سیاست های دسترسی آزاد، مسولیت های تنظیم شده و تبعیض غیر قیمتی در ارتباطات راه دور

Open access policies in telecommunications, including interconnection and unbundling, are implemented by regulators in an effort to increase competition in the sector. Lack of cooperation from incumbents is pervasive, given their incentives to engage in non-price discrimination and the moral hazard resulting from the inability of regulators to monitor the contract. We build a relationship between the access price and non-price discrimination, neither assuming a pre-determined market strategic interdependence or a specific demand function format. When the access charge is liberalized, the incentive for non-price discrimination disappears. It may be optimal for the regulator to set a second-best regulated access price to avoid non-price discrimination.

Open access policies are designed to foster competition in the telecommunications sector, and require adequate incumbent input supply to work properly. This is the case with interconnection, unbundling, carrier-selection and pre-selection. European Directive 2002/19/EC sets out the provisions governing open access: “making available of facilities and/or services, to another undertaking, under defined conditions, on either an exclusive or non-exclusive basis, for the purpose of providing electronic communications services.” The stated purpose of the regulation (art. 1) is “to establish a regulatory framework, in accordance with internal market principles, for the relationships between suppliers of networks and services that will result in sustainable competition, interoperability of electronic communications services and consumer benefits.” Direct intuition and the experience in the sector suggest that it is difficult to persuade incumbents to voluntarily assist regulators in this objective. The intended input price the incumbent charges the entrant is often higher than the price considered reasonable by the regulator. Armstrong (2002) offers a theoretical approach to this point, demonstrating that in a unit demand model the access price chosen by the incumbent, with or without the possibility of bypass, does not differ from that the benevolent regulator would choose to foster entry efficiency. By contrast, in a competitive fringe model, Armstrong finds an unregulated access charge that is higher than the regulator’s. The conditions that determine the cost and quality of the entrant’s service are commonly deemed by regulators as inappropriate or discriminatory against entrants. The strong technical interface required between both networks (entrant’s and incumbent’s) raises the possibility of a deliberate deterioration in the entrant’s cost conditions and/or service quality. Noll and Owen (1995) state that this practice was explicit in the interconnection between MCI and AT&T in the long distance service segment in the 1970s. Reiffen and Ward (2002) offer a survey on the empirical evidence of discrimination by regulated firms since the AT&T case.1 Lack of voluntary agreements on interconnection and unbundling has been prevalent in Europe, as evidenced by the Implementation Reports of the European Commission, 2000, European Commission, 2001, European Commission, 2002, European Commission, 2003, European Commission, 2004, European Commission, 2005 and European Commission, 2006. Likewise, Hausman and Sidak (1999) point out with regard to unbundling in the US that “by the fall of 1996, entrants and Incumbent Local Exchange Companies (ILECs) were unable to reach any voluntary agreements on the pricing of resale and unbundled network elements. As a consequence, hundreds of arbitration proceedings began in the fall of 1996. In most cases, each arbitration was a one-on-one proceeding between a single entrant and the ILEC.” Although the battles over input charges for the different types of open access policies are usually difficult, the regulator has significant capacity to enforce, ex-post, negotiated or regulated input prices. However, this does not hold for what is referred to as “non-price discrimination,” which can assume very subtle forms, from direct incumbent action to simple omission. Incumbents may simply be less concerned with receiving off-net call failures than on-net call failures. Yet, this may also be the result of the poor internal resources incumbents devote to entrant interconnection points. Further, it may be difficult for the regulator to verify whether delays in repairing interconnection points are due to a lack of incumbent goodwill. As Crandall (2002, p. 27) observes, Competitive Local Exchange Carriers (CLECs) have accused ILECs of failing to implement services on the agreed dates, of call drops and of poor call routing. According to EC reports, progress in unbundling across Europe is basically constrained by issues of technical complexity, requiring the direct involvement of regulators. Borreau and Dogan (2004) enumerate a number of complaints from entrants in Europe concerning quality degradation. In Germany and Denmark, entrants have accused incumbents of providing unbundled lines that are of poor quality or non-functioning. In short, quality issues can arise before or after a given line is unbundled.2 The general pattern of unbundling worldwide has consisted of implementation delays after enactment of the policy. Borreau and Dogan (2004) report that UK regulators required British Telecom to compensate entrants based on their estimated losses any time the company failed to meet contractual obligations, particularly with regard to line delivery. In summary, the regulator encounters significant difficulties in verifying claims of non-price discrimination against incumbents, requiring increased regulatory intervention. There is ongoing discussion among incumbents and entrants as to the suitability of regulated wholesale price levels. Not surprisingly, entrants, claim they are too high for their survival and market expansion, regardless of the actual regulated price, while incumbents claim they are too low to recoup their investments.3 Ultimately, the incumbent’s incentives to implement a non-price discrimination strategy are intimately bound to the regulated access price level. Beard et al., 2001, Bustos and Galetovic, 2003, Economides, 1998, Sibley and Weisman, 1998 and Mandy, 2000,4Mandy and Sappington (2007) – cost and demand sabotage, Weisman and Kang, 2001, Mattos, 2002, Mattos, 2007, Laffont and Tirole, 2000 and Sappington and Weisman, 2005 – self sabotage – and Sand (2004) explore non-price discrimination in vertically integrated companies providing inputs to downstream competitors. In general, the authors restrict their analyses to specific hypotheses about market strategic interdependence (Cournot or Bertrand) and demand format (mainly linear demand). Most of the research associates non-price discrimination with the low profitability of the access business, which results from the LRIC (Long Run Incremental Cost) price regulation5 methodology. A common feature of most of these models is that access price regulation constitutes a principal cause of non-price discrimination. These models involve a basic trade-off: non-price discrimination helps the incumbent’s retail business, but harms its access business. Our purpose in this paper is to assess the relationship between regulated access pricing, whether in interconnection, unbundling, carrier-selection or pre-selection, and non-price discrimination, neither assuming a pre-determined market strategic interdependence or a specific demand function. In addition, we do not consider sabotage costs, efficiency differential and the double marginalization problem6 contemplated in other non-price discrimination models. In this sense, our objective is to develop a pure demand model of non-price discrimination. Section 2 examines the basic non-price discrimination model and addresses the ILEC’s incentives to raise rival costs. Section 3 describes a game between the regulator and the ILEC aimed at deriving the optimal behavior of these agents in a sub-game perfect equilibrium with open access policies (OAP). In Section 4, we offer some concluding comments.

