Land Supply Strategy – Missing Places in the Jigsaw

Executive Summary

This paper attempts to analyse and provide observations on the land supply in Hong Kong in the medium and long term.

The potential residential land supply of Hong Kong announced by the government up to 2020 has been studied. Based upon the currently available details and the assumption that there will be no major policy change, the announced target of 20,000 private residential units land supply per year would be unlikely achievable in this period.

Would the long term supply be more optimistic? The government has started the consultation of reclamation and has also proposed resumption in the New Territories area to facilitate development. This paper will examine the government assumption of development density which drives the assessment of development lands required to meet the needs of Hong Kong. Examples of development density of selected new towns and islands and also new development areas have been studied. In the light of these analyses, the development density assumption seems to be overly optimistic. The importance of MTR connection for these new areas has not been made clear.

Factors restricting the land supply through Lease Modification / Land Exchange not told by the government have also been examined. The better use of the privately owned lands is a key “piece” of land supply which currently is missing out from the Land Supply “Jigsaw”.

The 2010/11 Policy Address: the Beginning of the Change

The 2010/11 Policy Address marked the beginning of the change to address the shortage of lands for development in Hong Kong. The shortage involves different sectors including but not limited to residential, office and shopping areas. The then Chief Executive, Mr Donald Tsang announced the target for private residential units land supply would be 20,000 units per year for the coming 10 years.

The government since then has introduced various measures in government land sales to boost up the number of residential units land supply. These include the required number of units or the restrictions of sizes in land grants beginning from March 2011. Then, there is a shift from auction to tender sales of government lands.

The public consultation of “Enhancing Land Supply Strategy – Reclamation outside Victoria Harbour and Rock Cavern Development” (“Reclamation Consultation”) is another major step which started in November 2011. The government had suggested that both the redevelopment and rezoning land supply options were not suitable for land reserve purposes. Both options would need the involvement of private owners and developers. The first stage of consultation was completed by March 2012.

Then there have been various development studies including but not limited to The North East New Territories New Development Areas (“NENT Development Areas”)Planning and Engineering Study. This study is controversial as the government would like to re-use the ‘Conventional New Town Approach”. This would confirm the exclusion of private land owners in such large scale development areas and this could have far-reaching impact onto the development lands market.

Whist the government has taken up the land supply responsibility seriously, her various actions suggest that she is not keen in seeking the co-operation and/or participation of the private sector land owners.

Hong Kong Residential Land Supply: 2011 to 2020

A news report of Ms. Carrie Lam, the then Secretary for Development, on 14 June 2012 revealed the latest official estimate of private residential land supply for the period of 2011 to 2020. The projected supply units would be marginally less than 60,000 in the first 5 years (2011 to 2015), whilst in the second 5 years (2016 to 2020), the number would be around 105,000. Such figures have not included the potential units which might come about from re-zoning of industrial lands and green-belt, etc which impact in the short term would not be significant.

The following table summarizes the sources of land supply and the relevant number of units in the two 5-years period:

Table 1: Forecast of land supply of private residential units in 2011 to 2020

  2011 to 2015 2016 to 2020 Total % of total
Application List 21,500 53,000 74,500 46%
West Rail Line & MTR 8,230 21,700 29,930 18%
URA 5,000 5,000 10,000 6%
Private Redevelopment 10,000 10,000 20,000 12%
Lease Modification /

Land Exchange

15,000 15,000 30,000 18%
Total 59,730 104,700 164,430 100%

Source: Hong Kong Economic Times report on 14 June 2012

These figures should be realistic if not optimistic assessment by the government as at the date of publication. If there are further major changes to the planning and land use policies in Hong Kong, then these figures could be re-visited accordingly.

The above supply figures represent new residential units, but have not deducted the existing residential units that would need to be demolished for development. This consideration is material for those URA and private redevelopment projects which will account for some 6% and 12% of the projected new supply respectively.

I have relied upon these figures for the analysis in this paper.

2011 to 2015

The supply figures in this period should be fairly accurate given its short term nature and that the government is in good control of various sources of supply including the application list, railway projects and the URA projects. Less certain would be the private redevelopment and land exchange cases (including lease modification).

The first 5 years supply figures have pointed to a significant shortfall of the 20,000 units per year target. Only about 60% of the target could be provided.

2016 to 2020

The supply figures of some 105,000 units suggested that the target of 20,000 units per annum should be achievable.  A close look of the supply would reveal the real challenges.

