International Institute of Communications

Shaping the policy agenda: TELECOMMUNICATIONS • MEDIA • TECHNOLOGY
Tel:+44 (0)20 8544 8076
Fax:+44 (0)20 8544 8077

social twitter sm  social linkedin sm  social youtube sm  social facebook sm

Taming the oligopolists

Sumit Sharma extends the argument for using competition policy rather than regulation for convergent networks by looking at oligopoly models.

Taming the oligopolists

Two consequences of recent technological and market developments in electronic communications networks are increased intermodal competition, and a transformation of networks from one to two-sided platforms that are part of a broader internet and media ecosystem delivering services to end users.

These technical and market developments include convergence of network technologies; consumer demand for service bundles; competition to provide these bundles; and the growing volume of data traffic driven by consumer demand for digital services. 


 "IP has led to the separation of services and content from the physical network."


The convergence of network technologies (and the adoption of technologies allowing for greater data transfer speeds) means that different communication infrastructures increasingly compete with each other, leading to greater intermodal competition. This convergence is based on the adoption of internet protocol (IP) technology, which means previously disparate communication infrastructures (fixed copper, fibre, CATV and mobile) use the same underlying transport protocols to supply various services to the end user. IP can be used to transport voice, various instant messaging services (which are essentially data traffic), and video in addition to data used for internet services. For example, to transmit voice, VoIP can be used over fixed networks and VoLTE over mobile 4G networks. IP based video on demand (VoD) is an over the top (OTT) service, ie. provided over the internet, and IPTV is a specialised service usually provided over reserved broadband capacity so as to meet certain quality of service parameters.

 

rise of two - and multi-sided platforms

The move to IP has also led to the separation of the supply of services and content from the underlying physical network. This means electronic communications networks now play the role of two-sided platforms (with consumers on one side and service providers on the other) in a digital ecosystem consisting of many interacting participants. In this ecosystem, voice and media services are just another broadband/internet service, and OTT service providers compete with both fixed and mobile network operators.

OTT service providers, such as Netflix and Skype, provide their services over telecoms networks but exercise no control over the networks themselves. End-user demand for access to communication networks is ‘derived’ – ie. based on the demand for the services provided over networks, rather than the networks themselves. So the supply of end-user services is based on vertical relationships between network operators and various content and service providers. These include TV broadcasters, OTT companies such as Netflix, and internet companies such as Google.

Indeed, electronic communications networks are transforming into multi-sided platforms and are keen to add advertisers to the mix, as Verizon’s push into mobile video and advertising via its proposed acquisition of AOL shows.

 

scope of article

This article discusses the implications of growing intermodal competition for ex-ante wholesale access regulation to encourage intramodal competition. I propose to discuss the implications of the development of the internet and media value chain in which electronic communications networks act as two-sided (or multi-sided) platforms in a forthcoming article.

Ex-ante wholesale access regulation is prevalent in Europe, and focuses on the regulation of access bottlenecks controlled by significant market power (SMP) operators (and see the European Commission recommendation on relevant product and service markets susceptible to ex-ante regulation). As such, this regulatory framework explicitly accounts for the possibility of more than one operator having SMP (captured by the notion of collective or joint dominance in the form of tacit coordination). However as noted by the body of European regulators, BEREC, whereas national regulatory authorities (NRAs) in Europe are experienced in regulating markets characterised by single-firm dominance, there is little precedent of findings of joint dominance and the subsequent design of regulatory remedies.

Indeed if there are two or more large networks serving end-users (based on some combination of cable, copper, fibre and mobile networks) then is regulation of these markets warranted? In other jurisdictions such as the US, the FCC moved away from strict wholesale access regulation, stating in its 2003 Triennial Review Order:

“We have come to recognise more clearly the difficulties and limitations inherent in competition based on the shared use of infrastructure through network unbundling. While unbundling can serve to bring competition to markets faster than it might otherwise develop, we are very aware that excessive network unbundling requirements tend to undermine the incentives of both incumbent LECs and new entrants to invest in new facilities and deploy new technology.”

The FCC continues this policy of regulatory forbearance for next-generation fibre access networks, a position also adopted by UK regulator, Ofcom, for next-generation access networks.

Against this background, BEREC’s ongoing work to discuss whether ex-ante regulation of oligopolies in the electronic communications sector is required and, if so, under what circumstances is timely. I explore whether ex-ante regulation of oligopolies in the electronic communications sector is required by revisiting the basics of oligopoly theory in the context of the competitive dynamics in the sector. These dynamics include the increase in cross-sector competition, the growing popularity of bundles (such as triple- and quad-play offers – a triple-play offer will typically comprise fixed voice, fixed broadband and TV. A quad-play offer will also include mobile services), and the entry of OTT service providers.

