Finding Alpha in Emerging Markets: A Conversation with Baillie Gifford's Andrew Keiller

In this episode of Forthlane Features, Vanessa Hui, Senior Client Advisor at Forthlane Partners, sits down with Andrew Keiller, Partner at Baillie Gifford, for a thoughtful discussion on the investment landscape across emerging market equities.

Andrew breaks down what “emerging markets” really encompass, why recent returns have been concentrated in a handful of semiconductor giants, and whether that concentration is likely to broaden.

The conversation explores the critical role emerging markets play in the global AI supply chain, the outlook for semiconductor demand amid unprecedented data centre investment, and whether today's environment represents a lasting supercycle or another cyclical upswing. 

They also examine the geopolitical forces shaping investment opportunities, including energy security risks surrounding the Strait of Hormuz, the challenges and opportunities of investing in China today, and where Andrew believes investors are overlooking compelling long-term growth opportunities. 

WHAT TO LISTEN FOR

1:36 What are "emerging markets", and why is the label misleading?

4:34 Will returns broaden beyond TSMC, Samsung, and SK Hynix?

9:18 How emerging markets are powering the global AI buildout?

17:18 Are we in a semiconductor super cycle?

19:44 Why is China the overlooked coiled spring of emerging markets?

GUEST: ANDREW KEILLER, PARTNER & INVESTMENT SPECIALIST, BAILLIE GIFFORD

Website | LinkedIn 

CONNECT WITH VANESSA HUI, SENIOR CLIENT ADVISOR, FORTHLANE PARTNERS

Website | LinkedIn 

Transcript

Vanessa Hui (00:08):

Welcome to another episode of Forthlane Features, a podcast where we explore the people, the ideas, and the themes shaping global wealth and asset management. I'm Vanessa Hui, senior client advisor at Forthlane Partners and today I'm joined by Andrew Keiller, partner at Baillie Gifford. We're going to discuss investing in emerging market equities, an increasingly important segment of the global investment landscape. For more than a century, Baillie Gifford has been investing through wars, technological revolutions, financial crises, and dramatic shifts in global economic leadership. Baillie Gifford was founded in 1908 in Edinburgh and currently manages approximately 180 billion pounds and remains one of the world's oldest and most respected independent investment partnerships. Through the Emerging Markets Fund, Baillie Gifford is one of the specialist managers within Forthlane's global equity strategy where Andrew and his team focus on identifying compelling opportunities across emerging markets. Andrew, thank you so much for joining me this evening.

Andrew Keiller (01:20):

Thank you very much, Vanessa, for having me.

Vanessa Hui (01:22):

So before we dive into specific countries and companies, I want to start with a foundational question. When people use the term emerging markets, what exactly does it refer to and how is it defined by index companies and really how did the term come

Andrew Keiller (01:41):

To be? Yeah, it's a good starting point because the honest answer here is actually that emerging markets I would say is a slightly clumsy label now for a very diverse set of countries and companies. If you ask the index companies, they define it across several measures and some of those are actually practical market tests. So they're about market liquidity, accessibility for international investors, foreign ownership ability, foreign exchange, how settlements works and so on. It's certainly not simply a GDP label. And if you look at countries like South Korea and Taiwan, these are in emerging market indices, even though many people would not think of them as emerging in a normal economic sense and indeed in many ways they're more advanced than many Western economies. I think the phrase itself came into use in the early 1980s, partly because calling something a less developed country didn't sound very good, but I think the term is actually now struggling because what it does is it puts Taiwan's semiconductor ecosystem together with India's banking system together with Brazil's energy companies together with Saudi Arabia's capital markets, China's consumer internet platforms, and tries to sort of bucket those together.

(03:02):

That might be useful for an index, but it's not very useful for understanding the world. About 25 countries have classified as EM countries. Some of them got 10 times the GDP per capita than others. I'd say that the thing that frustrates me with it is that the word emerging, it sort of implies inferior and that's what I think is most wrong and I'm sure we'll get into this in this discussion, but that the emerging markets of old was definitely seen as risky, as volatile, as not dependable, as full of governance disasters. And that is not the EN that I see today, in fact, far from it.

