اشتراک سرمایه گذاری در شبکه های پهن باند

This paper presents a model of competition between an incumbent firm and an Other Licensed Operator (OLO) in the broadband market, where the incumbent has an investment option to build a Next Generation network (NGN) and it can do so by making an investment sharing agreement with the OLO, or alone. Two different kinds of investment sharing contractual forms are analyzed, a basic investment sharing, where no side-payment is given for the use of the NGN between co-investors, and joint-venture, where a side-payment is set by the co-investing firms. Results show that investment sharing can potentially be beneficial in terms of competition and investments, but the number of firms involved matters and so does the choice of the NGN access price, for insiders and outsiders of the agreement. Even when the presence of firms outside of the agreement force insiders to compete more fiercely, there might be a concern with the potential exclusion of the outsiders from the NGN.

The telecommunications markets currently face a period of widespread debate about the deployment of the so-called Next Generation Networks (NGNs). These networks represent a decisive progress in the telecommunications technology, due to their enhanced possibilities in offering faster transmission, and thereof services which demand more capacity and faster connectivity, such as interactive TV-centric and gaming broadband services, IP-based and high definition TV. The issue of NGN deployment acquires a status of social interest at European and national level, due to the recognized importance of telecommunications infrastructures for economic growth (Czernich et al., 2011, Koutroumpis, 2009 and Roller and Waverman 2001).1 The European Commission dedicated a special effort to the development of digital markets in the European arena, setting ambitious targets in the document “A Digital Agenda for Europe” (COM, 2010a and COM, 2010b).2 The issues related to the roll-out of new access fiber networks and the replacement of the existing copper networks regard mainly the high sunk costs for the construction and the uncertainty of returns due to market and regulatory risk. For these reasons, telecommunications operators are deterred from investing in such new technology. The European Commission and the national telecommunications authorities face new regulatory challenges with respect to the previous market scenario with only copper networks. The challenge is now not only to ensure that viable competition is working in the market, but also that conditions are such that investment incentives are stimulated. The classic trade-off between static efficiency and dynamic efficiency emerges. Much of the debate on supply side policies focuses on what kind of access regulation should be set for the NGN and the rules regarding co-investment agreements between firms (Bourreau et al., 2012a and Bourreau et al., 2012b). In fact, given the market and regulatory risks and the extensive investment requirements for the NGN roll-out, the opportunities of cooperative joint investments have recently become a prominent topic of discussion. Such co-investments are believed to possibly be a solution to the asymmetric risk allocation, which slows down the NGN roll-out, and to the financial constraints faced by firms. The discussion over co-investment agreements regards their effective superiority in terms of social outcomes and what rules for such agreements should be set to avoid potential anti-competitive consequences or, more generally, to maximize their social benefit. To this aim, a special attention must be devoted to the access conditions between partners: compensations mechanisms, exchange of information, non-discrimination clauses. It is also important to consider what is the number of players in the market compared to what is the number of co-investment partners.3 European national authorities are adjusting their regulatory frameworks to include provisions regarding such co-investment agreements between telecommunications operators to avoid inefficient investment duplications and at the same time potential anticompetitive consequences from cooperation between firms. One of the most sophisticated set of rules in this respect is represented by the French regulation. In France, in the “high-density areas”, the regulatory authorities have set the following procedure for the NGN roll-out: (i) the initiator should first identify which other market operators are interested in co-funding the NGN investment; (ii) if no other firms participated in the investment effort, the investing firm is forced to give access at “reasonable and non-discriminatory condition”; (iii) if at least one other firm participated in the investment effort, they are forced to give access to late entrants, but the access price should be inclusive of a “risk premium” (Arcep, 2009 and Bourreau et al., 2010). A thorough analysis of these relevant factors regarding co-investment agreements for the NGN roll-out is still missing in the economic literature on regulation and investment in telecommunications; this paper aims at filling this gap. Existing papers address the impact of access regulation on NGN investment in different perspectives (see for a survey, Cambini and Jiang, 2009).4Nitsche and Wiethaus (2011) compare different regulatory regimes regarding the effects on investment incentives, the competition intensity and the resulting consumer surplus in a two-stage Cournot model with two firms, a vertically integrated incumbent and an access seeking entrant, in a context with demand uncertainty. Risk sharing is one of the regimes they look at in their analysis, but it is considered in a reduced-form fashion, in which the two co-investing firms share the cost of the investment and then do not have to make any further side payment for the use of the NGN. However, they conclude that risk sharing can be particularly beneficial both in terms of investment incentives and consumer welfare. Cambini and Silvestri (2012) use a similar model to Nitsche and Wiethaus (2011), but analyze a dynamic framework with vertically differentiated firms. The paper extends the previous analysis and finds similar results regarding the potential benefits of risk sharing agreements. Nitsche and Wiethaus (2010) also wrote a White Paper that contains a discussion over possible extensions of their basic model, and an overview of the results when considering alternative approaches to risk sharing. In particular, they focus on different compensation mechanisms between the co-investing firms, each time considering separately: the presence of asymmetry between co-investing firms (market share asymmetry or risk commitment asymmetry); the presence of outsiders without access