Forty-six days to launch. Phase One of MPF full portability begins on 21 December 2026, covering about one million employees who joined the workforce or changed jobs on or after 1 May 2025. Buried in the Mandatory Provident Fund Schemes Authority (MPFA) rules is a sentence that deserves more attention than it has received: employer mandatory contributions and employee mandatory contributions may be transferred together or separately, to the same scheme or to different schemes. From 21 December, you will not face a single move-or-stay decision. You will face two decisions that can be taken apart.
This is not an argument for moving for the sake of moving. MPFA chairwoman Ayesha Macpherson Lau’s warning stands: MPF is a long-term investment, and members should not transfer simply for the sake of transfer. This article does something different. It takes the split-transfer tool apart, shows what problems it genuinely solves, and names the new problems it creates.
Reframing: two accounts, one balance sheet
Start by correcting an intuitive error. Splitting a transfer is not, by itself, diversification. Two personal accounts still add up to one person’s retirement savings. Risk is measured on the combined portfolio, not account by account.
The rules, taken from the MPFA’s public transfer pages, are as follows:
- Eligibility: your current employment must have begun on or after 1 May 2025. Employees who started before that date and stayed in the same job must wait for Phase Two, for which the government plans to submit an amendment bill in mid-2027.
- Separate annual quotas: the employer portion and the employee portion each carry their own once-per-calendar-year transfer quota. Moving the employer portion does not consume the employee portion’s quota.
- Whole-portion transfers only: when you transfer a portion, you must transfer the whole of that portion. You cannot move half of the employer portion and leave the other half behind.
- New contributions do not follow: a transfer moves accrued benefits already built up. Employers will keep contributing through their own scheme, and new contributions will keep flowing into the contribution account under that scheme. A split transfer is a snapshot, not a permanent redirection.
- Voluntary contributions are a separate matter: whether an employer’s voluntary contributions can move depends on the governing rules of the original scheme. Tax-deductible voluntary contributions follow a different, anytime-transfer route and should not be confused with mandatory contributions.
Only once those five points are clear does splitting make sense. Without them, you merely replace confusion in one account with confusion in two.
The sandbox: a HK$30,000 salary, HK$60,000 split in two
Take a member earning HK$30,000 a month. Mandatory contributions total HK$3,000 a month: HK$1,500 from the employer and HK$1,500 from the employee. From mid-2025 to the end of 2026, accrued benefits reach roughly HK$60,000, about HK$30,000 from each side. These are rounded illustrative figures, before investment returns and fees. Three ways of splitting them compare as follows:
| Configuration | Employer portion, HK$30,000 | Employee portion, HK$30,000 | Combined equity weight | Annual fee difference* |
|---|---|---|---|---|
| Move both to a low-fee scheme | Low-fee Default Investment Strategy-type portfolio | Same low-fee portfolio | About 60% | About HK$354 saved a year |
| Split by role: growth and anchor | 100% North American Equity Fund | Default Investment Strategy-type portfolio (about 60% equities) | 80% | Depends on the fees on both sides |
| Move only one half | Move to a low-fee scheme | Stay in the original scheme | Depends on the original portfolio | About HK$177 (on the moved half only) |
*The fee difference uses the market average of 1.36% against the Default Investment Strategy’s actual average of about 0.77%, a gap of 0.59 percentage points. On HK$60,000 that is about HK$354 a year; on HK$30,000, about HK$177 a year. Fee data are from the mpf.hk fee-history database, as compiled on 22 October 2026.
The arithmetic in the second row is worth pausing on: HK$30,000 at 100% plus HK$30,000 at 60%, divided by HK$60,000, gives a combined equity weight of 80% — not 100%, and not 60%. That is the real use of a split transfer. Two accounts can be combined to produce a total weight that neither account provides on its own: a growth sleeve in one account, an anchor in the other. But the combined weight must be set first, and each account’s holdings derived from it. Choosing two sets of funds first and discovering the combined weight afterwards is the most common mistake this new freedom invites.
The fee maths also needs translating into human terms. HK$354 a year sounds trivial, but a fee gap is charged every year. On an illustrative basis of HK$5,000 contributed monthly at a gross return of 7%, a 0.59 percentage point fee gap amounts to about HK$26,000 over 10 years, HK$160,000 over 20 years and HK$559,000 over 30 years. That illustration assumes the gap persists unchanged for 30 years; real fees will change. It is a statement of direction and scale, not a forecast.
Forward-pricing protection: split the move, split the transit risk
Splitting carries one underrated benefit and one underrated cost, and they are two sides of the same coin.