Vanessa Hui (03:44):

We're definitely going to get into that and thank you for that really honest and helpful perspective. It's a really interesting backdrop and with that context in mind, let's turn to what's happening in emerging markets today, specifically performance. Emerging markets, as clumsy as a term that is, we're going to use that term in this discussion. So emerging markets have performed well recently, but it seems like much of that performance has really been concentrated in a handful of companies, particularly TSMC, Samsung Electronics and SK Hyneks. Do you expect performance to broaden out as investors take notice about performance, perhaps leading to structural increases in allocations to EM, or would you expect a handful of companies to continue driving returns?

Andrew Keiller (04:38):

Yeah, that's the question on everyone's minds at the moment. Has EM performance been too concentrated? There's no getting away from it. Yes, it has. The recent return story has been dominated by those companies that you just highlighted. They are all now trillion dollar plus market caps. The share price moves in the Korean companies in particular over the last year or two have been enormous. Samsung and Hyneks have moved by several multiples over a short period. But the interesting thing is they haven't become more expensive because their earnings have kept up with share prices and some and we keep seeing future upgrades. So this performance is not unwarranted. It's not just an expansion of multiples. We do own all three of these companies in our portfolios. We first did so in the 1990s, so I'd hope we have enough experience across the team to understand them well.

(05:32):

I'd say we've seen at least seven boom and bust cycles during our ownership. We've been through multiple technology changes and multiple leadership changes. At the moment, we're holding our nerve with them because we think the fundamental strength is there and these companies are solving for a huge global bottleneck, but complacency is a risk and we have been trimming them to fund ideas elsewhere because to your direct question, my instinct is that performance should broaden, but I don't think it will happen in a neat way. Markets very rarely will pass the bat on politely. We might see leadership move from memory and foundry into areas like power and equipment, copper, batteries, grid infrastructure, and much, much more. At the moment the beam is very bright but very narrow. It's shining on Taiwan and Korea. But to be honest, that presents an opportunity because it means things elsewhere are potentially being overlooked.

(06:27):

And for us, that's something we're continually focused on.

Vanessa Hui (06:30):

So on factor that doesn't seem to be overlooked is geopolitics. Everybody seems to be talking about the risk that geopolitics has on shaping the outlook for many emerging market economies. Specifically, if we look at the US Iran conflict, that has highlighted the vulnerability of many emerging economies to energy supply disruptions. How are you thinking about geopolitical risk today in your portfolio, particularly around energy security and the jeopardy of critical trade routes like Strait of Hermus?

Andrew Keiller (07:12):

Yeah. Look, geopolitical risk is not new to emerging markets. So I think experience is important here. It's hard to say that this particular one is only an emerging markets issue, of course. We've seen very clearly that when the strait comes under stress, the effect is not a local one. It moves through shipping costs, insurance, into oil prices, into inflation expectations, into central bank policy, into consumer confidence across many different countries. I think there is arguably a bigger investment point here about resilience and this goes for not just this geopolitical issue, but potentially other geopolitical shocks that we've seen and may see in the future is to say that governments and companies are no longer just optimizing purely for the cheapest possible supply chain. They're asking, can we get the power of the metals, the pipelines, the ports, the ships that we need when the world is stressed?

(08:10):

And we don't want to operate just in time. We want to operate just in case. And that's a marked shift I think we will see in the coming years and many of the suppliers for those problems that governments are trying to address are likely to be EM companies. So again, if you think about this with an optimistic lens, it's something that should present opportunities.

Vanessa Hui (08:32):

So another theme that I want to talk about, and I think it's hard not to talk about if we're speaking about emerging markets is artificial intelligence. This is one theme that really seems to sit at the intersection of geopolitics, industrial policy and economic development. And it's interesting because it seems like at least from my perspective in North America, investors often associate AI with companies like Nvidia, Microsoft, and OpenAI, yet many of the critical picks and shovels are the companies that you're focused on. So companies that are based in emerging markets. How do you think about the role emerging markets play in the global AI ecosystem?

Andrew Keiller (09:20):

Yeah. I think we've touched on the market concentration and concentration of performance and this is very clearly about AI development. I think that the easy mental picture of AI is somebody sitting typing into ChatGPT and helping them to solve problems. The investment reality underneath that is much more physical and that's where emerging markets plays the most important role AI needs chips, memory, packaging, power. It needs batteries. It needs raw materials like copper and steel, cooling systems, all sorts of energy. The software that we use may feel a bit weightless and words like cloud don't help that, but the infrastructure is anything but it's very easy for us to gravitate towards talking about TSMC, Hynex, and Samsung, but there are several other companies across the supply chain solving big problems. Take something like Montage that's a Chinese company and one of the very few that can deliver the tech needed that allows memory chips to connect to GPUs within data centers.

