Home > Media and Analysts > Speeches and Presentations > 2005 > Dec > Economic Regulation

15|12|05

Beesley Lecture - A Strategic Approach to the Economic Regulation of Spectrum, Telecoms and Broadcasting - 29 November 2005

Ed Richards, COO, Ofcom

Section I

Introduction

Good evening.

I would like to thank the organisers for inviting me here this evening to talk about our strategic approach to regulation of spectrum, telecoms and broadcasting.

I would like to cover four main areas:

Consider Ofcom’s general approach to economic regulation.^(-1-)
Review some of our key regulatory actions from an economic perspective.
Set these in the context of an underlying approach to economic regulation.
And conclude by identifying some challenges which still remain and which we shall have to address in consultation with others in the near future.

Ofcom’s General Approach

In preparing for this lecture, I thought it would be worth rereading some of the articles written by Michael Beesley, who gave his name to this important annual series of lectures on regulation.

In one of them^(-2-) I came across a succinct and valuable summary of the core tasks of regulation as Michael saw them. With his co-author, Stephen Littlechild, he placed promoting competition at the heart of a regulators' purpose:

Slide 1

Slide 2

This seems to me to be a very good place to begin. It is an encapsulation which has a strong sense of regulatory strategy at the heart of its message.

Understand the context in which you work.
Identify all the levers at your disposal and their relative efficacy.
And make choices between competing approaches which in sum amount to a strategic approach.

Ofcom was conceived many years ago, but born only just under two years ago, on 29 December 2003. As the newest of the sectoral regulators, our job was indeed to consider exactly what approach to regulation we were going to take – interestingly, not with a single Chief Executive or Director General, but as a Board of nine people.

For Ofcom of course, there is a significant additional dimension to those objectives that I described above. We are not purely an economic regulator of the kind that Michael Beesley had in mind. Our duties require us to be concerned with both the citizen and the consumer.

Our task is to navigate a path between these sometimes complementary but occasionally competing conceptions of the purposes and ends of regulatory activity.

Our approach can be characterised as seeking to apply economic principles wherever we can, even to the darker recesses of the cultural pole of our twin duties. This is not to stage a ‘takeover of the economists’ but rather the more limited objective of illuminating complex problems using the tools of economics, as a means of making a rounded, informed judgement.

We begin always by seeking to embed our approach in sound general principles of regulation; evidence based, transparent, consultative, proportionate and as predictable as possible;

In developing our regulatory strategy we have sought:

To be engaged with consumers and industry to understand these changes and developments, thereby avoiding the role of passive observer;
To reflect the highly dynamic nature of industry and technological evolution, remaining therefore at all times cautious of an excessively static view of markets;
To take a broad view, seeking to look at the connections and similarities between obstacles to competition across a range of markets rather than to focus exclusively on each economic market in isolation; and
To make choices between competing approaches, based on evidence and analysis and thorough review.

In a sense we have no option but to take this approach. Media and communications markets are highly dynamic, more so today than arguably at any time in their history. They are driven by great technological change and a speed of change which means that decisions made on the basis of static analysis, assessing a narrow set of evidence, will be overtaken by events.

We are in a period of unprecedented change. From analogue to digital, from narrowband to broadband, from passive to interactive services, from circuit switched to packet switched IP networks and many more rapid technological changes.

It is precisely for these reasons, for the tumultuous change that these developments represent, that one further principle is at the heart of our approach.

That is the wise counsel which advises that regulatory intervention needs to be weighed against the option of doing nothing or at least doing less.

There are at least two cases where this holds.

Firstly, where market failure may exist in some form, but where the costs of regulatory intervention outweigh any benefits from taking action.

Secondly, where the risk is specifically to nascent competition, and the cost of regulatory intervention is to inhibit the development of that competition by creating new or higher costs of doing business.

As ever with strategy, an organisation's approach is defined as much by what it does not do as much as by what it does do.

Our approach is not:

A process by which we try and determine which products, technologies and firms should succeed, as was the case for so much of the history of spectrum management for example;
Nor is it a process which looks to create specific market outcomes, with predetermined market shares, levels of switching or quantities of investment

It is a structured approach: that seeks to be anticipatory, informed and analytical.