The benefit is that transit risk can be staged. A transfer is technically a redemption followed by a subscription, leaving the money out of the market for about one to two weeks. MPF forward pricing (T+1/T+2 net asset values) means you cannot lock in the exit and entry prices. Move only the employer portion and the other half stays invested, so the combined portfolio’s amount in transit is halved: a 2% swing on HK$30,000 is about HK$600, against about HK$1,200 on the full HK$60,000. Splitting one large move into two smaller ones reduces the single-event shock.
The cost is the number of moves itself. Two moves mean two transit windows, two rounds of forward pricing and two sets of administration to follow up. A single 2% swing of about HK$1,200 is already 3.4 times the first year’s fee saving of HK$354 on a HK$60,000 balance. A swing is a risk, not a certain loss — its expected value is close to zero — but the comparison makes the point: splitting repeatedly to chase a few hundred dollars of annual fee saving gets the priorities backwards. A split transfer should be a one-off allocation decision, not an annual habit.
Two further practical costs must be stated plainly. First, two accounts mean two fee schedules and two statements, and there is no automatic rebalancing across schemes. Your combined weight will drift with markets, and rebalancing must be done manually in each scheme. Second, the Default Investment Strategy’s automatic de-risking applies only inside holdings that are in that strategy. Holding a Core Accumulation Fund and an Age 65 Plus Fund independently is not the same as holding the full strategy, and birthday de-risking will not operate across accounts.
Where splitting genuinely earns its keep
First, asymmetric fees. If both portions sit in the same expensive scheme, moving or not moving is a single calculation. But if your aim is to move the most expensive half first, splitting lets you act without waiting for a decision on the other half. Move one half now; the other half’s quota is not wasted by waiting.
Second, a deliberate division of risk. A younger member wanting more equity exposure, or an older member wanting less, can set a combined target weight first and then use a split to build it. The rule is target first, account holdings second — never the reverse.
Third, quota management. Because the two portions’ annual quotas are counted separately, you can move one portion in the 21–31 December 2026 window and keep the other portion’s 2026 quota in reserve, or hold it for a full review when the new calendar year’s quota opens on 1 January 2027. A quota is an option. Spending it early is giving up the option early.
None of these three cases includes chasing transfer bonuses. Bonus units typically must be held until they are credited, and leaving early can forfeit them. A bonus is a one-off amount; a fee gap recurs every year. The two cannot be compared without looking at the time horizon.
This week’s checklist
- Check eligibility and both balances. Confirm that your current employment began on or after 1 May 2025, then look up the employer and employee portions separately on the eMPF platform, with their fund expense ratios. Do not rely on the combined total alone.
- Write down the combined weight first. Before choosing any scheme, write down the combined equity weight you want — say 60% or 80% — and then work backwards to what each account should hold.
- Make one allocation decision, not a routine. If you split, write down where both portions are going, what each costs and what the combined weight will be, in a single decision. Do not turn transferring into a twice-yearly routine for a few hundred dollars of fee saving.
A split transfer is not simply extra freedom to move house. It is a larger board on which to place two separate pieces. The larger the board, the more it rewards deciding the whole position before moving the first piece.
- MPFA “
- Transferring MPF”
- public page, checked 29 October 2026 (the 21 December date is footnoted as subject to completion of Legislative Council scrutiny of the relevant legal notice)
- MPFA chairwoman’
- s blog press release, 25 October 2026 (about one million people in Phase One
- 2026 window 21–31 December
- 2027 transfers from 1 January)
- MPFA consultation conclusions on full portability (employer and employee portions handled separately, with annual quotas counted separately)
- mpf.hk fee-history database, 22 October 2026 (market average 1.36%
- Default Investment Strategy actual average about 0.77%
- BEA 1.0738%
- Manulife 1.4188%). Illustrations assume a gross return of 7% compounded monthly
- actual returns and fees are as stated in fund fact sheets and offering documents.
Related articles
View all
Behind the HK$1.67 Trillion Record: The Chairwoman’s 7.3% Is Both a Report Card and a Fee X-Ray
On 30 August 2026, MPFA Chairwoman Lau did something unusual: she wrote the...

MPF fee reform ahead: cap, consolidation and digitalisation
This article is a rewrite of a report from August 2013. MPF fee reform was...

Five years of the MPF’s Employee Choice Arrangement: Japanese equity funds up 81% — a tidy portfolio could already be up 80%
The MPF’s Employee Choice Arrangement — the...