Transcript
Speaker: Yeah.
Speaker: Hello, and welcome to Uncovering Trusts, a podcast by Edison Group. I'm your host, Liam O'Byrne, and today I'm joined by Milos Paps, Director of Investment Company Content at Edison, to talk about Invesco Asia Dragon Trusts, ticker IAD.
Speaker: Milos, as always, great to have you with us. Milos Paps, Thanks, Liam. It's good to be here, as always. So you published an update note on IAD a few weeks ago. What makes now a particularly interesting time to revisit the trust?
Speaker: Yes, I would highlight two things. ah First, Asian equities have been very strong, but market leadership has become unusually concentrated in a handful of AI-related technology stocks, particularly in Taiwan and South Korea.
Speaker: The trust has delivered very strong absolute returns, but its valuation discipline means it hasn't fully participated in that increasingly narrow rally. Second, the trust itself has changed significantly.
Speaker: It has now completed its first full financial year since the February 2025 merger with Asia Dragon Trust, which transformed its scale and corporate proposition.
Speaker: And there is a managed transition within the investment team, with Mark Yeh joining as co-manager and Ian Hargreaves due to retire in March 2027. So I think the interesting question today is how a valuation of an active manager navigates and Asian markets increasingly dominated by a small number of AI winners.
Speaker: Now, we discussed IAD on the podcast back in October, but before we dive into the you know the portfolio, the merger and AI in more detail, could you perhaps remind listeners of what IAD does and what differentiates its approach?
Speaker: Yes, with pleasure. um The trust invests in across Asian and Australasian equities with the objective of delivering long-term capital growth and income and outperforming the MSCI AC Asia ex Japan index.
Speaker: The board describes the investment approach as valuation, not value, which I think captures well. ah Fiona Young and the team are not simply looking for you know statistically cheap businesses. They want high quality, well-managed companies where the share price is meaningfully below their assessment of intrinsic value. That can include growth companies, technology leaders, or businesses going through a turnaround.
Speaker: They combine broader views on countries and industries with, to a greater extent, bottom-up fundamental research, and company meetings are an important part of the process. It is also quite concentrated, as there were 59 holdings at the end of August, and the top 10 accounted for just over half of the portfolio.
Speaker: Historically, that approach has had some all-weather characteristics, I would say. um Our analysis to N2025 showed a 10-year upside capture ratio of around 104% and downside capture of Put simply, over that period, the trust captured slightly more than 100% of the benchmark's upside, but only around 94% of the downside.
Speaker: And from your recent note, it's clear that approach has been somewhat tested as the benchmark has become very concentrated itself. Just how concentrated is it? Very. At the end of August, um Taiwan Semiconductor Manufacturing Company, so TSMC, Samsung Electronics and SK Hynix together represented about 31% the um MSCI AC Asia X Japan Index.
Speaker: For comparison, the three largest companies in the um MSCI USA Index represented just over 20%. um Information technology alone was about 46% of the Asian benchmark.
Speaker: um There is obviously a fundamental explanation as several years of underinvestment in memory capacity coincided with an nine unexpectedly rapid increase in the amount from AI.
Speaker: Samsung and SK Hynix have seen very strong performance. pricing and earnings, while TSMC has benefited from the wider AI infrastructure build-out. But Fiona's concern is increasingly about expectations rather than the immediate fundamentals. um A company can still produce exceptional earnings and disappoint investors if the market was already expecting even more. So is IAD essentially taking a negative view on AI?
Speaker: but No, and that distinction is important. um Technology was still 40.5% of the portfolio at the end of August. TSMC was the largest position at 15.1%, and Samsung Electronics was second at 10.5%. So the trust has substantial AI exposure. um um It is underweight technology only because technology has become such an enormous part of the index.
Speaker: ah What the managers have been doing is gradually taking profits in some of the initial AI winners as valuations have increased and looking further down the beneficiary chain.
Speaker: One example is companies using AI rather than selling the infrastructure. um For instance, Tencent and NetEase can potentially use generative AI to lower game development costs, reduce content more efficiently, and personalize the gaming experience.
Speaker: There are also second-order beneficiaries. um The enormous profits being generated by Korean semiconductor companies can feed through into employee buts bos bonuses bonuses, consumption, housing, and wealth management. That creates opportunities in businesses such as Samsung Fire & Marine, KB Financial, and even companies involved in the construction of semiconductor infrastructure such as Samsung E&A.
Speaker: So this is not an anti-AI position. It is a view that the equity opportunities created by AI may be broader than the three or four companies which have attracted most of the market's attention. And South Korea has been right right at the center of that rally. But I understand that Fiona is becoming more cautious there.
