A Century of Housing Laws and Subsidies

A century ago, South Africa’s first national scheme for public housing finance passed into law. The Housing Act came into effect on 19 August 1920 ‘to provide for loans of public moneys for the construction of dwellings’ (Union of South Africa 1920f: xxiii; all such citations abbreviated to ‘Union’ hereafter). The Act created a fund from which municipalities could borrow to support construction of houses at a lower interest rate than available elsewhere. It thus created a housing subsidy system. The Act passed in the aftermath of the First World War and of the devastating Spanish influenza pandemic of 1918–1919, which caused the deaths of many hundreds of thousands in the country (Phillips 1990).

One hundred years later, in March 2020, the South African Minister of Human Settlements told a housing conference that ‘The Department of Human Settlements is in the process of developing the best possible funding and financing model that will enable the achievement of a vision of sustainable human settlements and improved quality of household life’ (Sisulu 2020: 2). New legislation might result (Vawda 2020). Three weeks after conference, the first phase of countrywide ‘lockdown’ began, to combat the worst pandemic to occur since 1918. A hundred years after the first Housing Act and millions of government subsidized houses later, after over a decade of a slumping economy and rising unemployment, and in the midst of a global viral pandemic, the occasion arises to reflect on this century of public housing laws and practices.

The South Africa of the early twentieth century was not unique in facing deepening housing problems, also addressed in some other countries by new laws and practices. For example, New Zealand’s Workers Dwelling Act of 1905 and Housing Act of 1919 sought but largely failed to resolve the housing question, nevertheless establishing a trajectory of public housing support (Hargreaves et al. 1985). The British ‘Addison Act’ on housing and town planning (1919) laid the foundations of ‘council housing’ which over many decades accommodated up to a third of that country’s people (Malpass 2000). The twentieth century brought related approaches to many parts of the world, including both colonial and post-colonial parts of Africa (Freund 2020). As in other places, thinking back a century involves grappling with vast differences that exist between world and South African conditions in 1920, and those of 2020. The unfortunate commonality between the two dates lies in situations of deprivation, desperation, anxiety and lack of resources for individuals and households to take care of their own shelter needs and desires.

This article seeks to contribute to understanding issues involved in public housing in South Africa and perhaps more generally, by comparison between two eras a century apart. Sources include official reports and similar documents, newspapers, secondary literature and archival material. It begins by reviewing the circumstances of health, housing, economy and politics in the 1918–1920 period and the shaping of the Housing Act. Themes emerge of public health, social control, paternalism, racism, poverty and segregation, but also social democratic ideas. The text very briefly touches developments between the two moments of 1920 and 2020. It considers the circumstances that prevail in the field since the democratic election of 1994. It demonstrates that similar issues continue to confront the housing question—understood here to mean, fundamentally, growth of a large poorly housed population and the challenge especially for governments of what, if anything, to do about that (Engels 1935; Hodkinson 2012; Larsen et al. 2016).

Framing the 1920 Housing Act

South African public housing practice pre-democracy, including the 1920 Act, has often been described as fundamentally racist. Nkambule (2015: 41–42) wrote that ‘Soon after the formation of the Union in 1910, a housing act was legislated … subsequent spatial settlement in urban spaces and specifically housing policies and programmes [were] linked … to the broader geo-political system of racial domination in the country.’ South African history is most certainly one of racial domination and oppression, a crime against humanity in the resolution of the United Nations in 1973. South Africa was already (or always) a divided or ‘plural’ society by the early twentieth century. The 1919 Housing Committee terms of reference called for considering improvement of housing conditions ‘including coloured and native people’ (Union 1920a: 72). The 1920 Act, which flowed from that Committee’s report made only one reference to race, in Section 7 (3), requiring ‘reasonable provision for dwellings for the poorest section of the population including the coloured and native people’ (Union 1920f: xxvi). The history of publicly subsidized housing in the country is nevertheless a history of the use of public funds to divide, segregate and set apart ‘groups’ as categorized by government. This perversion of non-racial rhetoric to racist practice requires interrogation (cf. Parnell & Mabin 1995).