Some 30,000 units or 30% of the projected supply would hinge upon the timely delivery of developable lands in the NENT Development Areas. Readers may by now appreciate the importance of this source of development lands. The government seems to have felt the need to initiate the resumption so as to control the timing for delivery of the developable lands. The government apparently cannot have the luxury to wait for the private sector to initiate the relevant land exchanges. By doing so, the perception by the public of the government co-operating with the land owners and/or the riches can also be taken away. The use of resumption powers would, however, also mean a very costly exercise for Hong Kong.

Land Supply: Contribution by the Public Sector and the Private Sector

For the 10 years forecast period, the public sector land supply (including land application list, railways projects and URA projects) would account for some 70% of the total supply, with the remaining to be undertaken by the private sector.

The land supply from the land application list and railways projects has a common characteristic: back end loaded. Only about 28% of the total units in this category would be available in the first 5 years and the bulk would come in the second 5 years.

The projected supply from the URA, private redevelopment and lease modification/land exchange would be fairly consistent in the two 5-years period.

The Public Sector Land Supply

The supply of some 51,700 units in the second 5 years would hinge upon the timely delivery of lands in:

  • North East New Territories: 30,000 units;
  • Railway projects in Pat Heung Maintenance Centre and Siu Ho Wan Depot: 21,700 units.

These units are all in the northern part of the New Territories and Tung Chung area where the availability of employment opportunities, amongst other things touching upon the livelihood of the residents, would require further effort from the government.

The Private Sector Land Supply

The land supply from private redevelopment and land exchange routes will account for about 18% of the estimated total supply in the government forecast for the 10 years period. How realistic are these figures is difficult to gauge.

Land Exchange

Land exchange (including lease modification) supply would hinge upon many factors including but not limited to:

  • Relevant private lands have been acquired/controlled by private developers;
  • Relevant town planning applications applied and/or approved;
  • Relevant land exchange applications applied and/or approved in principle;
  • Property market cycle and land premium negotiation between government and the private owners.

Table 2 below showed that the land exchange revenue varied significantly over the last 5 financial years from HK$3.5 billion to HK$20.1 billion or 10% to 56% in terms of percentage to the total land revenue. On average, the sum would be HK$11.2 billion per annum or 22% of the land revenue.

Table 2: Land revenue from land exchange and the total land revenue

  Financial Year 5 years
  07/08 08/09 09/10 10/11 11/12 average
Land Exchange revenue (Billion) 10.2 3.5 20.1 6.7 15.6 11.2
Total Land revenue 53.9 13.7 35.8 65.2 88.1 51.4
Land Exchange revenue as a % of the total land revenue 19% 25% 56% 10% 18% 22%

Source: Lands Department

The above analysis revealed that land exchange is no longer a major supply of residential units. There were only a total of 12 exchange cases with a premium amount of over HK$ 1 billion in this 5 years period. How sustainable are these big land exchange cases is a question.

Resumption under the “Conventional New Town Approach”

The government announced in June 2012 that it will re-use the “Conventional New Town Approach” for implementing the NENT Development Areas. The private lands in the area will be resumed and cleared. Site formation works will then be carried out and infrastructure to be provided. Afterwards, land for various purposes like those planned for private development will be supplied to the market.

How developers/private land owners would react to such threat of resumption in the subject area and the wider New Territories? Would developers be deterred to assemble lands in the New Territories and future land supply through land exchange would become insignificant? Would developers just sit and wait for the resumption without taking further actions to protect their property rights? These questions are yet to be answered.

Private Redevelopments

There have been many private redevelopments in the urban areas in recent years. The contributing factors include:

  • The appreciation of property values;
  • The incentives available under the sustainable development policy; and
  • The lowering of threshold of ownership to apply under the law of Compulsory Sale for Redevelopment.

How sustainable would this route of supply be then? With the government actively pushing lands to the market, this would bring down the hope for property value appreciation. The change in the rules of the sustainability development has such impacts including: reduction of floor area incentives, increase in costs for environmental requirements and underground car parking. The reductions in building height limits through town planning zoning also restrict redevelopment values.

Hong Kong Residential Supply : 2021 and Beyond

When the government announced the Reclamation Consultation, the proposed relocation of the Shatin Water Treatment Plant and similar facilities did light up some hope of significant land supply then. Such hopes quickly died down after it had been revealed that the earliest available time of the land of Shatin Plant would be by 2027, i.e. some 15 years later.

The public acceptability of large scale reclamations, the choice of actual locations and connectivity of these sites to the main urban areas are major issues yet to be resolved. The building up of land reserve would still be a long way ahead.

The Assessment for Demand of Extra Land

The forecast of demand of extra land supply has been linked to the population estimates. The population forecast of Hong Kong in the Reclamation Consultation in 2039 would be up to 8.9 million.[1]Such population forecast (including the underlying assumptions) had attracted suspicions and criticisms from various groups and commentators at the time of consultation.

