A Study to Estimate the Economic Impact of Government Policies Towards Digital Television

Final Report for Radiocommunications Agency and Department for Culture, Media and Sport

January 1998
London

Project Team:

NERA

Phillipa Marks

Mark Shurmer

Ivan Viehoff

Smith System Engineering

Richard Womersley

Table of Contents

1.   INTRODUCTION

1.1. Issues and Approach

1.2. Costs and Benefits of Digital Television

1.3. Report Structure

2. THE NO SHUT DOWN SCENARIO

2.1. Introduction

2.2. Equipment Availability and Costs

2.3. Approach to Projecting DTT Take-Up

2.4. Results

2.5. Comparison with Other Studies

2.6. Prospects for Analogue Closure

3. ACCELERATING ROLL-OUT OF DTT

3.1. Introduction

3.2. Roles of Market Players in Assisting Roll-Out

3.3. Government Policy to Accelerate Roll-Out

3.4. Spectrum Auctions as a Route to Accelerate Roll-Out

3.5. Potential Impact of Early Shut down Date

3.6. Conclusions

4. OPTIONS FOR SHUTTING DOWN ANALOGUE SERVICES

4.1. Introduction

4.2. Overview

4.3. Terrestrial UHF Transmission Options

4.4. Satellite Transmission

4.5. Conclusions

5. COSTS AND BENEFITS OF ANALOGUE SHUT DOWN

5.1. Gross Costs

5.2. Costs and Benefits of Early Analogue Shut Down

5.3. Conclusions

ABBREVIATIONS AND TERMS

1. Introduction

1.1 Issues and Approach

This is the final report of a study for the RA and DCMS on the Economic Impact of Policies for Digital Television and the Closure of Analogue Transmissions. The main issues to be addressed by the study are:

• Estimation of the cost of moving all TV households from analogue to digital television reception.

• An evaluation of how Government might encourage a mixture of consumers, industry and the taxpayer to bear the cost of switching from analogue to digital transmission.

• An evaluation of how Government might deliver analogue switch-off in 5, 10, 15 years’ time. This is to include analysis of the costs of key timing issues, such as the naming of an analogue switch off date 5 or 10 years in advance.

There is a general lack of "hard" data on which to base answers to these questions. Digital terrestrial television (DTT) has not been implemented elsewhere, meaning that there is no firm data on consumer reactions, on equipment costs, on the effects of different government policies or on the practical difficulties in implementing DTT (e.g. the difference between predicted and actual coverage likely to be achieved). While there have been earlier changes in television transmission standards in the UK (e.g. the moves from 405 to 625 line and black and white to colour transmission) the Government has not before attempted to accelerate the change from one transmission standard to another by announcing a shut down date well in advance of 100% penetration of new reception equipment. Rather the transition has largely been left to market forces.

The absence of "hard" data has meant that our analysis has had to rely heavily on information and views given to us by market participants in interviews conducted in the second half of 1997. (A list of organisations interviewed is given in Appendix 1.) In instances where we have been given conflicting or different views on a particular issue, we have had to make a judgement concerning the appropriate assumption for our analysis. In general, we have been conservative in our assumptions. This means that somewhat larger cost estimates result, as compared with the situation in which a more optimistic assumption might have been made. In those instances where there has been no evidence or views on which to base assumptions, a range of plausible assumptions has been chosen.

Throughout this report we have made the following important assumptions concerning the roll-out of DTT services:

• DTT services are expected to commence in the last 4 months of 1998. For convenience we assume that the first year of DTT services is 1999. Digital satellite services are expected to start by Summer 1998 and digital cable services are likely to start around the same time.

• the expected population coverage of DTT services three years after the start of transmissions (using 81 transmitter sites) varies from around 90% for the largest multiplexes carrying the simulcast versions of the existing terrestrial services down to around 70% for the smallest commercial multiplex. In practice, it is also possible that there will be "black spots" of no DTT coverage within the projected coverage area.

• "free to air" DTT services (e.g. services provided by the BBC, ITV and Channel 4) will not be encrypted.

• Government policy will be that viewers must continue to be provided with access to BBC1, BBC2, ITV, Channel 4 and Channel 5 services free of charge. Moreover, a high proportion of the population (99.4%) must be capable of receiving them and locations where service is lost must be limited to scattered properties. In this report these are referred to as the qualifying services and the obligation on broadcasters to provide these services is referred to as the universality requirement.

• analogue services will be protected from "harmful" interference from digital services. For example, where a viewer may receive a significantly worse analogue TV picture because of interference from digital services, then the multiplex operators will be obliged to rectify the problem e.g. by installing better aerials at consumer properties.

1.2 Costs and Benefits of Digital Television

By way of background, we briefly describe the costs and benefits of DTT to each of the main interest groups or players involved in DTT, namely TV households, equipment manufacturers, multiplex operators, service providers (e.g. broadcasters, providers of subscriber management services etc.) and transmission operators.

In addition to the benefits to these parties, there is the more general benefit to society from the potential release of UHF spectrum for new services. It is thought that the uses for this spectrum might include mobile television and mobile communications services.

TV Households

The costs of DTT for households are the costs of purchasing new equipment. The benefits are the availability of an expanded range of free-to-air and pay TV services, and other data services. Although many of these services will also be available on satellite and cable systems, DTT will offer this choice to those not able to subscribe to satellite services largely because of planning restrictions or reasons of terrain or those who are not passed by cable services (currently around 50% of the population is not passed by cable services). A further advantage of DTT is that many of its services will be available free-to-air with the only prerequisite for reception being a one off capital purchase of reception equipment. In contrast even reception of free-to-air services in the cable and satellite environments requires a commitment to meet ongoing costs of equipment rental in the case of cable and of the costs of subscriber management, including provision of a smart card, in the case of satellite. The ability to avoid significant ongoing costs could prove an important driver of DTT take-up amongst households less willing to pay for TV services.

To receive DTT services, households will need to purchase either:

• a set top box (STB), to decode the digital signal back into analogue format so it can be viewed on an analogue television; or

• an integrated digital TV set (iDTV), which is likely to be both analogue and digital compatible.

This equipment will also contain software allowing encrypted signals (say for pay TV) to be decoded, and may also contain a modem allowing the consumer to use interactive services, with the telephone line as the return path. In multiple set households a separate STB or new iDTV will be needed for/to replace each TV set.

Two further pieces of new equipment may also be required. For those households with VCRs (about 80% of households) once analogue transmissions are switched off only limited functionality will be possible - that is, the viewer will generally be restricted to recording the channel currently being watched (see Appendix 2). Note that this position is the same as that currently faced by viewers of cable and satellite channels. If full functionality is required, then the household has two choices - either to purchase a STB for the VCR or to purchase a digital VCR when these become available.

Not all viewers who currently receive analogue services using a set top antenna will be able to receive digital services with this antenna. Furthermore, in some areas the frequencies selected for the transmission of DTT fall outside the frequency group used by existing analogue services in that areas. In these instances households will need to purchase new antennas. The extent of the first problem is not known, however, the second problem is thought likely to affect less than 30% of the population.

Transmission providers

DTT transmission will generate a new source of revenues (and profits) for transmission providers, although if in time analogue services are shut down there will be some offsetting loss of revenues. Because digital transmission allows more efficient use of spectrum, transmission operators could be net gainers as more services will be able to use their infrastructure.

Multiplex operators

MUX operators bear much of the risk of rolling out DTT services. They are expected to play a major role in stimulating take-up of services both through advertising and marketing initiatives and, in the case of BDB, subsidising STBs purchased by pay TV subscribers. The MUX operators are also all major programme providers and as such can expect to benefit from the increased demand for programming generated by DTT.

Service providers

DTT offers service providers a new outlet for their services. Television and related services will be funded either through consumer subscriptions or advertising. Many of these services will also be distributed by DSAT and cable systems. DTT is also likely to provide a new cost effective outlet for many retailing services (e.g. home banking, home shopping, booking tickets for entertainment services). The return path for placing orders is likely to be through a modem connection to the telephony network.

