Tech and the future of the world economy
01:02:06 Share

Welcome to the LSE Events podcast by the London School of Economics and Political Science. Get ready to hear from some of the most influential international figures in the social sciences. Well, good evening. Good evening and welcome to the LSE for tonight's event. This event forms part of the LSE Festival, which is our week-long celebration of social sciences. It runs from Monday up until Saturday.

We have a series of events this year on the threats and opportunities of the near and distant future and what a better world could look like. So my name is Neil Lee, I'm Professor of Economic Geography at LSE. And today we're talking about, I guess, a topic which is vital to our futures, a topic which is vital to our economic futures and maybe for the future of the world economy. Because what we can see is that the tech sector is changing fast, new tech firms are reshaping the world economy,

In the Draghi report, Draghi claimed that the entire difference in productivity between the United States and Europe came from the tech sector. I'm not sure that's true, but it certainly shows the significance of what we're talking about today. So we'll be asking questions about what holds advanced economies from forming world-class tech firms.

can and should they try to catch up? And I'm very interested to think about what policy can do, what entrepreneurs can do, and what academia can do. And reflecting this, we have representatives, we have a brilliant panel today, who represent each of these different constituencies.

So first off, we have Stan Boland. Stan is a British entrepreneur in the information technology sector. He has founded, led and sold three technology startups. I'm going to get the numbers right here. Raising $330 million of venture capital, which feels like a lot, and building global businesses bought for $1.3 billion.

So we're very pleased to welcome Stan today. I actually bought him lunch in the LSE staff canteen not knowing this. This is like taking him to Gregg's or something like that. But he's very, very generously agreed to come back. He was CEO of the pioneer computer company Acorn, helped take ARM public in 1998. Over the following 25 years, he founded and led two of Europe's most successful venture-backed semiconductor companies, selling them to Broadcom and NVIDIA.

He has led an AI startup that built a self-driving car and has pivoted to simulation technology for self-driving software, which he sold to Bosch in 2022.

He's a Cambridge physics graduate and has worked at Rolls-Royce and ICL. He's an active angel investor in technology startups, a board member of AI semiconductor company Fractile, and at Venture Accelerator Deep Tech Labs. We're very, very pleased to welcome him. Also very pleased to welcome Dr. Robin Klingler-Vidra. She's a reader in political economy and entrepreneurship at King's Business School. She's the author of Startup Capitalism. It's a new book. It's so good, I bought two copies. Okay. Okay.

and also the venture capital state, the Silicon Valley model in East Asia. Robin's research focuses on entrepreneurship, innovation and venture capital, and she's led teams, cross-country research studies for Innovate UK, UNDP, the Asian Productivity Organization. She'll be bringing a sort of global view to the discussion today. Also delighted to be joined by Kanishka Narayan, Labour MP for the Vale of Glamorgan.

But more than that, he's had a career which bridges public service, technology and finance, working in Silicon Valley and the civil service, which must be quite a contrast, I would imagine. In government, he's advised ministers on domestic public policy, a senior advisor to the Cabinet Office, an expert advisor to Britain's environment secretary. He's advised the Labour Party front bench as head of tech policy.

And, you know, more than that, worked with advised FTSE boards, worked with startup founders at Lazard. And he grew up and lives in South Wales, studied at Oxford and Stanford, and is passionate about social mobility and social justice. And he's very vocal and very optimistic and an interesting speaker on the tech economy. So for those of you who are social media users in the audience, the hashtag is LSEfestival.

Please can you put your phones on silent if you haven't already? That's the panel as well as the audience. The event is being recorded and will hopefully be made available as a podcast. A few points on what we're going to be doing today. First of all, each of our speakers is going to be opening with a sort of brief discussion. Then we'll have some discussion amongst ourselves and then there will be plenty of time for questions with the audience. So if you're in the room...

There will be a microphone roving. And if you are online, you can submit questions via the Q&A. Please include your name and affiliation, and we will make sure that those are also read out and responded to. So thank you very much. We start off with Stan.

Well, thanks, Neil, and thanks for the invite today. I probably wouldn't have had a pudding, to be honest, if I knew you were going to bring it up, actually, the lunch that you bought me, but it was very good. So I think the UK obviously faces a number of challenges as an economy, from the lack of investment over decades in infrastructure to the

demographics that we got to the availability and cost of housing. But there's no challenge more serious, I think, than the decline and anemia in our industrial base. And, yeah, the real question is, yeah, where are the large, successful, profitable businesses

uh, companies in the UK. Uh, how are we going to create jobs? How are we going to create household income? How are we going to create tax base? How are we going to create national wealth unless we have them really? Um, so this is not really about building the past. I mean, I think it's quite hard to imagine building a ship

building industry or reconstitute a large car industry in the UK or a steel industry. But I think it is really about building companies that are going to create competitive products, innovative companies based on technology and the assets that we've got in the UK that allow the UK to reach its potential, allows us to sort of create the lifestyle and the living standards that we all want in the country.

So the UK government is not blind to this, really. It's investing something like, yeah, I calculated, at least $25 billion a year in supporting industry, the technology industry. But I'd argue that quite a lot of that is kind of wasted, unfortunately.

So it spends something like 12.5 billion in UKRI grants and so on, about 10 billion in R&D tax credits in dollar terms, CDIS, EIS, VCT, patent box, about 3.5 billion.

And you'd expect from the 250 billion it's going to spend over 10 years, assets that we can point out that are worth like trillions of dollars in the UK in the tech space. When I look at the permanent UK tech scene, I can only add up companies worth about 86 billion.

So we seem to be short compared to the US, about 3.9 trillion of value in the UK. Our biggest tech company is Sage, which is worth about $16 billion. And I have, of course, excluded Arm from that. Arm is listed in the States and is owned by Japan. So it's not really easy to count it as a UK company anymore.

