Geographic access markets and investments

https://doi.org/10.1016/j.infoecopol.2015.04.003Get rights and content

Highlights

  • We study the adoption of access regimes based on the degree of infrastructure competition.

  • We compare differentiated access prices with a uniform price regime.

  • Geographically differentiated access prices improve welfare and incentivize investment.

  • When access in areas with infrastructure competition is deregulated, welfare may decrease.

  • Deregulation in access leads to multiple inefficient equilibria on investment level.

Abstract

We analyze the adoption of access regimes that differ according to the prevailing degree of infrastructure competition in different geographical areas of a country. Our results show that, compared to a uniform access price, geographically differentiated access prices improve welfare and incentivize investment. However, when access provision in areas with infrastructure competition is deregulated, welfare might decrease, because multiple inefficient equilibria at the wholesale level emerge, with either too little or too much investment.

Introduction

A typical feature of network industries is that, due to high investment costs, few players roll out infrastructures. Sectoral regulators then attempt to create competition at the retail level by imposing access obligations to existing networks. The traditional approach, at least in the telecommunications sector, has been to define “uniform access obligations”, where the same obligation to provide access and the same price of access apply nationwide.

Recently, some policymakers have argued that regulation should instead be based on local competitive conditions, such as the number of competing networks. The rationale behind this proposal is that most countries are composed of heterogeneous areas: Installing a network for a given number of consumers tends to be much cheaper in densely populated areas, such as urban centres, than in more sparsely populated ones, such as rural areas. As a result, multiple networks can emerge in urban areas, while at most one network is built in less populated ones. Thus, there is potential for competition at both the retail and wholesale level in urban areas, while without access regulation there would be none in rural areas. The question that we wish to address in this paper is how regulators should account for geographical differences in competitive conditions when they regulate access to networks. To the best of our knowledge, this issue has not been analyzed so far in the academic literature. The aim of this paper is to start filling this gap.

Our paper is inspired by recent decisions of national regulators and the European Commission that forcefully push for the adoption of geographical access rules in broadband access markets. For example, the 2009/140/EC Directive (“Better Regulation Directive”) explicitly considers the possibility of defining different geographical markets. It then invites national regulatory authorities to examine the differences in the degree of infrastructure competition across geographical areas, in order to determine whether the imposition of differentiated access obligations or even the lifting of obligations in specific areas are warranted.1

An important side-effect of access regulation is that it tends to reduce the incentives for investment in new infrastructures. This is of particular concern in markets such as broadband access, where legacy (copper) networks are considered outdated and where investment in (fiber) networks with much higher capacity is deemed necessary. Geographically-differentiated regulation has the potential to improve the trade-off between more intense retail competition and more infrastructure investment. We analyze whether this potential can be realized, and how it depends on the type of geographical regulation that is adopted.

We compare two different access regimes. First, in the “duplication-based” regulatory regime, the regulator maintains an access obligation everywhere, but sets different access prices in the areas with a single infrastructure and in the areas with multiple competing infrastructures. Setting two different access prices increases welfare, compared to a uniform access price regime. We discuss the relevant trade-offs, and show that the optimal access charge is higher where only one infrastructure is present than where multiple infrastructures exist. The intuitions are twofold: First, a higher access charge in areas with only one infrastructure provides stronger incentives to cover more of the most expensive marginal areas. Second, when access provision is an alternative to infrastructure duplication, the latter is optimal only in the most densely populated areas. A low access charge in competitive areas then guarantees that incentives for infrastructure duplication are weak enough.

Second, in the “competition-based” regulatory regime, the regulator sets the access price in the areas where only one infrastructure is present, but “leaves it to the market” everywhere else. The idea is that in the presence of multiple infrastructures, competition at the wholesale level arises. Contrary to what one might expect, though, we find that market outcomes turn out to be neither easily predictable nor efficient. First, the wholesale game between access providers has a natural tendency towards multiple equilibria. Second, none of the resulting equilibria is efficient if investment incentives are factored in: Either wholesale competition results in very low access prices which destroy investment incentives; or wholesale competition does not take off, and access and retail prices remain high. Thus, partial deregulation of access tends to lead to less certainty about market outcomes and lower welfare. Our general conclusion is that regulators should weigh very carefully the advantages and disadvantages of either regulatory regime.

Our paper draws from three different strands of literature. The first one studies the interaction between access regulation and investment, the second one analyzes competition at the wholesale level between vertically integrated firms, and the third one deals with universal service obligations (USOs), uniform pricing constraints and geographical service coverage.

The first strand of literature analyzes the impact of access regulation on firms’ investments.2 Some papers study the incumbent’s investment incentives (e.g., Faulhaber and Hogendorn, 2000, Foros, 2004, Nitsche and Wiethaus, 2011) or the alternative operators’ (e.g., Bourreau and Doğan, 2006) as a function of the access regime. In a companion paper, Bourreau et al. (2013), we study the relation between access regulation and incentives for cooperation in network investment, using a network coverage model similar to the present paper. However, none of these papers – including Bourreau et al. (2013) – looks at the introduction of geographically differentiated access rules, which is the key topic of this paper.

Our paper is also related to the literature on wholesale competition between vertically-integrated firms (Ordover and Shaffer, 2007, Brito and Pereira, 2010, Bourreau et al., 2011). These authors study whether a competitive wholesale market can emerge, or on the contrary, whether there is partial or complete foreclosure in equilibrium. We contribute to this literature by studying the additional issues of geographic access regulation and investment. We also introduce a specific regulatory tool – the dispute resolution mechanism – to avoid foreclosure, and show that it generates additional equilibria in the wholesale market.3

The third stream of literature concerns uniform pricing constraints and their impact on infrastructure coverage and market competition (Anton et al., 1998, Valletti et al., 2002, Foros and Kind, 2003, Hoernig, 2006). These papers focus on uniform pricing constraints at the retail level. However, they do not address the geographical differentiation of wholesale regulation, nor do they consider investment.