Manufacturers

Transmission and consumer electronics equipment manufacturers benefit from the introduction of DTT, because of the additional equipment that will need to be purchased initially and because DTT may stimulate an acceleration in consumer equipment purchases. This is, of course, a cost to consumers. However, analogue equipment sales will be lost.

There will be a longer term benefit to manufacturers only if there are higher profits in DTT as opposed to analogue equipment. There may be a one-off benefit if the equipment replacement cycle speeds up during transition.

Many of these costs and benefits are offsetting in the sense that they represent transfers from one group to another. In Section 5, we focus on the net additional benefits and quantify the effects of different shut down options.

1.3 Report Structure

The remainder of this report is structured as follows:

In Section 2, we present forecasts of the projected take-up of DTT assuming there is no government intervention to accelerate the shut down of analogue services (including no announced shut down date).

In Section 3, possible ways in which market players and government might accelerate the take-up of DTT are discussed.

In Section 4, we assess possible approaches that might be used to meet universal service requirements while at the same time shutting the transmission of the existing analogue services.

In Section 5, we assess the costs and benefits of shutting down analogue services in 5, 10 and 15 years time.

More detailed supporting material is provided in the following Appendices (presented in a separate volume):

Appendix 1 contains the list of organisations interviewed.

Appendix 2 provides a description of issues concerning DTT and the use of VCRs.

Appendix 3 provides the assumptions underlying our DTT take-up projections.

Appendix 4 provides a discussion of policy issues concerning the provision of qualifying DTT services on a universal basis.

Appendix 5 contains a description of the technical characteristics of different delivery media for digital television services and discusses technical aspects of different delivery media for universal provision of qualifying DTT services.

Appendix 6 discusses the role of market players in the provision of DTT services.

Appendix 7 contains analysis of approaches that might be used to extend terrestrial coverage of DTT services. Issues raised by the clearance of broadcasting spectrum for new uses are also discussed.

Appendix 8 provides a description of the US approach to implementing DTT.

Contents

2. The No Shut Down Scenario

2.1 Introduction

This section outlines our projections for the take-up of DTT over the next fifteen years in the absence of any announcement date for the closure of analogue services. There is a general lack of "hard" data upon which to base such projections. DTT has not been implemented elsewhere, meaning that there is no firm data on consumer reactions, on equipment costs or on the practical difficulties of implementing DTT (e.g. the difference between predicted and actual coverage likely to be achieved). To deal with these uncertainties we have adopted a scenario-based approach describing likely upper and lower bounds to consumer take-up. The scenarios draw on the experience of the transition from black and white to colour transmissions where appropriate. The resulting analysis provides a counterfactual against which the impact of an accelerated analogue shut down date can be measured.

The remainder of this section is structured as follows.  Section 2.2 sets out the key assumptions concerning the availability and cost of DTT reception equipment. Section 2.3 provides a brief overview of our approach to modelling the demand for DTT. Section 2.4 presents our projections of the take-up of DTT, while Section 2.5 compares our results against other projections. Finally Section 2.6 considers the implications of our results for the prospects for analogue closure. A fuller account of our approach, together with details of the key assumptions and data used are provided in Appendix 3 to this report.

2.2 Equipment Availability and Costs

In the analysis given below we have made the following assumptions concerning the availability and costs of DTT reception equipment:

• iDTVs will not be generally available across all screen sizes and price ranges until the year 2000 when they will enter the market at a price premium of around £275 compared with their analogue equivalents. It is assumed that this price premium will fall to about £100 after 5 years and then fall to a mature market price comparable to that of analogue sets by year 10. This is a conservative assumption as some sources believe that the price differential between analogue and digital sets will be negligible in 5 years’ time.

• STBs will be sold at an unsubsidised price of £400 initially, although after subsidy will retail at £200. The subsidy is assumed to be paid by BDB only to those consumers who subscribe to BDB’s services. After 5 years the unsubsidised STB price is expected to fall to about £200. It is possible that the price could fall to £100 - or less, comparable to that of the current analogue satellite STBs, by year 10.

We assume that when digital equipment prices are about the same as those for analogue equipment i.e. in year 10, consumers will purchase only digital equipment.

2.3 Approach to Projecting DTT Take-Up

Figure 2.1 presents an overview of our approach to projecting the take-up of DTT.

The starting point is to segment potential digital purchasers into two broad groups based on the value placed on the reception of multichannel TV (which we consider to be the main driver of digital TV take-up) as indicated by the willingness to pay for TV services. In order to segment the market we have pursued two approaches: bottom-up, focusing on pay TV penetration rates; and top-down, analysing the growth in consumer expenditures on pay TV and related services. Reconciling these two approaches, we project that the penetration of pay TV could reach in excess of 40% of television households (TVHH) by 2007 and that subscription and pay per view revenues could total in excess of £2.5 billion.

For those households placing a high value on multi-channel TV reception, the decision to adopt digital TV is likely to be an active one involving a choice between the digital cable, satellite and terrestrial platforms. In the short term, the choice of DTT may involve the purchase of an STB. In the longer term most purchases will be of iDTVs. For the remaining households, perhaps up to 60% of TV households given our analysis, digital TV is likely to remain of little interest no matter what the market developments. For these households the switch to digital may only occur as part of the normal television set replacement cycle, in which case these households will only purchase iDTVs.

We have developed two scenarios describing likely upper and lower bounds to consumer take-up of DTT. These scenarios differ in terms of the assumptions made about the rate of digital take-up across the two market segments. Potential pay TV subscribers who elect to take DTT will comprise:

• households willing to pay for multi-channel TV but currently unable to receive cable and satellite signals. In both scenarios we assume that the take-up of DTT across such households over time approaches that achieved in the market as a whole;

• a small number of existing analogue cable and satellite subscribers who elect either to adopt additional DTT capabilities when upgrading to digital cable or satellite or to convert to DTT. In the absence of further information we have assumed that this number amounts to 5% of analogue convertees in the Lower Scenario and 10% in the Upper Scenario;

• new pay TV subscribers. Again there is little firm data upon which to predict consumers’ adoption decisions between the alternative digital TV platforms. Given that DTT is likely to have a greater range of free-to-air services and lower subscription costs than digital satellite or cable and assuming the new subscribers have a lower propensity to pay for services than existing subscribers, we expect that DTT will compete effectively against the other digital delivery platforms. We have conservatively assumed that 33 per cent of new purchasers choose a DTT box in the lower case scenario and 50 per cent in the upper case scenario.

For the remainder of the population, the main driver of DTT take-up is the set replacement cycle. In each scenario we assume that the average time between set replacements remains at the current level of eight years. It is important to note that the use of an average replacement time is likely to lead to an underestimation of the time taken to replace all analogue sets with digital ones. In practice some households will replace their main sets less frequently than every eight years. Ideally the modelling would have been based on a distribution of set replacement times, however, such data is not available.

The difference between the two scenarios lies in the growth in the number of digital sales relative to total receiver sales. Figure 2.2 sets out our assumptions which are based on analogue technology analogy with the earlier colour television experience (see Appendix 3). The growth path of main colour sets relative to total replacement TV sales falls mid-way between our assumed digital sales profiles for the Upper and Lower Scenarios.

2.4 Results

Table 2.1, below, presents the projected take-up of DTT across the different hardware markets for the two scenarios over the fifteen year period ending 2013. The figures net out multiple STB/iDTV sales to focus on the number of households that have at least one means of DTT reception. These projections are consistent with the experience of take-up of colour TV, in the sense that the projections lie either side of the colour TV experience.

It is important to note that the result that 100% of households will have a digital means of reception by 2013 in part arises from our use of an average set replacement time, rather than the actual distribution of set replacement times. In reality, there will be a small number of households who replace their TV set very infrequently and who will not have a means of digital reception by 2013.

2.5 Comparison with Other Studies

A number of other organisations have developed projections of the likely take-up of DTT, especially in the run-up to the award of DTT licences earlier this year. The results of these studies provide a useful "reality check" against which our projections can be compared. Table 2.2 illustrates the various projections for DTT households in 2002 and 2007 (or for the nearest dates where these were not reported).