And so I'd argue that something's seriously wrong with our abilities to sort of create and to nurture and to mint and to keep technology companies in the UK. And that we really ought to address what the kind of causes of that are. So Neil very kindly sort of mentioned my resume. The first company I did, yeah, raised venture capital for, yeah, did have a US investor there. And in the process of getting a US investor, I think I got like 70 introductions to VP, CEO,

engineering or CTOs in my first day from this investor and did 26 trips to the States in a single year to go visit customers to try and figure out what the product is. Lifted teams, they gave us money, they gave us confidence.

And we started to get interest from customers, endorsement from customers. Within about a year, companies were offering to buy our fledgling company for $100 million. Then we got an offer to sell it for $300 million, a point to the bank, and ended up selling it for $640 million after about a year and a bit, returning 32X to investors. So, I mean, not surprisingly, first-time founder, we decided that we'd bank that 32X because we could always create another one.

So the second company was much more focused on let's create something that's here permanently as a UK permanent entity that is a firmament in technology in the UK. And, you know, that did take longer. It took about nine years, raised about 250 million, had all sorts of challenges competing with the US. So we competed with a US company called Qualcomm.

And I think in the process of doing that, it became obvious to me that all of our investors were tapped out. So we'd raised like 50 million from each of the major investors. And that's pretty much the limit of what they could put in. And that we really needed to raise another 250 million to sort of

sort of punch all the way through and keep this thing. But yeah, like many startup entrepreneurs face the prospect of either we migrate everybody to the States or we sell the company really. And, uh, and, and, and so for a completely different set of reasons, um, end up selling a company. Um, and,

And the third company I did, I won't go through the whole story, but we ended up selling that company as well. So I've actually kind of built and ended up selling three companies. So in a way, it's a kind of microcosm of the kind of challenge we've got really in founding but also scaling companies. So I think, I mean, the reflection I've had is really comparing what the UK looks like compared to the US, right?

I think if you look at the U.S., there are four things I think the U.S. is doing particularly well. One is it's paying a lot of attention to talent pipeline, not only its own talent pipeline, but attracting the very best in the world to come to that country, stay in the country, and making it desirable and easy for them to do that, really. The second thing that the U.S. is doing well is putting very early stage programs

around companies. So getting experienced operators and entrepreneurs around the table at the very early stage. And it could be in a campus, could be Stanford, but it could also be having partners in venture firms that are actually practically helping those companies early on.

The third thing that the U.S. is doing well is once companies are finding product market fit, it's able to write big checks into those companies. And the writing of a big check at the point of reaching product market fit is an anointer of success. So whereas in the U.K., a Series A company might raise $30 million.

In the US, that company is probably going to raise 150. And compared to the US company, there's only kind of one winner, and it's not the UK company. So, you know, we have to get real with the world the way it is. And weight of capital at the right point is a critical success factor for us here.

And then I think the fourth thing that the US does well is it doesn't put very many props in place for companies that are not succeeding. So there is no helicopter money of any sort, really. There is some R&D tax credit, but it's roughly half the level of the UK. And there are no schemes that are particularly designed to kind of prop up

companies that actually should fail because they're tying up precious resource in things that are never going to be successful and the lack of props is a clear US lesson there. And if I compare the UK, there's just two of those I'll just pick up on. One is the availability of capital and

The U.S. venture capital firms raised $76.1 billion last year in new money, commitments from LPs into new funds.

And the UK is roughly a fifth of the size population to the US. So the UK should be raising about $15 billion a year in venture capital. But UK firms only raised just over 5 billion last year in the UK. So we're running about a third of the level of venture capital in the UK compared to our US counterparts. And that is one major reason why we're not able to anoint the winner and to sort of give companies the legs to be successful and given the growth capital.

And the second thing is the props that we put in place in the UK. It has the effect, I think, of keeping companies going when they ought to be stopped, really. And therefore, there is a strategy which is open. The great thing about all these analysis is that they're actually very, very easy to fix if we're determined to fix them.

but the strategy you can imagine adopting is that we switch something like at least $5 billion a year, but maybe it's a bit more than that, from power

passive money into active money so in other words we reduce the props that we give to companies and instead we put that money into fund to fund investments into venture in the uk so not only would that nominally reduce public spending and because we can count the fund to fund investment as a national asset so it's not departmental spending anymore

But secondly, it can be put in on a basis that effectively doubles its size, really. So we put 50% into a maximum interest fund and the other half, we were creative in how we bring the rest of that capital in. But we can therefore mint about $10 billion a year of new capital.

And in doing so, we will have fewer zombies. We'll have the ability to write checks into companies at the point they reach product market fit. And we'll end up creating something that looks a lot like the Boston or the New York or the Silicon Valley ecosystem here in the UK. And that strategy seems to me to be wide open for the UK to adopt, especially with fiscal rules that have just been starting to be exercised by Rachel Reeves and the Labour Party team.

So I'll leave it at that and we'll hopefully cover some of these points in discussion later. Wonderful. Thank you very much, Robin. Great. So I'll pick up on some of the points and try to weave in to what Stan has sketched. So, I mean, first, I suppose when we talk about the role of tech and technology,

the model or the reference point. And as you mentioned in referencing the U.S., we think about this sort of stylized version of the Silicon Valley narrative that's about, you know, the founders and garages and the sort of David going up against Goliath. But my book, which is sort of featuring on different tables up here, focuses on

models around the world. And so maybe to make links to here in the UK, my co-author and I have spent the last 10 years looking at the approach to supporting startup ecosystem growth, especially in Northeast Asia, and comparing that and contrasting that with this stylized version of Silicon Valley. And I keep saying stylized because Silicon

What we think of as Silicon Valley and this disruptive outsider, sort of Steve Jobs type of model of narrative, I argue hasn't been the case, at least for the last 30 years, but probably longer.

