I.       Introduction

Even after the inception of competition in the interstate long distance telecommunications industry, the Federal Communications Commission has subjected AT&T to price regulation.  AT&T’s rivals, however, have been subject to a lower level of regulation since 1983.  This asymmetric treatment of AT&T was based on its “dominant” position in the interstate toll market. [1]   In October, 1995, the Commission granted a petition by AT&T to reclassify the firm as non-dominant in its provision of domestic interstate toll service.  In effect, this action removed domestic interstate toll service from price cap regulation and decreased the advance notice period for tariff filings for domestic-only services. 

Before reaching this decision, there was vigorous internal debate at the staff level.  To help inform this debate, the Competition Division of the General Counsel's office developed the domestic portion of the market analysis which is presented below.  The memorandum we presented recommended that the petition for non-dominance be granted on the grounds that AT&T cannot exercise individual market power to restrict output in order to raise prices for a sustained period of time.  This recommendation was consistent with the views of the decision-makers within the Commission.  The Commission also recognized that it continues to face a broader policy question:  even if AT&T lacks individual market power, should the Commission nonetheless impose some form of price regulation on the oligopoly of which AT&T is a part.  We believe that the analysis that follows will also be useful for informing that debate.

The discussion that follows is broken into three parts.  Part I briefly examines the general public policy reasons for imposing economic regulation on a market.  Part II summarizes some key elements of the structure and performance of the interexchange market.   Finally, Part III looks at the costs of price regulation.  We conclude that while the domestic interexchange market is certainly far from perfectly competitive (indeed, perfect competition is rare), the performance of this market is consistent with the vast range of unregulated American industries and, therefore, the prices charged by interexchange carriers should not be controlled by this Commission.  We, however, take a somewhat more agnostic view to the similar question for international interexchange service.

II.    Discussion

A.    Public Policy and Economic Regulation

Historically, economic regulation attempts to improve market performance -- and thus consumer welfare -- in areas of market failure, such as where there are natural monopoly characteristics of production  (e.g., through price regulation) or the presence of bottleneck facilities. (e.g., by requiring interconnection or equal access).  Following the Execunet decision in 1978, the Commission decided that the supply of interstate long-distance communications was not a natural monopoly and that an open entry policy that fostered competition with AT&T should improve market performance relative to a closed-entry policy complemented by economic regulation.  The decision reflected the Commission's judgment that competitive forces could replace economic regulation once actual competitors were both established and ubiquitous, thereby offering consumers an effective alternative to AT&T's interstate MTS.  The question now before this Commission is when that regulation of the interexchange market should end.

The evidence is clear that AT&T is no longer dominant in the overall, domestic interstate, interexchange market, and thus must be reclassified as non-dominant under Commission rules.  However, it is argued that the interexchange market is now a tight, three-firm oligopoly with AT&T as the price leader, and therefore AT&T, post-reclassification, would be able to achieve the practical benefits of a monopoly without any regulatory constraints through “tacit collusion.” [2]    Therefore, this reasoning goes, the Commission should either: (1) continue to price regulate AT&T as the oligopoly’s leader; or (2) directly price regulate the entire oligopoly.  The evidence, however, does not support either assertion.  Some of the salient aspects of the interexchange market that support this conclusion are discussed below.

B.     Structural and Behavioral Aspects of the Long Distance Market

In order to determine whether a market is performing well, it is necessary to examine both the structure of the market and the conduct of firms within that market. [3]   By viewing the interplay between the two, it is possible to assess whether there is sufficient rivalry to make it unlikely that industry participants will engage in collusive oligopolistic strategies that have the effect of raising prices and reducing output. [4]

1.       Structural Characteristics

The organization of the long-distance telecommunications industry today is complex.  While AT&T, MCI, and Sprint are an oligopoly of nationwide, facilities-based interexchange carriers (although there are at least nine other facilities-based interexchange carriers, one of whom, Worldcom [5] , has revenue approximating $4 billion annually and more than one-half the fiber-optic plant of MCI and Sprint), literally hundreds of regional and resale carriers compete for certain groups of customers also served by these major carriers. Certain structural attributes of the long-distance telecommunications industry directly affect the pricing of long-distance services by the facilities-based interexchange carriers in general and AT&T in particular.  More specifically, four aspects of industry structure effectively constrain AT&T's or any other carrier’s possible exercise of market power, namely: (1) lack of control of bottleneck facilities by AT&T; (2) only modest impediments to market entry by competitors in certain industry segments; (3) excess network capacity; and (4) the trend in AT&T's market share.

