Home > Spectrum > Strategic Spectrum Framework Review > Progress on key spectrum initiatives

Introduction

Why we are publishing this note

1.1 The Spectrum Framework Review (SFR) was published in 2005 and set out an outline plan for spectrum management until 2010. We are now just over halfway through this period and at a sensible point to assess how well we are achieving the keys aims of the SFR to move towards market mechanisms, to assess any changes that might be needed in the light of events and to update the actions included within the SFR as appropriate. It is intended that this note will provide a useful update for stakeholders including spectrum licence holders and those interested in using spectrum in the future.

A reminder of Ofcom’s spectrum strategy

Classes of spectrum

1.2 In the SFR we discussed how spectrum management fell into three different classes – command & control, market forces and licence exempt.

• Command & Control is where the regulator makes most of the key decisions about the manner in which the spectrum is used including either the usage (eg mobile) or the technology (eg GSM) or in some cases both.
• Market forces is where the market is able to decide on the manner in which the spectrum is used. The regulator sets a few basic restrictions to prevent interference to others and then licence holders can decide upon the use and technology that is optimal.
• Licence exempt is where anyone is allowed to access the spectrum without a licence as long as their equipment
conforms to certain conditions such as limits on transmitter power.

1.3 We argued that a mix of these different classes was needed in order to cope with the wide range of applications and uses that derived value from using the spectrum. We showed that it had to remain the responsibility of the regulator to determine what the most appropriate mix was.

1.4 In determining an appropriate mix we firstly considered the optimal amount of spectrum for licence exempt usage. We showed that although licence-exempt access enabled a range of innovative new ideas and generated consumer value that there was little need for any additional allocations because the current provision was relatively lightly used, especially at 5GHz. However, we promised to continue to monitor utilisation and to consider providing additional spectrum if congestion seemed likely. We discuss this in more detail below.

1.5 For the remaining spectrum we argued that there should be a strong preference towards market forces. By this we meant that spectrum should be tradable with minimal restrictions on usage and technology. Our arguments for this are summarised below.

Preference for market forces

1.6 The command & control approach used widely to date “works” in so much as it licenses spectrum to particular users and ensures that excessive interference is avoided. This allows a range of uses of the spectrum in a stable and predictable environment. However, it is unlikely that it achieves the full objective of a spectrum manager of maximising the value derived from the spectrum. To do this, the regulator would need to make sure that spectrum was divided up between all the different possible uses and users in a way which maximised benefits to end users of spectrum using services . Since it is almost impossible to predict the value that each different service provides under any given spectrum allocation it is difficult to see how a “command and control” approach to managing the radio spectrum could maximise value.

1.7 In times where supply of spectrum exceeded demand, or where there were a relatively small number of services, it was more plausible that the regulator might approach this goal. However, increasingly, demand for the spectrum has grown as has the number of spectrum-using services. There have been many pieces of evidence that suggest that regulators are failing to maximise value under such circumstances. For example:

• Some regulatory decisions, such as the allocation of spectrum to the ERMES paging system or the TFTS in-flight phone system in Europe, have resulted in spectrum being unused for over a decade. Clearly, it could have been put to an alternative use which would have resulted in some value.
• Widely differing valuations for the spectrum at auction suggests that the balance between different uses is incorrect - for example, prices in the 3G auctions were over 100 times higher on a comparable basis than the spectrum auctions at 3.4GHz despite the two bands being relatively similar in their physical properties.
• Many new applications or technologies have had difficulty in gaining access to spectrum – for example the iBurst cellular technology or more recently Mobile TV systems. While it is not certain that these would increase the value of the spectrum, their difficultly in entering the market may be a symptom of an excessively rigid system.
• Some applications which have historically been granted spectrum free, especially in the public sector, have not modernised their systems or reviewed spectrum planning for many decades despite the availability of more efficient technologies and the increased demand for spectrum, suggesting there are insufficient incentives for some users to optimise their use of the spectrum .

1.8 In addition, the increasingly blurred line between different services such as broadcast and mobile communications is making it difficult to operate a command & control approach where certain uses are allowed and others not.

1.9 Economists have long argued that market mechanisms should be applied to radio spectrum. Seminal papers in this area start with Coase in 1959 . The essential idea here is to allow pricing mechanisms to act as an incentive for holders of spectrum to optimise their use – buying more if their business case can justify it, selling spectrum if they have excess, and adopting new technologies that can use spectrum more efficiently where economically viable. Economic theory and practical experience suggests that in a market which is performing well, this will lead to an allocation of spectrum that maximises economic value. Under such an approach the regulator sets out rules that enable markets to function while ensuring that interference is controlled and then takes a back seat, leaving it to the market to determine the use and users of the spectrum.

