European equities: Conflict, energy and AI winners and losers

It has been a very eventful year thus far in 2026, so I am going to lay out our thinking on what has happened year to date, what we’ve changed in the funds and the reasons for the changes we’ve made.  We have made a handful of significant stock shifts, some of which occurred in January, and as per my January note I promised I’d come back with more detail on what we’ve been doing and why.  I’ll also start with a short update on our thoughts on the conflict in Iran and what it means for our investment views.   

The Iran conflict

I have already provided some initial thoughts on the Iran conflict, where I noted that this conflict was not a surprise as it had been well telegraphed in advance by the US Administration, although it has since become clear that not everyone was paying attention.  Whilst we were (and still are) cautious about making definitive statements about how this conflict evolves, our base case for investment purposes (partly informed by some of our advisors) has been that this campaign is likely to last for several weeks where the goals are:

1. To destroy Iran’s nuclear capabilities so that the “clerical- security oligarchy’’ controlling Iran is unable to develop nuclear weapons and use such weapons to destroy their neighbours
2. To destroy Iran’s ballistic missile capabilities (and supply chains) and to severely degrade conventional military capabilities so that the oligarchy is incapable of inflicting material damage on neighbouring countries, whilst also curtailing their capabilities to supply other rogue states with military hardware such as drones and missiles
3. To seriously weaken and degrade the leadership capabilities of the oligarchy so that their grip on Iran weakens, reducing their ability to continue destabilising the Middle East.

We have not seen anything that changes our base case, and evidence suggests that the US and Israel defence forces have been successful in materially degrading Iranian regime military capabilities, ballistic missile resources, and significantly weakening the leadership of the oligarchy.  There is also testimony that this operation, as well as that of last July has materially degraded Iran’s nuclear capabilities.  Nevertheless, as noted, we state (with low conviction) our base case that the conflict may last a few more weeks, and we remain vigilant to the need to change our view as such circumstances dictate. We will also not hesitate to make whatever portfolio changes are required because of that.

The key economic and market event has been the effective closure of the Straits of Hormuz by the Iranian regime with an inability thus far to reopen it.  Gas prices were the biggest risk to European economies, and the doubling of gas prices in the last month has brought that to the fore.  The rise in oil prices is a further factor that will constrain growth, if sustained, but the aspect we had underestimated was the damage caused by disruption of “product flow’’.   A significant amount of refinery assets are stuck behind the Straits of Hormuz pinch point and the lack of product such as jet kerosene, helium, fertiliser etc will have significant implications, particularly for Asian countries if sustained.  The longer the Straits of Hormuz remain closed the greater the risks to global economic growth and inflation with Asia impacted the most followed by Europe and the US the least (given its self-sufficiency in most energy areas).

Energy and resilience

What the spike in oil and gas prices, as well as the spike in refined product prices, should focus investor and policymakers’ minds on is a lack of resilience, especially in some large European countries, where no lessons seem to have been learned from 2022.  In the UK’s case, to have only a few days’ supply of gas storage points to policy failure, but Europe as a whole needs to think a lot more clearly about energy security; the costs of energy and supply chain resilience for key products and materials.  All of this points to the continued need for a major step up in investment in the capital equipment and material that underpins supply chains, energy resilience, and defence capabilities and this will require a deep change in policymaker mindset which is something that the leaders of Germany, Italy and France seem to grasp but not the UK.

Year-to-date market performance, stock dispersion and AI winners, losers

Although a short period, we note that European equity markets were sharply down from 1-9 March (the MSCI Europe index was down 6%)¹ in what appeared to be a “within-market’’ de-grossing / deleveraging’ most likely driven by hedge funds.  We would also note that this is in the context of having performed strongly in the first two months of the year – in which there was some significant volatility in markets and stock dispersion as discussed below: Jupiter European Select was slightly behind its benchmark.  We added to a number of positions that sold off heavily in the week including Arcelor Mittal, Kingspan, BBVA, Eurobank, Unicredit, Intesa SanPaolo.  We have bought back our positions in TotalEnergies and Shell, which is a change in tack. 

