Telecommunications (Security) Bill – in a Public Bill Committee at 2:44 pm on 19 January 2021.
We will now hear from Doug Brake, director of broadband and spectrum policy at the Information Technology and Innovation Foundation. We have until 4 o’clock for this session.
Good afternoon, Mr Brake. Will you introduce yourself for the record, please?
Doug Brake:
As you mentioned, my name is Doug Brake. I am the director of broadband and spectrum policy for the Information Technology and Innovation Foundation. We are a think-tank based in Washington DC, focused on policies that we believe advance innovation, with the basic belief that innovation is the key to economic growth and human flourishing over the long term.
Thank you. I call Sara Britcliffe.
Q I have a quick question that I asked the previous witnesses. What is it that you support in this Bill?
Doug Brake:
At a very high level, I would say cyber-security generally. The goal of Government intervention should be to make it easy, cheap and desirable for the private sector to do cyber-security well. I have some vague concerns that some increased costs might come from the Bill—the compliance costs—but identifying this as a serious issue that needs to be looked at and giving Ofcom the tools that it needs to investigate security challenges, especially with regard to the equipment and working with the private sector to mitigate those risks, is a big step forward.
On the diversification strategy, I think it is a very wise document. That to my mind is one of the best opportunities that we have to mitigate long-term risks, particularly where there are high-risk vendors in the area. So I think the diversification strategy is quite wise and would make the UK a real leader in this space in terms of policy.
What is your view on the implications of the Biden Administration for potential American involvement in the diversification strategy in telecoms more broadly?Q
Doug Brake:
That is a good question. A lot of people are asking that question and trying to figure out exactly where this will go. I think that at a high level we have passed through the confrontation with Huawei and China over some of these innovational mercantilist policies that we have seen, which have undermined the global innovation of wireless equipment. I don’t think that will change at a high level. No politician in Washington in the US wants to be seen as soft on China. I think there will continue to be policies that attempt to roll back some of the innovation mercantilism that we have seen in the wireless equipment space. I expect and hope that it will be done with a more measured and co-ordinated effort with like-minded allies such as the UK and with less scattershot policies across the US Government.
What we have seen over the last several years in the United States is a variety of different agencies doing what they can to mitigate the risks. It is less a co-ordinated whole of Government approach in the US and more a disjointed and fragmented policy response across different agencies, so I am hopeful that under a Biden Administration we will see a much more co-ordinated effort and one that is more co-operative with allies.
Q Regarding that co-operation between nations, clearly it has not just been the UK that has responded to the threat from Huawei; we were forced to go down our present track by the American sanctions against China over semiconductors. Going forward, how do you see that co-operation developing and how do we prevent a situation whereby, with its scale and obviously its resources, the United States dominates any future standards and also technology areas? They might talk about co-operation, which they do in a lot of areas, but when it comes down to brass tacks, it is America first in terms of business. And that is not a Trump comment.
Doug Brake:
It is a good question. To start with, I will take the first part of your question, with regard to the export controls that the Administration put in place with the aim of trying to kneecap Huawei; I think it is fair to say that.
First, from our perspective, ours was not a very well-thought-through strategy—right? Without co-ordination and without a broad coalition to address those sorts of trade practices, in effect in the US we really only shot ourselves in the foot. It undermined any of the technology companies or equipment providers that were attempting to sell components and chips to Huawei. So to my mind, if you are not going to succeed in killing Huawei, or if there are ineffective strategies that undermine your own industry, I am hopeful and expectant that we will see a change in the policy going forward.
That said, if there was a desire from a broader coalition internationally to make some more extensive efforts—something like a NATO for trade, to address these unfair practices—that could be a very effective strategy, if it was done with a broader coalition.
In answer to your second question, the long-term goal of diversification of the radio access network supply chain is to allow for a much more diverse and modular system, in which any number of companies can compete within different niche areas of the market. Admittedly, there are some areas of that—high-performance, generic server infrastructure, as well as software—that the US does quite well. However, I think that opening up the supply chain would allow for a number of companies internationally to compete quite strongly.
Also I think there is a question about the extent to which different countries are willing to aggressively pursue an industrial strategy to support the sort of change that could give them a potential comparative advantage in pursuing this sort of transformational change to the telecommunications supply chain.
