cable datacom news case no. (tlm-7) of

>

< http://www.cabledatacomnews.com/cgi-bin/printer.cgi ._ - '.

Cable Datacom News

OCTOBER 2003

Case No. U-13796

of 3 Exhibit - (TLM-7)

Backup Power Reemerges As Issue for Cable VolP Service Northeast Power Outage and Hurricane lsabel Stir Concerns About Network Reliability

OCTOBER 2003

By Alan Breznick, Editor, Cable Datacom News

It looks like cable operators preparing to roll out voice-over-Internet-Protocol (VoiP) service over their broadband plant may need to beef up their plans for emergency power backup, thanks to the back-to-back punches of the massive US. Northeast blackout in August and Hurricane isabel last month.

The national Network Reliability & lnteroperability Council (NRIC)--a group of wireline phone, wireless phone, cable, satellite and Internet executives that advises the FCC on technical issues and develops voluntary "best practices" standards for its industries--is looking at tightening its emergency power recommendations for all telephone providers. At their last meeting in Washington, D.C. in mid-Sept., NRlC members discussed the need for enhanced power backup because consumers are increasingly relying on all their communications networks during emergencies.

"We need to think more diversely about customer communications, such as cable phone or cable data service." said Pam Stegora Axberg, a senior vice president of Qwest National Network who chairs the NRlC Steering Committee. That's "because during times of crisis, people are using multiple channels." she noted.

Axberg said NRlC members are also looking at setting a higher standard for the amount of backup power needed. While the council's best practices now call for a minimum of eight hours of backup power, that standard could climb to 12 hours or beyond because of the great length of the outages caused first by the blackout and then by the hurricane.

"Now we're wondering whether that (the current standard) is long enough," she said. "We've never seen a power outage last this long previously."

1 of3 12/11/2003 4:42 PM

."

F . ' .. http://www.cabledatacoinnews.c,)~cgi-bi~printer.cgi

Case No. U-13796

What this means is that cable operators may soon be expected to provide emergency power backup for their fledgling IP telephony services at all times, even when the electricity shuts down and the lights go out. That, in turn, could require MSOs to conduct more expensive upgrades of their HFC plant, which is something that they've resisted doing so far.

It also means that cable operators may be expected to provide backup phone power for much longer stretches of time than before That would likely mean higher costs for them as well.

Although such leading cable phone players as Comcast and Cox Communications now provide "network powering" of customer premises telephony devices for their older circuit-switched services, MSOs have been balking at doing the same for IP telephony because of the high cost. Instead. many big cable operators have been touting the idea of using a less costly backup battery, which would keep the embedded multimedia terminal adapter (E-MTA) going for up to 10 hours in case of a power outage. And unlike network powering, which requires an upfront network investment before installing subscribers, the capital cost for a battery must only be incurred when a new customer is activated for service.

In the most prominent example, Cox plans to use backup batteries in its E-MTA units when it conducts a market trial of its new VolP product in Roanoke, Virg. near the end of the year. Bill Dame, senior manager of network switch engineering for Cox, said the batteries will provide 10 hours of standby power and four hours worth of talk time.

"The batteries have come a long way in the last two years," said Dianna Mogelgaard, director of product development for Cox. She noted that the batteries have evolved from large lead varieties to much smaller yet more powerful lithium ion models.

Even with cheaper battery systems available, though, several major MSOs are not exactly rushing out to offer emergency power backup for their IP telephony deployments right away. Take Time Warner Cable, the cable pioneer in residential VolP service. Although Time Warner launched commercial service in Portland, Maine last winter, now has more than 4,000 VolP customers there and plans to introduce service in three other markets this fall, the cable operator is not yet employing a backup power solution.

"We're still talking to people," said a Time Warner spokesman. "It's still kind of early. We're still looking for the right solution."

Likewise, Cablevision Systems, which is readying a rollout of its OptimumVoice IP telephony service to more than 4 million homes in the New York metro area this fall, will not initially use back-up batteries. Instead, the company spokesman said, Cablevision is mulling over offering battery backup as an optional, add-on service by the end of next year.

Time Warner and Cablevision executives defend their companies' lackadaisical power backup approaches by arguing that consumers don't expect lifeline phone service from their cable provider, even

Exhibit (TLM-7) of 3

2of3 12/11/2003 4 4 2 PM

http://www.cabledatacoinnews.com/cgi-bi~printer.cgi

Case No. U-13796

of 3 when it's marketed as a primary-line service. Company officials insist Exhibit __ (TLM-7) that cable telephony customers are much more interested in other service aspects, such as cheaper rates, unlimited dialing and the various calling features.

"Most of our customers have cell phones and powered cordless phones so it's not a priority item for them," the Cablevision spokesman said. To the point. during the East Coast blackout and the hurricane, many RBOC customers found themselves without phone service, not realizing that their cordless phones wouldn't work during a power outage.

In Cablevision's VolP trial on western Long Island earlier this year, the company spokesman said, customers did not stress the need for emergency power backup. "It was not the pressing issue you might think," he said.

The Time Warner spokesman concurred. "The RBOCs seem to be trying to make a big issue out of it (powering)," he said. "But that's just not ringing true with our customers."

Cox officials say their cable customers aren't making a big deal about backup power either. But, based on their years of experience with circuit-switched service, they still think it's an important service reliability feature to provide right now.

"Our perception locally is that it does make a difference, at least in Roanoke." Dame said. "If you launch the service and lose the customer, you're never going to get him back."

Mogelgaard put it somewhat differently. "I think we're all seeing similar trends," she said. "But we want to start conservatively."

Fortunately for the MSOs, it doesn't appear that either the big Northeast power outage or Hurricane Isabel disrupted their nascent IP telephony operations too much. In Time Warner's case, for instance, both the power outage and the hurricane spared Portland so emergency backup did not become an issue there.

Nevertheless, Time Warner, Cablevision and other MSOs may ultimately have to follow Cox's lead. As cable operators transform themselves into phone companies, consumers may expect them to perform just like phone companies, no matter what technologies they use or other services they deliver.

"I would say that this is a new area for them (cable operators) to be thinking about," Axberg said. "They've never thought of themselves as a lifeline service before. But that's changing."

Cable Datacom News is published by Kinetic Strateries lnc. Phoenix, Arizona USA. 0 Copyright 1996-2003 Kinetic Stratepies Inc. AN Rights Resewed.

12/11/2003 4 4 2 PM 3 0 f 3

?.''2UJGAN PUBLIC SERVICE COMMISSION

Exhibit - (TLM-8) Discussion of Claimed Triggering Companies

Public Version

FILE u - - I ~ ~ F ( - -~ EXHIBIT N 0 . m

Page 1 of 27 DATE 0 3 I l'?!o U E P O R T E R

1 SBC claims lo have identified eleven' companies as counting toward the retail

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

trigger in the seven Michigan MSAs in which it seeks a finding of no impairment. As

shown in Exhibit A-- (JPL-12) to Mr. Lube's testimony, these companies are: AT&T,

Choice One Conmlunications, Comcast, Climax (shown as CTS Telecom), KMC

Telecom, LDMI (with Mpower), MCI (shown as WorldCom), McLeodUSA, Mich Tel,

TDS and XO. I discuss each of the companies in this exhibit, focusing on attributes

relevant to the determination of whether the competitor should count toward the retail

trigger for mass-market switching.