 

Regulating oligopolies?

The range of oligopolis tic outcomes

The outcome of oligopolistic competition can be either the same as in a competitive market; somewhere between a competitive outcome and a monopoly; or similar to a monopoly depending on the nature of competitive interactions between operators. A monopolist, as the only operator in a market, can set the market price. A firm operating in a ‘perfectly’ competitive market is a ‘price-taker’ and cannot influence the market price. An oligopolist may have some market power (ie. some ability to influence the market price), depending on the market structure. The oligopoly range is:

  • At one end of the spectrum, a homogeneous goods Bertrand model, in which firms compete on price, is an oligopoly with a competitive outcome (ie. where market prices are equal to the marginal costs of production, or – in the case of network industries – a measure of long-run average incremental costs). This assumes that there is no collusion among firms (or that there is a low likelihood of tacit collusion).
  • A homogeneous goods Cournot model in which firms compete on capacity; a heterogeneous goods Bertrand model; and a dynamic model in which firms compete on both price and capacity – these are examples of oligopolistic markets with an outcome somewhere between a competitive and a monopoly outcome (ie. where prices are above marginal/incremental costs but potentially below those that would be observed under a monopoly). This also assumes that there is no explicit collusion among firms (although instances where firms might be tacitly coordinating cannot be ruled out).
  • At the other end of the spectrum, an oligopolistic market in which firms collude (whether explicitly or not) is likely to result in a monopoly outcome, depending on how successful the firms’ collusive strategy is.

In the first case, no regulatory intervention would be required, as the result would be the same as in a competitive market. In the third case, market outcomes are likely to reduce consumer and social welfare, and some form of intervention may be required. Explicit collusion (or cartelisation) is already covered by competition/anti-trust law and so ex-ante intervention would not be required in this case.

The middle cases (situations where there is no explicit collusion, but there could be tacit coordination) are potentially contentious. Is regulation of these markets warranted, and what factors and evidence need to be assessed before regulators decide to intervene?

 

The potential for tacit coordination

There are established economic criteria for assessing the existence or likelihood of tacit collusion, such as those set out in the EU horizontal merger guidelines. These criteria relate to the incentives and ability to collude, including factors such as transparency, stability of the market, disciplining mechanisms, and external competitive pressure.

The need for ex-ante regulation on the basis of a joint dominance finding should be assessed according to these criteria. If it is found that there is limited scope for tacit coordination, or there is a low likelihood of such coordination arising, the rationale for regulatory intervention in the market would be weak. The case for intervention would be stronger if the assessment reveals that either tacit coordination has already been taking place in the market, or market conditions are such that there is a high likelihood of this happening in the market review period.

However, current market developments mean that the electronic communications market is constantly changing, which potentially reduces the likelihood of tacit coordination and provides incentives for network operators to compete aggressively. Recent trends that reduce the scope for sustaining tacit coordination include the growing popularity of bundles, the growth of OTT services fuelled by consumer demand, and asymmetries in technology cycles and investment costs across network operators. I discuss these briefly below.

 

The popularity of bundles

Retail broadband is increasingly supplied as part of bundle offers. This means that assessing ‘the price’ of the broadband offered in a dual-, triple- or even quad-play bundle is not straightforward, especially given the many possible variations with respect to the other bundle elements. This complicated retail structure reduces clarity in the market, making tacit collusion less likely. If prices are difficult to compare (or, as in this case, obscured, with operators unable to tell which particular package a consumer switches to), a firm losing sales and observing churn rates cannot determine why this is happening. For example, it cannot identify whether it is due to an unexpected change in demand, or a deviation from the coordinated outcome by the other parties.

In such cases, a punishment strategy might be mistakenly employed in instances of naturally decreasing demand, thereby destabilising the coordination. (Such a strategy would be setting lower retail prices to force the party deviating from the coordinated outcome to (further) reduce its prices and suffer losses, or make lower profits than it would otherwise have done.) 


 "The changes in the way that services are consumed add instability to the market."


Evolving customer demand and dynamism introduced by OTT services

Retail market developments and evolving consumer demand are an important motivation for technical development. As discussed above, consumers increasingly buy broadband services in bundles that include other services, such as media and voice. At the same time, consumer demand for higher broadband access speeds is increasing, so operators are coming under increasing pressure to deliver service upgrades to support this. The demand for ever-increasing internet bandwidth is a derived demand stemming from wide-reaching changes in the communications and entertainment technologies that consumers use. For example, voice over internet protocol (VoIP) and videoconference technologies, OTT video and music streaming, and online gaming all require significant amounts of bandwidth.

