AFFIDAVIT OF ALFRED E. KAHN AND TIMOTHY J. TARDIFFBEFORE THE FEDERAL COMMUNICATIONS COMMISSION

In the matter of Application of SBC Communications Inc.,Pacific Bell,

and Pacific Bell Communicationsfor Provision of In-Region, InterLATA Services in California

TABLE OF CONTENTS

I . INTRODUCTION 1

II. THE HISTORICAL TRADE-OFF IN THE LINE-OF-BUSINESSRESTRICTIONS 2

A. The Issue During the Pre-divestiture Period 2

B. The Balance of Advantages and Disadvantages has 3

III. PBCOM'S ENTRY IS IN THE PUBLIC INTEREST 3

A. The Current State of InterLATA Competition 41. Long distance prices, access charges and margins, overall 52. Distribution of the benefits of competition between large and small users 11

B. The Benefits of PBCOM's Entry into the InterLATA Market 141. A more effective competitor 152. Economies of scope 16

IV. THE RISKS OF ANTICOMPETITIVE CONDUCT AFTER ENTRY ARESMALL AND OF SUCCESSFUL SUPPRESSION OF COMPETITION NIL19

A. Existing Safeguards Against Discrimination: audited service quality standards, monitoring by customer/competitors, penalties and opportunities for retaliation1. The established regime for competition 202. The anti-discrimination requirements of the Telecommunications Act 233. Incentive regulation 244. Accounting safeguards 26

B. The Unbundling, Resale and Interconnection Provisions of the 1996 Actand the Section 271 Checklist 27

C. Successful Competition between Vertically Integrated RBOCs and FirmsRequiring Access to Their Facilities in Other Markets 401. InterLATA corridor traffic 412. InterLATA service offerings by non-BOC LECs 373. Cellular 424. Paging 385. Voice Messaging Service (VMS) 396. Customer premises equipment 397. IntraLATA toll 40

- 2 -Draft

V. THE IMPORTANCE OF SYMMETRY IN EXTENDING THE FREEDOM TOCOMPETE 46

A. Blurring of Boundaries between Markets; the Importance of One-StopShopping 41

B. The Adverse Consequences of Asymmetrical Restrictions on the Ability toCompete Reciprocally 50

VI. SUMMARY AND CONCLUSIONS 52

BEFORE THEFEDERAL COMMUNICATIONS COMMISSION

WASHINGTON, D.C. 20554

In the matter of ) )Application of SBC Communications Inc., )Pacific Bell, and Pacific Bell Communications ) CC Docket No. _____for Provision of In-Region, )InterLATA Services in California )

AFFIDAVIT OFALFRED E. KAHN AND TIMOTHY J. TARDIFF

Alfred E. Kahn and Timothy J. Tardiff, being duly sworn, depose and say:

I. INTRODUCTION

1. My name is Alfred E. Kahn. I am the Robert Julius Thorne Professor of Political Economy,

Emeritus, Cornell University and Special Consultant with National Economic Research Associates, Inc. (NERA). I

have been Chairman of the New York State Public Service Commission and of the Civil Aeronautics Board; and in

my capacity as Advisor to President Carter on Inflation, I participated actively in the successful efforts of his

Administration to deregulate both the trucking industry and the railroads. I am the author of the two-volume The

Economics of Regulation, reprinted in 1988 by MIT Press, and have written and testified extensively in the area of

direct economic regulation, and particularly of the railroad, trucking, airline and telecommunications industries. Of

particular relevance to my statement here, I have also been a member of the Attorney General’s National Committee

to Study the Antitrust Laws (1954-56) and the National Commission on Antitrust Laws and Procedures (1978-80); I

am the co-author of Fair Competition, The Law and Economics of Antitrust Policy and have published numerous

articles in that area. I attach a copy of my full resume as Appendix A.

2. My name is Timothy J. Tardiff. I am a Vice President at National Economic Research

Associates. I have specialized in telecommunications policy issues for about the last 15 years. My research has

included studies of the demand for telephone services, such as local measured service and toll; analysis of the market

potential for new telecommunications products and services; assessment of the growing competition for

- 2 - Draft

telecommunications services; and evaluation of regulatory frameworks consistent with the growing competitive

trends. Most recently, I have participated in interconnection arbitrations, pursuant to the Telecommunications Act of

1996, in twelve states. I attach a copy of my full resume as Appendix B.

3. SBC Communications Inc. (“SBC”) and its subsidiaries Pacific Bell (“Pacific”) and Pacific

Bell Communications (“PBCOM”) seek authority for PBCOM to provide in-region interLATA services in the State

of California. The purpose of this affidavit is to assess the public interest implications of such entry.

II. THE HISTORICAL TRADE-OFF IN THE LINE-OF-BUSINESSRESTRICTIONS

A. The Issue During the Pre-divestiture Period

4. The progressive introduction of competition into the telephone business, dating back to the

FCC’s Above-890 decision in 1959 and to MCI a decade later, and AT&T’s evolving responses precipitated intense

controversy at the FCC, Congress, the Antitrust Division of the Department of Justice and the courts over how best

to reconcile the dominant position of the comprehensively integrated Bell System, on the one side, and the evolving

national policy of encouraging competition, on the other.

5. We make no effort to recount that history.1 We think it is not an oversimplification, however,

to say that once the commitment to competition was reached at the Federal level, the central issue was the extent to

which regulatory restraints on AT&T would be sufficient to ensure fair and efficient rivalry between it and its

challengers or whether, instead, it would be necessary to break up the Bell System, imposing line-of-business

restrictions on the successor companies, in order to deprive them of the power and motive to frustrate achievement of

that goal. In these intense debates, AT&T and its supporters in government resolutely proclaimed the benefits of the

comprehensive horizontal and vertical integration of the Bell System, and its adversaries tended to minimize those

asserted benefits to the point of denying their existence entirely.

6. What ultimately tipped the scales on the side of complete divestiture of local telephone service

1 A particularly thorough history is presented by Peter Temin in The Fall of the Bell System, A Study in Prices

and Politics, New York: Cambridge University Press, 1987; for an account of developments and the underlyingeconomic issues up to 1970, see Alfred E. Kahn, The Economics of Regulation, New York: John Wiley &Sons, 1970-71, reprinted by MIT Press, 1988, Vol. 2, pp. 126-152, 290-306.

- 3 - Draft

from the other operations of the Bell System—notably toll—was the developing view of the Department of Justice

that all the proposed protections against cross-subsidization, predation and exclusionary practices would be excessively

“regulatory” and ineffective, and that only a total separation of the putatively “naturally monopolistic” local telephone

service from the other potentially more competitive services would be consistent with the preservation and promotion

of competition in the latter markets.

B. The Balance of Advantages and Disadvantages has Shifted

7. The terms of the trade-off between the respective benefits of integration and divestiture have

changed drastically since the entry of the MFJ. In fact, whatever one’s evaluation of the net advantages and

disadvantages of the line-of-business restraints on the BOCs during this interval, they clearly must be reconsidered in

the light of (a) the dramatically changed factual circumstances; (b) our experience with the way competition has

worked in the interLATA market and increasing recognition of the important contribution that BOC entry is likely to

make in intensifying that competition and extending its benefits more broadly; (c) the changes in both regulatory

practice and in the market that have tended to dilute whatever power the BOCs may have had to handicap competitors;

(d) the extensive experience we have actually had since 1982 with competition between the putatively monopolistic

BOCs and rivals dependent upon them for certain inputs and (e) changes in the mix of national policies and goals

articulated most clearly in the Telecommunications Act of 1996. In our judgment, all these factors have shifted the

balance of the public interest—wherever it was in 1982—unequivocally over to elimination of those absolute

restrictions.

8. This proposition has now been endorsed, in both general terms of national policy and in

highly specific ways, by the Telecommunications Act of 1996. We have now made our choice. The Act clearly

concludes that the balance of advantages and disadvantages has shifted in favor of abandoning the line-of-business

restraints on the BOCs. It makes that abandonment conditional upon a public interest finding by the FCC; in our

opinion that criterion has been satisfied.

III. PBCOM’S ENTRY IS IN THE PUBLIC INTEREST

9. Whatever may be said in its favor, the current prohibition on interLATA entry by the RBOCs is also,

undeniably, inherently anticompetitive. In the name of preserving competitive opportunities for some, it prohibits

- 4 - Draft

others from competing entirely. The only possible justification for its continuance would be that the gains to society

from protecting the former outweigh the costs of excluding the latter and that those protections could not be achieved

by other means less costly to consumers. The remaining portions of this statement consist of an amplification of our

reasons for concluding that the balance of advantages and disadvantages has shifted in favor of abandoning the line-of-

business restraints on the BOCs.

10. The costs are great. The excluded competitors are large and potent. The market from which they are

excluded—a market whose boundaries have been defined entirely arbitrarily, so far as the relevant technology and

economics are concerned—has distributed the benefits of rapidly improving productivity and competition imperfectly

and incompletely. The customers that have benefited disproportionately little are precisely the ones that the excluded

BOCs would have the greatest comparative advantage in serving: those companies will therefore be the most logical

and effective competitors for residential and small business services initiated within their own regions. Unlike MCI

and Sprint in 1984, they already serve all of these customers. Supplying additional services to an existing customer

is far easier—and less costly—than establishing a commercial identity and presence before new ones.

A. The Current State of InterLATA Competition

11. The most fundamental change in interstate long-distance markets since the 1984 divestiture and the

one most relevant in the present context is that this portion of the industry is not only dominated by AT&T, MCI

and Sprint, but consists exclusively (apart from a few corridor areas that were exempted from the long-distance

restriction) of companies entirely separate from—indeed antagonistic to—the successor Bell Operating Companies. In

addition, the rapid development and expansion of fiber-optic technology has radically altered cost structures, much as

advances in microwave technology did a decade or two earlier, and contributed to a dramatic expansion—approximately

a trebling—of total network capacity in just 11 years. Whereas previously there was only the one nationwide long-

distance network, totally integrated with companies accounting for some 80 percent of all local service, there are now

nearly four backbone long-distance networks, fully separated from the BOCs. Those four clearly do compete with one

another, as well as with a large fringe of much smaller rivals, facilities-based and resellers. That competition is,

however, far from fully effective; and its deficiencies are ones that competitive entry by the BOCs is most likely to

remedy.

- 5 - Draft

12 . Long distance prices, access charges and margins, overall

13. Since divestiture—that is, between the beginning of 1984 and the end of 1997—long-distance prices

have declined about 26 percent in nominal dollars and about 53 percent relative to the Consumer Price Index (CPI).2

What is at least equally striking, however, is that these decreases have been more than fully “explained” by FCC-

mandated decreases in the prices that the long-distance carriers pay to the local exchange carriers for access to their

networks. (We emphasize that the evidence we present here is of the change in prices alone, for given volumes of

usage. This is not the same as average revenue per minute, as we will explain presently.) Turning first to the

reduction in access charges, according to the FCC, the average interstate switched access charge per conversation

minute fell about 69 percent from May 1984 to December 1997—a decline of about 12 cents per conversation

minute. 3 To make this reduction possible, the FCC imposed monthly subscriber line charges directly on telephone

customers, shifted costs to the intrastate jurisdiction through changes in separations rules and adopted price cap

formulas that mandated reductions over time in those charges. We show below that the long-distance carriers failed to

reflect fully in their prices these reductions in their direct costs of providing long-distance services.

14. According to a recently-published estimate, AT&T’s annual carrier access bill dropped by about $10.3

billion between 1984 and April 1995 (holding volumes constant, in order to reflect the pure change in price), while

over the same period of time the bills that its customers received fell by about $8.5 billion (once again holding

volumes constant).4 If anything, the divergence between the changes in access charges and long-distance prices has

become even more striking in the last few years: while access charges have continued to fall, prices net of access

charges have risen.5 Access charges per conversation minute decreased by about 23 percent between April 1995 and

2 As measured by the consumer price index for interstate long-distance. U.S. Department of Labor, Bureau of

Labor Statistics, Office of Publications, Division of Information Services.

3 Federal Communications Commission, Trends in Telephone Service, February 1998, Table 1.2, p. 4.

4 William E. Taylor and J. Douglas Zona, “An Analysis of the State of Competition in Long-Distance TelephoneMarkets,” Journal of Regulatory Economics, May, 1997, pp. 227-256.

5 Even when average revenue per minute is used to measure prices—a measure which we demonstrate belowoverstates the decline in toll prices—AT&T’s own data show that between the end of 1994 and the end of 1996,prices fell by no more than the reduction in access charges. (Affidavit of R. Glenn Hubbard and William H. Lehron behalf of AT&T Corporation, AT&T Exhibit G, In the Matter of Application by SBC Communications Inc. for Authorization Under Section 271 of the Telecommunications Act of 1996 to Provide In-Region, InterLATA Services in the State of Oklah oma , CC Docket No. 97-121, Figure 3).

- 6 - Draft

December 1997.6 In the same period, interstate toll prices (as measured by the CPI-telephone, interstate) increased by

0.4 percent. Obviously interstate toll rates and access charges have not changed to the same degree since 1995—they

have not even changed in the same direction. Those who argue that IXCs have in fact flowed through all of the access

charge reductions in retail prices typically base their contentions on changes in average revenues per minute (ARPM).

But the decline in ARPM over the last several years almost certainly overstates the actual decline in prices, as the

following considerations demonstrate:

• Suppose AT&T customers demand ten minutes of message toll service (MTS) for each minute of wide

area toll service (WATS) (and no other services) and that the price of MTS (per minute) is twice that of

WATS. If MTS and WATS prices increase by 1 percent but demand for WATS grows at 50 percent per

year while MTS demand grows at 10 percent per year, then the ARPM of usage decreases by about one

half a percent. In other words, ARPM declines despite the fact that both of the component usage prices

have increased.

• Suppose the prices in the discount plan remain fixed, but customers are able to receive lower effective

marginal prices when their usage expands (e.g., because they have installed fax machines). In that case,

ARPM would decline not because the price of usage declined but because customer demand increased.

In fact, in this example, ARPM could decline even when the prices for low and high volume users

increase if volume growth is sufficiently large. For example, suppose that at the starting point the

price schedule for an average user is 20 cents for the first 100 minutes and 15 cents per minute

thereafter; that those two price brackets are increased to 21 cents and 16 cents respectively, but that the

usage of the average customer grows from 150 to 250 minutes. In those circumstances, the ARPM

would have been 18.33 cents per minute before the price increase and 18.0 cents afterward.

