Short answer: On 24 July 2026, China introduced a 20% individual income tax on offshore trusts connected to Chinese tax residents. The tax bites at several points in a trust’s life, and there is a short window to settle historic amounts. For families and advisers in Singapore, the pressing problem is not whether to move money somewhere else. It is finding cash to pay a tax bill on assets that were never meant to be sold.
What actually changed
China’s Ministry of Finance and State Taxation Administration jointly issued Announcement No. 21 of 2026, with a procedural companion from the tax administration, Announcement No. 15 of 2026. Together they set out, for the first time, how China’s individual income tax applies to trusts set up outside China.
Chinese tax residents have always been taxed on their worldwide income in principle. What was missing was a rulebook for trusts. That gap is now closed.
The rule in one line: an offshore trust is treated as a see-through container for tax purposes. What the trust earns is traced back to the person who put the assets in.
Where the 20% is charged
Based on the published analysis of Announcement No. 21 by international law firms, tax arises at four moments:
- When assets go in. Transferring property into the trust is treated as if it had been sold. Tax is charged on the gain, meaning market value less original cost and reasonable expenses, not on the full value of the asset. The cost base is then reset to market value.
- Every year the trust operates. For trusts funded by residents, income is attributed to the contributor annually, whether or not a single dollar is paid out.
- When money is distributed. Distributions to residents from trusts funded by non-residents are taxable, and certain benefits such as year-end outstanding loans, guarantees or rent-free use of trust property can be treated as distributions even when no cash moves.
- When the trust ends. Liquidation gains, or the market value of property received, are taxed.
The clock
There is a 90-day statutory filing and payment window for historic liabilities. For resident contributors it covers property put into trusts between 1 January 2023 and 31 December 2025. For non-resident contributors it runs from 1 January 2023 to 24 July 2026. Pre-2026 income of resident-funded trusts is caught regardless of whether it was distributed.
The window runs to late October 2026. Reported dates vary by a day depending on the source, so the exact deadline should be confirmed with a qualified adviser rather than taken from press coverage.
One point is widely misreported. Morgan Lewis notes that the window is not framed as a general amnesty. Paying on time avoids late-payment surcharges. It is not a blanket waiver of penalties.
Why a second passport does not fix this
This is the part that has surprised people most, and it is the simplest to explain.
The tax follows the person, not the address of the structure. Holding a foreign passport or permanent residence elsewhere does not, on its own, end Chinese tax residency if the individual’s principal economic interests remain in China.
Loh Kia Meng of Dentons Rodyk put it plainly in comments to The Straits Times, describing a foreign passport as not, by itself, a tax plan. Ryan Lin of Bayfront Law made the same point from the other direction, telling the Business Times that residence rather than location is what triggers the charge.
So a fully compliant Singapore trust, administered by a Singapore trustee, holding assets custodied in Singapore, can still generate a Chinese tax liability. Nothing about it is defective. It simply was never the thing that determined the tax outcome.
Beijing also has the data. China has received automatic account information under the Common Reporting Standard since 2018, including from Singapore, and can cross-check it against domestic filings. Local tax bureaus in Shanghai, Shenzhen and Jiangsu had reportedly begun examining offshore trusts and applying 20% charges in selected cases before the national rules appeared.
The provision almost nobody is pricing
Most coverage stops at the headline number. The more consequential detail sits in the annual attribution rules.
According to Morgan Lewis’s reading of Article 4, income inside a resident-funded trust must be sorted into two boxes: property transfer income, and interest, dividends and bonus income. Losses on property transfers cannot be carried forward. Offsetting one box against the other is not permitted. Trust management fees, legal fees and investment advisory fees are not deductible.
Read that again slowly, because the arithmetic matters more than the rate.
A worked example
Illustrative only. Tigris interpretation of the rules as reported by law firm analysis. Individual outcomes depend on professional advice.
Imagine a trust that sells two holdings in the same year. One is sold at a gain of US$5 million. The other is sold at a loss of US$5 million. Economically, the family is flat. Nothing was made.
Under the framework as described, the gain is taxable at 20%, so roughly US$1 million becomes payable. The loss is stranded. It cannot be carried forward and it cannot be netted against the other category. Layer on non-deductible management and advisory fees, and the family has a real cash tax bill in a year with zero economic return.
