A thematic ETF offers an easy way to invest in megatrends such as artificial intelligence, defence or clean energy. However, a compelling growth story does not automatically make a good investment. In this article, we examine what the data reveals about thematic funds and how to evaluate an ETF’s structure, timing and role in your portfolio.
This content has been produced by Kvarn Investment Services Ltd, a licensed investment firm supervised by the Finnish Financial Supervisory Authority. The content is intended for informational purposes only and should not be interpreted as investment advice or recommendation. All investing involves risks, and past performance is not a guarantee of future returns.
Artificial intelligence. Defence. Cybersecurity. Clean energy. Space. Today, investors can tap into the defining stories of our time with a single click: cheaply, passively and in a way that feels compelling. But in the stock market, a good story and a good investment are two different things, and the gap between them can cost investors billions every year. In this article, we examine what the data says about thematic funds, why timing undermines so many investments, and how to distinguish a genuine structural driver from a marketing story using three product checks and one portfolio check.
Imagine that the core of your portfolio already consists of a broad global index fund, perhaps with some bonds and cash. Then the news starts repeating the same words: artificial intelligence, data centres, defence spending, electrification. You begin to wonder whether you, too, should own a piece of the very change everyone is talking about.
The idea seems sensible. Why settle for a broad market index when you can buy a direct stake in the future?
This is exactly the appeal of a thematic ETF. It packages a complex global shift into a single button: the product looks passive, diversified and modern. But that is precisely where the trap lies, and that is what we will examine next.
A thematic ETF is an exchange-traded fund that invests in a specific investment theme. A traditional ETF may track, for example, the S&P 500 Index, the European equity market or global equities.
A thematic ETF, by contrast, focuses on a specific future-oriented phenomenon. Themes may include:
From an investor’s perspective, the idea is clear: if a trend grows strongly in the future, the companies associated with it should benefit.
But the market is not that simple.
Two differences from a traditional index fund are worth remembering. First, a conventional index fund owns hundreds or thousands of companies and does not try to predict the winners in advance. Thematic ETFs are often more concentrated than ETFs that track broad market indices. The number of holdings varies by product, from a few dozen to several hundred companies, and the selections are based on the index provider’s own definition of the theme. In other words, the product does not really track the market. It makes active choices inside a passive-looking wrapper, and the index it follows is often created specifically for that fund.
Second, it costs more. Broad index ETFs typically charge 0.05–0.20 per cent a year, while thematic ETFs generally charge 0.30–0.75 per cent. The difference may sound small, but it is a hurdle the theme must clear every year merely to keep pace with a conventional index.
Investors rarely think in terms of index methodologies, market-cap weights or factor premia. People think in stories. A doctor understands the importance of diagnostics, a programmer sees the transformative potential of artificial intelligence, and someone who has worked in defence understands the role of the defence industry. Dry risk-premium theory is much harder to make compelling.
Stories make investing tangible, which is precisely why thematic ETFs have grown into a vast market. According to ETFGI, around 468 billion dollars was invested in thematic ETFs worldwide at the end of November 2025, across a total of 1,679 products. Thematic investing is no longer a marginal phenomenon; it is a major part of the competition among fund providers.
Assets under management in thematic ETFs have grown from a few billion dollars to almost half a trillion. Popularity, however, is not a mark of quality.
Popularity tells you that a product attracts interest. It does not tell you whether the product makes sense for your portfolio. The essential question is not whether the trend is real, but who will profit from it and at what price an investor can gain exposure.
For many thematic ETFs, the greatest weakness is not the theme itself but the timing of the launch. A fund provider does not usually launch a product when a theme is unknown, cheap and boring. The product appears only once the trend has attracted enough public attention to make it marketable. By then, the market has often already priced optimistic expectations into share prices.
The pattern is usually familiar:
A thematic ETF is therefore rarely a leading indicator. More often, it is a delayed echo of a move that is already well under way. A recent example emerged ahead of SpaceX’s June 2026 listing. Anticipation of the listing sharply increased interest in ETFs offering SpaceX exposure, shares linked to the company’s supply chain, and other space companies. Space stocks had already been rising for months before the listing, but part of the rally unwound on SpaceX’s first trading day as investors took profits.
The broader thematic fund market shows the same pattern. Morningstar’s dataset includes conventional mutual funds as well as thematic ETFs. According to the data, assets in global thematic funds grew from around 255 billion dollars in 2018 to around 806 billion dollars by the end of 2021. A record 589 new thematic funds were launched that year. The best-known example is Cathie Wood’s ARK Innovation, which returned more than 150 per cent in 2020 and attracted substantial new capital after the rally. When the market turned, assets in thematic funds fell sharply, and in the United States fund closures exceeded new thematic fund launches in the first half of 2024 for the first time since 2008.
