Towards a vertically separated broadband infrastructure: The potential role of voluntary separation

Abstract

Public policymakers face the challenge of maintaining incentives for investments in fiber-to-the-premises (FTTP) infrastructure while simultaneously enabling effective competition. In that regard, the ability of vertically integrated operators to discriminate against downstream rivals is regarded as a major regulatory problem, especially as policing nondiscrimination obligations is notoriously difficult for vertically integrated operators. The European regulatory framework mandates functional separation to eliminate discrimination incentives. However, this measure comes with significant drawbacks and caveats and resembles a strong market and ownership intervention. The present article discusses how voluntary vertical separation can achieve the positive effect of a vertically separated industry structure without the need for strong market interventions. We use a discounted cash flow model to compare the financial attractiveness of wholesale-only and integrated business models for a greenfield FTTP rollout. We further discuss and test how public policymakers could positively affect the profitability of a wholesale-only business model. Based on our findings, we recommend public-policymakers and national regulation authorities to proactively define separation scenarios and respective ex ante regulation reductions precisely, to make adjustments to broadband state aid programs that favor wholesale-only providers, and to critically reflect the hampering effect volume and time discount wholesale tariff structures of incumbents have on the emergence of wholesale-only networks.

Keywords

Wholesale-only vertical disintegration voluntary separation broadband

Introduction

The social benefits of broadband infrastructure are hardly disputed within academia and politics. Broadband infrastructure is broadly seen as a key enabling technology in today’s knowledge economy. Users can access information and entertainment services online and communicate with each other at lower costs than ever before, and businesses can enhance their efficiency and innovate faster and more easily. Numerous studies find empirical evidence that broadband infrastructure has a positive causal effect on economic growth, employment, and regional development, as well as on firms’ productivity and performance (see Bertschek et al., 2016, for an overview). There is also evidence suggesting a stronger effect for broadband infrastructures providing higher bandwidths (Abrardi & Cambini, 2019; Hasbi, 2017; Lee & Brown, 2008).

Against this background, it is unsurprising that there is a plethora of debates within politics, academia, and the industry on how to expedite and expand the geographical scope of investments in broadband infrastructure. Among other already widely adopted measures, such as state aid, other less prevalent ideas are evaluated concurrently. These do not necessarily depict direct measures, but rather aim to facilitate roll-out models that are seen as beneficial for successful broadband rollouts.¹ This article regards a vertical separation of wholesale and retail activities in the fixed broadband industry. Such structural separation, under which a firm either owns the network or retails broadband services to customers, opposes the current prevalent industry structure in Europe, mainly comprising of vertically integrated firms. This article presents the merits and caveats of a vertical disintegration alongside the existing literature and answers the question, how public policymakers could facilitate the emergence of voluntary separation while refraining from direct market interventions such as mandatory separation. Starting from the premise that market participants would increasingly adopt a wholesale-only model if it is at least as attractive as the integrated model or even superior, we test if, and under which circumstances this could be the case. To do so, we first model the investment decisions of integrated and wholesale-only operators using a discounted cash flow (DCF) model. As a second step, we test sensitivities to identify and potential levers to increase the financial attractiveness of the wholesale-only model. Lastly, we discuss what measures can be taken by public policymakers to affect the levers identified. Our article builds on broad research exploring the impact of vertical separation but is unique to the extend that firm-level investment decisions are modeled² to draw conclusions for the regulatory and public policy domain. This approach allows to take the current market structure into account to yield actionable conclusions.

This contribution is structured as follows: We first provide a brief literature review to provide context. Accordingly, the second section presents the industry and regulatory background, while the third section reviews the relevant literature, the fourth section presents the DCF model and the sensitivity analysis, and the final section concludes by discussing regulatory implications.

Industry and regulatory background

In the light of its aforementioned relevance, access to and adoption of high-speed broadband has become a top political priority in many countries. Nevertheless, incentivizing broadband investments is not the only objective public policymakers pursue. Most notably, national regulation authorities (NRAs) also strive to enable effective competition to optimize static efficiency. Achieving both aims, incentivizing investments while enabling competitive markets, is a challenging task. This is especially true for nonurban regions, where investment costs per household rise—in many instances, above the level at which more than one infrastructure is economically viable (Feijóo et al., 2018). In such regions, natural monopoly characteristics are exhibited by such broadband networks. To avoid consumer harm, the current regulatory regime in Europe mandates access to the infrastructure of so-called significant market power (SMP) operators.³ Mandated access allows downstream competitors to be active on the retail market without their own (access) infrastructure and therefore enables service competition. This separates the market vertically into wholesale (infrastructure) and retail (service) segments. Essentially, regulators seek to mandate access to allow for competitive retail markets, especially in regions in which multiple broadband access networks are not sustainable.

The regulatory approach applied today comes with significant drawbacks. Firstly, SMP operators have incentives to discriminate against retail competitors in favor of their own retail arms (Cadman, 2010). This can take the form of price and non-price discrimination (de Bijl, 2005; Laffont & Tirole, 2001). Supplying its own retail arm with cheaper and higher-quality wholesale products allows an SMP to capture additional profits on the retail market. Regulators address these concerns by defining wholesale product prices and quality parameters, usually in the form of price caps, nondiscrimination requirements, or provisions for equivalence of input (Cadman, 2019).

Such extensive ex ante regulation, however, comes with a number of caveats, as regulatory decisions are oftentimes drawn out, and there is a danger of status quo inertia, unjustified intervention, and a reduction of freedom to compete and innovate (Vogelsang, 2017). Moreover, the SMP designation requires thorough market reviews that come with a need for extensive data and qualified personnel to cope with analyzing complex market environments. While this task is already challenging for NRAs on a national level, the markets in most European countries tend to have heterogenous operators and thus competitive landscapes throughout the country.⁴ However, market reviews with regional market definitions or different remedies by region are the exception rather than the rule to date.⁵ Conducting extensive market reviews on a more granular level, for example, regional, or state-by-state in federal countries, would require even more NRA resources and more data, which is often not available. Moreover, such granular market reviews would need to be reviewed quickly and often, as market power can be subject to sudden changes due to market dynamics.⁶ Against this background, it is not surprising that the SMP designation in European telecommunication markets is so far limited to former state monopolies (incumbents) and cable operators with near-national coverage (Netherlands and Belgium). Consequently, local or regional networks of alternative operators (altnets) that have built out their own access networks in sparsely populated areas operate a nonduplicable infrastructure in the absence of any access remedies. This is problematic, as it enables these altnets to foreclose the retail market, or set prices above those a competitive market would bear, potentially harming consumers.

To summarize, the current SMP regime suffers from downstream discrimination incentives, high regulatory oversight costs due to extensive ex ante regulations, and potential unregulated local monopolies of non-SMP-designated altnets. These caveats could potentially harm consumers and detriment static efficiencies. To address this, the current regulatory toolbox encompasses a variety of countermeasures, including cost-oriented wholesale price caps, regulatory oversight, and open-access obligation in broadband state aid programs (see Bauer, 2010). The new European Electronic Communications Code (EECC)⁷ further extends those by allowing for ex post measures, symmetric regulation, and other measures. Public policymakers and NRAs now have the task of incorporating the new EECC recast into national law and need to choose instruments that preserve investment incentives while enabling effective competition. This article deals with the potential contribution that a vertical separation (also referred to as a wholesale–retail split) could play in this pursuit.

