Former BBC Chairman Gavyn Davies thinks the Government should leave the BBC alone, after the completion of the Charter Review process.

Mr Davies, who resigned in the aftermath of the Hutton Report, makes an economist's case for keeping the BBC in its present form.

This is the text of his Lubbock Lecture to the Said Business School in Oxford:

Economics and the BBC Charter

2004 Lecture by Gavyn Davies at the Said Business School, Oxford on 10 June

This evening I am going to ask what we can learn from economics to inform the debate on the renewal of the BBC's Charter in 2006, and in particular I will ask what economics can contribute to the basic question of why the BBC should exist at all. The truth is that we can learn a lot from economics. I know there will be resistance to this idea from many sides of the debate, perhaps especially from those most disposed to be friendly towards the BBC. They often seem to prefer arguments couched in social or cultural terms, and I understand that. But I ask them to approach tonight's lecture with an open mind, for there is much in the unappealing language of economics which can help justify the role of the BBC.

Economics, to the annoyance of many, gets everywhere nowadays. Ten years ago, Mervyn King and I agreed that there was a glass ceiling in the UK, above which economists could not penetrate. We were considered nothing more than technicians, focused only on concepts which could be measured in monetary terms. Surely no-one in their right minds would entrust a really serious job to an economist?

Yet, by the beginning of this year, professional economists occupied the Chairmanship of the BBC, the Chair of Ofcom, the Governorship of the Bank of England, and the Permanent Secretary's seat at the Treasury. Perhaps that was over-promotion of the dismal science. Alastair Campbell and Lord Hutton certainly must have thought so, and together they succeeded in reducing the roll call of economists by one - me. And while I am on the subject of the noble Lord, now sadly retired, I must commend to your attention Lord Hutton's forthcoming speech on "how to run a public inquiry". Should be good!

But back to something much more significant - the debate about the BBC Charter. Many would claim that the core case for the BBC should rest not on money and markets, but on culture and citizenship. And I would agree with them. At a recent lecture at Hertford College, Oxford, I argued that the concept of public value should lie at the heart of the BBC's charter bid. Some elements of public value - such as the price the consumer would pay to view a premiership football match - can be readily measured in monetary terms in the marketplace. But other elements - such as the value placed by society as a whole on an informed electorate - cannot be so easily valued. In fact, they probably cannot be sensibly valued at all.

This has led some observers to argue that economics is only relevant to a part of what the BBC should be expected to provide. The consumer value of the BBC, we are told, is the rightful subject matter of the economist. The citizen value, on the other hand, lies outside the purview of the economist. Ofcom, among many others, is prone to this type of assertion.

If people feel comfortable with this distinction, then I have no major objection to it, especially if it makes them eager to defend the right of the BBC to flourish in its present form. But, as a professional economist, I am slightly offended by it, since it entirely forgets that economics is not simply about what can be measured in market exchanges, but is also about all those things which cannot be so measured, but which need to be considered when designing policy. Much of the subject matter of economics concerns the problems that arise when markets are missing, or when important relationships between individuals are not captured by monetary exchanges. These causes of market failure form the core justification for public intervention in private activity, and always have done, from Adam Smith onwards. It can be frustrating for economists to listen to confident opinions being expressed by commentators who appear blissfully oblivious to the fact that two centuries of detailed economic thought has already been devoted to precisely the matters under discussion. It would surely be perverse to ignore these economic principles during the Charter debate.

In order to justify the existence of the BBC in its present form, there are at least three large questions which need to be addressed. These are:

1. Are there market failures in the private broadcasting market which imply that the market does not function efficiently, and therefore justify government intervention?
2. Are there distributional failings in the free market system, implying that the market (though possibly efficient) is unable to allocate information and entertainment satisfactorily to all citizens?
3. If public intervention in the market is justified on either of these grounds, are there alternative forms of intervention or regulation which would be preferable to the BBC?

