ACCC Chairman Rod Sims on the ACCC’s rethink on the way it approaches spectrum ahead of 5G auction
Good afternoon ladies and gentlemen and thank you for the opportunity to speak here today.
And I will begin by congratulating Nerida on her very recent and well-earned appointment as ACMA Chair.
Our two organisations have a long history of working closely together and this will certainly continue.
Today, I would like to address three issues:
- First, why the ACCC cares about spectrum
- Second, what else is happening in mobiles regulation, and
- Finally, 5G from my perspective as a competition regulator.
1. Why the ACCC cares about spectrum
The ACCC’s primary interest in spectrum is to ensure that the markets which rely on spectrum as an input are as competitive and therefore as effective as possible.
Under the current regime, we provide advice to the Minister of Communications, when asked, on whether competition limits should apply to spectrum licence allocations.
We also assess spectrum acquisitions under section 50 of the Competition and Consumer Act; the mergers and acquisitions provisions.
Our past approach to competition limits has been quite simple; we sought to prevent dominance of spectrum by any one licensee in particular bands.
We are re-thinking this approach.
The upcoming 3.6 gigahertz allocation is a fascinating case study of why this approach is becoming increasingly fraught. Instead, and this is a key point, we want to consider spectrum holdings holistically rather than in particular bands.
While international band plans are not yet settled, there appears to be general agreement that 3.4 to 3.6 gigahertz is going to be the first global “cab off the rank” for 5G spectrum, or as Nerida described it this morning, is going the “pioneer” band for 5G.
In Australia, the holdings in the 3.4 gigahertz band look very interesting.
Optus and NBN have significant holdings. Optus holds all 100 megahertz in Sydney and Melbourne metro. NBN has 100 megahertz in outer Sydney and Melbourne. Telstra has smaller holdings across a range of capital cities and regional areas.
Telstra has some holdings up the top of the band and I think we can expect that it will be seeking to add significant additional holdings in the auction.
In an unrestricted auction, could we expect to see Telstra buying 100 MHz in the metropolitan areas and NBN buying in the outer metropolitan areas and regional centres?
Where does that leave the others?
Auctions are, of course, seen as the best mechanism we have for allocating spectrum.
The theory is that auctions promote allocative efficiency; the firms that value the spectrum the most will win at auction.
However, as every commercial entity in this room knows, companies also put a high value on preventing their competitors or potential competitors getting access to spectrum.
Stefan Zehle of Coleago Consulting, an advisor to many international telcos in spectrum auctions, has said that, in many cases this is the largest source of value, meaning that a large part of the value in acquiring spectrum is actually in reducing competition.
We would have to be naïve in the extreme to think that this is not the case in Australia, too.
And also with a new entrant in the Australian mobiles market, there is an imperative to prevent incumbents from purchasing spectrum for the purpose of preventing the new entrant from competing effectively in the downstream markets.
Mechanisms are being used overseas to prevent this from happening such as the imposition of spectrum caps, where an operator is prevented from holding more than a specific amount of spectrum across bands, or spectrum set-asides where spectrum is set aside for a new entrant with the remaining spectrum allocated at auction.
I do not want to see some players locked out of markets, or a new mobiles entrant prevented from competing with incumbents, because they cannot get access to sufficient spectrum. This means it is vital to consider the impact of spectrum allocation on competition in spectrum-using markets.
I’m going to end this section by saying something I’ve said before and that I will keep saying; the value of spectrum lies in the economic and social benefits it can provide to citizens and consumers, not in financial returns to the Budget.
I’ve said it before as we’ve seen the evidence: when governments downplay competition to sell monopoly assets for the highest price the economy loses out.
2. What else is happening in mobiles regulation
We have heard a lot recently about how Australia’s mobiles market has benefited from a lack of government intervention.
But as the representatives of the MNOs in this room are very much aware, it is certainly not ‘unregulated’.
First, we have the Mobile Terminating Access Service, or MTAS.
The ACCC will start a declaration inquiry for this service next year. MTAS, of course, is used to terminate calls and SMS on mobile networks whether they originate from fixed or mobile networks.
The central issue we’ll be considering in the MTAS inquiry is whether MTAS is still fit-for-purpose; that is, promoting any-to-any connectivity.
