The Telecommunications Act of 1996 (Act) was expected to open telecommunications markets to competition, thereby substantially benefiting consumers. In particular, it was expected that the market for residential basic local exchange service, traditionally supplied only by incumbent local exchange carriers (ILECs), would become robustly competitive. There was an equally strong expectation that long distance markets would be opened to the Regional Bell Operating Companies (RBOCs) and large independent local exchange companies (LECs), with consumers further benefiting from increased competition in the toll market as well.
Unfortunately, the Act has not worked too well so far. The basic design of the Act and the way it has been implemented have led to a paralysis or gridlock that has resulted in a great deal of allowed local exchange competition but limited entry into long distance by additional carriers. Though large LECs such as GTE have entered the market for so-called in-region interLATA long distance, the RBOCs still have not been granted permission to enter these markets. So far, parties as diverse as the Federal Communications Commission (FCC), consumer group leaders, interexchange carrier executives, and RBOC executives have all expressed disappointment with the results of the Act on telecommunications competition.
Why hasn't the Act generated the robust competition its supporters assumed it would? This article argues that there are two primary reasons. First, there are basic design flaws in the Act that render it an ineffective public policy, no matter how well implemented. Second, the subsequent implementation of the Act and the way resale discounts were set by regulators caused a Catch-22 situation to develop, which has forestalled the benefits of the Act so far.
More specifically, the designers of the Act did not examine the economic attributes of the very market they sought to correct through governmental intervention, the residential basic local exchange service market. For this reason, the Act did not tailor the right public policy tools to local exchange markets. Instead, the Act empowers and encourages the FCC and the state public utility commissions to use policy tools that would work for various market failures, despite having no evidence that local telephone service consumers actually are being harmed by any such market failure. Specifically, the comprehensive unbundling of network elements and resale are not necessarily the correct public policies for resolving the observed lack of entry into local exchange markets. Much simpler policy tools could have been implemented first to see if unbundling and resale were really necessary.
In addition, the implementation of the Act was carried out in ways that deviated from sound economic principles and for that reason the results have been ineffective so far. Perhaps the best example to illustrate this is the FCC’s implementation of the so-called "avoided cost rule" resulting from section 252(d)(3) of the Act. The avoided cost rule was used to determine the rates at which the RBOCs could resell their retail services to other service providers (if negotiation failed to yield an agreed-upon rate). The FCC’s avoided cost rule and its further implementation by the state public utility commissions led to low resale rates, and hence made it easy for new firms to supply local service markets via resale of ILEC services. However, entry by resale did not count when the RBOCs sought entry into long distance markets by demonstrating their local service markets were open to competition. Only a preponderance of facilities-based competition in local exchange markets would have allowed the RBOCs to enter long distance markets.
These and other examples will be discussed in more detail subsequently. This article is organized as follows. Section II discusses the major design flaw of the Act: the use of policy tools mismatched to the observed lack of entry into residential local exchange telephone markets. Section III discusses specific issues in the implementation of the Act that muted its effectiveness. Sections IV and V discuss the other major design flaws of the Act that made ineffective public policy inevitable. Section VI provides policy conclusions.
II. The First Major Design Flaw: How the Act Fosters Competition in Local Exchange Service Markets