This paper would look instead at the assumption made by the government of the population density of development lands and what would this mean by comparison to existing development areas.

The existing population of some 7 million has been served by some 17,500 ha of land. The density is approximate 400 persons per hectare of land (“Government Assumed Development Density”). The projected population increase of 1.8 million by year 2039 would require some 4,500 ha of lands. The government considered that existing major land supply resources can provide a total of 3,040 ha of land. A shortage of approximately 1,500 ha of land has then been drawn.[2] This is also the rationale for reclamation from the sea.

Such land requirement assessment had not addressed two major aspects:

  • The demand for more space from existing population and business needs as a result of improved economic conditions.
  • How the new development areas would be able to achieve the Government Assumed Development Density?

The Reclamation Proposal

Reclamation is subject to various statutory and non-statutory constraints and the competing uses of the harbour and the sea. Two maps in the Reclamation Consultation provide vivid pictures of the constraints onto reclamation.[3]

A total of 25 reclamations sites have been put up for preliminary discussion/consultation. These sites are of different scale from some 10 ha to over 1,500 ha. Nine of the sites are below 29 ha, eight of them are between 30 and 99 ha and eight of them are over 100 ha.

Further assumptions are made to these site areas to analyse the reclamation proposal:

  • For sites fall within 10-29 ha: an average size of 20 ha
  • For sites fall within 30-99 ha: an average size of 65 ha
  • South Cheung Chau : 1,500 ha and Lamma North : 400 ha

Table 3: Analysis of Reclamation Sites

  No. of Sites [A] Total reclamation area (ha) [B] (%) Average reclamation area (ha)

[B] / [A]

Artificial Island 3 2,000 47% 667
Reclamation to Connect Islands 2 1,000 23% 500
Reclamation upon artificial or disturbed shoreline 13 645 15% 50
Reclamation upon natural but not protected shoreline 7 655 15% 94
 Total: 25 4,300 100%  

This could mean some 4,300 ha of new land if all reclamations would go ahead as proposed. This figure is significantly more than the 1,500 ha of land shortage figure as set out in the Reclamation Consultation. About 30% of the areas will be along the shoreline and the remaining 70% of these areas will be in the form of large artificial islands and/or reclamation to connect islands.

We shall look at the population density in some examples of existing built up areas of new towns, and islands and new development areas for comparison. .

Existing built up areas and new development areas

For the existing built up areas, the selections include Tseung Kwan O, Tsing Yi and Ma Wan with the latter two being islands with road network.

Table 4: Planned population density of Tseung Kwan O, Ma Wan and Tsing Yi

  Tsueng Kwan O Tsing Yi Ma Wan
Total area (ha) 1,732 1,067 101
Area after deducting Green Belt(ha) 972 644 69
Planned population 450,000 203,700 15,000
Planned population density

(per ha) (Total Area)

260 191 149
Planned population density

(per ha) (Deducted Green Belt)

463 317 218

Sources: Outline Zoning Plans / Planning Department

Tsueng Kwan O and Tsing Yi are served by MTR while only limited vehicular traffic has been allowed for Ma Wan Island.

The inclusion of green areas/mountains around these two areas would have significant impact onto the population density. Only when the green belt areas excluded from the calculation, then the population density in Tseung Kwan O area would exceed the Government Assumed Development Density, whereas the other analyzed density would all be well below.

This paper also studied the development density in various new development areas which public/community engagements and feasibility consultancy study are ongoing/will be undertaken. The analysis revealed that only Fanling North and Tung Chung (assuming the green belt areas being excluded) will be able to attain the Government Assumed Development Density or exceed the same. Both areas are served by MTR. Whereas for the rest of the areas (including Kwun Tung North which is planned to be served by MTR, Ping Che, Anderson Road Quarry and Kong Nga Po) will not be able to reach the Government Assumed Development Density.

Table 5: Development Density in New Development Areas

  Develop-able Area (Hectare) Resident-ial units to be provided Planning Population Planned population density (per hectare) Support of MTR (current / future) Remarks
Public /Community Engagement
North East New Territories Stage 3 – NDAs
Ping Che 153 6,500 17,600 115 No First population intake from 2022 – 2031 (Phase)
Kwu Tung North 251 28,700 81,900 326 Yes
Fanling North 129 18,600 52,100 404 Yes
Total 533 53,800 151,600 284    
Tung Chung New Town Extension Development Stage 1 206


Not yet determined 108,000


524 Yes Study target to end in 2014



(Whole New Town)

Anderson Road Quarry Stage 2 863 8,650 23,000 267 No Land availability: 2019-2020
Engineering Feasibility Study
Kong Nga Po 18.9 1,200 3,600 190 No Land availability: 2020


1 Area excludes existing green belt

2 Total areas includes 280 ha of extension area

3 Area of Anderson Road Quarry

 Source: OZP, relevant consultation documents and tender document.