In Northeast Asia, especially in Japan and Korea, the model is fundamentally different from this stylized version, but then also from the approach that the UK is taking. So effectively, David and Goliath are thought of as collaborators, right? And it's a whole of system, whole of state approach.

where startup policies are written and the idea of venture capital and scale-ups is doing so in partnership with national lead firms. So in Korea, you think about Samsung, Hyundai, LG, the Chebel. In Japan, you think about the Kuratsu and you think about also the large banks like, say, Mitsui, Sumitomo.

And the idea is that the corporate venture capital money is coming from these large firms and that they are key anchors. They are underwriting effectively the startups in the ecosystem. But there's also a catch.

It's not a sort of altruistic, these large firms are playing a national interest or social story here. One of my favorite interview quotes is hearing from a government official in Seoul. We ask in all of the interviews over the course of about a decade, you know, what are your KPIs? Why are you supporting startups? And one of my favorite quotes, I'll tell you the other one in a minute, but one of my favorite quotes was,

We are supporting startups because they are able to inject innovative DNA into the large firms, into the chubble.

And so it's not the startup as the benefactor of government support. In fact, startups are resources for the large firms and they're going to bring new talent, new ideas, a new model, a new way of thinking, of organizing. And that's institutionalized across government policy to support startups, especially in Japan and Korea, globally.

Where you have, for example, in the J Startup Initiative, so METI, which is one of these government agencies that's famed in academia because it's this bureaucracy that helped to lead the post-war miracle.

METI in 2018 said, "We're going to work with other ministries to create 20 unicorns," privately held high-growth companies that achieve a valuation in excess of a billion dollars. The judges for who is accepted to the J Startup initiative are the large firms and the main banks.

And so the idea that the startups that the government is going to support and then invest in through the GPIF, right, the massive public pension funds, the idea that the startups are going to disrupt or cannibalize or challenge, as you would imagine a typical stylized Silicon Valley narrative or David versus Goliath is sort of a fallacy, right? It's almost, well, why would we do that? We are competing against

in a global market. We are competing internationally. And for us to have an advantage for us to win, the national champions, along with the startups, are much stronger if we are working together. So it's a story of David and Goliath sort of, you know, arm and arm, if you will,

Not to say that that's not without problems, right? There's issues of this potential sort of who's setting the direction of innovation and thinking about technology and the future of the world economy. If startups are not challenging the incumbent firms and their technologies and they're complementary, then they're at the sort of behest or at the service of complementary. So if you think from a really disruptive sort of Schumpeterian perspective, you

Maybe one of the reasons that we're not seeing that business dynamism that you sketched, Stan, and that we're not seeing these breakthrough technologies at the same pace is perhaps because the ecosystem, even though the language is one of disruption and unicorns and innovation and startups positioned in this way, they're ultimately not necessarily trying to drive a revolution or to change

challenge the prevailing narratives or power. I mean, interestingly, especially in the U.S., the business dynamism and people like Ryan Decker and others have shown that business dynamism is down in the last 30 years. So I think that idea that the sort of KPIs of why we're supporting technology startups, a lot of it is

A lot of it is this idea that we want to encourage entrepreneurial culture. We want more startups. And this rising tide maybe will raise all boats. And relatedly, I'll tell you my second favorite interview quote, which was not a quote that I expected to hear in an interview asking about government support for startups. I heard from an entrepreneur in Korea that in Korea, to be an entrepreneur, you have to kill two women.

I'm like, did I ask the question that I thought I asked? And then he went on to explain that the sort of adage or the idea is that culturally the idea of, you know, coming home and telling your parents that you're going to be an entrepreneur, not a doctor, not a lawyer, not working at Samsung would kill your mother.

And for your future wife, the social shame and the missed income will be social sort of problems for you. And he explained that this is what the government really is working on, right? And trying to drive this cultural change. But I think part of the problem

Maybe the problem of that is creating or sort of fostering this idea that, you know, being an entrepreneur and maybe being an entrepreneur in an existing orbit is to be lauded.

But then perhaps that there's more of that need to sort of drive the upper out almost to use the McKinsey expression. Another thing that I'd like to touch upon is the social purpose, let's say, of startups and technology startups. In our work, we...

in addition to hearing interesting quotes about killing people. We tried to map on policies, media statements, and then in the interviews, is it about domestic aims? Is it to boost economic growth, innovation?

Maybe increasingly about sort of inclusion and distribution, either geographically or bringing populations that are otherwise sort of marginalized or excluded in these ecosystems, about bringing them in, so sort of domestic populations.

Or is it about the external, the sort of national security imperative? And I think this is something that, especially in the last 10 years and maybe even much more recently, we see this rise of that external imperative. What's the point of technology for a country? One, it makes

maybe long been about innovation and growth and creating good jobs, but increasingly it's about a sort of zero sum. We need it to enable our stature from a national security perspective to

And you see that in the U.S. increasingly also with the, I mean, Andreessen Horowitz, I think, has the most explicit fund objective now. And I just wrote down, not to read for long, but their American Dynamism Fund sort of goes back to the origins of Silicon Valley in the military imperative. So the National Dynamism Fund supports the national interest. Fine.

And then they go on to say that they invest in aerospace, defense, public safety, education, industrials and manufacturing. And then to go back to the Northeast Asia, this idea of David and Goliath together, especially around, you mentioned smart shipbuilding technology.

cutting edge, smart manufacturing. This is a business where the whole of system approach works really well. The small, you know, small capitalized, let's say, or, you know, seed funded startup is

can't and won't compete in chips and the sort of sectors that you know very well, right? So it's thinking about how do we invite and how do we involve startups and innovative talent in a way that they can bring this innovative DNA into the prevailing or towers of the economy. And I just wanted to close on, I've tried to map it onto three Ds. We'll see what you think.

