The Fund returned -2.5% over the quarter. It was a challenging quarter for the strategy with markets focused on artificial intelligence – both the hyperscaler platforms and increasingly the wider providers of infrastructure and hardware required to support them. The gravitational pull of this theme sucked liquidity from other parts of the market. At a veneer level markets appeared to broaden, with Emerging Markets and Smaller Companies indices performing strongly, for example. But in reality these were derivatives of the AI theme, with markets like South Korea and Taiwan dominated by companies that produce the semiconductors required to power datacentres.

Reasons to be cautious

We continue to believe a defensively positioned portfolio is warranted:

• The earnings and cashflow generation of the hyperscalers are increasingly out of alignment. Accounting conventions have boosted earnings shorter term, but the real story is evident in the declining free cashflows. We are approaching the point where cash generated will no longer cover expenditure in some instances. Given investors historically often cited the capital-light nature of the hyperscalers and strong cash generation as key attractions, this raises serious questions over the market’s willingness to fund future capital requirements at current equity and debt prices.

  
  

• Without a significant positive inflection in economic growth the trajectory of US Government debt is unsustainable. Pressure is mounting in various other countries around the world as well.

  
  

• Geopolitical risks remain high. Rising defense budgets must be paid for either by increasing Government debt, increasing taxes or cutting expenditure in other areas such as healthcare or education.

  
  

• Private market valuations are just not credible in our opinion. Once the intended banner initial public offerings of SpaceX, Anthropic and OpenAI are in the rear-view mirror it is hard to see how public markets can digest the follow-on wave of exits that private markets need.

  
  

• Overall, public equity market valuations are expensive by historic standards even though corporate profit margins are healthy and earnings growth has been unusually strong.

All of these points are well understood, which goes some way to explaining the extreme volatility we are seeing when stocks change guidance or miss numbers. Some commentators have been vocal about the apparent death of fundamentals and valuation as a driver of stock market returns as a result. But momentum and sentiment have always played a part in the stock market. Given the magnitude of the potential impact that AI is going to have it is entirely understandable that markets are focused on it. Predicting the impact that AI is going to have on company fundamentals is difficult, and heightened sensitivity to that is justifiable. But fundamentals and price will drive share prices over the long term, as they always do.

Fundamentals do matter

The table below shows some simple output from what we call “perfect foresight” analysis. We use this to help us understand which factors matter for predicting absolute share price performance in the future, and for comparing the expected return across companies that have differing business models and accounting profiles. It can help us compare “growth” companies with “value”, or seek to answer questions like: how much more we should be willing to pay for “quality”.

Perfect Foresight Analysis - 5 years to 30 June 2026

The analysis looks at 12,770 companies listed in developed markets and with a free-float adjusted market capitalization of over $1bn as at 30 June 2021. It shows the median returns (in local currency) for those stocks over the following 5 years to 30 June 2026 across two vectors – sales growth over the 5-year period and the forward 5 year PE i.e. the price as at 30 June 2021 divided by earnings 5 years later. The analysis essentially shows what returns could have been generated with “perfect foresight” on company earnings and sales over a 5-year period in which many observers seem to be saying fundamentals have not driven share prices….which is not correct.

The highest returns over the 5 years to 30 June 2026 were for companies that were very cheap on a forward 5-year PE basis because they grew sales strongly (top row, second column from right). Cheap companies on a forward PE basis that don’t grow fast (top left) outperform the median (bottom right) but returns for companies that are a little more expensive on a forward basis but grow faster are better. Returns for companies that grew sales fast but not enough to make them cheap on a forward PE basis are poor (second from bottom, second from right), with returns even worse if the forward PE is high but growth is low (second from bottom, far left). It is quite clear from this analysis that fundamentals strongly impacted share prices over the last 5 years.

The next table aggregates the same analysis for every quarterly rolling 5-year period from the start of 1990 to the end of June 2026, which is 40,882 unique securities with a free-float adjusted market capitalization of over $1bn at the start of each respective 5 year period. The median return overall (bottom right) is a little higher over the longer time period, which is a reflection of how much market cap weighted indices in recent years have been driven by a small number of larger companies. But the relationship between forward value (PE) and sales growth in segmenting stock returns remains very effective over this much longer time period.

