones, and to block subscriber 800 calls from payphones when the associated compensation amounts are not agreeable to the carrier.
98. Previously, in the access code call compensation proceeding, we relied upon AT&T 0+ commissions as a measure of the fair value of the service provided by independent payphone providers when they originate an interstate call. Data presented above, however, suggest that the 0+ commission rate exceeds the market rate for local coin calls while the costs of access code and subscriber 800 calls are less than the costs of local coin calls. Furthermore, commissions may include compensation for factors other than the use of the payphone, such as a PSP's promotion of the Operator Service Provider (OSP) through placards on the payphone. Accordingly, we conclude that a market rate based on 0+ commissions would result in a default rate that overcompensates payphone providers for access code and subscriber 800 calls. Moreover, our approach is based on the costs of a low traffic location that does not support commission payments.
b. Analysis of Record Evidence of Payphone Costs
99. Although we reject suggestions that we set the default rate based on the long run
costs of providing service, our analysis of the record evidence indicates that an estimate of the long run
costs of providing access code and subscriber 800 service, including an equal per call share of joint and
common costs,(256) is not significantly less than the market-based rate determined above. Over time, the
marginal cost associated with new entry (adding a payphone) may be an important determinant of the
market rate for access compensation. For comparison, we estimated costs of the installation and operation
of a payphone at a low traffic location; that is, at a location that would be expected to generate sufficient
calls so that the payphone provider could earn only a normal return on investment and could not pay
commissions to the premises owner.
100. We calculated a rate for access code and subscriber 800 calls by estimating the cost
of a typical multi-use payphone that is capable of being placed outdoors. We then subtracted all costs
directly attributable to coin and access code calls to determine the amount of joint and common costs
associated with a multi-use phone. We then determined the amount of joint and common costs attributable
to each call by dividing these costs by an estimate of the number of calls placed at a location where a
payphone will earn a normal return on investment. Three parties, Peoples, CCI and AT&T provided
relatively consistent cost data that could be used to estimate joint and common costs. The following sub-sections summarize our category-by-category estimation of costs.
101. Maintenance. Data presented by Peoples indicates maintenance cost of 4.7 cents
per call.(257) Sprint suggests 3.6 cents per call.(258) CCI data suggest 6.6 cents per call(259) and Robinson's data
for AT&T suggest a total of between 2.5 and 4.0 cents per call.(260) Based on the information presented by
the parties,(261) we estimate that joint and common maintenance costs at a low traffic location would amount
to between 4.0 and 5.0 cents per call.(262)
102. Line costs. Data for Peoples suggests line costs of 5.9 cents per call.(263) Data for CCI suggests line costs of 7.9 cents per call.(264) Sprint suggest 8.0 cents per call.(265) Robinson's study suggests line costs of 6.5 cents per call.(266) We estimate that joint and common line costs at a low traffic location would amount to between 6.5 and 7.5 cents per call.(267)
103. Sales, General & Administrative. Data for Peoples suggests SG&A of 5.4 cents
per call.(268) Data for CCI indicates SG&A costs of 5.3 cents per call.(269) Sprint suggests 1.57 for SG&A.(270)
Sprint, as a LEC and an IXC, has a significantly different organizational structure and payphone base from
hat of independent payphone providers. Accordingly, little weight was given to Sprint data for SG&A.
Robinson did not develop an independent estimate of SG&A.(271) Accordingly, we use the estimates based
on data for Peoples and CCI as the high and low estimates, respectively. We conclude that joint and
common SG&A at a low traffic location would amount to between 5.3 and 5.4 cents per call.
104. Capital and Equipments Costs. Most parties recognize that payphone providers
should have an opportunity to recover depreciation costs and earn a return on investment. Joint and
common investments for a new payphone should include not only the costs of purchasing and installing
a payphone, but also a normal increase in leasehold improvements, spare parts and inventory, and cash
working capital.(272)
105. Robinson estimated the average outlay associated with adding a new smart coin
telephone as $1,050 for the instrument,(273) $300 for a pedestal and enclosure, $395 for installation of the
telephone, pedestal and enclosure, and $150 in local exchange carrier connection charges, for a total
investment of $1,895.(274) Some PSPs claim that Robinson underestimated pedestal and enclosure and
related installation charges.(275) The Robinson estimates do not include other investments, such as
maintenance vehicles and office equipment, needed to support a payphone business. Several PSPs
estimated average capital costs per call, but did not provide sufficient detail to allow these estimates to be
used to estimate the direct capital costs of adding a payphone.