• Suppose, next, that the own-price elasticities for different services are different, even when the

percentage price change for each is identical. For example, suppose (1) the price of service A is one

dollar per minute, ten minutes are sold, and the A own-price elasticity is -0.2, and (2) service B has a

price of 50 cents per minute, a demand of ten minutes and an own-price elasticity of -5.0. If the price

of each of the services decreases by 10 percent, ARPM will decrease by 17 percent. Observe that the

6 Federal-State Joint Board Staff, FCC Monitoring Report, May 1997, Table 5.12, p. 616.

- 7 - Draft

anomalous result is not caused by substitution of lower-priced service—their demands are assumed to be

independent in this example—but reflects the inadequacies of the index itself.

• ARPM (as measured by the IXCs) goes down when facilities bypass is initiated by the end user, but

this decline will overstate the effective reduction in price or cost savings enjoyed by the customer. For

example, when a large customer builds a private network bypassing LEC access facilities, AT&T’s

ARPM from that customer could go down (relative to its MTS rates, which include the carrier access

charge) but the cost per minute to the customer would have to reflect both AT&T charges (ARPM from

AT&T’s perspective) and its own network costs.

15. These examples illustrate two general tendencies for ARPM to exaggerate recent price reductions. First,

when different prices are charged to different customer groups or for different services, differences in the rates of growth

of their sales (whether or not caused by the change in prices) can cause aggregate ARPM to overstate price reductions.

Second, ARPM from any one IXC will misstate end user costs when end users assemble services through a variety of

vendors.7

16. The long-distance carriers have strongly criticized the high charges they typically have to pay the LECs

for access to their networks. Those charges have indeed been set by regulators far above cost, deliberately, in order to

perpetuate the subsidy that had, before AT&T was split up, flowed from similarly inflated long distance charges to

hold down the rates for basic residential service. By any measure, however, AT&T’s own average markups above

those access charges and above its own long-run incremental costs continue to be at least as large as the markups in

the access charges themselves.

17. In 1996 AT&T’s average revenue per minute was about 17.3 cents and its access charges averaged about

5.3 cents per minute, thus producing revenue net of access charges that averaged 12 cents—a figure virtually

7 Changes in average access cost per minute (AAPM)—reflecting changes in payments to LECs—will likewise

exaggerate the reductions in access costs that IXCs or their customers have actually realized when they bypass LECfacilities. The tendency that we have described for ARPM to overstate price reductions is therefore offset to someextent by the similar tendency for AAPM to overstate reductions in access charges. At most the errors cancel oneanother. What is far more likely is that ARPM net of access overstates the actual reduction in prices or costsborne by customers: ARPM is likely to err by more than AAPM, because in every situation where AAPM isbiased downward (i.e., when LEC access is bypassed) so is ARPM, and there are a variety of other situations inwhich only ARPM is biased downward.

- 8 - Draft

unchanged from 1994.8 Incremental toll cost estimates range from 1-2 cents per minute at the low end to about 6-7

cents at the high end. The higher values include non-network costs such as overhead, customer and marketing costs,

not all of which are likely to be properly includable in the incremental cost of offering the service. Incremental costs

for access to LEC networks are only 1 cent per conversation minute or less.9 Adding the access charge and the

incremental production cost produces a range between 6.3 and 12.3 cents per minute for long distance and margins

that range between 5 and 11 cents, compared with margins of 5 cents for access charges.10

18. Thus, while a group of economists assembled by AT&T are correct in asserting that

8 AT&T’s average revenue per minute was 18 cents in 1994 and its access charges averaged 6 cents per minute

(AT&T ex parte letter in CC Docket No. 94-1, March 21, 1995). Data presented by Hubbard and Lehr (op. cit.,Figure 3) show that while AT&T’s average revenue per minute declined by about 0.7 cents between 1994 and1996, this decline was almost identical to the decrease in access charges over the same period.

9 AT&T economists cite incremental costs of carrier access between 1/3 and 1/2 cents per minute. They are silenton the question of long-distance incremental costs. D. Kaserman, J. Mayo, M. Crew, N. Economides, G.Hubbard, P. Kleindorfer and C. Martins-Filho, “Local Competition Issues and the Telecommunications Act of1996,” prepared on behalf of AT&T, July 15, 1996, p. 27.

10 As the ranges of the figures for AT&T demonstrate, this comparison is highly sensitive to the estimate weemploy for the LRIC of its operations.

The sources of the 1-2 cent per minute figure are Lewis J. Perl and Jonathan Falk, The Use ofEconometric Analysis in Estimating Marginal Cost, Presented at Bellcore and Bell Canada Industry Forum, SanDiego, California, April 6, 1989, Table 2; R.W. Crandall and L. Waverman, Talk Is Cheap, Washington:Brookings, 1996, p. 92;Paul W. MacAvoy, The Failure of Antitrust and Regulation to Establish Competitionin Long-Distance Telephone Services, MIT and AEI Presses, 1996, p. 115, citing an estimate by WhartonEconometric Forecasting Associates; and Lehman Brothers, Telecom Services: Buy the Bundle Builders, Get theGrowth, March 18, 1996: “Large customers and large resellers can purchase transport at close to long-runincremental costs, or at about the $0.02 per minute in average depreciation and network engineering costs of themajor players (this is the rate that the federal government recently negotiated on its multiyear FTS 2000 contractfor POP-to-POP transport).” (p. 28) It seems likely, however, that these various figures fail to include suchmarketing, customer service and overhead costs as would be properly part of the LRIC of the total service.

The bases for the 6-7 cents at the high end are the statement by Crandall and Waverman that “theincremental cost of long-distance service is probably no more than 5 cents and surely no more than 10 cents perminute” (pp. 276-277), considered together with their citation of the Company’s reported operating, marketingand customer service and general and administrative costs, which they take to be on the order of 3.9, 3.7 and 2.9cents per minute, respectively (p. 142). While all or a large portion of the first two categories are probably partof its TSLRIC (as contrasted with the LRIC of smaller increments), it seems highly unlikely that that would betrue also of the general and administrative costs. The estimate of 11.4 to 12.4 cents, including the access charge,is from their Joint Affidavit on behalf of Ameritech Michigan, In the Matter of Application of Ameritech Michigan Pursuant to Section 271 of the Telecommunications Act of 1996 to Provide In-Region, InterLATA Services in Michigan , CC Docket No. 97-137).

Clearly, some of these last costs are average, rather than marginal in nature. Since average costsfor this industry tend to exceed incremental cost, this last Crandall and Waverman estimate does not contradictour range. Even their upper limit produces a 5.6 cent markup, one just as large as is incorporated in the accesscharge.

- 9 - Draft

(i)f there is one factual issue in the telecommunications industry upon which there is virtuallyunanimous agreement, it is that carrier access services are currently priced well in excess of theirincremental costs11

they are wrong in their selectively pejorative treatment of those particular prices. While the LECs mark up their

carrier access prices over incremental cost by (say) a nickel, AT&T marks up long distance prices over incremental

cost by at least a nickel and perhaps several cents more. Thus the AT&T economists are wearing blinders when

they condemn the former markup—whose explicit purpose is to contribute to achieving the public policy goal of

universal subscription—as a “regulatory-sanctioned pricing distortion”—“clearly an anathema to economic

efficiency,” with cumulative social costs “certain to run into the billions of dollars per year” (at 27)—while AT&T

itself extracts for its shareholders at least as high a markup per minute in its own retail toll rates, in a market it

claims to be fully competitive.

18. As the foregoing discussion has demonstrated, cost measurement in these markets is fraught with

difficulty and contentiousness, with small absolute differences in estimated costs producing large differences in the

estimated percentage markups. Moreover, there may well be large differences between the short-run marginal costs

that it might be sufficient for individual sales, to individual customers, to recover for the transactions to be

profitable and the long-run incremental costs that companies would have to recover if they were to continue to

provide service to large groups or the totality of their customers. There is probably a large difference also between

the total long-run incremental cost of serving large business customers, on the one side, and residential, on the

other. Even with all these qualifications, the prices at which sales are actually being made provide at least a

suggestion of the upper bound of incremental costs. And this type of information tends to corroborate our upper

bound of 11 cents per minute. For example, long distance carriers are offering intraLATA toll service for 8 cents

per minute in California—a figure roughly equivalent to an interLATA price of 10 to 11 cents per minute, because

in-state access charges in California are lower than the national average. AT&T offers in-state toll calls at 5 cents

per minute in Connecticut.

19. There are two possible interpretations of these low prices of intraLATA offerings. One is that the

IXCs’ costs for these services may, in fact, be higher—as they almost certainly are in the Connecticut case—but the

carrier is able to offer these prices profitably because it expects to bundle the intraLATA with compensatingly higher-

11 Op. cit ., supra note 9, p. 26

- 10 - Draft

margin interLATA offerings—thus confirming our basic point. Alternatively, the prices are compensatory in

isolation, implying that costs are no higher than that. Either interpretation supports our conclusion that the IXCs’

prices contain healthy margins.

20. Distribution of the benefits of competition between largeand small users

21. Large business customers have benefited greatly from the new competition in the long distance

business. The combination of the large volume of their business, on the one side, and, on the other, the very wide

gap between the incremental costs of the IXCs and their average rates has forced the IXCs into intense competition in

offering special contractual arrangements, incorporating both special prices and new and superior service offerings. As

the FCC has observed, large customers now solicit proposals from multiple vendors and negotiate terms directly with

the interexchange carriers.12

22. The price reductions have been dramatic: the average charge for a minute of long-distance service for a

large corporation appears to have fallen by about 80 percent (nominally, and even more in inflation-adjusted dollars)

since 1983. 13 Prices in 1983 were at about 35 cents per minute and are now at about 7 cents per minute for the

largest business customers.14

23. Small residential subscribers have not benefited to anything like the same degree.15 In contrast with

the estimated 80 percent decline for large business customers, long distance prices for residential consumers (as

12 Report and Order, In the Matter of Competition in the Interstate Interexchange Marketplace , CC Docket No. 90-

132, FCC, 6 FCC Rcd. 5880, 5887, Adopted: August 1, 1991, Released: September 16, 1991, par. 38.

13 Michael T. Felix, “Preparing the Market for Enhanced Service Implementation,” Telephony, Vol. 230, No. 13,March 25, 1996, p. 40.

14 David Rohde, “VPN Rates On The Way Down,” Network World, December 2, 1996, Vol. 13, No. 4g, pp. 1,14-15; Table 7.12, Statistics of Communications Common Carriers, Federal Communications Commission,1988/1989 Edition, p. 286; Felix, “Preparing the Market...,” Telephony, p. 40; Crandall & Waverman, Talk IsCheap, p. 125; “GSA Tells Congress FTS 2000 Prices Beat Market Rates,” Telecommunications Reports,March 8, 1993.

15 The FCC therefore recently admonished participating parties in Sec. 271 proceedings: : “...in determining theextent to which BOC entry into the long distance market would further competition, we would find it morepersuasive if parties presented specific information as to how such entry will bring the benefits of competition,including lower prices, to all segments of the long distance market. Memorandum Opinion and Order, In the Matter of Application of Ameritech Michigan Pursuant to Section 271 of the Communications Act of 1934, as amended, To Provide In-Region, InterLATA Services in Michigan , CC Docket No. 97-136, Released August 19,1997, par. 16. (“Ameritech Order”)

- 11 - Draft

measured by the CPI) have declined by about 24 percent since 1984, which implies a price decrease of about 7 cents

per minute for them.16 During the same period, access charges declined 12 cents.

24. About one-half of the apparent 5 cents per minute increase in residential rates net of access charges

occurred after the beginning of 1994. Low volume users, who typically buy from the basic tariff, fared even worse.

AT&T increased the basic rate for residential interstate calling in January 1994 by an average of 6.3 percent—an

increase targeted at low-volume subscribers as well as ones under its residential calling plans indexed to the basic

rate.17 It increased rates further by 3.7 percent in December 1994,18 and 4.3 percent and 5.9 percent, respectively, in

February and December 1996.19 In each instance, MCI and Sprint followed in lock step.20 These increases over that

three year period occurred in the face of a continued drop in carrier access charges—by about 10 percent—during the

same period.21 During 1997, AT&T reduced its basic rates in mid-year, in fulfillment of its promise to the FCC to

pass through the latest reduction in access charges.22 This reduction was offset somewhat, however, by AT&T’s

restructuring of the discount periods in November of that year, the effect of which was to increase the average price.

Our colleagues at NERA estimate that the combined effect of the two 1997 rate changes was a reduction in basic rates

of 3.4 percent, or 0.65 cents. Significantly, this reduction fell short of the 0.78 cent reduction in access charges that

occurred in the middle of 1997. Over the entire six year period from the beginning of 1992 to the close of 1997,

AT&T increased these basic rates—paid by the majority of its residential customers—by 21.1 percent, while access

16 AT&T reported an average revenue per minute (ARPM) for its Consumer markets of about 23 cents in 1994. Ex

Parte Presentation in Support of AT&T’s Motion for Reclassification as a Nondominant Carrier, Attachment I,Letter from C.L. Ward, AT&T, to William F. Caton, FCC, dated February 8, 1995. We used changes in thepertinent CPI index to estimate ARPM for other years, in particular, 1984 and 1997.

17 “AT&T Proposes $750 Million Rate Hike, New Calling Plan Aimed At High-Volume Residential Users,”Telecommunications Reports, January 3, 1994.

18 John J. Keller, “AT&T and Rivals Boost Rates Further,” The Wall Street Journal, November 29, 1996, p. A3.

19 “AT&T Follows MCI, Sprint with Long Distance Rate Increases,” Telecommunications Reports, December 2,1996.

20 See par. 30, below, on the FCC’s expression of concern about this pattern of price leadership.

21 Access charges per conversation minute declined by 9.34 percent (from 6.66 cents to 6.04 cents) between July1993 and May 1997, although there occurred a brief intervening increase from 6.66 cents to 6.89 cents, or about3.4 percent, in July of 1994. (See Table 5.2 in FCC Monitoring Report, May 1997, p. 616.) On the furtherreduction in mid 1997, see note 22, immediately following, and par. 29, below.

22 According to the press account, this amounted to an 8 percent decrease in those basic rates. New York Times,July 1, 1997.

- 12 - Draft

charges declined.23

25. We must of course consider the possibility that this dramatic difference in the trend of long-distance

charges to large business and small residential customers represented a correction of a previous distortion—

specifically, cross-subsidization of the latter rates at the expense of the former—such as would be expected to take

place with the introduction of effective competition. This is the claim of Bernheim and Willig—that the costs per

minute of serving low-volume customers is significantly higher than of serving high-volume ones because of the

presence of fixed customer costs, such as billing, collections, fraud and customer service, that do not vary with usage

for any given subscriber.24

26. For purposes of testing this possible justification of the increase in AT&T’s long-distance charges,

net of access fees, to small residential users, we use the Company’s own definition of low-volume residential

customers as ones with long distance charges of $10 per month or less: these are the people who pay the basic rates

that have been subject to the recent increases.25 AT&T says that more than half of its customers fall in this category.