So first, there's this direction, right, the point of supporting startups. I would argue increasingly that national security imperative is changing what we think about as the military-industrial complex, that it's not about the sort of, you know, in the American setting, you have four or five companies like Lockheed Martin and others who are sort of

around NASA, for example, and around the sort of military imperative that is increasingly porous and supported and enabled by startups. And the Andreessen Horowitz example, I think it's a good one, how venture capital is increasingly explicitly there as well. So there's a direction to innovation and to the support of technology.

Second, this sort of dedication, if you will. And I wanted to touch upon this because I think this is increasingly in the public domain and debate here in the UK and also in the US. Part of what we look at in my work, we look at, you know, government support and the startup culture.

and outcomes in China. And, you know, now the idea, the Chinese saying for the tech work-life balance, which is not about balance, right? The 996, right? That you work 9 a.m. to 9 p.m. six days a week. That's sort of caught on here. And now there's maybe some push to sort of work harder and be more intense, that dedication. But then also equally a debate around, you know, maybe

mental health and balance? And is this something that we want? But I would say, I mean, actually, in the last five years already, China has moved on to 007, right? 24 hours a day, seven days a week. So it feels like we're sort of passé. And then the third D, so direction, dedication, and then the third one,

disruption piece, and it sort of brings me back to the point of technology startups. Is it about disrupting? Is it about driving a sort of revolution of the prevailing incumbent firms and technologies? Or rather, is it about enabling and keeping the persistence of national competitive advantage and that international stature of your lead firms? Wonderful. Thank you. Kanishka?

Thanks, Neil. And thanks, everyone, for making time. Let me keep it relatively brief and focus on three points. The first of which is that to jump to the prescription, you have to have the right diagnosis. And I think as part of that, it's really important to think of a range of different things that we would want a technology sector, and in particular, a subset of that, a venture-backed technology sector to achieve specifically. And I think these are a lot of the points that Stan and Robin have picked up particularly. But there are four

objectives, some of which at times have trade-offs in them, that we might want a technology sector in the UK to pursue for us. The first of those is a material impact on the primary crisis in this country, which is a crisis of productivity. We know that from the United States, venture-backed financing in particular is a great tool for higher R&D intensity of firms. It is also vastly overrepresented in public market exits.

So we know that firms which do really well in terms of scale and in particular in terms of research and development intensity, ultimately a driver of productivity, but those firms are often venture-backed. And so we might think that venture-backed technology is a great way of pursuing what Draghi saw as the mission in the European context.

The second reason is a different one, which is that in the UK in particular, the reality of it is that we haven't had a long track record of venture-backed financing of technology companies materially moving the needle for the R&D intensity of this country. And so the first order effect hasn't really been present by virtue of depth and longitudinal sustenance.

But the second order effect is another thing that we might be interested in, which is that we want technology not just to make a small number of firms and a small number of people better off, but to move the needle on productivity overall. And to do that, the adoption of technology, the ability for startups to partner with national champions and drive scale of adoption such that we're more productive working together is a really important part of it. And so you might want that. And if you want that, then the model might be less startup disruption. It might be a bit more industrial collaboration.

There's a third objective, which is that we want to make sure that technology in this country doesn't just do the things that it has done elsewhere, but does the thing that our values demand of it. And in particular, the question of technology, I think the central political challenge for it is to make the argument that technology is a good thing, both for growth and for distribution.

distribution across place, because the last wave of software innovation and enterprise software was very focused on agglomeration effects in a very small number of places for a relatively small cluster of people. I think the opportunity in a world where energy and land and water and some of these other resources are the biting constraints is to spread the opportunity more widely in terms of geography.

but also distribution in terms of wealth. If you think about the United States, there's some evidence that once you adopt employee stock option ownership and you adjust for that, actually technology in some areas can have a flattening effect in a localized context if you are spreading the base of stock option ownership widely enough and those firms are doing well and many people do well with them, not just a concentrated set. So thinking about distribution, I think, is a really important part if we want to make sure that the future of technology is a British

for us. And then the final one, exactly as Robin said, is that increasingly technology is not just a question of economics, it's a question of resilience and national security as well.

And I just said that out in terms of what we're shooting for, because there are times when the model you might want to pursue varies depending on the weighting you have across those different things. A lot of this conversation is focused on kind of the first two of those things, the drivers of growth and productivity, both first order directly and second order through adoption. And so I'll focus some remarks briefly on that.

I just have two things to say on that. The first of which is that maybe in the slight interest of provocation, I'd say that it is my belief, I think I've said this to Stan directly as well, it is my belief, I used to run the European part of a US venture firm, so I say it as someone who's valued the role of finance, and it might be heretical to say it in a building that I

I think I understand teachers finance, but finance is contingent. The fundamentals are almost never the availability of capital in a internationally mobile capital market like this. The fundamentals are often more fixed to do with culture, people and geography.

If you look at the fact that in our public markets, for example, valuations across sectors, but including in technology, materially lag the return on equity for those sectors and certainly US comparators, the biggest driver of that isn't really the amount of liquidity in the context. That's definitely part of it. But the liquidity is a symptom of something deeper. And that is a growth rate nationally that has been extremely low relative to comparable countries where people can invest their capital. If you just do a basic analysis

DCF model, if you're doing a tech investor model, you'll realize very quickly that if you toggle between the US and UK growth rate, that pretty much largely explains most of the valuation difference that you'll be willing to pay. So for me, a fundamental starting point is what we can do to move the needle on the underlying growth rate in this country. Capital supply might be something that kind of gets the flywheel going, but we've got to be very clear that that is not the fundamental. It is a contingent variable that is dependent on the other fundamentals. That

There are three other things that I think are really critical. The second of which is, in addition to national economic growth rate, the second of which is this perception and, frankly, reality that the TAM, the market size in this country, is constrained. The United States is really big in large part because of this virtual cycle, not just in terms of scale of country, in terms of population, as Dan mentioned, but the scale, the revenue pool that people can go after is radically different. Now, I don't think TAM is a necessary constraint. Market size is not a necessary constraint.