The stated rate is 20%. The effective rate on economic return, in a volatile or high-turnover portfolio, can be considerably higher. In a flat year it is mathematically infinite, because you are dividing a real tax bill by a return of nothing.
This is the point a purely legal reading misses. The rules do not merely tax wealth. They tax a particular style of investing.
The real squeeze is cash, not location
The consensus question in the market has been whether money will leave Singapore. On the evidence so far, that is the wrong question.
Offshore trusts have typically been used to hold exactly the things that are hardest to sell in a hurry. Pre-IPO stakes. Concentrated founder shareholdings. Property. Interests in private businesses. These were placed in trust for succession and asset protection, not for trading.
A tax bill, by contrast, must be settled in cash, on a fixed date. Lawyers in Singapore and Hong Kong have described clients preparing to liquidate parts of their portfolios to fund it.
Tigris interpretation: the assets that get sold first will not be the assets that created the liability. They will be the liquid ones, because those are the only ones that can be sold inside 90 days without a discount. Listed equities. Bonds. Money market holdings. The illiquid positions that generated the taxable gain on contribution stay exactly where they are.
The net effect is a portfolio that emerges from this process less liquid than it went in, with a thinner cash buffer and a higher concentration in assets that cannot be sold quickly. That is a portfolio construction problem, and it will outlast the filing deadline by years.
What Singapore’s own numbers actually show
It is worth separating the noise from the data.
At the second quarter economic survey briefing on 11 August 2026, Ministry of Trade and Industry permanent secretary Beh Swan Gin said no reports had come in from the wealth management sector indicating an impact, while adding that the sector was being watched closely. Edward Robinson, deputy managing director at the Monetary Authority of Singapore, said fund flows remained strong and no unusual volatility had been detected in recent weeks.
Bank of Singapore has said it has seen no significant outflows. UOB described the development as recent and said it was assessing. DBS chief executive Tan Su Shan, speaking at the bank’s earnings briefing on 6 August, framed the rules as reinforcing a trend the bank had already flagged for a decade, namely the growing importance of onshore wealth centres closer to clients.
The structural backdrop is also strong. The MAS Singapore Asset Management Survey 2025 recorded assets under management of S$6.7 trillion, up 10% year on year, with net inflows growing 29%. Around 76% of that money was sourced from outside Singapore and 88% was invested globally, across 1,320 licensed and registered managers and 1,406 Variable Capital Companies.
Singapore’s wealth franchise is not built on a single client segment. That is the reason a China-specific tax change does not read as an existential threat to it.
What this means for investors and advisers
| What changes | Why it matters | Practical implication |
|---|---|---|
| Tax follows residence, not structure | Jurisdiction of the trust is no longer the deciding factor | Structuring alone stops being a strategy. Substance and residence facts become the primary variables. |
| Income attributed annually, distributed or not | Tax falls due before cash arrives | Distribution policy and liquidity planning need to be designed together, not sequentially. |
| Losses stranded, fees non-deductible | Effective tax on economic return can exceed the headline 20% | High-turnover and high-fee strategies carry a structural drag that lower-turnover strategies do not. |
| 90-day historic settlement window | Cash required on a fixed date against illiquid holdings | Liquidity sleeves are the first thing sold. Rebuilding them should be planned in advance, not after. |
| Trustee and intermediary obligations | Record-keeping and reporting burden rises | Administration quality and documentation discipline become selection criteria for service providers. |
Tigris interpretation and forward view: we expect demand from China-linked clients to shift towards mandates with three characteristics. Lower portfolio turnover, so fewer taxable realisation events. More visible and predictable income, so tax liabilities can be forecast rather than discovered. Transparent, fee-efficient structures, given that fees no longer shelter anything. This is a forecast, not a stated policy position of any regulator.
There is a second-order consequence for Singapore itself. If structuring is worth less, then the part of the value chain that survives is genuine investment management. Licensed managers, real discretionary mandates, independent fund administration and regulated vehicles such as the VCC. That is precisely the layer Singapore has spent a decade building. Our view is that the city does not lose from this shift. The structuring intermediaries lose. The managers gain.
What could still go wrong
Honest analysis requires naming the gaps.