The above analysis revealed that for development lands to achieve the Government Assumed Development Density, the MTR connection would seem to be a necessary condition.

If the hope of land reserve lies with reclamation and the large artificial islands and connections of islands would be the major source of supply, then the connectivity issue by mass transportation means to the main urban areas would be the major challenge.

In the light of these constrains of land supply, it would be prudent for the government to re-examine other alternatives to enhance land supply.

The missing jigsaw in the land supply strategy

Land Resumption and Land Premium Assessment Policy

A major reason for the government to re-use of the new town resumption model is the avoidance of the “collusion between the government and the business”. With this government mindset continue, the efficient use of existing private land resources would be much restricted.

For an existing piece of private land which is restricted to a certain use under the government lease, a land premium is required to be paid for the change of land use which is permitted under the town planning zoning. This is an important route to meet with the changing needs of Hong Kong. The underlying principle for charging the land premium is for the government to take away all the benefit in land value appreciation arising from the land exchange/lease modification. In other words, a private land owner cannot share part of the gain in value created in the land exchange process. It is unfortunate that the current land premium policy in land exchange cases has been perceived by some commentators as to be beneficial to the private land owners.

In principle, only those land exchange projects which land costs (including the existing use value plus the government assessed land premium) are below the market value of the lands after the change of use would be worthwhile to undertake. In the market, land premium negotiations usually would take long time to come to an agreement and land owners have to choose carefully the right time to commit so as to achieve profitable developments.

The government has introduced the industrial re-vitalisation policy which nil premium is charged if the relevant buildings meet with the stipulated criteria and some impact is noted in the market. If the government is prepared to re-visit its land premium policy and to allow part of the value created in the process to the private market, this should promote more lands to be converted to meet the needs of Hong Kong.

Change of Land Use

Another area requires further consideration would relate to town planning. Has the planning process been serving its function well as a facilitator/promoter of change of use? A recent incident in Tsuen Wan Industrial area would provide some clues: various land owners had objected to the proposed CDA zonings which have been perceived to facilitate the government lands to be redeveloped into residential use. How then would the government expect that such zonings would be implemented by the private owners?

Given the many challenges including resumption, planning and infrastructure provisions that need to overcome, the outlook of the land supply would still be uncertain.

The use of existing private land resources have been restricted by such factors as: mindset of avoidance of the “collusion between the government and the business”, full land premium charged by the government and certain planning intentions not supported by the private owners.

Without this important piece of jigsaw of private lands in the land supply map, the development land creation exercise of Hong Kong would have to rely upon the government effort alone.



Appendix –Residential supply (2011 to 2020) suggested by Ms. Carrie Lam

2011 to 2015   No. of units
Application List Land in application list not yet sold 13,500
Kai Tak new development area 8,0001
West Rail Line Tsuen Wan 5 (Bayside) 2,384
Tsuen Wan 6 894
Long Ping Station (North) development 832
Long Ping Station (South) development 720
MTR Tin Shui Wai LRT Terminus development 1,600
LOHAS Park (Phase 4) 1,800
URA 5,000
Lease modification / Land Exchange 15,000
Private redevelopment 10,000
  Sub-total 59,730

2016 to 2020

Application List Kai Tak new development area 8,0001
North East NT new NDAs 30,000
Anderson Road Quarry development 8,000
Lamma Island Quarry development 2,000
Land zoned as Industrial use TBC
Land zoned as GIC use 5,000
Green Belt TBC
  Kam Sheung Road Development & Pat Heung Maintenance Centre Development 8,700
  Siu Ho Wan Depot development 13,0002
URA 5,000
Lease modification / Land Exchange 15,000
Private redevelopment 10,000
  Sub-total 104,700


1 There will be a supply of 16,000 units before 2019.

2 Estimated figures.

  • Land supply of approximately 19,500 units was provided in 2011 to 2012.
  • The above figures exclude future supply from reclamation, development of rock cavern and quarry sites.

[1]Population forecast of Hong Kong made by the Census and Statistics Department in 2010. Recent forecast released on 1 August 2012 for 2039 would be 8.47 million, i.e. – a reduction of some 0.43 million.

[2]The Media Briefing documents of the Reclamation Consultation, Civil Engineering and Development Department and Planning Department dated 4 January 2012.

[3]See P.10 and P.11 of the Media Briefing documents of the Reclamation Consultation.


Research Paper for the HKIS Annual Conference 2012:

HKIS Annual Conference 2012