But it is a real constraint for European firms. The reason why our biggest company in technology is still Sage, decades old, rather than a more modern company is because we haven't been able to start by selling internationally in many cases in the way that others have. If you look, for example, at Israel, you can see that market size is not a necessary part of your destiny if you're selling internationally day one. And so I feel a big part of how we've built up a culture of

here is by starting in this relatively mid-sized market and then a Series C or so graduating to selling internationally, that is not a necessary constraint. And I think that is a constraint that we should do everything to try and tackle. I think the third thing is, in addition to the pool of revenue, the market size, is the propensity of the buyer to buy, take

technology quickly. And this is partly to do with capital investment in private firms and scale firms more generally, but it is specifically to do with the culture of productivity enhancements in British firms. And that is a sectoral thing. If you've got lots of software companies, they're more likely to buy them in the first place. It's a cyclical thing. But in part, it's also just a feature of how we've built up management culture in this country as well.

So again, I think that's a fundamental challenge, but not a necessary constraint. And then the fourth thing, more generally on that point of culture, is the individual level thing, which is that when I was sitting in a room like this as an undergraduate, I

Almost no one in my class of 2011 graduating from undergrad wanted to go and run a startup. Most of us wanted to go and work in either services or the civil service or kind of a context which wasn't really fundamentally focused on taking risk and creating technological innovation. Now, I know that has moved on a bit, but I'd be really interested in drinks afterwards to get a sense of what proportion of the people in this room are willing to do what I think a very, very large proportion in my business school class at Stanford

And so if you take as given the capitalist contingent, the four fundamental constraints are our economic growth rate, the size of our market, the propensity to buy new technology, and at the very heart of it, the ability to buy new technology.

in each of our individual hearts, our willingness to take risk in this country. I think if we focus on those, we can very, very radically move the needle for technology more generally. And so the final and third point of evangelizing I'll end on is that I think we have everything in this country to massively punch above our weight. We have an exceptional talent base. Our university context continues, not least in this room and beyond.

to massively outperform other universities, not least on a per capita spending basis. So we have in the context, if you look at, for example, AI, we have second or third largest number of significant machine learning systems developed in this country relative to any other country. Only the United States and China are frankly comparable.

People often say to me, you don't have a Frontier Lab company in the LLM space. Well, if you speak to the Gemini team, you'll realize most of them are sitting in King's Cross. And so I am more focused on where the talent is, where the second order ecosystem effects are, rather than where the necessary headquarter or the capital supply ends up being. And then frankly, even on capital supply, you know, one of the things I'll recall is that

I came back in 2021 to start the European practice for the firm. And in that same week, about 10 US firms came here as well. And if you today want to raise the scale of capital that Stan was raising at a much more difficult time. So huge kudos to him for building the scale that he built in times which were much tougher. But if you're trying to do that today, you can go to, you know, I'm just counting SoftBank, Mubadala, MGX, Temasek, GIC, CPPIB, Zipod.

Ontario teachers, Daniel Eck has today invested 600 million in health. I mean, you're going to be extremely spoiled in terms of the choice you have to raise scale of capital. And by the way, you can go and do all of that within a 20 minute walk of each other. And so I think what we have is genuinely exceptional in terms of capital, in terms of community, in terms of compute, the three fundamental inputs into AI in particular. Yeah.

And the missing piece, to the extent there is one, is a willingness across each of our individual parts to raise the volatility we're willing to endure in our lives. Wonderful. Thank you, all three of you. I'm going to go straight to questions. I have lots of questions, but I'm not going to ask them. I'm going to give you guys a chance. So if you could put your hands up if you have questions. Hi, I'm interrupting this event to tell you about another awesome LSE podcast that we think you'd enjoy.

LSE IQ asks social scientists and other experts to answer one intelligent question. Like, why do people believe in conspiracy theories? Or, can we afford the super rich? Come check us out. Just search for LSE IQ wherever you get your podcasts.

Now, back to the event. When you are giving your question, please keep it short. No biographies, please. And please try and make sure there is actually a question there as well. So we'll start off with the woman who had her hand up at the first start, which is, yeah. What's your advice for people, for founders, entrepreneurs, pre-product market fit? Because I know you've talked a lot about when you find product market fit, the US is great. But a lot of my friends have literally left to go to the States. But what's your advice for them?

But like, just frankly, what's your advice for pre-product market fit founders? That's great. And we'll go gentleman at the back. And this is, we'll take the one at the back, which is closer to you. Thank you, Ian Brown. I'm a self-employed consultant. Nobody mentioned antitrust. And I wonder how important you think that is and where you would situate the UK now between the US and say the EU and some of the East Asian states on antitrust and the contribution to competition in the economies.

Okay, thank you. And I'll just go for the gentleman on the edge here. Hello, Ben Hawes. I'm a tech policy consultant. Hi, Ian. And I just want to pick up on the last point about adoption of innovations across the economy. You know, we have particular sectors where there's huge demand like construction, but productivity improvement has been really, really slow. What is the role for government?

in improving uptake, not improving the source of innovation. And actually, what works? Because I'd say I was a technology policy civil servant, mainly under governments that were very laissez-faire and just expected that competition would drive uptake of innovation far more than it actually does. But I think the interesting thing is that obviously, if you don't do that for government for a long time, you don't know what works. Thank you. Admirably concise. Stan, you're first.