ATgrT

Although AT&T provides local exchange service to both residential and business

mass-market customers in Michigan, it provides residential service exclusively via UNE-

P.' In response to Staffs First Data Request, AT&T indicates that it has ***BEGIN

PROPRIETARY x x x x x x x x x x x x x x x x x x x x x x x x x x x END

PROPRIETARY*** that serve customers in Michigan. AT&T states that those

switches serve a total of ***BEGIN PROPRIETARY x x x x x x x x x END

PROPRIETARY*** business customers spread across ***BEGIN PROPRIETARY x

x x x x x END PROPRIETARY*** SBC wire centers, with a maximum in any one wire

center of ***BEGIN PROPRIETARY

served via UNE-L. AT&T also explains that it serves ***BEGIN PROPRIETARY x x

END PROPRIETARY*** business lines

' Lube Direct at 66, A.136. AT&T's response to ACNE-TelTTalk Data Request 1.Ol.b.i and 1.02 confirms that

AT&T does not use UNE-L to provide service to residential customers. This data response, and all of the other CLEC data responses cited in this Exhibit, are attached hereto in Exhibit - (TLM-21).

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21


Public Version Page 2 of 27

x x END PROPRIETARY*** “Prime family” voice-grade-equivalent lines using

those switches, but that the “vast majority” of those customers “are served by high-

capacity lines” instead of DS-0 lines.’

Therefore, I eliminated AT&T from the analysis based on my trigger screen that

determines whether a carrier offers retail local exchange service to residential mass-

market customers.

SBC’s data, which I have used throughout my “trigger” analysis, confirm that

AT&T’s mass-market UNE-L entry is minimal. SBC identified a total of ***BEGIN

PROPRIETARY x x x x END PROPRIETARY*** supposed mass-market loops for

AT&T in ***BEGIN PROPRIETARY x x x END PROPRIETARY*** wire centers

in the SBC seven MSAs. The supposed AT&T mass-market loops range from a

minimum of ***BEGIN PROPRIETARY x x x x x x x x x x x x x x x x x x x x x x x

x x x x x x x x x x x END PROPRIETARY*** mass-market loops per wire center in

which SBC found AT&T to have any mass-market loops. Given SBC’s apparent data

quality problems, such low numbers could fall within a reasonable “margin of error” in

many wire centers (ie., it is unclear that these numbers are sufficiently different from

zero to be confident that AT&T actually has any UNE-L customers in those wire

centers).

Even if I had not eliminated AT&T on the basis of the residential screen, AT&T

would have dropped out of ***BEGIN PROPRIETARY x x x x x x x x x END

PROPRIETARY*** based on the 1% market share screen. There is no wire center in

AT&T response to Staffs First Discovery Request at 5-6, 3

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21



which AT&T’s market share is as high as ***BEGIN PROPRIETARY x x x x x END

PROPRIETARY***

Choice One

Choice One’s Web site indicates that it primarily markets to business customers.

For example, the first paragraph in the page on which the company introduces and

describes itself (httu://www.choiceonecom.coni/ourconipany/) states that Choice One

serves “small and medium sized companies in the Northeast and Midwest United States”

and that “[wle have chosen to target small- and medium sized-business because they are

typically underserved by the incumbent local exchange provider (Baby Bell or Local

Independent).”

A Choice One general manager recently confirmed the logic of avoiding the

residential market entirely “because there is not a lot of room for growth and it is a huge

drain on resource^."^

However, a January 8,2004 press release on the site also notes that the company

is introducing residential service in its New York service areas. Choice One’s reply to

Staffs First data request indicates that Choice One has one switch serving loops in

***BEGIN PROPRIETARY x x x x END PROPRIETARY*** Michigan wire

centers, with ***BEGIN PROPRIETARY x END PROPRIETARY*** residential and

***BEGIN PROPRIETARY x x x x x x END PROPRIETARY*** business loops;

however, the residential loops may well be residual lines from its merger with US.

Xchange and entirely unrelated to Choice One’s current business plan for Michigan.

4 1i~p://1nilwat1kee.biziournals.corn/1nilwaukee/stori~/2003/06/02/f~~~1~2.htm1

1

2

3

4

5

6

I

8

9

10

11

12

13

14

15

16

17

18

19

20

21



SBC’s data confirm that Choice One’s supposed mass-market UNE-L entry is

minimal. SBC identifies a total of ***BEGIN PROPRIETARY x x x x x END

PROPRIETARY*** supposed mass-market loops for Choice One in ***BEGIN

PROPRIETARY x xEND PROPRIETARY*** wire centers in the SBC seven MSAs.

Choice One’s supposed mass-market loops in those wire centers range from a minimum

of ***BEGIN PROPRIETARY x x x x x x x x x x x x x x x x x x x x x x x x x x x x x

x x x x END PROPRIETARY*** massmarket loops per wire center in which SBC

found Choice One to have any mass-market loops.

SBC’s reported total mass market loops for Choice One are only ***BEGIN

PROPRIETARY x x x x END PROPRIETARY*** of the total loops that Choice One

reported in response to Staffs First data request. Thus, it appears that the level of “mass

market” customers Choice One serves is close to incidental to its overall operation and

may well be merely lines that it acquired in a merger.

Nonetheless, I have not excluded Choice One based on my screen that requires a

carrier to be actively serving residential customers. My treatment of Choice One on this

screen is conservative in the sense that it tends to overstate the company’s active

participation in the mass market as a whole, given Choice One’s own public statements

concerning its disinterest in serving residential customers.

Choice One, however, does drop out of the analysis for ***BEGIN

PROPRIETARY x x x x END PROPRIETARY*** SBC wire centers based on the 1%

market share screen. ***BEGIN PROPRIETARY x x x x x x x x x x x x x x x x x x

1

2

3

4

5

6

7

8

9

10

11

12

13

I4

15

16

17

18

19



END PROPRIETARY*** in which Choice One counts toward the retail trigger, based

on my screeniug criteria.

Comcast

Comcast is a traditic a1 cable mpany that offers telephon service in some, but

not all, parts of its existing cable footprint. Comcast’s limited geographic scope is ample

reason to disqualify the company as a trigger. Moreover, Comcast does not make any use

of UNE-L to supplement its own cable loop plant and does not offer its facilities to other

potential market entrants.’ Thus, Comcast’s actual deployment does not demonstrate that

any other companies could enter SBC’s Michigan service territory and serve mass-market

customers without access to UNE switching.

Moreover, it is unclear that any cable telephony offering can be considered to be

entirely comparable to SBC’s retail voice service in terms of cost, quality and maturity.

Backup power issues are a concern, as documented in Exhibit - (TLM-7) to my direct

testimony. Comcast’s own Web site indicates that its cable telephony offering has

limited backup power.6

Cable companies such as Comcast typically initiate local exchange service when

they already have an embedded base of facilities constructed for another purpose and an

embedded base of video and cable modem customers. The ability to add telephony to an

existing cable network does not indicate that competitors without the “first-mover”

SBC’s own data show ***BEGIN PROPRIETARY x x x x x x x x x x x x x x x x x x

See Exhibit - (TLM-19).

x x x x x x x x x x x x x x x x x x END PROPRIETARY***

1

2

3

A

5

6

7

8

9

10

11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 21 28

29



advantage of a cable franchise would be able to compete with the ILEC without access to

unbundled local switching.