At the time of the Act, parliament was elected almost exclusively by male white citizens, who made up about a tenth of the population of the Union of South Africa created in 1910 from several settler colonies. Electoral politics pitted the governing South African Party with mostly urban support spread across the two large White language groups of Dutch-Afrikaans and English speakers, against the National Party of mostly rural Afrikaans speakers less enthusiastic about the British connection and more concerned about the position of ‘poor Whites’. A much smaller Unionist Party connected a pro-British line with large business interests, whilst a Labour Party with social democratic (and sometimes socialist) ideas drew support from the then substantial White working class: it would in time increasingly favour protecting White workers against Black competition. The constitution largely excluded the 78% Black majority, including those described at the time as ‘natives’ (African descended), ‘Asiatics’ and ‘coloureds’. What would ultimately become the largest party in the country, today the governing African National Congress (ANC), had begun in 1912 as the South African Natives National Congress (SANNC) but failed to gain much purchase in its defence of majority rights. It had no presence in parliament (Davenport and Saunders 2000: 270–275).

The world had been radically affected by the 1914–1918 war and many new institutions and approaches came to the fore. Parliament passed the Reserve Bank Act to create a central bank (Gelb 1989: 48), which is again debated 100 years later, on the very same day as the 1920 Housing Act. ‘In most general textbooks on South African history … the structural and other less visible changes wrought by the war are absent’ (Grundlingh 2014: 167). Agricultural boom, doubling in scale of manufacturing and industrial employment, and inflation all created volatile social conditions, including galloping housing problems particularly in the cities (cf. Union 1919b).

Shortages, rack-renting and dreadful housing conditions in South African cities appeared frequently in the pages of the press (e.g. Rand Daily Mail, hereafter RDM, 28.01.1919, 11.02.1920, 13.04.1920, 08.04.1920). Such reports were not confined to the established media and it is possible to find voices of Black commentators, scarce as the records are. The Abantu-Batho, a critical newspaper established in 1912 and close to the South African Natives National Congress (Limb 2012), reported in July 1918 that ‘The necessaries of life have risen in price, the rent in towns is enormously exorbitant, and the pre-war wages are inadequate to meet the situation’ (Abantu-Batho, ‘A Strong Hand’ 04.07.1918). After the first months of the flu pandemic, the article added in language typical of the time that ‘conditions under which our people live, more especially in Johannesburg, are a disgrace to Christianity and civilization … our people live in crowded yards, the sanitary conditions of which are not conducive to good health. … yet the landlord charges exorbitant rent …’ (Editorial, 08.11.1918).

South Africa became one of the worst flu-affected countries from September 1918. The first major legislative consequence of the war addressed public health, and such responses were closely linked to housing questions. Even before the Influenza Epidemic Commission (appointed on 4 December 1918) reported in February 1919 (Union 1919c), a Public Health Bill had been drafted. It was ‘a typical example of the growing body of interventionist social legislation and administrative law then developing in Europe and the USA’ (Phillips 1990: 204 and footnote 45). The new Department of Public Health came into being on 1 July 1919 (Phillips 1990: 204–206; Union 1919c) and other provisions of the Act commenced in January 1920.

Which Sphere of Government for Housing (and Planning)?

Meanwhile, the housing question began to receive direct attention, due both to the impact of the war in hugely increasing the need for housing and reducing materials supply; and due to the fear that bad housing conditions formed a key factor in the spread of flu. Municipalities simply did not have the financial resources to ‘solve’ the housing problem; when they acted in the field, it was in an ad hoc fashion, in parallel with those in many other places around the world, as in New Zealand, for example (Hargreaves et al. 1985). Before Union in 1910, colonial governments funded various kinds of housebuilding, famously after 1904 plague epidemics in several parts of the country—Red Location in Port Elizabeth’s New Brighton, Ndabeni, in Cape Town and Klipspruit (Pimville), outside Johannesburg in what became an origin of Soweto (Baines 2005: 245; Saunders 1984: 145; Kagan 1978: 25). In a few cases, municipalities also built houses that were occupied by coloured and White municipal employees (e.g. Maitland Garden Village, Cape Town: Union 1920a: 13–14).

National government had ignored housing questions, except for accommodation for railway workers employed by the state-owned South African Railways, which by 1920 had already provided almost 10,000 houses (Pirie 1982: 147). But ‘In the months after the (flu) epidemic, town councils, trade unions, private sector, churches and the Railways jostled with one another to undertake investigations into housing, to make representations and to debate the merits of various schemes’ (Phillips 1990: 223).