We try to take account of the trade-offs inherent in regulation. For example, the problems faced by market entrants against the legitimate rights of incumbents;
We try to consider regulatory activity in the round, identifying options as part of an overall set of policies influencing an industry and group of markets in ways that serve the interests of consumers, recognising that changes in one place may give rise to consequences in another; and
We try to identify clear aims, such as focusing regulation at the real economic bottlenecks, targeting action on these root causes and ensuring that this is coupled with deregulation in areas where effective competition can be established.

I would like to turn to how this approach has manifested itself in some key areas of policy.

However, before doing so, I should acknowledge Ofcom's other important role as a national competition authority. I'm focusing tonight on ex-ante economic regulation, but we should not forget Ofcom's ability to deal with competition concerns through the application ex-post of competition law. Something that we always consider before embarking on ex-ante regulation.

Section II

Review of key economic regulation by Ofcom

I would like to begin with fixed link telecommunications.

In fixed telecoms we have had to address the combination of economic bottlenecks and vertical integration.

Unlike traditional utilities, much of the fixed telecoms network is potentially competitive. There are plenty of long distance networks in the UK. But many customers only have access to one supplier for the line into their house. Cable covers around 50% of homes, and there are other operators in some dense urban areas, especially for business consumers. But anyone wanting to offer anything close to a national service faces this bottleneck: no-one other than BT can viably supply the line between the customer and the exchange, and in many cases from the exchange further back in the network, too.

Anyone wanting to compete with BT in the parts of the network that are competitive needs access to this bottleneck. Yet BT is a vertically integrated company. Its wholesale customers, to whom it sells access to this bottleneck, are also its retail competitors. So the combination of vertical integration and market power in this bottleneck part of the network gives BT an incentive to discriminate against its wholesale customers.

The traditional remedy to this was to require BT to provide open access to these bottleneck facilities, and to do so on cost-based terms. That was combined with further general requirements not to discriminate, and not to engage in margin squeezing.

These remedies don’t take away the incentive on BT to discriminate. They restrict its ability to discriminate – but only to the extent that such discrimination can be detected, and demonstrated. As a regulator, we can fairly easily prove price discrimination when it happens. But non-price discrimination is a different story. Interconnection of networks requires agreement at a very detailed level to ensure that services work smoothly. Local loop unbundling, for example requires rules about issues such as where competitors equipment can be located in an exchange, who can have access to the key to the exchange, what power supply is available, and a host of other matters.

All of these are real barriers to competition, but if everyone had an incentive to solve them, they could be solved pretty quickly. That alignment of incentives did not exist and the outcome was endless disputes and allegations of discrimination. The problem is not confined to Local Loop Unbundling. It applies also to a whole range of other wholesale products, such as wholesale line rental, carrier pre-selection, partial private circuits and wholesale leased lines.

To avoid regulatory micro-management one obvious solution would be structural separation: the separation of ownership of the bottleneck access network from the competitive parts of the network. In our Telecoms Review, we looked closely at whether this was the right answer.

This solution has arguably worked in other regulated network industries. But there are more difficulties with structural separation in telecoms than in other network industries – and let me remind you at this point of the earlier injunction to ensure that the benefits of regulatory action are indeed higher than the costs of that intervention.

There are two main problems. One is that telecoms is so fast-moving. The dividing line between the bottleneck and the competitive part of the network might be in one place today, but in a completely different place tomorrow. If you had a fibre network for example, you might not even bother with local exchanges at all. If you break up the network for good, it can’t adapt; you fossilise your break point in one place for ever.

Secondly, the bottleneck isn’t at the same level in the network everywhere. In the City of London, for large business customers, some would argue that there may not be a bottleneck at all. But for a small telephone exchange in Cornwall, the bottleneck isn’t likely to be limited to just the line between the customer and the exchange; it’s all the equipment in the exchange, plus the line from the exchange 50 miles back to the nearest big town.

Slide 3

So our review settled on a concept that hasn’t been used in telecoms networks anywhere else in the world: what we call Equality of Access. This involves two things. First, Equivalence of Input on a product-by-product basis. If BT uses the bottleneck part of its network, it must make exactly the same wholesale product available to its wholesale customers, on exactly the same terms, and using exactly the same systems and processes, as it supplies it to itself. Not a similar product, the same product. The reason for this is the incentive that it gives BT. If a wholesale product or system isn’t up to scratch, then its own downstream business suffers too.