Yeah. Well, the first question was about what advice would you give to pre-product market fit founders, I think. I'd like to say find another potential co-founder and spend time incubating possible ideas and then go and research

raise some pre-seed capital from angel investors that allow you to sort of leapfrog that into a company and a plausible idea that you can get ready to kind of pitch to investors in the seed round and get building those companies here. I do think, however, though, we do need to work harder at

fixing the VC ecosystem in the UK because it is not as well developed or as capable as the US. And there is an opportunity, I think, for us to do significantly more here at this time, really, to bring investors in US firms across the Atlantic and to sort of, you know, if we could...

demonstrate the talent that's here and we can also demonstrate the ability to raise funds here, I do think actually we can transform the ability of the UK to be the place that people decide to sort of set up their companies. And I think in the space of AI, for example, I think we are just about holding our own in terms of talent here in AI, but that is very much a kind of net-net. We're losing a lot of talent to the US now

And we are gathering other talent from other parts of Europe. So we are roughly net-net. But in my view, we're a tenth of what we should be in these markets. And I do think availability of strong partners in venture firms...

and the capital that allows them to write checks into those companies is one of the kind of key ingredients that is not really in place here and needs to be pretty urgently. I actually think our hair's on fire on this issue and that it's too complacent to think that we can just raise capital from anywhere in the world because I actually think...

Actually, the causality is actually the other way around a bit surprisingly. So I think the lesson we should learn is really from China. And actually what China did to build a vibrant tech sector is it studied the U.S.,

And rather than say, oh, we all know that capital will flow to good ideas and therefore our challenges, you know, we just need to generate those ideas to show energy behind them. And then the capital will flow, which is a traditional way of thinking. I actually think the reality is that if you put the capital in place, I think great founders and great ideas will rise to the occasion. It's what actually happens.

And the quid pro quo is if there is not enough capital in place or the capital is too constrained, I guess a classic would be our EIS funds, for example, where the mantra is, for God's sake, don't lose my capital. This is not swinging by the fences. This is, you know, give me a 1x return, I'm happy. That's not the capital that we want. That if we...

Yeah, we have, I think, the potential, I think, to sort of build some very large companies here and be the magnet in which this gets done, this side of the pond, actually. And the time to do that, I think, is now before it gets too late for us. Great. Thank you. Robin?

Especially on the government policy and what works. I mean, and just on that, we have the potential to be the magnet. I often think that this idea that we increasingly hear echoed in government in the UK is that we have a scale-up problem. And I think the framing is all wrong. And I think that it's been mentioned across the panel. The UK is tremendous, and this room is a great example of attracting world-class talent. We don't have a shortage of

great ideas and great founders, the big capital for them to grow. And to your point about the Israel and the sort of born global mindset. So I want to reframe it as the UK is a springboard and it's sort of USP globally is that the UK is a phenomenal springboard economy rather than having to scale up problem. Don't fight to hold back from the US, but

emphasize the ability to attract and to build and that UK is a phenomenal place to come and build. On the government policy and maybe to segue on the China policy,

I think that the Made in China 2025, and to your point about the role in sort of, you mentioned construction and boosting productivity, I think setting that direction and saying these are the technologies that we want to be world-class in. China did that with Made in China 2025 10 years ago, and incredibly across sort of the board, they're right at the leaderboards. And I don't think it's because it was top-down government sort of

but rather setting that direction and mobilizing and organizing resources, the government guidance funds and others, but then peddling

paired with what Kai-Fu Lee calls the gladiatorial entrepreneurs, right? That you let entrepreneurs compete, you know, sort of violently at the sort of ground level and then you support as they grow and as they scale. And I think there's a lot to learn from the combination there. In terms of policies that work, also just to mention, I really like Startup Chile. And I think that in that springboard framing for the UK, the

Chilean government, since it launched Startup Chile in 2010, has sort of put Santiago on the map as a place that you would go and start up. And it's an easy, cheap program to run. And you bring world-class talent and you change and you boost your local ecosystem. So I like that a lot. And just to say the...

And then the last one that I'll close on, the examples of what the UK could do now to boost what's happening. The Israeli Yosemite Fund, which launches in 1993, before Yosemite, which means initiative in Hebrew, there was almost no venture capital in Israel. By the year 2000, Israel was second in absolute terms only to the US. It's recently, in some years, sort of lost that position to China. But

But the Yosemite Fund was an incredible policy mechanism because it created a cohort of venture capitalists who went and also raised money from international investors. They earned a phenomenal return. And then you created that market mechanism. And then there's that vibrancy. And one of my worries about the mansion house sort of approach here, which I think has changed tack,

But it was a sort of a force, more of the sort of the stick than the carrot. And I much prefer that sort of building the market so that you have VCs who are writing big checks, who are good also, and who are going to help the startups raise that next check or two, not just the passive investors and not helping to identify that great talent.

Fantastic. Can you share? I'm going to do a whistle-stop tour of all the questions. So, firstly, there's a slight subtext of capital supply debate going on here. Let me just put two facts on the table. If you look at one of the biggest long-term LPs in venture capital, both investing in the UK and US, Vencap, they invest at about 6% in terms of inflows into the UK. Their return is about 3%. So, they're over-invested in the UK on a kind of returns basis. And that

pattern is broadly if you look at longitudinal data. I know Stan and I have a slightly divergent view on the data and we should try and actually get together and fix this problem because I'd be genuinely interested in it. The second thing to say to Robin's point is that the British state has been the single largest limited partner in British venture capital for three decades. We are pumping a huge amount of public money consistently into venture capital.

