2012 A Sequel to the 2009 Research Report on

Poor Housing Conditions in Hong Kong – the Extent, Distribution and the People Affected

November 2012

Contents Introduction ............................................................................................................................... 3 Scope and definition .................................................................................................................. 5 PH definition .............................................................................................................................. 7 Methodology............................................................................................................................ 15 Summary findings .................................................................................................................... 17 List of tables in Appendix 5 ...................................................................................................... 24 Poor Housing conditions in the Hong Kong SAR and its District Council districts ................... 25 Change from 2006 to 2011 .................................................................................................. 25 Geographical differences in PH composition....................................................................... 26 Rent for private permanent housing and rent-to-income ratio .......................................... 27 Bedspace/cockloft dwellers ................................................................................................. 28 Room/cubicle renters .......................................................................................................... 29 PH whole quarters ............................................................................................................... 30 Rooftop dwellers .................................................................................................................. 31 Temporary housing dwellers ............................................................................................... 31 Characteristics of Poor Housing in Sham Shui Po, Yau Tsim Mong and Kowloon City ........ 32 Limitations of the study ........................................................................................................... 34 Acknowledgements.................................................................................................................. 36 Appendix 1: District Council districts ....................................................................................... 37 Appendix 2: Census variables and their data items ................................................................. 38 Appendix 3: Definition of Census variables ............................................................................. 41 Appendix 4: Note on index values ........................................................................................... 47 Appendix 5: Tables ................................................................................................................... 48 Page | 2

Poor Housing Introduction This research report is a follow-up to our previous report of the same title dated September 2009 which was then based on the results of the 2006 Population By-Census. BPF has been an early bird to tap the detailed results of the 2011 Population Census (the Census”) now available and extracts form the main part of the quantitative results in this report. This report pinpoints the places in the SAR where living conditions for residents can only be described as poor. It does not include public housing in which living conditions are well managed and the buildings are well maintained. Many of the buildings the report covers are 40 or more years in age and are getting older. Much work has already been done by the Urban Renewal Authority and the Housing Society on investigating and reporting on the conditions of these ageing buildings. This report, based upon the latest available information, is about people, their living quarters, the rent they pay, how long they have been living in Hong Kong, their ethnicity and so forth. It rounds out the picture and hopefully will assist all those concerned with coming to grips with the problem of poor housing in all its aspects. It has been noted that a trend has emerged over the last decade in which landlords have been renovating and sub-dividing their existing units into new, self-contained, smaller units comprising a room with or without a ‘living room’ along with very basic sanitary and cooking facilities. Landlords of these units usually demand much higher rent than that for non self-contained ones. These units, complete with their own entrances are treated as an ordinary whole unit of living quarters by the Census & Statistics Department, Hong Kong SAR (C&SD) and thus there is no separate delineation of such split units. The BPF has been at the fore-front in reporting such phenomena in its previous September 2009 Poor Housing Page | 3

Report as well as in its September 2010 “Discontent” Report. Here we have formally incorporated as part of “Poor Housing” an estimate of the “poor” or “low-quality” flats, many of which are sub-divided. This category has now emerged as the dominant type of “Poor Housing”.

Page | 4

Scope and definition Hong Kong has an extensive public housing scheme which takes care of nearly half of its population, a large proportion of which belongs to the low income group. However, there is a section of the population with an income level approximate to or below that of public housing renters and lives in private sub-standard housing. This study is intended to identify such poor housing (PH), its extent and distribution and the profile of the people affected. It will also look into the most affected urban districts. The Census contains a whole range of data and the following are of most relevance to the study, namely: 1.

type of housing, e.g. public/private, permanent/temporary, whole or shared quarters, etc.

2.

tenure situation, e.g. whether the premises is owned or rented by the occupier, whether it is shared

3.

household/individual information, e.g. age demographics, household composition, income, educational attainment, ethnicity, etc.