Bottlenecks:  The Commission’s regulation of bottleneck facilities has been directed primarily at local exchange facilities that control access to residential and business customers.  AT&T as a long distance telecommunications carrier does not operate bottleneck facilities that restrict access to end users and has not since the 1984 divestiture that split-off local exchange “bottleneck” facilities.

 

Excess Capacity: For the 12 facilities-based interexchange carriers that voluntarily report fiber optic capacity to the Commission, AT&T, MCI, Sprint, and  Worldcom have over 95% of the fiber miles in the long-distance industry (see Table 1).  AT&T’s individual share is just over 40%.  The remaining 5% of facilities is operated by a number of smaller carriers. The record indicates that capacity is not a binding constraint in this market.  Indeed, AT&T states, and no one disputes, that within only 90 days, its largest competitors could take nearly one-third of AT&T's switched traffic, using their existing equipment, and without any need to upgrade the electronics on their existing lit fiber facilities.  Similarly, the record indicates that within twelve months, AT&T's largest competitors could absorb another 31% of AT&T's total switched traffic (making a total of almost two-thirds), by using currently lit fiber and adding switched ports at a cost of about $660 million. 

Market Share: At the time of the First Report and Order, [6] AT&T had approximately 90% of the overall long-distance industry revenues.  During the 1980's, this market share fell approximately three percentage points per year.  Since 1990, when AT&T's market share was 65%, its market share has steadily fallen at about half that pace.  Currently, AT&T's market share is 56.6% by revenue and 58% by minutes, with MCI and Sprint at about 17% and 10% by revenues, respectively.  See Figure 1.  Although AT&T's market share is large, it is steadily declining, reflecting increased rivalry in the industry. 

Impediments to Entry: The most significant barrier to entry is the MFJ's prohibition on RBOC entry into the interLATA long-distance market, apparently soon to be eliminated via Congressional action.  All other potential barriers to entry in this market can be overcome.  The main problem identified in the record is the cost of building a network.  A ubiquitous nationwide fiber network requires an estimated $3 billion in capital investment.  That cost, however, is declining as the cost of fiber optic cable falls.  In addition, an entrant can develop its business as using resale even before it builds its own network.  More to the point, there are a number of firms, such as Cable and Wireless and Worldcom, that maintain large facilities-based networks, but are not focusing on the residential segment of the long-distance market.  For those firms, the decision to enter the residential segment of the market is determined by the question of whether the price and available volume in the residential market would be sufficient to support profitably the cost of marketing to households.  Thus, AT&T's advertising budget of 2.7% of revenues and its “brand name” may represent impediments to entry.  On the other hand, for some segments of the industry, entry is relatively easy.  At present, there are just over 450 resale providers in the domestic interexchange market, with a total revenue of $4.4 billion.  By year-end 1995, revenue in the resale industry is expected to reach $5.6 billion.

2.      Behavioral Characteristics

Volume (Minutes of Use): The introduction of competition and the implementation of equal access helped promote the rapid expansion of interstate switched access minutes.  As shown in Figure 2, industry total minutes continue to grow, with total minutes expanding 8.2% in 1994.  Over the past decade, long distance traffic (measured in minutes of use) has more than doubled. [7] Since the last quarter of 1990, long distance traffic has increased by 30%, with an average annual growth rate of 7.2%. 