Total lines in service and/or market share for a cable telephony provider using its

own loop plant also do not provide useful evidence that the national finding of

operational impaimient has been overcome. Only UNE-L providers that pass this screen

can provide such evidence because only these companies deal with hot cuts and the other

operational issues associated with the attempt to use SBC facilities in a UNE-L

arrangement to serve mass-market customers.

There are also signs that Comcast may not be aggressively pursuing traditional

phone service at all. For example:

Comcast will reverse AT&T Broadband’s aggressive telephony acquisition policies and implement its own corporate policy of trialing and then deploying voice over IP services, a senior executive said today. AT&T enlisted more than 1 million telephony customers using conventional constant bit rate [CBR] phone technology. Comcast will maintain these customers, but it won’t go looking for more, John Alchin, Comcast’s executive vice president and treasurer, said during a luncheon presentation at the Warburg Media day in New York City. ‘There is an element of cutback on telephony’, said Alchin, discussing Comcast’s plans to spend more than $2 billion to upgrade AT&T Broadband plant next year. While we haven’t yet shared with you the details of the capital plans for 2003, you should not expect us to take the telephony product into a whole host of new markets. It will be a case of supporting the product where it is today without expanding. ’7

Also,

’ “Comcast Curtailing AT&T Telephony Deployments,” Dec. 12,2002, Telephony Online, at http:/ltelephonvonline.comiar/telecom coincast curtailing att/index.htm

1 2 3 4 5 6

7

8

9

10

11

12

13

14

15

16

17

Sxhibit - (TLM-8) Discussion of Claimed Triggering Companies


As a result of the Company’s reduced marketing efforts and focus on telephone service profitability, Comcast now expects to lose approximately 175,000 Comcast Cable phone customers this year, a modest adjustment from the original expectation of up to a 150,000 telephone customer decline [announced in the February 27,2003 g u i d a n ~ e ] . ~

For these reasons, I have disqualified Comcast from counting toward the retail

trigger.

Climax/CTS Telecom

Climax Telephone Company (a wholly owned subsidiary of CTS

Communications) operates as both an ILEC and a CLEC in Michigan. In 1996, Climax

became “the first case in which a licensed incumbent provider of basic local exchange

service has filed an application to provide service in another licensed incumbent

provider’s temtory.”’

Since then, Climax, by its own report, has not grown beyond ***BEGIN

PROPRIETARY x x x x x x xEND PROPRIETARY***, with which it claims to serve

a total of ***BEGIN PROPRIETARY x x x x x END PROPRIETARY*** analog

Comcast Third Quarter 2003 Earnings Release, October 30,2003 (Iittp://www.c1iicsk.com/~lioenix.zlitml?c=l1859 1 &o=irol- iiewsArticle&t=Reaular&id=464588&).

services include: Josh Long, “Marketing for Voice Put on Hold, Telephony Not a Priority for Comcast in 2003,” Xchange, 2/1/03 (http://www.x-changernag.com/articles/321window2.hhl); “Corncast Phone Falling Fast,” Telephony Online, August 1,2003, littp://teleulioiivoiiline.comlar/teleco~n coincast phone fallindindex.htin; “Cable Telephony Surging,” Teleplioriy Online, November 26,2003, l~ttp://teleuho~ivotili~ie.coin/ar/telecon~ cable telephonv sureindindex.htrn.

Order Approving Application, Case No. U-11143, 10/7/1996, at 2.

Other articles about Corncast’s less-than-aggressive approach to digital telephone

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19



loops.'o Because Climax relies entirely on ILEC-provided switching, I have eliminated

Climax as an ILEC affiliate.

Climax also states that it does not actively market its service to either business or

residential mass-market customers.'' According to Climax, it serves only ***BEGIN

PROPRIETARY x x x x x x xEND PROPRIETARY*** residential customers from its

switch.I2 Thus, one could also disqualify Climax from being counted toward the retail

trigger because it does not actively use UNE-L to serve the mass-market customers. As a

conservative measure, I did not screen out Climax on this basis.

SBC's data indicate that Climax's supposed mass-market UNE-L entry is

minimal. SBC identified a total of ***BEGIN PROPRIETARY x x x x x x END

PROPRIETARY*** supposed mass-market loops for Climax in ***BEGIN

PROPRIETARY x x x END PROPRIETARY*** wire centers within MSAs in which

SBC is seeking a finding of no impairment for mass-market switching. The supposed

Climax mass-market loops in those wire centers range from a minimum of ***BEGIN

PROPRIETARYxxxxxxxxx x x x x x x x x x x x x x x x x x x x x x x x x x x x

END PROPRIETARY*** mass-market loops per wire center in which SBC found

Climax to have any mass-market loops. If I had not eliminated Climax as an ILEC

affiliate or because it does not serve all mass-market customers using UNE-L, I would

still have disqualified Climax from counting toward the retail trigger in all but

Climax Responses to ACN/Z-Tel/Talk, requests 1.01 and 1.03, respectively 10

' I Climax Responses to ACN/Z-Tel/Talk, request 1.07.

l 2 Climax Response to ACNE-Tel/Talk, request 1.02.



1

2

***BEGIN PROPRIETARY x x xEND PROPRIETARY*** SBC wire centers based

011 the 1% market share screen

3 KMC Telecom

4 KMC’s supplemental response to MCI’s First Set of Discovery speaks for itself

5 6 7 8 9

10 11 12 13 14 15

16 17 18

19 20 21 22 23 24 25 26 27 28 29

KMC states that it should not be considered a Self Provider of analog Plain Old telephone Service (“POTS”) to the mass market segment utilizing our switches in Michigan. At this time, the principal business of KMC is to serve Enterprise customers and not Mass Market customers in the areas in Michigan where our switches are located. Today, KMC actively markets only to medium and large business enterprise customers, who have a high demand for a variety of sophisticated data-centric telecommunications services and solutions. KMC’s success had been in serving Enterprise customers and not Mass Market customers.

Today, KMC actively seeks to serve customers who plan to purchase digital services at capacities that justify the use of DS1-level loops. ....

There are two specific instances in which KMC may offer DSO level service while serving Enterprise customers. First, existing Enterprise customers who order additional voice services from KMC may, on occasion, be at capacity on their existing DS1 facility, necessitating the provisioning of individual DSO level facilities at an existing location. The second instance occurs when a prospective or existing customer wishes to include other locations into their service package, but those locations do not have sufficient volume to justify a full DSl. KMC would also provision individual DSOs to such locations.

30

31

32

33

KMC is clearly not an active participant in Michigan for mass-market customers

and should not be counted as a triggering company for that reason.

KMC provides an illustration of how easily a handful of UNE-L services

provisioned to an apparent mass market location (according to SBC) can be



misinterpreted as mass-market entry. Instead, such loops may be merely the byproduct of

an abandoned strategy or incidental requirements of service to an enterprise- market

1

2

3 customer.