It also asserts that customers with average monthly bills under $3 are below the “break-even point.”26 This claim

suggests that, to the extent these last customers can be segregated, rates charged them would indeed be expected to

increase under real-world competitive conditions, even though presumably the marginal costs of their long-distance

calling would be no higher than for higher-volume customers.27 But it would neither explain nor justify the increases

in basic usage rates undiluted by discount offerings that at least half of residential users were forced to pay on grounds

of either average cost per customer or marginal cost of usage: the group in the $3 to $10 per month range, with four

times the usage of the ones below AT&T’s claimed $3 break-even point, must have been making a very large

23 Calculated from contemporaneous news accounts and NERA’s analysis of AT&T’s 1997 basic rate changes.

24 B. Douglas Bernheim and Robert Willig, “An Analysis of the MFJ Line of Business Restrictions,” December 1,1994. Attachment G, Ex Parte Presentation in Support of AT&T’s Motion for Reclassification as aNondominant Carrier, CC Docket No. 79-252, April 20, 1995.

25 Letter of C.L. Ward to W.F. Caton Dated March 9, 1995 Re: Ex Parte Presentation CC Dockets Nos. 79-252,93-197, 80-286; D.J. Quinn, The Light User Segment of the Long Distance Market, March 8, 1995, p. 8.

26 Ibid.

27 That is to say, under theoretically pure competition, under which rates for usage would be held to marginal(usage-sensitive) costs, the higher average costs of the very low-volume users would not be reflected in usagerates higher than those charged heavier users. Since, however, the former particular customers would in thosecircumstances not be worth serving at all in these circumstances, providers of long-distance service to themwould have to be compensated for the fixed per-customer costs either by levying a flat charge on them or byfinding a way of charging them discriminatorily higher rates for usage.

- 13 - Draft

contribution to company profits.

27. The only possible explanation for their having fared so much less well under competition than

large business customers is that the long-distance carriers serving them have found it easier to resist the temptation

to engage in price competition for their patronage than for that of the big users. We observe repeatedly in AT&T’s

pricing behavior the kind of price leadership that denies low-volume customers the full benefits of competition,

once the adoption of alternative regulation permitted it to increase its basic rate schedule.

28. Incumbent IXCs have recently introduced discount plans that they claim provide lower prices to

smaller users.28 For example, Hubbard and Lehr claim that users making $5-$10 per month of long-distance calls

are benefiting from AT&T’s one rate (15 cents per minute) plan. In fact, their own data (Figure 5) show that

customers in the $5-$10 range experienced price increases between 1992 and 1996.29 The only way they are able to

show a price decrease thereafter is by constructing an artificial price of 15 cents per minute for 1997, on the ground

that customers “need pay no more” than this amount. And even their artificial 1997 price does no better than

restore users in the $5-$10 category to the price that their own data (in Figure 5) show those customers were

charged in 1990, despite the continued decline during this period in the access charge their suppliers paid.

29. Marybeth M. Banks makes similar arguments on behalf of Sprint, in response to the 1997 271

application of SBC Communications Inc., et al. in Oklahoma.30 Although her own Figure 1 shows that prices for

offerings that were available throughout the period she examined increased steadily after 1992, she claims that

prices have recently declined because of the introduction of new discount plans. It is clear, however, that (1) her

conclusion depends on what proportion of customers have actually chosen these plans (she claims, without

documentation, that a majority have actually done so; even if this assertion is correct, it clearly means that the

average price will have been higher than the price available under the most attractive plan) and (2) the deep

discounts are a recent phenomenon, introduced in 1996 or later. Thus, whatever may be said of the current

competitiveness of the long distance market, Sprint, like AT&T, has in effect conceded that until 1996 or 1997

28 Affidavit of R. Glenn Hubbard and William H. Lehr, op cit. Statement of Marybeth M. Banks, Sprint

Communications Company, L.P., Submitted to the Federal Communications Commission, CC Docket No. 97-121, May 1, 1997.

29 They do not even attempt to show what has happened to customers in the $0-$5 range, a group that hasdisproportionately experienced the undisputed increase in the basic tariff.

30 Op. cit.

- 14 - Draft

prices were not consistent with the kind of intense competition that they claim prevails at the present time.

Moreover, Ms. Banks concedes that most subscribers paying the basic rate (i.e., small users), which she says are

disproportionately served by AT&T, have not benefited from Sprint’s asserted price reductions because it tends not

to serve them.

30. The price data offered by the witnesses for AT&T and Sprint therefore confirm our conclusion that

small users did not, in the prices they actually paid, receive the full benefit of the reductions in the access charges

that their service providers were forced to pay the local telephone companies, let alone from any compression in

those providers’ margins above costs. AT&T’s promise to the Consumer Federation of America and the FCC in

the spring of 1997 that it would pass through to small residential customers promised reductions in access

charges—and the implied promise of the Commission to monitor its compliance—was itself in effect a concession

to the previous complaints both of these parties expressed before that time about the sufficiency of competitive

pressures alone. And of course a promised rate reduction equal only to a further reduction in access charges would

do nothing to reverse the large price increase net of those charges in the preceding five years.

B. The Benefits of PBCOM’s Entry into the InterLATA Market

30. The FCC, having previously expressed “serious concern” about the apparent pattern of price leadership

that had emerged in the long-distance business,31 has recognized that “the Act provides the best solution to any

problem of tacit price coordination...by allowing for competitive entry in the interstate interexchange market by the

facilities-based BOCs....”32 PBCOM’s entry will promote effective competition by reducing the ability of the IXCs

to engage in such quasi-collusive pricing at the expense of small residential customers.33 It will have this effect not

31 Order, In the Matter of Motion of AT&T to be Reclassified as a Non-Dominant Carrier , Federal

Communications Commission, FCC 95-427, Adopted: October 12, 1995; Released: October 23, 1995; par. 81-83.

32 Notice of Proposed Rulemaking, In the Matter of Policy and Rules Concerning the Inte rstate, Interexchange Marketplace Implementation of Section 254(g) of the Communications Act of 1934, as amended , CC DocketNo. 96-61, Federal Communications Commission, 11 FCC Rcd 7141; 1996 FCC Lexis 1472, FCC 96-123,Adopted: March 21, 1996, Released: March 25, 1996, par. 81, footnote omitted.

33 Our assertion is supported by the Affidavit of Professor Marius Schwartz, dated May 14, 1997, filed on behalf ofthe DOJ in response to Southwestern Bell’s Oklahoma petition (CC Docket No. 97-121), in his discussion ofthe significant efficiency advantages a BOC would bring to the long distance market. He states specifically that“A BOC would be especially well placed to address lower-volume customers.” (Schwartz, p. 33, par. 96) OnProfessor Schwartz’s later contention that the potential benefits of competitive entry into local Telcom markets

- 15 - Draft

merely because it will increase the number of large, well-positioned competitors in its region by one; more

important, it will introduce a large competitor that (a) begins with a zero market share, (b) is, by virtue of the likely

large volume of its purchases, in a particularly strong position to take advantage of the large excess transport capacity

of the present market-dominating IXCs34 and (c) has cost characteristics and (d) a current mix of services uniquely

promising to remedy the single greatest deficiency in the performance of the long-distance market.

31. While they would be a powerful market presence within their respective LATAs from the outset, the

RBOCs would begin interLATA competition with a zero market share—much like the competitive position of the

interexchange carriers when they were permitted to offer service within the LATAs; they would have to offer attractive

prices to break into that market.

32. A more effective competitor

33. There are of course already hundreds of resellers of long distance service. Typically purchasing these

services from a facilities-based carrier under long term contracts, such as AT&T’s contract tariffs or Tariff 12 options,

they have mainly targeted business customers with monthly long distance bills of several hundred dollars. While

PBCOM is considering deploying some facilities, it will probably first provide services on a resold basis, because the

current capacity of the combined interLATA networks is sufficiently large to meet immediate needs. PBCOM is well

positioned to be highly effective in obtaining underlying services from IXCs and more effective than pure resellers in

reaching small customers.

34. The present resellers have been disadvantaged by lack of a strong brand name. They have therefore had

to rely heavily on direct personal approaches and negotiations, which are unlikely to be economical in soliciting the

business of low-volume users. Pacific, in contrast, already serves those customers and has a strong established brand

name.35

35. At the buying end, PBCOM is likely to be able to negotiate much more favorable terms with the

IXCs than the present resellers, because of the large volumes of purchases to which it is likely to be able to commit

itself. (And of course volume discounts at favorable prices are a common phenomenon across the entire economy in

would be even larger, see par. 42, below.

34 FCC 95-427, par. 57-62.

- 16 - Draft

such circumstances.) Within the last two years, SBC, NYNEX, Bell Atlantic, Ameritech, BellSouth and GTE have

entered into agreements with such IXCs as AT&T, Sprint and LDDS WorldCom to resell their long-distance services,

at prices in the 1-2 cents per minute range.36 The addition of switched access charges (by imputation at the

originating end and by payment to other LECs at the terminating) of about 5-6 cents per conversation minute and any

additional costs associated with customer service, marketing, billing and collections would evidently produce total

LRICs below 10 cents per minute37—leaving wide room to under-price the IXCs. (Recall that AT&T’s average

revenue per minute was 17.3 cents in 1996.38)

36. Economies of scope

37. We have already referred to AT&T’s claim that its service to residential customers does not break even

until their long-distance bills reach $3 a month, because there are fixed costs of serving each customer regardless of

his or her volume of usage. The BOCs already incur many of those costs because they already serve most of those

customers in-region: this means the incremental customer costs of adding long-distance to their present mix of

services would be very small.

38. Another way of characterizing this positive case for removing the barriers to BOC provision of

interLATA services is the fundamental proposition that they—like all other potential entrants—possess strong

capabilities and potential comparative advantages in the interLATA markets and are therefore well-positioned to make

important, welfare-enhancing contributions that they are currently prohibited from making. These advantages are an

example of the economies of integration or of scope. They arise whenever conduct of two or more activities in a

single firm is more economical than if they were conducted by separate firms—generally because of the presence of

facilities, personnel or capabilities that can be shared by them. The achievement of these economies is frustrated

35 Ameritech Order, par. 15.

36 “NYNEX To Resell Sprint Service,” Bloomberg LLP, April 24, 1996. “Bells, GTE Lay Out MarketingStrategies, Swap Success Stories at New York Conference,” Telecommunications Reports, September 26, 1996.C. McElroy, “BellSouth To Buy Time On AT&T’s Long-Distance Network,” Bloomberg LLP, June 19, 1996.

37 See the various estimates summarized in note 10, above. Observe that the 6-7 cents per minute upper limit onnon-access incremental costs that we cite (and question) there are as high as they are because they are estimates ofthe incremental cost of the entire long-distance service. The cost (particularly marketing and customer servicecost) of a local telephone company adding long-distance service to its offerings would clearly be substantiallylower than that.

38 See note 8, above.

- 17 - Draft

when markets are balkanized and firms in one arbitrarily excluded from another—arbitrarily in the sense that the

reasons have no grounding in technology or economics—as occurred when the MFJ excluded the RBOCs from the

interLATA business and state regulators had yet to allow entry by the interexchange carriers into the intraLATA

business. Except if they are prevented by statute or regulation from availing themselves of these opportunities, the

cost to the BOCs of adding the provision of interLATA toll to a plant and organization already designed to provide

only intraLATA services would be lower than the cost of setting up a separate entity to serve that market alone—

because, obviously, they already possess a large portion of the capabilities of taking on those additional functions.39

39. These at present incompletely utilized talents or facilities represent opportunities for the firm as a

potential supplier of those other common products or services. To minimize their total costs and to recover their

fixed and common costs—particularly under the constraints of competition—telecommunications firms must

constantly seek out and develop services or lines of business that generate economies of scope with their current

service mixes.

40. Both the BOCs and IXCs are precluded from taking full advantage of those economies today—the

former because they are prohibited from combining interLATA traffic with their current intraLATA offerings, the

latter to the extent they are unable to add local exchange service to their long-distance offerings. That of course is

why the Telecommunications Act requires the LECs to make those services available to other providers at wholesale

and links the removal of the first handicap, symmetrically, to the removal of the other.

41. That a central purpose of the Telecommunications Act is to achieve a symmetrical lifting of

restrictions on the exploitation of economies of scope by any and all participants in the market is manifest

throughout its text. One particularly apt illustration is the provision that restricts large IXCs from jointly marketing

their own interLATA services with local exchange services purchased from a BOC until the latter company is itself

authorized to provide interLATA services in region or until 36 months from the date of its enactment.40 This clear

intention of the Act to free all competitors, symmetrically, to take advantage of their own particular economies of

scope is further reflected in its explicit stipulation that the requirement on the BOCs to offer their interLATA services

39 The Act’s requirement that the RBOCs offer interLATA services via a separate entity during a transitional period

would of course temporarily deprive them and the public of some of those benefits.

40 Telecommunications Act of 1996, Pub. L. No. 104-104, February 8, 1996, Sec. 271(e). The same time periodapplies to the separate subsidiary requirement on the BOCs under Sec. 272. Sec. 272 (f) (1).

- 18 - Draft

through a separate subsidiary be limited to three years from the date of authorization, unless extended by FCC rule or

order. Clearly that requirement, which cannot but entail sacrifice of some of those economies in the case of the

BOCs, should be removed at the earliest practicable date.41

42. BOC entry is almost certain to produce benefits besides driving and holding prices closer to present

incremental costs. Competition is a dynamic process. It exerts powerful pressures on suppliers to improve

productivity and to be innovative. The incumbent interexchange carriers may contend that competition is already

sufficient for these purposes. Such an assertion cannot possibly be valid, especially when the prospect is one of entry

by a new type of competitor—different in the mix of services it provides, the distinctive features of its technology and

the depth of its customer base. Elimination of the barriers to this new and powerful source of competition will

introduce new, additional stimuli to improving productivity and innovation as well as to genuine price competition

for the patronage of residential and small business customers—a protection essential now that the FCC’s approval of

AT&T's petition to be classified as non-dominant has virtually eliminated previous regulatory restraints.