A vertically separated industry structure, with a subset of operators exclusively active in the wholesale segment and another subset exclusively active in the retail segment, would eradicate downstream discrimination incentives, relax the need for extensive ex ante regulation and supersede SMP designation for altnets owning non-replicable broadband networks. While the benefits are apparent and widely acknowledged (European Commission, 2010), the current market structure is highly vertically integrated, with just a few exceptions within Europe of a significant scale (with Openreach the most prominent example). Generally, this separation can be a product of either mandated vertical separation, voluntary separation, or new entrants opting for a wholesale-only business model. The first option has been widely discussed in the literature, including theoretical research and in-depth case studies (Cadman, 2019; Mancuso, 2012; Nucciarelli et al., 2010; Ovington et al., 2017; Teppayayon & Bohlin, 2010). Mandated separation was already considered as a remedy during the liberalization phase around the 2000s in Europe. It has been incorporated in the European regulatory framework since 2007 when the European Commission (EC) allowed for functional separation to be mandated in cases “where all other regulatory tools have proved inadequate to address market and competition failures” (European Commission, 2007). Under functional separation, the regulated firm is required to separate its retail and wholesale functions, employees, and information; while it may maintain common shareholding, it may not maintain ultimate ownership.

Mandatory separation has received increased attention in the political debate as related measures have been adopted in a handful of countries (see Crandall et al., 2008, for a summary of cases in the United Kingdom, Italy, Sweden, New Zealand, and Australia), as well as in the context of the new EECC. The new EECC does not expand mandatory separation provisions to forms beyond functional separation (see Article 77, new EECC⁸). Instead, Article 78 addresses the case of voluntary separation of an SMP-designated operator and lays out a process for NRAs to analyze whether any obligations induced by regulation should be imposed, maintained, amended, or withdrawn after the separation took place. Moreover, NRAs are put in a position to commit to binding agreements related to voluntary separation of an SMP-designated operator (see new EECC⁹) to increase transparency and predictability. Article 80 further specifies under which circumstances wholesale-only providers designated SMPs could benefit from ex ante reductions in regulation.

However, mandatory separation is perceived as a strong market and property rights intervention and is widely criticized as ineffective and potentially detrimental to the telecommunications sector performance (see Gonçalves & Nascimento, 2010). Thus, it should be seen as a measure of last resort that should only be considered when other measures fail to achieve regulatory goals. Voluntary separation constitutes another option to attain an increasingly vertically separated industry structure. While voluntary separation, also referred to as a wholesale-only business model, has been a rare exception in the decade following liberalization, it has seen rising practical relevance in the past few years. A rising number of new entrants, publicly funded entities, and private operators voluntarily chose to adopt a wholesale-only business model without retail operations (see Table 1).

Table 1.

Examples of voluntary vertically separated network operators in Europe.

Firm	Country	Company type
CETIN	Czech Republic	Voluntary spin-off of O2 Czech (owned by PPF)
Reggefiber	Netherlands	New entrant financed by financial institution
NöGiG	Austria	New entrant—privately and publicly funded
RuNe	Slovenia, Italy, Croatia	Publicly funded initiative
Siro	Ireland	Joint venture between fixed operators
Stokab	Sweden	Utility
Open Fiber	Italy	Telecommunications arm of energy provider ENEL
City Fibre	UK	Privately owned by financial consortium

Such voluntary separation requires a deliberate decision by private companies, which is not directly influenced by governmental decisions. Nevertheless, these private business strategy decisions take place in a highly regulated market environment. Consequently, public policymakers do have the ability to indirectly influence private players’ market decisions (Huigen & Cave, 2008; Peitz & Cave, 2008). This situation represents an opportunity to strengthen the aforementioned encouraging trend seen in a growing number of wholesale-only providers throughout Europe. Accordingly, it seems the right time to systematically assess (regulatory) measures that could further increase incentives for operators to adopt a wholesale-only business model while preserving incentives to invest.

Literature review

A broad strand of the literature has evolved that explores how various factors impact market players’ willingness to invest in broadband infrastructure. In this regard, the impacts of regulation, competition, macroeconomic factors, and vertical integration have been subject to extensive research (see Abrardi & Cambini, 2019; Cambini & Jiang, 2009, for an overview). This article contributes to the strand dealing with a vertical separation, also referred to as a wholesale–retail split. Research on such vertical separation can be divided into two parts. Firstly, a large number of papers scrutinize the regulatory option to mandate separation on vertically integrated incumbents, which has been discussed intensively in academia and politics in the past and has been implemented in the United Kingdom and Sweden.¹⁰ Secondly, a much smaller number of papers deal with voluntary vertical separation or a wholesale-only business model.

Mandatory separation

Mandatory separation has been a vibrantly discussed topic since the liberalization of telecommunication markets and continues to be the subject of extensive research to date. Although our article is primarily focused on voluntary disintegration, we also survey the literature on mandatory separation because those findings relate strongly to our research. The academic literature on vertical integration mainly concerns the efficiencies that come along with vertical integration and the effect of separation on investment incentives. The following section presents these arguments and discusses the extent to which the findings apply to the present market environment and the case of voluntary disintegration. Moreover, the research on voluntary wholesale-only networks is surveyed.

The regulatory option to mandate functional separation was integrated into the European regulatory framework as a last-resort remedy in 2002 by the European Parliament & Council (2002). Since then, many papers have been written on that chapter, dealing mainly with the effects of mandatory separation on (i) efficiency and (ii) investment incentives:

Efficiencies in vertical integration

The discussion on the efficiency of vertical integration regards the economic question of what defines company borders. In the absence of vertical integration, production is coordinated through market transactions between firms, with price and quality the main coordination mechanisms. Integration leads to a structure in which the firm has to coordinate the production instead. Coase (1937) addresses the question of why the latter exists at all, given the market takes over the allocation mechanism. He argues that the market approach adds costs associated with defining and enforcing contractual property and contractual rights. Those costs include, among others, search and information costs, bargaining costs, policing costs, or legal costs and are summarized under transaction costs. He concludes that vertical integration takes place when the transaction costs exceed the organizational costs. Williamson (1985) extends this idea and argues that transactions with uncertain outcomes that are conducted frequently and require investments in highly specific assets are most efficiently organized under vertical integration.

Another argument put forth against separation is the risk of double marginalization (Bolle & Breitmoser, 2006; Chen & Sappington, 2009; Höffler & Kranz, 2011). Double markups occur when the wholesale buyer pays more for the input than the wholesaler’s marginal costs and raises prices above marginal costs (Spengler, 1950). Thus, the welfare loss which results in pricing well above marginal cost is duplicated for the wholesaler and the wholesale buyer. Pricing above marginal cost, however, is not feasible if the market is perfectly competitive. Thus, the double marginalization problem only occurs if there is imperfect competition in downstream markets. The double mark-up problem also vanishes even in imperfectly competitive environments if the trading parties engage in efficient bargaining with nonlinear pricing structures, such as two-part tariffs.