In January 1919, the government introduced an ‘Unhealthy Areas Improvement Schemes Bill’ (Union 1919a). Sections 14 and 15 would have required ‘accommodation in suitable dwellings … provided by the local authority or otherwise’ for people displaced by ‘improvement schemes’. The idea was to empower municipalities to intervene where crowded housing appeared to create health risks (an issue that recurs in 2020).Footnote 1 The notion was that a scheme could then be elaborated to ‘improve’ the area concerned and that the municipality could borrow money—presumably privately—to undertake the work (RDM ‘No insanitary areas’ 28.01.2019: 7). The bill was welcomed as a ‘Housing Bill’ in some quarters (cf. RDM 28.01.1919: 5, ‘No insanitary dwellings: Government’s Housing Bill’). But it was strongly opposed by municipalities since it offered no central government financial support.

At the time, the then four provinces possessed the power to regulate municipal government but lacked their own taxation powers. In two of their provincial councils—bodies elected along the same lines as the national parliament’s House of Assembly—draft ordinances aimed at empowering and even compelling municipalities to undertake housing schemes were debated. Pushed by the City of Cape Town, the Cape Province adopted a Municipal (Provision of Homes) Ordinance 23 in 1919, but some months later, the Housing Committee (see below) commented that ‘It appears to us that this ordinance will have no material effect in getting houses erected’ (Union 1920a: 13). The Transvaal provincial government was keen to ‘deal as soon as possible with the question of granting municipal councils powers to undertake housing schemes’.Footnote 2 Holding a majority in the provincial council, the Labour Party introduced a draft ordinance in February 1920. Members from the Unionist Party, described by a Labour member as ‘representing the capitalistic interests which crowded people into houses so that they should get the highest possible rents and reap the greatest rewards’, opposed the measure. One of the Unionists claimed that ‘private enterprise had done well and would continue to do well’, arguing against ‘people of thrift’ being ‘taxed to pay for the less provident’ (RDM 18.02.1920: 7 Housing Bill passed … Provincial Council Debates). But the province lacked the funds to support municipal initiative; whilst some municipalities pursued housing projects, for significant funding they had to await change at national level.

Laissez-Faire Versus Public Intervention: ‘the Duty of the Nation’

The most radical feature of post-war legislation lay in adoption of a much larger role for the state in general, against laissez-faire arguments for leaving matters to markets and private initiative (Phillips 1990: 223–4). A Public Health Bill clause removed before it became law sought to make landlords responsible to maintain ‘healthy’ rented premises. The heated opposition to this ‘socialist’ regulation came particularly from a Johannesburg member of parliament, Louw Geldenhuys (landowner and subsequent developer of most of northwestern Johannesburg). By contrast, the strongest voices calling for public action came from the Labour Party. Usually noted for its campaigns around the interests of White workers, those politics enfolded a notion that could more generally be termed social democracy in common with parties working through elected parliaments elsewhere, pushing for improved social welfare and regulation of market or capitalist economies. South Africa’s parliament became an arena for agitation for social democratic programmes on the part of the Labour Party’s members.

Such activity became increasingly focused on the ‘poor White’ issue through the next two decades (cf. Morrell 1992) and caused division in the Labour Party between those prepared to compromise with Afrikaner nationalism and others who wanted a more substantial social democratic position. After its success in the election of 10 March 1920 when its number of parliamentary seats rose from 4 to 21, Labour pushed for, and achieved, the Tenants Protection (Temporary) Act in April 1920 (Union 1920b). That outlawed eviction except on narrow grounds, as long as tenants continued to pay agreed rent, even after a particular lease expired. Similar law provides some protection against outrageous rent increases and can be seen in action today in jurisdictions from New York City to Sydney. Such legislation tends to be effective only in well-regulated circumstances, meaning as less formal settlements expanded in subsequent years, their residents enjoyed little such protection.