4

5

6

SBC's data further confirm that KMC's mass-market UNE-L entry is minimal

and incidental. SBC identified a total of ***BEGIN PROPRIETARY x x x x x END

PROPRIETARY*** supposed mass-market loops for KMC in ***BEGIN

7

8

9

10

11

12

13

14

15

16

PROPRIETARY x x xEND PROPRIETARY""" wire centers within Michigan MSAs

for which SBC seeks a finding of no impairment. The supposed KMC mass-market

loops range from a minimum of ***BEGIN PROPRIETARY x x x x x x x x x x x x x

x x x x x x x x x x x x END PROPRIETARY*** mass-market loops per wire center in

which SBC found KMC to have any mass-market loops.

JSMC does not achieve ***BEGIN PROPRIETARY x x x x x x x x x END

PROPRIETARY*** market share in any wire center based on SBC's data. Thus, if it

had not been eliminated as failing the active market participant screen, it would fall out

***BEGIN PROPRIETARY x x x x x END PROPRIETARY*** from the triggering,

CLEC list based on the 1% market share screen.

17 LDMI 18

19

20 own switch."

LDMI is not currently offering service to mass-market customers and does not

intend in the future to market to mass-market customers for services provided over its

I 3 LDMI Response to ACN, Z-Tel & Talk America data request 1.01, B-i.

1

2

3

4

5 6 I 8 9

10 11 12 13

14

15

16

17

18

19

20

21

22

23

24

Exhibit - (TLM.4) Discussion of Claimed Triggering Companies


Instead of being an active participant using W E - L for mass-market customers,

LDMI serves a smattering of mass-market UNE-L customers that it inherited from

bankrupt CLEC Mpower. LDMI implies that Mpower’s bankruptcy and its pursuit of

mass-market customers may be related.

LDMI, the identified competitive switch provider, is not actively providing voice service to mass market customers in the market, and has not been actively providing voice service to mass market customers in the market. The lines noted were provisioned by the previous CLEC, Mpower, from whom LDMI purchased the switch, collocations and customers noted in this spreadsheet. It should be noted that Mpower went bankrupt in the proce~s . ’~

Thus, LDMI cannot be considered an active participant for mass-market

customers and should be eliminated on that basis. LDMI’s business focus is

demonstrated on the “About LDMI” section of its web page, which begins: “Founded in

1992, LDMI has emerged as a world-class, full-service regional telecommunications

company serving thousands of businesses in the Great Lakes Region.” One of the five

primary headings on that page is “We’re All for Business.”

That focus is confirmed in LDMI’s response to Staffs First data request, in which

LDMI indicates that it serves ***BEGIN PROPRIETARY x x x x x x x x x x x x x x x

x x x x x x x x x x x x x x x x x x x xEND PROPRIETARY*** lines from its switch.

Thus, I also eliminated LDMI as a trigger CLEC because it does not provide service to

the entire mass market.

Letter from Gary L. Field to Mr. Steven D. Hughey, Assistant Attorney General, 14

Michigan Public Service Commission, 11/26/03.

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19



Even if LDMI were not eliminated as inactive, it clearly is not operationally

ready for mass-market entry via UNE-L. LDMI indicates that it has at most a minimal

electronic interface with SBC or other CLECs.” LDMI also does not have any internal

automated systems to process own retail orders when it uses its own switch.I6

Moreover, LDMI indicates that it serves more than ***BEGIN PROPRIETARY

x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x

END PROPRIETARY***’7 LDMI is thus another example of a case in which SBC’s

approach to trigger analysis would have a drastic and unreasonable effect, robbing

***BEGIN PROPRIETARY x x x x x x x x x x x x x xEND PROPRIETARY*** of

Michigan customers of the competitive choice they have made because of a handful of

loops inherited from another carrier’s failed business plan.

LDMI illustrates that mere ownership of a switch that serves some mass-market

customers does not necessarily relate to a strategy to enter any well-defined market at all

(let alone an MSA). LDMI states it has no idea why its switch does or does not connect

with any given SBC wire center as those decisions were made by someone at Mpower for

reasons that LDMI “cannot speak to.”’*

SBC’s data further confirm that LDMI’s mass-market UNE-L entry is minimal

and incidental. Although SBC identified a total of ***BEGIN PROPRIETARY x x x x

END PROPRIETARY*** supposed mass-market loops for LDMI in the seven MSAs at

Is LDMI response to MCI Data Requests MCIC-11, and MCIC-16 through MCIC-18, 12/9/03,

l6 LDMI response to MCI Data Requests MCIC-19 and MCIC-22, 12/9/03. LDMI response to ACN/Z-TelITalk Requests 1.01 and 1.02. LDMI response to MCI Data Requests MCIC-47, 12/9/03.

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

Exhibit - (TLM-8) Discussion of Claiined Triggering Companies


issue in this proceeding, those loops are spread across ***BEGIN PROPRIETARY x x

END PROPRIETARY*** wire centers. The supposed massmarket loops in those wire

centers range from a minimum of ***BEGIN PROPRIETARY x x x x x x x x x x x x x

x x x x x x x x x x x x x x x x x x x x x x x END PROPRIETARY*** mass-market

loops per wire center in which SBC found LDMI to have any mass-market loops. LDMI

does not achieve even a ***BEGIN PROPRIETARY x x x END PROPRIETARY"""

market share in any wire center and achieves greater than a ***BEGIN

PROPRIETARY x x x x x x x END PROPRIETARY*** market share in ***BEGIN

PROPRIETARY x x x x x x END PROPRIETARY*** wire center, based on SBC's

data. If it had not been eliminated because it is not an active market participant and is not

serving all mass-market customers, LDMI ***BEGIN PROPRIETARY x x x x x x x x

x x x x x x x x x x x x x x x x x x x END PROPRIETARY*** based on the 1%

market share screen.

Thus, the Commission has multiple bases on which it could exclude LDMI

entirely from the retail trigger count for massmarket switching.

McLeodUSA

SBC claims McLeodUSA as a triggering carrier. McLeodUSA's responses to

Staffs First data request indicate that McLeodUSA serves Michigan customers using

***BEGIN PROPRIETARY x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x

x x x x x x x x x x x x x x x END PROPRIETARY*** McLeodUSA's Web site

indicates that it generally serves both residential and small business mass-market

22 customers.

1

2

3

4

5

6

I

8

9

10

11

12

13

14

1s

16

17

18

19

20

Exhibit - (TLM.3) Discussion of Claimed Triggering Companies

Public Version Page 14 of27

SBC identified a total of ***BEGIN PROPRIETARY x x x x x x x x x END

PROPRIETARY*** supposed mass-market loops for McLeodUSA spread across

***BEGIN PROPRIETARY x x xEND PROPRIETARY*** wire centers in the seven

MSAs for which SBC is seeking a finding of no impairment. The supposed McLeod

mass-market loops range from a minimum of ***BEGIN PROPRIETARY x x x x x x x

x x x x x x x x x x x x x x x x x x x x x x x x x x x x x END PROPRIETARY***

mass-market loops per wire center in which SBC found McLeod to be serving any mass-

market loops.