A. Introduction
This section discusses the general attributes of local exchange markets, and lays the groundwork for the arguments made subsequently in this article. It argues that in a general sense, Congress did not correctly diagnose market failure in residential local exchange markets, hence the policy tools that it prescribed through the Act were inappropriate because they were premature. Because these policy tools were premature, it is generally true that they are not capable of yielding the benefits to consumers Congress envisioned. The prescription of these inappropriate policy tools constitutes a basic design flaw in the Act that makes it incapable of expedient benefits, though effective public policy may still be possible eventually.
Prior to the passage of the Act, local exchange markets were largely the exclusive bastions of the local exchange carriers (LECs). Since the early 1990s, sporadic inroads to local exchange competition were being made, notably in Illinois and New York, but residential customers usually had no practical alternative to the dialtone service offered by the LEC.[1] Local exchange markets were (and still are) stringently regulated by state public utility commissions as to price, quality of service, and availability. This led to two observations that would hold true for virtually any local exchange market: prices that were quite low (perhaps even below cost), and a lack of entry into these markets by other firms.
It was long maintained that local service prices were below cost, though little publicly available evidence of this emerged in regulatory hearings.[2] The policy of residual pricing[3] did not require detailed regulatory scrutiny of local exchange service costs, and even when such costs were computed, they were held as proprietary and hence shielded from the public record. The observed lack of entry was caused primarily by a direct proscription of entry due to the public utility law in most states or the low prices themselves, which made even allowed entry unprofitable.
B. What Problem Was Congress Trying to Solve?
Given the general attributes of local exchange markets, it was easy to consider these markets a perfect target for the Act’s competition-enhancing policies. However, the drafters of the Act seem to have glossed over the following two basic questions: To what monetary extent are local exchange markets problematic in ways that harm consumers? What specific policies should be enacted to alleviate these problems (if any)?
Normally, the overarching economic reason for fostering competition-enhancing policies is some type of market failure (caused by conditions limiting competition) which makes consumers worse off than they otherwise would have been (e.g., due to high prices, low quality, etc.). However, because retail prices in residential markets had been kept at very low levels to pursue universal service objectives,[4] it begs the question of exactly what problem Congress sought to solve in the first place. Policies designed to enhance competition in a market make sense only if the following conditions hold: (1) prices in that market are too high (due to the possession of market power by the incumbent firm); (2) the competitive interaction between firms resulting from a competition policy would curb that market power; (3) prevailing service quality levels or the number of choices available to consumers are deficient in ways detrimental to consumers; and (4) the direct regulation of retail prices is an ineffective means of correcting this problem. Unfortunately, two of these conditions (the first and the fourth) did not hold in local exchange markets at the time the Act was passed, and they do not hold now; the third condition is certainly arguable. Thus, at the time the Act was passed, residential local exchange markets were not yet ripe with unrealized consumer benefits.
C. Competition-Enhancing Policies Must Be Adjusted to Subsidized Markets
At the time the Act was passed, the stringent regulation of the residential local service market was already doing an effective job of muting any market power an ILEC may otherwise have had for local service. This regulation did so by keeping prices below market levels. A market in which the retail price has been set at below-market levels is not a market in which prices are too high due to market failure, and hence which can readily benefit from a policy designed to enhance competition. It is just the opposite, a subsidized market. If retail prices are already below efficient market levels and residence local service penetration levels meet government objectives, then the direct regulation of downstream prices is a presumptively effective policy tool. The reason that competition from alternate suppliers does not take place is most likely because prices have been set so low in the first place, with regulatory sanction.[5] This means that there are no positive profits an entrant can expect to earn in this market. Even if entry is not proscribed by law, at the subsidized retail price the lack of a profit opportunity makes entry rather unlikely to occur. Hence, if low regulated prices are the reason entry is not observed, then any policy designed to enhance competition[6] probably has little ability to make consumers appreciably better off in the short run. This is because it is unlikely that competition will make prices appreciably lower than they are currently.
The demand characteristics of local exchange service also make it difficult for the Act or any other public policy to deliver significant incremental benefits to consumers. The price elasticity of demand for local exchange service is extremely low, practically zero.[7] This may seem like the type of arcanum only economists appreciate, however, it has much to say about how competition in the local exchange (a primary objective of the Act) can result in additional benefits to consumers within a reasonable time. The lower the price elasticity of demand for local exchange service, the greater the retail price decrease needed (as a result of the increased competition due to the Act) to produce a significant increase in consumer benefits in the form of increased consumer surplus.[8] Because of the very small price elasticity of demand for residential local exchange service, a considerable price reduction is necessary to produce a small benefit to consumers after switching to a supplier offering a lower price.[9]
Unfortunately, the way the Act fosters competition in local exchange markets makes significant efficiency-based price decreases unlikely. The Act provides for two major competition-enhancing policy tools for local exchange markets: resale and the purchase of unbundled network elements. These allow new entrants to purchase the components needed to provide local service that they cannot or will not provide themselves. In other words, new entrants are required to provide only the components of production (e.g., retail sales operations) that are most likely available to all firms anyway. It is unlikely that for this latter component of production, any single firm will have a significant cost advantage. However, for the Act to result in price reductions that lead to significant increases in economic efficiency (due, in large part, to increases in retail market consumer surplus), it is this area (e.g., retail sales operations, billing) in which entrants must have a significant cost advantage over the LEC. These are the only costs new entrants not possessing their own networks can control.
D. Summary
In passing the Act, Congress seems to have relied on a blind allegiance to competition, assuming that because local exchange service markets were served solely by ILECs, competition must have been foreclosed, and consumers must have been harmed. Yet these assumptions ignore the nature of existing regulation and other economic attributes of local exchange markets. In taking a closer look, it is clear that residential local exchange markets do not suffer from the type of market failure that the competition-enhancing policies of unbundling and resale are designed to remedy (although the symptoms are quite similar). This fundamental failure of Congress to recognize the characteristics of local exchange markets, and tailor the policy tools to the problem at hand, makes the Act far less effective than it otherwise could have been.
This is not to say that the policy tools of unbundling and resale are useless. The problem is that policymakers never had a proper basis for knowing if such policy tools were really necessary. A better approach would have been to rebalance carrier access charges and local exchange service rates by pursuing access charge and universal service reform. Rate rebalancing in conjunction with access charge and universal service reform would have allowed local service prices to most consumers to rise to higher levels (e.g., the amount of current prices plus the amount of subsidy per line). Once rates were rebalanced, Congress could have set aside all remaining proscriptions to local exchange entry that existed in state public utility law, as the Act did. Given such policies, the problem that prices may be too low to attract even efficient entrants is removed, new entrants could decide whether to enter the market based on proper financial criteria, and policymakers could monitor these markets to see if a lack of efficient entry persists. If entry still did not take place, then the more stringent policies of unbundling or resale may, in fact, have been appropriate.
III. Problems in the Implementation of the Act

A. Implementation of the Avoided Cost Rule
Under the Act, section 252(d)(3) establishes a standard for the pricing of resold services, when no agreement on pricing can be reached, through the so-called "avoided cost rule:"
The Federal Communications Commission (FCC) issued guidelines on the implementation of section 252(d)(3) in August 1996 in its so-called Interconnection Order.[11] Unfortunately, the Interconnection Order defined the term "avoided costs" quite broadly and, from the perspective of mainstream economics, incorrectly. The FCC defined "avoided costs" in Appendix B-Final Rule § 51.609(b) as follows: "Avoided retail costs shall be those costs that reasonably can be avoided when an incumbent LEC provides a telecommunications service for resale at wholesale rates to a requesting carrier." This definition has been stayed pending appeal. The Interconnection Order also stated that:
Using economic criteria, avoided costs for section 252(d)(3) should be defined as the differential between: (1) the ILEC's total costs when offering the retail service only to end users, and (2) the ILEC's costs when functioning as both a seller of wholesale services to resellers and a seller of retail services directly to end users. This differential is the costs avoided by no longer providing the pre-resale level of retailing functions, net of any additional costs incurred in the provision of the wholesaling functions.[13]
Given the above definition, the avoided costs of resale will include two major components: (1) the costs of providing the service at retail which cease to be incurred due to the wholesaling of that service to a reseller in lieu of providing the service at retail directly to end-users, minus (2) the costs of wholesaling the given service to resellers. [14]
Because the FCC defined avoided costs expansively and the state public utility commissions in large part followed the guidelines of the FCC, the discounts off retail prices for resellers were higher than they should have been. This resulted in deep-cut discounts that gave entrants an economic incentive to favor resale over the purchase of unbundled network elements. Unfortunately, resale encourages entry by prospective local service providers (LSPs), but it does not lead to strong incentives for firms to engage in price competition. There is no compelling reason for firms to lower prices under this type of industry structure. What do economically rational resellers prefer -- low retail prices or high ones? Obviously, they prefer high prices combined with high discount rates. This is not the combination that expediently leads to significant downward pressure on retail prices via increased competition.
Unfortunately, the preference entrants would have for resale did not mesh well with the magnum of proof required by the FCC for RBOC entry into long distance markets, as the next major section discusses.
IV. The Second Major Design Flaw: The Magnum of Proof for Entry into Long Distance Markets