In this lecture, I shall focus primarily on the first of these questions, since this is where the most serious intellectual attacks on the BBC are now to be found. It seems to me incontrovertible that a necessary condition for the BBC to exist is that there is a market failure in the private broadcasting market in Britain. Otherwise, why would we want to go to all the trouble of collecting a licence fee, and using it to create a massive public body like the BBC? Wouldn't it be much simpler just to leave all this to the private sector, if we believed that the free market produces an optimal out-turn? Of course it would.

In years gone by, it used to be widely accepted that there was a clear market failure in broadcasting. There was a severe shortage of spectrum, leading to a restricted choice of channels, and there was no way of charging people directly for channels or programmes individually. But now there is virtually no shortage of spectrum, and many homes have access to the encryption technology which implies that channels can be bought and sold like any other free market service. This has led many to conclude that the television market is now just like any other, and that the case for market failure has therefore dropped away.

I was chief economist at Goldman Sachs for almost two decades. I hope therefore that my free market credentials cannot be seriously questioned. Free markets are generally, indeed almost always, the first and best default option for the organization of human activity. But that does not mean that they are always and everywhere optimal. Sometimes, whatever the market zealots of New Labour might now want to believe, there are reasons for believing that markets can fail.

Is this the case in the broadcasting market? Before answering that question, I want to define more precisely what service I am discussing. For simplicity, I will today omit any commentary on BBC radio and BBCi, both of which are generally acclaimed, and rightly so. I will focus instead on the more controversial service which BBC television seeks to provide. This needs to seen as a package of channels, which are together intended to inform, educate and entertain the viewer. The BBC package is explicitly intended to reach the mass market, not the tiny niche which is reached by PBS in America. I shall call this, in order to annoy some of the BBC's detractors, the market for Reithian services.

Now we need to define more accurately what we mean by a market failure in the area of Reithian broadcasting. Market failure exists when there is an under-provision of Reithian broadcasting services under free market conditions, relative to the socially optimum level. As in any other field of micro-economics, the socially optimum level occurs when the marginal social benefit is exactly equal to the marginal social cost. If the marginal benefit is greater than the marginal cost, then more Reithian services should be produced, and vice versa.

When I was at the BBC, I noted that many of my colleagues were extremely reluctant to hinge any argument for the BBC's existence on the concept of market failure. Since this concept is so widely misunderstood and misrepresented, I could readily understand their reluctance in terms of public relations. But as a rigorous economist, their reluctance made no sense at all. Without market failure, the government should not intervene in the commercial market-place, and the BBC should be privatized. That is all there is to it.

The confusion about this arises, I believe, because people misunderstand what economists mean by market failure. The term is frequently over-interpreted. So it is vital to say explicitly what I do not mean when I use this particular term of art.

First, I do not mean that Reithian broadcasting will be entirely absent from a free market system. Of course it will not be. If the BBC did not exist, there would still be a great deal of Reithian television produced by ITV, Channel 4, the excellent Sky News and other commercial providers. But there may not be enough to attain the social optimum level.

Second, I do not mean that the BBC should produce only those programmes which are absent in the market place, while focusing solely on unique, minority content. Since practically everything is produced to some degree by the free market, such a rule would soon leave the BBC producing absolutely nothing at all.

I therefore reject the fear that the market failure case automatically implies that the BBC will become an increasingly irrelevant provider of up-market niche channels. What it does imply is that the BBC should top-up the supply of mass market Reithian broadcasting which comes from the private sector.

The subject of welfare economics has established that a free market will result in a socially-optimal level of production of any good or service under certain conditions. These conditions include perfect competition, the absence of externalities, decreasing returns to scale, no missing markets, and no asymmetries of information. If these and a few other conditions hold, we would expect the free operation of supply and demand to result in an optimal or "efficient" level of output. Economists call this a Pareto-optimal situation, and it implies that no-one in the economy can be made better off without making someone else worse off. It represents the beautiful result of the free working of the price mechanism. For most goods and services, governments should get out of the way and watch the price mechanism do its magical work.

