Funds · Specialised funds
Sector and thematic funds: concentration, timing and the record
What sector and thematic ETFs hold, how concentrated they are, how far sectors diverge each year, and what happened to theme funds bought after their run-up.
On this page
A sector fund holds the companies of one part of the economy: technology, energy, banks, utilities. A thematic fund holds companies tied to an idea that cuts across sectors, such as clean energy, robotics, cybersecurity or "disruptive innovation". Both take a slice of the market instead of the whole of it, and both let an investor act on a view that the slice will do better than the rest.
The two differ in how the slice is drawn. Sectors come from a standard classification that assigns every company to exactly one group, so the sector funds of one family add up to the index with nothing left over. Themes are defined by whoever builds the fund. Two robotics funds can hold quite different companies, and one company can turn up in a dozen themes.
What sectors are and how the classification works is covered in Stock market sectors and industries: GICS and sector ETFs explained, and a tested strategy for switching between them is in Sector rotation strategy: momentum across nine S&P sectors, tested. This lesson looks at the funds as products: how concentrated they are, how widely their results spread, and what the record says about buying them after they have risen.
The idea
The Global Industry Classification Standard, GICS, sorts companies into four tiers, with eleven sectors at the top. State Street's Select Sector SPDR funds divide the S&P 500 along those lines, one fund per sector. Nine of them have traded since December 1998. Real estate was separated from financials in 2016, and the communication services fund began in June 2018.
Because the sector funds are just the index cut into pieces, owning all of them in index proportions is the same as owning the index. Owning one, or more of one than the index holds, is a bet that the piece will beat the whole. That is an active decision made with passive tools. The funds themselves are cheap: XLK, XLC and XLY each list a gross expense ratio of 0.08%. The cost of a sector position is not the fee. It is the risk of the bet.
A thematic fund makes the same kind of bet with a looser definition and usually a higher fee. Its index or manager chooses companies by their exposure to the theme, across sectors and often across countries.
How it works
Concentration inside the fund
A market-capitalization-weighted sector fund inherits the lopsidedness of its sector. When one or two companies dominate, they dominate the fund.
| Fund (as of Nov 17, 2026) | Holdings | Largest positions | Top companies' weight |
|---|---|---|---|
| XLK, technology | 76 | Apple 17.15%, Nvidia 12.56%, Microsoft 12.55% | 42.3% in three |
| XLC, communication services | 22 | Meta 20.83%, Alphabet 18.82% (two share classes) | 39.7% in two |
| XLY, consumer discretionary | 47 | Amazon 22.53%, Tesla 17.61% | 40.1% in two |
A holder of XLY has two-fifths of the position in two companies. A holder of XLC owns 22 stocks. These are sector bets in name and large single-company bets in practice, and the weights have to be read from the fund's holdings list to know which.
The concentration also stacks on top of whatever else the investor owns.
How far sectors spread
The chart shows, for each year from 2000 to 2025, the range between the best and worst of the nine original sector funds, with SPY marked.
The average distance between best and worst was 40 points and the median 34. The narrowest year was 2006 (11.9 points). The widest was 2022: energy rose 57.6% and consumer discretionary fell 36.8%, a gap of 94 points in a year when SPY lost 19.5%. In 2020 the gap was 79 points the other way round, with technology up 41.9% and energy down 36.9%.
| Year | Best sector fund | Worst sector fund | SPY |
|---|---|---|---|
| 2000 | Financials +24.1% | Technology −41.9% | −10.7% |
| 2008 | Staples −17.1% | Financials −56.7% | −38.3% |
| 2020 | Technology +41.9% | Energy −36.9% | +16.2% |
| 2022 | Energy +57.6% | Consumer discretionary −36.8% | −19.5% |
| 2025 | Technology +23.8% | Staples −1.2% | +16.4% |
With a spread that wide, picking the right sector is worth a great deal. The question is whether it can be done from what is known in advance. The simplest rule, buy last year's best, has no support in these 25 years. The year's best fund ranked 5.4 out of 9 on average in the following year, the middle of the pack. It finished in the top three 7 times and in the bottom three 12 times. The year's worst fund ranked 5.3 on average the next year and finished in the top three 10 times. Energy led in 2021 and 2022 and has been last six times. Technology led seven times and was last in 2000, 2001 and 2002.
How long a sector can stay down
| Fund, Jan 1999 – Oct 2026 (price) | Yearly rate | Volatility | Deepest fall | From → to | Regained its high |
|---|---|---|---|---|---|
| XLK, technology | 8.4% | 23.1% | −80.5% | Mar 2000 → Sep 2002 | Oct 2017 |
| XLF, financials | 3.9% | 21.0% | −80.0% | May 2007 → Feb 2009 | Feb 2021 |
| XLE, energy | 6.9% | 25.2% | −71.3% | Jun 2014 → Oct 2020 | Jan 2026 |
| XLV, health care | 7.1% | 14.3% | −37.4% | ||
| XLP, consumer staples | 4.4% | 12.7% | −35.0% | ||
| SPY | 6.6% | 15.1% | −52.2% |
Rates exclude dividends, which matter more for staples, utilities and energy than for technology. Three of the nine funds lost more than 70% at some point, against 52% for the index, and each took between 11 and 18 years to get back to its previous high on price. The technology fund that has led the market for a decade spent the seventeen years before that below its 2000 peak. Diversification across sectors is what keeps a broad index from doing the same.