43. In sum, the costs to consumers of a continued prohibition of the BOCs offering interLATA service

are very large: this proposition is, we submit, indisputable. What remains to be considered is whether the historical

purpose of that restriction—to encourage the transformation of the long-distance business from monopoly to

competition by eliminating any incentive on the part of the local telephone companies to use their monopoly power

to exclude rivals from a fair opportunity to compete—continues to justify consumers continuing to bear those heavy

costs. While we propose seriously to assess the dangers of the BOCs engaging in such exclusionary tactics, it need

not bias that discussion if we point out at the outset that our preceding analysis has already definitively answered that

question: there is not the slightest doubt that entry by the BOCs into the long-distance business will intensify

competition. The converse of that proposition—which we now proceed to expound—is similarly indisputable: there

is not the slightest possibility that they could so expand the zero market share with which they begin, debilitate such

competitors as AT&T, MCI and Sprint to such a point as to drive them from the market, and thereby restore the

vertically integrated monopoly of local and long-distance service that it was the purpose of the consent settlement of

1982 to dissolve.

41 The FCC’s implementation of the Act principally limits joint activities to the areas of sales and customer

support.

- 19 - Draft

44. In his Supplemental Affidavit in support of the recommendation of the Department of Justice that

applications by RBOCs for lifting of the prohibition on their offering interLATA service be denied until they have

more fully satisfied the burden of demonstrating their cooperation in establishing the conditions for competition at

the local level, Professor Schwartz contends that the potential benefits of that local competition greatly outweigh

the incremental benefits of RBOC entry into the interLATA market, mainly, it appears, because the market for

local services is much the larger of the two and local markets today are “rife with distortions.” (p. 8)42 What his

argument almost totally ignores is the fact that those distortions—which we agree are enormous—are regulatorily

imposed and are unlikely to be eliminated by competitive entry, so long as regulators persist in their present

policies of geographically averaging and suppressing prices of basic residential service far below economically

efficient levels, except if and as the ILECs are deprived of a reasonable opportunity to recover the costs of that

cross-subsidization. Unless and until that occurs, local competition will consist, preponderantly, of cream-

skimming, concentrating on the services whose prices have been and continue to be grossly inflated to finance that

cross-subsidization.43 There can be no assurance that such competition, artificially encouraged, is efficient. Any

insistence on the enormous benefits to be achieved by more local competition that simply ignores these regulatory

barriers and distortions, as Professor Schwartz essentially does, amounts in effect to writing Hamlet without

mention of the Prince of Denmark.

IV. THE RISKS OF ANTICOMPETITIVE CONDUCT AFTERENTRY ARE SMALL AND OF SUCCESSFUL SUPPRESSIONOF COMPETITION NIL

43. The contentions that the RBOCs should not be freed of the line-of-business restrictions on them because

they may be expected, if freed, to engage in anticompetitive conduct have been preponderantly (a) hypothetical—

42 Exhibit 2, dated November 3, 1997, Evaluation of the United States Department of Justice, In the Matter of

Application by BellSouth Corporation, BellSouth Telecommunications, Inc., and BellSouth Long Distance, Inc., for Provision of In-Region, InterLATA Services in South Carolina , CC Docket No. 97-208, November 4,1997.

43 Contrary to Professor Schwartz’s claim, it is not even clear that residential customers have enjoyed lower pricesfor the imperfectly competitive interLATA services than they have for the still regulated local services. Afteraccounting for the regulatorily-imposed shift in costs from toll to local prices, represented by the $3.50 federalsubscriber line charge, we calculate that local prices (adjusted for inflation) actually decreased more than didinterstate toll prices—both of which we measure with their respective consumer price indices.

- 20 - Draft

reasoning from their asserted continuing monopoly power and incentives to engage in such conduct and (b)

anecdotal—citing asserted instances of such conduct. The combination, however, of (1) the regimes already in

existence under which the BOCs had been providing equal and non-discriminatory access to their facilities by the long-

distance companies with which they now seek the opportunity to compete, even before passage of the

Telecommunications Act of 1996 and (2) the comprehensive requirements of the Act itself and of the FCC’s rules

implementing its provisions makes the likelihood of discrimination against competitors remote and of any such

discrimination effectively precluding their successful competition nil. This prediction is confirmed by the long

history—typically ignored by opponents of the present petitions—of successful competition between the BOCs and

competitors dependent on them for essential inputs. We proceed in this section to review these several considerations

and that history.

A. Existing Safeguards Against Discrimination: audited service qualitystandards, monitoring by customer/competitors, penalties andopportunities for retaliation

44. The quality of interexchange access is closely monitored by both competitors and regulators. In order to

discriminate successfully in the future, Pacific would not only have to explicitly violate the new Act; it would have

to do so in such a way that the differences in quality would be sufficiently detectable by customers to induce them to

shift their patronage to it while going undetected by sophisticated competitors and regulators—an eventuality so

unlikely as to border on the impossible.

45. Nor are those competitors dependent solely, for the redress of such discriminations as they do detect,

on the protections of government authorities. On the contrary, in keeping with the central purpose of the competitive

policies at the State and Federal levels, they already have the protection of being able to obtain from competitors

direct access to a substantial percentage of their total business (although still only a minority of subscribers) of any

offending BOCs.

46. The established regime for competition

47. PBCOM’s potential competitors are already operating in the interLATA market and dominate it.

AT&T, MCI and Sprint are large and powerful competitors, certainly more formidable rivals than MCI and Sprint

were to AT&T in 1984. Their nationwide optical fiber networks are in place; their costs are sunk; and their networks

- 21 - Draft

can be quickly turned to provide nearly any telecommunications service that appears to be profitable. In this world,

competitive strategies involving predatory pricing (e.g., cross-subsidization or a vertical price squeeze) by a new

entrant or an incumbent are doomed to fail. For a strategy that sacrifices profits (effectively investing in the

destruction of a rival) to succeed, the would-be predator must be able actually to drive the targeted competitor from the

market and, by so doing, be in a position to recoup its current losses at some future period. It cannot possibly

succeed in this endeavor if the rival is a telecommunications network-based carrier, because there is no way of driving

such a network out of the market; the costs associated with it are preponderantly sunk. Whenever the local carrier

tired of earning less money in the interstate toll market than it could earn in the carrier access market, it could raise its

toll price, but it would find competitors still in place and ready to compete.44

48. Arguments about the danger or likelihood of vertical squeezes and transfer of inputs among

affiliates at discriminatorily favorable prices completely ignore the statutory requirements of imputation and the

44 These relationships are complicated by the fact that within their own regions the LECs receive revenues both as

direct providers of long-distance services and in providing access to other IXCs. This has the consequence thatany losses that they might suffer by reducing their retail rates, with anti-competitive intent, would be cushionedby additional sales of access services stemming from the increase in end-market demand stimulated by those pricereductions and would therefore not have to be recovered entirely by retail price increases after competitors hadbeen driven out. There are at least two counter-considerations. First, to the extent this dual relationship of theLECs to this market gives them an additional incentive to reduce prices to ultimate customers, that is in itselfnot necessarily a bad thing, to put it mildly, particularly in a market that is inadequately competitive. In fact,Hinton, Zona, Schmalensee and Taylor have recently demonstrated that while under conditions of imperfectcompetition, such as do clearly prevail in the long-distance business, a less efficient ILEC might as a result takesome business away from a more efficient IXC, the social costs of the consequent productive inefficiency—evenin that remote eventuality—would be far outweighed by the social gains accruing to consumers in the form ofreduced prices. (Paul J. Hinton, J. Douglas Zona, Richard L. Schmalensee and William E. Taylor, “An Analysisof the Welfare Effects of Long Distance Market Entry by an Integrated Access and Long Distance Provider,”Journal of Regulatory Economics, Vol. 13, 1988, pp. 183-196.) Second, these additional access revenues wouldcompensate or more than compensate for the retail price reductions only under special and improbablecircumstances: the LEC would have to have a share of the interLATA market sufficiently large to depress theoverall market price, if it were to try to put a squeeze on competitors, yet not so large that its consequent loss ofrevenues from its own retail sales would not be offset by the increase in its revenues from the sale of access tocompetitors. Moreover, any price reductions encouraged by the complicating consideration we have justdescribed would pose a threat to competition only if the resulting relationship between the access and the retailcharges of the BOCs violated the imputation requirements of the Act, as well as all the other safeguards in placeat the state levels designed to prevent predatory pricing.

In his Supplemental Affidavit, Marius Schwartz likewise minimizes this unique incentive of the RBOCsto cut their interLATA prices, on the ground that they would in most circumstances lose more revenue directlyfrom those reductions than they would gain from increased sale of access, as our own exposition suggests. Wepoint out, however, that in this discussion he ignores the pressures on the RBOCs to offer discounts on theprices of the incumbent IXC’s—as other ILECs have indeed done on entering this market—as a means ofbreaking into it, entirely apart from the incidental benefit to them of increased access revenues, to the extentthese discounts expand the total market.

- 22 - Draft

sufficiency and efficacy of efficient component pricing rules in ensuring the ability of equally efficient competitors

to survive and to prosper in competition with the ILECs. The FCC has just reached the same conclusion:

275. Price Squeeze Concerns Are Adequately Addressed. Several parties have argued thatcurrent access charge rate levels create the conditions for an anticompetitive price squeeze when aLEC affiliate offers interexchange services in competition with IXCs....

278. We conclude that, although an incumbent LEC’s control of exchange and exchange accessfacilities may give it the incentive and ability to engage in a price squeeze, we have in placeadequate safeguards against such conduct....

279. The Fifth Competitive Carrier Report and Order separation requirements have been in placefor over ten years, and independent (non-BOC) incumbent LECs have been providing in-region,interexchange services on a separated basis with no substantiated complaints of a price squeeze.Under these separation requirements, incumbent LECs are required to maintain separate books ofaccount, permitting us to trace and document improper allocation of costs and/or assets between aLEC and its long-distance affiliate, as well as to detect discriminatory conduct. In addition, weprohibit joint ownership of facilities, which further reduces the risk of improper allocations of thecosts of common facilities between the incumbent LEC and its interexchange affiliate....[T]heprohibition on jointly-owned facilities also helps to deter any discrimination in access to theLEC’s transmission and switching facilities by requiring the affiliates to follow the sameprocedures as competing interexchange carriers to obtain access to those facilities. Finally, ourrequirement that incumbent LECs offer services at tariffed rates, or on the same basis asrequesting carriers that have negotiated interconnection agreements pursuant to section 251 reducesthe risk of a price squeeze to the extent that an affiliate’s long-distance prices would have toexceed their costs for tariffed services....

281. Furthermore, even if a LEC were able to allocate improperly the costs of its affiliate’sinterexchange services, we conclude that it is unlikely that the LEC’s interexchange affiliatecould engage successfully in predation.45

48. Moreover, whereas the rules for entry by competitors into the local exchange market are still in the

process of being hammered out, the arrangements for fair access by the long distance carriers to the facilities of the

BOCs—dealing with problems of provisioning, repair, billing, segregation of proprietary information and the like by

even-handed negotiations between the BOCs and interexchange carriers—have been in place for upwards of a decade.

Those same unbiased interconnection arrangements would continue if the BOCs were permitted to carry interLATA

traffic and, as we will point out presently, any deterioration would be all the more visible because of the long-standing

nature of the arrangements. Hence there can be little concern that the terms and conditions of interLATA

interconnection could distort the competitive process henceforward.

45 First Report and Order, In the Matter of Access Charge Reform , CC Docket No. 96-262 et al., Adopted. May 7,

1997, pars. 275, 278, 279, 281 (footnotes omitted).

- 23 - Draft

49 . The anti-discrimination requirements of theTelecommunications Act

50. Irrespective of whether they enter interLATA toll markets, the Telecommunications Act requires

RBOCs to provide interconnection and access to unbundled elements on a non-discriminatory basis.46 The language

could not be clearer. Incumbent local exchange carriers have the duty to provide interconnection

at any technically feasible point within the carrier’s network; [and] that is at least equal in qualityprovided by the local exchange carrier to itself or any subsidiary, affiliate, or any other party towhich the carrier provides interconnection. (Section 251 (c) (2) (B) and (C).)

Further, ILECs have

[t]he duty to provide, to any requesting telecommunications carrier for the provision of atelecommunications service, nondiscriminatory access to network elements on an unbundledbasis at any technically feasible point on rates, terms, and conditions that are just, reasonable,and nondiscriminatory in accordance with the terms and conditions of the agreement and therequirements of this section and Section 252. An incumbent local exchange carrier shall providesuch unbundled network elements in a manner that allows requesting carriers to combine suchelements in order to provide such telecommunications service. (Section 251 (c) (3)).

50. Further amplifying this nondiscrimination standard is the “checklist” contained in Section 271 spelling

out the prerequisites for interLATA entry.

51. Notwithstanding this clear Congressional directive for non-discriminatory interconnection, we must

confront the counter argument that the BOCs will have every incentive to flout it, once they are permitted to offer

interLATA service themselves, and all sorts of possible devious ways of doing so.

52. The access arrangements that have been in place for over twelve years, as we have already pointed out,

were developed during a period in which the RBOCs had strong incentives to provide high quality service, undiluted

by competitive (or anti-competitive) considerations. Moreover, those incentives have been intensified by the

emergence of competitors for access traffic: every major metropolitan area in the country—embracing a large share of

the BOCs’ total business—is now served by at least one CAP. And, it is important to bear in mind, it is

discrimination in the provision of access, not of other local exchange services, that opponents claim the RBOCs will

practice, with adverse competitive consequences, if they are permitted to compete with the incumbent IXCs.

46 The unbundled elements include and can be combined to include, all components of currently-used carrier access

services. Coupled with the resale requirements of the Act, these provisions encourage entry without requiringlarge amounts of investment, the implications of which we discuss more fully below.

- 24 - Draft

53. Explicit oversight by both regulators and purchasers of access provides strong additional guarantees of

good service. Quality standards are often built into tariffs or other administrative rules, and they are regularly

monitored and audited by IXCs and regulatory agencies alike. The FCC routinely monitors them through quarterly

and semi-annual reports of measures such as installation and repair intervals, post-dial delay, transmission quality and

service quality complaints. Finally, RBOCs have worked out monitoring and auditing programs directly with their

IXC customers, programs that may include financial penalties for failure to meet quality standards.

54. Our conclusions about the sufficiency of these safeguards have had the explicit endorsement of the

FCC. The Commission states in its recent order on non-accounting safeguards,

We believe...that sufficient mechanisms already exist within the 1996 Act both to deteranticompetitive behavior and to facilitate the detection of potential violations of section 272requirements.