The second aspect of Equality of Access is organisational change within BT. It’s no good BT’s wholesale customers sharing their commercial plans with BT – which they have to, if they want to buy its bottleneck assets – only to see these plans gradually leak out to the parts of BT that are their direct competitors. So the organisational change in BT has set up a new, ring-fenced division called Openreach, which controls the bottleneck asset. Openreach is independent in many respects: it has its own brand; significant discretionary capital expenditure and its staff’s bonuses are based on Openreach’s performance, not BT’s.

To reinforce these changes, BT has established the Equality of Access Board which will monitor, report and advise the BT Group plc Board on BT's provision of products on an Equivalence of Input basis and on the operation of Openreach.

All of this sounds like a lot of regulation, but our aim is that this should permit significant deregulation in time. This is for two reasons.

Slide 4

First, by focusing regulation on the bottleneck part of the network we expect to be able to regulate elsewhere. You can already see examples of this: we’re deregulating aspects of retail business telecoms services and examining the possibilities for deregulation of large business markets and relaxation in the retail line rental price cap and the case for changes in leased line regulation in major urban areas. In the next 12 months we will also look at the potential for deregulation in areas including wholesale international calls, wholesale broadband access and interconnection services and the need for a residential price cap at all - all assuming the necessary competition concerns can be met.

Second, if you can give BT the incentive to build fit for purpose wholesale products first time around, then you have a good chance of avoiding being drawn into detailed product design – an area unequivocally best left to companies rather than regulators if at all possible.

A key feature of this approach is that it was not a market review, but a strategic review of what we wanted our over-arching policy approach to be. We looked at a series of markets together not in isolation and tried to establish an approach that would be effective in general as well as in specific cases. We were able to do this by accepting undertakings from BT under the Enterprise Act – the key part of our competition law framework which enables us to deal not just with specific instances but where features of a market are preventing, restricting or distorting competition.

Broadcasting

I want to describe three areas which come together as the components of what we are trying to achieve in broadcasting. The first is the move towards digital switchover, second is the range of measures that are deployed to support competition in broadcasting and third, the case for continuing market intervention in the form of public service broadcasting.

The broadcasting market is becoming a more competitive and open market. One of the key changes taking place in broadcasting is the rise of digital television driven through satellite, cable and terrestrial platforms. This process, which brings dozens or hundreds of channels to viewers, will become universal as a result of switchover.

At present this is market driven, as consumers choose to adopt digital TV to take advantage of the great increase in choice offered by the multi channel environment. But in due course it will involve some compulsion (at least for those who wish to continue watching television.)

We have supported this approach:

It offers the prospect of a far more efficient use of the spectrum;
It hastens the development of a well functioning broadcast market by eliminating many of the barriers to entry that characterised the analogue age; and
And it offers a significant increase in choice to viewers and listeners which would otherwise not be available.

Alongside this we also have a range of measures which support competition in the broadcasting market. These cover two broad areas; access to distribution and access to content or the commissioning of content. Our work in this area is very much work in progress but the approach includes:

Cross promotion rules intended to prevent the leveraging of a position in one market into a separate one;
Intervention to address the imbalance in the negotiating positions between independent producers and vertically integrated broadcasters; and
A regulatory regime for electronic programme guides, conditional access and access services to ensure that provision is made on fair, reasonable and non-discriminatory terms.

I might also mention work on premium rights, DTT capacity and minimum carriage requirements.

Finally, we have made no secret of the fact that we believe that a fair system should be applied to all broadcasters, which means including the BBC under the regime of ex ante powers in the broadcasting area.

Competition in broadcasting and the evolution towards a more open market is of course part of a wider debate. We must also consider the role of public service broadcasting.

The origins of PSB of course lie very much in the analogue age. A key economic feature of broadcasting in the analogue world is its public good characteristics. Broadcasts in this world are both non-excludable and non-rival: when a programme is broadcast to enable one person to receive it, it is available to all who have a TV set in the broadcast area (i.e. it is non-excludable), and consumption by one person does not lead to a reduction in the amount available to others (i.e. it is non-rival in consumption). This leads to a situation where the market would under provide, because there is no means to charge viewers for what they watch. Advertising funded channels are to some extent an answer to this problem, but under those arrangements there is no direct way in which viewers can express their preferences.

