Skip to main content

The Fraud You Can See, the Numbers Nobody Measures

Public welfare isn't the problem; the problem is statistics that dilute it

@Mythogen Engine · Oct 23, 2026

“Snatching Rice from the Beggar's Bowl”, Part 3 | Previous: “Global Top, Local Bottom” | Next: “Those Who Leave and Those Who Arrive”

Not resenting the rich, not hating the poor: struggle is not competition, and in the end it reaches every floor.

There is no box for that on the form​

At the Housing Department counter, Ah Yan hands over her payslip: HK$22,000 a month. The clerk glances at it. “Above the income limit. Not eligible.”

She asks: “What about people who earn cash somewhere else and declare zero income?” The clerk flips through the form. “There is no box for that on the form.”

She asks again: “What about assets outside Hong Kong?” The clerk says: “There's a box. You fill it in yourself.”

Finally she asks: “How long is the queue?” “Five years and more, on average. But you don't need to queue. You're not eligible.”

The dialogue is fictional, but every line has a source — the income limit, the 5.4 years, offshore assets resting on self-declaration — and each is set out below. Ah Yan is the office administrator from Part 2 of this series: on the global ranking she sits in the top quarter; on the Housing Department's form she cannot even join the queue.

The core argument​

Public housing and social security worked in Hong Kong for decades; today they seem to have failed across the board for no stated reason: working people are ineligible, those who qualify wait more than five years, and rents are at record highs.

Public welfare itself is not at fault. Two things are: the means test can measure only payslips, not cash or offshore assets; and in response to abuse it cannot measure, the system tightens on the people it can. People feel that welfare cheats are everywhere, yet no one can produce a number — not because there are none, but because no one counts. Why no one counts is the subject of Part 4.

1. Decades that worked​

The benefit of public housing reaches beyond the people living in it: it holds down housing costs for the whole of society.

  • Scale: in 2024, 29.4% of the population lived in public rental housing and 15.3% in subsidised sale flats, together nearly half (Housing Bureau).
  • A social wage: Manuel Castells and colleagues, The Shek Kip Mei Syndrome (1990): public housing lowered housing costs, so wages did not need to be high for people to get by, and export industry stayed competitive.
  • A ceiling on private prices: when half the population can live in public housing, private housing cannot raise prices without limit. This is an inference, not a law; Singapore's HDB flats are another example worth comparing.
  • A ladder onto the property market: the Home Ownership Scheme (HOS) meant young families did not have to compete with investors for private flats from the start.
  • Room in the treasury: with living costs held down by structure, the government did not have to hand out cash to subsidise every household.

The counter-example: after the 2002 “nine measures” of then housing secretary Michael Suen, HOS construction stopped for nine years (HKU interpreting glossary), and the ladder and the ceiling disappeared together.

The first newly built HOS flats were not presold again until late 2014, 2,160 units completed in 2016/17 (Transport and Housing Bureau, 2015). The ladder was broken for more than a decade.

Taiwan offers another comparison: government-built social housing totals about 127,000 units, under 1.5% of the housing stock, and protection relies on a home-ownership rate above 80% (CNA, Liberty Times, Economic Daily News). Hong Kong once did what Taiwan never did, then took it apart itself over the past two decades.

2. Symptoms of a system-wide failure​

Today's symptoms cannot be explained by “insufficient supply” alone: the door is narrowing, the queue is lengthening and rents are at record highs, all at once.

  • A median-wage worker is ineligible: from April 2026, the public housing income limit is HK$13,230 for one person and HK$20,680 for two; Hong Kong's median monthly income is HK$23,600 (Sing Tao).
  • Those who qualify must wait: general applicants wait 5.4 years on average; at end-March 2025 there were about 202,000 applications, about 116,000 general and about 86,000 under the quota and points system (Housing Bureau).
  • The ladder has broken before: HOS construction stopped for nine years (section 1).
  • Rents at record highs: the Rating and Valuation Department's private domestic rental index stood at 205.8 in June 2026, an all-time high (Sing Tao); the Centa-City rental index rose 3.87% in the first half of this year, the biggest half-year rise in nearly nine years (World Journal).
  • Queuing before seven years' residence: in June 2019, of 147,900 general applications, about 14,000 were frozen because they “did not meet the residence requirement” — about one in ten waiting to complete their years of residence (Housing Authority paper SHC 62/2019). Once the years are met, these applications cannot be told apart from any others.

Official explanations mostly stop at supply; where demand comes from, and through which entry routes, is not broken down in official figures. The next two sections cover the means tests for welfare and public housing; the blind spots in entry routes and statistics are covered in Part 4.

3. CSSA: the one-year residence rule​

Eligibility for CSSA, Hong Kong's main welfare payment, depends only on residence status. In 2013 the Court of Final Appeal ruled in the Kong Yunming case that the seven-year residence requirement for CSSA was unconstitutional, and the government restored a one-year rule: applicants must have been Hong Kong residents for at least one year and lived in Hong Kong for one year; under-18s are exempt (Social Welfare Department). After the ruling, applications from people resident for less than seven years totalled 6,754 approvals by December 2016, involving about HK$2.5 billion over three years (SCMP). This is one of the few published figures, and it is already a decade old; after seven years, these applicants are also counted as “local”. As for holders of talent visas, the Immigration Department says holders of the Top Talent Pass Scheme (TTPS) are “not subject to any other condition of stay”; whether they can apply for CSSA after one year is not made clear in Social Welfare Department documents (see Part 4).