An additional efficiency loss that comes with vertical separation is seen in the loss of economies of scale. It is argued that economies of scale are higher without separation because a single provider on one infrastructure reduces overall investment costs and avoids duplication in consumer acquisition efforts, billing, and operational support systems (OECD, 2001). This argument originates from a time before the positive effects of service competition on price and innovation were widely acknowledged (Cambini & Jiang, 2009). With today’s open access interfaces, mandated access regulation, and vividly competitive European retail markets, these arguments become less relevant.¹¹

Some researchers point out that economies of scope between the retail and wholesale segments can be substantial, to the point that an integrated market structure is more effective (Cave & Martin, 1994; Kahn, 2004; OECD, 2001). It is apparent that certain facilities can be used jointly and some cost categories can be spread between both segments. However, these associated cost reductions are unlikely to have a substantial effect, as retail and wholesale operations have limited overlaps. Moreover, the economies of scope in operating infrastructure and delivering the services (broadband or telephony) to which these authors referred apply to the days of Public Switched Telephony networks (PSTN), in which the service layer was tightly coupled with the infrastructure layer. Today, networks have developed toward an all-IP infrastructure, which realizes fixed telephony services and TV services using a common IP platform (see (Gonçalves & Nascimento, 2010; Kirsch & von Hirschhausen, 2008). The IP platform is fully decoupled from the infrastructure layer and can be offered irrespective of the communication infrastructure (mobile, fiber-to-the-curb (FTTC), fiber-to-the-premises (FTTP), or cable). This constitutes the decoupling of the infrastructure and service layer. Consequently, the potential for economies of scope is substantially reduced for all-IP networks. This argument is in line with empirical results obtained by Bruno and Manello’s (2015) analysis of recent financial data from 13 European incumbents. Their results suggest “that probably vertical economies of scope are not so relevant in the Telecommunications sector, or, at least, they are more than compensated by efficiency gains” (Bruno & Manello, 2015, p. 136).

Summing up, we still see potential efficiencies associated with vertical integration, but they are likely of limited magnitude. They are composed of transaction costs between wholesalers and retail operators that exceed organizational costs, as well as economies of scope through joint usage of facilities or personnel. Most transaction costs, however, are not specific to vertical disintegration, as they are also incurred as part of a transaction between the SMP operator and a service provider making use of mandated wholesale access provisions. Accordingly, these transaction costs represent a necessary cost to enable competitive retail markets.

Reduced investment incentives

High upfront costs in network upgrades and rollouts require significant investment from private market players. Thus, incentives to foster investment are one key objective policymakers consider when designing regulatory measures. Mandatory separation has been widely criticized as detrimental to investment incentives. Numerous papers model several institutional setups to compare a vertically integrated institutional structure against a disintegrated structure in network industries.

In particular, two papers (Buehler et al., 2006; Buehler et al., 2004) conclude that a disintegrated provider invests less as profits from the service provisioning are not considered in investment decisions. Furthermore, Sarmento (2015) finds that investment incentives are lower under disintegration, irrespective of whether access price regulation is in place. Moreover, the findings of Milliou (2004) suggest that vertical disintegration furthermore leads to less investments in research and development. Similarly, Cremer and De Donder (2013) compare integration without regulation with ownership separation and find that investment levels and welfare gains are higher under functional than under ownership separation.

However, a paper by Avenali et al. (2013) challenges those findings. The authors model a situation in which an upstream monopolist provides network access to all downstream firms using a regulated linear access charge. They conclude that investment incentives may increase under ownership separation while also increasing social welfare. Along the same lines, Cadman (2019) finds that integration gives investments an additional strategic value, as it allows operators to benefit from a retail market foreclosure, and thus the net present value (NPV), and in turn, the incentive to invest, is higher than under disintegration. However, he points out that effective regulation rules out the foreclosure option, eliminating the difference.

The literature leans toward the notion that vertical disintegration reduces investment incentives. Therefore, and due to other disadvantages of imposed structural separation, including irreversibility, implementation costs, and the potential for a lower quality of service (Brito et al., 2011; Teppayayon & Bohlin, 2010), we follow the recommendation of Gonçalves and Nascimento (2010) to mandate separation only as a remedy of last resort.

Voluntary separation

The research on voluntary vertical disintegration is far less abundant. Most papers explore different levels of opening the network and discuss the factors leading to success or failure based on experiences with wholesale-only providers (Brennenraedts & Maltha, 2008; Lehr et al., 2004; Magnago, 2004). Our article, in contrast, relates to the strand of literature comparing the integrated business model with a wholesale-only model in regard to the ability to recover the investments necessary to rollout FTTP networks.

In an early paper, Banerjee and Sirbu (2005) develop a techno-economic model to compare the profitability of a wholesale-only operator with that of an integrated operator. They conclude that municipalities or communities building out FTTP wholesale-only networks can realize sustainable prices, are likely to create greater welfare, and are just as likely as an integrated operator to recover costs. While Banerjee and Sirbu consider an urban setting with 10,000 households, Felten and Langer (2016) differentiate between regions with deviating cost structures. They use a business case model to compare the percentage of the population that can be covered by a wholesale-only and an integrated operator without subsidies. They argue that wholesale-only operators could achieve higher penetration rates and lower financing costs and therefore are significantly better positioned to recover FTTP investments than an integrated operator. Additionally, they point out the sensitivity of capital costs, and thus draw the conclusion that public policymakers should aim to derisk infrastructure and promote the emergence of wholesale-only networks.

In summary, the literature on voluntary open access leans toward an optimistic view on wholesale-only networks, while the literature on mandatory separation finds efficiencies of vertical integration and argues that incentives to invest are higher under vertical integration. However, the existing research discussed here along with numerous case studies on vertically separated industry structures (Cadman, 2019; Mancuso, 2012; Nucciarelli et al., 2010; Teppayayon & Bohlin, 2010) do not address the question of how public policymakers can incentivize market participants to voluntarily adopt a wholesale-only business model, thereby fostering voluntary disintegration while preserving incentives to invest. We aim to close that gap by developing a DCF model that compares the profitability of an integrated operator and a wholesale-only operator and discussing potential (regulatory) measures that can serve the aim outlined above.

The model

When the EC introduced mandatory separation provisions for the first time in 2007, the concern was raised that under strong forms of vertical separation,

the largest part of the potential reward for taking the risk of investing in […] new networks would accrue to external service providers rather than an internal services division. According to the EC, only direct regulatory incentives for the network operator (in the form of guaranteed rates of return) might overcome the reluctance of a separated access provider to make such investments, which then puts the burden of determining the pace of innovation onto the regulator rather than the market. (European Commission, 2007, p. 43)

This argument presumes absent or impaired investment incentives for network operators without a retail arm compared to integrated operators. However, the demand and supply of broadband services have changed dramatically since 2007. Bandwidth-intensive services, such as high definition video-on-demand (VoD) or over-the-top video conferencing, have reached mass adoption (Cisco, 2019), increasing the demand for new networks providing higher bandwidths (Stocker & Whalley, 2018). Given this change in demand, providing guaranteed rates of return through regulation may not be necessary to provide for necessary investment incentives. The wholesale-only business model may allow network providers to recoup investment costs just as they would in an integrated business model.