Using it: thematic funds after the run-up
A theme usually reaches the public as a fund after it has already performed. The chart measures seven well-known thematic funds from their highest month-end value, with distributions reinvested, to October 2026.
| Fund (theme), since | Yearly return, fund / SPY | Peak | 12 months before peak | Since peak, fund / SPY | Deepest fall |
|---|---|---|---|---|---|
| ARKK (innovation), Dec 2014 | 13.0% / 13.3% | Jan 2021 | +169.7% | −40.2% / +120.4% | −77.1% |
| TAN (solar), Jun 2008 | −8.5% / 11.6% | Launch | −80.4% / +654.1% | −94.4% | |
| ICLN (clean energy), Aug 2008 | −4.3% / 12.3% | Launch | −55.9% / +717.5% | −86.1% | |
| LIT (lithium), Sep 2010 | 5.2% / 15.0% | Nov 2021 | +78.4% | −32.9% / +77.3% | −59.9% |
| BOTZ (robotics), Nov 2016 | 8.4% / 15.3% | May 2026 | +30.8% | −19.7% / +0.4% | −51.9% |
| HACK (cybersecurity), Jan 2015 | 14.4% / 13.5% | Oct 2026 | at its high | −34.0% | |
| KWEB (China internet), Sep 2013 | 0.9% / 14.2% | Feb 2021 | +84.3% | −69.7% / +114.5% | −77.1% |
Returns include distributions. One of the seven, HACK, beat SPY over its life. ARKK roughly matched it over twelve years, but only for someone who held throughout: the fund had returned 170% in the twelve months before its January 2021 peak, and anyone who bought near that peak has lost 40% since while SPY more than doubled. The two clean-energy funds launched in 2008 have never been above their first month-end value.
This is a small, hand-picked sample of survivors. The systematic evidence points the same way. Ben-David, Franzoni, Kim and Moussawi studied the whole US ETF market in the Review of Financial Studies (2023) and separated broad, cheap funds from specialized ones, meaning sector and thematic funds. They found that specialized ETFs are launched in segments that have recently had high returns and heavy media attention, hold stocks that are already expensive, and go on to lose about 30% in risk-adjusted terms over their first five years. Their explanation is that issuers compete for investors' attention by packaging whatever is popular.
Where it fails or misleads
The name is not the portfolio. A consumer discretionary fund that is 40% Amazon and Tesla is not a play on shopping. A theme fund's holdings depend on one index provider's definition. Read the top ten.
Yesterday's sector map changes. Classification changes move companies between funds. In 2018 the new communication services sector took large holdings out of technology and consumer discretionary, so a sector fund's history is not a record of one fixed set of businesses.
A hot theme can be right and the fund still lose. Solar power grew enormously after 2008. TAN's value fell 94% by November 2012. Growth in an industry is not the same as returns to its shareholders when the growth was already in the price and competition takes the profit.
Timing has to be right twice. A sector position needs an entry and an exit. The rule tested in Sector rotation strategy: momentum across nine S&P sectors, tested worked in some periods and not in others, and anything less systematic depends on judgment.
Overlap. Sector and theme funds added to a core index fund mostly add more of the largest companies the index already holds.
Compared with its neighbours
| Fund type | Defined by | Typical holdings | Fee level | Main risk |
|---|---|---|---|---|
| Broad index fund | The whole market | 500 to 4,000 | Lowest | The market's |
| Sector fund | Standard classification (GICS) | 20 to 80 | Low | One sector's long cycles; top-heavy weights |
| Thematic fund | The issuer's or index's definition | 30 to 100 | Higher | Bought after the run-up; narrow, expensive stocks |
| Factor fund | A measurable trait (value, size) | 100 to 1,000 | Low to moderate | Long stretches behind the market |
| Single-stock ETF | One company | One exposure | High | Everything (Single-stock ETFs: TSLL, NVDL, CONY and MSTY against their stocks) |
Holdings counts and fee levels are typical ranges, not rules. How an index's weighting method changes concentration is covered in Index weighting methods: cap, price, equal, fundamental and capped.
How Finabase does it
The Sector page shows the S&P 500's sectors. Each sector or thematic fund has a symbol page with a quote, chart and news. Its Holders tab can show a fund card with the issuer, expense ratio, assets, top holdings and sector weights when the fund-data feed has it; when it does not, the issuer's page has the list.