The direction reversed again in 2025: over three quarters, more than twice as many new thematic funds were launched as in the previous year, and 19 billion dollars of new money flowed into US thematic funds. Demand has therefore not disappeared, even though the long-term statistics give cause for caution. Nor does being late automatically mean making the wrong decision: some structural changes last for decades. But even a genuine trend can disappoint if expectations are already too high. Clean energy, electric vehicles and, in their time, the internet were all genuine structural shifts, yet many investors earned poor returns by buying during the hype phase.
Fund research firm Morningstar has tracked thematic funds for many years, and the results make sobering reading. Over the fifteen-year period to June 2024, around 15 per cent of thematic funds managed to do two things at once: survive and outperform the broad global equity index. More than half, around 55 per cent, were closed or merged into another fund.
The easiest way to understand this is to look at one hundred funds:
Of one hundred thematic funds, 55 were closed or merged, 30 survived but underperformed the global equity index, and 15 both survived and outperformed.
The picture is similarly weak over shorter periods. Over the three-year period to June 2024, only around 9 per cent of thematic funds both remained in operation and outperformed the global equity market. Success rates were also below one fifth over five- and ten-year periods.
When examining fund performance statistics, investors should be alert to survivorship bias. This occurs when the analysis includes only funds that are still operating, while closed or merged products are excluded from the dataset. Historical fund performance may then appear better than it really was. Comparisons should therefore also account for products that did not survive the full measurement period.
A short life cycle creates its own problems. In Morningstar’s historical data, around 94 per cent of thematic funds survived their first year, but fewer than 15 per cent reached the age of 15. A fund closure forces investors to realise or transfer their investment and find a new product. This can create tax consequences, trading costs and reinvestment risk, even when the assets can be invested elsewhere.
The data does not make the themes fictional or thematic ETFs automatically bad. It shows that thematic investing is difficult: you must choose the right theme, the right fund and the right timing simultaneously.
The problem becomes more pronounced when we examine the investor return gap. This does not measure what the fund earned on paper, but the return investors actually received as a result of their own buying and selling decisions. An investor who buys after a steep rise and sells during a decline earns less than the fund’s reported return would suggest.
The gap is exceptionally wide in thematic products. According to Morningstar, thematic funds returned +7.3 per cent a year over the five years to June 2023, but investors received only +2.4 per cent because of poor timing.
Fund returns and investor returns are two different things. In thematic funds, the gap is almost five percentage points a year.
The gap is not created inside the fund but by investor behaviour: investors enter at the peak of the hype and exit at the bottom of the disappointment. This is worth recognising before pressing the buy button, because it is the single most expensive mistake in thematic investing.
Timing, however, is only the first obstacle. The second is more fundamental: an accurate forecast of the future is not, by itself, a good investment strategy. Capital markets are a discounting machine: they price expectations about the future, not the present. That is precisely why even a company in a growing industry can prove to be a disastrous investment.
When you buy a theme, you are not buying its importance to the economy or society. You are buying the difference between today’s valuation and future reality, above all future profitability. This is the most common conceptual error in thematic investing: confusing a megatrend with an investment case.
A megatrend answers the question: what is growing or changing in the world? An investment case answers a different question: which company can benefit from the change profitably, with a sustainable competitive advantage and at a reasonable price?
Solar energy is a textbook example. Installed solar power capacity has multiplied over the past fifteen years, yet many solar energy shares and funds have still lagged far behind the global equity index. The explanation is simple: where there is a great deal of growth, there is also a great deal of competition. And where many companies compete with similar products, prices and margins fall precisely as volumes rise.
The same set of questions should be asked of every theme:
A theme can therefore be hugely significant to society and still be difficult as an investment product.
The third, often underestimated issue is that a thematic ETF looks passive, but major active choices are made inside it. The most important choice is made when the theme itself is defined. What counts as an artificial intelligence company? Is it Nvidia, which makes the chips required for computing? Microsoft, which sells cloud services and AI tools? A consultancy that helps other companies adopt AI? Or an industrial company that uses AI in its own production?
In narrowly defined or new themes, the investment universe is often small, and the fund provider typically has two options.
The first is to offer pure exposure to the theme through so-called pure-play companies. These companies are closely tied to the theme, but they are often small, loss-making, highly valued or even privately held. The result is a narrow portfolio of only a few companies, with high single-stock and sector risk.
The second option is to broaden the portfolio by including companies with only a loose connection to the theme. A space ETF may therefore include defence companies, satellite operators, semiconductor manufacturers and telecoms companies. There is nothing inherently wrong with this, but it reveals that the product does not buy a pure idea. It buys one methodological interpretation of that idea.
Structure matters. Two ETFs with similar names can conceal entirely different risks: one may be concentrated in a handful of high-growth companies, while the other may ultimately resemble a technology index with a new label. A fund’s name often tells you less than you might think, because the real answer lies in its holdings and the index methodology. Remember this as well: in a narrow theme, even an ETF with a few dozen shares can represent a sector concentration. When every company is supported by the same story, a crisis can bring down not just one company but the entire narrative.