Following the analytical framework developed by Cadman (2019), we apply a discounted NPV model to compare the ability of wholesale-only and integrated providers to recoup investment costs. As Cadman described, a firm would choose to invest if the NPV is greater than zero, given a discount rate greater than its opportunity cost of capital. Using this analytical approach, we compare the discounted NPV of an integrated operator, ${NPV}_{int}$ , for investing in FTTP with the NPV for a wholesale-only operator, ${NPV}_{WO}$ . The integrated business model is more financially attractive than the wholesale-only model if ${NPV}_{int} > {NPV}_{WO}$ , and vice versa.

In this chapter, we develop an NPV model for an integrated operator and a wholesale-only provider. In a second step, the model inputs for the wholesale-only provider are altered as a sensitivity analysis, which allows us to identify potential levers that could increase the financial attractiveness of an NPV model.

Base case model

Modeling the NPV basically requires simulation of the business case of an FTTP rollout. We do so by adopting a calculation methodology that is applied in several state aid programs throughout Germany, including the Bavarian subsidy scheme “Next generation network for commercial and accumulation areas in Bavaria.” That program is financed with an overall budget of €1.5 billion. It employs the gap-funding model, in which a municipality provides a grant for covering the so-called profitability gap of a network operator that builds, operates, and owns the new broadband infrastructure (see European Commission, 2008, for program description). The profitability gap is calculated as the sum of all anticipated discounted revenues subtracted by discounted investment and operating. Table 2 shows the relevant cost and revenue categories used within the Bavarian subsidy scheme.

Table 2.

Revenue and investment categories.

Category	Position
Revenues	Direct service revenues from newly acquired customers
	Additional revenues deriving from existing customers’ upgrades
	Indirect revenues, e.g., wholesale
Investment	Civil engineering
	Passive infrastructure
	Active infrastructure
Operating costs	Wholesale access (last mile access, etc.)
	Direct operating costs
	Capital costs

Our model is based on three modules: The upfront cost module calculates investment costs for a nationwide greenfield FTTP network. The demand module forecasts penetration on the newly built FTTP infrastructure while considering competing operators and infrastructures. The last module is based on the demand model and calculates revenue and operating costs.

FTTP investment costs per household vary widely between regions. The main causes are differences in population density and different drop segment lengths. Our investment cost per household estimation is based on the cost modeling obtained by Jay et al. (2012) and a cost modeling survey by Feijóo et al. (2018). In our model, we apply the cost per households estimates shown in Figure 1 ranked by clusters, each of which represents 5% of all households. Cluster 1 includes all households located in the densest urban areas with lowest per household FTTP investment costs, while households in cluster 20 are located in the most rural areas and require the highest capital expense to connect to an FTTP network.¹²

Figure 1.

Investment costs per household model input.

Using above household cost estimates, investment costs for each cluster, c, are calculated as follows:

{Inv}_{c} = \sum_{t = 1}^{T} \frac{T_{c} \times (H_{c} \times S_{t})}{{(1 + i)}^{t}}

where ${Inv}_{c}$ is the investment net present value for c; i is the weighted average costs of capital; T_c is the investment costs per household in c; H_c is the households in cluster c; and S_t is the share of households which are connected in period t

The variable S_t is used to reflect the constraint of civil engineering capacity, making it infeasible to build out nationwide FTTP networks in only 1 year. Thus, we assume that costs are split equally between the first 4 years in each cluster. H_c equals the number of households in the respective clusters.

Demand model

The take-up (defined as households connected divided by households passed) on the infrastructure is driven by three major determinants: (1) migration of customers on legacy copper infrastructure to FTTP services, (2) the degree of inter-platform competition stemming from parallel FTTP or Hybrid Fiber Coaxial (HFC) infrastructure, and (3) the service providers’ decision to migrate to FTTP networks from copper networks.

The diffusion trajectories of new technologies are often found to follow an S-curve pattern. Rogers (2003, pp. 136–167) highlighted that the S-curve consists of three sections: A slow growth incubation period that is characterized by a lack of demand for a newly introduced technology, followed by the fast growth stage, and concluded by a slow-growth maturity stage, in which adoption rises slowly as it reaches the market’s capacity.

In our context, FTTP does not represent new technology. The ubiquity of mobile networks and IP-based services such as subscription VoD has led to a surge of demand for fast broadband (see e.g. CERRE, 2017), while other quality parameters, such as delay, jitter, and packet loss also profoundly affect customers’ quality of experience (Stocker & Whalley, 2018). Early empirical evidence also shows that bandwidth demand and the valuation of Fiber to the Home (FTTH) has grown significantly in recent years (Grzybowski et al., 2018). Considering this, we do skip the incubation period, and therefore model the adoption curve over time as a limited exponential growth function, defined as follows:

{Take-up}_{c} (t) = F_{c} - F_{c} \times e^{- k t}

where ${Take-up}_{c} (t)$ is the adoption in cluster c in year t; F_c is the maximum take-up in cluster c; and k is the growth factor.

The value of F_c in year T reflects the circumstance that other competing infrastructures (cable, mobile, and other FTTC or FTTH networks) are likely to account for substantial market share. These are likely to be found in urban and suburban areas, while rural areas throughout Europe have little or no HFC coverage and it will most likely be too costly for more than one FTTP network to become established (Feijóo et al., 2018). Thus, we assume the maximum take-up to be limited to 55% in urban cluster 1 (with the highest population density) and 80% for rural cluster 20 (lowest population density).¹³ Adoption is modeled over an 8-year period and depicted in Figure 2.

Figure 2.

Take-up curve for all 20 clusters. The top line shows the modeled adoption curve for the most rural cluster, 1, while the lines below reflect clusters 2 to 20 in descending order.

Revenue and operating costs model

The revenue and operating costs model is based on the revenue per household connected and the EBITDA margins. We assume the effective monthly revenues per household (ARPU) to be €45, corresponding to the average EU28 least expensive price for a 200+ Mbps tariff including voice telephony (European Commission, 2018a). The wholesale price is set €10 lower,¹⁴ reflecting the difference between retail prices and wholesale products. Wholesale-only providers lack the low margin and asset-light retail portion of the value chain and therefore generate higher operating margins. Our EBITDA margin assumptions are sourced from a European industry benchmark carried out by Barclays (2018a), which reports 45% as a best-in-class EBITDA margin for integrated operators, while wholesale-only operators achieve an average of 55%. Furthermore, we add an efficiency markup, ${eff}_{int}$ , of €2 per month per connected household for the integrated operator to reflect the potential efficiencies of vertical separation described in the third section. Using these model inputs, operating expenses (OPEX) are calculated as follows:

{OPEX}_{t} = (ARPU \times {Take-up}_{c t}) \times (1 - EBITDAmgn) + {eff}_{int}

Full model description and results

The DCF is calculated for each cluster representing differently populated clusters. The full model is represented as (see Online Appendix for model input summary):

{NPV}_{c} = - {inv}_{c} + \sum_{t = 1}^{T} \frac{({Take-up}_{c t} *ARPU) - {OPEX}_{t}}{{(1 + i)}^{t}}

where ${NPV}_{c}$ is the net present value for cluster c and T is the business case horizon.

