HK$74.7 Billion of MPF Fund Switching: a Record Year of Chasing Winners and Dumping Losers

Hong Kong’s MPF members are switching funds at a record pace. GUM’s third-quarter report, published on 7 October 2026, puts switching volume at about HK$49.8 billion for the first eight months of the year, projecting a full-year figure of HK$74.7 billion — up more than 25% year on year and a five-year high. GUM is candid about the drivers: every MPF scheme is now on the eMPF platform, so switching takes minutes, and double-digit performance gaps across asset classes have convinced members of “the importance of active management”.

Yet the same report’s numbers tell the opposite story. Hong Kong equity funds returned -3.8% over the first three quarters — the worst of any category — and suffered more than HK$10 billion in net outflows. But that very same category returned +8.3% in the third quarter alone — the best-performing category of the quarter. The selling happened near the bottom; the rebound happened right after the selling. That is not active management. That is a behaviour tax being settled in real time.

Through the data: money chases past returns

The direction is perfectly consistent: the worst performers were sold, the hottest performers were bought. Asia equity’s +24.3% was powered by AI infrastructure, memory and semiconductor demand — but GUM itself notes that semiconductor stocks pulled back in Q3 on AI valuation doubts, “dragging the quarter’s return”. Buyers paid for the first nine months’ scorecard and are facing the fourth quarter’s valuation anxiety.

Even GUM’s own chief investment officer, in the same report cycle, poured cold water on it: single markets are expected to trade in narrow ranges, and globally diversified equity funds should have the edge in this “rotation market”. The house reports the switching as a headline, then advises against doing it.

The timing irony is sharper still. MPF Ratings’ September survey, also released 7 October: -1.47% for the month, about HK$5,060 lost per member — the third losing month of the year. Year-to-date members are still up HK$20,161 per head — but that is the buy-and-hold members’ money, not the market-timers’.

The compounding toll: a 1.2 percentage-point behaviour tax

Morningstar’s Mind the Gap 2026 (ten years to end-2025, nearly 23,000 funds) measured a brutal constant: funds reported 9.9% a year; investors actually earned 8.7% — a 1.2 percentage-point annual gap, entirely the cost of buying high and selling low. Hong Kong’s live version aired in September, when BCT disclosed that members in the same scheme ranged from +32% to -19% in the first half — a gap of over 50 percentage points — with the only difference being 110 fund switches in six months.

Scale 1.2pp to Hong Kong’s system: with about HK$1.7 trillion in total assets, 1.2% is roughly HK$20.4 billion a year left on the table; at the average balance of HK$343,242, that is about HK$4,119 per member per year — roughly one month’s contribution, paid purely to “gut feeling”.

The monthly-contribution sandbox spreads the toll across 30 years. Contributing HK$5,000 a month at an assumed 7% gross (illustrative):

• 7.3% (DIS Core Accumulation Fund, annualised since 2017) → about HK$6.47m over 30 years
• 6.1% (7.3% minus the 1.2pp behaviour tax) → about HK$5.10m
• Gap: about HK$1.36m — nearly 4 times the average balance, equal to 22 years of contributions

The gap is about HK$57,000 at 10 years and HK$360,000 at 20 years — the behaviour tax, like fees, is compounding’s friend; it just stands on the other side.

The way out: turn “doing nothing” into discipline

1. Replace manual rebalancing with automatic rebalancing. GUM’s August report noted the DIS Core Accumulation Fund “continues to be sought after” — its members make no choices, so they pay no behaviour tax. The statutory reference portfolio rebalances quarterly; its 60% global equity / ~40% bond split never drifts because you read a headline.
2. Run every switch through “stop, think”. The MPFA chairman said in late 2025 that “timing the market easily leads to buying high and selling low”, and promised in-platform prompts against frequent switching. Until the prompt arrives, be your own prompt: write down the reason and date of each switch, and review it 30 days later — the member who switched 110 times will not remember the reason for switch number 47.
3. Split large switches; avoid rate-decision weeks. MPF trades on forward pricing (T+1/T+2) — the execution price is only known in the future. Q4 still has two FOMC meetings (27–28 October, 8–9 December); splitting execution at least halves the variance of “blind buying and selling”.
4. Treat HK$74.7 billion as a contrarian indicator. When the whole system is switching at a record pace, that is usually when chasing and dumping are most crowded. GUM’s own CIO says diversified equity funds have the edge in narrow ranges — which is exactly what the DIS reference portfolio was designed for.

eMPF cutting switching costs to a few minutes is a good thing — provided it is execution friction that got cheaper, not thinking. The HK$74.7 billion record proves the former; the missed +8.3% Hong Kong equity rebound proves the latter has not caught up.