However, this is not always the case. Slide 8 shows a stylized version of how deviations from the assumptions of welfare economics might affect the market for Reithian services. In the diagram, the private demand and supply curves intersect at Qp, which is the amount of Reithian television which would be produced in a pure commercial market. However, let us now assume that some of the benefits of such television are not paid for directly by consumers, but are valued only by society as a whole. Such benefits might be the average level of information possessed by all citizens, or the general availability of cultural programmes. If these items are generally valued, but not directly purchased by individual subscribers to TV channels, then the social demand curve for Reithian television will be to the right of the private demand curve, and the optimal quantity Qs will exceed the amount produced in a free market. In such circumstances, the role of the BBC is to fill the gap between the amount of Reithian services that would be produced by the commercial sector, and the higher quantity which is implied by the social optimum.

There are in fact four main reasons why there is market failure in the UK broadcasting market. These represent clear deviations from the assumptions which are required in welfare economics to ensure that the free market produces a socially optimum result. The reasons are:

1. Broadcasting remains a public good, which implies that it should be provided without making a charge at the point of use.
2. Broadcasting involves the creation of inter-personal relationships, called externalities, which are not fully reflected in market transactions.
3. Broadcasting involves increasing returns to scale, which tends to lead to the existence of private monopolies in a free market.
4. Broadcasting involves informational deficiencies which lead to sub-optimal levels of demand for quality products from consumers in the free market.

None of these assertions is remotely new. In fact, they are all as old as the hills. They have been widely discussed in the literature, including in previous work by myself and Andrew Graham, and also in my government report on BBC funding in 1999. I would be so bold as to defy anyone to think of any major cause of market failure which does not fit logically into one of these four headings, or at any rate into a departure from the list of the assumptions required for Pareto optimality listed earlier. Quite often, an eager new author will come up with a fresh twist on these old favourites (such as the invention of the importance of citizenship, which is simply a particular type of externality), but the novelty usually turns out to be more apparent than real. If this were not the case, then my earlier confidence about the work of market economists over the past two centuries would be quite misplaced. It hasn't happened yet.

Why therefore do I labour these causes of market failure? I labour them simply to point out that they are still around, and that they have not disappeared just because technology has gone digital. It is important to consider them each in a little more detail, in order to accept that they have not been abolished by technical change, and indeed will be with us for a very long time.

First, is broadcasting still a public good? Public goods or services have two characteristics. The first is that they are non rivalrous, which means that consumption of the service by one individual does not preclude consumption of that same good by another individual. Street lighting is an obvious example. This implies that the marginal cost of supplying the good to successive individuals is effectively zero, once the original costs of production have been incurred. Furthermore, it implies that the price charged for the service should also be zero, since any positive charge will prevent some consumers from enjoying a product which could be supplied to them for nothing. I do not see how this can be described as optimal by anyone who has studied welfare economics.

The second characteristic of public goods is that they are non excludable, which means that consumption by one individual makes it impossible to exclude any other individual from having the opportunity to consume the same benefits. National defence is a prime example. The implication of non excludability is that it is impossible to make a charge for the product, since no-one will choose to pay for a product that is freely available to all. Therefore no private market will develop for the product, and it has to be provided through the public sector.

Traditional analogue broadcasting fulfils both of these requirements completely. Once the analogue signal has been provided for a single user, there is no extra cost for providing it to everyone in the same locality, so the product is clearly non rivalrous. Furthermore, once you have provided the signal for any household, you cannot exclude all other households, so it is also non excludable. It is in fact almost the perfect textbook example of a public good. To the extent that analogue broadcasting still exists (and by far the majority of television sets in the UK are able to receive only analogue signals), then television remains an archetypal public good.

However, this is not so obviously true of digital broadcasting, which is presumably one reason why some commentators believe that the arrival of digital services undermines the case for treating broadcasting as a public good. Digital broadcasting can be made excludable, via encryption of the signal along with set-top boxes inside each home. This means that a free market in television services can develop, since people can be charged for their use of individual channels and maybe even individual programmes. However, this will not become feasible until analogue switch over, and even then most TV sets may not be equipped with the necessary boxes, since many Freeview boxes are being sold without a charging facility.