Our model inputs consider the main arguments put forward by critics of separation. Vertical efficiencies decrease operating costs and the ability to capture profits on the retail market increases the economic value of a connected customer.¹⁵ Figure 3 visualizes these results per cluster using the formula presented above.

Figure 3.

DCF results for the base model. The rows reflect the clusters, c, while columns show the respective business case horizons, T. Red cells indicate that neither a wholesale-only provider nor an integrated operator could recover their investment. Yellow cells are profitable only for an integrated operator. Green cells indicate that wholesale-only and integrated operators can both recover their investment costs. DCF: discounted cash flow.

The integrated operator reaches the break-even point one to two years earlier than the wholesale-only operator. Moreover, the NPV for each cluster and each business case horizon is higher under vertical integration. These results explain the predominantly vertically integrated industry structure that can be found in European markets. Private players currently maximize their profits under a vertically integrated structure rather than by adopting a wholesale-only business model. The assumption chosen in this base model incorporates arguments against vertical integration while not considering factors that might have a beneficial effect on the financial performance of wholesale-only providers. Therefore, the following subsection will discuss such factors and their prerequisites and analyze their effects on the financial attractiveness of a wholesale-only business model by adjusting the DCF model inputs and testing sensitivities.

The up-side case for the wholesale-only model

As described in the second section, there is a rising number of network providers voluntarily choosing to either spin-off their network business (e.g. O2 Czech/Cetin) or entering broadband markets with a wholesale-only business model (e.g. Austria’s NöGig). The rationales these private firms communicate toward the financial markets hint of potential upsides of the wholesale-only model in regard to risk factors affecting costs of capital (often defined as weighted average costs of capital (WACC)) and anticipated take-up.

Firstly, the wholesale-only provider business model could benefit from lower WACC levels due to a lack of activities within the competitive retail segment. Infrastructure investors seek stable, predictable, and long-term cash flows and are willing to accept lower rates of returns on their invested capital. The lack of a retail arm benefits wholesale-only providers in that regard. Anecdotal evidence is already available in capital markets. O2 Czech, owned by private equity firm PPF, spun off its network division in 2015 to account for “different types of investments and horizons” between the network and retail business (CETIN, 2017). Furthermore, Windstream, an integrated operator in the United States, pursued a similar strategy, spinning off its fiber and copper assets into a separate company that was later listed on the stock exchange. Windstream defended the spin-off by arguing it was expected to “create [the] opportunity to unlock shareholder value by creating two independent public companies with distinct investment characteristics” (Windstream, 2014).

The results obtained by Schaeffler and Weber (2013) point in the same direction—they find that integrated utilities do have higher risk levels when compared to pure utility network providers without their own retail operations. Lower costs of capital, however, require that regulation does not interfere. Regulatory costs affect the risk assessment profoundly and therefore affect capital costs (Gentzoglanis, 2007). Regulatory authorities directly influence capital costs, for example, through pledging to reduce ex ante regulation for wholesale-only providers, as in the new EECC.

However, to the best of our knowledge, regulators throughout Europe have not proactively published guidelines on regulation reduction that would be granted in case of voluntary disintegration. Thus, the potential positive effect of reduced regulation is still subject to uncertainty, and therefore unlikely to play much of a role in private firms’ decision-making process. Thus, predefining reductions in ex ante regulation in advance of any voluntary separation could be a measure to derisk a separation decision and reduce capital costs.

Another important factor concerns the anticipated take-up of infrastructure. A nondiscriminatory wholesale-only provider will, at any given price, aim to attract as many service providers to sell products on the network as possible. If this aim succeeds, a wide variety of consumer offers will emerge, as service providers seek to differentiate themselves. In contrast, an integrated operator maximizes profits by acquiring as many profitable end-customers as possible while wholesale revenues are mostly seen as a regulatory requirement rather than a strategic priority.

The limited existing research on factors driving the adoption of ultrafast broadband indicates that intra-platform competition has a positive effect (Fourie & de Bijl, 2018; Ovington et al., 2017). A similar argument can be made with respect to tariff diversity. An increase in service providers selling products on a network will likely lead to a broader diversity of customer acquisition strategies. This, in turn, increases the need to differentiate from fellow service providers and thus ultimately leads to an increase in tariff diversity. Consequently, it is likely that customers will have a wider choice of tariffs on wholesale-only networks compared to a network operated by an integrated provider and therefore would choose to migrate to the FTTP network more quickly. This theoretical argument is aligned with empirical results obtained by (Haucap et al., 2016) using cross-sectional data and by Lange (2017) using time-series data. Both empirical analyses find that an increase in tariff diversity provides a significant impetus to broadband adoption.

Another important factor is the expected degree of inter-platform competition. Operators face two options: to enlarge the addressable market or to offer higher-quality products to their customers. Either they invest in their own networks or instead buy network access wholesale from other operators. The latter increases inter-platform competition but also duplicates costs and reduces take-up compared to the former option, where there are fewer networks competing for customers. Consequently, the anticipation of upcoming inter-platform competition is a key aspect affecting firms’ investment decisions.

If operators expect not to be challenged by another infrastructure in a given region, they would anticipate take-up rates substantially higher than in another region in which they expect another operator to build out their own, separate infrastructure. Under vertical disintegration, there is no possibility for a wholesale-only provider to discriminate against its own retail arm. Accordingly, operators presented with the two options described above do not face the risk of being discriminated against. In contrast, this does not hold true for a case in which network access would be supplied by an integrated operator. Thus, there is reason to expect that there will be less inter-platform competition if a wholesale-only provider is present in the market. Nonetheless, wholesale-only providers might also have an incentive to discriminate to maximize profits (Krämer & Schnurr, 2014). Thus, a nondiscriminatory market is key to achieve positive effects on anticipated take-up mentioned above.

While more intra-platform competition and less inter-platform competition represent potential upsides to wholesale-only take-up, there is a factor that prevents this upside from materializing. In many European markets, incumbents’ wholesale tariff structures include significant quantity rebates.¹⁶ Service providers buying access products from incumbents do have an incentive to stick to the access products, even if other regional wholesale-only networks offer FTTP and the incumbent inferior xDSL-based access, as sourcing from both providers would effectively increase the access costs for all households. Another similar factor is the existence of long-term access pricing tariff structures, where wholesale buyers receive additional discounts for committing to longer periods.¹⁷

To summarize, a lower WACC and faster take-up evolution are two potential upsides for the wholesale-only business case. To analyze the impact on the comparison to an integrated operator model, we alter the input parameters for the wholesale-only provider by reducing the WACC by 1%, increasing the maximum take-up by 10% in cluster 1 (and clusters 1 to 19 accordingly—see “Base case model” section), and reduce the time until the final take-up is reached by 1 year.