Performance for February and year to date

Figure 1 below lays out performance this year for the Jupiter European Select. The fund outperformed the benchmark in February and over the first two months of the year. European equities had a strong start to the year, following a strong end to 2025.  

Figure 1: Performance, Jupiter European Select SICAV

Past performance does not predict future returns. Source: Morningstar, Jupiter. NAV to NAV, gross income reinvested, net of fees. Jupiter European Select D EUR Acc, to 28.02.26. SFM is since fund manager start,12.05.25, when Niall Gallagher joined Jupiter. 

Performance
Jupiter European Select SICAV
Rolling 12-month performance %

Past performance does not predict future returns. Source: Morningstar, NAV to NAV, gross income reinvested, net of fees, Jupiter European Select D EUR Acc, to 28.02.26. Niall Gallagher took over the fund when he joined Jupiter on 12.05.25.

AI winners, losers and stock dispersion

There are a few points worth drilling down on.  First, there has been a very strong and almost “overarching” thematic of “AI winners’’ and “AI losers’’ that has driven wide stock dispersion within the market even though this has been narrow in terms of sectors.  Our funds have benefitted from this thematic driven dispersion with large positions in the perceived  beneficiaries of AI driven investment such as the semiconductor and semiconductor capex equipment stocks that benefit from hyper scaler build- out (ASM International, ASML and Infineon), as well as electrical equipment stocks that supply the electrical equipment that goes into datacentres and electrical grids (Schneider, Prysmian and more recently Siemens Energy).  Second, we owned few of the stocks that got “whacked’’ based on a perception that they are losers from the adoption of AI and AI driven innovation in areas such as enterprise software, vertical software, B2B services and wealth management.  We do have some ownership in information businesses such as LSEG and wealth managers such as FinecoBank and St James’s Place that sold off – on concerns that their “moats’’ faced breaching from new AI driven startups – but on balance we had more ownership of the perceived “AI winners” and a relative underweight in the so- perceived “AI losers”. 

Another interesting phenomenon has been the strong performance of businesses perceived to be neither winners, nor losers from AI but simply unaffected.  Stocks like KPN (telecoms), AB Inbev (beer) and some pharma names have performed strongly for this reason – after all people will still drink beer in an AI world – which might suggest a level of confusion within markets into the face of such violent thematic behaviour but such outperformance can be rationalised if one believes these companies will be able to apply AI to increase efficiency and operating profit margins. 

It is important not to get too drawn into the market narrative, and too reductive:  a number of the stocks that got supercharged by their strong AI winner status have other driving factors (e.g., decarbonisation driven electrification) whilst many of the AI loser stocks overlap with “quality growth’’ factors that have been de-rating for some time, and were simply too expensive before the market decided they were about to lose their moats (e.g., RELX).     

Some further thoughts

The year so far has been a thematic driven market:  we don’t want to underestimate the pace of capital investment to deploy AI capabilities, which is unprecedented, and the likely pace of innovation and disruption which is both profound, fast moving, and hard to predict.  However, we do have to have a framework for structuring what the build- out and deployment of AI might do to companies and sectors, and we also must pay attention to valuation, or what the market is asking us to pay for future cash flows.  Some points:

1.       Valuation matters.  I don’t want to succumb to my favourite pastime of quoting Warren Bufffett but there is only so much cashflow we can withdraw from a business over its lifetime so there is a value for everything based on what we think the business can do.  A case in point is ASML.  We’ve been following ASML intensively over the past 15 years and have been structurally positive on the stock for almost all of that period based on its role in the build out of compute & cloud computing, memory, AI etc, as well as the strategic competition between the US and China and the desire to win the compute/AI race and have self-reliance on semiconductors.  We’ve seen the stock priced on 50x earnings and 22x earnings within a period where the market has largely agreed with us – that is quite a range!  In periods where the market gets carried away it has made sense to lighten the position and when the market swerves towards pessimism it makes sense to increase the position.  The same is true on RELX (see point 2 below) and I’ll come back to that in more detail.   In the context of valuation mattering, we reduced our positions in ASML and ASM and our now underweight the sector – our perspectives on the businesses have not changed, this is purely a valuation call.  If the stocks were to fall back, we would likely buy back our positions but if the valuation were to continue to expand, we would likely reduce the positions further.