- Enforcement against trustees outside China is undefined. The announcements impose duties on offshore trustees but do not specify a mechanism for enforcing them against trustees with no presence or assets in China.
- One lookback period is unclear. Practitioners have flagged that the rules require residents receiving distributions from non-resident-funded trusts to file within the 90-day period without expressly stating a start date for that lookback.
- Double taxation is a live risk. A foreign tax credit is available, but it is capped on a country-by-country basis and depends on the legal character of the foreign tax and on documentation. Timing mismatches between jurisdictions can produce tax in two places on the same economics.
- Scope may widen. Caixin has reported that some local tax bureaus began levying personal income tax on returns from insurance policies within two weeks of the announcement. This has not been confirmed by official sources and should be treated as unverified.
- Behaviour may lag. Singapore lawyers currently assess the likelihood of clients unwinding trusts as low. That assessment is based on the first weeks of the new regime. It is not a settled conclusion.
The question worth holding
For two decades, a great deal of cross-border wealth planning rested on a quiet assumption: that once an asset sat inside the right structure in the right place, the question of where the owner lived became secondary.
Announcement No. 21 does not attack trusts. It does not invalidate them or deny their effect under the law that governs them. It does something more durable. It moves the trigger from the asset to the person, and it does so with data good enough to make the move stick.
The families most exposed are not the ones with the most aggressive structures. They are the ones whose portfolios were built on the assumption that they would never need to raise cash quickly. That assumption has just been tested, and the test carries a date.
The open question is whether the wider industry treats this as a compliance exercise to be completed by late October, or as a signal that portfolio design, not structural design, is where the real work now sits.
Frequently asked questions
What is China’s offshore trust tax?
It is a 20% individual income tax that applies to Chinese tax residents in connection with trusts established outside China. It was introduced by Announcement No. 21 of 2026, issued jointly by China’s Ministry of Finance and State Taxation Administration on 24 July 2026, with procedural rules in State Taxation Administration Announcement No. 15 of 2026.
Does it apply to trusts set up before 2023?
The 90-day historic settlement window covers property contributed from 1 January 2023 onwards, and pre-2026 income of resident-funded trusts. Older structures are not automatically outside the framework going forward, and treatment depends on the facts. This is a question for a qualified tax adviser, not for general commentary.
Will a foreign passport or permanent residence protect me?
Not by itself. Acquiring foreign nationality or permanent residency does not end Chinese tax residency if the individual’s principal economic interests remain derived from China.
Is money leaving Singapore because of this?
There is no evidence of that so far. As of the 11 August 2026 economic survey briefing, MAS reported no unusual volatility in flows and the Ministry of Trade and Industry said it had received no reports of impact from the wealth management sector. Bank of Singapore has reported no significant outflows.
What should affected families be doing first?
Two things in parallel. Establish the factual position, meaning who contributed what, when, and at what cost base, and who is a Chinese tax resident on the current tests. Then model the cash requirement against the liquidity actually available in the portfolio. The second exercise is often more revealing than the first.
Does this change how a portfolio should be built?
In our view, yes. Where realisation events, fees and losses all receive unfavourable treatment, the case for lower-turnover strategies with predictable income improves relative to high-turnover strategies with similar gross returns. That is an investment design question, and it sits alongside, not inside, the legal advice.
Sources
Ministry of Finance and State Taxation Administration of the People’s Republic of China, Announcement No. 21 of 2026 and State Taxation Administration Announcement No. 15 of 2026, as analysed by Morgan Lewis (28 July 2026); STEP; The Straits Times, “Singapore bankers, China-linked clients gear up for Beijing’s scrutiny of offshore trusts” (14 August 2026); The Business Times, via IFC Review (11 August 2026); CNBC (5 August 2026); Caixin Global (25 July 2026); MAS Singapore Asset Management Survey 2025.
More Tigris analysis on Singapore structures, regulation and cross-border capital is available on Tigris Insights.
This material is for informational purposes only and does not constitute an offer or solicitation to buy or sell any investment product, nor does it constitute tax or legal advice. Past performance is not indicative of future results. Investment products are available only to accredited and institutional investors as defined under the Securities and Futures Act 2001 of Singapore. Tigris Asset Management Pte Ltd holds Capital Markets Services Licence CMS101520 issued by the Monetary Authority of Singapore.