Not every popular investment theme is long-lasting, and even a genuine structural shift does not automatically become a good investment. The first task is therefore to determine whether the theme is already visible in the real economy: in investment, electricity consumption, supply chains, regulation or public budgets.
It is useful to borrow the thinking of major investors. BlackRock Investment Institute refers to mega forces: large structural shifts that shape the economy and investing both now and for decades to come. BlackRock identifies five: demographic divergence, digital disruption and artificial intelligence, geopolitical fragmentation, the transformation of finance and the low-carbon transition.
The central point is this: a mega force is not a product name but a structural change that affects growth, inflation and corporate profitability across sectors. A genuine structural driver can be recognised because it is already visible in hard data: money, electricity and budgets. Let us look at two examples.
It is easy to think of artificial intelligence as merely software or a chatbot, but the figures tell a different story. J.P. Morgan Asset Management estimates that cumulative investment related to artificial intelligence and data centres could reach around USD 5.3 trillion by 2030. The combined capital expenditure of five major US technology companies—Alphabet, Amazon, Meta, Microsoft and Oracle—is meanwhile estimated to rise from around 416 billion dollars in 2025 to more than 730 billion dollars in 2026.
At the same time, the International Energy Agency (IEA) estimates that data-centre electricity consumption will more than double by 2030, with artificial intelligence as the main driver.
When a theme is visible in power grids, it is no longer merely a story. Source: IEA.
Artificial intelligence is therefore not merely a software theme for investors. It is equally a semiconductor, electricity, power-grid, data-centre and infrastructure theme: a physical investment wave, not just a narrative. This is the hallmark of a genuine structural driver.
According to the Stockholm International Peace Research Institute (SIPRI), global military expenditure rose to around USD 2.718 trillion in 2024. Real growth was 9.4 per cent, the sharpest annual increase since the end of the Cold War, and as much as 17 per cent in Europe. In NATO’s Hague Summit Declaration, member states committed to raising defence and security spending to five per cent of GDP by 2035. Investors do not need to guess whether money is moving: it is directly visible in government decisions.
This is the decisive difference from a marketing story: you do not have to believe a promise about the future. You can already see the structural driver in investment, electricity consumption, supply chains and budgets.
Now the theory becomes practical. Before adding a thematic ETF to your portfolio, work through three product-related checks. Then check how much the product overlaps with your existing investments.
For a practical comparison of a fund’s fees, size, spread and other key figures, read our guide How to Choose an ETF in Practice.
Is there a concrete force behind the theme that compels money to move: a budget, regulation, demographics, demand for electricity, a supply-chain bottleneck or falling technology costs? If the theme cannot withstand a ten-year test without first requiring you to believe in a buzzword, it is not a megatrend but a cyclical trend or a fad. Be especially wary of conditional claims: “when artificial intelligence finally takes off”, “when the metaverse eventually materialises”, “once hydrogen replaces oil”. Every “when” conceals timing risk that you bear.
Next, examine how closely the businesses held by the fund are actually connected to the theme. Do the holdings already generate a significant share of their revenue from the theme, or are the selections based mainly on expectations of future growth?
Some thematic indices require a company to derive, for example, at least 50 per cent of its revenue from the theme. In others, selection may be based on broader classifications, expert judgement or an expectation that the company will benefit from the theme in the future. The less visible the connection is in current revenue, and the more the selection relies on forecasts, the greater the risk that the fund’s name promises more than its holdings actually deliver.
Always examine the index methodology: how companies are selected, how much revenue exposure is required and whether companies whose connection to the theme is still indirect may also be included. The real answer is found in the index rules, not in the fund’s name or marketing copy.
Even when the theme and methodology are sound, the product can still wither away. Look at the fund’s size, trading activity and fees. A fund with less than one hundred million euros combined with a cooling theme is a classic sign of closure risk, while a high fee on top of a small size and narrow theme is a toxic combination. Also check daily trading volume and the difference between the bid and ask prices, or the spread: thin trading in a small fund can erode returns by more than the management fee itself.
Many investors already own Nvidia, Microsoft, Alphabet, Amazon, Broadcom and ASML through a broad global index. Open the thematic ETF’s ten largest holdings and compare them with the ten largest holdings in your own index fund. If you find the same names—typically these very giants in an artificial intelligence ETF—you are not buying new exposure. You are paying for the same shares a second time, at a higher fee, and concentrating your portfolio instead of diversifying it. This can be thought of as hidden beta: the name promises a new investment angle, but the actual exposure is the same market risk already present in the portfolio.
Overlap with a global index varies by theme and product. European defence or rare earths may introduce a new type of exposure to a portfolio. A semiconductor ETF, by contrast, may increase the weight of companies already held by a broad global index.