Figure 2:  ASML Price/Earnings Multiple – FY1

Figure 3: RELX Price/Earnings Multiple – FY1

1.       How to frame AI driven change.  This is perhaps the toughest task facing equity investors at the current time – and will likely remain so for a long- time to come.  Given the pace of change and all-encompassing nature of AI, we don’t want to be complacent and underestimate the extent of disruption.  However, we should also recognise that there are actors within markets and economies with something to sell so when we read Substack articles telling us the world is going to totally change by 2028, we might want to question the motivations of such writers and, more generally, some of the hype around this topic in the context of some mega- IPOs coming down the track.  Here is a simple framework and set of questions we ask ourselves when thinking through the opportunities or risks from AI, fully accepting that we are going to have to continuously refine the questions:

a.       Does AI directly drive demand for this company’s products and services?  This is the most obvious and simple part of the framework.  It plays to the semiconductor (Infineon) and semiconductor capex (ASML, ASM International) I discussed above and to the companies supplying the electrical equipment to data centres and electrical grids seeing greater electricity demand, based on AI driven demand growth.  But we can extrapolate this further into the underlying materials required to build this equipment (copper, aluminium) as well as some of the building components used to build datacentres (Kingspan) etc. 

b.       What are the processes, and with what costs attached, that this company might be able to apply AI to replacing humans with code?  Great examples here reside within the banking system, and this was very much the topic du jour for full-year results presentations made by bank management teams: loan origination and risk underwriting, AML / KYC (anti-money laundering and Know Your Customer), account openings etc are all areas with great (theoretical) automation potential.  The more thoughtful banks are also building functionalities where AI increases employee productivity, for example by aiding wealth managers/advisors to build wealth plans more quickly for customers so that the advisors can service more clients.  However, we also need to remember that banks are highly regulated entities with Prudential and Conduct of Business regulations, and they can only move as fast as regulators and central banks will allow them to. Also, Europe has strict employment laws – whilst AI may allow positions to be eliminated it is not clear that the employees freed up will be able to be made redundant, without very significant cost implications, so the cost take out is likely to rely on “age pyramids’’ with a gradual reduction in workforce as older workers retire.  In thinking through this “frame’’, there are many other companies and sectors where AI applications may help to produce cost and capital savings; for example, in the case of consumer stocks, AI could help with demand forecasting, supply chain optimisation as well as reducing the costs of external marketing – something Nestle management discussed in recent meetings.

c.       For businesses that could be attacked by AI – the “AI losers’’ what are the “moats’’ that might protect the business from AI native start- ups?  The key aspects we focus on here: are the extent to which a business has proprietary and unreplaceable data; the extent to which data has been augmented in a way that requires expert judgement that an AI LLM cannot do in a way that will be accepted (at least for now); the extent to which binding regulation touches/governs a business and its processes; and, the extent to which a business has embedded its products and services into customers workflow.  These three aspects are by no means complete and there will be other aspects we should ‘frame’ but for the moment they are the ones that seem most important to us.  To bring this to life, I’m going to use two real world examples where we have made portfolio changes / additions:

i.       RELX.  As can be seen in Figure 3, this stock has de-rated very significantly from a price of over 30x earnings to under 15x earnings in the space of a year, but it is important to remember the starting point; this is a good business with decent growth and historic defensiveness but the valuation became excessive, and the stock was something of a “poster child’’ for “quality growth’’ so a part of the de-rating journey overlaps with the significant de-rating of the quality growth factor in 2025.  The market began to get progressively concerned about the AI risk to RELX towards the end of 2025/early 2026, and this hit an apex with the announcement of Claude add-on functionality to legal in February – over the course of nine months the stock halved without any change in earnings expectations.  RELX fits the profile of a business where we think it is unlikely to get disrupted anytime soon:  much of the content in the Risk business is proprietary, industry standard and required by regulation, whilst the legal business contains content collected over decades and centuries that very often is not publicly available in digital form.  Beyond this, the legal content is often augmented by expert legal opinion and is deeply embedded into the client (law firm) workflows.  Most importantly, it is not a major cost to law firms, but it is absolutely mission critical as no litigator can take the risk of appearing in court without the most complete, and up to date legal precedents.  RELX management have been developing and applying AI for years and have thought through the consequences of what AI might mean for the industry and have been careful to ‘gatekeep’ access to their content.  We used the opportunity in the RELX share price cash to buy a new position as we saw significant upside to intrinsic value  