Based on this information regarding mass-market loops served via UNE-L and its

own switches, I did not entirely “screen out” McLeodUSA. However, McLeodUSA’s

market share equals or exceeds 1% in only ***BEGIN PROPRIETARY x x x x x END

PROPRIETARY*** wire centers in the SBC seven MSAs. Hence. in all but those

instances, I have excluded McLeodUSA from the retail trigger count for mass-market

switching for SBC.

My treatment of McLeodUSA is, like much of the rest of my analysis,

conservative in the sense of giving SBC’s trigger claim the benefit of the doubt. For

example, although McLeodUSA does serve some residential accounts, its focus appear;;

to be on business customers. McLeodUSA itself reports only that only ***BEGIN

PROPRIETARY x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x

x x x x x x x x x END PROPRIETARY*** serve residential customer^.'^

McLeodUSA response to Staffs First Data Request, column AH and AI totals. 19

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20



McLeodUSA also appears to provide a good illustration of the potential ham1 to

facilities-based competition of preniaturely eliminating the UNE-P option for mass-

market customers in an overly broad market, such as on an MSA-wide basis.

McLeodUSA’s Michigan tariff includes a wire center by wire center listing of which

delivery platform McLeodUSA currently uses in Michigan. Specifically, the

McLeodUSA tariff indicates where it provides service using “the McLeodUSA Class 5

Local Switch in combination with a local loop leased from the incumbent carrier” as

opposed to using “a combination of network elements.”’” Based on that information for

McLeodUSA’s residential offering, I note that McLeodUSA continues to use “network

elements” (primarily UNE-P and hence UNE switching, but perhaps also some resale) to

deliver service in a significant number of wire centers. For example, even in the Detroit

and Grand Rapids MSAs, McLeodUSA indicates that it continues to use UNE-P as its

delivery vehicle in a number of SBC wire centers. In Grand Rapids alone,

McLeodUSA’s tariff states that it relies on UNE-P in at least five SBC wire centers

***BEGIN PROPRIETARY x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x

x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x

x x x x x x x x x x x x x x x x x x END PROPRIETARY***. Thus, McLeodUSA’s

tariff indicates that it does not provide UNE-L service at all in an area in Grand Rapids in

which SBC has more than ***BEGIN PROPRIETARY x x x x x x x x x END

PROPRIETARY*** loops. This information suggests that, even in MSAs in which

2o McLeodUSA Telecommunications Services, Inc., Tariff M.P.S.C. No. 7R. Definitions are from the First Revised Sheet No. 18. Listings of McLeodUSA’s delivery platform by wire center are included in section 5.0, at sheets 113-129.



1

2

3

4

5

6

7

8

9

they serve mass-market customers via UNE-L in a nontrivial number of wire centers,

facilities-based carriers may rely on a UNE-P option in other wire centers to enable

nbiquitous coverage, despite the econoniic incentive to maximize switch utilization.

Some of the specific loop counts noted above may be the result of mismatched

timing between SBC’s data and McLeodUSA’s tariffs; nevertheless, the point is

important. McLeodUSA’s recent public statements make clear that, although the

company is working on “continued migration of customers from resale to higher margin

platfonns, as well as installation of new customers on-switch,’’ it still relies on a

substantial amount of UNE-P to achieve a viable customer base as it expands its facility-

10 based network.2’

11

12

13

14

15

16

17

At most, McLeodUSA’s actual deployment demonstrates tbat, in a handful of

SBC wire centers, McLeod has been able to provide UNE-L services at a volume that

meets my minimum threshold for demonstrating that the company has had some SucceS::

in overcoming barriers to entry. However, McLeodUSA’s financial condition suggests

that the cost to McLeodUSA for attempting this entry approach may have been too great,

making it debatable whether McLeodUSA can be considered “likely to continue”**

offering service using its existing approach. Based on McLeodUSA’s own statements,

*’ McLeodUSA Reports Third Quarter 2003 Results (press release), 10/22/03, at 2. Also see the “Selected Telecommunications Statistical Data” provided as part of that press release, which indicate that McLeodUSA is gradually decreasing the level of UNE-P in its network but still relied on UNE-P as its delivery platform for 32% of its lines as of September 30, 2003. Exhibit A to McLeodUSA’s response to MCI’s First Set of Discovery reports numbers that diffk from SBC’s and that also do not match the designations in its tariff. However, that attachment confirms that McLeodUSA relies on UNE-P to provide for ***BEGIN PROPRIETARY x x x x x x x x x x x x x x x x x x END PROPRIETARY*** of Michigan wire centers.

Triennial Review Order, 7 500

1

2

3

4 5

6 7 8 9

10 11 12 13

14 15

16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

Exhibit - (TLM-8) Discussion of Claimed Triggering Conipanies


taking away other entrants’ access to W E - P based on McLeodUSA’s success as a

“trigger” would be high-risk regulation indeed. Consider, for example, the following

statements from McLeodUSA’s 2002 Annual Reportz3

Fiiiancial Iiiforritation Related to Our Post-Emergence Operations Is Limited

Because we emerged from bankruptcy on April 16, 2002, and divested non-core assets in 2002, there is limited operating and financial data available from which to analyze our operating results and cash flows. As a result of fresh-start accounting, a comparison of information reflecting our results of operations and financial condition after our emergence from bankruptcy to prior periods may not be meaningful.

Executing Our Business Strategy Involves Substantial Risks

There are substantial risks in implementing our

* difficulties arising from a slower economy, which business strategy, including:

has adversely affected demand for competitive telecommunications services and may delay revenue growth and the length of time required to achieve profitability and positive cash flow;

other governmental developments during the course of implementing our business strategy;

restructuring that increase the risk that we may lose or fail to attract customers;

potential difficulties in retaining quality management to execute our business strategy; and

invest in improved business systems to support a larger enterprise.

the chance for adverse regulatory, legislative and

* reputation issues related to our bankruptcy and our

we may not have sufficient capital to continue to

23 As of Februaty 8,2004, McLeodUSA did not have a copy of its 2003 Annual Report available on its Web site.

1 2 3

A

5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

21 28

29 30 31 32 33 34 35 36 37 38

39

Zxhibit - (TLM-8) Discussion of Claimed Triggering Companies


One or more of these factors, individually or combined, could affect adversely our ability to conduct our operations.

Our BUS~IWSSCS Do Not Generate Positive Cash Flow

Our operations consist primarily of our retail telecommunications business. To date, our retail telecommunication operations have not generated positive cash flow in any quarterly period. While the strategic initiatives we have undertaken to improve our business are designed to result in the telecommunication business eventually generating positive cash flow, there can be no assurance that these steps will be successful in the time and of the magnitude expected. Even if we achieve our targeted level of Earnings before Interest, Taxes, Depreciation and Amortization, or “EBITDA,” we may continue generating negative cash from operations until fiscal 2005 as a result of capital expenditures and projected interest on the remaining amounts outstanding under the Credit Agreement dated as of May 3 1,2000, among McLeodUSA Incorporated, the lenders party thereto, and JPMorgan Chase Bank, (the “Credit Agreement”). Retention of access to the amounts that remain available under the Credit Agreement, dated as o f April 16,2002, among McLeodUSA Incorporated, the lenders party thereto, and JPMorgan Chase Bank, as Agent, or the “Exit Facility,” is critical in funding our future operations.