The process of comparison is complicated because of differences in the time period considered and the way in which results are presented. It is not always clear, for example, whether the figures take into account digital set replacement sales in addition to pay TV subscribers. The studies which produce estimates towards the lower end of the range appear to focus largely on the take-up of DTT subscription services. Leaving this caveat aside, analysis of Table 2.2 suggests that our projections are within the range of the other studies.

2.6 Prospects for Analogue Closure

Our analysis suggests that the take-up of DTT by the general population will be a gradual process driven largely by the set replacement cycle. Even in the upper bound scenario 100% penetration of DTT reception equipment is not reached until 2013. This would suggest that an earlier closure date for analogue transmissions is unlikely unless methods can be found to accelerate the take-up of DTT.

While our estimates assume that DTT transmission infrastructure is built out to provide universal coverage, in practice the current plans provide for only 90% coverage for the largest multiplex and around 70% for the smallest. If universal coverage of qualifying services is to be maintained, then some method of extending coverage either by terrestrial or perhaps alternative delivery platforms will need to be found. The issues of accelerating DTT take-up and of achieving universal coverage are considered in the next two sections.

Contents

3. Accelerating Roll-Out of DTT

3.1 Introduction

This section examines the prospects for accelerating roll-out of DTT.

In Section 3.2, we examine what the various market players are already committed, or appear likely, to do to assist.

In Section 3.3, we examine a variety of government initiatives which might accelerate roll-out to greater than what the market might provide, within the current spectrum management framework.

In Section 3.4, we examine the potential for a new form of spectrum management under which it is possible to auction spectrum to future users, so that they might pay to clear out the existing users.

In Section 3.5, we consider the potential impact of announcing an early shut down date for analogue transmission.

Finally, we draw some conclusions in Section 3.6.

3.2 Roles of Market Players in Assisting Roll-Out

BDB

BDB has a greater commercial interest in DTT than any other single party. It is likely to be the main party marketing DTT to the customer. BDB’s prospects are limited if DTT becomes only a third route for delivery of pay-TV by purchase of special equipment, in more or less equal competition with cable and satellite. Rather its success depends upon DTT becoming the main mode of transmission for free television, so that pay-TV delivered by DTT is more accessible to the general customer than cable and satellite. Thus BDB’s growth is likely to depend upon the period when general customers are replacing their analogue TVs with iDTVs.

However, in the initial phase of the market, DTT will mainly be taken up by customers seeking pay-TV, perhaps assisted by the fact that the existing terrestrial broadcasters will also be available by DTT. It is perhaps unfortunate in this context that BDB has the multiplexes with the poorest coverage, as it implies poor sales in areas where only the high coverage multiplexes are available.

BDB’s main efforts to assist roll-out comprise the following:

Subsidy: Its plans are not finalised, but the favoured option is to offer cash back to customers who purchase digital set top boxes and take out a subscription to BDB.

Digital Multiplex Group: BDB is a leading player in this grouping for co-ordinating the efforts of all the multiplex providers.

Technical Assistance: BDB is providing technical assistance to equipment manufacturers, thus saving them research costs.

BDB’s main alternative to subsidy to the customer is direct subsidy to the manufacturer. Whilst subsidising the manufacturer is attractive because it has the potential to be multiplied up, by reducing the gross amount of the retail mark-up and VAT, in practice electronic goods are sold at the market price and the subsidy may be absorbed in increased retailer mark-up. Further, it is difficult to avoid distorting competition between manufacturers.

BDB is concerned about the low coverage of its multiplexes. This is both in terms of the fact that certain major economic areas are not planned to be covered at all, and also that coverage may turn out to be patchy in areas which are thought will be covered. It is willing to invest money in extending the coverage area to major economic areas where coverage is imperfect or unavailable, and is considering multi-point video distribution systems (MVDS) as a possible technology if DTT frequencies are not available, but MVDS frequencies are available.

Other multiplex operators

The BBC is committed to DTT and is making every arrangement to have a service ready. Cross-promotion will be made from existing terrestrial services. Likewise, ITV and C4 have committed to one minute of cross promotion per day for DTT services.

Pay-TV services from providers such as SDN and Channel 4 may make contributions to BDB’s costs, e.g., subscriber subsidy, conditional access costs, to reflect the relative degree of benefit they obtain.

The BBC and SDN are committed to putting their DTT services on D-SAT. Whilst this may appear to be in competition with DTT, in practice it assists in making free-to-air DTT services available in areas where DTT does not reach, thus facilitating universality objectives.

Major retailers

Major retailers are committed to marketing DTT. However retailers expect analogue TV to form the main market for a period, and will be just as pleased to sell satellite systems. They expect D-SAT to enter the market before DTT. So DTT must be able to stand on its merits against the alternatives. The advantage of DTT to the retailer is that it offers the potential to sell an entire new TV receiver to the customer, not just a STB. However retailers are concerned that confused messages may leave customers unwilling to make their normal replacement purchase with any type of receiver, and similarly damage the market for analogue VCRs.

We understand that major retailers have placed orders with manufacturers for iDTV sets to be ready at the same time as STBs, i.e., shortly before DTT broadcasts start. Manufacturers should be able to comply with this, at worst by putting a STB and receiver in the same cabinet. We understand that retailers have found sources willing to supply receivers which include an integrated conditional access system (CAS), although this contradicts the BREMA position that major manufacturers will not supply such sets. An integrated CAS is a rather cheaper solution than manufacturing a plug-in CAS module, which will therefore assist in keeping the cost down to the consumer. A plug-in CAS module also raises greater concerns about pirating.

Manufacturers

Good products from manufacturers will be critical to getting DTT into the market, and manufacturers have advanced plans to supply products, even though the final specification is not yet known. Early iDTVs will almost certainly all be dual standard DTT/analogue compatible, as this is a necessary condition to maintain the functionality of VCRs and retain access to analogue transmissions, given the partial coverage of DTT multiplexes.

3.3 Government Policy to Accelerate Roll-Out

Announcement of analogue shut-down date

The announcement of a credible analogue shut-down date should do much to improve the credibility of DTT in the UK market. All interested parties are in favour of it, and moreover think it the single most important thing the government could do, short of spending money or reducing tax to assist sale or rental of DTT equipment. Moreover, they would wish to hear an announcement as soon as possible, not delaying to a point in time when the government would be forced to make an announcement. The reason for this is that an announcement makes the purchase of an iDTV as a replacement set for an analogue set credible. As noted above, it is a process of making iDTVs the replacement set of choice which will drive conversion to DTT.

BDB considers that the cause of DTT will be helped by any announcement that gives a feeling of confidence in analogue switch-off and broad timescale, even if it is not completely definitive as to mechanism and timetable. However, excessively hedged or conditional statements could be damaging if they lead to confusion. Confusion would arise if a statement suggested that the government has left itself open to be persuaded to change its mind, or to make a long delay. The current statement about re-examination at a later date does nothing to instill confidence in analogue switch-off.

The US provides an example of how not to do it. The Federal Communications Commission made an early announcement of switch-off in 9 years, without obtaining political approval. They have subsequently been forced to make switch-off subject to conditions which are plainly at risk of not being fulfilled, thus considerably diluting confidence that the timetable will be adhered to. (See Appendix 8.)

Warning stickers on analogue TV sets

A requirement could be made to label analogue televisions with a warning to the effect that it is an analogue device which will need additional equipment to continue functioning beyond a certain point. This, it would be hoped, would encourage take-up of digital television. An alternative policy would be a requirement to provide advice at point of sale in some form, verbal or written.

This suggestion has met with a mixed reaction from interest groups. Comments made include:

This risks damaging the market for all television receivers, as there is a valid market for analogue televisions for an extended period, given that STBs are likely to become quite cheap by switch-off date, and customers will therefore not incur large financial losses if they still have an analogue TV at switch-off.

It is preferable to sell iDTVs through their positive features, rather than negative marketing of analogue receivers, as this will depress the market. If "digital" can become a positive brand, then this might be more effective than negative marketing of analogue.

At some point in time it becomes irresponsible not to provide a clear warning when selling an analogue set. Whilst retailers with a reputation to keep are likely to take care to give the customer good advice, the role of reputation in other parts of the market, given the infrequency of purchases, is limited. A substantial number of customers are therefore at risk of not having the relevant information drawn to their attention unless there is a policy of compulsion.