In the age of satellite and cable this has changed. While broadcasts are non-rival in consumption, there is the scope for excluding non-payers and for the market to reflect viewers’ preferences. This means that it is possible for a market to operate, with suppliers charging consumers according to their level of consumption. The non-rival characteristic remains, which is essentially the same as there being a zero marginal cost. This potentially leads to inefficiency, as the costs of making programmes need to be recovered. Prices need to be significantly above marginal cost to recover these costs, and this leads to the exclusion of some viewers whose valuation exceeds the marginal cost of supply. Pricing strategies of bundling and discounting ameliorate this inefficiency but it remains to some degree.

While prices which depart from marginal cost are common and are not usually regarded as a reason for intervention more weight might be given to this issue in broadcasting. This is for two reasons: First the inefficiency is potentially large, as it is equal to the total valuation of the programmes of all excluded viewers. But second, and I think more importantly, because the inefficiency arises in the form of exclusion, this might be regarded as a socially, as well as an economically, undesirable outcome.

This concern is underlined by a second feature of broadcasting – the presence of externalities in the form of benefits accruing to, or costs being borne by parties other than the producer or the consumer involved in the market transaction. To take one very obvious example, broadcasting often has an educational dimension, and this may generate benefits to parties other than the viewer of the broadcast. There are also broader benefits to society which can be generated by programmes with a strong element of information – news, current affairs and documentaries being obvious cases. Content which makes people aware of wider issues in society, the way in which a healthy democracy works and so on may lead to our society generally being one in which people prefer to live, compared with an alternative where there is no, or less, broadcasting of this type.

Quantifying this type of effect or value is no easy task and it is not one that I would claim we or anyone else has run to ground. However, there is some interesting evidence. Research carried out by the BBC and Human Capital in 2004 asked people how much they would be willing to pay to avoid being excluded from watching BBC programmes, and then how much they would be willing to pay to avoid the BBC being closed down. The first of these values is an indicator of their private valuation of BBC programme that they watch themselves, while the second includes also a broader benefit to society – how much they believe we benefit from the BBC being available to others. In the survey, the second valuation was significantly greater than the first – by 10 to 20%.

So there appear to be a significant difference in private and social valuations in this area.

Does this matter, in the sense of there being a reason for intervention? Well, individuals, when they decide what to purchase and at what price, will reflect only the first of these values in their decisions. They will not reflect the broader social value because they will believe that they will receive these benefits whether they make that additional payment or not. This is similar to the free rider problem with public goods. Thus, a subscription type service would not fully reflect the benefits generated, and there would be too little PSB type content produced. So while it is difficult to quantify the relevant factors accurately, it is clear that such effects do exist.

High quality broadcasting is also sometimes argued to be a merit good, whose value exceeds that which an individual consumer would place on it. This leads to an under-provision by the market left to its own devices and this under-provision justifies intervention to raise the level of provision of this type of programming. This type of intervention is more likely to be more effective in achieving its aim of promoting the viewing of high quality programming where viewers have restricted choice over what they view. This was the situation in the old analogue five channel world, but the situation is very different in a multi-channel world. Of course, this is a challenge for the externalities argument as well.

There is one further argument that warrants consideration. This is that the underlying economics of broadcasting might tend to lead to a concentrated industry structure in which consumer preferences are not always fully reflected in the operation of the market. In particular, in an oligopolistic broadcasting market, the most likely detrimental consequence is that producers would attempt to improve profitability by reducing costs through offering low quality but cheap to produce programmes.

Oligopolistic market structures are not easily dealt with under competition law and, in an area as important to people and society as broadcasting, there is a case for addressing this quality issue by ensuring the existence of broadcasters operating on a not for profit basis to set quality standards. This is an argument which is likely to become less persuasive as the market functions more effectively, with reductions in barriers to entry and the ability to offer a range of different products and services reflecting different viewer preferences.

Slide 5

So the arguments for public service broadcasting intervention from an economic perspective are many and varied. Our own view is that there remains an enduring case for public service broadcasting into the digital age but that, in the future, the case will rest more on the broader citizenship benefits than the conventional consumer market failure arguments of the analogue age.

How do these three strands – digital switchover, the range of competition policy and the case for PSB - come together?

Well, we can observe that one major consequence of the move to digital broadcasting and a more open market is that the historic model of commercial PSB is being irreversibly undermined.