Speaker: The strength of the cycle itself is not really in dispute. um Fiona describes the current environment as one of the strongest memory cycles we've seen. Supply is tight because there were several years of underinvestment, and ah as I've just mentioned, while AI demand increased much faster than expected.
Speaker: And semiconductor capacity can't be added overnight. You need clean room construction and new equipment, so those tight conditions could potentially persist for another couple of years.
Speaker: Her concern is less about the immediate fundamentals, more about how much good news is already reflected in expectations and share prices. The MSCA Korea index roughly doubled in US dollar terms in 2025 and rose by more than another 90% in the first eight months of 2026.
Speaker: um History also tells us that semiconductor cycles tend to turn earlier than investors expect. So it is perfectly possible for memory pricing and earnings to remain very strong while semiconductor shares start performing less well.
Speaker: Fiona also questions some of the assumptions behind demand as hyperscalers are spending extraordinary amounts on data centers. So eventually the returns on that capital need to justify the investment.
Speaker: and much cheaper Chinese large language models introduce another variable. um One interesting example she gave was an Asian software company that uses cheaper Chinese models for routine tasks and Western frontier models only where it really needs the additional capability, a practice known as model routing.
Speaker: That suggests AI adoption can continue strongly without every application requiring the most expensive computing infrastructure. And how has that positioning affected performance? Well, absolute performance has remained very strong. In the year to April 2026, the Trust delivered and an EV total return of 39.6% and a share price total return of 44.7%.
Speaker: ah But the benchmark returned 45.7% with a shortfall mainly reflecting the technology underweight, I've discussed, and stock selection in China. um That became even more visible in the June quarter. um The Trust's NAV rose almost 20%, which ah in most circumstances would be an excellent result, but the benchmark rose nearly 27%. Not owning SK Hynix alone detracted around 6 percentage points from relative performance.
Speaker: On the other side, stock selection in Taiwan was very strong. Yajio was a particularly large contributor alongside MediaTek and Lagan Precision. And it is worth keeping the recent relative weakness weakness in context. um At the end of August, the Trust and AV had returned about 66% over five years, compared with roughly 53% for the benchmark, and about 190% over 10 years versus around 150% for the benchmark. Now, if investors are crowding into Korea and Taiwan, does that you know create opportunities elsewhere? That is increasingly how the managers see it. um China is probably the clearest example.
Speaker: ah Fiona argues that China has effectively become a source of funding for the AI trade as investors sell existing Chinese positions and move the money to Korean and Taiwanese semiconductor stocks.
Speaker: That is quite attractive for a contrarian manager because valuations in parts of China have become increasingly appealing as investors have shifted their attention as well as elsewhere. um The macro backdrop is clearly difficult. The property sector remains weak and consumer confidence is subdued.
Speaker: But um the managers think investors can become too focused on those aggregate indicators and overlook individual businesses capable of prospering despite them. Tencent and NetEase are good examples.
Speaker: At the same time, the trust has been quite disciplined about selling when the underlying thesis deteriorates. um it For instance, it exited TNG as competition increased while Wulian Yeh was sold after a restatement of its 2025 financial statements, weakened confidence in its reporting and governments. And I guess ah India has almost the opposite problem, you know, an attractive structural growth story, but expensive equities. Has the opportunity improved there?
Speaker: Gradually, although the trust remains under with India, which represented 7.3% of the portfolio at the end of August, compared with roughly 12.6% of the benchmark. um The market had become extremely popular, with investors often willing to pay very high valuations for structural growth. um More recent relative weakness has started create individual opportunities, although Fiona still considers the overall market expensive.
Speaker: Shreeram Finance is one example. um The company provides finance to truck owners and smaller businesses that are underserved by traditional banks. So there is scope to benefit from the continued formalization of credit.
Speaker: MEFG's 20% investment may also help improve its funding profile. Another example Fiona highlighted is Delivery, the logistics operator. It offers exposure to the growth in Indian e-commerce, but without paying the valuations attached to some of the consumer-facing platforms.
Speaker: but And I think that that is a recurring theme the portfolio. So if the obvious way of accessing a structured growth story becomes expensive, look for a less obvious beneficiary. Beyond AI and Asian consumption, are there any other themes the managers are looking at? For example, energy and infrastructure?
Speaker: Yes, so there's a broader structure theme around energy security, resource investment, and supply chain resilience. ah COVID, the Russian-Ukraine war and subsequent geopolitical disruption have encouraged governments and companies to focus less exclusively on the low lowest cost supply chain and more on security of supply.
Speaker: um ai reinforces that because data centers require enormous quantities of electricity, copper and other physical infrastructure. There is one reason the managers recently initiated a position in Worley, the Australian engineering and project services company.
