The Compounding Punishment of Themes: Morningstar Shows Chasers Lost 7% a Year — and MPF’s Next Theme Is Already on Its Way

Track 1 Data X-Ray | 2026-10-06

The core proposition

The most profitable stories of the past decade often came with the most expensive tuition. Morningstar’s September 2026 study of thematic ETFs (by Jeff Ptak) delivers a brutal number: the average thematic ETF returned 10.5% annualised over ten years — roughly 5 percentage points a year behind the S&P 500. What investors actually earned was worse still: over the three years to November 30, 2024, the ETFs themselves lost about 1% a year while the average dollar invested in them lost more than 7% a year. What the fund earned is one thing; what you earned is another — and theme-chasers lost on both counts.

The more important finding is Ptak’s pattern: the longer the measurement period, the wider the gap between investor returns and the fund’s published total return. The mistake is not a one-off; it compounds. Poorly timed purchases and sales from prior years stack up, and compounding punishes them year after year.

MPF members may assume this is a story about American ETF investors, nothing to do with them. Look at the local league table: the Haitong Korea Fund (Class T) is up 77.42% year-to-date, while the BOC-Prudential China Equity Fund is down 10.11% — a spread of 87.53 percentage points (MPFGo September ranking, reported by on.cc on 2026-10-02). The halo of the top performer is precisely the bait for theme-chasing. And the next theme — the gold ETF — already has its gate open: the MPFA opened it on July 7, 2026 (Guidelines III.10, 15th edition: SFC-authorised, HKEX-listed, physical-backed only, capped at 10% of NAV). The chasing script is being rewritten locally.

X-ray: published returns versus investor reality

The gap is manufactured by flows arriving exactly when they shouldn’t. In 2020–21, investors poured a net US$95 billion into thematic ETFs — buying at the peak of a year in which the average thematic ETF rose 58%. In the 12 months that followed, the average thematic ETF fell 13%. In the 12 months to February 2023, investors yanked US$12 billion out — selling at the bottom of a year the average ETF lost 21%. In the year that followed, the average ETF gained 14%. Buy high, sell low, with precision.

Since 2024, they have come back: US$76 billion of net inflows, led by artificial intelligence, energy transition (smart grids) and security (defence tech). AI-themed ETFs routinely posted 20%+ rolling 12-month returns, and money followed the scent — a rerun of the 2020–21 script.

Ptak’s other verdict deserves to be carved on a wall: “The hotter a fund is to handle, that is, the more volatile its returns, the less of that fund’s total returns investors capture. They get rattled.” The hotter the fund, the wider the gap — and the relationship holds even after controlling for fund type.

Now the local lens. September’s MPF league table spans 7.83 percentage points (Haitong Korea -T at +3.37% versus Fidelity’s Hong Kong equity fund at -4.46%); year-to-date the span is 87.53 percentage points. GUM’s August report shows Asia equity funds up 27.9% in the first half, leading all categories — that is where the AI-semiconductor theme lives inside MPF. When a +77.42% number makes the headlines, the next step is money chasing it at the top. This is not a forecast; it is human nature, verified by Morningstar with US$95 billion of flows.

The compounding cost: the gap is not deducted once — it is deducted for thirty years

A behaviour gap is a tax, levied on the compounding engine every year, just like a fee. Illustrative calculations (monthly contributions of HK$5,000):

• A 2.5-percentage-point behaviour tax (the empirically measured gap for international equity ETFs): over 30 years, a 7% annual return grows to roughly HK$6.10 million; at 4.5% it reaches only about HK$3.80 million — a gap of roughly HK$2.30 million, 6.7 times the average MPF balance of HK$343,242 (MPF Ratings, 2026-09-24), or about 38 years of total contributions.
• The extreme version of theme-chasing: the same monthly contribution over 10 years at 7% a year grows to roughly HK$865,000; at the thematic-ETF investor’s average of -7% a year, it shrinks to about HK$432,000 — half simply evaporates, a gap of roughly HK$433,000.
• Per-member view: HK$343,242 times a 2.5% annual gap equals about HK$8,581 a year — 1.7 months of contributions. Every year of chasing is a silent withdrawal.

Thematic investing charges you twice: first, the theme itself underperforms the market (5pp a year behind the S&P 500 over a decade); second, your own timing loses again (the average dollar lost 6pp a year more over three years). Double taxation, compounded over thirty years.

The way out

First, recognise the theme’s disguise inside MPF. MPF cannot buy ETFs, but themes change costume: Korea equity funds, Asia equity funds (with AI-semiconductor exposure), and soon, gold-ETF constituent funds. Any number that reads +77% year-to-date should trigger your alarm, not your switching instruction.

Second, use the system against your own nature. Morningstar found that funds on retirement-plan menus show significantly narrower investor gaps — constrained choice and automatic contributions protected returns. MPF’s closed contribution system and forward (unknown-price) dealing, settling T+1/T+2, are exactly that kind of structural guardrail. The DIS Core Accumulation Fund has returned about 7.3% annualised since April 2017 (MPFA chairman’s blog, 2026-08) at an actual average fee of only around 0.77% — it is itself an “anti-theme” bet: you buy no story, and you pay no story tax.

Third, discipline the gold ETF. The gate is open (10% of NAV cap, physical-backed only); the first constituent fund is a matter of time. Gold rose 64% in 2025 and hit a historic high of US$5,589.38 on January 28, 2026 — the temptation to chase the top is the same script as AI in 2021. If you want exposure, treat it as a satellite: set a cap, rebalance on schedule, never add on a rally.

Fourth, remember the unknown price. MPF switches execute at forward prices — the instruction you place after seeing +77.42% will never execute at the price you saw. Chasing inside MPF cannot even buy at the quoted price. That is the system’s final warning to chasers.

Themes change; chasers don’t. Morningstar’s data proves it: the more moving the story, the wider the gap; the longer the holding period, the heavier the punishment. MPF gave you two guardrails — monthly contributions and unknown-price dealing. Don’t dismantle them yourself.