Churn: Evidence indicates that in 1994, the churn rate was 27 million (i.e., consumers changed carriers 27 million times).  Of the 27 million changes in 1994, over 19 million were by customers who made only one change during the year.  Thus, about 1 in 5 residential customers changed carriers at least once last year.  For 1995, consumer churn is running at an annual rate of 30 million.

 

Decreasing Prices: The pattern of price movements for long distance telecommunications services is mixed, although trending downward, with differences depending primarily on the volume of use. On average, however, prices have fallen.  Basic residential MTS rates, generally paid by the lowest volume users, have risen over the past four years while volume discounts by AT&T have become deeper.  Specifically, the prices available to AT&T’s residential customers with bills over $10 per month are down substantially.  As shown in Table 2, prices for the highest volume users (500 to 1,000 minutes) are down by about 28% since 1991. [8]   Even residential customers with monthly long distance bills of about $20 have seen prices fall by nearly 15%. 

Setting aside differences in volume and prices among residential customers of long distance services, Table 3 shows that AT&T’s domestic average interstate revenue per minute fell 10% between 1991 and 1994, declining from 17.8˘ to 16.2˘ per minute. Thus, declining prices for higher volume users, those spending over $10 per month on long distance services, have more than offset the increase in basic rates.

Table 3 also illustrates that while access costs have fallen only 0.5˘ from 1991 to 1994, the domestic average interstate revenue per minute fell by almost 1.6˘.  This evidence shows that reductions in prices have outpaced reductions in access costs. [9]   Table 4 confirms this assessment, showing that AT&T’s best prices, for all but the lowest volume residential customers, have fallen faster than access cost. [10]   Indeed, best prices for the highest volume users (500 and 1,000 minutes) have fallen in excess of 30% more than reductions in access costs. Therefore, the assertion that rivalry in the long distance market has been insufficient to force prices down faster than access charge reductions is incorrect.

While not all customers have seen prices fall as dramatically as others, as discount plans become increasingly price competitive and as basic prices inch upwards, more customers are switching to the cheaper discount plans. A recent survey shows that between May, 1994, and May, 1995, the percentage of AT&T customers on discount plans increased from 35 to 42% (see Table 5). [11]    The survey also shows that in 1995, 62% of AT&T's domestic residential interstate revenues were received from customers on discount plans (see Appendix).  Since revenue per minute is less for customers on discount plans, it follows directly that the percentage of minutes on discount plans far exceeds the minutes for which the basic rate is paid. [12]   Finally, the survey reveals that the average long distance bill over the past year fell 10% (see Appendix).

 

Robust Resale Market:  As noted above, there are now approximately 450 providers in this market, with a total revenue value of $4.4 billion.  By year-end 1995, this value is expected to reach $5.6 billion.  A considerable number of conduct-related complaints have been filed by resellers against AT&T.  However, just over 25% of resale capacity was provided by AT&T in 1994 and, by year-end 1995, AT&T will no longer be the largest supplier of resale capacity in this market.  Rather, Worldcom will be the largest with 23.2%, followed by AT&T (20.3%), MCI (13.8%), and Sprint (13.2%).

3.       Nature of Market Performance

Although observed market performance does not resemble that predicted by the model of perfect competition, the conduct of interexchange carriers suggests that the quality of market performance, while mixed, is satisfactory and consistent with rivalrous competition among few firms. The evidence on prices for long distance telecommunications services demonstrates that the major carriers are not pursuing a joint profit maximization strategy. [13]   Specifically, a primary characteristic of an oligopoly where the firms attempt to collude and maximize joint profits is the stifling or lack of price competition.  Under such conditions, prices should go up and volume should go down.  The performance here is the reverse.  As shown in Table 2, the prices available to AT&T’s customers with bills over $10 per month are down substantially. Figure 3 shows that these customers (those paying over $10 per month for long distance services) represent 93% of all residential minutes.  Thus, an overwhelming majority of residential long distance services sold by AT&T are being offered at prices substantially below their 1991 levels. 

