New Survey on Right of Abode Casts Doubt on Government's "Taxi Method" Survey

Press release

18 October 1999


A new survey casts serious doubt on the reliability of the Government's claim that 1.603 million "additional" Mainlanders are eligible for right of abode on the basis of the Court of Final Appeal's judgments of 29 January 1999. Results of the new survey indicate that the Government has overestimated by more than 1 million the number of "additional" Mainlanders.1

The survey was conducted, on behalf of the Hong Kong Human Rights Monitor, by Pam Baker and Company, Solicitors ("PBC"), on the PBC's clients claiming right of abode in Hong Kong. Between late July 1999 and 14 October 1999, more than 5000 clients were asked to complete a questionnaire and provide supporting documents -- including a marriage or notarial certificate -- evidencing the marital status of their Hong Kong resident parent(s).

The survey showed that among PBC's client base there were 4686 additional Claimants. Of these 4686:

4211 (89.9%) were born in registered marriage but before either parent became a permanent resident.

475 (10.1%) were born either in a traditional marriage that has not been registered or outside of wedlock. (Persons born in traditional unregistered marriages were classified in the Government's survey as "illegitimate". PBC and Human Rights Monitor used the same criteria.)

Accordingly, in our large survey of additional actual right of abode Claimants in Hong Kong, there were 8.87 "legitimate" Claimants for every "illegitimate" Claimant. This gives a ratio of 8.87 : 1 (4211 : 475 ).

This is a similar result to that which the Government arrived at in a survey it conducted in March and April 1999. That survey indicated there were 9.4 additional legitimate Claimants for every illegitimate Claimant; a ratio of 9.4 : 1. This survey was conducted by the Census and Statistics Department using the "direct question method" which involved asking Hong Kong residents how many Mainland children they had born out of registered marriage. The results indicated there were approximately 20,000 additional illegitimate children eligible for right of abode now. The Government's most recent estimate of additional legitimate children eligible for right of abode now is 188,000. This gives a ratio of 9.4 : 1 (188,000 : 20,000).

The results of these two surveys differ dramatically, however, from the highly publicised Government survey that was conducted using the "taxi method". That survey concluded that for every additional legitimate Claimant there were 2.686 illegitimate Claimants; a ratio of 1 : 2.686 (188,000 : 505,000). Some statisticians have suggested that this taxi method ratio is so high it defies common sense.

The taxi method survey involved respondents answering either how many taxi rides they had taken in the past week, or how many illegitimate children they had in the Mainland. The person recording the answer did not know which question the respondent was answering, and so various assumptions were made. It was by using the taxi method that the Government arrived at its overall estimate of 1.603 million additional Claimants within 10 years, more than two-thirds (1.086 million) of whom are said to be illegitimate.

The Government only used the taxi method survey to estimate the number of potential illegitimate Claimants. It used a more regular method of asking Respondents directly the number of legitimate children they have in the Mainland. The new survey does not suggest that the Government's estimate of 188,000 additional legitimate Claimants eligible for right of abode now is an overestimate.

Similarly, the Government's estimate of the ratio of children born to additional legitimate and illegitimate Claimants is not disputed by this survey. The Government estimates that each 1st generation illegitimate Claimant will have had an average 1.15 children born in the Mainland by the time s/he has lived in Hong Kong for 7 years.

Applying the above ratios of each of the 3 surveys, and based on the assumption that the Government's estimated number of 1st generation legitimate Claimants (188,000) is accurate, the surveys provide the following estimates, rounded to the nearest 1000:

Government's taxi method survey
Category 1st Generation 2nd Generation Total
Reg. Marriage 188,0002 329,000 517,000
Unreg. Marriage 505,000 581,000 1,086,000
Total 693,000 910,000 1,603,000

Government's direct question survey
Category 1st Generation 2nd Generation Total
Reg. Marriage 188,000 329,000 517,000
Unreg. Marriage  20,000  23,000 43,000
Total 208,000 352,000 560,000

Actual Claimants survey
Category 1st Generation 2nd Generation Total
Reg. Marriage 188,000 329,000 517,000
Unreg. Marriage   21,0003   24,0004 45,000
Total 209,000 353,000 562,000

Caveat

There is one point about the new survey which might cause some bias in its results. To register as a client of PBC the Claimant had to be in Hong Kong at some point between May - July 1999, and had to be willing to register. It is possible that illegitimate Claimants found it more difficult than legitimate Claimants to come to Hong Kong during that period, or were more reluctant to register than legitimate Claimants. However, while the survey is not perfect in this respect, its results are based on the only hard data of actual Claimants that has been released publicly (the Immigration Department has its own records on actual right of abode Claimants in Hong Kong too) and the evidence suggests that this possible bias in the data has not had a substantial practical effect. This evidence includes:

  1. There is no evidence that any illegitimate Claimant who wanted to come to HK to claim right of abode was unable to do so. The evidence is in one sense to the contrary - curbs were being placed on the entry to HK of legitimate Claimants during the registration period precisely because officials thought they might claim right of abode.

  2. The new survey result tallies fairly closely with the Government's own "direct question" survey result.

  3. The new survey result accords with common sense concerning the ratio of legitimate to illegitimate Claimants, unlike the Government's taxi method whereby more than two-thirds of all additional Claimants are said to be illegitimate.

  4. Applications for right of abode for illegitimate Claimants born after either of their parents became a permanent resident were accepted for the first time ever starting 16 July 1999, 3 months ago. Based on its taxi method survey, the Government announced in June 1999 that it expected 200,000 illegitimate Claimants to be eligible. The media went to different points in Guangzhou on 16 July 1999 expecting a flood of illegitimate Claimants making applications. Instead, only a tiny number of illegitimate Claimants appeared to register on that day; the overwhelming majority of applicants were apparently legitimate. Since 16 July there have no reports indicating that the Public Security Bureau has been swamped with applications from illegitimate Claimants.

30% of Respondents to Government Survey Indicated their Children Would Not Exercise their Right of Abode

It was widely reported that the Government's survey also indicated that some 30% of eligible additional Claimants would not seek to live in Hong Kong. If that is taken into account, the number of estimated Claimants is substantially reduced. See graph 4 as an example, although the 30% figure could be applied to the figures from any of the surveys.


  1. The word "additional" is used throughout this document and in the graphs. It means those right of abode Claimants who (a) were born before either parent became a HK permanent resident; and (b) those born out of registered marriage, described as "illegitimate". They are "additional" to the Claimants expected by Government prior to the CFA judgments.

  2. This figure does not include the right of abode Claimants who were born in wedlock and after either of their parents became a HK permanent resident - persons who the Government has always accepted have right of abode. The Government's estimate of the number of such Claimants is 98,000. This makes the total number of legitimate 1st generation Claimants used in each of the 3 surveys to be 286,000 (188,000 additional Claimants + 98,000 "original" Claimants.) See notes 2 and 3 of the Government's press release dated 25 June 1999.

  3. The figure of 21,000 unregistered marriage Claimants is calculated using the PBC survey ratio of 8.87 legitimates to each illegitimate (8.87 : 1) and applying that ratio to the Government's survey estimate of 188,000 legitimate Claimants.

  4. The figure of 24,000 unregistered marriage Claimants among the 2nd generation is calculated by applying the Government's estimate that each 1st generation illegitimate Claimant will have an average 1.15 children born in the Mainland.


1999 (c) Hong Kong Human Rights Monitor


Back to the homepage in: English Traditional Chinese Simplified Chinese