Thematic ETFs vs Building Your Own Basket: What the Research Shows

A thematic ETF gives you one ticker and someone else's holdings list; building the basket yourself gives you control of the holdings, the weights and the concentration, at the cost of doing the maintenance. The reason this choice deserves more thought than it usually gets is that the packaged version has a genuinely poor track record. Morningstar's research found that only about one in ten thematic funds both survived and beat a broad global equity index over the fifteen years to mid-2024, and that the average thematic fund kept almost half its assets in just ten holdings. This article compares the two routes honestly, including where the ETF is the better answer. It is educational and is not investment advice.

What the evidence says about thematic funds

Start with the numbers, because they are worse than most people expect and they shape everything that follows.

Read those together and the failure mode is obvious. Thematic funds get launched after a trend is already visible, which is when the constituent stocks are most expensive. Morningstar measured funds launched between 1996 and 2018 trailing the global index by 3.3 percentage points a year over their first two years. Then attention moves on, assets never arrive, and the fund closes. Funds that liquidated held a median of just $5 million a year before being wound up.

Why do thematic ETFs close so often?

Because a fund needs assets to survive and themes have short attention spans. Morningstar found that thematic funds which liquidated or merged held an average of $25 million twelve months before closing, with a median of $5 million, and that funds with at least $50 million in assets had far higher five-year survival rates. When management fees stop covering the cost of running the portfolio, the sponsor closes it. As of June 2020 only 45% of the 141 US thematic funds cleared that $50 million bar.

Closure is not just an inconvenience. You are liquidated on the sponsor's schedule, not yours, which means the proceeds and any realized gains land in whichever tax year they choose. For a US taxable account that is a real cost imposed by someone else's business decision.

The concentration problem nobody mentions

The second issue is more subtle and, for most themes, more damaging. Nearly every thematic fund holds the mega-cap incumbent that touches the theme, and because most of them weight by market cap, that incumbent becomes the largest position. You wanted exposure to the emerging pure plays; you got a portfolio where the trillion-dollar company you already own through your S&P 500 fund is 9% of it.

The average thematic fund put almost half its assets in ten names against about 12% for the MSCI ACWI Investable Market Index. That concentration is not necessarily wrong. Concentration is the point of a theme. What is wrong is that you cannot choose where it sits. If you already hold the incumbent elsewhere, the fund quietly doubles you up in it while diluting the exposure you were actually paying for.

What you get by building the basket yourself

Four things, and they map exactly onto the four problems above.

Holdings control. You decide whether the incumbent belongs. Excluding it is often the single biggest difference between your basket and the fund tracking the same theme, and it is the change a fund can never make for you.

Weighting control. Cap weighting hands the basket to the biggest name. Equal weighting gives the pure plays real influence. Custom weights let you express conviction that is not proportional to size. Running the same list under all three tells you how much of the fund's result was construction rather than theme, and the answer is usually more than expected. The mechanics are in equal weight index.

No closure risk. A list you maintain does not get wound up because it failed to gather assets. You exit when you decide to, in the tax year you choose.

You can test it first. This is the one a fund structurally cannot offer. Before you own anything you can run the exact holdings list against the S&P 500 over several windows, read the drawdown next to the return, and find out whether the theme survived the last bad year. That is what backtesting a portfolio is for, and it is free to be wrong at that stage.

What building it yourself actually costs

Being fair about the other side of this matters, because the DIY route is routinely undersold.

You place every trade, so a twenty-name basket is twenty orders instead of one, and rebalancing means another twenty. In a taxable US account each rebalance realizes gains you owe tax on that year, and there is no fund wrapper absorbing the internal turnover for you. You maintain the list, which means noticing when a company sells the division that put it in the theme, and deciding whether it stays. You need enough capital that twenty positions are each meaningful after commissions, though fractional shares have largely solved that. And you carry the discipline burden: a fund holds its losers mechanically, whereas you will be tempted to quietly drop yours.

None of that is prohibitive, but it is work, and if you will not do the work then the ETF is the better product for you even with the odds it carries.

Is it better to buy a thematic ETF or individual stocks?

Buy the ETF if you want the theme as a small satellite position, do not have a view on which companies within it are the real exposure, and value the single-ticker simplicity more than the roughly 0.60% a year it costs. Build the basket if you have specific views on which companies belong, if you already own the mega-cap incumbent and do not want more, if concentration limits matter to you, or if you want to test the construction before committing. The middle path many people miss is to design the basket first and then compare it against the available funds. If a fund matches what you designed, buy it and save yourself the work.

How many holdings should a thematic basket have?

Most self-built theme baskets land between 12 and 30 names. Below about ten, a single earnings miss decides your year and you are stock-picking rather than theme-investing. Above about forty, you have almost certainly diluted the theme with adjacent large caps and would be better served by a broad index at a fraction of the cost. The wider research on this is in how many stocks you need to diversify.

Whatever the count, write the theme down as a single sentence first and hold every candidate against it. If you cannot say in one sentence why a company belongs, it does not belong. That rule alone prevents most of the drift that turns a theme basket into an expensive large-cap fund.

Picking the theme itself

The construction advice above assumes you already have a theme. Choosing one well is a separate skill and the evidence suggests most people choose late, after a trend has been reported enough times to feel obvious. The useful signals tend to be commercial rather than narrative: which companies are actually winning contracts, hiring in the relevant function, or being talked about by customers rather than by analysts. Watching how a theme's companies are discussed in the wild, using something that tracks how brands are mentioned across the web and social, is a more honest early signal than a fund launch, which by definition arrives after the trend is established.

Then test the basket through a bad year. A theme that only works from a 2020 start date is not a theme, it is a window. Run it from several starting points, look at the worst drawdown as carefully as the return, and decide with the numbers in front of you. The full approach is on the thematic investing page, with worked cases in thematic investing examples.

The Morningstar figures cited here come from published research with the period end dates stated in the table, not from current-year data. Fund landscapes change. Indexes is educational and informational software for designing, backtesting and tracking an index. It never places trades, connects to a brokerage, or holds assets, and backtested results are hypothetical and do not predict future returns.