The paper offers some key messages. When a benevolent regulator maximizes its welfare function and accepts the fact that it is unable to monitor the behavior of the ILEC, non-price discrimination is never an SPE equilibrium. Moreover, it is possible that in order to induce the ILEC to play “Does Not Discriminate” the regulator may at times be forced to increase the ILEC’s rents to a certain minimum level. However, depending on the parameters of the problem this may not be required, since it may still be to the ILEC’s advantage to provide open access to the CLEC, even at t = treg, given the profitability of the access business. There are two possibilities when the ILEC’s incentives for non-price discrimination are incompatible with the first-best10 access charge given in (3). First, the regulator may have to give up rents to the ILEC by raising the access charge to a second-best optimum, provided the ILEC is willing to cooperate with the entrant and the regulator in the OAP. This is the finding emerging from the majority of studies cited in the introduction. As long as the regulator allows higher profitability in the access business, through a higher t closer to the profit maximizing access charge, the opportunity cost of non-price discrimination for the ILEC rises to a point where the expected gains in its retail business do not compensate for the losses incurred from reducing the access sought by the CLEC. Second, in anticipating that a non-verifiable strategy of non-price discrimination is inevitable at any value of the access charge, t, the regulator may opt against enforcing the OAP. The benevolent regulator is able to achieve its first best regulated access charge only under certain specific conditions linked to the relative effects of c, and thus p, on the ILEC ‘s retail (X) and access (x) businesses. The puzzle here is why non-price discrimination occurs in the real world. One possibility is that the actual behavior of regulators is not benevolent. Their insistence on TELRIC hypothetical ideally efficient cost rules to price access, as pointed out by Laffont and Tirole (2000), is clearly an explanatory factor. Why the regulator does not accept that its capacity to monitor non-price discrimination is considerably limited is a subject warranting study. In view of the asymmetric information of the public in regard to the regulator, the latter may prefer to give the appearance that it is technically capable of enforcing the OAP, in order to derive the political dividends of such a strategy. Given that non-price discrimination seems to occur in the real world, the regulator does not maximize a (benevolent) welfare function by insisting on the OAP. Therefore, a normative analysis might serve to more accurately reflect its true behavior. Our model departs from the literature by not imposing any specific oligopoly market interdependence or restricting the shape of the demand function. The degree of substitutability between the networks implicit in the first and second derivatives of X and x with respect to p are the key variables for assessing the ILEC’s incentive to discriminate. The benevolent regulator may be obliged in certain cases to raise the access charge from its first-best level (3) to address the moral hazard problem. The ILEC’s rents are increased up to its participation constraint so as to secure the required cooperation with the OAP. It is always the case that if the regulation of access charges is considerably lax or absent altogether (t = tunreg), the incentive for non-price discrimination disappears completely, a finding that is in line with the current literature. Finally, there are some extensions worth mentioning that could enrich this simple model. We could add the CLECs to the game by (i) offering the ILEC an open access contract, even when the regulator prefers to not implement the OAP (16 does not hold) or (ii) enabling a facility-based entry. It is possible that in certain cases the ILEC and the CLEC would be willing to sign a contract for a particular access price the regulator would nonetheless choose to block. This could occur when the regulator deemed the facility-based entry preferable to the OAP based entry insofar as it implies a more robust competitive regime.