1.10 A market instrument that has already been widely adopted is the use of auctions to distribute spectrum. Auctions are valuable because they are more likely to identify the best user of the spectrum than a regulator using administrative allocation. But auctions on their own still ‘freeze’ the assignment of spectrum. They need to be accompanied by mechanisms to trade and change the use of spectrum as market conditions change and new services become available. Together these mechanisms are what we term “market forces” and they enable spectrum to be used in the manner that generates the greatest value.

1.11 However, in the SFR we noted that there were some areas where it was difficult to make full use of market mechanisms or where restrictions have to be imposed, in particular where spectrum usage required international harmonisation in applications such as aeronautical use and satellite transmission and here an appropriate mix of market forces and command & control should apply. In particular, we believe that the opportunity cost of spectrum should be clear in such cases because there is often scope for some change of use, or additional usage in the short term and without a clear opportunity cost it may be difficult for appropriate longer-term decision to be made.

Spectrum for licence-exempt applications

1.12 Licence-exemption is an important usage of spectrum. It enables many applications that add substantial value to users including WiFi distribution of data around the home and office, BlueTooth wireless links between devices such as cellphones and headsets, cordless telephones, baby monitors, remote car key fobs and much more. However, it adds relatively little economic value to the use of the spectrum – recent studies suggest that perhaps only around 1% of the value generated by using the spectrum comes from licence-exempt applications. Hence, while it is important to provide licence-exempt spectrum, the success of applications such as WiFi should be placed in context with the applications that dedicating spectrum to licence-exemption might displace, such as cellular.

1.13 As we promised in the SFR, we have attempted to characterise the congestion in the licence-exempt spectrum. In 2007 we embarked on a series of measurements in a range of locations and across all of the key bands . We recognise that these measurements are not perfect and that it is extremely difficult to characterise fully congestion in licence-exempt bands, nevertheless, the measurements do provide some useful indications. Broadly, they show that most of the bands are relatively lightly used. Even the highly popular 2.4GHz band is only around 20% used on average and 40% used in the busiest locations. The 5GHz band, seen by many as an “overflow” when 2.4GHz becomes overly congestion, is almost completely unutilised. This leads us to believe that there is no pressing need for additional spectrum to be found for licence-exempt allocations, although there may be sound spectrum management reasons for exempting the spectrum in any case, as discussed in our Licence Exempt Framework Review (LEFR).

1.14 Some also believe that licence-exempt spectrum is a key area for innovation. While we agree that there is innovation in these bands, we believe that in general innovation in licence-exempt devices occurs in licence-exempt bands whereas innovation in licensed devices occurs in licensed bands. So, for example, the innovations that led to the Blackberry-style wireless email or GPS navigation-based applications occurred in licensed spectrum. They did not start in licence-exempt spectrum and then migrate across to licensed spectrum. Hence, an appropriate balance of licensed and of licence-exempt spectrum is needed to encourage both forms of innovation.

1.15 In summary, we remain fully committed to the provision of licence-exempt spectrum, but believe that it must be carefully balanced with the provision of licensed spectrum, taking into account the likely economic value under both approaches across a range of frequencies. Given that licence-exempt allocations are generally lightly used this suggests to us that there is currently no need to allocate more. However, this may change in the future and we will keep this issue under review.

Spectrum as a policy tool

1.16 We also argued that spectrum should not be used as a policy tool. By this we meant that spectrum should not be set aside for particular applications in order to achieve a policy goal such as the provision of high definition television. Nor should additional licence conditions be placed on spectrum awards to achieve social objectives such as requirements to provide coverage to rural communities. This is somewhat different from the policy approach adopted by our predecessor, the Radiocommunications Agency and some other regulators and hence it is worth restating our thinking here.

1.17 Spectrum is an input to a process that delivers some end service. For example, spectrum is one input used by cellular operators in their delivery of a cellular telephony service. But spectrum alone is insufficient; operators also need staff, equipment, power, sites for masts and so on. If spectrum is provided in a manner that is designed to achieve policy goals – for example a larger allocation of spectrum in return for delivering coverage to rural communities – this tends to result in an overall reduction in the value that the economy derives from the spectrum.

1.18 The reason for this is that users of spectrum, such as operators, need to make trade-offs between all the different inputs they use. For example, it is generally possible to use less spectrum by building additional cell sites and installing additional equipment. Operators will tend to seek the optimum balance for themselves by selecting the point which minimises their costs while generating the outputs, or services, they believe necessary. However, if the price of spectrum is distorted, for example by being reduced in return for rural coverage, then they will make a different trade-off. With less expensive spectrum they will now choose to use more spectrum and less equipment in order to deliver the same output.

1.19 While this change in spectrum use will be desirable for the operator, it will be sub-optimal for the economy as a whole. Because the operator is using more spectrum than they would under a market approach, there is less spectrum available for other applications and the overall value that can be derived from the spectrum is lower. Given that spectrum generates in excess of £40bn per year of value to the UK economy, even a small percentage reduction in efficiency could have a very large financial value.