Those first regulating provisions were renewed and strengthened by the Rents Act 3 months later (Union 1920d; Maass 2012: 58–65), the first in a long line of Acts seeking to control rents. During the debate on the Rents Bill at the end of March, Morris Kentridge, newly elected Labour member for Fordsburg, a working class area in Johannesburg, ‘was glad that the government was at last alive to the fact that something must be done to deal with this evil … however the government appeared to be very half-hearted … and he hoped (the law) would lay down that the rights of property should not come before human lives’. Completely to the contrary, the Unionist member for the much wealthier constituency of Rondebosch, Cape Town, ‘said that the gist of (Kentridge’s) arguments was the nationalization of houses and property which was merely crying for the moon’. Walter Madeley representing mining town Benoni near Johannesburg said that ‘private enterprise having everywhere broken down in dealing with the shortage of houses, it was the duty of the nation to take it in hand’. His colleague JH Gow representing central Johannesburg added that ‘the only cure lay in a state housing scheme’ (RDM 30.03.1920: 8, ‘Rents Bill in the Assembly’). South Africa’s housing legislation certainly drew on ideologies that could be identified around the world at the time (Freund 2019b).

The Housing Committee 1919–1920

On the passage of the Public Health Act, the government withdrew the Unhealthy Areas Bill (RDM 21.06.1919: 9) and initiated a government investigation into housing questions. A Housing Committee was appointed by the Minister of Public Works (who was also the Minister of Health at the time) ‘to consider and report whether it is advisable for the Government to give financial or other assistance to local authorities and others in providing housing accommodation in urban areas for persons of limited means, including coloured persons and natives, and, if so, the best method of doing so’ (Union 1920a: 7 – hereafter ‘Report’). The government appointed four men (only!) to the committee: Victor Robinson (Chair), retired Natal civil servant; Fred Wiener, Cape Town City councillor; Johannes (not Jan) Smuts, registrar of deeds in Pretoria; and Piercy Eagle, chief architect of the Department of Public Works. A government official, W.J Gyde, became the secretary (RDM, ‘The housing problem. Committee sitting at Cape Town’ 03.09.1919; Eastern Province Herald, 17.09.1919 ‘Government Housing Committee: local sittings commenced’).

The Housing Committee took verbal evidence in private from 218 witnesses, in nine cities. It sent out a letter requesting all 193 municipalities then existing to respond to the questions implied in the terms of reference and ‘about half … sent replies’. Committee members ‘had reference’ to a long list of then available literature on ‘housing and planning questions’, almost all of it British in origin (Report: 29, 72–78). The Committee reported on 19 December 1919.

The tone of the Committee’s Report showed that its members thought of the population as divided into three parts—‘native’, ‘coloured’ and ‘European’ (sometimes adding ‘Asiatic’), though at times all were included as ‘members of the community’. The language of the report is paternalistic, sometimes in the extreme. Yet the Committee expressed concern at the risk that municipalities would continue to neglect decent housing for Black people:

We earnestly hope that interest in the housing problem which has been awakened in South Africa may not exhaust itself in efforts for the housing of the white people only … The interest in the problem must be extended to its existence amongst and its effect upon the coloured, native and Asiatic people who form so large a proportion of the population of our towns, and if the white section of the public is indifferent to the needs of the other section, the authorities have a greater responsibility in the matter. (Report: 36 Para 101)

What influenced the Committee to develop such an emphasis? Witnesses (Report: 69–72) included Dr. Abdullah Abdurahman in Cape Town (long-term president of the African Peoples’ Organization (APO) and both Cape Town City Councillor and Cape Provincial Council member). He seems to have been the only person of colour to appear in Cape Town; in several places where the Committee sat, it heard only White witnesses, some of whom may well have argued against neglect of the majority population. In both Kimberley and Bloemfontein, further APO members appeared; in Pretoria, five members of an ‘Indian Defence Association’ gave evidence. In East London, Dr. Walter Rubusana led a ‘native deputation’ that included eight others. He was a co-founder of the SANNC (later ANC) and a founder of Izwe Labantu newspaper (Ngqongqo 2008: 40–41). Like one of the Bloemfontein witnesses, he chaired the ‘location’ (Black residential area) advisory committee in his city. In the latter city, among others, Selby Msimang appeared: he too was a co-founder of the SANNC who went on to a remarkable political career. He edited the short-lived Messenger/Morumioa at the time, a newspaper that cannot (yet) be traced today (Mkhize 2015: 13).

