While many of the Orbis SICAV Global Balanced Fund’s long-held convictions have been rewarded over the last quarter, it is important to remember that the path to returns is never predictable. Portfolio manager Alec Cutler reflects on the drivers of the Fund’s strong performance and highlights some of the long-term opportunities that may deliver returns in years ahead.
2025 was a pleasing year for the Orbis SICAV Global Balanced Fund, which delivered strong absolute returns and again outpaced its benchmark and peers. As a team, we take pride in this performance. But good results are also a source of consternation for us, because we know a time will come when we look stupid. Indeed, many of our biggest winners in 2025 were once painful detractors. That is the nature of our work. We never know what path returns will take, only that they don’t come in a straight line. Stock markets illustrated that well last year. From mid-February to early April, world stock markets fell by more than 15%. Since then, they have roared ahead to new record highs.
What didn’t drive performance
With that backdrop in mind, note that we did not outperform this year by taking more stock market risk. Net of hedging, the Orbis SICAV Global Balanced Fund’s average stock market exposure was 58% – less than a passive 60/40 mix of stocks and bonds, and much less than some of our peers. Broadly, our preference for value shares and mid-cap companies was also a headwind.
Security selection also provided its share of humility, as only half of our stock selections outperformed. Several of our highest-conviction holdings lagged, including Burford Capital, Kinder Morgan, Cinemark, RXO and ANTA Sports. Our largest bond position, in long-term US Treasury Inflation Protected Securities, suffered in April and has barely recovered since.
What did drive performance
While only half our stock selections outperformed, we put more capital behind our winners in 2025, and some of those winners were substantial. Defence contractors continued to perform well as the reality sets in that Europe must defend itself. Energy infrastructure providers outperformed as investors came to appreciate the demand growth from ageing grids and power-hungry data centres. Semiconductor manufacturers rose strongly, as the worst memory downcycle since the global financial crisis gave way to an extreme – and extremely profitable – supply crunch.
More broadly, in 2025, non-US stocks and currencies outpaced the mighty S&P 500 and US dollar for the first time in years – a tailwind for relative results given our low exposure to US assets. Favouring gold, and increasingly gold miners, over government bonds contributed to performance as markets echoed our concerns about governments’ disregard for fiscal discipline.
What we did about it
As contrarian investors, when the Fund looks greatly discounted, we’re probably frustrated about performance, which is often the culprit for the discounts. When performance looks great, we worry about the Fund, fearful that attractive discounts have narrowed.
Fortunately, the latter problem has a straightforward solution: Rotate the Fund. In 2025, turnover was much higher than normal, as we rotated capital from appreciated winners into neglected ideas trading at what we viewed as deeper discounts.
On the other side of that rotation, we added to three areas: healthcare, high-conviction detractors and artificial intelligence (AI) consumables.
Our global analyst teams have unearthed compelling ideas in healthcare; spanning biotech drug developers, clinical testing businesses and equipment companies. Many of these were growth darlings just three years ago, but sentiment has soured post the COVID-19 pandemic, knocking valuations down to attractive levels. Having bought up these businesses over the last six months, healthcare now represents 10% of the Fund.
We have also added to many of our laggards, including all five of the equity detractors mentioned earlier. Where our assessment of the company’s worth remains high, we are happy to build larger positions at lower prices.
Finally, we have added to what we call “AI consumables”. AI spending continued to rise in 2025 as big tech companies vied for dominance in the war for AI supremacy. While our analysts believe that Alphabet’s AI advantages over Meta and OpenAI are underappreciated, within a moderate risk Fund, we believe that we can find names that sidestep that clash of titans altogether.
This thinking isn’t new. We’ve long believed that our AI infrastructure companies could benefit from the rising capital and competitive intensity of the tech giants, and we still do. However, looking forward, as valuations for critical energy infrastructure names increase, we have incrementally moved towards the manufacturers of AI consumables.
Counterintuitively, our consumables include computer chips, which are generally considered long-lived assets. But the bleeding-edge chips populating data centres are not forever assets. Most companies buying them pencil in depreciation over five or six years. But this obscures the economic cost of inefficiency. As the latest chips are more power-efficient than their forebears, running a data centre with old chips will incur higher power costs, so for some uses, only the best will suffice. With Nvidia designing new AI chips on an annual cadence, sales for Taiwan Semiconductor Manufacturing Company, which makes all of them, should be very healthy. So should sales for memory makers like Samsung.
Our other consumables are more obvious. Data centres have a voracious appetite for energy, and they need it 24/7. This is now reflected in the valuations of infrastructure providers and nuclear operators but not in the valuations of natural gas producers and transporters. As accessing grids becomes tougher, we expect that the tech giants will change their approach. Why go through the hassle of bringing energy hundreds of miles to a data centre when you can bring the data centre to the energy?
Marcellus Shale gas producers in Pennsylvania could benefit, as companies consider building data centres and gas turbines near gas fields. They’ve struggled since the 2010s, when shale oil drilling flooded the market with cheap byproduct gas. With oil prices down, drilling has slowed, tightening gas supply just as data centre demand rises – an attractive setup.
Through dogged research and opportunistic price-taking, we seek an elusive balance of being happy about performance and the Fund at the same time.
In the last quarter, we re-established a position in Alphabet, trimmed Nebius Group into share price strength and exited PDD Holdings to reallocate capital to higher conviction ideas.