The overall picture is inconsistent with tacit collusion, joint profit maximization, or the lack of price competition.  The evidence is also consistent with the conclusions of the Commission's January, 1995 price cap order for AT&T. [14]    In that order, the Commission re moved AT&T's commercial services from price cap regulation, [15] because it found that there are enough competitive alternatives to AT&T's commercial long-distance service to constrain AT&T's exercise of monopoly power for these services. [16]   To support this decision, the Commission held, inter alia, that several factors would prevent the largest interexchange carriers from engaging in oligopolistic coordination resulting in prices above competitive levels.  For example, the Commission held that because the networks of the interexchange carriers have significant excess capacity, the cost of serving additional traffic is very low and, conversely, the cost savings associated with a traffic reduction are also "quite low."  Thus, concluded the Commission, these carriers have a great incentive to protect their substantial investment in these facilities by keeping their prices low to attract and keep customers.  Moreover, the Commission also found that the possibility of oligopolistic coordination to raise prices is limited by the demonstrated willingness and ability of commercial long distance customers to move among various commercial long distance services offered by the numerous interexchange carriers. [17]

Similarly, in the Mobile Services Order, [18]   the Commission found that there was sufficient competition for cellular services to forebear from the tariffing requirements of the Communications Act in markets with only two suppliers. [19]   Specifically, the Commission recognized that while duopolists may be able to sustain a shared monopoly -- with the attendant higher prices -- by either tacitly agreeing not to price aggressively or restricting the amount of investment in new capacity, there could be reasons why it may be difficult or unprofitable for cellular providers to coordinate their actions in this manner.  First, the Commission recognized that collusion could be limited by competing services, i.e., the cross-price elasticity of supply. [20]   Second, the Commission concluded that in addition to present, actual competition, the threat of potential competition in the future could affect current pricing and investment.  That is to say, a cellular provider may choose to invest in additional capacity today in anticipation of gaining future market advantage rather than to restrict output through tacit or explicit collusion with a fellow duopolist. [21]   Finally, the Commission recognized that other factors, such as rapid changes in the nature of the product, or even quality and price competition, may also diminish the likelihood of collusion.  Indeed, the Commission specifically found that "[c]omplex pricing structures, such as are used in the cellular industry, make it difficult for a carrier to sustain collusive pricing." [22]

Finally, in the California Petition R&O, the Commission, applying an structure-conduct-performance analysis, rejected the contention that market performance of cellular duopolies in California warranted continued rate regulation of intrastate rates.  The Commission held that to justify such continued regulation, the State was required to show that the conduct and performance of cellular providers ill-served consumer interests by producing rates that were not just and reasonable or that were unreasonably discriminatory. [23]   After analyzing the structure, conduct and performance of two-firm cellular markets, the Commission concluded that the evidence did not justify continued rate regulation.  The Commission found that: (1) cellular rates were declining; (2) entry by PCS would result in direct and fundamental changes in the cellular market; (3) there was no evidence of "systematically collusive or other anticompetitive practices"; (4) there was no evidence of widespread consumer dissatisfaction with cellular providers; and (5) there was no evidence that firms were withholding investment as a means of restricting output and thus boosting prices.  In general, in rejecting California's bid to continue its regulation of the cellular market, the Commission emphatically rejected California's contention that "any evidence of market imperfection [is] proof of the need for continued rate regulation." [24]   The parallel to the interexchange market is obvious.

C.     Inefficiencies of Continuing Price-Cap Regulation and Degraded Market Performance

The brief discussion above demonstrates that the interexchange market, while an oligopoly, has sufficient rivalry to preclude joint profit maximization or tacit collusion.  The question therefore becomes: should the Commission price regulate an oligopoly that nevertheless produces satisfactory market performance?   Even if the performance does not equate to perfect competition, the answer is clearly no.