These small changes in input produce quite a different picture (Figure 4). The wholesale-only provider reaches the break-even point for clusters 1–18 one year earlier than an integrated operator. This result indicates that even a small reduction in capital costs and improved take-up could lift the financial attractiveness of a wholesale-only business model to a point at which rational market participants would choose such a model over an integrated structure. Our DCF model is limited by the fact that we assume a greenfield rollout scenario and do not include one-time break-up costs. Thus, it can only represent the rationale for incumbents to a limited extent. Nevertheless, the potential positive effect of reduced costs and higher take-up for a separated structure will be comparable, even in a brownfield scenario and if one-time break-up fees were included in the model.

Figure 4.

DCF results for the upside case.¹⁸ The rows reflect the clusters, c, while columns show the respective business case horizons, T. Red cells indicate that neither a wholesale-only provider nor an integrated operator could recover the investments. Blue cells are profitable only for a wholesale-only operator. Green cells indicate that wholesale-only and integrated operators can both recover their investment costs. DCF: discounted cash flow.

Discussion and policy implications

Ex ante regulation in the telecommunications sector has succeeded in enabling retail competition by creating market structures that allow alternative operators to effectively compete with incumbent operators on the retail level as well as providing the stepping stone to build-out their own networks in urban and semi-urban environments. However, investment costs in semi-rural and rural regions do not allow more than one network to be economically viable.

This article discussed a vertically separated industry structure as a potential answer to issues arising in such regions, most notably, potential downstream discrimination. We surveyed the existing literature and presented mandatory separation opponents and proponents’ main arguments as well as reviewing the limited existing research on voluntary separation. Based on that review, we concluded that mandatory separation comes with significant drawbacks and issues, and therefore should remain a measure of last resort.

In regard to voluntary separation, we contribute to the literature by developing a DCF model to provide an analytical framework that allows a comparison of the financial incentives for firms to adopt either a wholesale-only or an integrated business model. In our base model, which aims to incorporate the existing literature findings and current market players’ performance indicators, we found that an integrated business model prevails as beneficial throughout all clusters. As a next step, we argued that NRAs and public policymakers do have levers and options to decrease the regulatory risk for wholesale-only providers and to positively affect prospective take-up for wholesale-only providers. Feeding these arguments into our model, we found that even slight improvements (1% WACC reduction and +∼10% take-up up-side) are sufficient to tilt the financial attractiveness toward the wholesale-only business model.

Based on these analytical results, as well as the existing literature, we recommend public policymakers and NRAs to jointly and coordinately incentivize market participants to adopt a wholesale-only business model. Doing so will allow society to reap beneficial effects of a separated industry structure while superseding the need for strong structural market intervention. In practical terms, we propose that to reduce uncertainty and regulatory risk, public policymakers and NRAs should adopt, refine, and make use of the new EECC provisions on voluntary separation.

Although EECC Articles 78 and 80 list precise procedural and organizational requirements (e.g. 3 months’ notice, no common control or ownership, no exclusive agreements), they do not define the extent to which ex ante regulation is reduced for voluntarily separated operators. SMP-designated operators can generally be subject to the full scope of ex ante regulation—obligations of transparency, nondiscrimination, access or price-control, and cost accounting (see Article 78 new EECC¹⁹)—but may be subject only to nondiscrimination, access or fair, and reasonable pricing obligations (Article 80 paragraph 2). These provisions represent guidelines rather than precise provisions for the extent to which regulation would be reduced in case of a voluntary separation, and thus do not reduce uncertainty and achieve the effect discussed in section “The up-side case for the wholesale-only model.”

To address this gap, we propose that regulators proactively define the ex ante reduction if a vertically integrated SMP operator voluntarily separates its wholesale and retail business under the fulfillment of the requirements of Article 80 Paragraph 1 (sections a and b). This can take the form of binding commitments given by NRAs (already proposed in the new EECC²⁰). NRAs could commit to binding agreements to proactively initiate consideration of such a move at the SMP operator, even if operator is not yet considering a voluntary separation.

Secondly, regulators need to consider the hampering effect of wholesale tariffs featuring volume or long-term discounts. These tariff structures incentivize retail providers to buy access products from the incumbent in regions even if a wholesale-only provider also provides access to FTTP networks. In such cases, buying from the incumbent would be rational for the retail provider, as it is contractually obliged (long-term discount), or doing so decreases the average cost per subscriber. To preclude associated negative effects on the anticipated take-up of a wholesale-only provider, NRAs should carefully weigh the benefits of volume and long-term wholesale tariffs.

These approaches are intended to nudge private firms toward a wholesale-only business model. Public policymakers also have also the option to more directly influence the business model with broadband subsidy schemes. In recent years, governments throughout Europe have increasingly funded broadband deployment with state aid (see Feasey et al., 2018, for a European overview). These programs represent an opportunity to foster the emergence of additional wholesale-only networks, as public policymakers could tailor the programs accordingly. Italy can be examined as a rather extreme example in that sense; the government’s state aid program strictly excluded integrated operators from public tenders (European Commission, 2016). This approach, however, risks decreasing the competitiveness of the tender process and therefore decreases the intervention’s efficiency. As a result, public policymakers should extensively analyze the market structure and potential effects of such restrictions. Conveniently, the European guidelines for state subsidies in broadband deployment²¹ allow favoring wholesale-only proposals by awarding them additional points. Granting additional points in the tender would not exclude vertically integrated operators but could further nudge operators toward voluntary vertical separation.

Our findings show that mandatory separation is not the only route toward a separated industry structure. Public policymakers and NRAs do have many levers to strengthen the observed trend toward voluntary separation. We suggest they proactively and coordinatively use those opportunities both in state aid programs as well as precise, binding commitments defining the potential reduction in ex ante regulation in cases of voluntary separation.

Supplemental material

Supplemental Material, Model_altered_model - Towards a vertically separated broadband infrastructure: The potential role of voluntary separation

Supplemental Material, Model_altered_model for Towards a vertically separated broadband infrastructure: The potential role of voluntary separation by Fabian Queder in Competition and Regulation in Network Industries

Supplemental material

Supplemental Material, Model_base_case - Towards a vertically separated broadband infrastructure: The potential role of voluntary separation

Supplemental Material, Model_base_case for Towards a vertically separated broadband infrastructure: The potential role of voluntary separation by Fabian Queder in Competition and Regulation in Network Industries

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Fabian Queder

Supplemental material

Supplemental material for this article is available online.