A. Local Exchange Entry by Resale Does Not Hasten RBOC Entry into Long Distance Markets
Under section 271 of the Act, an RBOC is allowed to pursue entry into the in-region interLATA long distance market via two separate tracks. The Act's so-called "Track A" requires the RBOC to show that it is "providing access and interconnection to its network facilities [to] one or more unaffiliated competing providers of telephone exchange service . . . to residential and business subscribers."[15] The Act states that "such telephone exchange service may be offered by such competing providers either exclusively over their own telephone exchange service facilities or predominantly over their own telephone exchange service facilities in combination with the resale of the telecommunications services of another carrier."[16] The intent of Congress was clear,[17] and the Act has been interpreted by the FCC as meaning that entry occurring by pure resellers does not entitle an RBOC to pursue in-region interLATA long distance under Track A.[18]
The Act's so-called "Track B" allows an RBOC to pursue in-region interLATA entry if, after ten months past the date of enactment of the Act, no provider has requested access and interconnection at least three months before the RBOC applies for permission to provide in-region interLATA long distance, and a statement of the terms and conditions the RBOC generally offers to provide such access and interconnection has been approved or permitted to take place by the state public utility commission.[19]
The above requirements are not problematic unless new entrants to local exchange markets prefer pure resale to the provision of local service through their own facilities-based networks. If resale is the preferred method for entering local exchange markets by a preponderance of entrants, the above requirements constitute a serious design flaw. This is because pure resale reduces entry barriers into the local exchange market (and, in fact, mutes ILEC market power), but entry by facilities-based carriers is required to allow RBOC entry into in-region interLATA markets. Thus, a Catch-22 exists in the basic design of the Act, depending on how the Act is implemented. If the FCC and the state public utility commissions mandate deep-cut discounts on resold services to enable entry into local exchange markets, then prospective entrants would have a strong preference for entering via resale rather than through the purchase of unbundled network elements or investment in their own facilities. Yet this type of entry does not allow the RBOC to obtain permission to enter the in-region interLATA long distance market.
As Section III of this article has pointed out, the avoided cost rule and its implementation led to low resale rates. Because of this, prospective entrants to the local exchange market have a clear preference for resale over other methods of entering this market. The various state public utility commissions, in their roles as arbiters of interconnection rates and resale discounts, have allowed prospective LSPs to enter local service markets via resale policies that feature deep-cut discounts off existing retail prices. Because the rates for unbundled network elements and resale discounts have not been set via a systematic approach (which would synchronize these rates with each other and minimize arbitrage opportunities), the Catch-22 cited above has emerged. Resale has led to entry into local exchange markets; yet, to be allowed into the interLATA market, the RBOCs must demonstrate that entry by facilities-based carriers has occurred.
This has forestalled the benefits that otherwise would have accrued to purchasers of long distance. In economic terms, because long distance expenditures constitute a larger household budget share of income than local exchange services, and the price elasticity of demand for long distance is far higher than that of local service, the foregone benefits of increased long distance competition most likely far outweigh any gains experienced so far from local exchange competition.
B. The Moral Hazard Problem
The greater the interexchange carrier (IXC) entry into local exchange markets, the greater the likelihood that RBOCs will be allowed into long distance markets; but IXCs may prefer the status quo. As was explained above, before an RBOC can enter the in-region interLATA long distance market, it must meet the requirements of section 271 of the Act. Compliance with section 271 of the Act can be demonstrated by meeting the requirements of either Track A or Track B. As described above, Track A requires the RBOC to show that it is "providing access and interconnection to its network facilities [to] one or more unaffiliated competing providers of telephone exchange service . . . to residential and business subscribers."[20] Track B allows an RBOC to pursue in-region interLATA entry if, after ten months past the date of enactment of the Act, no provider has requested access and interconnection at least three months before the RBOC applies for permission to provide in-region interLATA long distance. Under Track B, a statement of the terms and conditions the RBOC generally offers to provide such access and interconnection must be approved by the state public utility commission.[21]
No RBOC failed to receive requests for access and interconnection from prospective entrants in a given state, so Track B has been moot. However, under Track A, an interexchange carrier (IXC) can apply for access and interconnection to function as an LSP. This gives an IXC an opportunity to argue that the 14-point checklist of the Act has not been met, and hence that the RBOC should not yet be allowed into the in-region interLATA long distance market. In other words, to the extent that the IXCs can discourage local exchange competition by expressing dissatisfaction with the terms of access and interconnection, they can delay the point in time at which the RBOCs can enter the in-region interLATA long distance market.
Hence, the difficulties of some of the major IXCs in getting into local service may be a form of rent protection in the interstate long distance market.[22] If IXCs believe that expected revenue losses from the future competitive inroads of the RBOCs will not be more than offset by the profits from entering the local exchange market, they have an incentive to argue that: (1) RBOCs have forestalled IXC entry into local exchange markets by not meeting the Act's "checklist" items, and (2) RBOC entry into interLATA long distance markets is not in the public interest. To foster local exchange competition, the Act makes a very important implicit assumption: that important prospective LSPs, the IXCs, desire to enter the local exchange market, knowing that in doing so they will hasten the ability of the RBOCs to enter the in-region interLATA long distance market.
V. The Third Major Design Flaw: The Expansive De Facto Definition of "Essential Facilities"