(10:30):

Take something like Acton technology. This is a Taiwanese company and one of the best providers of data center switches in the world. Take something like Chroma ATE, that's another Tairese company and it's one of the best testing providers in really intensive environments. So there are lots of companies beyond the ones that are the most well known. I think the other thing worth saying here is I do not think EM is just a duplicate of US AI exposure and that's a point that we're reading quite a lot. The US market is more often the model layer, the software layer and the hyperscaler demand layer. So if you own Nvidia or Anthropic or Alphabet, they have exposure to AI demand, of course, but they might still be underexposed if you're a US investor to the companies that are enabling that demand to be built in the real world and EM is much more about the physical supply layer, which extends beyond the companies that I've mentioned into things like energy and materials and much more that I'm sure we'll come onto.

Vanessa Hui (11:34):

Right. And you mentioned that everybody does talk about TSMC. So I am going to ask you about SMC. I mean, it is one of Baillie Gifford's largest holdings in the emerging markets funds. So what would you say that, from your perspective, what is it that makes TSMC such a unique business and what do you think people are still underestimating about its long-term opportunity?

Andrew Keiller (12:02):

Yeah. There's a common accusation in emerging markets that everybody owns TSMC. So do you really have a differentiated view? And ultimately having a differentiated view is how you make money for your clients. I would say first of all, we have owned it for much longer than most and own it in bigger size than most. Actually, the average manager is underweight TSMC today and we're not. The simple answer on why this is such a great company is that it does something that no others in its industry can. And I'm talking about extremely high-end semiconductor manufacturing at scale.

Vanessa Hui (12:38):

It's

Andrew Keiller (12:39):

Also unique from a financial perspective. There are very few companies in the world, regardless of industry that can invest over $50 billion a year and maintain the rates of return on that spend that it does. I'd say in a nutshell, TSMC is about technology, it's about customer trust and it's about advanced manufacturing. Now, most companies are viewed at one of those things. Very few companies are good at all three. TSMC is not just making chips, it's manufacturing at a level of precision where even a tiny defect can totally destroy the economics of their product. And so its own customers just cannot do this themselves. Its own customers are some of the most sophisticated technology companies in the world themselves and they are designing products around TSMC's capabilities years in advance. So we're talking about barriers to entry that very few others can boast

(13:35):

And I'd say that creates a very unique competitive advantage because it's not just scale, it's accumulated process knowledge and complete dependence by your customer. What do investors underestimate? Possibly the duration of that advantage. And people often think about semiconductors as a short product cycles. That might be true at the chip level, but at the level of the manufacturing ecosystem, there is a very long memory. Once you've got the design tools, the engineers, the suppliers, the customers orbiting around a single company becomes extremely hard to unpack that or to recreate it elsewhere. So despite the large size of this company, we still see a path to it doubling or more from here over the next five years or so, which is what our investment hurdle is. So we remain happy holders at the moment.

Vanessa Hui (14:29):

Great. And you mentioned Montage and Hynek's. So for listeners who don't follow semiconductors closely, why has memory become such a critical bottleneck in AI?

Andrew Keiller (14:43):

It boils down to the fact that memory demand is increasing very rapidly and there are very few companies in the world that can do it and those that can have not increased their capacity for doing so as quickly as demand has risen and that has therefore created a bottleneck. In our recent work, we've looked at the superchip that NVIDIA released back in 2024. It's called the GB 2000 and it yields an aggregate capacity of around 13 terabytes of high bandwidth memory. I wouldn't worry about the technicals there, but the point I want to make is that by 2027, the new version of that, which is called the Vera Rubin Ultra, it's expected to carry roughly 150 terabytes. So that's an 11 fold increase in the capacity of this in just a few years.

(15:38):

We're thinking very carefully about whether memory has moved from being what people used to think of as a commodity product to something much more strategic and that type of evidence from NVIDIA's demands would suggest so. Historically, memory was treated as something that was interchangeable between systems. In AI, what we're now seeing is the best memory is increasingly customized. It's qualified with specific customers and it's then physically integrated with them and that changes the economics of it. It becomes harder to deliver the fewer companies that have the capability, of course, the narrower that bottleneck becomes and that provides investment opportunities for those that are trying to help the world solve that.