We are starting to see a couple of interesting issues arising with mobile interconnection. We are aware of smaller providers of innovative services that may be being inhibited in interconnecting with mobile networks.
The ACCC would be very concerned if smaller networks were not able to interconnect with larger networks. We will keep a close eye on developments.
We are also interested in the evolution of mobile termination from voice calls to calls terminated on IP networks and whether this impacts on MTAS.
I would encourage those interested in MTAS to start thinking about these issues.
I don’t propose to talk much about mobile roaming. There are plenty of opinions on that in this room, I am sure. However, I do encourage you to read the ACCC’s final report, particularly beyond the headlines.
In the inquiry, we looked very carefully at competition in the mobile services market and incentives to make investments to expand coverage and improve networks, particularly in regional areas.
What we found was that declaration could actually harm the interests of consumers by increasing prices in regional areas and by undermining the incentives of mobile operators to make investments to compete with each other in regional areas.
We also put out a paper at the same time on regional mobile issues.
We heard from many regional consumers and businesses that inadequate mobile coverage (including quality and depth of coverage) in regional areas affects the social and economic well-being of regional communities.
We see this ourselves and I could provide you with many examples where my coverage has dropped out as I have travelled with my fellow Commissioners into regional Australia.
What we were able to do in this paper was identify a number of issues where we think improvements could be made that would deliver better outcomes for regional consumers.
These include better transparency about network coverage and quality, more accountability around network investments, and better information for regulatory and policy decision makers.
We are already acting on those issues within our purview. We will shortly commence a review of the Facilities Access Code, and just this week commenced a review of our Infrastructure Record Keeping Rule to improve the information that we collect about mobile networks.
We will also actively engage with industry and federal and state governments about the issues this inquiry has identified; I encourage you all to also engage on these issues.
I want to turn now to 5G.
A few years ago, we had mobile operators coming to us concerned that their future business model would be limited to pushing data through dumb pipes.
Other than faster speeds and capacity and lower costs, there was this new ‘Internet of Things’ on the horizon which could drive demand. Consumers wanted more and more data but were perhaps unwilling to pay higher prices. Over-the-top providers were cannibalising voice and SMS revenues. The ACCC was cutting regulated termination rates.
It all seemed a bit depressing for the mobile operators.
It sounds hyperbolic to say it, but mobile operators around the world are now at an extraordinary moment.
And Australian mobile operators are faced with a unique confluence of incentives which means that Australia is particularly well-placed to benefit from 5G.
I will discuss these in more detail a little later.
You don’t have to go far to hear all about the benefits that 5G will deliver; from the ‘tactile’ Internet enabling doctors to perform surgery remotely, to self-driving cars.
I am a competition regulator and so I am going to take this opportunity to look not only at the opportunities, but also the threats of 5G.
5G will certainly be a significant, even revolutionary, development. But where it will take us and what new markets will look like, is unknown.
As you all know, 5G is simply a technology upgrade, but it is one that requires extensive investment in network equipment, spectrum, backhaul and base stations.
We don’t yet have a technical standard but we can assume that 5G is going to have at least five basic features:
- Low latency, meaning very little delay in device response times
- Very high data rates
- Huge growth in system capacity
- Very low device cost, and
- It will integrate better with Wi-Fi and other networks, offering seamless, always-on connectivity
In terms of what is delivered to market, 5G networks will likely offer a number of things, but I am going to focus on the three most obvious:
- Wireless broadband services at gigabit speeds
- The ability to connect large numbers of devices and objects (the Internet of Things)
- The ability to provide a number of discrete ‘fit for purpose’ networks rather than ‘general purpose networks’.
The first point is receiving the most interest in Australia at the moment.
Wireless broadband at gigabit speeds
As our Telecommunications Market Study draft report, released on Monday notes, 5G deployment has the potential to support and accelerate the extent of mobile and wireless substitution for fixed line customer access networks.
Current 4G networks can technically achieve speeds at or above NBN fixed line broadband services. However, the increases in system capacity, speed and ability to integrate with other networks enabled by 5G means that it has a real advantage in competing against fixed line networks.
Potentially, 5G can be used as a last mile technology. We are seeing this in the United States where Verizon has announced that it will use 5G as the last mile in its fixed network, greatly reducing its network rollout costs.
