For discussion on
22 December 1998

Legislative Council Panel on Housing

Rent Control in Private Residential Housing


This paper briefs Members on the present position regarding the phasing out of rent control under the Landlord and Tenant (Consolidation) Ordinance.


2. The present rent control legislation, the Landlord and Tenant (Consolidation) Ordinance (LTO), was enacted in 1973. It applies to a small number of pre-war (Part I) and some post-war (Part II) domestic tenancies ( the latter being those tenancies in premises built before 19 June 1981 and created before 10 June 1983).

3. It is the Government's established policy to phase out rent control gradually. Back in 1993, legislation was passed by the Legislative Council to provide a mechanism for gradually adjusting controlled rents to prevailing market rents (PMR) (from a minimum of 60% of PMR before July 1994 to 90% after June 1996 for Part I and from a minimum of 75% of PMR before July 1994 to 90% in December 1996 for Part II). This phased programme has worked well and modest increases in rent have been spread out over time. However, this process was set back when the Legislative Council passed a resolution in December 1996 to defer the ending date of decontrol to 31 December 1998 and to make other changes to the rent increase mechanism under Part II.

The Policy

4. Phasing out of rent control is a long standing Government policy. Rents in the private sector should be freed from artificial controls and should be determined by market forces. For many decades, rents of a small number of controlled premises have been suppressed at artificially low levels at the expense of landlords. This is particularly unfair to landlords of older premises who are increasingly required to upkeep their premises in order to meet various requirements relating to building, fire safety and maintenance standards. Seen in present day circumstances, rent control has been applied arbitrarily to a selected group of tenants according to the age of the premises or the date on which a particular tenancy was created, regardless of the financial situation of these tenants. This special treatment creates an inequitable situation as the large majority of tenants in the private sector are paying market rents. Hence, Government is of the view that rent control should be removed according to a planned and gradual process as already laid down in the legislation. Upon decontrol, former Part I and Part II tenants will come under Part IV and will continue to enjoy security of tenure.

5. To assess the impact of decontrol on tenants, the Rating and Valuation Department conducted a survey on controlled tenancies in the middle of this year. The final report was completed last month (November). The findings and analyses are summarised below.


6. There are around 15 000 controlled tenancies (involving some 18,000 households) in the territory, constituting only about 1% of all private domestic premises. The majority of them fall under Part II, being 97.6% of the total.

Part II Tenancies

7. There are 14 661 Part II premises. The average controlled rent is $6,258 which represents 82.5% of the average PMR of $7,586. To obtain a more accurate assessment, the study has excluded those dormant cases where rents have not been increased for a long time or up to the minimum level as permitted under the Ordinance. If these tenancies are excluded, there are 11 996 premises and these tenants are already paying close to 90% of the PMR (87.8%). Their average rent is $6,610 and their average PMR is $7,528. With domestic rents falling steadily in the past year, the gap between market rent and controlled rent has narrowed. Decontrol will not have much adverse impact on tenants involved.

8. Naturally, Members will be particularly concerned about the impact of decontrol on sub-tenants living in small bedrooms, bedspaces and cocklofts. There are about 4 500 sub-tenants. They are on average paying more than 90% of the PMR. As a result, the rent increase arising from decontrol is fairly small -

Controlled RentRent to PMR Ratio

9. It is unlikely that decontrol in this gradual manner will create particular hardship for sub-tenants. Besides, assistance under the Comprehensive Social Security Assistance Scheme is available to tenants facing genuine hardship.

10. Rent control in the present form was first introduced to address the acute shortage of rental flats in the private sector. However, with increase in the overall supply of rental flats in the private market and the recent downward adjustment in rental levels, such control is no longer necessary or desirable. In fact, whilst the majority of unprotected Part IV tenants are paying about 30% of their household income on rent, tenants of controlled tenancies are paying proportionately less at the expense of the landlords concerned. This is clearly not fair.

11. Upon decontrol, a substantial number of former controlled tenancies ( 70%, 3,481 tenancies) would not face any immediate increase in rent either because they are already paying market rent or their rents have been adjusted this year. About 25% of tenancies due for rent revision in 1999 (1,235 tenancies) will face an increase of less than 10% of the existing rent which is already allowed by the present legislation. The remaining 5% (323 tenancies) will face an average additional increase of 10% ($676) upon decontrol.

12. Overall, the additional rent for these 1558 tenancies and 255 sub-tenancies upon decontrol in 1999 is small, as shown below -

Cases Subject to Rent Increase
Average Projected Increase in Rent
Part II
Expires as
Part II
34811558$6,236 $861$721$140(2.2%)
2490-- --
830--- -

Part I Tenancies

13. As Part I tenancies constitute only a small fraction of the controlled tenancies (2.4%), the key findings on Part I tenancies are summarised at the Annex for Members' easy reference. It can be seen that the additional rent increase as a result of decontrol is minimal.


14. Given that the existing legislation on rent control is not desirable and the impact of decontrol is minimal as revealed by the survey, we conclude that rent control under Parts I and II should be allowed to lapse on 31 December 1998 as resolved by the Legislative Council in December 1996.

15. Former protected tenants will be provided security of tenure in accordance with Part IV of the Ordinance under which rents may be determined by the Lands Tribunal in case of disagreement between the landlord and the tenant and tenants are protected from unlawful eviction and repossession.

Housing Bureau
Government Secretariat
December 1998


Findings on Part I Controlled Tenancies

There are 367 Part I premises. The average controlled rent is $3,198 which represents 61.3% of the average PMR of $5,213. If dormant cases (see Note) are excluded, there are 142 premises and these tenants are already paying 88% of the average PMR.

2. Out of the 367 Part I premises, there are 181 sub-tenants. The majority (167 or 92%) are occupying bedrooms and they are already paying 88.5% of the PMR.

3. Upon decontrol, the majority of former controlled whole-flat tenants (61%, 78 tenancies) would not face any increase in rent because they are already paying market rent. The remaining 39% (50 tenancies) due for rent revision in 1999 will face an average increase of 4.96% ($219) of the average existing controlled rent ($4,411).

Note : Those cases where rent has not been increased for over 30 months.