4.

type of accommodation with data items as follows:

A. permanent whole quarters, which can be further differentiated according to six types of housing: a) public rental; b) subsidized sale; c) private residential flats; d) villas/bungalows/modern village houses;

Page | 5

e) simple stone structures/traditional village houses; f) staff quarters B. Room, i.e. unit of quarters subdivided into cubicles sharing common facilities with other household/s in the same unit C. Bedspace/Cocklofts D. Rooftop structures E. Temporary excluding rooftop structures F. Others which include accommodation in non-residential areas with no area partitioned off for living purpose; unsheltered accommodation, etc. (Notes: “Room”, “Bedspace”, and “Rooftop structures” are all considered as private housing. Also there is a further distinction regarding whether the residential quarters are originally designated as domestic or non-domestic units)

Page | 6

PH definition 1. “Bedspace/cocklofts”, “Rooftop structures”, and other “Temporary” Structures: A living unit’s physical condition, its safety, the level of crowdedness and the quality of the surrounding neighbourhood all contribute to “poor” housing conditions for people living there.

Therefore, “Bedspace/cocklofts”, “Rooftop structures”, and other “Temporary”

housing, as in our previous studies, are included in this study’s definition of poor housing because of their less than desirable physical conditions.

2. PH Rooms/cubicles: C&SD defines Rooms/cubicles as having “more than one domestic household sharing the unit of quarters. This includes households occupying one or more rooms/cubicles together with other cockloft(s) and/or bedspace (s) in the quarters” (see Appendix 3) When two or more households share a single living quarters unit subdivided into “rooms/cubicles”, this could fall into the category of poor housing. However, for some households, this might also be a matter of choice in some circumstances such as those in which single adults want to move out of their parents’ home for some independence or to be closer to their place-of-work. These more affluent room renters should be segregated from the study’s target group. In this report we have used the $2,000 monthly household rent for private permanent housing as the threshold to exclude the more affluent room renting households. This represents an amended practice compared with our previous “Poor Housing” studies where a per capita household income below the SAR median for all housing types, i.e. $6,000 in the 2006 By-census was used. Given the shortage of affordable low-end housing in the private market, room rentals remain a popular choice for certain

Page | 7

sectors of the population, such as economically inactive persons or ethnic minorities, who are attracted by the comparatively low rent levels of rooms. 3. PH whole quarters: Meanwhile, some landlords, or their main tenants, have been shrewd in spotting and fulfilling the unmet demand for small, affordable housing units, particularly within crowded urban areas. They have done so by subdividing an original housing unit into several selfcontained units complete with individual sanitary facilities and sometimes even with cooking facilities and with a separate entrance in order to maintain individual household privacy. It is now estimated that around 40,000 of these self-contained small housing units were created between 2001 and 2006 as revealed by the reduction in the number of “room renter” household during the same period. We believe that most of such subdivisions go unrecorded officially. The increase in supply of such self-contained private quarters does not contribute to the official housing stock statistics as most of the alternations were done without application to the Buildings Department. Market forces have created, and continue to create, “solutions” which address the pronounced need, and consequent demand, for this lower standard housing. The aforementioned self-contained split quarters have not been separately enumerated by C&SD and thus have fallen outside of our previous classification of PH. However, these quarters are often substandard as normally they are crowded and severely short of space for communal or individual activities, such as studying, thereby leaving very little or often no space for privacy amongst members of the same family. This trend is confirmed by a number of third party sources, including media reports, field surveys and project team interviews with knowledgeable participants and experts in the housing sector.

Page | 8

This supposition is further supported by the substantial reduction recorded since 1996 in the number of shared tenancy households in private quarters while the number of sole tenant whole quarters households has increased substantially. Since most households in the lower income brackets experienced falling income during the period, it is a very reasonable assumption that the reduction in the prevalence of quarters sharing is actually due to the increase in the number of sub-divided flats whose occupants have been ‘elevated’ to sole tenant status albeit with no real improvement in the size of living space but with some improvement in privacy. Table A below indicates that the creation of such sub-divided flats has been rampant between 2001-2006 (but started in the prior 5 year period). This trend has continued from 2006 to 2011 but at a slower pace likely owing to a) gradual exhaustion in the supply of large and very old flats most suitable for subdivision, and b) growing public attention to the lack of safety in subdivided flats and the buildings containing them.

