Hong Kong publishes no household balance sheet — what it does publish is HK$20.6 trillion of deposits and HK$1.67 trillion of MPF
No Hong Kong authority compiles a household financial balance sheet, so there is no official answer to what share of household assets sits in deposits. The measured pieces are customer deposits with authorized institutions, MPF net assets and long term insurance assets — three different universes that cannot be added.
What share of Hong Kong household assets sits in deposits?
There is no official answer, and that is the finding rather than a gap in this article.
Hong Kong compiles no flow-of-funds accounts. The Census and Statistics Department’s national accounts cover GDP, GNI and expenditure components; there is no financial account and no institutional-sector balance sheet. The HKMA’s data and statistics portal has no household asset category. So the denominator that the question requires — total household financial assets — is not produced by anyone.
What exists instead is a set of separately measured pieces.
The pieces Hong Kong does measure
The pieces Hong Kong does measure
Customer deposits with AIs (HK$ m)
Shared linear scale from zero. Missing values are not plotted or treated as zero. Selected entries from the table below. Its dates, units, definitions and limitations still apply.
| End of period | Customer deposits with AIs (HK$ m) | of which HKD | of which foreign currency | MPF net assets (HK$ m) |
|---|---|---|---|---|
| December 2015 | 10,749,749 | 5,312,403 | 5,437,346 | 591,320 |
| December 2019 | 13,771,586 | 6,884,143 | 6,887,444 | 969,455 |
| December 2024 | 17,372,718 | 7,839,207 | 9,533,510 | 1,291,179 |
| December 2025 | 19,431,264 | 8,136,778 | 11,294,485 | 1,553,364 |
| July 2026 | 20,562,027 | 8,641,180 | 11,920,847 | — |
| June 2026 | — | — | — | 1,665,466 |
These two columns are not two parts of one total and must not be added. Customer deposits cover every depositor with an authorized institution — companies, non-residents and financial firms as well as individuals — and the HKMA publishes no breakdown by type of depositor: its three customer-deposit datasets split by currency, by deposit type (demand, savings, time) and by institution type (licensed banks, restricted licence banks, deposit-taking companies), never by sector. Part of the MPF total is itself invested in Hong Kong bank deposits, so the two figures overlap by an amount that is not published. The dates also differ: the deposit series runs to July 2026 and the MPF series to June 2026.
Two things the deposit series does show
First, the currency mix has moved. Hong Kong dollar and foreign currency deposits were within half a percentage point of each other at the end of 2015 and again at the end of 2019. By the end of 2025 foreign currency was 58.13% of the total, and in July 2026 it was 57.98%. The HKMA publishes the split, so this is observable even though the depositor behind it is not identified.
Second, the MPF total has grown faster than deposits. MPF net assets multiplied by 2.82 between the end of 2015 and June 2026; customer deposits multiplied by 1.91 over a slightly longer window to July 2026. That is what a maturing mandatory scheme looks like — contributions accumulate on a schedule — and it is not evidence about how households allocate money they can choose to move.
Long term insurance is the third measured piece. The Insurance Authority reported total assets under long term business of HK$5,398 billion at the end of 2025, of which net assets were HK$744.3 billion. At the end of 2024 the detailed return gives total assets of HK$4,695,723 million against insurance liabilities of HK$3,651,053 million. These are insurers’ balance sheets, not household claims, and they are not comparable with the 2015 and 2019 years because the risk-based capital regime changed the basis. They are mentioned here and left out of the table for that reason.
The three Asian markets that do publish the answer
| Market | Reference | Currency & deposits | Debt securities | Equity & funds | Insurance & pension | Other | Sector definition |
|---|---|---|---|---|---|---|---|
| Japan | end-March 2026 | 47.22% | 1.52% | 23.63% | 24.48% | 3.15% | households incl. unincorporated business |
| Taiwan | end-2024 | 34.78% | 0.25% | 26.10% | 26.12% | 12.74% | households & non-profit institutions |
| Singapore | 2026 Q2 | 32.36% | — | 20.19% | 47.45% | — | households |
| Korea | 2025 Q2 | 45.23% | 3.25% | 22.52% | 28.18% | 0.83% | households & NPISH |
| United States | end-March 2026 | 10.7% | 4.3% | 56.7% | 25.7% | 2.5% | households & nonprofits (BOJ basis) |
| Euro area | end-March 2026 | 31.1% | 2.8% | 37.6% | 26.5% | 2.0% | households & NPISH (BOJ basis) |
Hong Kong has no row in this table. That is a coverage gap, not a zero, and no proxy has been substituted for it. For Japan, the United States and the euro area the equity-and-funds column adds the "equity" and "investment trusts" lines of the same Bank of Japan comparison table — a sum of published items from one table and one vintage, not an estimate. Sector definitions and reference dates differ across the rows, so these are six separate readings rather than a ranking on one day.
Hong Kong's position in such a table, if it existed, would be pulled in two directions that are hard to net out by eye: the MPF places a mandatory block of assets outside deposits, as CPF does in Singapore, while the currency board arrangement and the size of the offshore banking business put an unusual amount of non-household money into the same deposit series. That is one reason a sector breakdown would be worth publishing, and it is interpretation rather than measurement.
What to check when reading these numbers
Check whether the figure has a denominator at all. “Deposits reached a record HK$20.6 trillion” is a level. “Households hold X% of their assets in deposits” is a ratio, and in Hong Kong the bottom half of that ratio is not published by anyone.
Check who the depositor is. A deposit total that includes corporate treasuries, insurers — who held HK$144.6 billion with authorized institutions at the end of 2024 — and non-resident money is not a household statistic, however often it is used as one.
Check stock against flow. Thailand, the Philippines and Indonesia publish flow-of-funds accounts that contain only transactions. Adding to deposits in a year and holding deposits are different measurements.
And do not stack the measured pieces into a total. Deposits, MPF assets and insurer assets overlap in amounts that are not published, so a sum of them is not a household balance sheet.
Questions & answers
What share of Hong Kong household assets is held in cash and deposits?
There is no official figure. Hong Kong compiles no flow-of-funds accounts and no household sector balance sheet, so the denominator needed for that share does not exist in any published Hong Kong statistic. The HKMA measures deposits, the MPFA measures MPF assets and the Insurance Authority measures insurer assets, but none of them measures total household financial assets.
How much is held in bank deposits in Hong Kong?
Customer deposits with authorized institutions were HK$20,562,027 million in July 2026, of which HK$8,641,180 million were Hong Kong dollar deposits and HK$11,920,847 million foreign currency. That total covers companies and non-residents as well as households; the HKMA publishes no split by type of depositor.
How large is the MPF system?
HK$1,665,466 million of net assets at the end of June 2026, across 4.97 million scheme members. It was HK$591,320 million at the end of 2015 and HK$1,291,179 million at the end of 2024.
Which Asian markets do publish this?
Japan, Taiwan and Singapore. Their currency-and-deposits shares of household financial assets are 47.22% at end-March 2026, 34.78% at end-2024 and 32.36% in 2026 Q2 respectively. Hong Kong, Malaysia, Thailand, the Philippines, Indonesia and Vietnam do not publish a comparable figure.