Vanessa Hui (16:24):

That's very helpful. Thank you. You mentioned something about the semiconductor industry typically being seen as cyclical, characterized by periods of overinvestment followed by sharp downturns. It feels like we're clearly in the midst of another investment boom. How are you thinking about this cycle? Are we building data centers at this rate forever or are we in a cycle? Help us understand that.

Andrew Keiller (16:54):

Yeah, it's a good question. We are in a super cycle at the moment. I do not think we are building data centers at this rate forever. Let's be very honest, very few investment booms will proceed in a straight line. This applies to semiconductors, but it's applied to railways in the past. It's applied to telecoms. It's applied to the internet. It's applied to shale energy. These have all had periods of overbuild. Some capital was wasted, some companies disappointed investors, but the infrastructure itself ultimately changed the economy. And I think we're in a big supercycle for semiconductors in particular for memory and foundry companies at the moment. So it's our job to think about that cycle and think about where we are within it. The current cycle has two features that I think are worth calling out. The first is that the demand shock is enormous. Though the hyperscalers are obviously spending at a scale that would've sounded implausible even just a few years ago.

(17:54):

And the second, as we touched on before, is that the bottlenecks are very much physical. So token usage within large language models can grow exponentially, but foundry capacity, memory capacity, packaging, power, that can't and they scale more slowly. So not every data center will earn a great return. I don't think that's controversial to say. The question is whether the world still needs more compute, more power, more memory than it has today. And on that, the direction still looks clear to us and so there are still very strong investment opportunities.

Vanessa Hui (18:30):

Great. So just pivoting from specific companies and the industry, I just want to pivot to one of the countries within emerging markets and that's China. Again, I think China illustrates how clumsy the term emerging markets is, but how are you thinking about China today and where would you see the greatest opportunities and risks?

Andrew Keiller (18:55):

Yeah, I think it's pretty strange for a country of the size of China that it kind of feels like it's out of the spotlight in emerging markets at the moment where everything is about Korea and Taiwan and that strikes us as odd, not least because China itself is developing a globally competitive AI stack of its own, albeit with a very different strategy given its ability to access advanced chips, given the energy that it has access to. But more broadly, China also boasts some of the same ingredients that investors absolutely crave elsewhere. So it's got engineering talent, it's got world-leading manufacturing. It's got leadership in globally relevant industries. Taiwan and Korea are very concentrated technology markets. China is not. So yes, it has internet and hardware companies, but it also has automation, batteries, healthcare, big domestic consumer companies, electric vehicle companies, financials, industrials. So it's not really one trade.

(19:56):

So I hesitate to give China a very broad view trying to cover it all together.

(20:01):

But the other thing I think we should acknowledge is the risks are very real. So consumer confidence has been weak. Property has damaged household wealth in recent years. Regulation can obviously be very abrupt. We talked about geopolitics before. That's certainly not a background issue for China. And for some investors, that's enough to stay away. One thing that I think is notable just now is just that there is a huge pool of latent domestic money. And what I mean by that is that Chinese households have got very large bank deposits. We've got about $23 trillion sitting in the Chinese banking system just now that is earning little to no return. If some of that moved towards consumption, if some of that moves towards the equity market, then the consequences could be huge. So in some ways it feels a bit like a coiled spring at the moment and in an environment where the spotlight is shining elsewhere, it's providing good valuation opportunities as well.

Vanessa Hui (21:05):

Can you speak to why there is so much cash sitting in bank accounts? What does it feel like people are waiting for?

Andrew Keiller (21:14):

Yeah, I think, and this is linked to a crisis of confidence in the consumer. So we can see on many metrics that consumer confidence is low and that's been the case since COVID actually. So I has never really recovered from COVID in terms of consumer spending getting back to where it was. And so people instead are choosing to save and the vehicles that they have to save are limited in terms of where they can actually generate a return. So you've got the auction of bank deposits or the bond market, you're not getting very much return in either of those. You can invest in property and that's not gone very well. So our contention is that the equity market for domestic investors looks increasingly attractive as a savings vehicle. And because of the magnitude of money we're talking about here, if some of that money moved out of the banking system into the equity market and the government were to add policy incentives to make that more likely, then I think that could be very meaningful for returns and that's something you want to be involved in before it happens rather than after.