TABLE A Domestic households by tenure of accommodation in private quarters, 2011 1996

2001

2006

2011

number of shared tenancy domestic households occupying private quarters

100,362 (5.4%)

73,463 (3.0%)

22,109 (1.0%)

11,491 (0.5%)

number of sole tenant domestic households occupying private quarters

191,229 (10.3%)

226,435 (11.0%)

266,562 (12.0%)

318,488 (13.4%)

Note: Figures in ( ) denote % in respect of the total domestic households excluding those living on board vessels.

Field research indicated that a $4,000 monthly rent for a whole quarters unit in private permanent housing could reasonably represent the universal threshold below which such quarters could safely be described as “poor” in terms of space rented or the conditions of the quarters and/or the conditions of the building/the immediate living environment and/or the location’s remoteness in respect of centres of jobs, basic services and social activities. Page | 9

The $4,000 threshold could be too low for sought-after districts such as Central & Western or Wan Chai or Eastern where job opportunities abound and where, according to the Census, the median monthly rent for private permanent housing stands at $10,400, $12,000 and $12,000 respectively. These median rents contrast sharply with their counterparts in Sham Shui Po or in Yau Tsim Mong which stand at $4,000 and $5,500 respectively. By using the lower threshold of $4,000 per month, we are generating a conservative estimate for poor housing in the urban areas. We believe that there could be additional unsatisfactory sub-divided flats say mostly in Central & Western, Wan Chai, Eastern and better parts of Yau Tsim Mong where some of the flats with a monthly rent between $4,000 and $6,000 could very well be adjudged as “poor”; in fact, there are almost 5,500 households paying such rents for private flats in Central & Western, Wan Chai and Eastern. On the other hand, some private permanent whole quarters in the northern New Territories with monthly rents below $4,000 might not be as “poor” except in terms of their remoteness. The Census shows that less than 10% of the residents in non-New Town New Territories work in the same district where they reside1 . With our estimate of PH whole quarters in the New Territories amounting to 29,000, 10% or 2,900 quarters can be discounted from the New Territories’1 PH. Thus, our probable underestimation in the urban areas can be closely balanced by our similar overestimation in the New Territories.

PH exclusions 1. Staff quarters In our enumeration of PH whole quarters, we have purposely excluded staff quarters even though the actual rent paid is frequently under $4,000 per month. This is because staff quarters are predominately owned by government or quasi-government organizations such 1

table 42 in Summary Results of 2011 Census

Page | 10

as the Hospital Authority and our various universities, all of which generally follow public rental housing or better standards for non-officer ranks. Therefore, housing in the staff quarters category has not been included as PH. 2. Zero rent or rent-free We have also excluded those households who pay zero rent or otherwise reside rent-free in their whole quarters or room/cubicle dwellings because the absence of rent in many of these circumstances is not an indication that these quarters comprise poor housing. Again these can contribute towards under-estimation.

3. Public subsidized flats As in our previous studies, this study also excludes “Public Rental” and “Home-Ownership Scheme” (HOS) accommodations from PH. Firstly, public housing normally provides a selfcontained living unit in which the family/individual tenant is the sole occupant and in which cooking/sanitary facilities are not shared with others. Secondly, public flat renters are more fortunate in that the Government undertakes to cap their median rent-to-income ratio at 10%. 4. Marine Population There are 434 households living on board vessels. These households are not incorporated in this report which addresses only the landed population. These 434 households are quite polarized in their income distribution and 121 of them have a monthly household income under $10,000. Policies and measures intended to alleviate the living conditions of this small but poor marine population could be quite different from those for the landed population.

Page | 11

5. Accommodation in non-residential quarters with no area partitioned off or unsheltered accommodation These accommodations have not been included in PH definition here for practical reasons as data collected is not specific enough and policies and measures to alleviate the living conditions of those say in unsheltered accommodation could be different from those of the other PH population. Thusly, we believe our overall estimate of PH for the whole territory is largely on the mark but on the conservative side by way of our Definition. For the purposes of simplicity and straightforward comparison, figures in this report incorporate the more universal definition of PH, i.e. based on accommodation types deemed as poor or on a benchmark monetary rent level irrespective of District Council district salient conditions.