ii.       LSEG.  This is another stock that has de-rated over the last year largely based on concerns that new AI startups will replicating and replace parts of the Data business within LSEG.  We believe this materially misunderstands what the LSEG Data businesses offer customers; as with RELX, much of the data is proprietary and cannot be replicated, the nature of data usage requires extraordinary accuracy and latency in provision, some/much of the data has regulatory underpinnings and the data provision is deeply embedded into client workflows in a way that is hard to replicate.  Beyond embedding into client workflows, the collection and construction of the datasets require real time connectivity into over 600 exchanges processing vast amounts of data per second underpinned by significant infrastructure.  As with RELX, management has thought deeply about how to apply AI into business processes and how to “gatekeep’’ data; we do not see AI LLMs, and new AI native start-ups as competitors to LSEG but as customers and distribution channels.

We have bought back our positions in TotalEnergies and Shell.  We had sold our positions in these stocks towards the end of 2025 on concerns of a gas glut emerging in the 2026-2030 window and concerns that the oil price would remain low for some time. With the benefit of hindsight this was a mistake and we should have recognised the “option value’’ of higher-than-expected oil and gas prices.

One final change I want to highlight is our purchase of shares in ArcelorMittal, which clearly does not fit into an AI narrative!  The logic for the purchase of this stock is based on fundamental sector change and a Return on Capital Employed (ROCE ) inflection. For most of the last 15 years, ArcelorMittal has struggled to generate a ROCE above the cost of capital and was out of scope for us.  The reason for the low ROCE was NOT due to poor management – this is a very well-managed company – but the flooding of the European steel market by cheap imports.  Very often, these imports were not subject to the same growing stringency in carbon emissions as European steel makers and therefore benefitted from a significant cost advantage, taking market share, pushing down the cost of steel reducing both pricing and capacity utilisation for European steel companies.  Two profound changes now stand to change the ROCE profile of European steel makers: 

• First, the application of the Carbon Border Adjustment Mechanism from January 2026 will add on the cost of embedded carbon emissions in the form of a carbon tax.  This tax is born by the importer, and the importer bears the risk for any inaccurate calculation in the carbon tax
• Second, the European Commission has announced the application of tariffs and quotas on imported steel to be applied later this year.  Whilst this will increase the cost of steel for domestic users of steel, it is at last a recognition amongst policymakers of the importance of maintaining a steel industry – tanks and artillery are made of steel – demonstrating a burgeoning understanding of the importance of “resilience’’.  We see no such recognition in the UK, and it is plausible that the UK ends up losing the remnants of its steel industry for good.

What these initiatives will do is raise the price of steel, crowd out imports, and replace them with domestic supply thereby increasing capacity utilisation.  We believe this will significantly raise ROCE at ArcelorMittal over the next few years and this is not reflected in the stock price.       

Fund risks

• Currency (FX) Risk - The fund can be exposed to different currencies and movements in foreign exchange rates can cause the value of investments to fall as well as rise.
• Share Class Hedging Risk - The share class hedging process can cause the value of investments to fall due to market movements, rebalancing considerations and, in extreme circumstances, default by the counterparty providing the hedging contract.
• Pricing Risk - Price movements in financial assets mean the value of assets can fall as well as rise, with this risk typically amplified in more volatile market conditions.
• Market Concentration Risk (Geographical Region/Country) - Investing in a particular country or geographic region can cause the value of this investment to rise or fall more relative to investments whose focus is spread more globally in nature.
• Derivative risk - the fund may use derivatives to reduce costs and/or the overall risk of the strategy (this is also known as Efficient Portfolio Management or "EPM"). Derivatives involve a level of risk, however, for EPM they should not increase the overall riskiness of the strategy.
• Liquidity Risk (general) - During difficult market conditions there may not be enough investors to buy and sell certain investments. This may have an impact on the value of the fund.
• Counterparty Default Risk - The risk of losses due to the default of a counterparty on a derivatives contract or a custodian that is safeguarding the fund’s assets.