Price-cap regulation of AT&T’s services involves both direct and indirect costs.  Direct costs include:  (1) expenses incurred by AT&T in complying with the Commission’s price-cap rules; and (2) the portion of the Commission’s budget that is reasonably allocable to handling price-cap regulation of AT&T, including reviewing its tariffs and enforcing price-cap rules.  Indirect costs of price-cap regulation of AT&T’s services include consumer benefits that are foregone as a consequence of price-cap regulation.  Although the direct costs of price-cap regulation of AT&T’s services are not trivial, the indirect costs are likely to exceed the direct costs over the long term.

By capping AT&T’s basic residential MTS rates, price-cap regulation may prevent AT&T from introducing more economically-efficient rate structures for its long distance service.  This distortion introduced by price-cap regulation of AT&T’s services may seem counter-intuitive, although entirely plausible once it is recognized that a movement toward economically-efficient rate structures may require some rates to increase while other decrease. [25]   Both the intensity of demand and the extent of price sensitivity will differ between and among different residential subscribers.  As a result, an economically-efficient rate structure that recognizes such differences in demand between different consumer groups should reflect different markups for different user groups, if setting price equal to service-specific marginal cost is infeasible. Moreover, given the inertia effects of many long distance customers in responding to changes in price or in selecting an improved price plan, price increases may provide just the required stimulus for long distance subscribers to consider other pricing plans that may actually reduce their total spending for their customary levels of long distance calling.   Thus, by constraining AT&T’s ability to redesign its rate structure, consumers lose the net economic benefit that more efficient rate structures can provide.

Further losses in net economic benefit to consumers may be attributable to the service and price plan innovations that notice periods discourage AT&T from offering to long distance telecommunications customers.  Innovation that improves consumer welfare is rewarded by transitory profits that an innovating firm can earn before the innovation is copied or imitated by competitors.  The existing notice period required before AT&T can introduce modifications to its residential MTS pricing plans provides sufficient time for non-dominant carriers competing with AT&T to emulate AT&T’s proposed innovation.  Thus, the notice requirement discourages innovation by AT&T and results in some loss in dynamic economic efficiency that harms consumers. 

Finally, the existence of price-cap regulation or other variants of government price controls represents a disincentive for unregulated firms outside the telecommunications industry to enter industry sectors subject to direct government intervention.  While the removal of price-cap regulation does not affect the Commission’s ability to investigate consumer complaints or remove the requirement to file tariffs, it signals nevertheless a willingness to reduce the extent of government intrusiveness in the long distance telecommunications industry.  This policy stance should be viewed favorably by both potential entrants into the long distance industry and the investors that must commit the funds to make market entry possible.

In sum, these static and dynamic economic efficiencies, or indirect costs, attributable to price-cap regulation of AT&T’s residential MTS can be completely avoided if such regulation is eliminated. As shown in Section B, there exists today sufficient competitive rivalry to constrain AT&T from exercising market power to an extent that would adversely affect residential customers of MTS.  As a result, eliminating price-cap regulation of AT&T’s residential MTS may be expected to improve market performance immediately by avoiding both the direct and indirect costs of price-cap regulation.

III. Conclusion

In sum, while the performance of the interexchange market is less than perfect, the interexchange market is still sufficiently competitive not to require price.  The structure, conduct, and performance of the domestic interexchange industry is comparable with many other American industries, yet these industries are not subject to price regulation. [26]   Where sufficient rivalry exists, perfect competition is not a prerequisite to the abolition of price regulation.  As the Commission held last month, “[a]lmost all markets are imperfectly competitive, and such conditions can produce good results for consumers.” [27]   The termination of existing price regulation of AT&T does not suggest a diminished desire for additional competitive entry into the interexchange market or an increase in the extent or intensity of price competition.  Indeed, the Commission should continue its policy of removing all regulatory barriers to entry and identifying other impediments that deter efficient entry.  Such pro-competitive scrutiny and competition advocacy by the Commission should greatly enhance consumer welfare over the longer term relative to a reactive focus on regulating prices in a fast‑paced market where the efficiency of such regulatory intervention may be doubtful.

 

Appendix:  Summary Tables of the Bill Harvesting Survey