We also find that, beyond the reporting requirements mandated under the 1996 Act, there are otheravenues by which a telecommunications carrier may obtain information relevant to detectinganticompetitive BOC conduct. For example, competitive telecommunications carriers, on theirown initiative, could seek to incorporate certain performance and quality standards into theirnegotiated or arbitrated interconnection agreements to ensure that BOCs satisfy their obligationto provide service in a nondiscriminatory manner.

And, it concludes,

We believe that the reporting requirements required by the 1996 Act, those required under statelaw, and those that may be incorporated into interconnection agreements negotiated in good faithbetween BOCs and competing carriers will collectively minimize the potential foranticompetitive conduct by the BOC in its interexchange operations.47

1 . Incentive regulation

2. As of the end of 1996, at least 30 states and the FCC have substituted price caps for traditional cost-

plus, rate base rate of return regulation. California was one of the first of these states. Price caps represent an

improvement over the traditional methods of regulation in two ways. First, they supply stronger incentives on the

part of the regulated firms to improve their efficiency, since they retain the benefits of any such cost reductions—

subject of course to reexamination of the price cap formulas. Second, and more directly pertinent in the present

context, they can eliminate the incentive of the regulated firms to engage in predation or otherwise cross-subsidize

47 First Report and Order and Further Notice of Proposed Rulemaking, In The Matter of Implementation of the

Non-Accounting Safeguards of Sections 271 and 272 of The Communications Act of 1934, as amended , FCC,CC Docket No. 96-149, Adopted: December 23, 1996, Released: December 24, 1996, pars. 321, 326 and 327.

- 25 - Draft

competitive services because, by breaking the link between the firms’ overall profits and regulated rates, they

eliminate—to the extent the price cap regimes are pure48—the opportunity to recover all of those costs or losses from

monopoly customers. Unsurprisingly, state regulators and Federal courts have ruled that price cap regulation can be

an effective safeguard against cross-subsidization and other such anticompetitive behavior.49

3. Pacific’s price performance regulation plan in California achieves a similar result. The California

Commission was one of the first to institute price regulation, when it ordered a price cap plan beginning in 1990.

48 That is to say, to the extent that they do not provide for sharing between companies and ratepayers of excesses or

inadequacies of profits and are not promptly “corrected” to eliminate excessive profits or losses, eitherretroactively or prospectively. The majority of the plans are indeed “pure” in the former sense: of the (at least)29 states we counted as having adopted some form of price cap regulation as of June 1996, only two hadprovisions for sharing with ratepayers either surpluses or deficiencies in achieved rates of return. . The same istrue of the recently revised FCC formula. Fourth Report and Order in CC Docket No. 94-1 and Second Reportand Order in CC Docket No. 96-262, In The Matter of Price Cap Performance Review for Local Exchange Carriers and Access Charge Reform , Adopted: May 7, 1997, Released: May 21, 1997.

As for “purity” in the sense of a complete abandonment of tests of the price cap formulas or freezesagainst achieved rates of return, no plan to our knowledge rigidly excluded the possibility of such a test—in thissense, no plan was “pure.” On the other hand, our survey, as of June 1996, of price cap plans adopted in theprevious three years disclosed that the commissions were typically planning on an approximately five yearinterval before subjecting the formulas to review. The periods (in years) were: Illinois—3; Iowa—4; Kansas—5; Kentucky—at least 4; Maine—5; Massachusetts—at least 6; Michigan—2; New Jersey—6; North Carolina—5; Ohio—6; Pennsylvania—5; South Carolina—1; and Wisconsin—6.

49 [A] well designed price cap plan insulates ratepayers from investment risk and subsidization ofnew ventures. Massachusetts Department of Public Utilities, NYNEX Price Cap, D.P.U. 94-50(May 12, 1995), p. 121.

A properly designed alternative regulation plan affords the opportunity not only for the Companyto transition itself to a more competitive environment, but allows this Commission toimplement safeguards and allocate risk in a fashion that protects the interests of all interestedparties. Illinois Commerce Commission, 92-0448/93-0239 Consol. (October 11, 1994), p. 19.

We find attractive many aspects of a pure price cap model for establishing revenue levels .…Theutility and its shareholders would be completely at risk for their operational decisions, andincentives to cross-subsidize more competitive activities with monopoly profits from basicservices would be greatly reduced. California Public Service Commission, Decision 89-10-031, In the Matter of Alternative Regulatory Frameworks for Local Exchange Carriers (October 12,1989), at 172-173

[T]he FCC has taken specific affirmative steps designed to deter and detect cross-subsidization byintroducing price caps as well as further strengthening its cost accounting rules. We conclude thatwith the implementation of these measures, the FCC … has demonstrated that the BOCs’incentive and ability to cross-subsidize will be significantly reduced. California v. FCC, No. 92-70083 and Consolidated Cases, 39 F.3d 919 (9th Cir. 1994) (“California III”) at 926-927.

[Price cap regulation] reduces any BOC’s ability to shift costs from unregulated to regulatedactivities, because the increase in costs for the regulated activity does not automatically cause anincrease in the legal rate ceiling. United States v. Western Elec. Co., 301 U.S. App. D.C. 268,993 F.2d 1572 (D.C. Cir.), cert. Denied, 114 S. Ct. 487 (1993) at 1580.

- 26 - Draft

Although the plan contains a provision for sharing excess earnings and therefore departs somewhat from pure price

regulation, the productivity targets in the plan have been sufficiently stringent that Pacific has not been close to

sharing during its seven-year history.50 Therefore, the plan has functioned in practice in the same way as a pure

price cap regime.

4 . Accounting safeguards

5. If there is one task at which regulators have proved themselves adept both before and since divestiture, it

is in allocating costs in such a way as to protect purchasers of regulated—and, in particular, basic exchange—services

and, by so doing, protecting competitors from cross-subsidization. Whether in so allocating costs as to set floors

under the prices of competitive services markedly above incremental costs or in setting ceilings on basic service rates

below incremental costs—and even farther below economically efficient levels—regulators have erred in the direction

of over-protecting the BOCs’ competitors from efficient competition and underpricing regulated services.51

6. In addition, the Commission has now officially found that its various accounting safeguards, including

its existing rules governing transactions between the LECs and affiliates, are fully sufficient to guard against

subsidization of competitive activities at the expense of subscribers to regulated telecommunications services.52

7. Inexplicably, the FCC’s recent Ameritech Order contradicts its conclusions, only a few months

50 The California plan employs a price index of inflation minus a productivity target. That target was 4.5 percent

from 1990 to mid-1994 and 5 percent up to the end of 1995. Beginning in 1996, the formula was suspended,which in effect sets the productivity target equal to the inflation rate.

51 Crandall and Waverman, op.cit.; A.E. Kahn, “The Uneasy Marriage of Regulation and Competition,”Telematics, September 1984, pp. 1-2, 8-17 and “The Road to More Intelligent Telephone Pricing,” Yale Journalon Regulation, Vol. 1, No. 2, 1984, pp. 139-157; and D.L. Kaserman and J.W. Mayo, “Cross-Subsidies inTelecommunications: Roadblocks on the Road to More Intelligent Telephone Pricing,” Yale Journal onRegulation, Vol. 11, No. 1, Winter 1994, pp. 119-147.

52 Report and Order, In the Matter of Implementation of the Telecommunications Act of 1996: Accounting Safeguards Under the Telecommunications Act of 1996 , FCC 96-490, CC Docket No. 96-150, adopted:December 23, 1996, Released: December 24, 1996:

our cost allocation and affiliate transaction rules, in combination with audits, tariff review, andthe complaint process, have proven successful at protecting regulated ratepayers from bearing therisks and costs of incumbent local exchange carriers’ competitive ventures, (par. 25)

and:

We have previously concluded that these [affiliate transaction] rules provide effective safeguardsagainst cross-subsidization. (par. 108)

See also pars. 1 and 275.

- 27 - Draft

earlier, about the efficacy of existing accounting and non-accounting safeguards in the long distance markets: “in

order for this potential [benefit of RBOC entry] to become a reality, local telecommunications markets must first

be open to competition so that a BOC cannot use its control over bottleneck local exchange facilities to undermine

competition in the long distance market.” (par. 388) We respectfully assert, and the evidence proffered in our

statement demonstrates, that (1) RBOC entry will be beneficial independent of the development of local

competition and (2) these benefits will not be diluted by any offsetting threat to the vitality of competition in the

interLATA market, because, as the FCC recognized earlier, existing safeguards are adequate and it is simply

inconceivable that the RBOCs could achieve monopoly power there at the expense of AT&T and the other

incumbent carriers.53

B. The Unbundling, Resale and Interconnection Provisions of the 1996 Actand the Section 271 Checklist

60. Manifestly, the provisions of the Act and the FCC’s Interconnection Order have had the intention, and

will have the effect, of strengthening the competitive safeguards previously instituted:

• The Act and the FCC’s Interconnection Order substantially expand and accentuate the degree of mandated

non-discriminatory access to essential inputs. Not only are more unbundled elements to be provided at a

greater number of points of interconnection, the prices charged for these elements must be approved

under an open process and comply with explicit rules designed to afford rivals a fair opportunity to

compete. If anything, these new terms and conditions are too restrictive, to the detriment of efficient

competitive initiatives and responses on the part of the incumbents.54

• The Telecommunications Act imposes restrictions and handicaps, for a limited number of years, on

the offer by incumbent LECs of services they were previously barred from offering at all—such as in-

region interLATA toll and manufacturing. These restrictions, in the form of required structural

53 Professor Schwartz, the DOJ’s economist, agrees with our conclusion (his original affidavit, par. 14). His

primary concern, fully spelled out in his Supplemental Affidavit, is with the sufficiency of the RBOCs’incentives to cooperate in opening local markets to competition once they have received interLATA authority.

54 A detailed critique of the Interconnection Order is unnecessary to this declaration. Our major concerns about itare that the prices and terms of access may be unduly favorable to entrants that choose to compete through resaleof incumbents’ services and use of unbundled elements. The adverse consequences of such an imbalance include:(1) diluting the incentives for facilities-based entry into local exchange services and (2) eroding the incumbents’incentives to upgrade their networks and offer innovative services.

- 28 - Draft

separations and limitations on the marketing of local exchange and other services, would limit their

exploitation of economies of scope and thereby handicap them in competing with rivals, to which

these restrictions would not apply.

61. The incumbents’ ability to offer in-region interLATA services is dependent on their satisfying a

“checklist” of requirements, which include nondiscriminatory access to essential inputs and the demonstrated

presence of competition for local exchange services.55 These requirements, in and of themselves, hasten the erosion

of the “local bottleneck”the basis for the historical concern about anticompetitive behavior—and make lifting of

the ban on interLATA services contingent on the FCC’s being satisfied that the possibility of such behavior has

been sufficiently minimized by the presence of local competition or its possibility.56 In our opinion, which we

will document below, the imposition of additional requirements, such as have more recently been recommended by

the Department of Justice or are imposed or implied by the Commission’s recent Ameritech Order, are neither

necessary nor desirable.

62. Those who oppose BOC interLATA entry systematically ignore or minimize the extent to which the

Act’s interconnection arrangements, once in place, make those markets highly contestable— a market condition that

Professor William J. Baumol, one of the major and original protagonists of contestability theory, recognizes “offers

public interest benefits virtually the same as those insured by powerful competitive forces.”57

63. The interconnection arrangements contemplated by the Act reduce the sunk investment costs of

entry—which Professor Baumol identifies as the primary barrier to contestability58—dramatically. This is the

consequence, in particular, of the resale obligations imposed on the ILECs. If, as the FCC and state commissions

all over the country have decided or are in process of deciding, any would-be competitor has the right to purchase

any and all of an ILEC’s present retail services at its retail prices less a discount large enough to enable equally

55 Alternatively, if no qualifying entrant pursues interconnection within the state, the BOC’s demonstration that the

necessary conditions for entry have been established will suffice (so-called “Track B” for interLATA entry).

56 In fact, the “local bottleneck” is already eroding, independently of the Telecommunications Act, as indicated bythe fact that revenues for competitive local exchange carriers grew by 80 percent in 1996. (“CLEC RevenuesGrow 80% in 1996, Report Finds,” Telecommunications Reports, February 5, 1997, p. 18)

57 Affidavit of William J. Baumol on behalf of AT&T Corporation, AT&T Exhibit B, In the Matter of Application by SBC Communications Inc. for Authorization Under Section 271 of the Telecommunications Act of 1996 to Provide In-Region, InterLATA Services in the State of Oklahoma , CC Docket No. 97-121, page 14.

- 29 - Draft

efficient retailers to compete with it, then all of the ILECs’ present retail markets are as close to perfectly

contestable as conceivable: rivals could at any time compete with them without having to sink a dollar into

equipment that might not be fully retrievable if they decided to withdraw.

64. To be sure, that characterization may exaggerate the perfection of the consequent contestability of local

telephone markets. Presumably the challenging reseller would have to put in place some sort of interfaces to

purchase services from the ILEC; it would have to make marketing contacts with customers and arrangements for

billing them, some of which costs would be irretrievable upon its withdrawal from the market. The notion of a

competitive entrant having to be spared even the costs of contacting potential customers and billing them, however,

would reduce the concept of contestability to an absurdity. Moreover, billing could always be purchased as needed and

therefore involve no sunk cost. Competitors could contract out for marketing as well, under terms that, similarly,

would make those costs avoidable. As the entry and continued existence of some five hundred resellers of long

distance services attest, these barriers to entry and exit must be very low indeed. And the incremental costs to a carrier

such as AT&T or MCI, already covering virtually the entire interLATA market, of adding such consumer contacts for

purposes of selling intraLATA and local services as well—adding some lines to their advertisements and bills—must

come as close to zero as can be conceived in the real world.

65. The requirements of the Act with respect to unbundled network elements are less likely to have such a

dramatic effect at the wholesale level, at least initially. Certainly, if challengers are to have the ability to purchase

from ILECs all the inputs necessary to duplicate the latter’s offerings, without having to make any investments of

their own, that is likely to make the supplying of those services at wholesale likewise highly contestable. On the

other hand, presumably the challengers would have to make some financial commitment, if they chose to lease

network elements from the incumbents. The ILECs might therefore continue, at least for a time, to enjoy

monopoly power at the wholesale level.

66. The implications of this new situation are nevertheless dramatic. What it means, specifically, is that

the typical requirements in governing statutes or regulations for reclassifying the entire range of retail local

telephone services as competitive will, as a matter of economics, be satisfied by these rules. In these

circumstances, deregulation of the retail operations of the ILECs becomes not just possible but mandatory:

58 Id., p. 16.

- 30 - Draft

effective competition demands that they have the identical freedom to compete at the retail level as is now enjoyed

by their competitors, subject—with the major exception of the imputation or efficient component pricing rule

(ECPR)59—to no obligations, handicaps or regulatory responsibilities that are not also borne by them.

