Auctioning off public land amid Cape Town’s housing crisis and land scarcity: What gives?

The forthcoming auction of 50 City of Cape Town-owned pieces of land on 26 February 2026 has created a public furore, attracted significant media attention and is the subject of a number of attempts to prevent it from going ahead. Incidentally, the properties actually total 47, not 50, and add up to about 282,000m² of land (more than 28 hectares or 40 soccer fields).

The City previously auctioned 44 properties on 2 October 2024 (33 of which were zoned for, or suitable for, residential use) and a further 13 on 29 May 2025 (11 zoned for, or suitable, for residential use). That’s 107 properties in just 16 months. Heritage concerns were also raised before the October 2024 auction regarding properties in Constantia which a civic association alleged were subject to land claims. The City hasn’t publicly released the revenue figures generated from these auctions, but it is likely to be in the billions.

James Vos, the City’s mayoral committee member for economic growth, claimed before the 2024 auction that the revenue generated would be used for service delivery, “ensuring lasting benefits for our communities”, and that “this approach maximises underused land, driving long-term economic sustainability”. He also claimed in the same article that “‘no government, including the City, should be a Land Bank”. However, the City hasn’t provided any proof that the revenue generated from that auction was in fact used for service delivery, so the public is forced to take it at its word.

Before the current February 2026 auction, Vos repeated similar claims about the benefits of auctioning City-owned land, noting that “municipal land release is a driver of jobs, investment, housing delivery and revenue that can be reinvested into City services. Government assets should work for the people, not sit idle.” He also explained that in the 2024/25 financial year, the City generated more than R262-million in revenue through the transfer of 70 City-owned properties to new owners. In addition to these sales, the City earned about R334-million in annual lease revenue from its property portfolio, which includes 233 new leases established during that same period. He claimed that “in this way, the City is lessening the burden on ratepayers by diversifying revenue… and delivering long-term value.” We’ll interrogate these claims, but first let’s turn to the broader context.

Amid a housing crisis, where land in Cape Town is scarce and the need for developing affordable housing, particularly in well-located areas, is very high, the City should not be selling scarce public land. As a housing rights-focused organisation and law centre, Ndifuna Ukwazi regularly represents clients who are under threat of being rendered homeless and displaced via eviction proceedings. When the City is joined to these court proceedings because of its legal obligation to provide alternative accommodation to the evictees, it consistently claims that it has no suitable land available to permit recipients of housing kits (wood and corrugated iron) to self-build structures, or for other temporary emergency accommodation. However, at the same time the City is selling off public land that would be suitable for these purposes.

The City states that all properties identified for the forthcoming auction have been thoroughly assessed and are not required for municipal purposes. Many have questioned the apparent lack of public participation around the release of these sites. It turns out the City is relying on public participation processes and in-principle approvals conducted as far back as 2019. While the City may claim that all processes have been followed according to the relevant legislation, specifically the Municipal Finance Management Act, Municipal Asset Transfer Regulations and the City’s Management of Certain of the City’s Immovable Property Policy; the public can rightly feel aggrieved that public input on land release is mostly treated as a tick-box exercise, with a lack of meaningful public input on such a contentious issue as the sale of public assets, which are ultimately owned by all the city’s residents, the City being only the custodian.

Why did the City decide that the properties on auction are not required? In the absence of publicly accessible documentation, it can be assumed that the properties zoned for, or suitable for, residential use if rezoned (23 of the 47 erven, or about half) are not viewed by the City as currently suitable for large-scale BNG (breaking new ground) public housing projects or cross-subsidised affordable housing development (publicly funded social housing cross-subsidised with market-rate housing, as is required in the current declining social housing funding environment), as they are either too small (generally under 1,000m2) or are not well located close to the city centre or any other economic nodes. It should be noted that it appears that none of the residential-zoned sites is being released with conditions obligating private developers to include affordable housing on-site – a dire missed opportunity in a city facing a housing crisis.