More importantly, digital broadcasting will still be non rivalrous, just like its analogue predecessor. This will never change. The marginal cost of providing a satellite digital signal to all homes is effectively zero, once the satellite has been launched. So broadcasting will remain forever a public good, in the sense that the marginal cost of provision to additional users is negligible. This is a sufficient condition to suggest that a charge at the point of use would be inefficient, for the reasons explained above.

Slide 12 compares broadcasting with other public services, based on the extent to which the relevant services exhibit the characteristics of non rivalry and non excludability. The further upwards and to the right you move in the chart, the greater the extent that the requirements of a public good are met. Analogue broadcasting is almost a pure public good, alongside national defence and street lighting. Encrypted digital broadcasting is at the top left, with the likely state of the actual broadcasting market in 2015 somewhere around the middle top. This would still leave broadcasting at the end of the next charter period exhibiting far more of the characteristics of a public good than many other services provided by the public sector - including health, education, vaccinations, the Forth road bridge, and the fire and police services. Based on this graph, it is not clear why so many commentators wish to shift the BBC outside the public sector, while leaving these other services firmly inside it.

In summary, then, broadcasting will always remain non rivalrous, so the optimal subscription charge at the point of use should be zero. Any attempt to exclude some users by levying a subscription may result in under-consumption and could be sub-optimal. And the technology required to put such a charging mechanism in place would be a waste of resources.

The demand curve for a public good is derived by adding vertically the individual demand curves for every member of society. Obviously this is impossible in practice, but it is possible to derive a national demand curve for BBC services by asking samples of the population how much they would be willing to pay for the entire package of BBC services, if they were available on subscription. Just before I left the BBC in January, we were in the process of doing this, with the research being carried out by Human Capital. Slide 15 shows the results of a preliminary study, which was only a pilot and the figures may have changed since January. Nevertheless, I believe that the results are extremely instructive.

About 17 million households, or more than three-quarters of the population in the UK, value the BBC at or above the £121/year which they are charged in the licence fee. About one third of the population would be willing to pay double the licence fee, and more than one quarter would be willing to pay treble the licence fee. Thus the vast majority of the population are gainers from the existence of the BBC, with many gaining by a very large amount.

However, around 6 million households say that they value the BBC at less than the current licence fee. Whether they would in practice refuse to pay £121/year if they actually tried to live without the BBC is not clear, but that is certainly what they say today. Of these 6 million, it is quite likely that a large number actually refuse to pay the licence fee, and are not caught in any given year. This might knock about 1.5 million off the total, leaving about 4.5 million households who actually pay the licence fee while attaching a lower value to the BBC's services than the cost of the licence fee. These are the losers from the current compulsory system of funding the BBC, though most of them do not lose by very much, since even they generally assign a positive value to BBC services.

It is right for the BBC to worry about these people, and to tailor some of their services towards addressing the problem. It does not seem to me impossible to eliminate this problem entirely by shifting BBC priorities somewhat towards those who are currently underserved, many of whom will probably be in the ethnic minorities, will be on lower incomes, and will be located outside the south east of England. Greg Dyke was very aware of this problem, and was addressing it very effectively. However, more needs to be done during the new charter.

When I made the mistake of pointing this out as BBC Chairman, I was roundly condemned as someone who favoured "dumbing down" the BBC. This was entirely incorrect, but I continue to believe that the problem should not be swept under the carpet. A disaffected minority is dangerous for the BBC and for the licence fee system. It has never been a feature of the BBC in the past, and it should remain a priority to eliminate the problem in the future.

The problem, however, must be seen in context. Those who gain from the existence of the BBC vastly outweigh those who lose, both in raw numbers and in monetary terms. I reckon that the amount gained exceeds the amount lost by well over £2 billion a year, equivalent to 0.25% of GDP. This amount of consumer surplus would be instantly lost if the BBC were closed down.

This may be what explains the extraordinary and enduring popularity of the BBC as a public service in the UK. After their punishing experiences in the past year, I do not expect New Labour politicians to try to meddle with the BBC and its services very much during the Charter debate, even though some of them might be itching to do so. The reality of politics should now prevent that. But these figures also provide a strong and enduring economic case for the BBC, whatever the incumbents of black limousines may happen to think at any moment in time.