1.20 Economic theory is very clear that where a particular output is desired then any interventions to achieve this should, where possible, be made at the output stage, not at the input stage. Interventions made to inputs distort the market and result in a less efficient use of inputs, reducing the overall value to the economy. The true economic cost of spectrum is its opportunity cost – the use of spectrum for one technology or service typically denies its use for others and the opportunity cost reflects the loss to society of this option. Decisions on choice of inputs should respond to price signals that reflect economic cost.

1.21 Unfortunately, the use of spectrum as a policy tool is seductive. It appears to many that policy goals can be achieved “for free” through attaching conditions to spectrum since no direct funding is required from the Government, or other similar body. The alternative of direct subsidy requires often difficult budgetary discussions and enables scrutiny and analysis as to whether value for money is being achieved. Hence, for politicians, public interest bodies and others, using spectrum to achieve their goals rather than direct intervention can be seen as simpler to achieve and easier to explain.

1.22 However, as we have set out above, what appears to be a cost-free intervention will actually result in a cost to the economy, which may in some cases be many billions of pounds. This cost is not immediately apparent because it is in terms of benefits that will not be received in the future due to inefficient allocation of spectrum. Nevertheless, it is very real and can have a noticeable impact on consumers, citizens and the economy.

1.23 Intervention at the output stage also makes the cost of providing the public goal very clear which invariably makes for better policy discussion and decision than when that cost is hidden.

1.24 It is for these reasons among others that, for example, we consider the free provision of additional spectrum for high definition TV (HD-TV) is inappropriate from a spectrum management viewpoint. Although on the face of it, it appears to deliver HD-TV to society at apparently none, or little cost, this is not so, and the value lost due to preventing other applications could be substantial. In the same way that the broadcasters do not expect, nor receive, free employees or free broadcasting equipment, neither is it appropriate for them to receive free, or discounted spectrum.

1.25 In some cases we have inherited policy conditions attached to licences issued by previous regulators. We will consider carefully whether these conditions should remain in force, or should be removed, taking into account the effect that the change could have on optimal use of the spectrum, as well as other issues such as competition and the availability of services to citizens and consumers.

1.26 The key message is that while the use of spectrum to achieve policy goals is seductive in that it appears to allow worthy objectives to be achieved at no cost, this is far from the truth. It is likely to be more costly than intervention at the output stage and it results in less clarity as to the cost of achieving the objective. It should generally be avoided.

Hoarding and speculation

1.27 We expect that as a market in spectrum becomes established we may see similar activities to those that occur in other markets, such as speculative acquisition. In general, we do not see this as a problem. For example, a speculator who conjectures that a new technology will shortly emerge and who buys spectrum but then leaves it unused as technology develops may increase the value derived from the spectrum. This is because, were it immediately put into use, it might prove more difficult to refarm to the new technology with resulting loss in value to consumers. Speculators may also play useful roles in consolidating fragmented spectrum holdings or otherwise adding value to the market. Given that this kind of activity can be very beneficial to consumers, we do not generally propose to employ “use it or lose it” clauses which would limit licensee’s flexibility about when and how to use spectrum.

1.28 We do, however, need to be wary of the risk that the acquisition of spectrum could be anti-competitive. For example, an operator might acquire spectrum purely in order to prevent the entry of a new competitor. Given the already competitive nature of most wireless services and the increasing amount of spectrum available in a wide range of bands we do not expect to see this happening, but it is an important issue that we will consider carefully. If we believe that the risk is large enough then there are powers available to take appropriate action.

Our vision

1.29 Based on this strategy, in the Spectrum Framework Review we set out a vision for spectrum which can be summarised as:

The Ofcom Spectrum Vision

1. Spectrum should be free of technology and usage constraints as far as possible. Policy constraints should only be used where they can be justified;
2. It should be simple and transparent for licence holders to change the ownership and use of spectrum; and
3. Rights of spectrum users should be clearly defined and users should feel comfortable that they will not be changed without good cause.

1.30 We noted that we would achieve this by:

• Providing spectrum for licence-exempt use as needed, but our estimates were that little additional spectrum would be needed in the foreseeable future, growing to 7 per cent of the total spectrum;
• Allowing market forces to prevail through the implementation of trading and liberalisation where possible. We aimed to implement these policies in around 72 per cent of the spectrum; and
• Continuing to manage the remaining 21 per cent of the spectrum using current approaches.

1.31 We noted that there would be circumstances when we could not fully achieve this vision. In these cases we promised that we would explicitly explain why we have departed from it.

1.32 We still believe this to be the correct vision and are not intending to change it at this stage.

• Progress on key spectrum initiatives [pdf]
  Full Print Version

Back to top