The historic model in television has been to trade PSB obligations (such as News, regional news, children's programming and so on) in return for access to scarce analogue spectrum rights at below full market value. As the value of these analogue rights declines to zero with digital switchover, so the ability to extract PSB obligations that carry a significant opportunity cost declines.

I think this requires us to reconcile the features of the preceding arguments. That firstly, there is an enduring (if hard to quantify) case for public service broadcasting even in a fully digital age. But at the same time that we wish to see the market to develop in the direction of more competition and more choice for viewers and listeners.

This combination is likely to lead us towards greater transparency in the economic support for PSB and some fundamental questions and almost certainly revisions to the model of how PSB is delivered, under what funding arrangements, and to what scale.

We have made a contribution to that debate – arguing that if PSB is to be maintained and strengthened (the remit we were given by Parliament) then overall funding levels should be broadly in line with today’s levels and that we need to consider whether a new form of intervention (a public service publisher) might be part of a revised set of institutional arrangements for delivering the aims and purposes of PSB in the post switchover world. We look forward to a new phase of that debate next year.

Spectrum

Finally, I want to turn to spectrum. I have left it until last in this particular triumvirate because it is typically underestimated on a number of fronts:

It is a crucial raw material for both telecoms and broadcasting; and
It is arguably subject to the greatest change in the direction of policy.

These changes to spectrum policy have the scope to influence profoundly and irreversibly both telecoms and broadcasting markets.

In thinking about spectrum, it is helpful to keep in mind that its two key economic characteristics are similar to those of land.

First, like land, spectrum varies in quality and for high value types of spectrum, there is scarcity and so a key question for economic efficiency is how the asset is to be allocated among competing uses and users.

Second, again as in the case of land, unlimited access to it can lead to excessively intensive usage. In the case of land this leads to the well known phenomenon of the tragedy of the commons, whereby individuals having free access to the land seek to maximise the value of their output, but in doing so fail to maximise the overall value of output.

In the case of spectrum the users interfere with each other, so attempts to increase output by one individual reduce the value derived by others. That is, there are externalities in the use of spectrum which have to be managed to allow it to be used efficiently.

In the past the problem of scarcity has been dealt with by a process in which spectrum has been assigned to users and uses on the basis of administrative judgement. Negative externalities have been dealt with by granting exclusive rights of use for particular blocks of spectrum.

Over time some economic principles have been introduced – for example, cost benefit analysis is often used to inform the assignment process, and auctions have increasingly been used to release spectrum into the market. In addition, forms of Administrative Incentive Pricing (AIP) have been used to confront spectrum users with the opportunity costs of the spectrum they use.

But the reach of these market mechanisms has been limited until very recently. As you will know Ofcom is committed to changing this and putting market based measures at the heart of the spectrum allocation process. This has the potential to yield very large benefits by promoting more efficient use of spectrum. Many studies have shown that the potential benefits of allocating spectrum efficiently could be very large, running into tens of billions of pounds (including by Tom Hazlett, another of this years Beesley lecturers).'

The measures introduced and proposed by Ofcom include:

Removal of unnecessary restrictions on type of use – or spectrum liberalisation;
Allowing free trade in rights of spectrum use;
The use of auctions to release new spectrum into the market - a raft of which are planned for the coming months and years; and
The use of AIP to create incentives for more efficient use in cases where public policy dictates an administrative rather than market allocation (e.g. defence).

In combination these measures should result in the allocation of spectrum to its most productive use and this is the heart of our approach.

However, we also recognise that the problem of the tragedy of the commons does not apply to all parts and uses of the spectrum, and that a commons approach, rather than the granting of individual rights of use might be suitable in some circumstances. Many of you will be familiar with the explosive growth of Wi-Fi technology for example which operates in the unlicensed commons of 2.4Ghz.

Essentially, the question of whether a commons model or an individual rights of use model should be preferred comes down to the relative sizes of the negative externality of interference and the transactions costs associated with an individual rights regime. High interference externalities favours an individual rights approach, whereas high transactions costs tends to favour a commons approach.

Indeed, an individual rights approach can bring its own problems. One example is the risk of hold up - where multiple ownership of a piece of spectrum required for a particular enterprise confers on some owners the ability to delay or prevent the venture going ahead unless they are receive a large share of the benefits. This outcome has been referred to as the ‘tragedy of the anti-commons’. Another potential problem is the issue of anti-competitive hoarding which also may not be straightforward to deal with.