A. A Parable About a Monopolist Brewery
Many years ago, in a mythical place called Gatesville, the Soft Micro-Brewery (SM) was the sole provider of beer. SM possessed all the latest, most technologically advanced and economically efficient brewing equipment available. Despite this, it dominated the market for one reason and one reason only: it owned the entire water supply of Gatesville. For this reason, no other brewer could compete with SM. No other brewer could get the water needed to brew beer, and the costs of transporting water in from other areas was prohibitively high.
As a result, SM had no competition, and the retail beer market in Gatesville did not perform well, in the very vocal opinion of local economists. A six-pack of average-quality beer was $8.00, and only one variety was available: the Dominant Pale Ale. SM essentially had a stranglehold on the market.
The Chancellor of Gatesville became jaded with this situation, for he loved beer and had a modest civil servant’s income. Seeking to take regulatory action, he formed a Regulatory Committee to study the problem. The committee determined immediately that the major problem was SM’s ownership of key productive inputs other firms did not have, and it sought a regulatory solution. The committee decided that access to these key inputs by other brewers would solve the poor economic performance of the retail beer market, yielding lower prices, greater quality, and more choices for consumers. The concept of "unbundling" was discussed as the way to ensure that other brewers had access to the key productive inputs. The committee would set access prices for the inputs and compel SM to sell them to other brewers. Local economists were consulted as to the efficient access prices, and, because Gatesville is a mythical place, there was consensus among them as to the appropriate access prices.
Almost immediately the question of where to limit the extent of unbundling came up. In other words, to foster competition, should the Regulatory Committee mandate access to only the water, or should it mandate access to all of SM’s manufacturing equipment? Predictably, there were two sides. One school of thought (the Riparians) held that though the water was not available to other brewers, such brewers could purchase all of the same equipment as SM from equipment suppliers. Thus, the government should compel access only to the water, nothing else. Providing access to the water would open the retail beer market to competition, and the competitive process would correct the obvious market failure. The opposing school of thought (the Comprehensivists) held that it was necessary to mandate access to SM’s water supply and manufacturing equipment, allowing prospective new entrants to purchase access to any of SM’s existing manufacturing processes, in whole or in part.
Eventually, the Regulatory Committee decided that the water was the only true "essential facility" needed by other firms to provide competition to SM in the retail beer market. All other brewing equipment and human capital were available in existing, well-developed equipment and labor markets, hence there was no need for government intervention to provide access to such facilities. The retail beer market was opened up to competition through mandated access to the water, and soon the market was operating efficiently.
The Riparians stayed to regulate SM and handle other regulatory matters; the Comprehensivists left to become telecommunications regulators.
The moral of this parable is this: unbundling should be confined to only those wholesale (or "upstream") facilities that, by virtue of their lack of availability, prevent an efficient competitive retail (or "downstream") market from operating. Governmental intervention in markets beyond this level of unbundling is unnecessary to yield a competitive outcome in the retail market. Intervention beyond this level cannot increase the probability of efficient entrants[23] entering the market, but it can introduce significant transactions costs.
B. What the Act Required as Access to De Facto Essential Facilities
The Act required all states to allow telecommunications competition, preempting all formal barriers to entry, state and federal.[24] All LECs have the duty to provide the resale of their telecommunications services, number portability, dialing parity, access to rights-of-way, and reciprocal compensation arrangements for the transport and termination of telecommunications.[25]
The ILECs have some additional, somewhat overlapping obligations. They must provide interconnection to their networks, access to unbundled network elements (such as transport, switching, loops, etc.), resale at wholesale rates of their services, and collocation of equipment necessary for interconnection or access to unbundled network elements at the premises of the ILEC (e.g., an ILEC central office).[26] In total, the Act’s requirements of the ILECs are the guts of a fourteen-point "checklist" of items that must be addressed before the RBOCs are allowed to compete in the in-region interLATA long distance markets.[27]
Thus, the Act required that the LECs and RBOCs provide access to a wide variety of de facto essential facilities. A problem with such a broad list of essential facilities is its effect on the magnum of proof required for RBOC entry into the in-region long distance market. The Act requires that the RBOCs meet all fourteen items of the checklist before RBOC entry into the in-region long distance markets will be allowed. However, from an economic perspective, in-region interLATA entry should be allowed once market power in the local exchange has been curbed,[28] and this may be the case after just some, but not all, of the fourteen checklist items have been met. The framers of the Act assumed that all fourteen checklist items were necessary before market power in the local exchange could be curbed, and they left the state public utility commissions and the FCC with no options if a less stringent subset of these checklist items was actually sufficient to curb RBOC market power. In this way, an expansive definition of essential facilities has served to forestall RBOC entry into the very markets in which the greatest consumer benefits could be realized: the in-region long distance markets.
If Congress had made the admittedly more nebulous concept of curtailment of RBOC market power in the local exchange the magnum of proof for RBOC long distance entry (in lieu of a detailed checklist), an expansive list of essential facilities could have been mitigated by state public utility commissions and FCC reviews of local exchange market power. It would have been possible to reasonably conclude that local exchange market power had been curbed even without compliance with all fourteen items of the checklist.
The situation we now observe is that all fourteen checklist items are assumed to be necessary to curbing RBOC market power in the local exchange, when in fact a much shorter list may be quite sufficient (and capable of curbing any RBOC market power that may exist in the absence of current regulation). Because there has been entry into local exchange markets without the RBOC compliance with all fourteen checklist items, this latter possibility cannot be discounted.
C. How Should Essential Facilities Be Defined?[29]
In economic terms, whether a so-called "essential facility" exists in a wholesale telecommunications market depends entirely on its effect on the competitive process in the adjacent retail markets. Thus, in arriving at a determination of the "essentiality" of an upstream productive input, the following questions need to be considered:
If the answer to this second question is yes, then according to the Larson-Weisman definition of "essentiality" of wholesale inputs (the economic efficiency criterion), there are at least four necessary (though not sufficient) criteria to hold true: (1) the absolute requirement that an entrant have physical access to the "essential" wholesale input to provide service at all; (2) a welfare-enhancing competitive process could not operate properly in the retail market unless efficient entrants have access to the wholesale input; (3) the "essential" wholesale input service is available only from a monopolist or consortium of firms acting as a monopolist, and no other source; and (4) prospective entrants can earn non-negative economic profits post-entry when paying the welfare-maximizing wholesale price for inputs.[30]
Unlike prior definitions of essential facilities appearing in the literature, all of which have assumed that the determination of "essentiality" and the pricing of facilities deemed "essential" (i.e., access pricing) were two separate matters that could be addressed sequentially, the Larson-Weisman definition merges the proposed pricing of an alleged essential facility with the ultimate determination of its essentiality.
In other words, the terms at which an alleged essential facility should be made available to prospective competitors in a downstream market become one of the determinants of essentiality. This is necessary as an incentive mechanism to ensure that entrants can improve the competitive process downstream and to screen inefficient entrants that cannot improve the public interest in the downstream market unless they are subsidized by the vertically integrated incumbent firm.[31]
D. Summary: How Efficient Pricing of Unbundled Network Elements Could Have Mitigated the Act’s Overly-Broad Definition of Essential Facilities
Nowhere does the Act refer to the concept of "essential facilities," though earlier versions of the draft legislation did so. However, the Act implicitly treats LEC local exchange services as de facto "essential facilities," in the antitrust sense. That is, the Act assumes that for true competition to take place, a regulatory agency must mandate open access to the LEC's local exchange services via resale, unbundling, or some other means, with no determination of whether the lack of open access actually impedes the competitive process in adjacent markets in the first place.[32] In other words, it performs no specific screening function to prevent inefficient interconnection, unbundling, or resale policies. To be fair, however, even if it did, it would most likely be difficult to implement such screens through adjudication, no matter how carefully the law was worded, and adjudicating essentiality would have added even more delays to the surprisingly protracted negotiations on terms of network access held by the state public utility commissions.