Perfect Foresight Analysis - 31 December 1990 to 30 June 2026

So why have we shown this analysis? Because we think it is very pertinent to the market’s current focus on AI. Whilst accounting conventions (writeups on non-realised gains on private investments, for example) can obscure “real” earnings over shorter time periods, and sentiment can drive stocks shorter term, it is comforting to note how important fundamentals are over the longer term. That said there is no single magic formula for companies. The mix of growth, value and quality characteristics that unlocks attractive returns is far more complex than the simple 2 factor analysis provided above. But it can be analysed. The tricky part isn’t so much proving that fundamentals and price drive share prices longer term as it is reconciling this with how difficult it is to actually predict those fundamentals on a forward basis. It is then far from trivial to build a portfolio that can cope with shorter term swings in performance of stocks and groups of stocks (themes) that are often not fundamentally driven.

Over the last 5 years the market has strongly favoured hyperscalers that have grown sales and profits quickly. They will have deserved their strong performance if sales and profits growth are sustained and it would be wrong to suggest otherwise. But if sales falter or strong sales do not translate into profits over the next 5 years because capital expenditure is too high, then stock returns are likely to be weak over the coming years. Our analysis of the hyperscalers suggests it is almost impossible for them to generate the levels of profit growth that would historically have been required to generate attractive share price returns over the coming 5 years. The hyperscalers have no choice but to undermine their profitability through capital expenditure and (likely to come shortly in our opinion) price discounting in order to gain or maintain market share because they are locked in an existential battle with each other. Barriers to entry for providing the infrastructure needed by the hyperscalers seems high in the short term amid component shortages and constrained supply but are likely to be low in the longer term as well. In aggregate, these companies are now a huge component of global indices and this drives exposure in most client portfolios. We believe this is one of those (rare) times when it is pretty obvious. We believe it is important to avoid the “centre of the storm” and focus where we are more confident in our ability to identify companies with attractive prospects for earnings, sales, and other metrics that impact share prices over the longer term. If that means we lag market indices shorter term, then so be it.

The correction that never was

The strength of markets and the AI theme over the quarter somewhat masks the fact that the MSCI USA Index fell over 15% from 1^st to 25^th of June 2026. Broadcom missed on earnings and the new Fed Governor sounded hawkish after a particularly strong US jobs report. That was enough for investors to start worrying about precisely the things we are focused on and for the US market to sell off aggressively. But very strong results from semiconductor companies towards the end of the quarter came at the same time as an apparent peace deal in Iran and the US market reversed its decline aggressively into the quarter end.

Avoiding the storm

We are finding more and more compelling opportunities emerging in parts of the market that are being overlooked. When the fund launched in November 2025, the relative valuation of different “styles” and “themes” had compressed significantly over the preceding year. This has reversed in a short period of time, which we believe is a big positive for this strategy.

Defense stocks provide a good example. Despite it becoming ever more obvious that demand for their products is growing, defense stocks sold off significantly over the quarter. We sold Raytheon the week before the Iran war precisely because valuations had become stretched and the whole sector was becoming a momentum trade. But 4 short months later, valuations are back to levels that make no sense to us given the high predictability of future earnings compared with other industries. It seems the strong retail flows into defense stocks in January and February this year reversed, in order to chase the semiconductor stock momentum. To us what is happening in the global defense industry isn’t a short-term fashion or trade, it is a complex and very substantial shift in industry dynamics, involving a new capital cycle that will last a decade and beyond. It is interesting to note how governments are changing the way they contract with defense companies, for example, in order to encourage more innovation. Long term profit margins within the defense industry should improve and this does not appear to be well understood.

Of course, the defense sector is now a space that has proven it can become “fashionable” and sentiment is likely to play a role in shorter term share price movements.  We think there is substantial value to be added from taking advantage of sentiment cycles around themes where we have built deep knowledge, as with defense. Understanding what price to pay for different types of company will be critical, particularly when focused on absolute rather than relative returns.

Increased market exposure

In June we upped the Fund’s equity exposure significantly. We maintained the (roughly 20%) hedge ratio on European equities but increased gross equity exposure from 40.1% to 60.5% over the quarter with net equity exposure increasing from 30% to 44.4%. It is perhaps counterintuitive to increase equity exposure in a portfolio that aims to be very defensively positioned, but our confidence in the stocks held and the portfolio’s ability to withstand a downturn is high. We have not changed our stated risk tolerance, namely to construct the portfolio such that, based on our internal modelling, we would expect beta in a falling market to be below 0.2.We added to the more defensive/resilient parts of the portfolio in particular rather than to all positions, and added 3 new stocks.