to the agreement; and the effect of changing the number of outsiders and insiders to the agreement on the final outcome of the model. Interestingly, Nitsche and Wiethaus (2010) also consider the potentially depressing effect on NGN investment incentives of a non-margin squeeze obligation, in case the investment turned out to be unsuccessful, while the other relevant papers assume no regulation at retail level. However, their discussion does not contain any analytical solution and the chance to have outsiders with access to the NGN is only discussed, because their model could not give an insightful numerical solution to the case. Among the vast literature on network investment and regulation in telecommunications, there are few papers addressing directly the effect of different forms of co-investment agreements. Inderst and Peitz (2012) analyze the role of different contract types and access regulation on innovation and competition in NGN investment. In their model, an incumbent and an alternative operator (other licensed operator—OLO hereafter) can possibly invest in building a NGN, cooperatively or on a stand-alone fashion. They show that access contracts signed after the investment deployment lead less often to the duplication of investment and a wider roll-out, compared to a market where it is not possible to sign access contracts. In comparison to such ex post contracts, contracts signed before the investment deployment lead to an even wider roll-out and to a less frequent duplication of investments. However, both types of contracts can be used to dampen competition. Bender (2011) examines a model inspired by Nitsche and Wiethaus (2011) but in a framework with horizontal product differentiation with price competition between an investing and an access seeking firm. In a context of uncertainty about the success of the NGN, he compares regulatory regimes with symmetric and asymmetric risk allocation, where the firms always have the opportunity to cooperate and jointly roll-out the NGN. Notably, he also analyses whether the firms are willing to cooperate in the investment deployment, as they cannot be forced to do so by regulatory authorities. However, he does not look into the different possible forms of compensation schemes for the use of NGN in the co-investment agreement, but rather, the risk sharing contract is modeled as a fixed transfer payment from the OLO to the incumbent. Kraemer and Vogelsang (2012) show the results of a laboratory experiment stating that co-investment increases investment with respect to access regulation but it also facilitates collusion. Finally, Bourreau et al. (2012b) analyze cooperative investment in a NGN and how it interacts with access obligations in presence of demand uncertainty. They show that co-investment only increases total coverage if service differentiation and/or cost savings from joint investment, in particular due to high uncertainty, are high. Mandated access reduces incentives for co-investment not only through lower returns but also by creating the option to ask for access instead. Voluntary access provision instead increases infrastructure coverage but reduces social welfare in local areas by softening competition if services are almost homogeneous. The issue of the co-investment agreements relates also to the R&D literature, although regulation is not a stake there. Lerner and Tirole (2004) and Choi (2010)5 consider a model where: (i) some firms form a research joint venture; (ii) the research joint venture offers licenses of its technology to the outsiders of the agreement. This literature analyses whether patent pools are welfare-enhancing, and hence, whether they should be authorized or not by competition authorities. The perspective though is ex post, i.e. innovations have already been done, and the question is whether pooling them is welfare beneficial. Also, usually the firms considered are not vertically integrated, as in the telecommunications markets, and they are not subject to regulation. This paper analyses two different approaches to co-investment compensation mechanisms between an incumbent and an alternative operator, and their effect on the final outcome, in terms of consumer welfare and investments: basic investment sharing, where the firms share the investment cost and do not pay each other any compensation for the use of the NGN; joint-venture, where the firms share the investment cost and then set an internal access charge for the use of the NGN that maximizes their joint profits.6 A multi-stage static model where firms are symmetric is adopted and it is also assumed – for simplicity and to complement existing studies by Nitsche and Wiethaus (2011) and Bourreau et al. (2012b)–that there is no uncertainty on the market success of NGN. Regulatory commitment is only partial: firms know that the regulator is able to commit to a certain regulatory regime, but not to a predetermined level of the access charge to the NGN. For this reason, if the internal access charge is regulated, the authority decides upon the level of the access charge after the investment is deployed. Though closer to the Nitsche and Wiethaus (2011)'s paper, this paper focuses on the implications of more complicated structures of the co-investment agreements, in terms of alternative compensation schemes for the use of the NGN. The results show that investment sharing is the socially preferable option in most cases. Basic investment sharing in particular seems to be beneficial as it ensures more competition than joint-venture and fairly high investment incentives. Joint-venture gives relatively higher investment incentives, but it carries more risks in terms of anticompetitive effects of the investment cooperation agreements. In particular, when all firms in the market participate to it, a joint-venture agreement can sensibly reduce competition downstream. When an outsider firm with right to ask access to the NGN is introduced in the final market, after the co-investment agreement has already been made and the investment deployed, the result shows that the risk of anticompetitive effects decreases, but there might be exclusion of the outsider from the NGN through high access price. If the regulator's objective is reaching a stage where all firms use the NGN, and so the copper network can be switched off, then a light regulation imposing no exclusion would be advisable. The remainder of the paper is organized as follows. Section 2 provides the setup of the model with only the two co-investing firms as a benchmark situation. Section 3 examines the model with the introduction of a firm outsider to the co-investment agreement and explains the main findings under various regulatory circumstances. Section 4 summarizes the paper and concludes.