106. We estimate joint and common equipment costs by: a) estimating the amount of assets that are likely to be added when a payphone is added; b) subtracting the amount attributable to the coin mechanism; c) calculating a monthly cost for the balance; and d) dividing the monthly cost per payphone by the low traffic location number of calls. Peoples 10K data indicate that Peoples depreciable net investment per payphone amounted to $1,617 as of December 1996.(276) CCI's 10K data indicate that CCI's depreciable net assets per payphone amounted to $1,704 as of December 1996.(277) Firms, however, add new assets rather than depreciated assets. Adjusting for depreciation, we estimate new depreciable investment per payphone of $3,234 for Peoples(278) and $2,799 for CCI.(279) As explained above, we impute $710 of new investment per payphone directly to coin calls. Accordingly, we calculate new joint and common investment per payphone of $2,524 and $2,089, respectively. These amounts of new investment would result in monthly investment costs of $43.94 and $37.07, respectively.(280) The carriers would also expect to earn a return on some other assets on the books -- pre-paid expenses and inventory. These items add $1.79(281) and $2.01(282) in investment costs per month, respectively. Summing the investment costs and dividing the low traffic location number of calls results in estimates of total investment costs of 7.2 cents per call and 8.4 cents per call, which we use as the likely range.
107. Other Costs. We concluded above that it was reasonable to include $0.01 in
adjusting the market rate for a local coin call to account for the cost of ANI ii deployment by the LECs,
passed through to PSPs in the form of higher access line charges, and include that figure in our analysis
here. We also concluded that carriers would receive access code and consumer 800 access compensation
approximately 3 months later than they would receive coin revenues, and thus included interest, based on
an 11.25% annual cost of capital the long run cost estimate. We use that same figure in our analysis here.
In addition, we explained earlier the positions regarding including commissions as a cost-factor, and thus
conclude that those costs are excluded properly from a cost-based analysis.(283)
108. Total Long Run Cost. The preceding analysis suggests that total long run cost of
access code and consumer 800 calls would range from 24.7 cents per call (based on a sum of the low
estimates) to 28.1 cents per call (based on the sum of the high estimates).(284)
109. Sprint's Motion. On September 16, 1997, Sprint filed a Motion asking that the
Commission require Bell Atlantic to submit a copy of the NET cost study filed before the Massachusetts
DPU and supporting papers to the Commission and to all parties of record in this proceeding. On
September 26, 1997, Bell Atlantic filed an opposition to Sprint's motion to require production of a
confidential cost study and conditional cross-motion for production of payphone cost data from Sprint and
AT&T. Bell Atlantic argues that Sprint's motion should be rejected because: (1) the study was prepared
for the Massachusetts DPU and Sprint should seek relief from that agency; (2) there is no justification for
requiring the production of the study because the study examines incremental costs, which, Bell Atlantic
argues, the Commission has rejected; and (3) the information is confidential.
110. We deny Sprint's motion and decline to require Bell Atlantic to submit a copy of
NET's cost analysis. We are not persuaded that the NET cost study, which Sprint indicates was submitted
to the Massachusetts DPU on a confidential basis, is necessary for us to reach a decision in this
proceeding. Furthermore, we note that there are differences of opinion regarding the NET methodology.
The NET study as well as other confidential studies filed in other states are not before us. We further note
that as Bell Atlantic states, the information is confidential, and therefore, should we require Bell Atlantic
to make such a filing, Bell Atlantic likely would require that we treat the study as confidential. Were we
to agree, the information would not be available to the parties. We note, moreover, information on the
record provides deregulated coin rates for several states. Because we are denying Sprint's motion, we
need not address Bell Atlantic's conditional motion for production of documents.
D. Per-Call Compensation Rate
111. In this section, we conclude that the default market-based per-call rate for
subscriber 800 and access code calls is $0.284, which reasonably accounts for the payphone costs that are
incurred solely in connection with local coin calls and costs that are specific to access code and subscriber
800 calls.
1. Comments
112. Parties filed comments that varied considerably, primarily depending on whether
they relied on a market-based or derived rate methodology. AT&T and ARCH argue that the
compensation rate should be $0.11 per-call, based on the costs of providing a subscriber 800 or access
code call.(285) AT&T arrives at this rate by estimating a cost of $76.85 per month for a payphone divided
by an average of 700 calls per phone per month.(286) AT&T contends that this rate is consistent with
NYNEX's local coin rate of $0.167. Alternatively, AT&T and MCI argue that if the Commission adopts
a rate based on an offset from the local rate, then the offset should be at least 50%.(287) AT&T further
argues that even using a adjusted market approach as suggested by the Coalition results in payphone
compensation in the amount of $0.1067 cents per call, which is in line with the rate that AT&T has
calculated for coinless calls based on its estimated monthly costs of a payphone.(288) AT&T further states
that even if adjustments have to be made for depreciation, overhead, general and administrative expenses
and taxes, the per-call cost for coinless calls would only increase to 12.2 cents per call.(289) AT&T
maintains that $0.35 is not the appropriate unregulated coin rate because it was based on a small and
unrepresentative sample of rural states, and the cost in those states could be higher than in other areas.(290)
The Commission ignored the deregulated rate in other rural states, where the rate is $0.25, which, AT&T
asserts, also is the dominant rate where the majority of payphones are located.(291) Borden suggests a rate
of approximately $0.133 per call, and Champion suggests a rate between $0.08 and $0.11.