Q Regarding the ban on semiconductors, obviously that is worth a lot to the US in terms of exports, mainly to China. Do you see that changing? Also, do you think there is a danger—not in the short term, obviously, but in the longer term—that China will then create a separate semiconductor market, which could be a threat not only to the US but to others, in that they will buy into markets and try to get standards that are different from those of the US?
Doug Brake:
I think it is absolutely right that there is a real risk if we cut off supply to China, particularly in semiconductors. We have already seen an aggressive action on their part to stand up an indigenous semiconductor industry. This is getting a little outside of my area of expertise; semiconductors is not some place that I know super well. However, I think that it is absolutely correct that there is a real risk that the extent to which we try to cut off Chinese companies will see them double down efforts to create their own indigenous supply chain. So—absolutely.
I am hopeful that we see either a change to that or a much broader international coalition to double down on those efforts. I think that it is more likely that we will see a Biden Administration ease some of those restrictions, or work through the current legal means to allow for licences for companies to sell semiconductors to Huawei and others.
During my previous career as a physics researcher many years ago, I was fortunate to work at places such as the Advanced Photon Source synchrotron facility in Argonne, Illinois. I worked extensively in the semiconductor space, looking at materials such gallium nitride and other group III nitrides. What I learned back then, working very extensively with American scientists and scientists from around the world, was the importance of that root-and-branch look at semiconductors, innovation and having a decades-long view. From your perspective, how much does that fit in with a joined-up international approach to create diversity, both at the end stage and at the really early research stageQ ?
Doug Brake:
That is absolutely right. This is a long-term effort. I worry about some who tout ORAN as something of a silver bullet that we can make a quick transition to, that it is a flash cut for existing equipment providers to an open RAN sort of system—a more modular and diverse ecosystem. It is something that is going to take a number of years. I honestly worry that it is late for ORAN to be incorporated into 5G, at least on a broad scale. For greenfield networks, it is a different story and it might make sense to go with these open and modular systems from the get-go.
I worry that this is much more a conversation about putting in the tools, resources, testing facilities, the labs, R&D, et cetera, to put us on a path for years down the road so that this becomes the industry standard. I do think, absolutely, that this is the time to be looking at those early stage investments to be driving further and, frankly, looking down the road to 6G, to be able to put in place the policies and efforts to transition the industry to this more diverse future, and put those in place now for years to come.
Thank you for coming in. A quick question: can you put in layman’s terms what the roll-out of 5G anywhere means in broadband terms? Can you also place that in terms of rural areasQ ?
Doug Brake:
I worry that sometimes 5G is conceptualised as a singular technology or a singular thing. It is not a monolith; there are a number of different component technologies and a number of different flavours. Depending on whether you are doing a fully 5G network, a stand-alone network or a non-stand-alone network, it is a very different sort of system. There are also a lot of differences between what spectrum is used to deploy the network—if you are using low-band, mid-band or high-band spectrum or a combination of all three. It is hard to answer that question in generalities.
A number of different component technologies and architectures will be rolled out over time. At a high level, the real advantage of 5G compared with 4G is in its flexibility. It is able to tailor its connectivity to a number of different applications’ needs. It can offer extremely high throughput and much faster speeds. It is very reliable, with very low latency. For example, if you want to stream a football match while travelling on a train, it can do that quite well, or quite a bit better than LTE and 4G today. At the same time, you can also change very obscure technical parameters to make for simple communications that require very little battery on the device side to be able to communicate. If you want to have massive deployments of sensors for smart agriculture, or something like that, that have battery life in the order of decades, it can do that. The hallmark is its flexibility.
Given that flexibility, it is anticipated that 5G is going to be much more deeply integrated within the economy and trade sectors, and will be a key tool to boost productivity. There is an important hope that we see a broad deployment, not just in urban areas but in rural areas. Again, I go back to that note on differences depending on the spectrum that is used to deploy—unless it is of interest, I do not want to get too bogged down in the details, but there are real differences in what we would expect to see deployed in urban versus rural areas. But, again, we would also expect to see very different use cases in those areas. Admittedly, there will likely be a performance difference between urban areas and more rural areas. But at the same time, like I said, the use cases look very different—you are not likely to have massive crowds of people all looking to share video from a stadium or something like that in rural areas. There will be a real difference in the roll-out, but I worry that sometimes the challenges with that have been overstated.