Those who deny that broadcasting is a public good usually argue that channels and programmes should be financed in the market just like any other service. In the old days, the recommended method for privately financing the BBC was usually advertising revenue, but since the Peacock Report in the late 1980s, the fatal flaws in this option have been widely recognised. I do not think I need to rehearse these flaws today. Instead, those who believe in privatising the BBC's revenue stream usually recommend that the BBC should become a subscription service. If this were to happen, the BBC might choose to charge a licence fee which maximised its overall revenue. I reckon this might be around £170/year, at which point about 13 million households would choose to pay the new charge, and about 10 million would choose to lose the services of the BBC. Clearly, the ability of the BBC to provide universal services for the whole nation would be lost overnight - a sacrifice which I for one would not be prepared to make.

Some people would, however, probably welcome such a shift in policy. These would be the 6 million households who currently value the BBC at less than the licence fee. Everyone else would lose part of their consumer surplus if there were a shift to a subscription system. About 4 million would lose all of their consumer surplus because they would no longer subscribe for the BBC at all, while about 13 million would subscribe, but would see their consumer surplus declining because they would now by paying about £170/year for the BBC instead of the current £121/year.

Overall, there would be a large drop of about £0.5 billion in national welfare if the BBC shifted from a licence fee system to a subscription system. This could only be justified if one placed a very high weight indeed on the welfare of those 6 million households who would welcome such a change, and placed rather a low weight on the welfare of everyone else. Rather than conducting intellectual distortions to do this, I would prefer to eliminate the problem of the 6 million by addressing the underserved directly, as discussed earlier.

So the existence of the BBC adds enormously to the total welfare of the country, compared either to a situation in which the BBC were closed down, or to a situation in which the BBC were privatised and allowed to behave like any other commercial company, financing itself through subscription revenues. Many people dislike the licence fee system on political grounds, and many others seem to have persuaded themselves that changes in technology should lead inexorably to the BBC becoming a subscription based service. These people should face up directly to the large welfare losses which would immediately be incurred if their ideas came to fruition. How, I ask them, could the nation be compensated for these losses?

If we accept that the BBC continues to produce a public good which adds to national welfare, this still leaves an important question about the optimum scale of the BBC itself. Has it grown too big? One way of looking at this is to define the core television service of the BBC to be BBC1, and then ask whether the marginal benefit derived from extra services exceeds the marginal cost of providing them. If so, then maybe the BBC is too big. It is easy to measure the marginal costs of extra television services, and these are shown on a per household and per head basis in Slide 17. Essentially, it is worth providing BBC2 if the average individual is willing to value the service at just under £7 per year - not per week or per month. The values that need to be placed on BBC3, BBC4 and News24 are on average less than £1 per year for each citizen. I would make the confident assertion that average individual would place at least that amount of value on these services, noting that the cheapest Sky package, excluding sport, costs about £150 per annum. If this is indeed the case, then I would argue that the recent expansion of BBC television into the digital world has already added further to national welfare.

What about the other causes of market failure? (Slide 18.) Many types of externalities clearly exist in broadcasting. These are linkages between the welfare of individuals which are never traded in exchange for money. For example, if an individual feels better off because other people are better informed or educated, this effect will never be captured in free market prices, and programmes which contribute to information and education will be under-provided in a commercial system. These effects are well known, as are the negative externalities which are triggered when some citizens choose to access pornographic or violent programming. Less familiar are the externalities which are created when television is the catalyst for communities of interest to develop. These allow individuals to derive greater enjoyment from their enthusiasms because they are shared with many other people. As Martin Brookes has recently pointed out in an important paper, these externalities apply just as much to football and Eastenders as they do to opera and politics. They offer a clear economic justification for the BBC to be involved in popular programming.

