During Q2 2026 the strategy outperformed the MSCI Emerging Markets index by 1.8%. The quarter was characterized by a strong equity recovery after a March correction in response to the war in Iran.
Key drivers of outperformance were technology stocks in Korea and Taiwan although this was entirely driven by stock selection rather than overall exposure to these volatile markets. Our underweight positions in India and Brazil also helped as these markets lagged the broader index. China proved a drag during the quarter as a number of industrial and consumer related stocks that we own faced profit taking.
Market volatility resulted in another fairly active quarter in terms of repositioning the portfolio. We continued to take profit in AI driven markets such as Korea and Taiwan, leaving us somewhat underweight the broader AI hardware space. We also used a recovery to further reduce exposure to the UAE as we wait for the economic impact from the Iran war to start feeding into the domestic economy. On the flipside we have been adding to Indian financials and to Chinese consumer and industrial stocks which have undergone a derating due to lack of AI interest.
Every conversation we have with investors starts with the AI space. While AI related hardware stocks account for around 40% of the MSCI Emerging Market index weight, they account for 90% of its returns due to extreme volatility. This means that directional exposure to the sector risks completely swamping any other active decisions on a portfolio. While we do not shy away from having a view on the future of AI (and most specifically for EM, of AI capital expenditures), we never want to have a single concept or theme dominate our portfolio and relative returns. This has led us to be quite careful in the degree to which we have been underweighting some of these “hot” markets as they have rallied.
To quote Scott Fitzgerald, “the test of a first-rate intelligence is the ability to hold two opposed ideas in the mind at the same time, and still retain the ability to function.” We are less concerned about rating our intelligence, but we always accept the possibility of outcomes diametrically opposed to our beliefs. With many of our team members having first-hand experience of the enormous misallocation of capital during the dot-com boom, we are naturally sceptical as to the sustainability of the current splurge. The fact that SpaceX is expected to require US$90bn annually of external financing over the coming years is, in itself enough to give us pause. Losses for SpaceX IPO investors could have a chilling effect on the ability of other AI firms to continue funding their gargantuan investments. Having said that, there is a stark difference to equity valuations in our markets compared to the turn of the century Nasdaq. Valuations have remained remarkably cautious with P/E multiples on EM memory stocks below 5x expected earnings. Such circumspection offers a degree of support if AI capex plateaus rather than collapsing or indeed, continues apace as sell-side analysts universally expect.
We can express a cautious view on AI by carefully underweighting the segment and its most optimistically priced corners. This still allows for stock picking in that and other areas of the portfolio to generate alpha. There is also the tantalizing possibility that a leverage-fuelled AI market washout could lead some high-quality hardware businesses to become deeply undervalued and set up buying opportunities. In any case we believe that active managers with a disciplined but common-sense approach should see the current environment as an opportunity to demonstrate their value add. Investors with passive exposures to the asset class are likely to suffer from concentration risk and changes in momentum.