This paper develops a model of competition between an incumbent firm and an OLO in the broadband market, where the incumbent has an investment option to build a NGN and it can do so by making a sharing agreement with the OLO, or alone. Different from previous theoretical research, this paper discusses two different kinds of sharing contractual forms—basic investment sharing, where no side-payment is given for the use of the NGN between co-investors, and joint-venture, where an insiders’ internal transfer is set by the co-investing firms—comparing them with the scenario in which the incumbent invests on a stand-alone basis. Then, the model is extended considering the entry of an outsider firm, with option to ask access to the NGN. The final purpose of the paper is to analyze how policy settings, particularly regarding network access rules, affect the firms’ investment choice. Consistent with the result in Cambini and Silvestri (2012), results show that the presence of a basic investment sharing agreement positively affects competition and it also gives fairly high investment incentives compared to the no risk sharing alternative, but also to the joint-venture case. For these reasons, the paper shows that basic investment sharing is the preferable option in most cases. As regards to the joint-venture case, when the agreement is made by all the firms present in the market, it is more likely to dampen competition downstream so much that, although the investment incentives are stronger, the results in terms of consumer surplus and social welfare tend to favor basic investment sharing. As a policy recommendation, drawn by the comparison between the case with two firms and the case with the addition of an outsider, it results that investment sharing can potentially be beneficial in terms of competition and investments, but the concerns shown by the authorities related to the inherent form of such agreements are not void. The number of firms involved matters and so does the choice of the NGN access price, for insiders and outsiders of the sharing agreement. Although eventually the regulators’ objective is having no more network duplication, it might not be an optimal strategy for a start to have all firms in the market involved in a sharing agreement, unless the insiders NGN usage transfer is constrained at zero, like in the basic investment sharing case. Even when the presence of firms outside of the agreement forces insiders to compete more fiercely, there might be a concern with the potential exclusion of outsiders from the ultra-fast broadband market. In this framework, NGN access regulation has a positive effect on competition, but it also largely reduces investment incentives. If the urgency of the regulator is to ensure that, once it is deployed, the highest possible number of firms can use the NGN, then this result says that ex-ante regulation is necessary to avoid foreclosure, even though it reduces investment incentives. On the contrary, if it is socially beneficial to have a fast and a wider deployment of an NGN infrastructure, then ex ante intervention should be at least partially relaxed. Many regulatory authorities have indeed introduced ex ante rules for NGN (such as obligations for granting access to dark fiber or local loop unbundling to fiber) but the economic conditions of wholesale services are much more debated, with several countries imposing above-cost access prices to both consider static and dynamic goals (Cullen International, 2012). The framework used in this paper abstracts from the presence of demand uncertainty in order to shed light on the contractual mechanisms of the investment sharing agreements, which has not yet been examined satisfactorily by the literature. One extension to this paper could surely integrate the context with uncertainty, making with any probability investment and welfare results lower in absolute values in all cases, as it is found in a companion paper to this one (Cambini & Silvestri, 2012). Indeed, uncertainty is probably one of the most relevant issue influencing investment incentives. Another interesting extension might be introducing an asymmetry between firms participating to the co-investment agreement with and without internal transfers. Finally, other alternative forms of sharing agreements could be analyzed, such as one in which the outsider can pay a share if the investment costs and be made part of the co-investment consortium even after the NGN deployment.