67. As we have already suggested, the characterization of these markets as close to perfectly contestable

applies most unqualifiedly to the performance of the retailing function, because of the sales for resale obligations

of the incumbents. If a competitor seeking instead to produce the services itself by assembly of inputs from the

ILECs would incur a greater volume of sunk costs, simple deregulation of the retail markets would expose

consumers to possible exploitation, as the incumbent companies raised their retail prices to which the prescribed

wholesale discounts would apply. The residual protection of consumers that these considerations suggest could be

ensured, however, by a continuing obligation of the wholesaling ILEC to make its pre-existing retail services

available at the regulated prices it charged at the time of transition less the prescribed discount—an obligation

explicitly accepted by the Southern New England Telephone Company in reorganization proposals essentially

approved recently by the Connecticut Department of Public Utility Control.60

68. The Justice Department, through an affidavit of May 14, 1997, filed on its behalf by Professor

Marius Schwartz in response to Southwestern Bell’s Oklahoma interLATA entry petition, recommended that

RBOC entry into long-distance be delayed until local markets are irreversibly open to competition (par. 19)—that

is, until new access arrangements have been clearly demonstrated to be working (par. 182). Professor Schwartz

also discusses how one would determine whether the local exchange had been irreversibly opened to competition.

His preferred metric is the presence of competition (par. 20).

69. The Justice Department’s proposals would impose a substantial burden on the ILECs beyond the

requirements of the Telecommunications Act. The Act requires only (1) the existence of one or more qualifying

interconnection agreements with predominantly facilities-based carriers (“Track A”) or (2) the establishment of

approved interconnection terms and conditions in the absence of an interconnection request within a specified time

period (“Track B”). The Justice Department would add a demonstration that competition has either started or been

59 The ECPR was first formulated by Professors Baumol and Willig. For a clear statement of it, see William J.

Baumol and J. Gregory Sidak, Toward Competition in Local Telephony, Cambridge: The MIT Press, 1994.

60 DPUC Investigation of the Southern New England Telephone Company Affiliate Matters etc., Docket No. 94-10-05, Decision, June 25, 1997.

- 31 - Draft

proven infeasible. But not only does it offer no proof that any additional requirements are necessary; it fails to

recognize that they may be positively harmful.61 Just as its proposal is intended to reinforce the incentives of the

RBOCs to cooperate in facilitating the entry of competitors, it gives the opponents of RBOC entry into

interLATA markets new opportunities to use the regulatory process to delay it.62 We see no reason to alter the

balance between these two incentives and opportunities established by the statute. Ascertaining whether entry has

been “irreversibly established” and sufficient to satisfy the proposed additional test would be extremely contentious.

For example, the filings of economists representing the IXCs in the Oklahoma case (and, to our knowledge, in

Connecticut) make it clear they will argue that no amount of entry via resale or even by purchase of unbundled

elements would be sufficient; yet, paradoxically, the most telling criticism of the FCC’s interconnection order is

that by prescribing excessively favorable terms—terms advocated by these same IXCs—for entry via those two

routes, it discourages the genuinely independent, facilities-based, entry that they insist (in other contexts) is

necessary. (In these circumstances, Professor Schwartz’s assertion that “the most reliable demonstration of such

opening [of local markets to competition] is…meaningful local entry of all three modes” (Supplemental Affidavit,

par. 5, stress supplied) has an unmistakable Catch 22 flavor.) The delays resulting from disputes of this kind

would prevent consumers from enjoying the benefit of RBOC entry that Professor Schwartz acknowledges. In

addition, the enhanced prospect of impeding RBOC entry by regulatory pleadings would provide intensified

incentives for the IXCs to pursue less vigorous local entry strategies themselves than they otherwise would,

because more vigorous local competition on their part would hasten the day of the RBOC entry into competition

with them that they obviously are exerting strenuous efforts to obstruct.

70. Conversely, and for the same reasons, actual entry by RBOCs into long distance service would be

likely to strengthen the IXCs’ incentives to compete vigorously in the local market, since it would withdraw the

reward for delay. As artificial and asymmetrical restrictions are removed for everyone, performance in the

marketplace, rather than success or failure in gaming the regulatory process, will and should determine which

61 Similarly, the FCC’s Ameritech Order provides no demonstration that the unspecified number of additional

“factors” (par. 391) would do more good than harm.

62 For a forceful argument, which accords with our own experience, that the major IXCs have been far moreresistant than the CLECs to concluding interconnection agreements without recourse to arbitration and lessaggressive in deploying their own facilities locally, see Peter W. Huber, “Local Exchange Competition Underthe Telecom Act,” Telcom Policy and Analysis Group, Washington, D.C., Nov. 4, 1997, pp. 36-40.

- 32 - Draft

companies succeed or fail.

71. Professor Schwartz mentions, but evidently fails to weigh in the balance, another aspect of the

symmetry envisioned by the Telecommunications Act that argues on the side of removing the interLATA

restriction on the RBOCs as quickly as possible. Under the Act, the extension of presubscription for intraLATA

service is made contingent on and simultaneous with permitting the BOCs to compete interLATA. Against the

incremental impetus that delay in lifting the latter restriction might give to the BOCs to encourage the

development of local competition must logically be weighed the cost to consumers of the consequent delay in

universalization of intraLATA presubscription. Since basic local rates to most subscribers are already repressed

below efficient levels, whereas intraLATA toll rates remain egregiously inflated in order to generate a contribution

toward holding down the former, it seems clear that the direct benefits of the introduction of presubscription to

competitive IXCs for the latter services offers a far greater potential for consumer benefit than such incremental

encouragement to competition in the provision of local services as retaining the ban would impart. This

consideration is even more compelling in view of the fact that the resale provisions of the Act already offer CLECs

a ready opportunity, by purchasing the subsidized services from the incumbent LECs, to bring the benefits of

competition to the provision of such overpriced local services as Custom Calling features.

72. Professor Schwartz claims that his proposed standard would promote the public interest, because he

believes that the benefits from the faster development of local competition that would be achieved by withholding

interLATA authority from the RBOCs until they had satisfied the more stringent conditions he recommends would

outweigh the costs of delaying RBOC interLATA entry. He bases this assessment, in his Supplemental Affidavit,

on essentially qualitative considerations that are, in our opinion, biased against RBOC entry:

• The first, to which we have already responded (par. 42, above), is that the local market is both larger

and has more room for improved performance, because it is “largely a regulated monopoly rife with

distortions.” (par. 18) As we have already pointed out, the distortions—whose dimensions and

egregiousness we concede—are the consequence of regulatory policies; the remedy is not necessarily

competition, some undeterminable proportion of which, responding to those distorted price signals, is

doubtless inefficient, but reform of those regulatory policies.

• In proffered rebuttal of our earlier comments on his first Affidavit that his recommendations would

- 33 - Draft

give the major IXCs a strong incentive to be uncooperative in negotiating with the BOCs, because

that would delay the latter companies’ entry into competition with them, he observes that, “to his

knowledge,” local competitive entry has been no more pervasive in SNET and GTE territories, where

those ILECs already have interLATA authority and the IXCs would, under our reasoning, have no

incentive to engage in dilatory tactics in negotiating with them. To this rejoinder, there are several

responses: (1) Following Professor Schwartz’s own reasoning—to the effect that once long-distance

entry is permitted an ILEC, it would have reduced incentives to cooperate in opening its local

markets—we would expect to see significantly less progress in the introduction of local competition

in GTE and SNET than in RBOC territories. Yet he makes no such claim. (2) If Professor

Schwartz’s “understanding” is correct with respect to GTE, it would probably be because of that

company’s typically rural and suburban service territories, where competitive entry is far less

attractive, precisely because of the distorting regulatory policies to which we have already alluded. (3)

So far as the experience in Connecticut is concerned, it seems to be true that, despite that state being

at the very forefront in seeking to open the local telephone business to competition, and some 36

applications to offer such service there having been approved,63 the actual inroads of competitors have

apparently been modest.64 The relevant question, apart from remaining uncertainties about the facts,

is one of responsibility. While AT&T and MCI publicly “blame SNET…for sabotaging the quality

of its service,” 65 an AT&T witness has in regulatory proceedings in that state explicitly praised

SNET for its cooperativeness and good faith efforts to establish conditions for CLEC entry,66 and a

63 Joan Muller, “Back to the future: SNET still dominates,” The Boston Globe , January 21, 1998, p. D1, D16.

64 According to the Company, it has lost a considerably larger share of the market than would be suggested by thestatement in Muller, loc. cit ., that SNET still “controls 99 percent of the phone lines in the state.” In view ofthe availability to competitors (in theory at least) of the option of leasing SNET’s lines, it is not clear whetherthat percentage accurately reflects the incumbent company’s share of the market defined in terms of the numberof customers. Since the implied 1 percent of lines in the hands of CLECs are almost certainly concentrated inmetropolitan business districts, their share of total revenues would presumably be several times higher.

65 Muller, op. cit.

66 Q. As I understand AT&T has recently provided a forecast to SNET. Without divulgingnumbers has SNET stated in any way that they will not be able to handle the amount oforders that AT&T has forecasted?

A. No, they have not.

- 34 - Draft

witness for TCG testified there that in his Company’s experience SNET had been the most

cooperative of all the ILECs in the country in those negotiations67—both of these encomia for a

company that already had interLATA authority and therefore in flat rebuttal of the rationale for the

strategy advocated by Professor Schwartz and his DOJ client. The Company, proclaiming its good

faith efforts, protests first, that it was geared to handle 10,000 orders a week for resales but never

received more than 3,000—an assertion confirmed by a DPUC Commissioner; and that the task of

preparing operations support systems capable of permitting its competitors to order electronically

transfers of customers and their provisioning, repair, billing services has proved to be far more

difficult than anyone had anticipated.

• Professor Schwartz claims that advocates of RBOC entry have overstated the extent to which prices

would drop if that were to occur. He appears to be satisfied that recent IXC offerings, which even

according to him are at least 10 cents per minute, do not leave much room for large price reductions.

Yet in the case of intraLATA competition, he cites claims of prices as low as 3 cents after

presubscription had been ordered. Because long-distance costs are not very distance-sensitive, the

potential for price reduction should be similar for interLATA services. Therefore, if his intraLATA

figure is accurate, current IXC interLATA prices leave considerable room for erosion in consequence

of RBOC entry; and, as we have already pointed out, his contention to the contrary notwithstanding,

the RBOCs would have strong incentives to offer substantial discounts to break into that market, just

as the IXCs have done when first offering intraLATA service—witness AT&T’s 5 cents a minute

offering in Connecticut. (See our reference to the SNET experience, in par. 77, below.)

73. In contradiction of the basic rationale of the Telecommunications Act, which is to leave the outcome to

the competitive market (subject to the antitrust-like protections incorporated in the Act), the rationale of the Justice

Q. In fact, isn’t it true, Mr. Curran, that AT&T has sent pretty nice complimentary letters to SNET on

behalf of the work they have done?

A. There has been some exemplary work between the companies in terms of cooperation and we thankyou for that.

Testimony of Kevin Curran, Connecticut Department of Public Utility Control Docket No. 94-10-05, April 3,1997, Transcript pp. 501-502.

67 The witness was Paul Karoupas, now TCG’s Vice President for regulatory and external affairs. Information from

- 35 - Draft

Department’s proposal is essentially regulatory. Rather than requiring regulators to satisfy themselves only that “the

requisite arrangements necessary to open the local market are made available” (Schwartz Affidavit, par. 70, stress

supplied), as the statute stipulates, it would require them additionally to assess the degree to which that availability

has proved effective—that is, whether “meaningful local competition” has “emerged,” and, if not, “why” (pars. 20,

80)—both complicated questions that Congress wisely intended to head off by establishing specific criteria for open

markets in the checklist. This is clearly a rationalization for continuing micromanagement of a process that,

Professor Schwartz and we agree, Congress intended to deregulate.

74. In his Supplemental Affidavit, Professor Schwartz disputes our contention that his proposals

constitute micromanagement, claiming instead that they will lead to less, rather than more regulation. We stand

by our assessment. He has clearly proposed an additional, results-oriented standard that would provide the occasion

for additional jousting before regulatory commissions such as we have already described, including assertions such

as the IXCs have already been making that the only “meaningful” competition would be facilities-based—thereby

giving them an additional reason to continue their own policies of emphasizing entry, instead, by using the

facilities of the incumbent LECs (another Catch 22). For example, despite the enormous amount of time, effort,

and money that ILECs have already expended in accommodating unbundling and resale and negotiating with

CLECs68—resulting in 1700 agreements as of late November 1997, according to the count of the U.S. Telephone

Association (currently 31 of them in California, SBC tells us)—Professor Schwartz believes that RBOCs “would

fight every requirement” (par. 42) imposed to open their markets. Given his belief that (1) some entrants will

depend on unbundled elements well into the future and (2) ILECs will be prone to discriminate in new access

arrangements (par. 70),69 the inescapable conclusion would seem to be pervasive regulation as far as the eye can

see. And with it would go the endless disputes over whether the RBOCs have gone far enough to satisfy the

desires of competitors seeking the most favorable possible deals and regulators under strong pressure to produce

results in the form of visible competitors.

the Connecticut Department of Public Utility Control.

68 SBC informs us that it has already spent $1.2 billion to accommodate local competition.

69 Thus, like other commentators, Professor Schwartz has ignored the extended history of successful competitionwith vertically integrated RBOCs that we describe below.

- 36 - Draft

C. Successful Competition between Vertically Integrated RBOCs and FirmsRequiring Access to Their Facilities in Other Markets

75. There has accumulated, over the last decade or more, a great deal of actual experience with competition

between the RBOCs—and LECs that are not BOCs—on the one side, and rivals dependent on access to their facilities.