Thus, while the individual rights approach is suitable in most cases and is certainly at the heart of our strategy, there should not be a presumption that it is always superior to a commons approach.

Nor indeed should it be assumed that there are no circumstances in which broader public policy concerns will have a bearing upon a particular spectrum decision. We have already proposed that spectrum in VHF Band III be used to allow further DAB compatible Broadcasting Act licences to be awarded in the interests of completing the public policy objectives associated with digital audio broadcasting, better known as digital radio. Similar debates will rage in relation to the spectrum released as a result of digital switchover.

So while our view is that markets are generally superior to regulation, our twin duties mean that we must always weigh both the interests of the consumer and those of the citizen. Equally, in those cases where administrative non-market assignments have been made, it is likely to be appropriate to apply AIP to the spectrum to encourage efficient use.

Finally, technology may well begin to allow us to have our cake and eat it. It might be possible to use some parts of spectrum for both licensed and commons applications simultaneously. This might be possible where an alternative spectrum commons use does not affect the technical use of the licensed holder and yet generates consumer value.

Accordingly, we are proposing that certain types of very low power use, such Ultra Wide Broadband (UWB) should be allowed to use spectrum on an unlicensed (i.e. commons) basis, subject to their staying within a specified emissions limit. This type of hybrid approach to spectrum licensing - an individual rights based approach running side by side with an unlicensed approach - has the potential to unlock large amounts of currently unrealised economic value in the way we use the spectrum. It is also a source of enormous potential innovation.

Slide 6

Let me conclude this section by trying to bring a little of this to life. Over the next year or so Ofcom aims to release to the market spectrum in a variety of bands, notably the GSM-Dect Guard Bands and part of the L-Band. We expect a number of exciting new technologies to compete for these bands, including those for broadband wireless access and mobile multimedia as well a new applications like GSM services for a campus or business premises.

In the slightly longer term there is a large block of spectrum - 190 MHz - which will be coming available for new uses across the EU from 2008. We have argued that this spectrum – the so called 3G expansion band - should be allocated on a basis that is as technologically neutral as possible. We have also recently announced the start of our work on allocating the spectrum released from switchover - again very valuable spectrum that could be used for mobile video, HDTV, or 3G services for rural areas to name but a few.

Section III

Underlying approach and common themes

So is there anything that ties all this activity together?

In a sense we start with a range of market structures that many would argue are inherent to the nature of the industries in which we work – industries which exhibit network effects, increasing returns to scale, economies of density and so on.

Equally, we have the legacy of historic policy making and technological limitations which in their own way have tended to underpin the monopolistic or oligopolistic characteristics of many of these markets. Anything to do with the historic approach to spectrum illustrates this point - from the allocation of mobile licences and scarce broadcasting rights - and of course in fixed telecoms the incumbent was originally created by the state.

Our fundamental aim in relation to competition is to make more and more of these markets either contestable or, better still, subject to effective competition.

The nature of our intervention is guided by two principles.

First, remedies should be concerned with making markets work more effectively and, second, market failures should be tackled at their root cause - in other words as far upstream as possible, since this allows other regulation to be removed and for innovation and competition to flourish downstream.

So in spectrum the approach to assignment is market based and the approach to correcting for externalities is to define individual usage rights. This approach reduces the intervention to a minimum and targeted level and allows markets to allocate resources.

In broadcasting it involves encouraging the development of competition in the market for content - a situation in which many organisations may test their ability to create attractive services in the market rather than just a handful approved by the regulator. It means backing this increase in competition by appropriate access requirements in the distribution of such services.

In telecoms, the key market failure is the existence of enduring market power caused by barriers to entry combined with vertical integration. I have explained how we are seeking to address this problem.

In a market such as fixed telecoms and indeed many communications markets, the realistic goal is not perfect competition, but contestability or better still effective competition.

Contestability does not strictly require there to be any actual competition: the threat of entry if customers are not properly served is sufficient to discipline firms. Perfect contestability is a rare thing in real life – perhaps as rare as perfect competition – but there is no doubt that making entry easier and less costly can have a beneficial effect, both through the effect of actual entry and through the threat of entry.

Effective competitive is the idea that might best be described as a state of rivalry amongst firms, striving to win customers. It does not require the large numbers of small, price taking competitors as in the perfect competition model, but it requires competitors to behave in an active, rivalrous, customer seeking way.