New purchases

AG Barr - UK listed drinks company with market capitalisation c.£700m, yield c.4%

AG Barr was added to the portfolio as a Resilient stock. This theme is intended to provide moderate absolute return potential over the longer term, with relatively low beta and greater resilience. The theme has not worked well at all in recent months. It hasn’t been defensive enough in falling markets and it hasn’t participated in rising markets. Sentiment towards Consumer Staples in particular has been difficult to understand given the obvious economic impact of a protracted war in Iran. AG Barr sold off on apparent concerns that the war in Iran would impact costs but we think that was overdone. After many years as a rather sleepy Scottish company that sells Irn-Bru, largely in Scotland, there is evidence the company is flexing its growth ambitions. It has bought a number of small, branded drinks companies and is coming towards the end of a capital expenditure cycle that should cut costs long term and increase capacity. It seems perfectly reasonable to us that it can grow earnings at 10% per annum over the coming 5 years and that puts it on a very undemanding forward multiple.

Katakura - Japan listed conglomerate with market capitalisation c.£350m, yield c2.3%

When we first looked at Katakura 16 years ago it was a traditional low quality but profitable Japanese conglomerate that owned a frankly ridiculous amount of undeveloped land for legacy reasons. Now it is a traditional low quality but profitable Japanese conglomerate within a property company whose key asset (a shopping mall in north eastern Tokyo) is worth (we estimate) double the market capitalisation of the whole company. Overall, we believe the stock trades at a 70-80% discount to its net asset value, with huge amounts of cash, securities investments and property assets. It has plenty of scope to continue to invest in developing its property portfolio and from an earnings perspective is priced at around 12x current year profits. Katakura probably won’t fully embrace the “governance reform” theme that is driving many companies in Japan to sell cross-shareholdings and increase profitability mechanically. It is playing lip service to reform but at this stage that is all it is. But Katakura doesn’t need to do anything dramatic really. It has a perfectly good long term property business that can grow and a fortress balance sheet. We think it could become an indirect beneficiary of governance reform simply because the valuation is ludicrous. The discounts to net asset value for a number of the regional banks in Japan have reduced significantly over the last year without them doing anything dramatic. A similar narrowing of the discount for Katakura would increase its share price very significantly. In the meantime, domestic Japanese deep value stocks are often uncorrelated to global indices and there is a reasonable chance Katakura wouldn’t fall meaningfully if at all in a global rout driven by falling AI stocks, or a broader bear market driven by rising interest rates, for example. It is a small company and so the position is lower than is typical for the Fund (around 50bps once fully invested), but its potential impact is substantial.

Rheinmetall – German listed defence prime with market capitalisation c.£39bn, yield c.1.2%

We bought Rheinmetall after it fell almost 20% in a day, having drifted down materially from its peak before that. Even after that it needs to grow significantly over the coming years to qualify as “cheap”, but we are confident it will. Germany is mobilising and it cannot do that without Rheinmetall. The specific catalyst for the share price correction was around the German Government’s decision to abandon the F126 Frigate programme. It was a particularly public loss as Rheinmetall had only just bought the shipbuilder that was supposedly going to take over as the prime contractor for the F126. It was indeed a setback, but we think the market over-reacted and after the share price has more than halved, it now does not discount the incredibly strong strategic position that Rheinmetall is in. The stock complements the Fund’s other defense positions well, adding higher operating leverage and direct exposure to Germany.

Outright sales

Alibaba – Hong Kong listed e-commerce, cloud computing a payments company with market capitalisation c.£190bn and yield 0.9%

We are painfully aware that selling Alibaba could be a mistake. It is probably the best placed non-US listed company within the broad AI infrastructure space. We would have expected it to benefit over recent months and it didn’t, largely we think because AI capital expenditure drove its cashflow negative before the same thing happens to its US-listed peers. The option value of this stock in the wider portfolio was substantial but predictability was equally low. As we increased the gross exposure of the portfolio to stocks we wanted to improve predictability and Alibaba was a casualty of this process. It is very difficult to sell a stock we like, that is cheap and has very significant upside. But it is just as important to manage a portfolio that is simple and coherent.

Current portfolio

The table below shows the whole portfolio over the 3 quarter ends since inception. Broady speaking Resilient, Inevitable Growth and Geopolitical Winners are all stocks we expect to have low beta in a falling market. In aggregate however this part of the portfolio should generate total returns of over 10% per annum long term, based on our forward estimates of sales and forward 5 year PE. Return expectations for Countercyclical Potential, Japan Reform and Cyclical Winners are higher but with less predictability on downside beta. That said each position has idiosyncratic elements to it that we think improve the potential for attractive returns in challenging markets and overall market exposure remains modest. The portfolio’s bond exposure is short dated and should largely be regarded as “dry powder” to be deployed in the future, likely on market weakness. The US treasury position we think provides some tail risk protection in certain circumstances and augments the overall currency exposure of the portfolio.

As ever please talk to us if you have any questions about what we have covered in this letter. We are grateful for your support.