An ounce of actual experience is surely weightier than a pound of speculation about possible misdeeds or, indeed, of

anecdotal claims about exclusionary practices. Assertions about the theoretical inadequacies of regulatory safeguards

against predation, cross-subsidy and discriminatory treatment of competitors simply ignore this historical evidence. In

practice, competition by non-vertically integrated firms with RBOC “bottleneck monopolies” has already succeeded in

other telecommunications markets that are at least as susceptible to anti-competitive tactics as the interLATA

market—geographic corridors in which the BOCs have been permitted to offer interLATA service, cellular, paging,

voice messaging services (VMS), customer premises equipment (CPE), intraLATA long distance and the offer of

long-distance service by LECs other than BOCs.70

76. InterLATA corridor traffic

77. RBOCs have routinely provided interLATA services since divestiture under exceptions to the AT&T

consent decree, the notable example of which is Bell Atlantic’s interLATA service between New York and New

Jersey. In that narrow market, the RBOC offers rates about 20 to 30 percent below AT&T’s;71 yet it has only a small

70 International experience lends further support to the argument that regulatory safeguards are effective. While the

United States was clearly the leader in opening long-distance markets to competition, it has been alone inrequiring divestiture and quarantine. And yet, despite their having removed their barriers to entry into thosemarkets well after the United States had done so and despite their having permitted the providers of essential localexchange services to continue to offer the newly competitive services, toll competition has made substantialprogress in other countries. For example, the incumbents in Canada have lost more market share sincecompetition was authorized in 1992 than occurred in the United States over the comparable period after 1984.Similarly, three facilities-based carriers have captured over 30 percent of the Japanese long-distance market since1987, despite the fact that the incumbent NTT remains vertically integrated. Willie Grieve and Stanford L.Levin, “Telecom Competition in Canada and the U.S.: The Tortoise and the Hare,” Selected Papers from the 25th

Annual Telecommunications Policy Research Conference, Alexandria, VA, September 27-29, 1997. Likewise,Spiller and Cardilli report that facilities-based local competition has progressed at a healthy pace in the smallercountries they examined (Australia, Chile, Guatemala and New Zealand), even though none of these countrieshas the extensive unbundling requirements for an indefinite duration that prevail in the United States or hasprevented incumbents from vertically integrating. Pablo T. Spiller and Carlo G. Cardilli, “The Frontier ofTelecommunications Deregulation: Small Countries Leading the Pack,” The Journal of Economic Perspectives,Vol. 11 (1997), pp. 127-138.

71 "Bell Atlantic Seeks Nondominant Status in `Corridor'," Telecommunications Reports, July 17, 1995.

- 37 - Draft

share of this traffic, despite purported overwhelming advantages stemming from its control over local service.72 Over

ten years have passed without adverse consequences for competition.73

78. InterLATA service offerings by non-BOC LECs

79. Large non-BOC LECs, such as GTE, SNET, United Telephone and Rochester Telephone (now

Frontier), have similarly offered interLATA services without apparent anti-competitive effect. The SNET experience

in Connecticut is quite informative.74 SNET began offering out-of-state service in April 1994 at rates 15 and 25

percent below AT&T’s undiscounted rates for peak and off-peak calling respectively. By the end of 1996, it had

captured about 30 percent of the market, thereby providing large benefits to consumers in the form of lower prices and

new service offerings. For all the reasons we have already summarized, PBCOM’s entry into interLATA service may

confidently be expected similarly to benefit consumers, without denying its rivals a fair opportunity to meet that

competition.

80. Cellular

81. LECs have participated in cellular telephony since 1983. The service is organized as a (largely)

unregulated duopoly in the United States, with entry limited by the availability of only two 25 MHz channels in each

geographic market. At its inception, one channel was allocated to wireline carriers (generally a BOC or GTE) and the

other to non-wireline providers. The simple fact is that the wireline licensees (the LECs) have not come to dominate

the market, as would have happened if they had been able to subsidize these services from their local telephone

services or otherwise discriminate against their competitors. Despite a late start, non-wireline suppliers have nearly

equal market shares.75 Indeed, the largest cellular company in the U.S. is AT&T/McCaw, a non-wireline supplier,

72 The same is true of NYNEX (now part of Bell Atlantic). The fact that these two RBOCS serve corridor traffic

only through 10XXX access may explain their relatively small shares in their respective corridor areas.

73 That the FCC is of the opinion that anti-competitive behavior has not been a problem in these markets issuggested by the fact that when, in September 1990, it placed these interLATA services provided by LECs underprice caps, it elected not to subject them to price floors, as it had in other such situations.

74 Joint Affidavit of Robert Crandall and Leonard Waverman on Behalf of Ameritech Michigan, In the Matter of Application of Ameritech Michigan Pursuant to Section 271 of the Telecommunications Act of 1996 to Provide In-Region InterLATA Services in Michigan , FCC CC Docket No. 97-1, Vol. 3.1, January 2, 1997, pars. 53-54.

75 Paul Kagan Associates, “Cellular Ownership,” Wireless Market Stats, August 31, 1995, No. 72; and Donaldson,Lufkin & Jenrette, The Wireless Communications Industry , Table 2A: Cellular Industry - Quarterly Subscribers,Summer 1996, p. 10.

- 38 - Draft

and seven of the top ten companies ranked by the ratios of subscribers to population covered are not BOCs.76

82. In rebuttal testimony in response to the 1997 271 application of SBC Communications Inc., et al. in

Oklahoma,77 Professor Robert E. Hall dismisses this record of successful cellular competition with the incumbent

LECs on the ground that only a finite amount of spectrum has been assigned to that service. That fact would be

relevant if its effect were to limit the ability of the two certificated incumbent carriers to take on additional

business, either individually or collectively. The fact is, however, that the expansion of cellular service has not

been substantially capacity-constrained. On the contrary, the growth in subscribership has averaged over 40 percent

annually, stimulated in important measure by competing promotional offerings by the wireline and wireless rivals.

83. Perhaps the best evidence, however, that participation by incumbent LECs in the cellular business does

not foreclose competition comes from the wireline carriers themselves. Though they are presumably most

knowledgeable about the real risks of anti-competitive conduct directed at them by the incumbent wireline carriers, the

number of territories in which telephone company cellular affiliates compete with one another has grown rapidly from

about 5 in 1986 to more than 30 in 1995.78 And a company as knowledgeable and sophisticated as AT&T has sunk

billions into this market through its purchase of McCaw and PCS licenses—powerful objective evidence that its

frequently expressed concern about LECs discriminating against it, in favor of their cellular affiliates in their home

territories, has not deterred it from entering into competition with them.

84. Paging

85. Paging markets tell a similar story. While LECs have been important participants, they have always

been far from dominant. The largest suppliers, Paging Network and MobileMedia, are not affiliated with an LEC and

another large firm, SkyTel, became the first to introduce two-way paging (in September 1995). All told, radio

common carriers provide the largest share of these services; LEC affiliates account for only about 20 percent of the

total. The paging market is characterized by successful entry (SkyTel’s satellite service in 1987) and exit (MCI’s sale

76 Paul Kagan Associates, “Cellular Industry Eclipses Projections (Again),” Wireless Telecom Investor, March 14,

1994, No. 73.

77 Affidavit on Behalf of MCI, In the Matter of Application by SBC Communications Inc. for Authorization Under Section 271 of the Telecommunications Act of 1996 to Provide In-Region, InterLATA Services in the State of Oklahoma , CC Docket No. 97-121, pp. 41-42.

78 The 1995 number reflects direct competition among the former BOCs except for Pacific Telesis, which spun offits cellular company (now known as AirTouch Cellular).

- 39 - Draft

of its paging and cellular interests to McCaw in 1986, NYNEX’s sale to Page America in 1990 and MobileMedia’s

acquisition of BellSouth’s paging subsidiary in January 9679). Again, however, concerns that the LECs might cross-

subsidize their offerings of these competitive services or otherwise exercise their control of local facilities to obstruct

their rivals have proved unfounded; many years of competition have not eventuated in their dominating the business.

86. Voice Messaging Service (VMS)

87. Many LECs have long been allowed to provide information services, and Pacific and the other BOCs

also have been allowed to enter those markets in recent years—all without evidence that competition has been

undermined.80 Since the BOCs and GTE began offering VMS, consumers have benefited in at least two ways. First,

the monthly charge has dropped from $30 in 1990 to $5-$15 in 1995.81 Second, the LECs began offering VMS to

residential and small business customers, a thitherto untapped market segment. In five years, the BOCs’ participation

in this market increased from zero to over six million subscriptions, yet competitors have thrived and the BOCs and

GTE together account for just over 15 percent of the total revenues nationally.82

88. Customer premises equipment

89. Though barred from manufacturing, Pacific and the other BOCs have been permitted to distribute CPE.

As in the case of interLATA toll, competitors of the BOC must interconnect with the incumbent’s network—

typically in the form of connecting to a BOC-provided access line. There is no evidence—nor have there, to our

knowledge, been even assertions—that they have attempted, by exercising their market power, to exclude

competitors,83 let alone succeeded. Indeed, their collective market share of CPE distribution is small, on the order of

79 “M&A: MobilMedia Corp.,” Telecommunications Reports, January 8, 1996.

80 In addition, the FCC has ruled that the Open Network Architecture (ONA) safeguards are sufficient to deterconduct that has been alleged to be anticompetitive in the past. (Bell Operating Companies Joint Petition forWaiver of Computer II Rules, Order, 10 FCC Rcd. 13764 , 1995, par. 32.)

81 J.A. Hausman and T.J. Tardiff, "Benefits and Costs of Vertical Integration of Basic and EnhancedTelecommunications Services," prepared for filing with the Federal Communications Commission, ComputerIII Further Remand Proceedings, CC Docket No. 95-20, on behalf of Bell Atlantic, Bell South, NYNEX, PacificBell, Southwestern Bell, and U S West, April 6, 1995.

82 Ibid., pp. 5, 10.

83 NERA staff reviewed complaints filed against the BOCs with the FCC between 1985 and 1991 and found nocomplaints about the offering of interconnection of CPE.

- 40 - Draft

10 percent.84

90. IntraLATA toll

91. The final and most directly relevant evidence is to be found in intraLATA long distance. Nearly all

states permit intraLATA toll competition; and in none of them have Pacific and the other LECs been required to

divest themselves of their toll businesses or even to create separate subsidiaries. When the IXCs entered these

markets, they (i) started with small initial market shares, (ii) had few facilities within the LATA, so that they were

heavily dependent on the LECs for access to subscribers, (iii) did not have complete dialing parity, and (iv) had to

compete against inexpensive local calling within the LATA and overcome initial ignorance on the part of subscribers

that they now had a choice of intraLATA long distance providers. Even under these circumstances, LECs are losing

significant amounts of market share, particularly for large business customers that combine interLATA and

intraLATA traffic on the same dedicated facilities. Despite the fact that dialing parity has not been universally

required, the IXCs have already captured 22 percent of that market nationwide. (Schwartz Affidavit, p. 11, see ftn. 4).

This amount of market share loss by incumbents is comparable to AT&T’s in the interLATA market by 1988 (four

years after divestiture) and is all the more remarkable in light of the fact that intraLATA toll competition was not

even authorized in the two states that generate the most intraLATA toll calling, accounting for 46 percent of all such

calling (California and New Jersey), until 1995. The success of competition for long distance intraLATA business is

strong evidence that the hypothetical dangers of discriminatory treatment of BOC affiliates and their competitors are in

fact adequately precluded by existing regulatory safeguards.

92. In his Supplemental Affidavit, Professor Schwartz implies that we missed the point of his discussion

of intraLATA toll—namely, that it has been an example of intentional ILEC foot-dragging in the form of

opposition to presubscription. Since we testified on behalf of some of the RBOCs in support of their position, we

can assert with authority that the sole basis of their opposition was that it would upset the symmetrical

arrangement under the MFJ that reserved the intraLATA market for them while excluding them from the

interLATA. The Telecommunications Act decisively endorsed that position, linking intraLATA presubscription

with the lifting of the ban on the RBOCs’ offer of interLATA service. Moreover, Professor Schwartz’s emphasis

84 NATA, 1991 Telecommunications Market Review and Forecast, p. 60.

- 41 - Draft

on the importance of presubscription—a proposition with which we agree—underscores the validity of the point

we were making and repeat here: the ability of the IXCs, in a short period of years, to “capture 22% of that market

nationwide”—even without presubscription—is a powerful rebuttal to his general proposition that if the RBOCs

were permitted to provide interLATA service, they would use their local monopolies to impede competition.

V. THE IMPORTANCE OF SYMMETRY IN EXTENDINGTHE FREEDOM TO COMPETE

93. As we have already observed, the provision in the Telecommunications Act prohibiting joint marketing

of local exchange and interexchange services by major IXCs (carriers serving more than 5 percent of the nation’s

lines) that resell the incumbents’ local exchange services until such time as the incumbent LECs qualify for

interLATA entry—at which time, both they and all competitors would be permitted to market services jointly—is

clearly intended to preserve competitive parity between those two. It also reflects the extent to which the markets for

the several telecommunications services are converging and the necessity, if all participants are to compete on equal

terms, for all to have equal freedom to bundle their various services and offer customers one-stop shopping.

A. Blurring of Boundaries between Markets; the Importance of One-Stop Shopping

94. The industry has reacted to the recent dramatic technological and regulatory changes with a kaleidoscopic

variety of new ventures, typically involving entry into other markets, preexisting and new—sometimes by companies

operating alone, at other times through partnerships or acquisitions—as each attempts to take advantage of these

exploding opportunities (or perceived opportunities). All give rise to the prospect of competition among them.

While, as we have observed, a main strategy, since passage of the Telecommunications Act, has been to negotiate

entry into local markets by reselling the services and using the facilities of the ILECs, the three major IXCs have

made large commitments of capital as well—preponderantly in wireless and in service to business customers in

concentrated metropolitan areas. AT&T’s purchase of McCaw Cellular allowed it access to local networks covering

about half of the United States and it has strengthened its position in these areas by winning licenses in 21 major

markets in the recent PCS auctions, with bids totaling approximately $1.7 billion.85 AT&T has also explored

85 AT&T last year announced a new wireless system that would link customers directly to its network. According

- 42 - Draft

alliances with such small non-Bell local providers as Metropolitan Fiber Systems,86 as well as cable companies, such

as Time Warner,87 and it has recently announced its intention to acquire Teleport, one of the largest operating CLECs.

In terms of local wire services, AT&T states that it already has installed “more than 100 local switches and special

computers for routing traffic;88 to this would be added Teleport’s 41.89

95. MCI and Sprint also have made major commitments to entering the local market, bypassing LEC

facilities—almost exclusively, again, in wireless and concentrated business markets. 90 MCI first entered into a

partnership with British Telecom, which involved an infusion of $4.3 billions of BT capital.91 Most recently, it

announced its intended merger with WorldCom, which should, among other things, strengthen its local exchange

presence, because of WorldCom’s ownership of the largest operating CLEC, MFS. Sprint has joined with cable

to a recent article, “The system, which consists of a small transceiver mounted on the side of a house, could giveAT&T lightening-fast entry into the local phone business,” John J. Keller, “AT&T Unveils New WirelessSystem Linking Home Phones to Its Network,” The Wall Street Journal, February 26, 1997, p. B4. GregoryRosston, then the FCC’s Deputy Chief Economist, commented on this development as follows: “AT&T Corp.recently announced that it is developing PCS technology to be what Chairman Reed Hundt has termed the‘Raiders of the Local Loop,’” RCR, March 3, 1997, p. 59.