Speaker: It is exposed to energy, chemicals and resources. And the managers like the combination of recovering industrial industry investment, a healthy project pipeline and relatively limited fixed price contracting risk. So again, this is another example of looking one or two steps removed from the obvious PI trade. And we should also talk about the management changes. What exactly is happening there? Fiona Young remains the lead manager. Mark Yeh became um co-manager from the 1st of August and Ian Hargreaves will remain involved with the trust until he retires from Invesco on the 31st of March, 2027.
Speaker: ah The key point is that this is a gradual transition. Fiona became co-manager in back in 2022 and moved into the lead role in May 2024. Mark has been part of Invesco's Asian and emerging markets equities team since 2021 and became a deputy fund manager in 2025.
Speaker: um The board in Invesco have been explicit that the investment philosophy process and portfolio construction remain the construction framework are unchanged. Moving on to the merger then, which has transformed the corporate side of IAT, what would you say are the main benefits for shareholders? I would say scale is probably the starting point. The trust has had had total assets of around £1.1 billion at the end of August, compared with roughly £250 million pounds for the old Invesco Asia Trust before the combination was announced.
Speaker: That greater size improves liquidity, raises the market profile and allows fixed costs to be spread over a much larger asset base. The indicated and ongoing charge is around 0.7%, which is the lowest among the comparable Asian equity investment trusts.
Speaker: There is also the enhanced dividend policy. um The policy is to distribute roughly 4% of the end April NAV over the following year, paid in four quarterly installments. Importantly, the portfolios not manage to generate that level of natural income. The distribution can be funded from revenue and distributable reserves, so Pion and the team do not have to distort the investment strategy by chasing high-yielding stocks.
Speaker: For FY27, he announced annual distribution is 19.2 pence per share, which is 21.5% higher year-on-year because the April 2026 NAV was substantially higher. And what about the discount?
Speaker: This has also improved maturely since the merger was announced. The current discount is around 7% compared with an average of 7.6% over the previous year and around over years.
Speaker: The board has said it believes the shares should normally trade at a discount of less than 10%, and it has continued to use share buybacks when the discount is unusually wide. More importantly, shareholders now have this three-yearly unconditional tender offer.
Speaker: The first is due in 2028 and will allow shareholders to tender up to 100% of their holding at a 4% discount to NAV. That does not guarantee where the shares trade between the tender dates, but it is a meaningful structural mechanism for addressing a persistent discount. And does gearing play much of a role?
Speaker: Not currently. The trust can use gearing, but it was only about 3.3% in the end of August, so the investment case is overwhelmingly driven by stock selection and portfolio positioning rather than leverage. Understood.
Speaker: Looking ahead then, what would you see as the main risks from here? The most immediate one is that the environment which has channeled the strategy simply persists. If AI hardware earnings remain exceptionally strong and investors continue concentrating capital in a handful of semiconductor companies, the trust technology and the weight could continue to weigh on relative returns, even if the portfolio produces good absolute returns.
Speaker: um There are also broader regional risks, continued weakness in Chinese property and consumption, geopolipolitic geopolitics, trade tensions, currencies and capital flows. And the nature of a contraint strategy means the managers are often early, they may identify a valuation opportunity before the wider market does, and sometimes the investment thesis will simply prove wrong.
Speaker: Against that, the trust is not avoiding the region's major growth themes. It still has very significant technology exposure, but combines that with positions in the Chinese internet, Korean financials, Indian finance and logistics, and Australian resources and infrastructure. Now, before we wrap up for today, Miloš, how would you summarize IED's investment case?
Speaker: I would highlight four points. First, the trust offers a well-tested valuation-aware approach to Asian equities that has outperformed this benchmark over five and 10 years, even though the recent highly concentrated market has been more difficult.
Speaker: Second, the AI debate illustrates what differentiates the strategy. The managers still own major semiconductor winners, but they are willing to take profits as valuations rise and look for second-order beneficiaries rather than simply following benchmark benchmark momentum.
Speaker: Third, there is a broad opportunity set beyond AI hardware, um like the ones I mentioned, so Chinese internet companies, Korean financials industrials, Indian finance and logistics, and Australian energy infrastructure. And fourth, the trust itself is a much stronger proposition after the merger. It now has greater scale, a low ongoing charge, an annual distribution of roughly 4% of NAV paid quarterly, active buybacks, and the 2028 tender offer.
Speaker: The trade-off is that being valuation conscious and prepared to move early can be uncomfortable and a small number of stocks dominate the market. But for investors who want genuinely active Asian exposure rather than something that increasingly resembles a concentrated semiconductor index, that differentiation that differentiation is really the central point of the trust investment case.
Speaker: Brilliant. Milos, thank you very much. Thank you. You've been listening to Uncovering Trusts, a podcast by Edison Group. To find out more about Invesco Asia Dragon Trust and read our latest research, please visit www.edisongroup.com.