If the FCC and the state public utility commissions had sought to ensure economically efficient (i.e., welfare-maximizing) prices for unbundled network elements (in lieu of entrant-maximizing prices), the broad de facto definition of essential facilities implicit in the Act would have been largely moot. The efficient access price to unbundled network elements would have served as a mitigating screen. Though it would not have contracted the mandated set of de facto essential facilities or confined it to those necessary for fostering competition in local exchange markets, it would have removed the economic harm from having an overly broad definition of essential facilities (aside from unnecessary transaction costs).
To see this, note that the Larson-Weisman criteria for essentiality add a fourth criterion to three rather well known ones. This fourth condition (i.e., prospective entrants can earn non-negative economic profits post-entry when paying the economically efficient wholesale price for inputs) is an important screen, since the Act does not determine essentiality of facilities explicitly. In theory, this fourth condition ensures that prospective entrants can engage in welfare-increasing competition with the incumbent firm if all other impediments to entry are relaxed.[33]
For example, suppose the efficient price of access to unbundled network elements is the well-known efficient component-pricing rule, or ECPR (or an efficient variant of it).[34] In this case, if firms cannot pay the ECPR price, then the unbundled network element in question (from an economic perspective) is not essential to downstream competition.
Now consider two scenarios. In the first scenario, prospective entrants cannot pay the efficient access price and expect to earn positive profits. In this case no downstream entry takes place, consumers are no worse off, and the ILEC is not harmed financially (aside from transaction costs it may have borne in hearings, etc.).
In the second scenario, assume that prospective entrants will pay the efficient access price, for they expect to earn positive profits in the downstream market. If there is no change in the equilibrium downstream market price due to competition, then, again, neither ILEC nor entrant nor consumer is made any worse off (ignoring transaction costs). However, if the interim equilibrium downstream market price declines due to competition (a key attribute of the competition Congress sought), the ECPR-based access prices ratchet down as well, consumers are made better off, and total profits in both the upstream and the downstream markets ostensibly increase.[35] The lower access prices may attract yet more entrants, who also contribute to lower interim equilibrium downstream market prices, and the cycle can repeat itself until a new long run equilibrium price is reached in the downstream market.[36]
Thus, if the FCC and the state public utility commissions were willing to set prices of unbundled network elements at the efficient levels, the harm from an overly broad de facto definition of essential facilities would have been mitigated, but the potential for welfare-improving downstream competition would have been preserved. This would have required the state and federal regulatory agencies to encourage politically unpopular "high" prices for unbundled network elements, and it would have required an iterative approach to access prices, using a mechanism that ensured access prices would decline expediantly as downstream market prices declined with competition.[37]
In summary, the Act should have used a more narrow implicit definition of essential facilities, and the FCC and state public utility commissions should have sought to price these essential facilities at more efficient levels. However, given that the Act’s working definition of essential facilities is overly broad, using efficient access prices still could have mitigated ineffective policies.
The economically efficient wholesale price is an important determinant of whether a lack of access to unbundled network elements foreclosed welfare-improving downstream competition (i.e., by preventing the entry of firms that were capable of lowering total industry costs and increasing the consumer surplus in the downstream market). Because the efficient wholesale price serves this function, it is an economic brightline defining the point at which wholesale prices can screen inefficient or opportunistic entrants to a market. Thus, lacking a specific definition of "essential facilities" in the Act, the efficient pricing of unbundled network elements could have served as an important determinant of whether a given element is truly "essential" and could have reduced the inefficiencies of an overly broad definition of essential facilities, yet it would have done nothing to prevent efficient competition from emerging.
VI. Conclusion and Policy Recommendations
So just how did the Act get becalmed in the horse latitudes of ineffective policy? Doubtlessly, there are many hypotheses of political stripe and both strategic and economic natures.[38] This article has endeavored to pinpoint, from an economic perspective, the major design elements in the Act that prevented the benefits of competition from reaching consumers.
First, Congress misdiagnosed local exchange markets and passed the Act, which required the ILECs to make access to local exchange networks available to competitors (primarily through resale and the sale of unbundled network elements). Other methods of open access were also required by the Act, such as collocation in ILEC central offices.
There was no need to require such comprehensive open access policies so soon. Local exchange markets were not markets in which ILECs had foreclosed entry by other firms and had harmed consumers by requiring them to pay overly high prices. Prior to the Act, most states enforced legal proscriptions to local exchange market entry, and due to universal service objectives, local service prices were so low that few or no firms could enter the local exchange market profitably. Thus, it stood to reason that entry was not likely to be observed in local exchange markets. To get to the root of the situation, Congress could have set aside all state entry proscriptions (as the Act did), and via access reform and universal service reform, encouraged the state public utility commissions to allow residence local exchange service prices to rise to efficient levels.[39] If entry then occurred, the problem was solved. If it did not, then policies such as resale and the sale of unbundled network elements could have been considered, though in a more limited manner than as prescribed by the Act.
Instead, the Act plunged right in with an aggressive solution to the problem as perceived by Congress, open access policies, mandating access to a large number of de facto "essential facilities" (in the antitrust sense). The Act’s overly broad implicit definition of essential facilities caused a much lengthier process of negotiated hearings on unbundled network element prices than need be. These hearings could have been confined to a much smaller set of essential facilities, such as loops in rural or high-cost areas, phone numbers, and a few other selected network functions. For example, unbundled network elements such as transport and switching have been generally available for years, so there was no need for the Act or the FCC’s interpretation of it to vest these elements with the de facto status of essential facilities. The expansive definition of essential facilities created protracted hearings that ultimately delayed the opening up of competition in local exchange markets.
Simultaneously, resale prices were set by the state public utility commissions mandating deep-cut discounts (following the guidelines of the FCC). These deep discounts off of ILEC retail prices were designed to encourage competition; however, they created a serious problem with bringing the benefits of the Act to the American public. These resale rates were out of synch with the prices for unbundled network elements, giving many prospective entrants to local exchange markets a preference for resale over the purchase of network elements. These low resale rates interfered seriously with the magnum of proof the RBOCs had to meet to demonstrate that local markets were open to competition--and that the RBOCs should therefore be allowed into long distance markets.
Unfortunately, the FCC did not consider resale strong enough proof that local exchange markets were indeed open to competition--facilities-based competition was what the FCC required. Thus, the RBOCs could show that they lost a large number of local exchange lines to resellers but a far smaller number of lines lost to facilities-based carriers. As a result, when the RBOCs tried to get long distance relief under section 271 of the Act, they were not successful.
In addition, Congress assumed that all fourteen checklist items were necessary before market power in the local exchange could be curbed, and left the FCC with no options if a less stringent subset of these checklist items were actually sufficient to curb such market power. The Act’s expansive definition of essential facilities has served to forestall RBOC entry into the very markets in which the greatest consumer benefits could be realized: the in-region long distance markets. If Congress had made the curtailment of RBOC market power in the local exchange the magnum of proof for RBOC long distance entry (in lieu of a detailed checklist), an expansive list of essential facilities could have been mitigated by FCC review of local exchange market power. It would have been possible to conclude in an economically correct way that local exchange market power had been curbed even without compliance with all fourteen items of the checklist.
The result is what we observe today: much local exchange entry by resellers, a lesser amount of entry by facilities-based LSPs, little or no significant price competition in local exchange markets,[40] and no RBOC entry into long distance markets. Thus, over three years after the passage of the Act, consumers in local exchange and long distance markets are not significantly better off than before the Act was passed.