Q Thank you very much for joining us, Doug. It is particularly interesting to have your American perspective. As you may have heard, my first job as a hardware engineer was with Nortel, a Canadian-American company that had just bought one of the last UK companies in this area. Nortel ended up seeking chapter 11 and having most of its IP bought by Huawei. What are your views on how we got ourselves into this position of not having a single UK or US supplier that can supply to our UK networks? How do you think we can work together to rebuild our telecoms sectors? You talked about pursuing transformational change and an industrial strategy to do that. What might you say to the Minister about how the UK Government should be working with the US, and about what sorts of vehicles there might be to work together and with other allies to achieve that?
Doug Brake:
That is a great question. We talk now about needing diversification and seeking entry of a US-UK equipment supplier, but the question and lessons from history are about why we need this in the first place. In the past, we had quite successful telecommunications supply companies, especially in the US. The president of our organisation, Rob Atkinson, set out to answer that question. You may have seen an article in the American Affairs journal, titled, “Who Lost Lucent?” It is a long and interesting article—I will not go into all the details of history. I would say that it is fair to characterise the failures and decline of Lucent as a complicated story, but it stems from a combination of unique challenges imposed by the Anglo-American economic system, systemic failures of US Government policy—particularly with regards to anti-trust and some of the regulatory policy throughout the 1990s—and very strong and aggressive foreign industrial policies, particularly with regards to China, to acquire market share.
I am happy to go through that in some detail, but feel free to cut me off if I go on too long. You are absolutely right to say that we had Lucent and Nortel. Lucent was absolutely massive—it was three times larger than Nortel—and originally spun off from AT&T’s equipment arm, Western Electric. It had the famous Bell Labs. Throughout the ’90s, it was the largest telecoms equipment company and was still growing dramatically overseas, but due to a number of strategic decisions within the company and decisions within the US Government, it ended up really suffering as a result of the dot.com bubble.
Setting aside all the competitiveness questions, particularly with regards to Chinese companies, a hands-off, free market globalised system reigned in the US and UK throughout the ’90s. It was finance-focused capitalism that saw Lucent and Nortel cut their R&D budgets and staff dramatically, particularly as a result of the 2001 crash—much more so than some of their international competitors. With that financial system, it was harder for those companies, which were designed to be growth companies—much more so than a valued company. They were focused on growing quarter after quarter and meeting their financial targets, which made it very difficult to focus on long-term growth. You can contrast that with Ericsson in Sweden, where the Wallenberg family control a lot of the voting shares. Ericsson was able to focus on much longer-term value creation, and they did not cut staff or R&D by nearly as much as Lucent did.
Before that, I think there are a lot of lessons to be learned from the aggressive anti-trust action that broke up Bell Labs and restructured the entire industry. Up until the restructuring of the US telecom market in 1984, Bell Labs had a fantastic situation in order to generate innovation. It had the commercial drive, focus and flexibility that is often lacking in a Government research lab. It also had a long-term focus and an interest in broad technological change, which many R&D efforts in industry do not see. It had steady revenue from telecom rates. There is a complicated story there. It is hard to tell what concentration is good for innovation and where competition is really the order of the day, but it seems clear that the decline of Bell Labs was a real loss.
Q Perhaps you could contribute the article you mentioned and the key points made as written evidence.
Q The second part of my question was about how we can work together.
Doug Brake:
Absolutely. I think the diversification strategy is a very strong document. I would say, when it comes to open RAN generally, there are clear benefits that you have heard a lot about, I am sure, including diversification and faster innovation when software is decoupled from hardware. Generally, lower margins on generic components eliminate the risk of the entire sector tipping to a single vendor or a gradual narrowing of trusted suppliers, but there are real challenges with this process. Again, this is going to be a gradual effort. There is not a need to transition immediately.