Notes

References

Abrardi

Cambini

(2019). Ultra-fast broadband investment and adoption: a survey. Telecommunications Policy, 43(3), 183–198. doi:10.1016/J.TELPOL.2019.02.005

Avenali

D’Annunzio

Reverberi

(2013). Bundling, competition and quality investment: a welfare analysis. Review of Industrial Organization, 43(3), 221–241. doi:10.1007/s11151-013-9378-8

Banerjee

Sirbu

(2005). Fiber to the premise (FTTP) industry structure: implications of a wholesale-retail split. In 2nd international conference on broadband networks, BROADNETS 2005, 2005 (pp. 599–606). doi:10.1109/ICBN.2005.1589786

Barclays. (2018a). European telecom services make up or break up. Retrieved from https://www.sipotra.it/old/wp-content/uploads/2018/09/EUROPEAN-TELECOM-SERVICES.-MAKE-UP-OR-BREAK-UP.pdf

Barclays. (2018b). The rise of the wholesale-only mode. Retrieved from http://www.astrid-online.it/static/upload/ee15/ee15dd20387f716c39d4823e71773e7c1.pdf

Bauer

J. M.

(2010). Regulation, public policy, and investment in communications infrastructure. Telecommunications Policy, 34(1–2), 65–79. doi:10.1016/J.TELPOL.2009.11.011

Bertschek

Briglauer

Hüschelrath

Niebel

(2016). The economic impacts of telecommunications networks and broadband internet: a survey. SSRN Electronic Journal. doi:10.2139/ssrn.2828085

Bolle

Breitmoser

(2006). A service of zbw on the allocative efficiency of ownership unbundling standard-Nutzungsbedingungen: On the allocative efficiency of ownership unbundling on the allocative efficiency of ownership unbundling. Retrieved from http://hdl.handle.net/10419/23824

Brennenraedts

Maltha

(2008). Demand aggregation for Fiber optical networks in the business market: a case study of the Netherlands. In 2008 1st international conference on infrastructure systems and services: building networks for a brighter future, INFRA 2008. doi:10.1109/INFRA.2008.5439657

10.

Brito

Pereira

Vareda

(2011). An assessment of the equality of access and no-regulation approaches to next generation networks. Telecommunications Policy, 35(9–10), 818–826. doi:10.1016/J.TELPOL.2011.04.015

11.

Bruno

Manello

(2015). Benchmarking and effects of reforms in the fixed telecommunications industry: a DDF approach. Telecommunications Policy, 39(2), 127–139. doi:10.1016/J.TELPOL.2014.12.009

12.

Buehler

Gärtner

Halbheer

(2006). Deregulating network industries: dealing with price-quality tradeoffs. Journal of Regulatory Economics, 30(1), 99–115. doi:10.1007/s11149-006-0011-8

13.

Buehler

Schmutzler

Benz

M.-A.

(2004). Infrastructure quality in deregulated industries: Is there an underinvestment problem? International Journal of Industrial Organization, 22(2), 253–267. doi:10.1016/J.IJINDORG.2003.07.004

14.

Cadman

(2010). Means not ends: deterring discrimination through equivalence and functional separation. Telecommunications Policy, 34(7), 366–374. doi:10.1016/J.TELPOL.2010.05.004

15.

Cadman

(2019). Legal separation of BT: A necessary incentive for investment? Telecommunications Policy, 43(1), 38–49. doi:10.1016/J.TELPOL.2018.05.004

16.

Cambini

Jiang

(2009). Broadband investment and regulation: a literature review. Telecommunications Policy, 33(10–11), 559–574. doi:10.1016/J.TELPOL.2009.08.007

17.

Cave

(2009). Regulating telecommunications in Europe present and future. In William

L. H.

Lorenzo

(Eds.), Internet policy and economics: Challenges and perspectives (pp. 15–28). doi:10.1007/b104899_2

18.

Cave

Martin

(1994). The costs and benefits of accounting separation: the Australian and British debates. Telecommunications Policy, 18(1), 12–20. doi:10.1016/0308-5961(94)90089-2

19.

CERRE. (2017). Demand-side policies to accelerate the transition to ultrafast broadband - EU agenda. Retrieved from https://euagenda.eu/publications/demand-side-policies-to-accelerate-the-transition-to-ultrafast-broadband

20.

CETIN. (2017). 2017 investor presentation. https://www.cetin.cz/documents/10182/124348/20170102+CETIN+Company+overview.pdf/ac291d02-0990-45dd-8219-5ea7c51191be

21.

Chen

Sappington

D. E. M.

(2009). Designing input prices to motivate process innovation. International Journal of Industrial Organization, 27(3), 390–402. doi:10.1016/J.IJINDORG.2008.10.008

22.

Cisco. (2019). Cisco visual networking index: Forecast and trends, 2017–2022 white paper. https://www.cisco.com/c/en/us/solutions/collateral/service-provider/visual-networking-index-vni/white-paper-c11-741490.html#_Toc532256804

23.

Coase

R. H.

(1937). The nature of the firm. Economica, 4(16), 386–405. doi:10.1111/j.1468-0335.1937.tb00002.x

24.

Crandall

Eisenach

Litan

(2008). Vertical separation of telecommunications networks: evidence from five countries. Federal Communications Law Journal, 62(3), 493–539.

25.

Cremer

De Donder

(2013). Network investment under legal and ownership unbundling. Review of Network Economics, 12(1), 27–59. doi:10.1515/rne-2012-0012

26.

de Bijl

P. W. J.

(2005). Structural separation and access in telecommunications markets. Competition and Regulation in Network Industries, 6(2), 95–114. doi:10.1177/178359170500600202

27.

European Commission. (2007). Impact assessment accompanying document to the commission proposal for a directive of the European parliament and the council amending European parliament and council directives 2002/19/EC, 2002/20/EC and 202/21/EC commission proposal for a directive of t. Retrieved from https://ec.europa.eu/transparency/regdoc/rep/2/2007/EN/2-2007-1472-EN-1-0.Pdf

28.

European Commission. (2008). Case decision: State aid N 266/2008 – Germany Broadband in rural areas of Bayern. Retrieved from https://ec.europa.eu/competition/state_aid/cases/225952/225952_885448_27_1.pdf

29.

European Commission. (2010). Digital single market: political agreement on the rules shaping the telecommunication markets in the 5G era. Federal Communications Law Journal, 62(493). Retrieved from http://www.repository.law.indiana.edu/fclj/vol62/iss3/1

30.

European Commission. (2016). State aid SA.41647 (2016/n) – Italy - Strategia Banda Ultralarga. Retrieved from https://federalismi.it/ApplOpenFilePDF.cfm?artid=32268&dpath=document&dfile=22072016110021.pdf&content=State+aid+SA.41647+(2016/N)+%25E2%2580%2593+Italy+-+Strategia+Banda+Ultralarga+-+stati+europei+-+documentazione+-+

31.

European Commission. (2018a). Fixed broadband prices in Europe 2017 (BIAC study). Retrieved from https://ec.europa.eu/digital-single-market/en/connectivity

32.

European Commission. (2018b). Study on broadband coverage in Europe 2017. Retrieved September 17, 2019, from https://ec.europa.eu/digital-single-market/en/news/study-broadband-coverage-europe-2017

33.

European Parliament & Council. (2002). Directive 2002/21/EC of the European parliament and of the council of 7 march 2002 on a common regulatory framework for electronic communications networks and services (framework directive). Official Journal of the European Communities (L 108), 45, 33–50. doi:http://data.europa.eu/eli/dir/2002/21/oj

34.

European Parliament & Council. (2018). Directive 2018/1972 of the European Parliament and the council of 11 December 2018 establishing the European Electronic Communications Code (recast). Official Journal of the European Union (L 321/37). https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32018L1972&from=EN

35.