A third form of market failure is that broadcasting tends to be an industry in which there are high fixed costs (for example, the cost of launching a satellite), and much lower variable costs. (Slide 19.) There are therefore substantial economies of scale and scope which result in marginal costs declining as the scale of the firm increases. The minimum efficient scale of a broadcaster will tend to be large, leading to a lack of competition and eventually to monopoly. Monopoly producers will over-charge for their products, and the quantity produced will be much less than the social optimum.

Those who think this is all rather far-fetched should consider the recent history of the television market in the UK. (Slide 20.) BSkyB, after knocking out BSB, grew its revenue from nothing to £3 billion in just over a decade. It is now larger than the BBC. ITV has struggled against this competition, and forces to consolidate ITV further, and then sell to a foreign buyer, are strong. According to some forecasts, Sky's free cash flow will rise several fold in the next few years, reaching well over £1 billion in 2007. (Slide 21.)

Sky is becoming a colossus compared to ITV, which now exhibits a business model which is clearly under threat. (Slide 22.) It would be surprising if ITV's advertising revenue rose at all during the current decade, since ITV's audience share has fallen by about one third over ten years. The only way in which ITV has been able to stabilise its revenue in the face of these plummeting audiences has been to increase its advertising rates substantially. But it is increasingly questionable whether advertisers will be willing to pay premium rates for access to a declining minority audience. This squeeze may pose an ever-greater threat to the ability of ITV to afford the kind of public service programming which it has been able to make in the past. Ofcom will no doubt fight strongly against these trends, but I have only limited faith in the power of outside regulators to hold back the powerful forces of the market.

In my opinion, these trends will increase the importance of the BBC as a provider of mass public service broadcasting, and as a bulwark against the possible future dominance of Sky. But even the BBC will find this a hard task. (Slide 23.) Many observers have suggested that the BBC has been over-funded in recent years, and it is true that its revenue has grown relative to ITV. But the growth of Sky has out-stripped everyone, including the BBC. While this is an eloquent testimony to the strength of Sky's consumer offering (and I yield to no-one in my admiration for Sky as a commercial endeavour), Britain needs to ask itself whether it wants Sky to be so much bigger than anyone else, including the BBC, in a few years time.

Far from expanding relative to its competition, the BBC's share of broadcasting revenue has continued to shrink in the past decade. (Slide 24.) And if the licence fee rises only in line with the RPI in the next charter period, this relative decline in the size of the BBC will accelerate. Its share of industry revenue will fall to under 20% by 2015, leaving it struggling to play its traditional role as the standard setter in British broadcasting.

The trend in the broadcasting market is towards fewer, bigger players (Slide 25), with Sky in particular being likely to exert increasing market power. Market failure on this score is increasing, not diminishing. And the BBC is shrinking in its relative size, making it harder for it to redress this market failure. Yet, in the face of all this evidence, the BBC's competitors and some politicians routinely castigate the BBC for being a gorilla of frightening scale.

I doubt if the BBC itself will have the temerity to fly in the face of this misguided tide of chatterati opinion by asking the government for a licence fee settlement in excess of inflation, for fear of being laughed out of court. But the truth is that no other public service in Britain - not the health service, not the schools, not the army and definitely not the police - would ever contemplate accepting a decade-long settlement in which its income is frozen in real terms. Why should such a miserly outcome be seen as a "good" settlement for the BBC, our consistently most successful public service throughout its eighty year life? If the BBC itself feels unable to complain about this, then its friends should certainly do so on its behalf.

The last source of market failure is informational deficiency. (Slide 26.) This is extremely familiar territory. Some goods and services need to be experienced by consumers before they can be fully appreciated. Education is normally thought to be such an "experience" good. Free market provision results in under-consumption, and therefore too little formation of human capital. This is close to the case for Reithian television, a complex set of services the benefits of which become clear only after considerable sampling and experience. The free market will not encourage such experience, which needs to be nurtured through marketing, scheduling and programme making skills. And this is hard to achieve through the attentions of a regulator like Ofcom, since it is so hard to define precisely in a contract. Hence the need for an institution like the BBC, which does all of this because it is built into its DNA.