However, five of the well-located sites show the lack of innovation in current City thinking. A large 25,000m² erf at 55 Century Boulevard in Century City, currently zoned for business use, could easily be rezoned to permit a mixed-use, residentially led development of affordable and market-rate housing, since it is well located close to schools, public facilities, public transport and job opportunities. The opportunity to bring affordable housing to Century City should not be missed. A 5,874m² erf at 40 Koeberg Rd, Durbanville, already zoned for residential use, could bring much-needed affordable housing to Durbanville.

The combined 791m² site at 138 Upper Canterbury Street, Gardens, and the 990m² site at 13 Paarl Street, Goodwood, both present opportunities for an innovative new model we propose for cross-subsidised social housing, where multiple smaller erven in reasonably close proximity can be released to, developed and managed by the same social housing institution to meet their preference for a minimum of 200 units for feasible development, maintenance and management costs. The Upper Canterbury Street site, ironically identified by the City in its own 2017 affordable housing prospectus as available for future affordable housing development and clearly reprioritised without public input or notice, is only 150m from the “Fruit and Veg” site which was finally released after eight years for mixed social and market housing development late last year, but which will likely still not be viable as a standalone site. The Paarl Street site is 450m from Vasco Station and about 1km from the existing Goodwood Station social housing development. Additionally, a 492m² site at 36 Picton Street is just more than 200m from Parow Station, where there is Prasa land suitable for a similar development to that at Goodwood Station, as well as close to other City-owned land.

The balance of the 47 erven are in industrial areas or not in well-located areas, mostly in various parts of the Cape Flats. Those that are not well located, but not in industrial areas, would not be able to attract public housing subsidies due to the current funding environment. Without state subsidisation for BNG housing (this funding has been reduced to minimum) the land would only be viable for self-build housing (via the nationally funded Enhanced People’s Housing Process or EPHP programme). The EPHP programme, introduced in 2008, empowers beneficiaries to actively participate in decision-making, design and construction of their own homes. It focuses on community involvement to improve housing quality, promote skills development and ensure sustainable, community-led development rather than just providing a finished house. However, this programme is currently highly undercapacitated and underfunded. So, in the current context, even if the pieces of land located across the Cape Flats were released to community beneficiaries, they would not receive state housing subsidy funding and support, and would have to rely on savings and cooperative and private funding. Communities, social movements and NGOs regularly call for public land to be released to community members, cooperatives and micro-developers, but the City appears not to consider this in its evaluations, and neither do other metros.

The social value of public land needs to become a stronger principle in legislation and policies governing public land release in practice. The national government should be prioritising social housing by increasing funding from the current 2% of the human settlements budget allocation, and sorting out governance issues at the Social Housing Regulatory Authority. It should also increase funding and support for self-build housing via the EPHP programme and potentially new forms of self-build support in the upcoming Human Settlements Act and Code. This would make these smaller pieces of land viable for social housing and for release to communities for self-build housing.

However, there are also broader questions at play here. Why should the City even be selling public land? The City, as highlighted earlier by Vos, believes that public land that is underutilised should be maximised by auctioning it off. The sale of the land can then, so the argument goes, generate jobs, investment, housing delivery and revenue that can be used for service delivery. He neglects to mention that selling public land is not the only way to generate revenue and “sweat” public land assets. The state, and cities in particular, here and elsewhere in the world, have always owned land to be used for public benefit. So no, alderman Vos, governments have been banking and managing land for many years.

Long-term leasing of public land, typically for 50 to 99 years, is a long-established practice globally, used as a strategic tool for urban development, infrastructure financing and agricultural tenure. In Hong Kong, revenue from land leases has historically financed a significant portion of annual public infrastructure investment – Cape Town has soaring land prices and could learn from this example. Canberra, Australia, is another example of a city with a long history of leasehold. There is precedent for long-term leasing in South Africa. The state frequently uses long-term leases (often 99 years) for agricultural land and large-scale renewable energy projects, while private precedent includes the 2,200-hectare mixed-use Waterfall City development in Johannesburg, where all land belongs to the Waterfall Islamic Institute and for religious reasons the property cannot be sold.