Effective competition and contestability therefore both describe market conditions which serve the interests of consumers, and which are realistic goals in many of the markets in which we regulate, even those which appear relatively intractable.

Our task remains to make markets more competitive and to increase the degree of contestability and effective competition by reducing barriers to entry and focusing regulation on the root causes of bottleneck power.

It is this pursuit of effective competition in communications markets which characterises our overall approach to economic regulation (which must, of course, co-exist with our broader regulatory remit, that goes well beyond economic regulation).

Are there any fundamental tensions in our approach? I think there is certainly one which we are alive to and reflects a core judgement that we have made.

A more pure Schumpeterian approach might suggest that the bottlenecks themselves could be susceptible to the forces of creative destruction, with high economic rents attracting technological development and competitive challenge. In this analysis cost based access would of course undermine such forces.

Our judgement has been that for the fixed link access network in telecoms this is highly unlikely within a credible planning horizon. There is little if any evidence around the world to support the alternative proposition. Competing technologies may offer genuine multi-player inter-platform competition for some well established markets but fixed link access will retain a structural advantage for the foreseeable future, particularly in markets requiring higher bandwidths. So on balance, we have taken the view that consumers are better served by a tough regime for fixed access regulation which enables downstream competition, innovation, and where possible deregulation.

That is not to say that for lower bandwidth markets we do not see significant scope for the development of inter-platform competition. Freeing up key input markets such as spectrum are crucial to such developments and are illustrated already by the potential for fixed/mobile voice call substitution and also for lower bit rate broadband. Nor does it mean we are pessimistic about the contribution from existing high bandwidth inter platform competition, notably cable. Look to the US to see what a healthy rivally between DSL and cable can deliver. And, of course, the success of a policy to increase competition in the provision of DSL, alongside cable, will itself introduce an interesting dynamic in the distribution of television and radio content.

Convergence – written off as a fad a few years ago - is an important part of the argument. Convergence allows many types of delivery systems to convey services that were previously deliverable only by a single access technology.

So now voice, data and video can all be carried over a cable, satellite or various broadband networks - or a mixture of such networks. The multi-product network is gradually becoming a reality and will in due course begin to change the boundaries of market definition.

Slide 7

While this has not happened yet, the signs are that it will be a major influence in the years to come. Technology has enabled us to begin to think of a wide variety of broadband platforms – mobile, wifi, wimax, DSL, cable – all capable of distributing the same or similar voice, data, video using common TCP/IP protocols.

Mobile operators are already offering video services, Sky has entered the DSL market through its purchase of Easynet, subject to regulatory clearance, BT has entered the television market and will offer on-demand video and of course cable has long provided a triple play service.

These developments sit alongside and not in opposition to our attempt to open up enduring bottlenecks to give multiple players the opportunity to provide the triple or even quadruple play services that are likely to become the focus of consumer demand in the years ahead.

This is likely to raise a number of complex but important issues in applying the regulatory and competition rules in the communications sector. There is the prospect of large players from one market exploiting technological convergence to enter adjacent markets and offer multi-play services to consumers leading to a greater number of players in the provision of each service and an increase in competition.

The prospect of large players competing aggressively is an attractive one, and one that should be encouraged – it certainly sounds like effective competition.

But there is a word of caution to note at the same time: large players providing bundled services might make life very difficult for providers of a single service. The test of whether this is a concern will rest with the consumer. If consumers generally want to buy services as a bundle, it will be difficult to argue that suppliers should not sell bundles. But if consumers continue to want to buy significantly on a separate basis, the question arises of whether bundling strategies by large players have an exclusionary motive.

Taken in the round we see this as an opportunity for more competition that should benefit consumers, but equally we must be alive to the potential for abusive behaviour.

Section IV

Some issues for the Future

Let me conclude by putting on the table some of the issues that I think will require some close attention in the coming months and which I hope may spark a little debate this evening.

The first is the interplay between competition policy and consumer policy.

In many ways, a lot of what I have talked about today has focused on structural rather than behavioural issues of regulation. But behavioural issues are becoming more not less prevalent as we see the development of more competitive markets and as technology opens up radical new opportunities for commercial exploitation. This I think is common to other sectoral experience too.