86 “AT&T Vows Battle to Offer Local Service,” The Wall Street Journal, October 27, 1995, p. A4.

87 John J. Keller and Eben Shapiro, "Time Warner’s Cable-TV Unit, AT&T in Talks," The Wall Street Journal,May 16, 1995, p. A3. Additionally, AT&T has filed plans with the FCC to bypass local connections, using anadvanced satellite communications system. B. Ziegler, J. Cole, Q. Hardy, "Satellite Plan Would Let AT&TBypass Local Networks," The Wall Street Journal, October 5, 1995, p. A6.

On the other hand, as but one prominent illustration of the rapidity with which these plans change, TCI—the country’s largest cable operator—has signaled its intention to abandon its previously announced ambitiousplans to offer telephone services, “Malone Says TCI Push into Phones, Internet, Isn’t Working for Now,” TheWall Street Journal, January 2, 1997, p. A1.

88 “AT&T Vows Battle to Offer Local Service,” The Wall Street Journal, October 27, 1995, p. A4. In fact, RobertAllen, AT&T’s chairman, stated on February 8, the day the Telecommunications Act became law, that it had theability to directly connect its large business customers to offer local exchange service. It helps put the 100switches into perspective to point out that the RBOCs currently have about 6,000 switches (not includingremotes). On the other hand, because (1) the switches of new local exchange entrants are likely to be placed inareas with higher volumes and (2) such entrants will be able to obtain unbundled switching from the ILECs, thissimple comparison of the respective numbers understates their importance.

89 Fitch IBCA, Inc., “Teleport Purchase Near-term Credit Negative for AT&T,” January 14, 1998.

90 MCI’s web page reports investments of $1.7 billion over the last three years, which have resulted in provisionof local service to large businesses in 21 major markets, plus residential service in California and Illinois.Considering the discrepancy between local prices and costs and the lack of full resolution of this problemthrough rate rebalancing and universal service funding reform, MCI’s pattern of investment in the apparentlymost lucrative areas is not surprising.

91 “BT Agrees to Invest $4.3 Billion for 20% of MCI; New Joint Venture,” Telecommunications Reports, June 7,1993.

- 43 - Draft

companies in a number of areas to offer basic telephone service through a joint wireless/wireline strategy,92 and,

although it recently spun off its cellular holdings, it has not abandoned vertical integration with wireless services.

Instead, like Pacific Telesis (recently joined with SBC) previously and U S West recently, Sprint has evidently opted

for the newer generation PCS as its main wireless platform, as its aggressive recent advertising campaign attests.

Sprint has also entered into partnership with Deutsche Telekom and France Telecom, which will allow the European

firms to obtain a large jump-start in the United States market and allow Sprint to do the same in Europe. This joint

venture, Global One, is aimed at serving the multibillion-dollar global communications market.93

96. Other firms similarly compete for position in these existing and emerging markets. Besides MCI and

AT&T, there are numerous other facilities-based carriers entering local exchange markets. A recent article reports that

as of the end of 1996, 27 CLECs (including MCImetro) had installed 139 switches throughout the U.S.94 In addition,

there are at least 23 ventures by electric utilities into telecommunications, making use of their rights-of-way, excess

fiber capacity95 and large capital reserves, which make the telephone and/or cable markets appealing to them. 96

97. These investments, partnerships and market interpenetrations are powerfully impelled by potential

economies on both the demand and the supply sides. The former spring from the attractiveness to consumers of one-

stop shopping—purchasing expanding bundles of services, at attractive prices, from single, familiar suppliers. On

the supply side, there are ubiquitous promised economies of scale and scope. The greater the capacity of switches and

transport facilities, the lower are unit costs: this means the incremental costs of adding capacity are lower than

average costs. Similarly, the use of common facilities permits the offer of additional services at incremental costs

much lower than if they had to be provided on a stand-alone basis. Entry into new lines of business at rates above

92 The trade press recently reported that there would be a limited delay in Sprint Spectrum’s plans to offer local

wireline services through upgraded cable facilities: “Sprint ... probably won’t rely on networks owned by Tele-Communications Inc., Comcast, and Cox Communications until the second wave of [PCS] rollouts scheduledlater this year.” (Vince Vittore, “Sprint PCS Launches in 6 More Markets,” Cable World, March 3, 1997.)

93 “With Variations, Sprint Announces European Pact, ”The New York Times, Late Edition, Friday, June 23,1995, p. D2.

94 Joan Engebreston, “The New Guys in Town,” Telephony, June 2, 1997, pp. 98-110.

95 For example, SCANA Corp., the parent company of South Carolina Gas and Electric, currently controls 2,500route miles of cable fiber through its subsidiary MPX Systems, Inc., and is planning to double that. “GrowingUtility Fiber Market Tempered by Considerable Hesitancy,” Fiber Optics News, Vol. 15, No. 19, May 15,1995.

96 One of us (Kahn) has listed these 23 in testimony on behalf of Boston Edison before the MassachusettsDepartment of Public Utilities in D.P.U. 97-96, Code of Conduct, November 21, 1997, Appendix Table 2.

- 44 - Draft

those low incremental costs provides the opportunity to earn contribution toward common and fixed costs and profits.

98. These economies have a dynamic as well as a static aspect. Complementary goods become more

plentiful and of higher quality as the number of users of any one of them—such as basic telephone service—increases.

Since consumers seem to prefer the supplier of communications services that gives them access to the largest number

of complementary services—internet access, information services, database access, video on demand—there is a very

strong incentive for the various participants in this industry, once freed from legal and regulatory barriers, to compete

in developing these new bundles of services.

99. Until recently, there had been (1) increasing expressions of concern that the pace of local competition

had been very slow, accompanied by (2) complaints, mainly by the IXCs, that the major impediment had been

foot-dragging by the ILECs in making the necessary interconnection arrangements with their challengers. Both

AT&T and MCI complained also of the high costs of their previously announced ambitious plans of large scale

entry on a facilities basis; the sharp reduction in the price British Telecom was to have paid for MCI and

uncertainty about changes in AT&T’s top management and company strategy were some symptoms of this more

pessimistic prospect. These various problems appear to be in process of resolution. AT&T changed its top

management, announced its intention to refocus its local entry strategy in the direction of using its own facilities,

along with unbundled elements of the ILECs, and has given that announcement credibility by its proposed

acquisition of Teleport, just as MCI’s announced merger with WorldCom appears to strengthen its local exchange

capabilities. Whether any of these efforts reflects a serious intention by the IXCs to serve the residential market

with terrestrial facilities97—whether, indeed, such an effort would make sense even if basic residential services were

not grossly underpriced—must remain a subject of profound skepticism.

100. As for the allegations of ILEC culpability, we have already pointed to the 1700 negotiated

interconnection agreements as of November 1997—31 of them in California. This is a remarkable achievement in

light of the complexity of these negotiations and the fact that the process is barely a year and a half old.

97 Significantly, in view of our previous references to the experience in Connecticut, the only substantial

competitors of SNET in that market have been Cox, Cablevision, and TCI (Cablevision recently announced itsattention to acquire TCI’s Connecticut assets)—because, of course, their cable already passes by the majority ofresidences in the State.

- 45 - Draft

B. The Adverse Consequences of Asymmetrical Restrictions on theAbility to Compete Reciprocally

101. As a general proposition, asymmetrical regulation attenuates both the incentives and the ability of some

providers to avail themselves of these scope and scale economies and to pass the benefits on to their customers under

the pressures of competition. As a result, large benefits are lost and costs incurred.98 Specifically,

• Stifling the incentives of RBOCs to offer new services costs society billions of dollars annually in

lost consumer benefits.

• As we have already observed, “one-stop shopping” can be worth a substantial part of the total value

of a product or service to consumers; competitors that can offer it have a considerable competitive

advantage over those than cannot.99

• The sacrifices of scope economies entails artificially inflated production costs.

102. The upgrading and modernization of the switched public network and the fullest exploitation of its

capability of offering a variety of sophisticated and innovative services—which are the central goals of the

Telecommunications Reform Act—depend not just on freeing the telephone companies and all others from restrictions

and handicaps on their ability to do so; it also requires offering all parties the full, undiluted incentives of a free

market system to undertake the requisite, typically risky investments.

103. Those incentives are of two kinds. The first is the stimulus of competition itself. The strongest case

for substituting the discipline of competition for that of regulation is the superior ability of the former to exert

pressures on all producers to be efficient and innovative, if they are to survive, let alone prosper. The second is the

self-interest of the telephone companies, freed from continuing restrictions on the services they are permitted to offer.

104. Particularly during the next several years, when we will necessarily continue to depend very heavily on

the ILECs for accelerating the deployment of an advanced telecommunications infrastructure, it is essential that we

98 Hausman and Tardiff, op. cit.

99 See for example, “Study Says Consumers Would Buy Bundled Services,” Telecommunications Reports, August12, 1996. That article reports that almost 80 percent of U.S. households would buy bundled services from asingle provider. Other studies have quantified the value of “one-stop shopping” to consumers. For example, seeTestimony of Arthur T. Smith on behalf of Southern Bell, Docket No. 930330-TP (Fla. P.S.C. July 1, 1994).This preference for one-stop shopping cuts across cultures: a study of Japanese consumers has estimated thevalue of the ability to obtain calling services from a single provider at about 14 percent of the average price.Timothy J. Tardiff, "The Effects of Presubscription and Other Attributes on Long-Distance Carrier Choice,"Information Economics and Policy, Vol. 7, 1995, pp. 353-366.

- 46 - Draft

not weaken the second of these incentives in a misguided effort to strengthen the first. Attempts to micromanage the

process of deregulation, we have found in other industries, are more likely to produce distortions than actually to

encourage efficient competition.100 Ultimately, both incentive systems require the shrinking of regulation and of all

such regulatory restrictions to the absolute minimum and entrusting protection of the public to deregulated

competition—subject, as always, to the constraints of the antitrust laws. But in the interim, delay in allowing

Pacific and the other RBOCs the opportunity to offer both local and interexchange services is not only unnecessary to

preserve equal competitive opportunities for equally efficient rivals. It would be blatantly anti-competitive, because it

would unnecessarily deny Pacific the ability to offer the same combinations of services, exploiting the same

economies of scope, as both Congress and the FCC have taken extraordinary pains to ensure will be available to their

competitors. And by weakening both the ability and the incentives of the BOCs to invest in modern infrastructure

and to innovate, it will tend to frustrate achievement of a central goal of the Act.

VI. SUMMARY AND CONCLUSIONS

105. The presumption in any system that is supposed to be governed by competition must be in favor of

permitting companies to enter whatever markets they want to enter, by integration or otherwise.

106. The desire of the BOCs to have the restriction on their ability to offer interLATA service lifted is a

desire to compete: that is clearly the place to begin in assessing their petitions. What they are asking for permission

to do is to integrate—to extend their operations from the supply of the “raw material”—local access—into the supply

of one of the major end-services making use of that input. Vertical integration of this kind is most likely to

recommend itself to companies—and, by the same token, to be socially creative and competitive in its effects—when

it represents a fuller use of existing capabilities—equipment, knowledge, managerial capabilities, marketing

facilities—of the integrating firm—that is to say, when it represents a fuller exploitation of potential economies of

scope.

107. That is undeniably the case here. The same Pacific facilities—switches, transport facilities, marketing

operations—as provide local exchange and intraLATA toll services can also supply long-distance services, which,

packaged with the others, are much more attractive to consumers than each or only some of them supplied separately.

100 Alfred E. Kahn, “Applications of Economics to an Imperfect World,” the Richard T. Ely lecture, The

- 47 - Draft

For exactly the same reasons, long-distance companies, cable and cellular operators are eager to use their existing

capabilities and facilities to add local telephone services to their offerings. Integration in both directions would,

manifestly, be competitive.

108. The ultimate economic question is whether Pacific and the other BOCs can possibly, by the exercise of

such diminishing but residual monopoly power at the local level as they possess, succeed in suppressing competition

as an effective force in the market they wish to enter—suppress competition, that is to say, as contrasted with

discommoding competitors. And this leads to our final and in a real sense definitive point. We find the ultimate

essential component of the successful strategy of cross-subsidization, predation or exclusionary tactics hypothesized

by opponents of BOC entry into the interLATA market—namely, the permanent removal or disabling of competitors

sufficient to enable the predator to recover the costs of those cross-subsidizations or other schemes by raising prices—

flatly inconceivable. The incumbent long-distance providers are in command of 100 percent of the market. They

have installed capacity that is not going to go away. The marginal cost of operating it is low, leaving its owners

with latitude for matching price reductions more than sufficient to dissuade any would-be predator. It is the present

long-distance companies that are the dominant firms in that market. In these circumstances, we find it simply

inconceivable that they would or could either be driven out of business or be so debilitated by discriminatory tactics

that might, hypothetically, be employed by the BOCs as to weaken the protection afforded consumers by their

continued competitive presence. In these circumstances, entry by Pacific and the other RBOCs could only be

beneficial to consumers in the interLATA market.

109. The Telecommunications Act, however, seeks also to encourage competition at the local level for its

own sake and not merely to ensure fair competition in the interLATA market. Setting aside the legal consideration

that that was not the purpose of the MFJ, which singled out the Bell successor companies, and therefore no more

justifies retaining the special MFJ-imposed restraints on them than imposing them afresh on all other ILECs, the Act

seeks to achieve this goal by requiring the RBOCs to make the stipulated required tools available to rivals. The

extension of that precondition recommended by the Department of Justice, however, could well be counterproductive,

even in terms of the achievement of that separate goal of local competition, because it offers the incumbent

interexchange carriers additional pretexts and inducements to refrain from negotiating in good faith the necessary

American Economic Review, Papers and Proceedings, Vol. 69, No. 2, May 1979, pp. 1-13.

- 48 - Draft

conditions of their entry—as the evidence we have cited suggests it has done—while unjustifiably extending the

period during which consumers are denied the benefits of additional competition in the offer of interLATA services.

- 49 - Draft

Alfred E. Kahn

Subscribed and sworn to before me this _____ day of March 1998.

____________________

My commission expires: ____________________

- 50 - Draft

Timothy J. Tardiff

Subscribed and sworn to before me this _____ day of March 1998.

____________________

My commission expires: ____________________