OTT services are also an emerging challenge for network operators, as they offer innovative services on the supply side and alter consumer habits on the demand side. For example, Netflix (an OTT player) argues that the proposed merger between AT&T and DirectTV would enable AT&T “to prevent or delay cord cutting and cord shaving” by consumers, ie. moving from cable and satellite TV subscriptions to cheaper online alternatives, reflecting the competition among these companies.

Such competitive pressure from OTT service providers on each network operator is independent of the presence of other network operators. From an economic perspective, these new services contribute to enhanced and more uncertain dynamics in the market.

These changes in the way that communications services are consumed, together with evolving customer demand (for service bundles and higher speeds), add instability to the market and may be an important driver of network operators’ quality and service upgrades.

 

Asymmetries in technology and investment costs

Asymmetry in investment costs and technology development cycles may also mean that the opportunity to gain a technical advantage through network upgrades is likely to come at different times for CATV, fixed and mobile operators. This allows each network technology in turn to enjoy a period of quality leadership, before another standard. This dynamic can be expected to incentivise network operators to capitalise on their position of quality leadership while it lasts, by aggressively attracting subscribers. This is likely to increase profits and help with the recovery of (often substantial) upgrade investment costs.

These asymmetries also create further instability for any hypothetical coordination among firms. technology catches up and exceeds the new standard. This dynamic can be expected to
incentivise network operators to capitalise on their position of quality leadership while it lasts, by aggressively attracting subscribers. This is likely to increase profits and help with the recovery of (often substantial) upgrade investment costs. These asymmetries also create further instability for any hypothetical coordination among firms.

 

Dynamic competition

Finally, it is not clear which oligopoly model would apply to the electronic communications industry, which is characterised by high sunk investment costs, substantial excess capacity, and negligible marginal costs of production. Oligopolistic competition in the sector may be closer to a Bertrand (price-based) model, and therefore a competitive outcome, than Cournot (capacity-based) competition, especially shortly after investment in network upgrades leads to an increase in capacity (such as investment in fibre access networks or the rollout of 4G).

Over the long run, these markets are probably best described by dynamic models of competition, in which firms must invest in upgrading networks and expanding capacity to keep up with the pace of technology change, but must price services competitively in order to grow or maintain market share. This raises the question: is two enough to arrive at a long-term competitive outcome?

To answer this, it is important to recognise that an assessment of long-term welfare should not focus only on the higher prices that may result from a more concentrated oligopolistic market structure. It must also take account of:

  • Potential changes or innovations in services offered as a result of demand trends and investment in new technologies. For example, price indices may fail to reflect the take-up of data-driven plans as much as minute-driven plans
  • Quality of service and benefits from competition on non-price features, which include broadband speed, reliability, security and privacy; customer service differentiators, such as access to technical support and installation professionals; innovative media services, such as on-demand catalogues or higher-quality broadcasts; and other initiatives such as WiFi hotspots and international roaming agreements.

 

From regulation to competition policy?

Given the development of intermodal (or infrastructure) competition, market developments which reduce the likelihood of tacit coordination, and the dynamic nature of competition in the sector, it may be appropriate to adopt a deregulatory approach (with appropriate transition periods) or ‘light touch’ regulation depending on specific national circumstances.

For example this could involve mandating supply (ie. commercial wholesale supply) and nondiscrimination but forbearing from setting prices based on cost orientation or charge controls.

Such an approach would be similar to that followed in the US (and the UK for next-generation fibre access networks), and would use competition law to ensure that firms do not engage in anticompetitive practices to foreclose entry. The advantages of such an approach would be stronger incentives for network operator to invest (and innovate) in broadband networks to compete with each other.

"These dynamics of the internet and media value chain are widely debated in various guises."

Examples of this dynamic at work include the competition between BT and Virgin Media in the UK, and between Google Fibre and AT&T in the US. The transformation of electronic communications networks from one to multi-sided platforms as discussed briefly above also suggests that using regulation to sustain competitive markets may be complex as it would have to account for:

  • Vertical relationships among platform operators and various content and service providers
  • Vertical integration between network operators, content providers and content producers
  • Changing market dynamics driven by fast paced innovation, and entry/exit across the internet and media value chain.

These dynamics of the internet and media value chain are being widely debated in various guises – the recent FCC open internet order and the European Commission’s digital single market initiative are two examples.

I propose to explore these complex set of interactions and whether a well-developed competition framework may be better suited to deliver the dynamic competition and investment outcomes that would provide long term benefits to consumers in a forthcoming article.

Stay up to date with the IIC

We will give you a monthly round up of up-coming events, where we’ve been as well as interviews and selected articles from InterMedia.

Subscribe to Policy World

Follow us on Twitter