PH summary PH includes the following types of accommodation: 1.

Bedspace/cocklofts

2.

Room/Cubicle, with a requirement to pay rent and a monthly rent below $2,000

3.

Whole quarters rentals (excluding staff quarters) with a requirement to pay rent and a monthly rent below $4,000

4.

Roof top structures

5.

Temporary, non-rooftop structures (hereinafter referred to as “Temporary housing” or “Temporary structure”)

Types 1, 4 and 5 are taken directly from the Census and types 2 and 3 have been re-defined by the project team. Note: “Whole quarters” is a generic term including all flats and apartments in multi-storey blocks or houses built mainly for residential purpose as well as all former subsidized sale

Page | 12

flats, i.e. those free from “alienation of restrictions”. The term also includes villas, bungalows, modern village houses, simple stone structures, and traditional village houses or staff quarters. In the urban areas, types 1-5 of the above-listed PH accommodation are normally situated in buildings which are of “poor” quality and are poorly maintained, if at all. These buildings have a very high probability of having no lift, thereby creating mobility problems for the disabled and the elderly and posing challenges for fire fighting. In addition, staircases are usually narrow and poorly lit and many so-called light wells that intend to provide light and air to apartments in the rear of the building do not function as such.

Page | 13

Poor Housing Classification Chart

Is the household’s accommodation a bed space/cockloft, rooftop structure or temporary structure?

Yes

No

Is the type of accommodation equivalent to “room/cubicle”?

Yes

No

Does the household have to pay monthly rent >$0 and < $2,000?

Is the type of accommodation equal to non-staff whole quarters and monthly rent>$0 and $0 and < $2,000)

PH whole Temporary quarters (rental housing with monthly rent >$0 and < $4,000, excl. staff quarters

PH Total

SAR total

945

1,084

2,841

38,999

1,965

45,834 1,151,116

42

123

714

27,702

15,305

43,886 1,217,246

987

1,207

3,555

66,701

17,270

89,720 2,368,362

Urban %

2.1%

2.4%

6.2%

85.1%

4.3%

100.0%

NT %

0.1%

0.3%

1.6%

63.1%

34.9%

100.0%

Total %

1.1%

1.3%

4.0%

74.3%

19.2%

100.0%

Urban NT Total

Page | 17

3.

The urban area (Hong Kong Island and Kowloon) accounted for close to 50%, or 100,000 persons, in PH, the predominant PH type there being PH whole quarters flats accommodating around 85,000 people and representing 87.4% of the total PH population there. The rest of the urban PH population, numbering approximately 12,200 people, were accommodated across the remaining PH housing types, i.e. about 4,500 in temporary housing, 4,300 in PH rooms, 2,400 in rooftop structures and 1,000 in bedspace/cocklofts.

TABLE B-2 Population in domestic households by PH type and District Council district, 2011 District Council districts:

Central & Western Wan Chai Eastern Southern Sham Shui Po Kowloon City Wong Tai Sin Kwun Tong Yau Tsim Mong

Bedspace/ Rooftop cockloft structures

PH room PH whole Temporary quarters housing

PH total

SAR total

PH %

243,636 147,481 573,882 267,230 369,338 364,185 412,142 610,552 299,892

1.57% 1.32% 0.96% 1.89% 8.14% 4.35% 0.69% 1.68% 7.33%

281 27 67 259 30 32 336

16 46 165 924 356 54 232 590

268 251 135 33 1,249 766 124 134 1,354

3,160 1,559 4,864 3,460 27,304 14,192 2,217 8,924 19,412

89 69 282 1,561 337 502 435 916 293

3,814 1,952 5,513 5,054 30,073 15,846 2,830 10,238 21,985

Urban sub-total

1,032

2,383

4,314

85,092

4,484

97,305 3,288,338

Urban horizontal %

1.1%

2.4%

4.4%

87.4%

4.6%

100.0%

Kwai Tsing Tsuen Wan Tuen Mun Yuen Long North Tai Po Sha Tin Sai Kung Islands

48 -

222 42 11 28 -

157 368 92 79 57 80 80 3 -

2,260 7,360 4,849 22,363 12,256 4,517 2,956 934 6,034

849 2,217 3,797 11,978 12,769 3,583 1,816 736 1,290

3,488 10,035 8,749 34,448 25,082 8,180 4,852 1,673 7,324

NT sub-total

48

303

916

63,529

39,035

NT horizontal %

0.0%

0.3%

0.9%

61.2%

37.6%

Total:

1,080

2,686

5,230

148,621

43,519

Horizontal %

0.5%

1.3%

2.6%

73.9%

21.6%

497,898 296,499 474,751 566,555 296,137 289,300 617,446 427,568 135,984

0.70% 3.38% 1.84% 6.08% 8.47% 2.83% 0.79% 0.39% 5.39%

103,831 3,602,138 100.0% 201,136 6,890,476 100.0%

Page | 18

2.92%

4.

The New Territories also has around 28,000 or about 60% of its PH households in PH whole quarters, indicating the generally lower rental levels there. 60% of these whole quarters are more likely to be in the form of modern village houses, sheds or simple stone structures and traditional village houses versus an estimated 40% in multi-storey flats. Nevertheless, the trends of vanishing space and smaller flats continue into the New Territories particularly into the New Town areas, although over crowdedness there is less severe than in urban areas.

5.

The New Territories have another 15,000 non-rooftop temporary housing which account for about one third or 35% of its PH households. In fact, the New Territories account for nearly 90% of all the SAR’s temporary structures. As expected, the prevalence of “Temporary” housing increases with the distance of the outlying/rural areas from the more densely populated city centre, particularly in the northern-most New Territories region. Thus “Temporary structure” was the second most common form of PH, followed by “room cubicles” and then distantly by “Rooftop” and “Bedspace” dwellings.

6.

The three urban DCDs below accounted for nearly 70% of both the PH households and population in the urban area, with the number of PH households and population respectively being:

Number of

Sham Shui Po

Yau Tsim Mong

Kowloon City

household

14,162

11,058

7,600

population

30,073

21,985

15,846

These three districts have the greatest clusters of poor or varied buildings according to a recent study by the Urban Renewal Authority which cites a total of 2,200 such building blocks. Page | 19

7.

In comparison with the findings of the previous BPF PH reports, Sham Shui Po (SSP) has replaced Yau Tsim Mong (YTM) as the district with the most egregious conditions of PH. This district has the lowest median rent level at $4,000 for those in private housing despite its being quite centrally located on the Kowloon Peninsula. Its residents have a diversity of ethnic backgrounds.

8.

Yau Tsim Mong has become the urban district with the second highest prevalence of PH. Following SSP, it draws even with Kowloon City (KC) in recording the second lowest rental levels amongst all the urban areas (Kwun Tong has the same lowest rental level as does Sham Shui Po but in the case of Kwun Tong, this is due to the presence of some 5,600 low rent staff quarters.)

9.

Yuen Long tops the PH list in terms of its number of PH whole quarters flats and temporary quarters with approximately 10,000 and 5,000 units respectively. Yuen Long is followed by North which has nearly 5,000 PH whole quarters flats along with about an equal number of temporary quarters. Thus, the northernmost part of the New Territories, which is situated farthest from city living, has a high prevalence of PH, a fact which is partly indicated by its lower rents.

10.

Households and residents of PH throughout the SAR were, in general, likely to have a per capita monthly income below that of the SAR median and generally on a par with those living in public rental estates. However, while their median income levels are comparable, PH households, apart from those in urban bedspaces or New Territories temporary structures, have a greater variance in terms of their income distribution and represent a more diverse group than the public rental households. In addition:

Page | 20

a. The education levels of the PH household heads are higher in comparison with those five years ago. However, there is a distinctly lower presence of persons with tertiary education (Table 11); b. PH households recorded a higher concentration of single-person households and single-parent households. PH households are twice as likely to be a single person household as compared with the SAR norm. Single person households have more difficulty in applying for public rental housing. Also, single parent families are three times likely to be found in PH households compared with the SAR overall (Table 5); c. The Census indicates that the average age of PH residents turned younger versus five years ago and there were more household heads in the age group of 15-29. When compared to the SAR overall, it is twice as likely to find household heads in this age group probably because more younger people opt for subdivided flats (Table 12). Moreover, further examinations show that that 935 PH households are headed by students with 80% of these residing in PH whole quarters flats (Table 14); d. PH residents are almost three times likely to be new immigrants with 17.5% of them having residence outside of Hong Kong 5 years ago versus 5.7% in the SAR overall (Table 9);

Page | 21

e. PH residences were occupied overall by a high proportion of Nepalese, Indonesian/Thai/Filipino and Indian/Pakistani nationalities in decreasing order of proportions. This reflects a strong demand for low-end housing from the Nepalese and Thai/Indonesian/Filipino communities, with the latter group often as non-live in foreign domestic helpers. Together with those of Indian/Pakistani origin, these three groups account for about 6% of the PH households in Hong Kong. Table C Percentage share of PH total by ethnicity of household head, 2011 % share of PH Total

% of SAR Total

Index of PH to SAR

Nepalese

2.0%

0.2%

839

Thai

0.4%

0.1%

463

Indonesian/Filipino

1.8%

0.4%

507

Indian/Pakistani

1.7%

0.6%

285

93.3%

97.1%

96

Other Asian

0.3%

0.5%

66

White

0.4%

1.2%

37

Total:

100.0%

100.0%

100

Chinese

Page | 22

Map 1: Poor Housing in Hong Kong (by districts), 2011

Page | 23

List of tables in Appendix 5 Table 1:

Number of Domestic households by PH type and District Council district, 2011

Table 2:

Population in domestic households by PH type and District Council district, 2011 (arranged in order of HK Island / Kowloon / New Territories)

Table 3:

Percentage of domestic households by rent-to-income ratio and PH accommodation type, 2011

Table 4:

Percentage of domestic households by rent-to-income ratio in broad accommodation type, 2011

Table 5:

Percentage of domestic households by household composition and PH accommodation type, 2011

Table 6:

Median per capita monthly domestic household income (HK$) by PH accommodation type and household size, 2011

Table 7:

Median household rent (HK$) by PH type, 2011

Table 8:

Percentage of domestic households by ethnicity of household head and PH type, 2011

Table 9:

Percentage of population (age 5 and above) in domestic households by 2006 domicile and PH type, 2011

Table 10:

Number of immigrants (age 5 and above) whose place of residence in 2006 was outside the Hong Kong SAR, by District Council district, 2011

Table 11:

Percentage of domestic households by education attainment of household head and PH type, 2011

Table 12:

Percentage of domestic households by age group of household head and PH type, 2011

Table 13:

Index of household head by age group and PH type, 2011

Table 14:

Percentage of domestic households by economic activity status of household head and PH type, 2011

Table 15:

Percentage of domestic households by PH type in the 5 districts with the highest PH prevalence, 2011

Page | 24

Poor Housing conditions in the Hong Kong SAR and its District Council districts Change from 2006 to 2011 1.

‧Partly due to the introduction of a minimum wage in April 2011, household income has generally risen from 2006. The median monthly domestic household income has grown 19% from $17,250 to $20,500 while inflation for the lesser half of the population, as measured by CPI(A), rose by 14%. Although the average real income for this group has increased by 4.4% in these five years, the average private housing market rent has TABLE D Consumer price index changes from 2006 - 2011 CPI(A) overall

Composite CPI - overall

CPI(A) private housing rent

Composite CPI private housing rent

2006

93.3

92.0

88.1

86.5

2011

106.4

106.0

107.9

107.7

inflated far more, registering an overall increase of 25% (or 10% in real terms) while, in particular, the average rent in the lower private housing market has climbed by 22% (107.9 over 88.1). Thus, the increase in household income is exceeded by a steeper rate of increase in rent. 2.