We think the last of these three points has considerable force. But taking into account the other points, it suggests that there is some art in deciding the correct timing. It is probably not appropriate to start such a policy until the point when iDTVs are becoming the replacement set of choice, since an earlier application of the policy would indeed have the risk of depressing the entire market, and having a negative political reaction. Further thought should be given nearer the time as to how to ensure the information is undistorted, whether there should be a sticker, or a requirement to give advice in another form, e.g., verbal or written.

Ban the sale of analogue sets after a certain date

At some point, one might ban the sale of analogue sets. This would be an unfair policy for people for whom analogue sets are adequate, unless digital sets came to be available at little or no premium. People requiring analogue sets would be forced to buy second hand, or import sets, which would increase the cost. It would moreover probably breach Community law unless there were a Community Directive requiring the promotion of conversion to digital.

We consider this to be an inferior policy to providing proper information on the risk of buying an analogue set.

Require dual mode digital/analogue sets to be sold

This is not necessary, as it appears the market will provide these by necessity so long as analogue VCRs are in common use and DTT coverage is patchy.

Allow degradation of the analogue service

We consider here the use of policies which have the effect of degrading analogue transmissions as a by product. We do not consider polices, such as deliberately adding snow to analogue transmissions, which would be difficult to justify politically.

Analogue service will be marginally degraded in some ways by the digital switch-on. For example, small numbers (<1%) of households will experience noticeable interference, and may have to have aerials repositioned or replaced at the expense of the multiplex operator in order to restore the service quality to that previously available. The fact that this is considered to be tolerable suggests that there are further valuable opportunities for extending DTT if the level of disturbance is sufficiently small.

More frequencies or higher powers might be supplied to DTT, thus increasing coverage by:

Allowing old analogue transmitters to fail, particularly if the size of the population losing an analogue service, as a consequence, is small;

The interference limits might be changed, allowing higher levels of interference to existing analogue signals by DTT;

Certain relays might be deliberately switched off, particularly if the size of the population losing an analogue service, as a consequence, is small.

It appears likely that there are many opportunities to make an increase in DTT coverage which is large in comparison to the loss in quality or loss in coverage experienced by some viewers. Perhaps free STBs might be provided where analogue coverage is lost - the willingness of the broadcasters to fund this would indicate whether the change was economic.

As more transmitters or higher powers are authorised to DTT multiplexes, it is likely that the broadcaster will obtain effective control of the spectrum occupied for the long term. Given that it is intended that substantial spectrum should be released for other services, then it is necessary to evaluate, in each case, whether an extension of the spectrum controlled by the DTT broadcaster effectively reduces the amount of spectrum which might be freed for new services. If so, then an informed decision must be made as to whether this is consistent with policy aims.

Allow lock-in in return for subsidised equipment

In practice the government has little alternative but to allow this. Commercial broadcasters are not willing to subsidise equipment for purchasers who do not intend to subscribe. Nor is it yet evident that the value of the franchise which they have been granted is so large that they can afford to act with such public largesse. OFTEL has limited the period of subscription lock-in in other cases, and policy in respect of DTT is not yet determined.

Make Multi-point Video Distribution System spectrum available to DTT broadcasters

The multiplexes with lower coverage, in particular those controlled by BDB, fail to extend to some economically important areas. Moreover there is a risk that coverage may turn out to be patchy in some economically important areas which are currently intended to be covered. The patchy coverage of DTT is a factor which makes it difficult to market as a universal service. Also it would be valuable to other multiplex operators if the key driver of the marketing, BDB, were able to market over a wide geographic area.

BDB has expressed an interest in using MVDS to extend coverage to economically important areas which fail to receive DTT. The most recently licensed cable franchisees have the right to use MVDS in their franchise areas, though none has yet requested or been assigned frequencies except for testing purposes. If BDB and/or other DTT licensees are willing to invest in MVDS to extend coverage, this is a highly spectrum-efficient option which could be highly advantageous to the government, given that it reduces the long-term demand for UHF spectrum, and transfers demand to less congested bands.

Financial assistance or tax privileges

Financial measures could certainly assist in roll-out, but at a cost to the government. The government may find it attractive to give some financial stimulus to DTT, particularly if it can be done relatively cheaply and in a targeted manner. We also consider in this section assistance to areas which will not experience a long simulcast period. We consider a few options, but further work would be required to be done to make any firm recommendation.

Assistance to rentals

There is precedent for the use of tax breaks to assist in a broadcasting change-over. During the switch-over from 405 to 625 line TV, the government gave tax breaks to TV rental companies to rent out 625 line receivers by way of an increased capital allowance against corporation tax. At that time, TVs cost a much higher proportion of the household budget, were less reliable, fewer households had access to credit, and many households, especially lower income, chose to rent TVs. This tax break assisted roll out of the technology though there is no way of knowing by how much. Renting was also common when there was rapid growth in VCR use for similar reasons. Reportedly rentals played a role in early roll-out of satellite television; since the customer has to pay a subscription for programming, an additional charge for equipment rental is easy to arrange. But few new subscriptions are based on rental, now that equipment is cheap.

Renting of TVs and VCRs has faded rapidly since the tax break ceased in 1984. It is difficult to know to what extent this is due to the loss of the tax break, or to the fact that simultaneously equipment became cheaper and more reliable, incomes rose and access to credit improved. Some anticipate that renting will become more important again with the advent of DTT, because of the higher price of an iDTV, uncertainty over the reliability of the early equipment, and also as a mechanism for trying the product out without committing to it. This is plausible, but to what extent renting will obtain a much higher market share than the current 15%, with or without a tax break, is speculation.

VAT reduction

News sources reports that the Italian Government recently passed a law providing for a charge of only 4% VAT on DTT equipment. We believe that a similar measure in the UK would be poor value for money. It will result in a substantial loss of government income, give the greatest benefit to those who can afford expensive equipment, and may result in higher retail margins.

Free equipment or vouchers

The cheapest way of allowing a customer to convert to DTT at the government’s expense is to issue a free STB, or give them a voucher which can be used to obtain a free set top box of approved type, or alternatively a voucher with a monetary value to be set against approved equipment, whether STB or iDTV.

Incentive issues arise (with households deliberately waiting for government support) and also possibly fraud problems (with households claiming multiple entitlements). Plainly if all households were issued with free equipment or the means to buy it, then it becomes possible to switch off analogue TV very quickly. However it appears more likely that value for money would be obtained if such measures were targeted at relatively small areas where assistance was most required.

UK experience of targeted capital assistance on the basis of need, (e.g., Social Fund) has not been encouraging: administrative costs are high (e.g., 40%), and targeting has been at best variable. Subsidies for capital schemes (e.g., assistance with house insulation) generally obtain higher take-up with higher income groups better able to pay their contribution. It therefore seems unlikely that a means-tested scheme, or one targeted at households who have not already purchased equipment, would meet objectives efficiently.

Any practical scheme for assistance is likely to be a blanket scheme in specified areas, e.g., a single STB or voucher for all households with a TV licence in a specified Postcode area known to meet specified criteria. The areas most likely to need assistance are those where there is no simulcast period (or a short simulcast period), or where DTT is unavailable and other technologies are used to reach households which may be more expensive than DTT.

3.4 Spectrum Auctions as a Route to Accelerate Roll-Out

One option which has been raised is that the broadcasting spectrum should be auctioned to one or more spectrum "landlords" with sitting broadcasting "tenants". The new landlords would be permitted to find their own ways to release spectrum for new purposes, whether more broadcasting or other purposes, while respecting the need to continue delivering a defined minimum free-to-air television service to a sufficient proportion of the population. In particular, the spectrum "landlord" would be able to make the decisions on how to reach fringe viewers purely on economic grounds, taking into account the alternative use of the spectrum.

The economic argument for such a solution is that the spectrum is exceedingly valuable in alternative uses, and the costs of transferring customers to more spectrum efficient reception methods (e.g., give everyone who needs it a set top box) may be rather smaller than the value of the spectrum released. A private spectrum landlord may be better placed to make the trade-offs and deliver an efficient broadcasting solution than the public sector forced to respond to political considerations. Furthermore, a private spectrum owner might be willing to make investments in order to release greater amounts of spectrum. This may give them the wherewithal and the incentive to release the spectrum more quickly than an institutional solution.