We Expect to Incur Significant Losses Over the Next Several Years

If we do not become profitable in the future, we could have difficulty obtaining funds to continue our operations. We have incurred net losses every year since we began operations. McLeodUSA incurred a net loss applicable to its common shares of $204.9 million for the period April 17,2002 to December 31,2002. We expect to incur significant operating losses during the next several years. If we are unable to generate an operating profit in the future, there may be adverse consequences to our business.

...

1 2

3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21

22

23

24

25

26

21

28

29

30

31

32

Exhibit - (TLM.-8) Discussion of Claimed Triggering Companies


Failure to Raise Necessary Capital Could Restrict Our Ability to Develop Our Network and Services

There can be no assurance that our capital resources will be sufficient to enable us to achieve operating profitability. Failure to generate or raise sufficient funds may require us to delay or abandon some of our plans or expenditures, which could harm our business and competitive position. We expect to meet our funding needs through various sources, including existing cash balances, our Exit Facility vendor financing and cash flow from future operations. Our estimated aggregate capital expenditure requirements include the projected costs of

completing our in-region fiber optic communications network and deploying network assets currently not in service;

switches;

telecommunications assets in our target markets; and

to support a more efficient telecommunications company.

- activating and augmenting voice and data

constructing, purchasing, developing or improving

improving the business infrastructure and systems

MCI - SBC cites MCI (WorldCom) as a triggering carrier. SBC identified a total of

***BEGIN PROPRIETARY x x x END PROPRIETARY*** supposed mass-market

loops for MCI spread across ***BEGIN PROPRIETARY x x x x x x x x x END

PROPRIETARY*** wire centers in the SBC seven MSAs at issue in this proceeding.

The supposed MCI mass-market loops range from a minimum of ***BEGIN

PROPRIETARYxxxxxxxxx x x x x x x x x x x x x x x x x x x END

PROPRIETARY*** mass-market loops per wire center in which SBC found MCI to be

serving any mass-market loops.

SBC’s claim demonstrates the fallacy of the company’s application of the trigger

test, MCI relies on UNE-P as its predominant vehicle for serving mass-market

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

Exhibit - (TLh”-8) Discussion of Claimed Triggering Companies

Public Version Page 20 o f27

customers. Indeed, MCI states that its only mass marketing is for its “Neighborhood”

product, which is UNE-P based.24 I understand that MCI does not serve residential

customers via UNE-L except in very specific conditions related to Brooks Fiber facilities

in limited areas-ie., in a manner that has nothing whatsoever to do with the company’s

“forward-looking’’ mass-market entry.25 Instead, MCI provides residential mass-market

loops where it has inherited those loops from a prior CLEC and the resulting customerr;

do not have an alternative local service provider available.

Moreover, I am infonned that MCI does not use mass media to market any UNE-

L services to mass-market customers. Instead, the media advertising from MCI

concerning residential and small business offerings is all directed toward MCI’s UNE- P

based product. To the extent that MCI serves any small business customers via UNE-I,

it currently obtains those customers through direct sales contact with medium and large

business customers. This information discredits one of SBC’s alleged bases for its MSA

market definition because MCI does not use mass media in conjunction with so-called

“mass-market” UNE-L; the economies of scale and scope for direct sales contacts are

quite different from those associated with media advertising and are consistent with a

more circumscribed market definition.

I also understand that the “small businesses’’ that MCI serves via UNE-L often are

not small businesses at all, Instead, in many instances, MCI provides three or fewer

MCI response to ACN/Z-Tel/Talk request, 1.07. 2s My understanding of this point and others concerning MCI’s service to residential and

small business customers is based primarily on my review of information provided by Sherry Lichtenberg, a Senior Manager in the Operations Support Systems and Facilities Development Business Unit for MCI. Ms. Lichtenberg is concurrently filing testimony that confirms this information.

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

Zxhibit - (TLM-8) Discussion of Claimed Triggering Companies


W E - L lines to a single business location as part of a package of telecommunications

services that includes DS-I and/or other high-capacity, high-volume services suitable to

enterprise customers, as the FCC has defined that term. The ability to serve a small

volume of such analog loops as part of a package of services does not provide probative

evidence of the ability to overcome economic and operational barriers with respect to

mass-market switching.

I eliminated MCI from the trigger analysis based on the ‘‘serves residential

customers” screen.

Had I not done so, MCI’s low volumes (which reflect the “incidental” nature of

much of its business services via analog UNE loops) as reported by SBC would have

caused the company to fall out of the analysis in all but ***BEGIN PROPRIETARY x

x x x END PROPRIETARY*** wire centers after I applied the 1% market share screen.

MCI’s data related to those wire centers raise further questions concerning that

data on which SBC has chosen to rely in its trigger analysis. In wire center ***BEGIN

PROPRIETARY x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x



x x x x x x x x x x x x x x x x x x x x x x x x x END PROPRIETARY***26 Both of

these examples suggest potentially substantial miscounts in SBC’s data. Until such

discrepancies can be investigated, the Commission should not rely on SBC’s claims to

26 Id.

dxhibit - (TLM-6) Discussion of Claimed Triggering Companies


1

2

3 line counts are exaggerated.

eliminate mass market UNE switching. As my “trigger” analysis is based entirely on

SBC’s self-reported numbers, it is further conservative to the extent that SBC’s reported

4 Mich Tel

5

6

7

8

9

10

11

12

13

14

15

16

17

16

19

20

21 serve.

SBC includes Mich Tel in its count of triggering CLECs in ***BEGIN

PROPRIETARY x x x END PROPRIETARY*** MSAs even though SBC’s own data

indicate that it would be generous to describe Mich Tel’s use of UNE-L as marginal.

SBC’s own data indicate that Mich Tel’s supposed mass-market UNE-L entry consists of

a total of ***BEGIN PROPRIETARY x x END PROPRIETARY*** supposed mass-

market loops in ***BEGIN PROPRIETARY x x x END PROPRIETARY***

Michigan wire centers within the MSAs for which SBC seeks a finding of no impairment.

The supposed Mich Tel mass-market loops range from a minimum of ***BEGIN

PROPRIETARY x x x x x x x x x x x x x x x x x x x x x x x x x x x END

PROPRIETARY*** mass-market loops per wire center in which SBC found Mich Tel

to have any mass-market loops. As with a number of other CLECs on SBC’s trigger list,

the Mich Tel mass-market loop counts are sufficiently low that they may be as much the

product as SBC data errors as of any real UNE-L entry.

To the best of my knowledge, Mich Tel did not respond to any party’s discovery

in this docket (including Staffs); therefore, I do not have the data necessary to determine

what, if any, portion of the mass market base Mich Tel may be serving or attempting to

Exhibit - (TLM-.8) Discussion of Claimed Triggering Companies


1

2

3

4

5

6

7 that impression.

8

9

10

11

12

The various Mich Tel-related Web sites that I was able to locate shed no

additional light.27 Instead, they indicate that the company is somehow involved in a wide

range of lines of businesses including long distance, Internet services, wireless service,

high speed data services, collocation and interoffice facilities and even PC and equipment

sales. The overall impression is that Mich Tel is largely oriented toward ISP services.

But, as no specific list of available phone services is provided, it is impossible to confirm

Hence, Mich Tel may be disqualified on a number of bases, including not offering

service to the entire base of mass market customers and not currently offering service.