Footnotes

[*] Director, Corporate Demand Analysis, SBC Management Services, Inc., 175 E. Houston, Room 10-S-30, San Antonio, TX 78205, Work Phone: (210) 351-3931, FAX: (210) 351-5177, e-mail: alarson@corp.sbc.com. The opinions expressed in this paper are those of the authors and do not necessarily represent the opinions, policies, or business plans of SBC Communications, Inc. or any of its subsidiaries. The authors wish to thank Tom Makarewicz, Terry Schroepfer, Dennis Weisman, and an anonymous referee for reviews of earlier drafts of this article. Rhonda Gravemann and Gloria Lu served as research assistants to the project.

[**] Economist, Regulatory and External Affairs Department, Southwestern Bell Telephone Company, One Bell Center, Room 38-E-05, St. Louis, MO 63101-3099, Work Phone: (314) 235-7794, FAX: (314) 235-7815, e-mail: dm2992@rea1.sbc.com

[1] See Alexander C. Larson, Reforming Telecommunications Policy in Response to Entry into Local Exchange Markets, 18 HASTINGS COMM/ENT L.J. 1, 7-13 (1995), for a description of the early state efforts at promoting telecommunications competition. See also Alexander C. Larson, Overview, 15 PUBLIC UITLITIES LAW ANTHOLOGY xvii (July-Dec. 1992).