This is different from what is proposed in the US. There, the plan is merely that spectrum will be auctioned off after it has been freed. This is much less radical than auctioning it in advance of it being freed, so that the purchaser has to make the changeover arrangements.

The essential elements of this solution are as follows:

• A new form of spectrum licence which provides for the concept of landlord and tenant, and which gives the landlord sufficient flexibility on the range of uses to which the spectrum may be put.

• CEPT/ITU co-ordination permitting a wider range of uses for existing analogue broadcasting spectrum.

• A clear definition of the minimum service obligations for broadcast customers.

• Competition policy decisions on the identity of landlord and tenant, and level of aggregation permitted.

These are discussed below.

A new form of spectrum licence

Current licence forms in existing and anticipated legislation do not admit the concept of a spectrum landlord and a spectrum tenant. Nor do they provide for spectrum to be allocated to a sufficient broad range of uses to give the landlord the ability to refarm to new uses, even within similar categories of use. Complete flexibility in allocation is not necessary nor desirable given the constraints of international spectrum co-ordination. This is because one of the most important factors adding value to spectrum licences is the existence of international co-ordinated bands. However, in order to refarm the spectrum, the licensee would need to be able to vary the use between the range of primary uses (which often allows some flexibility) and possibly secondary uses, allocated at international levels.

New radiocommunications legislation could be brought in within about four or five years to facilitate this new form of spectrum licensing. This is a complex area and sufficient time must be allowed for the ideas to be developed (about one year) and consultation with users (about a year), as well as for drafting and passing the legislation. In New Zealand, where legislative processes are generally much faster than in the UK, legislation of this nature took 2 to 3 years to come to fruition, and a longer period must be expected in the UK. Once passed, ideally some experience of it would be had in less economically valuable bands first so that any legal uncertainties can be clarified and confidence gained with the new system. Uncertainty over the level of rights granted would be a major concern of bidders in assessing the amount to bid, or the methods available to them for management.

European and International co-ordination

The potential uses the released spectrum could be put to are currently constrained by International Telecommunications Union (ITU) Radio Regulations. If spectrum were sold before this were clarified, it is likely that the value would be considerably depressed because of the political risk. It is unlikely to be sensible to auction the spectrum before the possible new uses were clarified. Likely time-scales for coming to agreement are discussed in Appendix 7. There we suggest that it will be several years before European policy becomes clear and that it would take at least a further 4 years to change the ITU radio regulations.

Minimum service obligations

In order for this radical idea to work, the level of broadcasting service to which the population had a right would have to be very clearly specified. A large proportion of the costs and spectrum occupancy of UHF broadcasting are attributable to delivery to the fringe areas comprising only 5-10% of viewers. A spectrum landlord would be tempted to release spectrum by reducing service in these communities without providing an alternative, or by attempting to find an alternative service that fell short of aspirations, or caused the recipients to incur increased costs. The interpretation applied to minimum service requirements could make an important difference to the ability to release spectrum for other services and to the cost that must be incurred.

The minimum service obligation could be placed on the broadcasters or on the landlord. At one level, it makes more sense to place it on the broadcasters as they are closer to the problem. In this case, the landlord would compensate the broadcasters for the costs of retaining their obligation given the changes made by the landlord.

It would be important to ensure that the minimum service specification was such that sufficient spectrum be released in order that the costs of providing the solution were much less than the value of the new uses permitted. For example, a minimum service requirement that all DTT multiplexes should be as widely available since current analogue coverage would leave little spectrum left, whereas a terrestrial solution restricting minimum service to the qualifying services in digital form would release much more spectrum, particularly if the qualifying services could be repackaged into, say, two multiplexes.

The nature of the solution available is not entirely in the landlord’s hands. For example, if satellite delivery is to be an arm or a major plank of the solution, then it must be made feasible by providing a route by which the minimum acceptable service is placed on satellite. The ITC, as licensor of the digital multiplexes, may be in a much better position to require the multiplex licensees to make their qualifying services available through additional media. However a spectrum landlord would be in a weak bargaining position when it came to compensating the last multiplex to put its qualifying services on additional media in a way which precisely satisfied ITC rules, thus enabling the spectrum landlord to clear the spectrum.

Competition considerations

It is probably unthinkable to allow a single buyer to obtain control of the entire released analogue broadcasting spectrum. Competition rules as to the level of aggregation of spectrum are paramount to safeguard future competition in new markets, but also affect the likely value of the spectrum. The difficulties this imposes are developed in the analysis below.

Analysis

The paramount importance of the minimum service obligations makes it difficult to sell the spectrum to several different landlords, because responsibility for a clear obligation becomes difficult to enforce on separate parties. For example, in any given area there could be up to five different spectrum landlords each holding spectrum for one of the five qualifying analogue TV channels. In practice, if any one landlord pays the full cost for clearing one channel, e.g., by distributing STBs, then all five channels will have been cleared by that action. It is unlikely that any single landlord would be so magnanimous, but rather would seek contributions from the other four. Given the ability of any one of several to refuse to co-operate, this bargaining situation is unstable. Thus the cost of the obligation that the landlord has taken on is uncertain, because he has no well-defined responsibility for the share of the costs which are to the common benefit of all the local landlords.

Given this situation, it seems unlikely that anyone would be prepared to bid very much for a share of the spectrum, unless their share of the clearance costs were clear. Indeed, it would appear necessary for them to create a single organisation for organising clearance, whose costs would be shared in fixed proportions. It is only a small step further to suggest that risk be removed more fully by the government mandating the existence of such an organisation, and stating how the costs are to be shared.

In essence, this is a "contracted out" solution to clearance, rather than devolving it to diverse spectrum landlords. The government would obtain competitive bids for clearance solutions from competing project management teams and recover the costs of the selected contractor from the proceeds of the auction. Whilst this is more interventionary than a fully privatised arrangement, it has the merit of providing a simple and clear mechanism by which new entrants to the spectrum pay the major refarming costs to beneficiaries of the existing system, and removes the free-rider problem of sharing the obligations in an ill-defined way among several parties. This may therefore enable a higher value and more rapid solution than a system which attempts to minimise side-payments.

We suggested above that the landlords may have the power to change the timescale of the switch-off and devise solutions for affected parties. We have also concluded that it is desirable for the government to announce the timescale of switch-off as soon as possible, and probably well before it becomes practical to auction the spectrum. Thus, in practice, the spectrum would be auctioned with a timescale already in place. So, in practice, this freedom is not available.

Thus we conclude that for the spectrum auctions there is little to be gained from doing much different from auctioning the spectrum after it has in effect been cleared (this is similar to the US approach). The possible difference that adds value is to use part of the proceeds of the auctions to pay clearance costs to a clearance contractor. With the knowledge that funds are likely to be available to pay clearance costs, e.g., compensating households, it may become politically possible to deliver switch-off more quickly and increase economic value of the spectrum more quickly than with traditional methods.

3.5 Potential Impact of Early Shut down Date

All of the interested parties we spoke to thought that Government commitment to an announced shut down date would assist the development of DTT. Specifically, it was thought likely that this would provide an incentive for retailers and multiplex operators to place large equipment orders and it would mean that manufacturers would reduce equipment prices faster than otherwise would be the case.

We have not been able to find an objective means of quantifying the effect of an early shut down date. The transition from analogue to digital mobile telephony does not provide any useful lessons. Reasons for this are that despite an announced shut down date for analogue services in the UK of 2005, operators have incentives to continue to maximise the use of the analogue network until near the end date. Operators are providing low cost analogue services as a means of competing with DCS1800 services offered by Orange and One2One, while not losing revenues from customers paying higher tariffs on the digital service. Thus the analogue/digital distinction provides a means for operators to segment the market. This situation does not apply in the case of DTT, because terrestrial pay TV services are not available in analogue format and such segmentation is not possible for advertiser financed services.