I'm not objecting to us doing more of it. In fact, we are going to do more of it. But the central challenge for us is to ask how much and to what end and when do we draw the line and think that it has had the ecosystem impact it's intended to. And so those are very brief thoughts on capital supply. In terms of the questions asked, pre-product market fit, Stan is the operator, so I'll leave it to him. The one point I'll just emphasize is that if you're pre-product market fit, think of gaining product market fit, not just in the UK, but internationally.

prematurely internationally too because uh if you're spending all your time just getting product market fit in the uk and focusing on the uk as the sole market uh you might fall foul of the the thing i was describing earlier and by the way there is no requirement to be in the us to be able to sell in the us in terms of the entire headcount you can have an amazing product team and engineering team here and sell in the us uh day one and the example of israel is very compelling in that regard antitrust i think uh in britain we have a problem with uh

business dynamism. And my diagnosis of the dynamism problem is if you think about the life cycle of a firm from entry through to consolidation growth and then exit, I think most of our challenge is at the point of entry.

We don't have enough entry into technology markets in the first place. And then, you know, maybe there's a slight problem at some points in terms of organic growth in the market. I don't think we have a primary problem in terms of inorganic acquisition driven innovation impacts, the kind of kill zone type stuff that happened at some points in the United States. I just don't think that is applied anymore.

in any material sense in the UK. So I think antitrust is a really important tool in a bunch of areas and in a bunch of sectors. But when it comes to technological innovation through venture-driven startups in particular, I think our challenge is much more the point of entry rather than at the point of consolidation that we might want to antitrust on. And then on adoption, you know, I think this is the central policy question at the moment.

because I actually genuinely don't think that we have compelling enough evidence on what really moves the needle on this. So I'll give you four hypotheses or four bets that I would consider making in terms of what can move the needle on it. One, there's some sense that government procurement and being the kind of bearer of first risk in some technologies can be a helpful thing in driving larger adoption. Two, we know that to an extent management quality is a really important driver of technology adoption. And

And so there might be things that we can do to both improve manager quality and also scale when we see very good managers scale them.

And relatedly, three, we know that some firms, there's huge firm level variation in adoption. Some firms, which are scale ups, which grow really fast, which drive a bunch of our productivity uplifts, are particularly good at technology adoption. And so we might want to try and find ways of turbocharging them. You know, I worry that the government trying to turbocharge those firms might lead to their decline and take out the thing that drove them in the first place. But that's another variable that I'm very interested in. And then the final thing here is that

In many areas, the thing that really drives adoption is to have a fantastic supply side. People used to say, because there's this whole debate between being the frontier or driving the tail, long-tail adoption, people used to say that PageRank was like a kind of really...

nerdy PhD thesis that Larry and Sergey did. And obviously, Google is now the most widely used web search application. In many ways, being at the frontier and the supply side is the thing that ends up driving demand side adoption. Because if you build very good products that integrate with people's workflows, and you have an incentive to do that, then people adopt it. And so those are my kind of four ways of thinking about what can drive adoption.

Wonderful. Thank you. I think we don't have long. I'm going to ask Catherine if there's any quick questions online. Yeah, we have a question from David Wood, who is chair of London Futurists.

He says, should the UK be trying to create world-class tech companies by itself, or would a deep collaboration with the EU be more credible, for example, by providing greater scale? We'll take a question in the room. This gentleman's been waiting patiently. My name's David Lowry. I graduated in sociology in 1978 from the School of Economics.

I was surprised that in a discussion on the future of the world economy that not one of the panelists mentioned either the environment or mentioned climate change. And it seems to me that's fundamental to what's going to happen in the world economy, how climate change goes. I wonder if you could comment on the energy dimension of climate change and the power that the information centers are going to require. Yeah.

In particular, I've heard that in the United States, they want to use small modular reactors, which are experimental. Do you think the future on experimental power generators is a good idea when you need to have power generators that don't go wrong? It's a great question, and welcome back. I will just go with those two questions for now. So, Stan...

So the first question was about should we be doing it? The UK is not too small a country to build its own companies, although I would strongly welcome collaboration with other European countries and with the European Union. So you've been part of Horizon, you're having access to EIF funds through fund funds.

is all extremely welcome and, uh, and you're reducing the barriers for people to move here, you know, make it possible to live, you know, you know, join the Schengen, having a single currency, all this would be great, uh, which I would, uh, strongly, uh, like to see, but, I'm not sure we're going to see it very anytime soon. Um, but, um, but, yeah, but I, but I think we can build successful companies here, uh,

They are exactly as Kanishka says, they have to be global, essentially. So one of the companies I'm invested in right now is, Neil mentioned, Fractal. Fractal is probably Europe's only chance, really, of building a large chip company that can start to take bytes out of NVIDIA. That's global from day one, has to be global from day one.

And, you know, there is just enough talent, I think, to be able to put that together, you know, from folks in the UK, but also people that are kind of tracked from other parts of Europe, actually, to kind of build that thing. So, you know, so I think it is plausible and possible to do that from here. I think in terms of energy, I think it's absolutely nuts building these, like, huge data centers that are going to, you know, take enormous amounts of power. Yeah.

Yeah, I can see why that's the case. GPUs are totally the wrong architecture to kind of build these sort of systems. If you take an NVL72 rack from NVIDIA, it's got 72 H100 chips in it, and it consumes 130 kilowatts in a single rack.

you put multiple of those in a data center, you quickly get up to half a gigawatt or something of power consumption. It's insane as a policy. Therefore, approaches that look to decimate or more than decimate the kind of power consumption

of power consumption of these data centers is urgent, actually, and should be a sort of national priority to get invested in those, really, which is one of the reasons that we're doing Fractal, is the ability to do that. But I think there are approaches that can be done that will have a massive impact that don't need to sort of cost the earth in terms of power consumption. Wonderful. Thank you. Robin?