Important information

This is a marketing communication. This document is intended for investment professionals and is not for the use or benefit of other persons, including retail investors. This document is for informational purposes only and is not investment advice. The value of investments and income may go down as well as up and investors may not get back amounts originally invested. An investment is designed to be held over a longer term. Initial charges may have a significant impact on returns if the investment is withdrawn in the shorter term. Exchange rate changes may cause the value of investments to fall as well as rise. Past performance does not predict future returns.

Where a benchmark is used for comparison, it is shown for illustrative purposes only and does not imply future performance. The views expressed are those of the individuals mentioned at the time of writing, are not necessarily those of Jupiter as a whole, and may be subject to change. This is particularly true during periods of rapidly changing market circumstances. Company or holding examples are for illustrative purposes only and are not a recommendation to buy or sell. Every effort is made to ensure the accuracy of the information, but no assurance or warranties are given.

The Company is a UCITS fund incorporated as a Société Anonyme in Luxembourg and organised as a Société  d’Investissement à Capital Variable (SICAV). Please refer to the latest Prospectus and to the Key Investor Information Document (KIID) (for investors based in the UK) and Key Information Document (KID) (for investors based in the EU) before making any investment decision.  Particularly to the sub-fund’s investment objective,  characteristics including those related to ESG (if applicable), and additional risk factors.

These documents are available from www.jupiteram.com or from www.eifs.lu/jupiteram or from:

France: CACEIS Bank France (Centralising Agent), 1/3 Place Valhubert, 75013 Paris, France. 

Italy: Allfunds Bank, S.A.U., Milan Branch, Via Bocchetto 6, 20123 Milano, Italy. CACEIS Bank, Italy Branch Via Piazza Cavour 2,20121 Milano, Italy. Société Générale Securities Services, Via Benigno Crespi 19, 20159 Milano, Italy. The sub-fund has been registered with the Commissione Nazionale per le Società e la Borsa (CONSOB) for the offer in Italy to retail investors. 

Luxembourg: the Company’s registered office: 31 Z.A. Bourmicht L-8070 Bertrange, Grand Duchy of Luxembourg. 

Spain: Allfunds Bank, C/ La Estafeta 6, Edificio 3, 28109 Alcobendas, Madrid, Spain. For the purposes of distribution in Spain, the Company is registered with the Spanish Securities Markets Commission – Comisión Nacional del Mercado de Valores (“CNMV”) under registration number 1253, where complete information, including a copy of the marketing memorandum, is available from the Company authorised distributors. Subscriptions should be made through a locally authorised distributor. The net asset value is available on www.jupiteram.com.

Hong Kong: Jupiter Asset Management (Hong Kong) Limited, Suite 1706, Alexandra House, 18, Chater Road, Central, Hong Kong. 

Switzerland: The representative in Switzerland is FIRST INDEPENDENT FUND SERVICES LTD., Feldeggstrasse 12, CH-8008 Zurich. The paying agent in Switzerland is NPB New Private Bank Ltd., Limmatquai 1, CH-8001 Zurich. The prospectus, the key information documents, the articles of association as well as the annual and semi-annual reports may be obtained free of charge from the representative in Switzerland.

United Kingdom:  Jupiter Asset Management Limited (the Investment Manager and UK Facilities Agent), registered address: The Zig Zag Building, 70 Victoria Street, London, SW1E 6SQ, United Kingdom, authorised and regulated by the Financial Conduct Authority.

Information is also available on how subscriptions and redemptions can be made, and arrangements related to investor rights and complaints handling.

Unless otherwise specified in this document, The Bank of New York Mellon SA/NV, Luxembourg Branch (the Company’s Depositary and Administrator) is responsible for processing subscription, repurchase and redemption orders and making other payments to Shareholders. The Bank of New York Mellon SA/NV, Luxembourg Branch, Luxembourg Branch,2-4,Rue Eugène Ruppert, L-2453 Luxembourg, email Distributors: JupiterLUXdistributor@bny.com, email Investors: JupiterLUXinvestor@bny.com