Feasey

Bourreau

Nicolle

(2018). State aid for broadband infrastructure assessment & policy recommendations. Retrieved from https://www.cerre.eu/sites/cerre/files/CERRE_StateAidBroadband_FinalReport_0.pdf

36.

Feijóo

Ramos

Armuña

Arenal

Gómez-Barroso

J. L.

(2018). A study on the deployment of high-speed broadband networks in NUTS3 regions within the framework of digital agenda for Europe. Telecommunications Policy, 42(9), 682–699. doi:10.1016/j.telpol.2017.11.001

37.

Felten

Langer

(2016). Structurally independent broadband infrastructure can solve perceived FTTH coverage issues. SSRN Electronic Journal. doi:10.2139/ssrn.2794850

38.

Fourie

de Bijl

P. W. J.

(2018). Race to the top: Does competition in the DSL market matter for Fiber penetration? Telecommunications Policy, 42(9), 778–793. doi:10.1016/j.telpol.2017.11.003

39.

Gentzoglanis

. (2007). Regulatory risk, cost of capital and investment decisions in the telecommunications industry: International comparisons. In ITS-15th Biennial Conference, September 4–7, 2004. Berlin.

40.

Gonçalves

Nascimento

Á.

(2010). The momentum for network separation: a guide for regulators. Telecommunications Policy, 34(7), 355–365. doi:10.1016/J.TELPOL.2010.05.008

41.

Grzybowski

Hasbi

Liang

(2018). Transition from copper to fiber broadband: the role of connection speed and switching costs. Information Economics and Policy, 42, 1–10. doi:10.1016/J.INFOECOPOL.2017.07.001

42.

Hasbi

(2017). Impact of very high-speed broadband on local economic growth: Empirical evidence. Retrieved from https://econpapers.repec.org/RePEc:zbw:itsp17:168484

43.

Haucap

Heimeshoff

Lange

M. R. J.

(2016). The impact of tariff diversity on broadband penetration—an empirical analysis. Telecommunications Policy, 40(8), 743–754. doi:10.1016/J.TELPOL.2015.09.005

44.

Höffler

Kranz

(2011). Legal unbundling can be a golden mean between vertical integration and ownership separation. International Journal of Industrial Organization, 29(5), 576–588. doi:10.1016/J.IJINDORG.2011.01.001

45.

Huigen

Cave

(2008). Regulation and the promotion of investment in next generation networks—a European dilemma. Telecommunications Policy, 32(11), 713–721. doi:10.1016/J.TELPOL.2008.08.004

46.

Jay

Neumann

K.-H.

Plückebaum

(2012). The cost of nationwide fiber access in Germany. Communications & Strategies, 1(85), 169–188.

47.

Kahn

A. E.

(2004). Lessons from deregulation: Telecommunications and airlines after the crunch. AEI-Brookings Joint Center for Regulatory Studies.

48.

Kirsch

von Hirschhausen

(2008). Regulation of NGN: Structural separation, access regulation, or no regulation at all? Communications & Strategies, 1(69), 63–84.

49.

Krämer

Schnurr

(2014). A unified framework for open access regulation of telecommunications infrastructure: review of the economic literature and policy guidelines. Telecommunications Policy, 38(11), 1160–1179. doi:10.1016/J.TELPOL.2014.06.006

50.

Laffont

J. J.

Tirole

(2001). Competition in telecommunications. Retrieved from https://books.google.de/books?id=ucS5F4lg6TkC

51.

Lange

M. R. J.

(2017). Tariff diversity and competition policy: drivers for broadband adoption in the European Union. Journal of Regulatory Economics, 52(3), 285–312. doi:10.1007/s11149-017-9344-8

52.

Lee

Brown

J. S.

(2008). Examining broadband adoption factors: an empirical analysis between countries. Info, 10(1), 25–39. doi:10.1108/14636690810850148

53.

Lehr

Sirbu

Gillett

(2004). Broadband open access: lessons from municipal network case studies. In Proceeding of the TPRC Conference. 2004.

54.

Magnago

A. C

. (2004). Open access-business models and operational costs. In Proceedings of the broadband Europe international conference 2004, Brugge.

55.

Mancuso

(2012). Regulation and efficiency in transition: the case of telecommunications in Italy. International Journal of Production Economics, 135(2), 762–770. doi:10.1016/J.IJPE.2011.10.009

56.

Milliou

. (2004). Vertical integration and R&D information flow: Is there a need for ‘firewalls’? International Journal of Industrial Organization, 22(1), 25–43. doi:10.1016/S0167-7187(03)00090-0

57.

Nucciarelli

Sadowski

B. M.

Achard

P. O.

(2010). Emerging models of public–private interplay for European broadband access: evidence from the Netherlands and Italy. Telecommunications Policy, 34(9), 513–527. doi:10.1016/J.TELPOL.2010.07.004

58.

OECD. (2001). Restructuring public utilities for competition. doi:10.1787/9789264193604-en

59.

Ovington

Smith

Santamaría

Stammati

(2017). The impact of intra-platform competition on broadband penetration. Telecommunications Policy, 41(3), 185–196. doi:10.1016/J.TELPOL.2017.01.002

60.

Peitz

Cave

(2008). Innovation, convergence and the role of regulation in the Netherlands and beyond. Telecommunications Policy, 32(11), 744–754. doi:10.1016/J.TELPOL.2008.09.003

61.

Rogers

(2003). Diffusion of innovations (5th ed.). The Free Press.

62.

Sarmento

(2015). Vertical separation and access pricing: effects on investment incentives. Actual Problems of Economics, 167, 228–241.

63.

Schaeffler

Weber

(2013). The cost of equity of network operators—empirical evidence and regulatory practice. Competition and Regulation in Network Industries, 14(4), 385–410. doi:10.1177/178359171301400404

64.

Spengler

J. J.

(1950). Vertical integration and antitrust policy. Journal of Political Economy, 58(4), 347–352. doi:10.1086/256964

65.

Stocker

Whalley

(2018). Speed isn’t everything: a multi-criteria analysis of the broadband consumer experience in the UK. Telecommunications Policy, 42, 1–14. doi:10.1016/j.telpol.2017.06.001

66.

Teppayayon

Bohlin

(2010). Functional separation in Swedish broadband market: next step of improving competition. Telecommunications Policy, 34(7), 375–383. doi:10.1016/J.TELPOL.2010.05.005

67.

Van Der Wee

Casier

Dixit

Lannoo

Verbrugge

Colle

Pickavet

(2015). Techno-economic evaluation of open access on FTTH networks. Journal of Optical Communications and Networking, 7(5), 433–444. doi:10.1364/JOCN.7.000433

68.

Vogelsang

(2017). Regulatory inertia versus ICT dynamics: the case of product innovations. Telecommunications Policy, 41(10), 978–990. doi:10.1016/J.TELPOL.2017.09.006

69.

Williamson

O. E.

(1985). The ECONOMIC INSTITUTIONS of CAPITALISM firms, markets, relational contracting. University of California.

70.

Windstream. (2014). Letter to shareholders. Retrieved October 8, 2019, from https://www.sec.gov/Archives/edgar/data/1620280/000119312514450718/d798710dex991.htm#rom798710_4

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.77 MB

0.79 MB