We are not aware of any relevant experience in terrestrial TV. Previous technology switches, such as the transition from 405 to 625 line transmissions, have been allowed to occur naturally without any closure date being announced in advance. However, the Dutch satellite TV market provides a recent example of an analogue to digital conversion. Holland Media Group (HMG) transferred its services from analogue to digital satellite (to save costs). Most of HMG’s viewers took the service by cable (6m), while 400,000 received the service DTH. Once the decision to convert had been taken and the appropriate contracts with Astra signed, less than 6 months remained before analogue switch off. Details were made public via press announcements, Teletext and publicity in satellite magazines. A phone line was set up to deal with enquiries and complaints. It was intended that DTH consumers would pay for the cost of transfer but that HMG would provide technical and financial assistance to cable companies for re-tuning. Because of the short time scales, only one manufacturer produced equipment before shut down and because of this the equipment was sold at a higher price than would have been the case in a more competitive market.

Up to 3 weeks before the switch off viewers refused to believe that the change would actually happen and many complaints were received. Two weeks before switch off the Dutch Consumer Association took HMG to court to stop the switch off but the court found in HMG’s favour. There was a 2 week simulcast period to give cable operators time to retune their equipment. Since the changeover (which occurred about 2 years ago) HMG has not recorded any noticeable change in viewing, suggesting that consumers converted when it was clear that this was the only available option.

One lesson that may be drawn from this experience is that viewers who do not see any advantage in DTT may be very reluctant to purchase DTT equipment until the last minute, believing instead that by doing nothing the change will not happen. For this reason and, in the absence of other evidence, we have made the following highly speculative assumptions concerning the impact of an early announcement of analogue shut down:

• industry responds to the announcement by bringing forward the date at which digital equipment costs fall to equal those of analogue equipment. Assuming an announcement is made in year 1 or 2, then equipment costs fall to analogue levels by year 5 (and not year 10 as assumed in our original scenarios).

• consumers react to the rapidly falling digital equipment prices, such that by year 5 all new TV purchases are of digital sets.

• consumers to do not react directly to the announcement of a shut down (the ostrich phenomenon).

The impacts of these assumptions on service take-up are discussed in Section 5.

3.6 Conclusions

We conclude that there are a number of useful policy options available to the government to accelerate take-up of DTT and reduce the likely time to analogue switch-off.

An early, credible announcement committing to switch-off over a clearly understood timescale. Whilst the announcement need not be definitive, it should not give room for major slippage or cancellation.

To provide clear information to purchasers of analogue sets of the consequences to them. This should not be implemented immediately, but at a carefully judged time once iDTV sets have become a practical alternative for the majority of the population. The format and medium of the message needs careful thought nearer the time.

To adopt a flexible approach to MVDS licensing, to give digital multiplex licensees the opportunity, if they are willing to invest in it, to ensure that at the least the core densely populated areas obtain full coverage of services available by DTT.

To use some of the money raised by spectrum auctions to fund a clearance contractor to make arrangements making it politically possible to carry through with switch-off. However it is unlikely to be effective to place the obligation to deliver universal television service and manage changeover on the new spectrum owners who intend to use spectrum for other purposes.

Contents

4. Options for Shutting Down Analogue Services

4.1 Introduction

Conceptually, the simplest way of ensuring universality requirements are met after analogue services are shut down would be to use the four (or five) frequencies broadcast from each transmitter to transmit the two (or three) digital multiplexes carrying the qualifying services. Any spectrum left over could be either used to increase the power and coverage of existing digital transmissions or recovered for other services. This approach would ensure over 99% population coverage for qualifying services and would allow remote communities to receive additional free-to-air digital services. It has the major disadvantage that very little, if any, spectrum would be released for alternative uses, such as mobile communications.

The key to releasing spectrum for other services is the closure of some, or all, of the existing analogue service. Following the roll-out of DTT services across the 81 planned sites (or the estimated 120 for which spectrum can currently be found), there will still be around 1,000 relay stations providing only analogue television transmission to around 5 to 10% of the population. By and large, these people live in rural communities. It is only by finding spectrally efficient solutions to providing continued service to these 5 to 10%, thus being able to reduce the considerable spectrum occupancy of these services, that substantial amounts of spectrum can be freed.

In this section, we assess different approaches that might be used to deliver qualifying services to the final 10% of the population. This section draws on the material presented in Appendices 4, 5 and 7.

4.2 Overview

The possible existing media that could be used to deliver qualifying services to the final 5 to 10% of the population are: satellite, cable, MVDS, transmission over existing telephone lines and UHF transmission. Cable, MVDS and existing telephone wires provide only incomplete solutions to the problem, though they may be appropriate for filling in coverage in certain areas. The main issues raised by these approaches are as follows:

Cable transmission would not be economic in rural areas, although might be an option in compact rural towns, or urban areas with coverage problems (such as, possibly, the south coast). However, MVDS is likely to be cheaper to provide in semi-urban environments. In either case, the services would need to obtain a programme feed either off air or from satellite. While a hill top site might be used to receive digital services off air, it is likely to be too distant from potential viewers for an MVDS site to provide coverage or for a cable connection to be economic. Thus these services may need to be fed by satellite, in which case DTH satellite may be more economic.

Neither cable nor MVDS transmission would be suitable for remote households, as cable transmission is too expensive and MVDS services operate over transmission distances of at most about 5km;

Transmission down existing telephone lines would require regulatory changes so that BT could carry broadcast services. It would require significant investment by BT and even then is only likely to provide access to at most an additional 1-2% of households. This is because the signal degrades as the distance traveled increases and many rural households live some distance from their nearest BT exchange.

Satellite transmission is understood to be feasible for most households (Astra claims up to 95%) in the areas concerned. However, at present only the BBC, S4C and Channel 5 appear to be planning to transmit their services by digital satellite; and

Satellite services are not (and cannot be) subject to "must carry" provisions for qualifying services whereas cable is subject to such provisions under the 1996 Broadcasting Act.

A difficulty with all of these, except UHF, is that there would be on-going infrastructure costs to cover. Cable, MVDS and telephone line delivery all have on-going costs for the investment in and maintenance of the local transmission infrastructure, which is generally charged to the household benefiting. Satellite transmission incurs costs for control of decryption by smart-card, as satellite transmissions must be encrypted for copyright reasons, also normally charged to the household benefiting. Such charges to the household would be inconsistent with the ambition that qualifying services should be available approximately universally and free from on-going charges. Broadcasters would have to cover these costs if the free-to-air concept is to be extended to properties only able to receive services in these ways. We note that there would be potential offsetting benefits to broadcasters/transmission operators through being the sole provider of services to these households (e.g. opportunities to sell other services).

Terrestrial transmission overcomes the free-to-air issues: however, there will be other costs to multiplex operators and consumers. Also, depending on the approach taken, varying amounts of spectrum will be released.

We conclude that the only two options which, in themselves, offer the possibility of reaching the majority of the final 10% of the population economically are direct-to-home satellite and UHF transmission. The other possibilities may have a role to play in providing infill in certain areas. Satellite and UHF are discussed further below and the most feasible variants are costed in Section 5.

4.3 Terrestrial UHF Transmission Options

There are three basic approaches to providing television programmes, via terrestrial means, to all those who currently receive terrestrial services. Within these approaches there are a number of options for achieving the same outcome, these options are discussed further in Appendix 5. The three basic approaches are:

1. simulcast: transmit both digital and analogue versions of the qualifying services from all transmitter sites. This would allow viewers to purchase digital televisions or STBs when they are ready to do so. It does, however, require additional spectrum to be found to simulcast the analogue and digital services and additional infrastructure to be installed at each of the relay stations to transmit the digital multiplex(es). The amount of spectrum released depends upon the number of multiplexes provided following switch-off. Little would be released relative to the present position unless service from such relays is limited to little more than the qualifying services.

2. change-over: modify analogue relay stations to digital operation. This could be done when the majority of viewers in the area served by a relay station had dual compatibility. It is estimated that around 75% of the existing relay stations could, following simple modification, be converted from analogue to digital operation. No additional spectrum needs to be found to implement this approach; but neither does it release much, unless service from such relays is limited to little more than the qualifying services.