Vanessa Hui (22:25):

So we've covered a lot of ground today and I would just want to zoom out. So Andrew, when you look out over the next decade, I know a decade in these days feels like an incredibly long amount of time given just how much volatility there is in the world. What are the most exciting opportunities in EM that investors are not paying enough attention to? You've touched on a little bit with TSMC, but just zooming out more from a broader thematic perspective.

Andrew Keiller (22:59):

Yeah. I think I could be here all day just asking some of these, but I guess to give that some structure, I might group it into three buckets. The first week we've talked a bit about, but it would be the physical side of the internet. So

Vanessa Hui (23:18):

We've

Andrew Keiller (23:19):

Talked about AI, but the cloud automation, these all sound digital, but they need physical supply chains underneath them. So it's semiconductors and memory, but it's also power equipment. It's raw materials like copper and steel. It's technology like cooling equipment. It's increasingly effective grids and many of the critical companies solving for these issues are in emerging markets. So it's the physical side of the internet, that would be the first bucket if you like. The second would be domestic scale and I'm talking here about some of the largest countries in the world are classified as emerging markets. The likes of India, Indonesia, Brazil, Mexico, China, these are not just export stories. There are some great export stories coming from these markets, but they have domestic champions that can benefit from income catch up that you just don't get elsewhere in the world. And there are many companies that are trying to play into that.

(24:25):

It might be the digital first banks, the innovative insurance companies, the logistics platforms, the consumer brands. So the opportunity here is about income growth and also about formalization of retail and of saving.

(24:41):

And then just in terms of what's the exciting opportunity for EM over the next decade, I guess the third bit's not really a bucket, but it might be a cheat answer to say that it's the opportunity for EM as a whole that comes from investors catching up to what it's really about. For all the excitement and vastly improved returns in the last couple of years, EM allocations remain low. EM flows for active funds remain negative. Global investors are underweight. So we have a job in trying to convince our clients and our prospective clients that we're only at the start of that promise period. I would frame that as to say that the EN of the past is extremely different to the EM of the future. It's no longer a homogenous asset class full of governance disasters and badly run banks and pedestrian consumer staples, highly cyclical commodity companies that people have associated with.

(25:39):

That framing 10 years ago might have been fair, but today EM is more important than developed markets in many ways in getting us towards solving two of the world's biggest shortages in resources and in hardware. Investing in EM should allow investors to capture the returns that will be generated along the way. And at the moment, it's nowhere near priced for that promise with discount to develop markets being around 40% or so. So maybe that's not a third bucket, but I think there's an overall under appreciation of the asset class that was worth raising.

Vanessa Hui (26:18):

I couldn't agree more. And what a great way to end. I have one more question for you and it's just one that I ask every guest on Fourthlean Features. And that question is, what is one lesson you've learned in your career that still continues to guide you today?

Andrew Keiller (26:35):

Yeah, good question. My instinctive reaction there is to say keep learning, but I think the answer is probably it's about how to hold your views. We're trying to invest in companies for the long term. By that we mean periods of five years or more. We're investing in companies where we expect earnings growth to increase over that time period and in long-term investing, you need conviction because if you change your mind every time a share price moves, you'll never own anything important for long enough. But the danger is that you fall in love with something, you start asking, "How do I defend what I already believe?" So I think the lesson is that when the facts change, you need to change. That might sound obvious, but it's very hard in practice. You need to be structured such that you can actually do that. You need to incentivize people so that that's okay.

(27:31):

In emerging markets, the old facts were that the asset class was more volatile, more fragile, often lower quality. That might gain true a point in history and we shouldn't just airbrush that out, but those facts have changed in several important ways because I would say the corporate qualities improve massively. I would say that central banks are much more credible than they were before. I would say that some of the world's most strategically important companies are now in emerging markets. Yes, you need to have strong views, but you should make those conditional because markets change a lot and the market's not going to reward people for being consistent if they are consistently wrong. It rewards people who can keep adapting. And I guess to the way I started this answer is keep learning.

Vanessa Hui (28:24):

Andre, thank you again for joining us. I know it's late into the evening in Edinburgh, so I could not be more grateful for your time and your insights and all the perspectives you've brought on behalf of your team at Baillie Gifford. And to our listeners, thank you for tuning in to another episode of Forthlane Features. If you've enjoyed today's conversation, please subscribe and share the podcast with someone who may find it valuable. Thank you, Andrew.

Andrew Keiller (28:51):

Thank you.