[2] GTE Corp. claims its residential local exchange rates are subsidized in the amount of $22.90 per access line per month from its other services. Residential exchange rates are subsidized about equally from interstate access revenues, intrastate access, and intraLATA toll. GTE Execs See Hope for Move to Universal Service Surcharge, 64 TELECOMM. REP. 35 (Nov. 16, 1998).

[3] Residual ratemaking is the setting of the "residually priced" service rates so as to yield closure to a revenue requirement, after the rates for all other services have been determined. Thus, for a given service priced residually, its price is set so as to cover the "residual" revenue requirement not recovered by all the other services whose prices have already been determined. Residual pricing is typically used as a means of setting basic local exchange rates at low levels so as to foster universal service.

[4] Federal regulators remain committed to achieving what they perceive as the critical goals of U. S. telecommunications policy, among which is to "maintain rates for basic residential service at affordable levels." As if to eliminate speculation regarding potential local telephone service price increases, the FCC further states: "We believe that the rates for this service are generally at affordable levels today." In re: Federal-State Joint Board on Universal Service, 12 F.C.C.R. 8776, ¶2 (released May 8, 1997).

[5] So far, even large, well-established IXCs have experienced losses in entering local service markets. For example, MCI indicated on July 10, 1997, that "it would likely sustain losses of $800 million this year from its entry into local phone services." However, it remains to be seen if such firms' local service market losses are not actually a normal cost of entry. John J. Keller & Gautam Naik, MCI, British Telecom Shares Tumble After Projection of Local-Market Losses, WALL ST. J., July 14, 1997 at A3. MCI President Timothy Price indicated: "Are we going to take losses in local? You bet we are. We took 'em in long distance and we'll take 'em in local to get in." Id. at A6.

[6] When entry does not take place because prices are low, then policies designed to foster entry through access to so-called essential facilities are misplaced. Having low terms of access to the essential facilities does not mitigate this situation.

[7] This has been estimated at about -.04 or less. See generally LESTER D. TAYLOR, TELECOMMUNICATIONS DEMAND IN THEORY AND PRACTICE (1994).

[8] Consumer surplus is an economic measure of consumer welfare defined as the difference between what a consumer is willing to pay for a given good, service, or commodity, minus what he must pay. See, e.g., DAVID L. KASERMAN & JOHN W. MAYO, GOVERNMENT AND BUSINESS: THE ECONOMICS OF ANTITRUST AND REGULATION 49-50 (1995).

[9] This applies to the ability of the Act to provide immediate benefits to consumers. It is not necessarily true in the long run if true competition takes hold in the local exchange market and the market elasticity becomes much larger than it is currently. However, this may take some time.

[10] 47 U.S.C. § 252(d)(3) (Supp. II 1996).

[11] See In re Implementation of the Local Competition Provisions in the Telecommunications Act of 1996, 11 F.C.C.R. 15499 (released Aug. 8, 1996) [hereinafter cited as Interconnection Order].

[12] Interconnection Order at ¶911.

[13] See Alexander C. Larson, Wholesale Pricing and the Telecommunications Act of 1996: Guidelines for Compliance with the Avoided Cost Rule, 8 U. FLA. J. L. & PUB. POL’Y 243 (1997) for a complete discussion of the avoided cost rule.

[14] To illustrate, suppose a government agency chooses to increase competition in the market for soft drinks, where existing firms both produce and retail their products. Assume that existing firms are very efficient producers but relatively inefficient marketers. A more efficient marketer could lower industry costs and make consumers better off through lower prices. If existing firms decide to produce the product and retail it, but also enlist resellers to assist in retail marketing, they will avoid some of the costs they now incur for retailing. Suppose the retail price of a can of soda is $1.00, but existing firms can avoid $.10 per can in retailing costs if they enlist resellers to help them. In this way, existing firms will offer the soft drink at a wholesale price computed as their retail price less the avoided costs of retail marketing, or $.90 a can. Resellers with retailing costs of less than $.10 per can will enter the market and make consumers better off with lower prices.

Now assume that these same existing firms can avoid $.30 a can in costs if they choose to cede all retail marketing functions to resellers. This is what the FCC’s definition of avoided costs has mandated: that existing firms behave as if they avoided $.30, even if they actually avoid only $.10 in costs. This opens the door for entry by inefficient firms that have marketing costs of more than $.10 but less than $.30 a can. Applying the FCC's logic to this example, existing soft drink suppliers would be expected to wholesale their soda at $.70 a can. This requires them to provide a subsidy of $.20 per can to resellers, which would make it difficult or impossible to compete with those same resellers.

[15] 47 U.S.C. § 271(c)(1)(A) (Supp. II 1996).

[16] Id.

[17] The Act as codified specifically states "Presence of a Facilities-Based Competitor." 47 U.S.C. § 271(c)(1)(A). The issues are over: (1) the proportion of its own facilities a carrier uses and (2) if leasing facilities in total, as a reseller, competition is not actually "facilities-based."

[18] See In re Application by SBC Communications Inc., Pursuant to Section 271 of the Communications Act of 1934, as amended, To Provide In-Region, InterLATA Services in Oklahoma, 12 F.C.C.R. 8685, ¶1 (released June 26, 1997) (citing the Act at § 271(c)(1)(A): "[W]e conclude that SBC has not demonstrated on this record that it is providing access and interconnection to an unaffiliated, facilities-based competing provider of telephone exchange service to residential and business subscribers, as required by . . . the statute.") (emphasis added).

[19] See 47 U.S.C. § 271(c)(1)(B).

[20] 47 U.S.C. § 271(c)(1)(A).

[21] See 47 U.S.C. § 271(c)(1)(B).