3. continue analogue service: provide an alternative means to getting analogue pictures to relay stations. By feeding the analogue relay stations with decoded versions of their digital equivalents, the analogue transmissions from the main transmitter sites could be turned-off. Such an approach is simple and cheap but does not release much spectrum nor provide a universal digital service.

The simulcast approach is the most complex and would require the greatest investment in additional transmission infrastructure. It is also uncertain whether sufficient spectrum could be identified to make such an approach feasible nation-wide. It does, however, offer the simplest conversion route for viewers.

>The change-over approach is relatively uncomplicated from a technical perspective. It would require modifications to relay stations and a small amount of additional infrastructure. The complications are in the management of the change-over. The main drawback is the need for viewers to purchase or be provided with dual analogue-digital compatible sets (or a stand alone set top box) for which there would be no digital service until the change is made.

The continuation of analogue service approach is possibly the simplest and cheapest and would potentially allow the earliest closure of main station analogue transmissions. The number of analogue transmitters that could be closed, is, however limited to those which provide a simulcast service initially (i.e. 81 to 120) and this approach would leave the 10% of the population without digital broadcasts incapable of ever receiving DTT services, except to the extent that they were made commercially available by other methods. It raises the question of why the change-over approach is not used, as the continuation of analogue transmissions takes up more spectrum and provides no more channels.

All of the above options would allow some analogue transmissions to be switched off, although the exact amount of spectrum cleared varies. Each offers viewers a different service ranging from as many as four or more multiplexes, down to one or two multiplexes carrying little more than qualifying services, or continued analogue service. In the long term, all except the "continue analogue service" option offer viewers at least some digital services.

A solution that involves continued local terrestrial transmissions of 4-6 channels, whether digital or analogue, from many local relays will release very little spectrum. The channels switched off at the main transmitters will be cleared, but new channels will be occupied by the DTT transmissions from the main transmitters. In essence, very little spectrum will have been cleared from the current situation. Although the transmissions from local relays to fringe populations are often in areas where the spectrum is less valuable, they are also in hilly urban areas. Even in remoter areas, they cause interference over a much wider area than the intended recipient population, and in some atmospheric conditions interference can be experienced over a considerable distance. Alternative uses, e.g. mobile telephony, are generally much lower power than broadcasting, and will be prone to interference from any broadcasting carried out on the same channel over considerable areas. The interference experienced by public access mobile radio services from continental broadcasting confirms this. We therefore believe that this type of solution is unlikely to be consistent with the policy aim of releasing material spectrum for new services.

Table 4.1 summarises the various options available to enable analogue transmissions to be closed. It has been assumed that a maximum of six multiplexes is available. However, once the analogue broadcasting spectrum has been released, this may change and each option may need to be reviewed were new services to be provided. Where reference is made to a dual mode receiver, this could alternatively represent an analogue receiver with a set top box.

The column headed "equipment required by viewer" indicates what is required in order for the viewer to receive a service both before and after the addition of digital services. In the limit, where analogue transmissions are turned off, each viewer will require either a dual-mode or a digital receiver (the exception is for the "continue analogue service" method where a standard analogue receiver would continue to operate).

4.4 Satellite Transmission

Satellite is the most important alternative to UHF for delivering public service digital programmes to fringe areas. Indeed, some remote areas without analogue service have already expressed considerable interest in the possibility of receiving public service channels by this route, as currently they are restricted to purely commercial offerings and C5.

We describe here the equipment needed, the coverage that would be obtained, and delivery issues.

New equipment

Digital satellite (D-SAT) broadcasting would be received from an Astra satellite in a different orbital location from the current analogue Astra satellites. Viewers will need new equipment, and also a new site for the dish if the new alignment cannot be served from the existing site. The interest of BSkyB in transferring to digital broadcasting provides a synergy which may assist in promoting D-SAT technology as a successful delivery medium.

Coverage

According to Astra Marketing, at least 95% of households are potentially able to receive satellite broadcasts. Groups of properties, normally blocks of flats, can be served with a shared dish by SMATV, at much the same cost per property as if they had individual dishes, and the figure of 95% includes the cases where this is practical. Thus the approximately 5% of exceptions cover situations where terrain makes it impossible to find a mounting site with line of sight to the satellite neither on or close to the property itself, nor on a neighbouring property where SMATV might be used.

The 95% figure does not exclude cases where planning restrictions, or contractual restrictions on leased property, may be an impediment to mounting a dish. According to Astra Marketing, a solution satisfying planning restrictions is usually available. We have also heard informal figures suggesting that the proportion of properties unable to receive satellite in practice is 20-30%, suggesting that significant obstacles impede fitting a dish quite widely. It is likely that in the current market, given the optional nature of satellite television, people are deterred by installation difficulties, even if in principle they are soluble. Some local planning authorities apply more rigorous constraints than others, for no apparent reason. However, we assume that there would be considerable political pressure to ease planning constraints in areas where satellite became the main route for reception of qualifying services.

Assuming that planning difficulties can largely be addressed, this implies that by using satellite (alone) as a supplement to the 90 to 95% potential DTT coverage obtainable by qualifying channels using currently available UHF frequencies, public service coverage could be extended to about 99.5%. This is comparable to the existing analogue coverage, though the 0.5% not covered by either DTT or D-SAT would probably be materially different from the current 0.6% not covered by analogue UHF. In particular, remote areas of Scotland would obtain good coverage, and reception black-spots would be transferred elsewhere, probably as much to England as anywhere else. Cable can be used for further in-fill where reception problems occur in urban areas, and MVDS in other semi-urban areas or reasonably compact settlements.

Delivery

In order for satellite to be a practical and acceptable method for delivery of qualifying services on a reasonably widespread basis, then the five basic channels would have to be all available on digital satellites (D-SAT) in the same orbital sector, so that they could all be picked up from a single dish. The BBC has reserved digital satellite space. Channel 5 is currently available on analogue satellite and is expected to be available on D-SAT. S4C intends to go on D-SAT. Thus there is an issue as to whether ITV and C4 will go on D-SAT. It could be made a condition of their licences.

It is plainly costly to deliver the complete regionalisation currently provided for analogue onto satellite, especially in view of the small in-fill needed in some regions. The BBC’s plan is to reduce the level of regionalisation provided for terrestrial transmission, but it will provide at least the four regions corresponding to the four countries of the UK. More difficult questions arise in the case of ITV regionalisation, as the level of regionalisation is set by legislation and licence. Further examination is needed of what is the appropriate level of regionalisation on satellite. In particular, given that additional provisions were inserted during the passage of the Broadcasting Act 1996 to protect regional programming and production on the Channel 3 network, Parliament may seek similar safeguards for the regional output of Channel 3 for digital satellite delivery.

D-SAT STBs will contain a modem and be required to be connected to the telephone system as a condition of access to subscription services, or at least to obtain subsidised installation. This can also be used to ensure the correct region is accessed for regional services, by cross referencing the telephone number to the post code, which then allows the STB to select only the service intended for the households’ post code area. For example, this prevents London viewers from receiving a Scotland service. However it could be an unreasonable restriction on a household seeking access only to the free-to-air services that it should connect its STB to the telephone system, and viewers in such a situation may in practice be able to make a free choice of the region they watch.

Qualifying services will need to be encrypted for copyright reasons if they are to be carried by D-SAT. This method relies on the use of a smart card. (The smart card might also be used to restrict viewer’s access to only their local ITV service). We understand that annual costs of £10 per household would be incurred to pay for the smart card and associated management costs. The alternative of soft encryption (currently used by Channel 5) in which a UK decoder board is installed in the STB is not thought likely to be feasible because it does not provide adequate copyright protection. For commercial satellite services, it is normal for the subscriber to cover this cost. If, however, the customer sought only free-to-air services, then it would be in conflict with the government’s policy objectives, as they are currently understood, if DTT was unavailable and the customer had to pay this on-going cost. This suggests that the broadcasters of qualifying services must pay for and distribute the smart card in these circumstances. A question for further consideration is if broadcaster funded smart-cards are available for free-to-air services, then to whom would it be distributed: to anyone who sought it, or restricted to households resident in areas where qualifying services are not available by DTT.