The comparison tab rebases a fund and SPY to percent change from the first bar in view and tabulates price returns from one week to two years, excluding dividends. It does not go back far enough to show a fund's whole history from its peak; the chart tab does.
Frequently asked questions
Are sector ETFs diversified?
Within the sector, only partly. XLK holds 76 companies, but three of them were 42% of the fund on November 17, 2026. Across the market, not at all: a sector fund removes the diversification between sectors that protects a broad index from one industry's decline.
What is the difference between a sector fund and a thematic fund?
A sector fund follows a standard classification in which every company belongs to one sector. A thematic fund follows a definition written by its issuer or index provider and can hold companies from any sector or country that fit the idea.
Does buying last year's best sector work?
Not in the nine Select Sector funds from 2000 to 2025. The previous year's best fund ranked 5.4 out of 9 on average the next year and finished in the bottom three more often than in the top three.
Why do thematic ETFs so often disappoint?
They tend to be launched, and bought, after the theme's stocks have already risen and become expensive. A 2023 study in the Review of Financial Studies found specialized ETFs lost about 30% in risk-adjusted terms over their first five years.
How many sector funds does an investor need?
None, to be diversified: a broad index fund already holds every sector at its market weight. A sector fund changes those weights on purpose.
Summary
Sector and thematic funds are cheap, convenient ways to make a concentrated bet. The concentration is greater than the labels suggest: two or three companies make up about 40% of several large sector funds. The pay-off for being right is large, since the best and worst sectors have finished 40 points apart in an average year, but last year's ranking has not predicted this year's, and a sector can spend more than a decade below an old high. Thematic funds add a looser definition, a higher fee and a tendency to arrive after the gains: six of seven large ones stood roughly 20% to 80% below their peaks in October 2026, and the academic evidence across all specialized ETFs shows the same pattern.
Check your understanding
5 questions on this lesson. Each answer comes with the reason, and nothing is scored.
An investor holds 20% of a portfolio in a technology sector fund in which one company has a 17.15% weight. How much of the portfolio does that company make up through this fund alone?
- 17.15%
- 3.4%
- 0.34%
Show the answer
B. 3.4% Multiply the fund’s share of the portfolio by the company’s weight in the fund: 0.20 × 17.15% = 3.4%. Any other fund that holds the same company adds to it.
From 2000 to 2025, how did the best of nine sector funds in one year rank the following year, on average?
- First or second
- About the middle: 5.4 out of 9
- Last
Show the answer
B. About the middle: 5.4 out of 9 The year’s best fund finished in the top three 7 times and the bottom three 12 times the next year. The previous year’s ranking carried almost no information.
The technology fund XLK fell 80.5% between March 2000 and September 2002. When did its price regain the 2000 high?
- 2004
- 2009
- October 2017
Show the answer
C. October 2017 A fall of 80% needs a gain of 400% to recover. XLK took about seventeen years on price. Financials and energy had similar falls and needed 11 to 14 years.
ARKK returned about 13% a year from December 2014 to October 2026, close to SPY. Why could a holder have done much worse?
- Its fee was higher than its return
- Anyone who bought near the January 2021 peak, after a 170% rise, has lost 40% since
- It paid no distributions
Show the answer
B. Anyone who bought near the January 2021 peak, after a 170% rise, has lost 40% since A fund’s return since launch is earned only by those who held throughout. Buying after a large rise is a different experience. From its peak to October 2026 ARKK lost 40.2% while SPY gained 120.4%.
What did Ben-David, Franzoni, Kim and Moussawi (2023) find about specialized ETFs?
- They outperform because they hold growing industries
- They are launched after segments have had high returns and attention, and lose about 30% risk-adjusted over their first five years
- Their underperformance is fully explained by fees
Show the answer
B. They are launched after segments have had high returns and attention, and lose about 30% risk-adjusted over their first five years The study attributes the underperformance to the overvaluation of the stocks the funds hold at launch, not to fees: issuers package what is already popular.
Sources and further reading
- State Street Global Advisors — Technology Select Sector SPDR Fund (XLK) fund page: expense ratio, number of holdings, top holdings as of November 17, 2026
- State Street Global Advisors — Communication Services (XLC) Select Sector SPDR fund page: inception June 18, 2018, holdings as of November 17, 2026
- State Street Global Advisors — Consumer Discretionary (XLY) Select Sector SPDR fund page, holdings as of November 17, 2026
- MSCI — The Global Industry Classification Standard (GICS): structure and the eleven sectors
- Itzhak Ben-David, Francesco Franzoni, Byungwook Kim and Rabih Moussawi, “Competition for Attention in the ETF Space,” The Review of Financial Studies, vol. 36, no. 3 (2023), pp. 987–1042
- Finabase chart data — month-end prices of the nine original Select Sector SPDR funds and SPY from January 1999 to October 2026 (price only); monthly prices and distributions of ARKK, TAN, ICLN, LIT, BOTZ, HACK, KWEB and SPY through October 2026; statistics computed by Finabase