It can be used as a fixed-mobile hybrid offering. Deutsche Telekom already offers hybrid modems for its ADSL customers who experience congestion in the evenings with the wireless service kicking in when necessary to increase capacity; 5G would make this absolutely seamless.
And as everyone in this room is aware, 5G networks can be used as a substitute for fixed networks.
What is most likely in Australia?
I mentioned that there is a confluence of incentives pushing Australian mobile operators to 5G.
First, we now have no fixed / mobile integrated operators in the traditional sense of what we had with Telstra and Optus; TPG is perhaps an exception with its fibre-to-the-basement network.
In the past, a dominant player would have had incentives to stomp on any new technology threatening its business model.
Traditional horizontally integrated networks have incentives to push traffic to their lower-cost fixed network. Our MNOs do not have that incentive and indeed are more likely to have current incentives to push their customers away from NBN.
Second, our MNOs have superb networks. All three Australian networks rate consistently among the best in the world. Network investment has meant that Australian MNOs are well placed to rollout 5G with the extensive small cell deployment and fibre required by these networks.
Finally, on the demand-side, as we note in our Communications Market Study Draft Report, the performance of NBN services in terms of price and service quality will have a significant impact on the degree of substitution sought by consumers.
The upshot? Follow the money!
Mobile services are relatively lightly regulated and we expect that this is where capital will head. 5G is an incredibly expensive proposition, but offers real payoffs for MNOs in the Australian environment.
Obviously the threat in this context is to the NBN. We have clearly said that we do not support any extension of the Regional Broadband Scheme to the mobile operators. The way to counteract threats is not to impose regulation to protect the NBN.
Clearly NBN Co is aware of this threat but there are signs that it is also aware of the opportunities as NBN Co has spectrum suitable for 5G also.
Internet of Things
Adding to these incentives, MNOs are increasingly well placed to take advantage of the burgeoning Internet of Things, or IoT.
I talked about MNOs being concerned a few years ago about being dumb pipes.
Well, we are now looking at a future where telecommunications is no longer merely a wholesale input for industries which needs to be supplied at a competitive price but actually, in the words of one commentator, the indispensable “technological backbone” for many industries for whom machine-to-machine communication is or will be necessary.
5G will give the telcos the opportunity to actually embed themselves in these industries. We will be watching these developments very closely.
Finally, the IoT is likely to also present the MNOs with new incentives for network deployment, including in regional areas.
As the role of mobile networks stretches beyond providing communications between people, to providing connectivity to millions of devices and objects, the business cases for network deployment will no longer be restricted to where people live and travel. Increasingly, the need for data to support businesses will justify new investments in regional areas.
The last thing that I want to mention is the potential of 5G to go beyond the ‘general purpose broadband network’.
While the technical standards are not established, it seems very likely that 5G will offer network operators the ability to divide their networks up into discrete slices. The same physical network will be able to provide varying levels of performance to cater for different user cases.
This means that an operator could offer a dedicated network for IoT which provides connectivity for a huge number of devices transmitting tiny amounts of data periodically.
It could, concurrently, offer a fit-for-purpose high-bandwidth network in times of natural disasters, dedicated to emergency services traffic.
The same network operator could offer a dedicated network supporting self-driving cars.
If this potential is realised, it will be very interesting to watch as the commercial arrangements for access to these networks are organised.
But I will make the point that Australia is well-positioned to avoid access issues because of our strong infrastructure-based competition in the mobiles sector. We expect that this will be further strengthened with the expected entry of a fourth mobile operator.
If there is to be just one ‘takeaway message’ from my talk this afternoon, it is this: the ACCC has a clear agenda for mobiles.
We want to engage all involved to think more deeply about the impact of spectrum allocations on competition, particularly where spectrum is critical for new entrants and to building next generation networks. We will be strong advocates on this.
We will closely monitor whether our regulatory settings promote interconnection between networks, particularly between smaller fixed networks and larger mobile networks.
We want to see incentives for investing in regional mobile coverage and network quality maintained, and we want greater transparency. We are examining regulatory settings around facilities access.
And finally, we want the full benefits of 5G to be realised, and we want these benefits to be realised by all Australians through a competitive market.