113. CompTel argues that a fair compensation amount based on incremental costs is
between $0.03 to $0.05 per call,(292) and that even under a direct cost approach, compensation should not
exceed $0.10 per call.(293) Frontier argues that a cost-based rate should be approximately $0.10 per call,(294)
but no higher than $0.11 per call.(295) ITA argues that the rate should be between $0.08 and $0.15 per
call.(296) MCI argues that the per-call rate for access code calls is $0.083 per call, and that the number for
subscriber 800 calls should be even lower.(297)
114. MIDCOM states that the rate should be $0.057.(298) Sprint argues that on a fully
allocated approach to costs, using an efficient bellwether provider, the default rate per call should be
$0.06.(299) TRA argues that the 35 cent rate is too high.(300) Excel argues that the Court decision
demonstrates that we cannot set the rate for subscriber 800 and access code calls at the same level as the
local coin rate, and thus the Commission must reduce the $0.35 rate.(301)
115. The Coalition states that, to truly reflect the market, the local coin rate needs to
be adjusted from $0.35 upward to $0.42 or $0.43 per call.(302) In a fully realized market, the Coalition
states, noncoin calls would be carrying a greater portion of the payphone costs than coin calls, and
therefore should be priced at a higher rate.(303) APCC alleges that the average per-call local coin rate is
$0.41, not $0.35.(304) IPTA and TEI state that the record supports a compensation level of no less than
$0.35 per call.(305) CCI requests that the Commission set the per-call compensation rate at $0.35.(306)
116. The majority of the IXCs argue that there should be one national rate,(307) because
a varying rate would be nearly impossible to administer, and could increase the costs to carrier-payers of
administering per-call compensation.(308) Furthermore, CWI argues that because not all carriers can block
calls, the Commission should not create a situation where carriers must block calls because they are
unaware of the rate to be charged.(309) MCI argues that if the Commission does not adopt one uniform rate,
then it should set parameters such as notifying carriers of the coin rate in advance and changing the coin
rate not more than once per year.(310) APCC argues that the Commission should not adopt a uniform
compensation rate, and although the costs associated with a non-uniform rate may be higher, the benefits
of directly market-based compensation are worth the extra costs.(311)
2. Discussion
117. We conclude from our analysis in Section B, that the market-based rate for access code and subscriber 800 calls, adjusted for cost differences is $0.284.(312) We further conclude that the market-based rate we establish herein as a default rate for per-call compensation promotes the goals of Section 276 of the Act, fair compensation, the deployment of payphones, and competition, and is a rate that is reasonably related to the market-based local coin rate. As discussed below, we conclude that the $0.284 default rate for per-call compensation rate, absent negotiations, should be in effect for two years to enable LECs, PSPs and IXCs additional time to transition efficiently and without disruptions to the deregulated payphone market structure created in the Payphone Orders.(313) Furthermore, we conclude that after the two year per-call compensation rate period, "fair compensation" for access code and subscriber 800 calls pursuant to Section 276 and an analysis of the record is the deregulated market rate for the local coin call adjusted for costs as discussed herein. Accordingly, the default rate for the first two years of per-call compensation is $0.284; after the first two years, the default rate is the market-based local coin rate minus $0.066 per call. We conclude that the default per-call rate falls within a zone of reasonableness that will provide fair compensation for subscriber 800 and access code calls as required by Section 276, while allowing the market to develop, and PSPs who desire, to negotiate a different rate.(314)
118. In adopting an adjusted market-based rate approach, we note that the Commission
has the authority to rely on market forces, and further, that "market predictions are within the institutional
competence of the Commission."(315) In adopting this approach, we are confident that market forces will
keep payphone prices at competitive levels, and that our default rate is in accordance with prevailing
market conditions adjusted for costs. Courts have upheld rates established by regulatory agencies that lie
within a "zone of reasonableness,"(316) particularly, in the context of ratemaking. While we do not consider
the development of the default rate established herein to be ratemaking, because market imperfections
currently exist, within the evolving competitive payphone market, we have set a default rate to ensure
competition.(317)
119. As discussed above, in response to the claims of parties on the record that only
a rate derived from cost data submitted in the record will provide a valid per-call rate, we have also
performed an analysis of those data for purposes of comparison with the market-based per-call rate we
establish in this order. In setting the default rate for per-call compensation at $0.284 based on our market-based analysis, we have also considered the results of our analysis of the record information concerning