In terms of collaboration, I would agree with Stan's point, but I would then point to three quick examples of governmental collaboration. I mean, the UK has Department of Business and Trade, has Venture Capital Unit, we have the Global Entrepreneur Program. So we have mechanisms for taking venture capital out and also bringing entrepreneurs and venture capital in.

But programs that have been around for a long time that I think work really well in terms of that collaboration. So Taiwan and New Zealand have had a venture capital fund. It's a fund of fund that's operated for years and it invests across one another's startups.

The Bird Foundation, which is the bilateral investment in R&D between Israel and the U.S. from the 1970s, where effectively it built R&D centers in Israel. And then the idea was that companies would go and do sales and marketing in the U.S. And that reinforced the sort of relative strengths. And then third,

Korea's Ministry of SMEs and Startups, which I think is the first but maybe still the only in the world that's explicitly a ministry for SMEs and startups using the language of startups.

They run the K Startup Grand Challenge, which brings entrepreneurs from around the world to come and to start up in Seoul for a year and to work with Samsung, Hyundai, and LG. So I would think of those as collaborative models, but not because you can't build big enough in your own context, but because increasingly it's about having that born global sort of presence and connection from the beginning.

Yeah, I think Robin and Stan have laid out a good manifesto for European collaboration. So I'll leave that politically aside. The climate question is a really important one. And two overarching thoughts on this. One, to completely agree with what Stan said, which is that it is an urgent and really important task to focus on how we make sure that we are building efficient, energy efficient data centers and infrastructure.

That is a really exciting nascent area of technology and research, and we should try and do more of it. But the second thing is to keep in mind the magnitudes that we're talking about, which is not to underplay it, but to think about the role that data centers and AI more generally can play in the climate context. If you, I think, project out even the most optimistic demand scenario for AI and commercial cloud, you basically get from 2% to something like 6%.

6% or so of electricity demand coming from data centers. I'm not saying that's immaterial, but it is very materially dwarfed by both existing sectors like petrochemicals and prospective sectors like what we're hoping to do on electric vehicles. And so, yes, you can try and do a lot on optimizing data center efficiency, and I think it's a really important thing to try and do. And by the way, that's data centers altogether. If you just look at the AI part of that, it's even half of that.

That said, it's important to do. But the central critical question is if we're focused on climate technology, how we can bend the cost curve on renewables, how we can make energy more efficient by, in part, not exclusively, but in part, deploying AI in the service of driving that cost curve.

cost curve down is the central question. So I'm focused on the data center energy efficiency question. But to me, there's an even bigger price when you just look at the relative magnitudes of demands on electricity over the course of the decade and two ahead. Fantastic. I'm going to ask a final question, which I realized this is going to be politically difficult because I've warned two of the panel, but not the panel for whom it might be politically difficult. So say you become chancellor for the day and you want to further this agenda. What's the first thing you do?

Stan, what would you do? I would increase the total amount of money that we put into British Business Bank from roughly fund to fund $400 million a year to roughly $4 billion a year so that we can mint new capital in the UK at a rate of knots that gets us to the same position as the UK. It's a single lever we can pull that will make the biggest single difference. And I think everything will flow from that.

Right. Thank you. I don't want to repeat that. I mean, I think that boosting the money that's actually so I'll continue in that, I think, and maybe it's a bit critical of the other LSE, the London Stock Exchange. I think that AIM isn't up to snuff. And I think if we want to have the ability for, you know, exits, which historically aren't necessarily the big thing.

sort of exit venue anyway, right? Most exits are mergers and acquisitions or trade sales, and increasingly so with deep pockets. But if you need an IPO exit venue that really works in the UK, I would focus on boosting AIM's ability to serve that function so that you have a British version of a NASDAQ. And Japan and Korea did that with JASDAQ and COSDAC. So exact coffee just changed the beginning.

Wonderful. I'm so sorry for asking. No, well, Neil, I'm going to slightly reframe your question. My three asks of the Chancellor, who's doing an excellent job on day one, would be the following. And the central framing would be that we need to make sure that Britain is not just at the frontier of technology, but that drives the fairness and distribution from it as well. So three things I would do. One, reform the Entrepreneurs' Management Incentive Scheme. It's a tax incentive for employee stock option ownership so that way more of us

are incentivized to work in companies at the frontier of technology because we think that it's a good thing for our prosperity because we have a stake in those companies. Two, change non-compete laws in particular, which means that if you're at, say, Google DeepMind and you're leaving, it's quite hard to start a startup for at least six to nine months unless you've been very lucky in your employment contract. The non-compete clauses stop you from doing that. We should just reduce that very materially, if not drive it down to zero.

And three, we're already doing a huge amount in terms of pooling pension pots across the country. I would love for a lot of those pension pots, not least the Welsh pension, pooled pension pot, to focus increasingly on capital supply in those places and have the deepest ambition, not just for what London can do, but for what each part of the UK can do.

Wonderful. Thank you very much. Great suggestions and a great answer. I thought very well done. So let me just finish off with a few quick announcements, a few quick shameless plugs. So first of all, we have a new MSC and innovation policy at the LSE. If you liked tonight and you want more, like a year's more of it, please.

Applications are open. Second of all, if you liked tonight, but you thought we were too nice on technology, there is a more critical panel on technology tomorrow night in the LSE Festival, breaking the Jeff Bezos model of new technology. I'm sure it'll be brilliant as well. Third, and probably more relevant to everyone,

There's drinks at the back after today. Hopefully our panel will be able to join us for a bit. Speaking of which, thank you all very much for attending online, for attending in person, and particularly, thank you so much to our panel. That was fantastic. Thank you.

Thank you for listening. You can subscribe to the LSE Events podcast on your favourite podcast app and help other listeners discover us by leaving a review. Visit lse.ac.uk forward slash events to find out what's on next. We hope you join us at another LSE Events soon.