[22] The IXCs, by their lack of entry into local service markets, do not have the ability to bar RBOC entry into interLATA markets absolutely. The RBOCs, in their applications before the FCC for section 271 relief, can still argue that other LSPs have entered the local exchange market. However, the IXCs are major market participants who can devote significant resources to intervening in state arbitrations and to lobbying efforts. In addition, policymakers may give far more weight to IXC entry than entry by a firm providing only local service.

[23] That is, those capable of lowering prices or raising product quality levels.

[24] "No State or local statute or regulation, or other State or local legal requirement, may prohibit or have the effect of prohibiting the ability of any entity to provide any interstate or intrastate telecommunications service." 47 U.S.C. § 253(a) (Supp. II 1996).

[25] See 47 U.S.C. §§ 251(a)(1)-(b)(5) (Supp. II 1996).

[26] See 47 U.S.C. §§ 251(c)(1)-(6).

[27] See 47 U.S.C. §§ 271(c)(2)(B)(i)-(xiv).

[28] Current state telecommunications regulation already does this, if the low prices in evidence are any indication.

[29] Portions of section V(C) have been adapted from Alexander C. Larson & Dennis L. Weisman, The Economics of Access Pricing, Imputation and Essential Facilities, with Application to Telecommunications Policy, 3 COMM. L. & POL'Y 1, 14-18 (1998).

[30] The efficient wholesale price for inputs here is assumed to be the wholesale price that will maximize total surplus (the sum of producer and consumer surplus) in the downstream market, taking into account the impact that entry will have on downstream prices.

[31] While such an approach may improve consumer surplus in the downstream market, it will decrease total surplus (producer surplus plus consumer surplus).

[32] Similarly, the major interconnection cases of the last decade uniformly accepted local exchange networks as essential facilities, most without significant analysis of the point. See Southern Pacific Comm. Co. v. AT&T, 740 F.2d 980, 1008 (D.C. Cir. 1984), cert. denied, 470 U.S. 1005 (1985) ("by using its control over access to these essential facilities [local distribution network], AT&T had the ability to extend its natural monopoly power in the market for local public switched telephone service to the competitive market for intercity private line service." AT&T's refusal to interconnect excused by legitimate business justification based on regulatory policy); Litton Systems, Inc. v. AT&T, 700 F.2d 785, 811 (2d Cir. 1983), cert. denied, 464 U.S. 1073 (1984) (noting, in passing reference, that AT&T's control over local telephone network is "a textbook example of a monopolist in control of an essential facility."); United States v. AT&T, 524 F.Supp. 1336, 1353 (D.D.C. 1981) (Greene, J.) (denying AT&T's motion to dismiss: "[I]t is clear that the local facilities controlled by Bell are 'essential facilities' within the meaning of [the listed] decisions.").

[33] The logic behind this fourth criterion is as follows: if wholesale inputs are truly essential, then a denial of access to these inputs impedes the competitive process in the downstream market. However, if the denial of access to these inputs truly impedes the competitive process in the downstream market, then prospective entrants that can foster the competitive process downstream have one very important attribute: they can earn non-negative post-entry profits when paying the efficient wholesale price for the inputs. By definition, firms that are irrelevant to the competitive process downstream cannot do so.

[34] The ECPR holds that the price of an upstream productive input should equal its average-incremental cost, including all pertinent incremental opportunity costs. That is, the optimal wholesale input price is the input's direct per unit incremental cost plus the opportunity cost to the input supplier of the sale of a unit of input. See WILLIAM J. BAUMOL & J. GREGORY SIDAK, TOWARD COMPETITION IN LOCAL TELEPHONY 94 (1994). This access pricing rule has its roots in two papers: Robert D. Willig, The Theory of Network Access Pricing, in ISSUES IN PUBLIC UTILITY REGULATION 109-152 (Harry M. Trebing ed. 1979); and William J. Baumol, Some Subtle Pricing Issues in Railroad Regulation, 10 INT'L J. TRANSP. ECON. 341 (1983). The ECPR was popularized largely by the papers of Baumol and Sidak. See William J. Baumol & J. Gregory Sidak, The Pricing of Inputs Sold to Competitors, 11 YALE J. ON REG. 171 (1994); and WILLIAM J. BAUMOL & J. GREGORY SIDAK, TOWARD COMPETITION IN LOCAL TELEPHONY 93-116 (1994).

[35] In fact, if local exchange prices do not decline, then the entire exercise of fostering competition was pointless, and the initial access prices set at ECPR levels did nothing to foreclose efficient competition.

[36] This assumes Cournot competition takes place from entrants. Notably, the FCC did not believe ECPR-based terms of access could change after initially being set. See In re Implementation of the Local Competition Provisions in the Telecommunications Act of 1996, 11 F.C.C.R. 15499, ¶¶ 709-10 (released Aug. 8, 1996) ("We conclude that ECPR is an improper method for setting prices of interconnection and unbundled network elements because the existing retail prices that would be used to compute incremental opportunity costs under ECPR are not cost-based. Moreover, the ECPR does not provide any mechanism for moving prices towards competitive levels; it simply takes prices as given.").

[37] Admittedly, this is a difficult function for any administrative agency. However, this approach may have surpassed the results we now observe from the implementation of the Act.

[38] For example, the Justice Department’s Assistant Attorney General-Antitrust, Joel Klein, has hypothesized that the RBOCs used litigation to get into interLATA markets while attempting to do as little as possible to open their markets to competitors. See Klein: Bells’ Strategy at Fault for InterLATA Failings, 64 TELECOM. REP. 37, 47 (Nov. 23, 1998).

[39] As hard as that would have been, the way the Act was implemented was probably much harder.

[40] Seventeen months after the passage of the Act, there were no significant reductions in local phone bills due to competition. See John Greenwald, Hung Up on Competition, TIME, Jul. 21, 1997, at 50-51.