The framework is easiest to understand by comparing products in the same way. The following four examples examine three questions: how the index defines the theme, how concentrated the implementation is, and what investors should monitor in the product’s viability or risk structure. The purpose is not to rank the products.
VanEck Semiconductor UCITS ETF provides concentrated exposure to semiconductor manufacturers and companies that supply production equipment. Artificial intelligence, cloud services, electric vehicles and industrial automation support long-term demand for semiconductors, but the industry is also highly cyclical.
The index requires companies to derive at least 50 per cent of their revenue from semiconductors or related business. This gives the fund a relatively direct connection to the theme.
The fund’s annual fee is 0.35 per cent. Individual company weights are capped, but it remains an ETF concentrated in a single sector. Some of its largest holdings are already included in broad global indices, so investors should check both the overlap and their total semiconductor exposure.
The fund provides Europe-focused exposure to companies that benefit from defence and cyber-defence spending by NATO countries. The underlying structural driver is already visible in government budgets and defence-industry order books.
The index limits the investment universe based on company location and the revenue generated from defence activities. As a rule, companies must have significant revenue exposure to defence equipment, defence technology or cybersecurity.
The fund’s annual fee is 0.39 per cent. The fund is still young and has a fairly concentrated set of holdings. Investors should therefore monitor its size, trading, spread and the development of individual company weights.
The fund invests in companies involved in areas such as hydrogen production, fuel cells, electrolysers and the integration of hydrogen into energy systems. However, the long-term potential of the hydrogen economy does not automatically mean that the current pool of listed companies is broad or profitable.
The fund’s investment universe is small, which is reflected in a limited number of holdings and large individual company weights. The theme’s commercial development also depends on factors including falling costs, infrastructure construction and industrial demand.
The fund’s annual fee is 0.50 per cent. A small fund size, a limited set of holdings and weak trading can increase concentration and closure risk. These factors should be assessed separately from whether an investor believes in hydrogen’s long-term role.
The fund provides global exposure to companies linked to the clean-energy transition. Its investment universe is broader than that of the hydrogen fund and covers several technologies and parts of the value chain.
The performance of these companies is nevertheless strongly affected by interest rates, energy policy, subsidy schemes, raw-material prices, competition and fluctuations in production capacity. A genuine structural change can therefore be accompanied by long periods of weak investment returns.
The fund’s annual fee is 0.65 per cent. Its large size generally reduces closure risk, but does not eliminate industry cyclicality, valuation risk or the importance of the purchase date.
This perhaps captures the article’s most important lesson: the wrong price at the wrong time can nullify even a good theme. A fund’s large size generally reduces closure risk, but it does not protect investors from weak returns or poor timing.
Four cases side by side. The annual fee is the management fee charged by the fund, while size refers to its assets under management; the smaller the fund, the greater the risk that the product will be closed.
A thematic ETF is best used as a satellite, not as the core of a portfolio. The core is the part on which long-term wealth building rests, often a broad, low-cost and well-diversified index fund or ETF. A satellite is a smaller, deliberate layer on top, allowing investors to express a view by overweighting areas such as semiconductors, European defence or critical metals.
Two rules of thumb go a long way. First, no single narrow theme should ever be capable of bringing down your entire portfolio. Second, unless you can explain aloud why the theme should still belong in your portfolio five years from now, it does not belong there today. In that case, you may own a news headline rather than an investment case.
A genuine megatrend is visible in companies’ current revenue, investment and value chains. A marketing story is far more visible in the fund’s name than in its holdings.
Morningstar’s data shows that many thematic funds are launched only after a theme has attracted substantial attention, that a large share of products are closed over the long term, and that the timing of investor cash flows can weaken realised returns.
The mega force framework helps identify which themes are genuinely structural. Figures from the IEA, SIPRI and NATO, meanwhile, reveal where money, electricity and budgets are actually moving.
The next time you encounter a thematic ETF that promises a megatrend cheaply and effortlessly, do not start with the name. Start with three checks: the structural driver, revenue exposure and viability. Then make sure you do not already own the theme, and keep it as a satellite. What matters is never the name, but what has actually been packaged beneath it.
The information and sources presented are for illustrative purposes only. While obtained from sources deemed reliable, their accuracy cannot be guaranteed.
Earnings season is one of the most important moments for investors to understand what is really happening inside listed companies. In this article, we explain what earnings season means, which figures matter most, and why share prices can react sharply even when the headline result looks strong.
What should investors watch in the second half of 2026? The first half showed that the same market forces can be both risks and opportunities. Kvarn’s mid-year outlook looks at both sides of fragmentation, inflation, artificial intelligence and diversification, and shows how a European investor can build a portfolio that does not depend on a single forecast coming true.
Start investing today
Create account