There are figures on CSSA fraud too, but only for the part that is caught: from 2010/11 to 2014/15, substantiated fraud cases ran between 674 and 1,117 a year, and the conviction rate among those prosecuted in 2014/15 was about 97% (LegCo Research Office, 2016). CSSA cases in the same period numbered in the hundreds of thousands; in 2023/24 about 240,000 cases were handled (SWD budget controlling officer's report). Recent figures are similar: in 2021/22 the department received 1,143 public reports and identified 550 cases itself, but only 323 were substantiated, under 0.2% of about 220,000 cases (LegCo reply, 2022); in a 2025 reply the government said substantiated cases in recent years were about 0.1% to 0.3% of the total. The same reply also admitted that the department does not keep a breakdown of cases involving assets or income outside Hong Kong (LegCo reply, 2025). What is caught runs to a few hundred cases; how many are not caught, no one knows; and the cross-border kind that draws the most suspicion is not counted separately even among the cases already caught.

4. The means test screens out the wrong people​

The means test can see a payslip but not cash or offshore assets; the result is that workers who honestly report their pay are screened out.

  • Income limits: from April 2026, HK$13,230 for a one-person household and HK$20,680 for two (Sing Tao). Hong Kong's median monthly income is HK$23,600: a median-wage worker is already ineligible.
  • Waiting time: general applicants wait 5.4 years on average (June 2025, Housing Bureau).
  • Residence: at the time of allocation, only half of household members need to have lived in Hong Kong for seven years (Sing Tao application guide).
  • Property: the only rule is that applicants “must not own any domestic property in Hong Kong”.
  • Offshore assets: in 2023, Housing Authority member Leung Man-kwong admitted that offshore and mainland assets rely on tenants' own declarations: “because of administrative and enforcement limits, there is no way to go abroad or to the mainland to ask for the relevant information”, and the system “relies heavily on residents declaring voluntarily” (HK01).
  • Flats recovered for abuse: from July 2022 to June 2025, about 9,000 flats were recovered, equivalent to saving over HK$9 billion in construction costs (Sing Tao). For a breakdown of reasons, only the investigation figures in the next item are available.
  • Investigations by type: of the abuse cases investigated by the Housing Authority in 2024, about 8,400 (52%) concerned occupancy, including flats left empty, occupied by non-Hong Kong residents, or used for non-residential purposes; about 7,700 (48%) concerned income and asset declarations, mainly concealed assets or income (LegCo Research Office, October 2025).

What was caught: 9,000 flats. How many were not caught: nobody has a number.

5. What people feel, and the loopholes​

For each loophole, one must first separate “possible under the rules” from “how many people actually do it”; the first can be verified, the second mostly has no numbers.

LoopholeCheckable sourceRuleScale
Cross-border cash income (parallel trading, working on the mainland) while declaring zero incomeSWD CSSA fraud cases, Housing Authority prosecutionsThe system relies only on declarationsOnly what is caught: about 700 to 1,100 CSSA cases a year (2010/11–2014/15)
Undeclared offshore and mainland assetsHousing Authority meeting recordsVerified: the authority admits it cannot checkOnly investigations: about 7,700 income and asset declaration cases in 2024
Not living in the flat; occupied by non-Hong Kong residentsHousing Authority anti-abuse workVerifiedOnly investigations: about 8,400 occupancy cases in 2024
Children of one or two non-resident parents count toward “half the household has lived here seven years”Housing Authority rulesVerified: children born in Hong Kong with established permanent residency are all deemed to meet the seven-year ruleNo figures (see Part 4)

6. Conclusion: welfare fraud is an enforcement problem​

Public welfare itself is not at fault; welfare fraud is an enforcement problem, not a problem with protection itself, which is consistent with the position in Part 1 of this series. Responding to abuse that cannot be measured by tightening protection punishes honest people: workers whose payslips show everything are screened out, while those who declare zero income may get through.

“No numbers” cuts both ways: this essay cannot prove how large the abuse is, and the government cannot prove that it is small. The conclusion is not “there is a lot of welfare fraud”, but “the people making decisions have not measured”.

To answer the question, only three numbers are needed: the distribution of public-housing household heads by place of birth and years of residence; the latest number of public-housing applications frozen for not meeting the residence requirement (the last published figure was about 14,000 in 2019); and the number of CSSA recipients by years of residence. Residence already has to be declared and checked when applying for public housing and CSSA; the government is not unable to produce these numbers, it has simply never published them. Anyone can request them under the Code on Access to Information; if the reply is “not kept”, that is itself the answer.

One more point is worth noting: the people measured are always the bottom. CSSA recipients must declare every dollar of income, public-housing tenants must declare income and assets year after year, and those in the queue must prove again and again that they are poor enough. As for who collects the land rent, why the treasury has long depended on land sales, and who sets land prices, there are no equally strict numbers. People look downward along the numbers they are given, so the object of suspicion is always someone poorer than themselves. Welfare fraud should be investigated; but spending all the effort of scrutiny on the poorest in public housing and CSSA, while land rent keeps flowing upward, is exactly snatching rice from the beggar's bowl.

The next essay asks: why does no one measure? The answer lies not at the Housing Department counter but further upstream: who comes in, by which route, and how the Census and Statistics Department lets those who leave and those who arrive cancel each other out.

“Snatching Rice from the Beggar's Bowl” series: Prologue | Part 1 “Suffer If You Don't Work?” | Part 2 “Global Top, Local Bottom” | Part 3 “The Fraud You Can See, the Numbers Nobody Measures” | Part 4 “Those Who Leave and Those Who Arrive” | Final part “After the Counting”

Sources​

Most sources are in Chinese; titles are translated.