The prevalence of PH increases significantly from 2006 to 2011 with the adoption of an extended definition of PH which now includes certain whole quarters resulting from subdivision, or self-contained units, in which there is no sharing of communal facilities but which are still considered to be poor based on a low rent level which reflects their living conditions/environment and state of repair. This adjustment is permitting the new set of PH data to knit up with previous PH data as the subdivision of flats has

Page | 25

become widespread from the late nineties and migration from Rooms to Whole Quarters can be observed. 3.

Some urban districts with the highest incidence of rundown buildings such as SSP, KC, YTM and Kwun Tong in the urban area have seen some degree of neighborhood redevelopment in recent years, including work by the Urban Renewal Authority. Urban redevelopment has probably contributed to the decrease in shared PH rooms and rooftop dwellings. However, the decrease in shared PH rooms is also likely due to a surge in the subdivision of flats and also a change to a more stringent definition of PH room in this updated study. Bedspace/cocklofts have increased but that could be due to the inclusion of the cockloft category while cocklofts were not separately specified in previous studies.

4.

Even though the government has entirely phased out public temporary housing, private temporary PH is still on the rise while there are significant reductions in PH Room and Rooftop when compared to 2006. TABLE E Number of households in conventional PH types in all districts, 2006 vs. 2011 2006

2011

% change

923

987

6.9

PH Room

12,856

3,555

-72.3

Rooftop Temporary

1,734 15,452

1,207 17,270

-30.4

PH type:

Bedspace/cocklofts

11.8

Geographical differences in PH composition 3.0% of Hong Kong’s population, or 3.8% of its households, lived in poor housing. The top five amongst the eighteen districts ranked by total number of PH households were:

Page | 26

TABLE F Districts with the greatest number of PH households, 2011 Yuen Long

Total households PH %

Sham Shui Po

Yau Tsim Mong

North

Kowloon City

15,022

14,162

11,058

9,952

7,600

7.9%

10.5%

9.8%

10.0%

6.1%

Table 1 also reveals a difference in PH composition: PH types and composition change with their district’s distance from Victoria Harbour, i.e. from older, more densely populated districts bordering or in proximity to the Harbour, moving steadily outward to the most rural areas such as northern New Territories and outlying islands. In the older, more densely populated areas bordering the Harbour, the dominant PH type was PH whole quarters in multi-storey buildings, a category accounting for 80% of all PH households. Almost all “PH Room/cubicles” and almost all “Bedspace/cocklofts” were situated in these same areas while Temporary Housing made up only 3% of PH households. By comparison, 60% of the PH households in the New Territories live in PH Whole Quarters and 60% of these are represented by village houses/stone structures rather than multistorey flats. There is also a dominance of temporary housing with around 15,000 such units accounting for 89% of the SAR’s temporary housing. (see Tables 1, 2)

Rent for private permanent housing and rent-to-income ratio Hong Kong Island has the highest median monthly rental level in the private housing market and all districts there have recorded a median monthly rent level above $10,000 whereas Sham Shui Po has the lowest level at $4,000. While Kwun Tong shares the same lowest rental level with Sham Shui Po, Kwun Tong also has a high percentage of staff quarters (5,576 out of the SAR total 19,338) the majority of which are under $4,000 per month. As a result, Kwun Tong’s rental levels as well as its rent-to-income ratio are the lowest amongst

Page | 27

all district council districts. Otherwise, median rent-to-income levels for private sector housing converge at around 25% for almost all districts. In public rental housing, nearly 80% of the renters paid less than 20% of their income as rent. Thus public rental households have a higher disposable income particularly when compared with PH households given that both households have comparable average incomes. (see Tables 3, 4)

Bedspace/cockloft dwellers Bedspaces are spread throughout all the urban districts. Out of the 987 households including cocklofts in this category, 105 or about 10% were made up of more than one person. The average per capita monthly income for these 987 households was around $4,650 and their median rent was $1,260, implying that on average their median rent-toincome ratio is around 28%. 42% of these households paid more than 30% of their income in rent and these households are probably the poorest overall in terms of their housing conditions

TABLE G

high

rent-to-income level. A

Household composition of bedspace dwellers, 2011 Household composition:

and

No. of bedspace households

One adult (aged 18 and over) only

882

Single adult with children aged