First, there is a real risk of bandwagoning, where this is seen as a silver bullet and even companies that might not be interested in pursuing this area, such as Nokia and Ericsson, are willing to join in these efforts, even if it is just for the sake of defence. So, there is a real risk of bandwagoning. There is real complexity with transitioning to this sort of system. It is not immediately clear how well open RAN will scale. Actual implementation at scale in urban areas is adding a tremendous amount of complexity. There is a much larger attack surface. It is worth keeping in mind SolarWinds, a US company trusted by many within the Government, which saw this massive damaging breach.
I think there is a real challenge that remains to be addressed in the manufacturing of stand-alone radios. I think that is a potential opportunity for real co-operation: identifying companies that are interested in focusing purely on radio. There is still hardware that needs to be provided that historically was integrated with the broader system, when you only have relatively small providers that are interested in scaling up manufacturing.
I am just going to interrupt you there. I am sorry, but I am conscious of time and I want to give the Minister a fair opportunity.
Q Thank you for that interesting evidence. This follows on from Chi’s question, in a sense. I think it goes without saying that we will be very keen to work with America in the future, and the opportunity to do so is significant. I have held conversations with my Swedish and Finnish counterparts, but when it comes to further international co-operation, if you were in my position, where would you look to first? Do you think the Bill is forward leaning enough to bring that collaboration up a gear, if that makes sense?
Doug Brake:
I think there are two different opportunities. First, in the efforts of diversification, this is necessarily a globalised sector. The incumbents are massive companies with huge global economies of scale, so in order to transform the industry structure, it is going to have to be a global effort. We need all the countries aiming in this general strategic direction.
I think the document is sufficiently forward leaning. At a high level, one of the most important first steps is identifying this as a strategic imperative—that this is a goal that is shared by Governments across the world—and taking a genuine interest and focus, especially on the level of venture capital investment. Just the creation of the document is a hugely important first step. As for continued research, the real focus is on research and development and test beds. They are the key tools that we need to test and scale up, to identify real challenges and complexity.
I am not sure if this quite fits the answer, but there is a challenge around systems integration. We need to identify real leaders in systems integration. When you have real risk in pulling together different components from different suppliers, into what is essentially critical infrastructure, the risk of failure—at least, the downsides of failure—is extreme, so operators are often eager to have a single company that they can go to if something goes wrong, which can integrate all the different components. There is an important opportunity, to the extent that policy can help support those efforts.
There is all sorts of opportunity for global collaboration and for rowing towards the direction of this diverse supply chain. I think you have put together a very thoughtful piece in moving that forward. Then again, I go back to saying that this is not a silver bullet in addressing the long-term challenges around innovation mercantilism from China and Chinese companies. I think there should be more co-ordination and collaboration, especially when it comes to trade policy. Again, this is outside my area of expertise—I am 5G, specifically—but the more we can co-ordinate to be honest and up front about the real challenges and work to scale back the problem, the better.
Q On that long-term point, do you think that both the Bill and the strategy allow Government to retain the kind of flexibility we will need for them to persist for the long term? We do not want to be revisiting primary legislation too often on this, even though we obviously need to keep aspects of it under review?
Doug Brake:
I think that this is absolutely the right direction to be moving in. Clearly, you need the tools to be able to analyse the risk, identify high-risk vendors and work away from potential security risks associated with that. So, absolutely, you need the tools, but there is always a broad challenge when it comes to cyber-security of the negative extra challenges, where private-sector providers might not always face all the downside of cyber-security breaches.
You can solve that by increasing the cost and increasing the downside to cyber-security risk. I think it is much wiser to help work with Government to lower the cost of doing cyber-security well. The UK, from what I can tell, is a real leader in this regard, setting up NCSC. To be able to work closely the private sector, to identify those risks and eliminate them, is much better than just turning up the dial on the downside to cyber-security breaches, or things of that nature.
I would tweak the Bill in that direction. I guess much of this can be done through implementing regulations, but, to my mind, focus more on collaboration and co-ordination with the private sector, rather than simply increasing the downside as well as the compliance costs with the legislation.
I think that brings us virtually to time. Thank you, Mr Brake, for your evidence. That was the final evidence session for the Bill, so I thank all the witnesses. The Committee meets again on Thursday morning for line-by-line consideration. I believe that will be at 11.30 am in Committee Room 14.