In the text book, competition delivers the right outcomes for consumers. We never get textbook competition, especially in the communications markets. Where the consumer is ill-informed about choices available the otherwise smooth mechanisms of the market fail to deliver optimal outcomes. Equally, where consumers do not feel protected against scams and exploitation they may face direct harm and, even where a consumer is not a direct victim, the perception of an unreliable, exploitative market may itself threaten to undermine the benefits that flow from healthy competition.

This suggests that we need to consider competition policy as a set of tools that go hand-in-hand with consumer policy. Successful consumer policy will need to consider explicitly the effect on competition, and the potentially longer term effect on consumer benefit. Equally, we need to make sure that competition policy initiatives are carried out in the context of an explicitly defined consumer interest, and delivery against that interest. Indeed, if we cannot define the consumer interest in any particular policy we should consider carefully the reasons for carrying through that initiative.

Secondly, the issue of new bottlenecks or what we might call risky bottlenecks.

There is a lot of literature on how to identify and deal with long established bottlenecks – it has been the stuff of these lectures over many years. But we also know that in general regulators should not interfere with risky investments.

But how do we proceed when both these elements are present together, as they are likely to be in the next generation of broadband access networks? The question of rewards for investment mattered much less when telecoms was a utility business with a low risk premium but telcos are now facing a very different set of challenges, with the transition of PSTN to IP. This presents the cost of upgrading networks when there is uncertainty about the level of demand for the new services which would justify the upgrade. In other words a major investment which by its nature carries substantial risks. This problem is increasingly recognised but there is less agreement at present on how regulators should respond.

Finally, challenges in the international arena. I have said little so far about policy developments at international, and specifically EU level. The latter in particular is of critical importance. The broad thrust of EU activity in communications is largely consistent with the approach I have outlined in this lecture. In communications policy, we do not recognise the ideological divide that is often said to characterise our relationship with Europe. In telecoms, broadcasting and spectrum management the trend in recent years in Europe has been towards much greater liberalisation and a break from the old models of industrial and cultural policy.

There are however one or two concerns worth noting. The first is that as we review the Framework Directives we need to look at not only their effectiveness in terms of a harmonised approach to the process leading up to regulatory outcomes but also perhaps in relation to the effectiveness and timeliness of remedies. In the absence of a debate of this kind the process will remain a sitting duck for those who say that regulation in some member states is still being used to inhibit competition, particularly from other member states or from beyond Europe.

A second international issue ripe for debate is the issue of the role of harmonisation of spectrum use. The debate about how spectrum should be allocated could sometimes be characterised as one between ‘liberalisers’ on one side and ‘harmonisers’ on the other. The former group emphasises the role of the market in achieving an efficient allocation, the latter argues that there are benefits from ensuring a common international approach. These benefits comprise the lower equipment costs that result from the exploitation of economies of scale and the user benefits of interoperability across countries.

We remain of the view that flexible deployment of spectrum according to market needs is the best approach to unlock its true economic potential but recognise that harmonisation arguments have some merit. We would like to explore whether a market based approach might actually produce the kind of harmonised outcome that some of our colleagues in Europe value or whether, with a proportionate level of intervention, it could be made to do so.

Closing Remarks

I have set out in this lecture how I believe Ofcom's approach to economic regulation seeks to be engaged, informed and attempts to take a broad view of the industries and markets in which we work.

I have summarised our approach in key areas of fixed telecoms, broadcasting and spectrum. This has been by no means exhaustive (I have not really mentioned mobile at all and have said little about radio) but has tried to illustrate our focus on developing effective competition throughout the value chain, or at least downstream from enduring economic bottlenecks. I have highlighted the scope for convergence and the evolution of communications networks to increase competition and deliver benefits for consumers.

Finally, I have noted three issues which will certainly concern us in the coming months. There are no doubt many others that the audience would now like the opportunity of raising.

Thank you very much for listening.

Footnotes

1.This Beesley Lecture, like others in the series, focuses on economic regulation. Ofcom's remit is much broader than this. The range of Ofcom's overall duties are reflected in the twin duties to serve the interests of both consumers and citizens.

2.The regulation of privatised monopolies in the United Kingdom, M.E.Beesley and S.C.Littlechild, RAND Journal of Economics, Vol.20, No.3, Autumn 1989

Download PDF presentation here
If you have any trouble downloading this file please right click and use the 'Save Target As' function