Regardless, Mich Tel is disqualified as a trigger ***BEGIN PROPRIETARY x x x x x

END PROPRIETARY*** based on the 1 .O% market share screen. Indeed, it has not

achieved more than ***BEGIN PROPRIETARY x x x x x x x x x x x x x x x x x x x x

13 X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X E N D

14 PROPRIETARY***.

15 - TDS

16

17

18

19

SBC claims TDS as a triggering carrier. TDS did not, to the best of my

knowledge, provide substantive responses to any discovery issued in this docket except

for Staffs.” However, even in response to Staffs request, TDS objected to providing

any data regarding the number of loops of any type it serves using its own switches.

’’ E.g., http://www.michtel.coin/index.shtiiil, htt~: / /www.michtel .com/host ing/colocat i~~~,

’* TDS’s reply to MCI’s discovery consists almost entire of objections and the claim that

and iittp://~~w.metroceli.corn/metrocel I/frames.aso.

its responses to Staff were sufficient.

1

2

3

4

5

6

I

8

9

10

11

12

13 14 15 16 17 18 19 20

21

22 23

dxhibit - (TLM-8) Discussion of Claimed Triggering Companies


Hence, there are no data at my disposal with which to evaluate SBC’s claims. SBC,

however, identified a total of ***BEGIN PROPRIETARY x x x x x x x x x END

PROPRIETARY*** supposed mass-market loops for TDS spread across ***BEGIN

PROPRIETARY x x x END PROPRIETARY*** wire centers in the seven MSAs at

issue in this proceeding

TDS’s primary business is cellular service through its U.S. Cellular operations,

with 4.1 million wireless lines in service in 25 states.29 Secondarily, it is an ILEC, with

ILEC operations split among 11 1 subsidiaries that serve “71 1,200 access line equivalents

in 28 states.”30 After that TDS is a CLEC that has two subsidiary CLEC operations with

“291,400 access line equivalents in five state^."^' TDS’s CLEC line of business is very

deliberately linked to and leveraged off of its ILEC operations. TDS describes the

strength of its CLEC operations as follows:

By carefully choosing its areas of operations near [ILEC] TDS Telecom’s existing operations, TDS Metrocom is able to leverage TDS Telecorn’s management and process infrastructures. TDS Telecom provides shared services for functions such as billing, accounting, regulatory affairs and human resources, allowing TDS Metrocom to focus its own resources on marketing, selling and serving the customer - critical areas to support rapid

TDS’s President and Chief Executive Officer recently reinforced this approach:

We are continuing to grow our CLEC businesses, TDS Metrocom and U. S. Link. Metrocom is being grown

29 Telephone and Data Systems, Inc. (“TDS”), 2002 Annual Report, at 2-4.

3o Id.

’I Id. I also note that lines reported as “equivalents” can be misleading as, for example, a

32 Id. at 4.

few physical DS-3 lines can be reported as many hundreds or thousands of “equivalent” lines.

1 2 3 4 5 6 7 S 9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

Cxhibit - (TLM-S) Discussion of Claimed Triggering Companies


aggressively using our distinctive fomiula for success-that is leveraging the strengths of our well established ILEC business to create a high quality, cost effective sales offering very quickly in the markets we choose to enter. Metrocom uses extensive support in many forms from TDS Telecom’s ILEC, leaving Metrocom free to concentrate its energies on marketing and selling to customers. Metrocom brou ht this successful approach to southern Michigan last year. $3

Thus, in Michigan as elsewhere, it appears that TDS is an ILEC leveraging its

assets into new markets. Although TDS does apparently target mass market customers it

only does so where it can use its ILEC resources to maximum advantage in doing so.

The Commission’s list of Regulated Local Telephone Companies Licensed in Michigan

(as of January 26, 2004) lists five distinct TDS-affiliated ILEC operations in Michigan--

Chatham Telephone Co., Communications Corp. of MI (“CCM’)), Island Telephone

Company, Shiawassee Telephone Co., and Wolverine Telephone C O ? ~ I have, therefore,

excluded TDS as an ILEC affiliate.

TDS does appear to serve mass-market customers and, based on SBC’s data,

appears to have more than a 1% market share in ***BEGIN PROPRIETARY x x x x x

x x END PROPRIETARY*** SBC wire centers. Thus, if it were not an ILEC affiliate,

TDS would count toward the retail trigger in those wire centers.

XO Communications

The final triggering company claimed by SBC is XO. SBC identified a total of

***BEGIN PROPRIETARY x x x END PROPRIETARY*** supposed mass-market

33 1itto://www.teldta.co1n/investor/invu~esentation05232002.1~tm, remarks of Ted Carlson, TDS Annual Meeting, 5/23/02, discussion accompanying slide 10.

httu://m.cis .state.mi.us/mpsc/comm/clec/newlocal.udf 34

1

2

3

4

5

6

7

8

9

10 11 12 13 14 15 16 17 18 19 20

21 22 23

24

25

26

6xhibit - (TLM-8) Discussion of Claimed Triggering Companies


loops for XO in the seven MSAs for which it seeks a finding of no impaimlent. The

supposed XO mass-market loops in those wire centers range from a minimum of

***BEGIN PROPRIETARY x x x x x x x x x x x x x x x x x x END

PROPRIETARY*** mass-market loops per wire center in the ***BEGIN

PROPRIETARY x xEND PROPRIETARY*** wire centers in which SBC found XC)

to be serving any mass-market loops.

It is my understanding that XO does not provide service to residential customers

and has not indicated any intention of doing

confinned that:

Moreover, like KMC, XO has

. . . it should not be considered a Self Provider of analog Plain Old telephone Service (“POTS”) to the mass market segment utilizing its switches in Michigan. At this time, the principal business of XO is to serve Enterprise customers and not Mass Market customers in the area in Michigan where our switch is located. Today, XO actively seeks to serve Enterprise customers, who have a high demand for a variety of sophisticated data-centric telecommunications services and solutions. XO’s success has been in serving Enterprise customers not Mass Market customers.

XO actively seeks to serve customers who plan to purchase digital service at capacities that justify the use of DS1-level

36 loops ...

The XO Web site confirms that XO does not actively market service to residential

customer^.^' Therefore, I eliminated XO from my analysis based on the “serves

residential customers” screen

XO response to Staffs First data request, which reports one switch with no residentid 35

customers. Also, XO response to MCIC-62 and MCIC-53, in which XO states that “. . .n6 residential customer has requested service from the Company.”

XO Supplemental response to MCI’s First Set of Discovery, 02/02/04. 36

1

2

3

4

5

6

7

8

9

10



Had I not applied this screen, XO would have failed the 1% market share screen

in ***BEGIN PROPRIETARY x x x x x x x x x x x x xEND PROPRIETARY*** in

which SBC identifies XO as providing mass-market loops. A carrier that fails this screen

in any geographic market is a carrier whose actual deployment provides no meaningful

evidence of overcoming economic and operational barriers to serving mass-market

customers via W E - L . Indeed, XO’s service volumes do not exceed a ***BEGIN

PROPRIETARY x x x x x END PROPRIETARY*** market share in any SBC wire

center. These low volumes provide a second, independent reason for the Conmission to

disregard SBC’s claim that XO should be counted toward the retail trigger for mass-

market switching

~ ~

The XO page (http://www.xo.com/ ) provides information about services for “Small & 37

Growing Business,’’ “Large Business” and “Carrier Services.” I was not able to find any reference to a residential service offering.