the long run costs of payphone service. We have calculated the long run costs per-call for a provider to
install a payphone to be in the range of $0.247 per call to $0.281 cents per call.(318) An estimate compiled
under this long run costs approach must be considered a lower bound when establishing a default rate. The
rate derived in this manner, by definition, just covers the cost of installing and operating a payphone at a
marginal location. As such, it will not encourage either the deployment of additional payphones or an
incentive for IXCs to negotiate with PSPs. Such minimal incentives are contrary to the goals of promoting
competition among payphone service providers and promoting the widespread deployment of payphone
services. Accomplishing these goals requires that we ensure that the default rate, in addition to covering
cost, provide sufficient incentives for PSPs to deploy additional payphones and tangible incentives for IXC
and PSPs to negotiate. Thus, the default rate we adopt for subscriber 800 and access code calls based on
the market-based local coin rate adjusted for costs differences is appropriately and reasonably at the high
end of the range compiled from the long run cost analysis.
120. We deny requests that we should mandate a uniform and fixed per-call
compensation rate for each compensable call. A fixed rate would not promote the statutory goals of
Section 276, because it would not encourage negotiations between IXCs and PSPs. It is our expectation
that IXCs and PSPs will build business relationships and create operating procedures to provide
compensation in an efficient manner. Given that we have adopted a deregulatory approach in this order,
we conclude that we should not establish those procedures. Under the approach we established in the
Report and Order, the market is allowed to set the compensation amount for calls originated by each
payphone. The court did not vacate that part of the Report and Order. For market-based pricing to
function effectively, it is not unreasonable that there be some variation in compensation amounts from
location to location. We also decline to delay the effective date of this order as requested by CWI. As
we discussed previously, we conclude that it is in the public interest to make this order effective
immediately.(319)
121. In this order, we extend the per-call interim compensation period subject to a
default rate established in the Payphone Orders for an additional year. Thus, the per-call compensation
period during which the default rate is $0.284 begins on October 7, 1997, and ends on October 6, 1999.
We established the interim compensation plan in the Payphone Orders in order to ease the transition to
market-based rates. We stated that it was necessary to observe over time how the payphone marketplace
would function in the absence of regulation. We noted that market imperfections had led us to establish
a default rate. On this record, we conclude that additional time is required to ease the transition to market-based rates and that continuing the applicability of the default rate for an additional year is in the public
interest. As we have summarized in this order, we have received comments from LECs, PSPs, and IXCs
regarding the problems and issues they face in transitioning to the payphone market compensation structure
we established in the Payphone Orders. For example, IXCs and their customers allege that after the first
year of per-call compensation established in the Payphone Orders, when the default rate will be the
deregulated coin rate adjusted for cost differences, PSPs will raise the coin rate in a manner that will raise
substantially the per-call rate for access code and subscriber 800 calls. They indicate that their systems
are not adequately prepared to respond to such situations. In addition, LECs have indicated problems in
providing the payphone-specific coding digits required to respond to calls from payphones on a real-time
basis for some payphones in their serving areas.
122. Although we conclude in this order that the marketplace, based on negotiations
between IXCs and PSPs, is where compensation decisions should be determined and that the default rate
after the per-call transition period should be the market-based local coin rate adjusted for cost difference,
we believe that this two year per-call compensation period subject to the default rate is necessary to afford
IXCs, PSPs and LECs the opportunity to adjust to and adequately prepare for the deregulatory market-based structure we adopted pursuant to Section 276.(320)
E. Other
1. Comments
123. AirTouch Plan. AirTouch suggests that the Commission explore a new method to resolve the compensation issue due to the wide divergence of views expressed in the replies, and its concern that call blocking options do not exist. AirTouch argues that the Commission should adopt a method that does not rely on call tracking or call blocking to place checks on the imposition of excessive charges by payphone service providers.(321) AirTouch proposes that the Commission adopt a unique 8XX approach that would be toll-free for long distance charges, but could be accessed from a payphone only if the caller deposits coins (presumably at a fraction of the local coin rate). PageNet and PCIA support AirTouch's unique 8XX approach and state that it merits further investigation.