So it seems to me clear that there are many sources of market failure for the BBC to address, and it is also clear that different genres have different degrees of market failure attached to them. (Slide 27.) Few market failures are apparent, you will be pleased to hear, in the fields of pornographic or violent conduct, and rather few in the area of imported mass entertainment. It would not be the end of the world if the BBC were absent from these genres. On the other hand, mainstream popular entertainment, produced at home, does exhibit several key elements of market failure, and there are therefore good reasons for a strong BBC presence in this genre. And of course, the most serious market failures are in the fields of news, information and education, so this is where the BBC's presence should be most pronounced. Given this pecking order of market failure, it is of considerable comfort that countless surveys of audience opinion show that the public believes that that the BBC's strengths relative to its competition correspond precisely to what this theory says is appropriate.

This is also re-assuring from the point of view of so-called distributional failures. (Slide 28.) These are not strictly in the same camp as the market inefficiencies which we have just been discussing, but they are nevertheless important further reasons for public intervention in commercial activity. If society is dissatisfied with the distributional consequences of the free market, it can legitimately decide to redistribute income and wealth, even though the required policies (taxation and public expenditure) may result in a breach of overall economic efficiency. A fairer distribution of resources, even in a less efficient economy, can sometimes be deemed to be a better overall outcome for society.

If redistribution is deemed desirable, this can be done via direct provision of specific goods and services. For example, most societies opt for a minimum compulsory of provision of education services. This direct provision usually occurs when goods are deemed to be so-called "merit goods". These goods are believed to be desirable in themselves, even if this over-rides market preferences. Reithian broadcasting, which promotes learning, the growth of human capital and good citizenship, can be argued to be a merit good which should be subject to a minimum national provision. Once again, distributional concerns are much more powerful in the traditional BBC genres than they are in genres usually more associated with the commercial sector.

Conclusions

So where does all of this leave us? (Slide 32) I would like to leave you with the following main conclusions:

1. Market failure is a necessary condition for the BBC to exist in its present form. It is not a sufficient condition, but those who hope to justify the existence of the BBC without placing market failure at the centre of their case are facing an uphill intellectual struggle.
2. Despite widespread assertions that market failure in the broadcasting market is being eliminated by technical change, the free market in the new charter period would continue to result in an under-provision of Reithian broadcasting services, relative to the social optimum. Broadcasting will remain a public good in the digital world.
3. The BBC is one way of addressing this persistent market failure, thus ensuring that an under-provision of Reithian broadcasting does not occur. It is not the only way, but it is the way that Britain has chosen, and it is probably the best way.
4. The BBC, funded by a licence fee of £121/year, creates well over £2 billion of consumer surplus, or national welfare, each year. In order to maintain its ability to contribute this amount to national welfare, the BBC should not be allowed to shrink in relative terms. This is exactly what would happen, notably relative to Sky, with a real terms freeze for the licence fee in the new charter.
5. A subscription-based BBC would exclude 10 million households from receiving the BBC, and would generate much less consumer welfare than the current licence fee system. There would be some net gainers from a switch to subscription, but they would be in a minority, and the nation as a whole would lose. Nevertheless, the needs of the minority - probably concentrated among the ethnic minorities, the lower income groups, and those outside the south east of England - are important and should be addressed by re-orienting parts of the BBC's output to take account of their particular requirements. This does not imply dumbing down.
6. While the BBC is not the only way of addressing market failure, alternative models for government intervention and regulation are at best unproven. Ofcom will be severely stretched to prevent the commercial sector from moving towards greater monopoly along with less public service broadcasting. It should focus its attention on this massive task, rather than seeking to extend its power over the BBC.
7. Market failure is in some ways much more obvious in broadcasting than it is in health and education, which are not public goods. (Slide 32.) Other public services like law and order and the road system also exhibit fewer of the characteristics of market failure than Reithian broadcasting. People should remember this before calling on the BBC to be privatised.
8. Those who argue that changes in technology have eliminated the case for the BBC are wrong.

Shortly, the new Chairman and Director General of the BBC will publish the BBC's own case for Charter Renewal. I hope and trust that it will not be expressed in the austere technical economics of this lecture. But this economist, for one, will be behind them.