.

Exhibit - (TLM-8) Discussion of Claimed Triggering Cornpanks

***REDACTED***

Exhibit - (TLM-9) Trigger Screen Summary Table

***HIGHLY SENSITIVE CLEC INFORMATION***

MICHIGAN PUBLIC SERVICE COMMISSION FILE u - i 3 7 9 L

EXHIBITNO. 1- zL

o m 0311 9 1 0 ~ U R EPORTER

Exhibit - (TLM-9) Murray Trigger Screen Summary Table

HIGHLY SENSITIVE CONFIDENTIAL DATA -- NOT FOR DISCLOSURE TO ANYONE WHO HAS NOT SIGNED THE PROTECTIVE ORDER IN MICHIGAN PSC CASE NO. U-I3796

Screen 4 I Screen5 I Screen 6

Notes: See Murray Exhibit - (TLM-8) for complete discussion by CLEC

(1) Refers only to the seven MSAs in Michigan for which SBC is seeking a finding of no impairment. Companies listed as failing at wire center level fail at every wire center--the other carriers fail at some wire centers, as well. Likewise, companies listed as failing on the MSA level fail in every MSA-other carriers fail at some MSAs as well.

(2) Choice One's residential loops may well be residual lines from its merger with U.S. Xchange and entirely unrelated to Choice One's current business plan for Michigan.

(3) Although the carrier has some residential customers, they were acquired from other carriers and are grandfathered; carriers are no longer actively adding residential CustomerS.

2/9/2004

Exhibit - (TLM-14) Excerpts from Deposition of SBC Ohio witness Dr. Timothy I. Tardiff

Public Utilities Commission of Ohio Case No. 03-2040-TP-C01 November 19,2003

.

I

MC GlNNlS & ASSOCIATES, INC. 614.431.1344 COLUMBUS, OHIO 800.498.2451

www.mcginniscourtreporters.com

1 BEFORE

2

3

THE PUBLIC UTILITIES COMMISSION OF OHIO

_ _ _ 4 In the Matter of )

Implementation of the ) 5 Federal Communications ) Case No. 03-2040-TP-COI

Conmiission's Triennial ) Attorney-Examiner Agranoff 6 Review Regarding Local )

Circuit Switching in ) 7 the Mass Market. 1

_ _ _ 8

9 Deposition ofTimothy J. Tardiff, a witness

I0 herein, called by the Intervenors for cross-examination

1 I under the statute, taken before me, Linda D. Riffle,

12 Registered Diplomate Reporter, Certified Realtime

13 Reporter and Notary Public in and for the State of Ohio,

14 pursuant to notice and stipulations of counsel

15 hereinafter set forth, at the offices of Public

16 Utilities Commission of Ohio, Hearing Room 1 I-D,

17 180 East Broad Street, Columbus, Ohio, on Wednesday,

18 November 19,2003, beginning at 4:OO o'clock p.m. and

19 concluding on the same day.

20 _ _ _ 21

LL

23

24

25

60

MC GINNIS &ASSOCIATES, INC. 614.431.1344 COLUMBUS, OHIO 800.498.2451

www .mcginniscourlreporters.com

1 going on there to -- to indicate what was going on.

2 Q. Have you done any such analysis?

3 A. Studying of CLEC's bankruptcies? No.

4

5 switch serving the entirety of the SBC service territory

6 in the State of Ohio.

7 A. Oneswitch?

8 Q. One switch serving iiiany different MSAs.

9

10 four MSAs throughout the state, how do you take into

I 1 account the scale and scope of that switch when, in

12 fact, it's serving other MSAs?

13 A. Well, 1 think you are. But, I mean, I think

14 you just described how it's happening. Again, the FCC's

15 order suggests that -- 1 mean, you know, the way that --

16 most conservative inipainnent analysis is to attribute

17 the total cost of the switch to only the areas you're

18 doing the impairment analysis.

19 And the effect of that would be to cause

20 the -- you know, the average cost to be higher, you

21 know. So if -- if the switch has the capability of

22 having a broader area and it's -- in effect, that's

23 lowered the cost of service in the CLEC's -- excuse me,

24 in the MSAs for which you're doing the impairment

25 analysis.

Q. Let's take a situation where you have a CLEC

If you are only looking at impairinent, say, in

61

MC GINNIS & ASSOCIATES, INC. 614.431.1344 COLUMBUS, OHIO 800.498.2451

www.nicginniscourtreporters.com

1 Q. Wouldn't another approach be to assume an

2 economic use of that switch throughout the four MSAs

3 being studied?

4 A. Excuseme?

5 Q. Wouldn't another way of assuming that proper

6 scale and scope of that switch would be to assume an

7 efficient use of that switch in the four MSAs being

8 studied as opposed to the 14 MSAs that that switch

9 serves?

10 A. I don't think I understand your question.

1 I Q. 1 think you're saying that your approach would

12 be to assume the total cost of the switch being applied

13 to the four MSAs when, in fact, that switch is serving

14 all 14 areas. My suggestion or question, I guess, is

15 couldn't the approach be just to assume an efficient use

16 of that switch in each of the four MSAs being studied?

17 A. Well, that's -- I mean, on -- I mean, that is

18 the cost of serving. I guess I'm not understanding

19 where you're going with this; 1 mean, what the question

20 is.

21 Q. Where I'm going might depend on what your

22 answeris.

23 A. Well, okay.

24

25 allocate properly the scale and scope of a switch when

Q. HOW do you allocate -- I mean, how do you

62

MC GINNIS & ASSOCIATES, INC. 614.431.1344 COLUMBUS, OHIO 800.498.2451

www.mcginniscourlreporlers.com

1 you are -- when the geographic market you are analyzing

2 is smaller than what the switch actually serves?

3 A. When you look at patterns of entry, you know,

4 I mean, that -- in my mind, that presupposes what the

5 CLEC is doing. So whatever costs they're incuming,

6 they find that they can serve, say, throughout an MSA,

I so they're not impaired in that MSA. So whatever causes

8 one to make that conclusion reflects, you know,

9 basically how they're deploying switches.

10 Q. Okay. So if I understand you correctly, then,

I I you can assume an efficient use ofwhatever switching it

12 is to serve the geographic market under consideration

13 if, in fact, any switch is being used to serve that

14 geographic market?

15 A. Makes _- Yeah, I think that makes sense.

16

17 today MI. Deere testified that in Ohio, there are

18 different bands, B, C and D, within certain MSAs. Do

19 you remember the testimony?

20 A. 1 understand those to be different _- areas

21 where different prices apply to UNE loops.

22

23 MSA, correct?

24 A. So, in other words, you could see all three

25 prices?

Q. In Ohio, did you -- I think -- believe earlier

Q, With -- And that could all be within the same

Exhibit - (TLM-19) Coincast Discussion of Backup Power

MICHIGAN PUBLIC SERVICE COMMISSION

cable datacom news case no. (tlm-7) of

Documents