Municipalities routinely lease land for residential, business and social purposes. Ironically, the City of Cape Town auctioned off a 99-year lease for a 25,000m2 bulk Tower Block development next to the Cape Town International Convention Centre in April 2024. Vos said at the time that the aim of the lease auction was to “unlock diverse business opportunities and generate valuable construction jobs” and “bolster the city’s economy with massive socioeconomic spin-offs contributing to the metro’s GDP, creation of employment opportunities and be a productive asset in perpetuity” while also “bolster[ing] the City’s ability to provide basic services to our citizens while helping to reduce unemployment”. Sound familiar?

Long-term leasing of public land can also be a strategic land value capture tool. Land value capture is a public financing mechanism that recovers the “unearned increment” or increased value of land resulting from public investments (for example, infrastructure, zoning changes) rather than private effort. It allows governments to recoup these value increases to fund public services, infrastructure and affordable housing. It is becoming a more prominent tool in South African cities and towns, particularly those that have inclusionary housing/zoning policies such as Johannesburg (since 2019) and Stellenbosch (since 2023) – Cape Town is yet to implement such a policy, despite first investigating it in 2018.

China promotes faster housing development and captures public land value via local government financing vehicles (effectively government-owned development companies). These use public land as collateral to borrow from financial institutions to build public infrastructure, and use that infrastructure to make nearby land more valuable. They then lease that land to developers, and use the lease income to pay back the loans, and the cycle repeats. A well-managed version of this model could be implemented in Cape Town.

Ultimately, long-term leases are a better use of public land without surrendering ownership of the asset that is effectively being held in trust for all citizens. Over time, the total long-term lease revenue is very likely to be greater than the short-term revenue from a sale. As Vos inadvertently highlighted above, the City generated more in lease revenue in the 2024/25 financial year (R334-million) than from property sales (R262-million). And lease revenue is a constant and escalating revenue stream over the term of the lease (up to 99 years), rather than the one-off cash injection of a sale. Once public land is sold it is not only no longer an asset that can generate future revenue (future value belongs to private owners), but the City loses control of the asset for publicly beneficial uses such as housing, community facilities or infrastructure, and limits future opportunities for the land.

There is also the broader question of municipal revenue generation (a pressing need for all municipalities) and how this should be balanced with service delivery. Questions around alternative sources of municipal revenue generation are part of the current review of the White Paper on Local Government driven by the national Department of Cooperative Governance and Traditional Affairs. The question of selling public land is also a value-based decision. Should local government focus purely on providing services and infrastructure, and leave development to the private sector? Or should local government use public assets more wisely for long-term public benefit? Once public land is sold, and short-term revenue is spent, the choices made by local governments today will have consequences into the future.

This auction brings up the need for strategic advocacy directed at structural reform of public land governance, including the promotion of leasehold models, redistribution-oriented disposal frameworks, statutory safeguards ensuring that land sale is a measure of last resort, and the development of standards regarding the validity period of public participation processes.

Ultimately, Ndifuna Ukwazi calls on the City of Cape Town (and other municipalities, levels of government and state-owned enterprises in South Africa) to not sell off state land, but rather make use of long leases, or release land to social housing institutions, affordable housing developers, micro-developers and communities for housing construction at a nominal value of R1.

The City of Cape Town must look for alternative sources of revenue, rather than selling off public assets that ultimately belong to all residents of Cape Town. This proposed auction should serve as a catalyst for broader policy engagement concerning the constitutional governance of public land and the alignment of municipal asset management practices with spatial justice objectives. DM

Jens Horber is a researcher and coordinator at Ndifuna Ukwazi.