The rest of the paper is organized as follows. In Section 2 we present the model setup. In Section 3 we analyze geographically differentiated access prices, comparing them to uniform access prices. In Section 4 we study competition-based access prices. Section 5 concludes the paper. Longer proofs can be found in Appendix A.

Section snippets

Model setup

We consider a country represented by a line of locations z0,z, each with identical demand. There are two incumbent operators (firms 1 and 2), who can invest in coverage of new network infrastructures, and an entrant (firm E), who does not invest.4 We assume that the fixed cost cz of covering a location z is differentiable and increasing in z, i.e.,

Duplication-based regulation

In this section we assume that the regulator anticipates that there will be locations with different degrees of infrastructure competition, and that he can implement what we call duplication-based regulation,that is, a regime with two different access prices, ã for single and a for duplicate infrastructure areas.

Competition-based regulation

An alternative regulatory regime proposed by some regulators (e.g., see Ofcom, 2007) is what we call competition-based regulation. In this regime, the regulator sets the access charge in locations with a single infrastructure (i.e., in the SIA), but does not regulate the wholesale market in the locations with infrastructure competition (i.e., in the DIA). In the DIA incumbents can therefore set the access charge to their networks on a commercial basis. The implementation of competition-based

Conclusion

One of the most hotly debated issues about the EU regulatory framework for electronic communications is the introduction of geographical regulation, that is, wholesale access rules that vary according to the degree of infrastructure competition in local markets. In this paper, we develop a model of geographical regulation that allows us to determine the impact of the differentiation of access rules on welfare and firms’ incentives to invest in new infrastructures.

We compare two alternative

References (23)

  • M. Bourreau et al.

    Upstream competition between vertically integrated firms

    J. Ind. Econ.

    (2011)
  • Cited by (14)

    • Rolling out of fibre optic networks in intermediate versus urban areas: An exploratory spatial analysis in the Netherlands

      2021, Telecommunications Policy
      Citation Excerpt :

      However, empirical studies have rarely focused on a more detailed analysis of the spatial factors driving the provision of very high capacity fibre-based networks. From the literature on very high capacity fibre-based networks (Bourreau et al., 2015; Bourreau & Cambini, 2012), we derive a model that examines the geographical development of these networks in greater detail. The paper utilizes basic data mining techniques in conjunction with exploratory spatial data analysis to examine the provision of these networks for about 7.8 million households in the Netherlands.

    • Ultra-fast broadband investment and adoption: A survey

      2019, Telecommunications Policy
      Citation Excerpt :

      From a regulatory point of view, this calls for a transition from country-wide uniform measures to a more locally tailored regulation, with different ex-ante access rules across areas depending on the degree of infrastructure competition. Bourreau, Cambini, and Hoernig (2015) focus on the implementation of geographically-differentiated regulation imposed on fiber infrastructure as a potential regulatory tool to improve the trade-off between more intense retail competition and more infrastructure investment. The authors analyze whether this potential can be realized, and how it depends on the type of geographical regulation that is implemented.

    • Strategic investments under competition for access provision

      2018, Telecommunications Policy
      Citation Excerpt :

      They discuss the impacts of several regulatory tools such as a wholesale price cap and vertical separation. Bourreau, Cambini, and Hoernig (2015) compare the effect of uniform access pricing on investment in rolling out infrastructure with that of deregulating access price in a geographical setting. This paper proposes a different model that relates to the second approach by shedding light on the following three aspects.

    • Access regulation and the entrant's mode of entry under multi-product competition in telecoms

      2016, Information Economics and Policy
      Citation Excerpt :

      Such coexistence stems from the deployment of fiber access infrastructures in some geographic areas, which implies that the copper access networks are still in operation in the remaining geographic areas of a given country. In this framework, the related literature studies the relationship between the firms’ incentives to extend their fiber deployment to more geographic areas and the access price of: (i) copper unbundling (Bourreau et al., 2012a); fiber unbundling (Bourreau et al., 2015; Lestage and Flacher, 2010); both copper and fiber unbundling (Bourreau et al., 2014a). Although the deployment of the fiber access networks in a given geographic market may allow the non-investing firm to choose between using the old and the new technology in that market (Brito et al., 2012; Inderst and Peitz, 2012b), the above-cited literature assumes that when a firm invests in better quality (e.g., fiber) access networks, it no longer provides the lower quality (e.g., basic broadband) services.

    • Exclusion and regulatory intervention in investment sharing agreements

      2016, Telecommunications Policy
      Citation Excerpt :

      Hence, seems that this extension simultaneously combines both theories at a glance. It should be then clear the relevance of the following mathematical formalization that, until now, has not yet been captured by previous theoretical contributions applied into the telecommunications sector, namely, neither by Foros (2004), Cambini and Silvestri (2013) nor Bourreau et al. (2015).30 Finally, the topic regulation of tight oligopolies is, currently, a major concern in Europe as expressed in BEREC (2015).

    View all citing articles on Scopus

    We would like to thank Giacomo Calzolari, Pedro Pereira, Patrick Rey, Yossi Spiegel, Tommaso Valletti, Thibaud Vergé and Ingo Vogelsang, as well as participants at various conferences and workshops, for their useful remarks and comments. Marc Bourreau acknowledges financial support from Orange. Steffen Hoernig acknowledges funding from the PTDC/EGEECO/100696/2008 grant of the Ministry of Science and Technology.

    View full text