Finally, we note that it is not possible to oblige satellite operators, such as Astra to carry the qualifying services. Thus, if there was not sufficient transponder capacity available satellite would not provide an adequate solution. It would appear that this is unlikely to be a problem in practice, as there is spare capacity on Astra 2a and 2b. Also it seems unlikely that, in future, the capacity owner would capriciously renege on its transponder leasing contract with one of the qualifying service providers.

4.5 Conclusions

The key issue in closing off analogue service is finding a delivery method which retains universality of service, and thus continues to provide service to around 99.4% of the population currently served by the analogue network.

The existing UHF transmission takes up large amounts of spectrum to reach the final 5 to 10% of viewers. It could be arranged that these viewers continue to receive analogue transmission once analogue services from the main transmitters are switched off, or that their analogue transmissions are converted to digital, however this would release very little spectrum for other services. Alternatively, they could be given a reduced UHF digital service, for example of one or two multiplexes only. This would release more spectrum, but imposes transition costs on the fringe community whilst receiving little or no benefit in terms of additional services.

To what extent UHF spectrum should be used, on economic grounds, for broadcasting to the fringe audience depends upon its relative value in other uses. Broadcasting is probably the most valuable use of the spectrum, but with rapidly diminishing returns once the core 90 to 95% of the population is served. It is possible that broadcasters would outbid other users for this spectrum, even to serve the remote communities. Whilst there may be a few areas where a UHF change-over might be appropriate, it appears unlikely that it is economic to serve all areas with UHF DTT. Thus other methods will almost certainly be sought to serve these communities.

All alternative methods (cable, satellite, MVDS, by telephone, etc.) include some ongoing costs, i.e. local infrastructure and/or smart-card, which are normally paid for by the viewer when buying subscription services. For those households not wishing to subscribe to pay-TV, UHF transmission would appear more attractive. If the qualifying services are to be provided free of on-going costs to the viewer, some of these local costs may have to be covered by the public service broadcasters when viewers do not wish to buy subscription services.

Cable will be useful for extending coverage where black-spots occur in urban areas. MVDS will be useful for semi-urban areas or compact rural settlements. Digital satellite (D-SAT), in conjunction with SMATV appears to have the potential for extending delivery to the greatest part of the fringe community, since it is applicable in remote areas and scattered communities. D-SAT services, including at least BBC, S4C and Channel 5, are likely to be available from the inception of DTT, thus providing extended simulcast in all areas. ITV and C4 must go onto D-SAT if qualifying services are to be available. However, there are coverage black-spots in certain terrains and urban situations, and planning controls can restrict installation. Using a portfolio of technologies, delivery can be extended to about 99.5% of the population.

In a world where viewers are expected to receive their qualifying digital services by a variety of technologies in addition to DTT, the route for qualifying service in each area needs to be identified relatively early to provide a simulcast period. Only if viewers are aware of their simulcast technology do they have the opportunity to make a convenient transition at minimum cost. If required to comply with free-to-air universality policy, free smart cards or other assistance must also become available during the simulcast period. Without simulcast, additional assistance in making a rapid switch-over may be needed.

Contents

5. Costs and Benefits of Analogue Shut Down

This section addresses the following questions:

• What are the gross costs of moving to digital transmissions in terms of investment in transmission infrastructures, reception and DTT programming equipment?

• What are the key issues and likely costs and benefits of switching off analogue TV in 5, 10 and 15 years’ time?

5.1 Gross Costs

Below we present estimates of the gross costs associated with the introduction of DTT. These comprise the costs of:

• investment in transmission infrastructure (Section 5.2.1);

• reception equipment (Section 5.2.2); and

• TV programming (Section 5.2.3).

Investment in Transmission Infrastructure

DTT will require new capital investment in order to upgrade the terrestrial transmission network to digital operation. The incremental costs will include the installation of new antennas and new transmitters, and the strengthening of existing masts. Estimates of the cost of network roll-out vary and in part depend upon the level of coverage achieved. The investment cost profile in terms of homes passed is highly non-linear, as a few large transmitters located in or near major conurbations can reach a significant proportion of the population. The ITC and BBC estimate that around 90-92% of the population can be reached using 81 transmission sites. To reach the final 8-10% about a further 1,000 transmission sites are likely to be required. Hence, the investment cost per household begins to rise as an increasing number of smaller transmitters and relay stations are required to cover smaller towns and dispersed rural populations.

Below we estimate the cost of extending universal coverage to the final 10% of the population under two different UHF transmission options and the option of satellite transmission of qualifying services. In order to provide a complete comparison of the costs of these two options we have also estimated the consumer cost of reception equipment for each option.

Reaching the 81 main sites

Based on various industry estimates, we expect the scale of capital investment required for transmission from the 81 main sites to be around £170 million. This figure comprises around £110 million for transmission from sites to the public, £50 million for encoding, multiplexing and compression at the studio and £10 million for distribution from the studio to the transmission sites.

In addition to these once off capital costs there are annual/operating costs which, based on discussions with industry, we estimate to be as follows:

• £25-35 million maintenance costs;

• £10-14 million multiplexing and encoding;

• £20-28 million distribution;

• implying a total ongoing cost of £55-77 million per annum.

If universal coverage of the qualifying services is to be maintained, and analogue services are to be switched off, then some means of extending digital coverage will need to be found. The main options are extending coverage using UHF spectrum or providing for carriage of the qualifying services on digital satellite. Other non-UHF transmission options such as cable or MVDS do not provide for universal coverage but could be used to provide infill in certain areas (see Appendix 5 for further discussion of this).

Extending UHF terrestrial coverage

Completing national roll-out of DTT services such that the qualifying services could be received by about the same number of viewers as currently receive an analogue service could be tackled in a number of ways (as described in Section 4). To provide a bound to the range of likely costs, we have quantified the costs of the following two options:

1. analogue to digital change-over; and

2. simulcast three multiplexes containing qualifying services.

The change-over option would involve modifying analogue relay stations to digital operation. There would be no simulcast period. Four or five digital multiplexes could be transmitted following changeover, with potentially six multiplexes later, once all the spectrum has been released from analogue services. In the simulcast option the existing analogue channels would continue to be broadcast together with at least three existing digital multiplexes.

Table 5.1 presents estimates of the total capital cost of extending terrestrial coverage using these two approaches to extending coverage. Full details of these estimates are provided in Appendix 5.

Table 5.1 Capital Cost of Completing National Roll-out of DTT Beyond the 81 Main Sites
(1997 £)

The perceived cost to consumers of switching from analogue to digital reception will differ between the two scenarios. In the case where services are simulcast, those consumers who obtain a net benefit from switching to digital reception will make the switch. In the case where the analogue to digital conversion happens "overnight", most if not all consumers are likely to perceive the change as imposing a cost on them, namely that of purchasing DTT reception equipment.

We have dealt with this issue by estimating the costs to consumers at year 10 (based on our low take-up scenario) of having to purchase a STB to receive digital services. It is assumed STBs have fallen to a mature market price of £100 by year 10. In the analogue to digital change-over scenario all 2.3m households are assumed to need a new STB. In the simulcast scenario only 1.5m require an STB, as the remainder are projected to have already converted to DTT reception. Thus the costs of converting at year ten are estimated at around £230m in the change-over scenario and £150m in the simulcast scenario.

Satellite

Completing the roll-out of qualifying services using digital satellite would require capacity for 21 digital channels. The total annual cost for uplinking and transponder capacity for the 21 channels is around £10.4 million. In addition services would need to be encrypted for copyright reasons. Industry estimates suggest that the annual cost per household for provision of smart-cards to decrypt the qualifying services would be in the region of £10 per household suggesting an additional annual cost of £23m assuming the final 10% of TVHH receive their services by satellite.

In order to receive services, consumers would need to purchase digital satellite reception equipment. As in the UHF transmission cases, we have examined the costs of transferring to satellite reception at year 10. Based again on our low take-up scenario projections, it is assumed that around 17% of households across the country will have already switched to satellite reception in order to receive pay TV services. The costs of consumer reception equipment for the approximately 10% of households lying outside the coverage area of the 81 main sites is estimated as follows:

• 17% of these households are assumed to have already adopted digital satellite;