‘Writing history from an amputated archive is risky’ (Limb 2012) and the difficulties are compounded when seeking to discover and convey voices of those outside the hegemonic classes. We know from the report that the minutes of evidence taken by the Housing Committee were transcribed and typed. Unless those minutes can be located in the archives, where they have thus far proved elusive, it can be imagined that Msimang’s contribution to the Housing Committee might have elaborated on what he told the Select Committee on Native Affairs 4 years later: ‘In regard to your question whether we want the Government to give the natives an opportunity to buy or whether we want the Government to give land free to the natives, I think we want both. The position is that to-day there are natives who are not in a position to pay …’ (Union 1923: 115). It seems likely that figures such as Rubusana and Msimang would have had much to say, and they probably influenced the concern that the report shows for Black citizens’ housing.

The Committee argued for the necessity of a public role in addressing the housing question:

As regards houses for the poorer classes of white people, coloured, native and Asiatics, there can be no doubt that, as private enterprise has not in the past met the need, it would be futile to rely on its doing so in the future … means must be devised for making good the shortage of houses within a reasonably short space of time, and It Is only the authorities, local or central, which can be expected to take action. (Report: 38, bold in original)

The Committee’s recommendations focused on creating a system in which municipalities would be key actors in building houses:

We consider that the local authorities have failed, through lack of adequate powers or from other reasons, to deal with the housing question in the towns, and the conditions … are … such as to justify and demand intervention and action on the part of the Central Government … the Committee considers that the local authorities should be empowered to act within defined limits in the matter of housing and that the Government should take power to compel action on the part of local authorities in certain eventualities to provide or secure proper housing for the people’. (Report: 40)

To equip government to do so, the Committee recommended establishing a ‘housing commission’ (43). It also called for ‘town planning principles and practice to be applied’, meaning good layout and urban design in later terms—something still present in the Housing Code today (59). And the Committee concluded: ‘It is practically impossible to dissociate the subjects of housing of the poorest classes from the question of a minimum wage …’ (p. 56 para 203).

The Committee split on one major issue—that of subsidizing intervention. The majority recommended against subsidizing housing, supporting this anti-interventionist argument with the equally anti-urban notion that ‘It would certainly not be wise or proper for the State or the local authority to provide yet another attraction (to leave rural areas for cities and towns) in the form of subsidised rents in town’ (p. 56 para 199). ‘Subsidy’ at the time meant that public finances would be used to keep rentals for houses built by municipalities lower than they would otherwise be. A minority report, signed by Wiener and Smuts, recommended the opposite. ‘We think the Government should consider whether it can equitably enforce … compulsion without consenting to share a possible loss … We therefore think that the Government may well consider subsidies to local authorities who provide housing accommodation … (Report: 65–66).

In the event, the government—dependent on Labour Party support—adopted the minority view in the Housing Bill introduced to parliament in June 1920. The means to subsidy lay in the low rate of interest to be charged by government on housing loans to municipalities.

Debate and Adoption of the Housing Act

The Rand Daily Mail described the Housing Bill (Union 1920c) as ‘a complex measure somewhat heavy to assimilate in its entirety’ (this remark and the subsequent paragraphs are drawn from RDM 18.06.1920: 4 ‘Economic topics in the Assembly’; 31.07.1920: 7 ‘Housing Bill in Committee’; and 03.08.1920: 5 ‘Nationalists and the Housing Bill’). ‘The first concern’, said Sir Thomas Watt, the responsible minister, ‘was to provide decent houses so that people might get shelter and live in some degree of comfort’.

In the circumstances of the time corporate interests particularly in mining and commerce dominated the country, although with an orientation towards state support for various economic activities (cf. Freund 2019a: 27–33). In that context, the minister’s speech reads as a radical proposal. Watt argued ‘no government in these times could sit still with folded hands and trust to the law of supply and demand’. Addressing the geography of housing, the minister used words that ring loudly today: ‘Houses for working people must be within a reasonable distance of their work … providing houses in a distant suburb, involving expense and also considerable time in travelling between the home and the working place, (is) not in favour …’. The Bill tried to address that problem by the ‘absolute necessity’ of ‘empowering local authorities to acquire land for the purpose of providing houses by compulsory expropriation’.

Emphasizing the extent to which laissez-faire had been overruled, the Bill also gave powers to the government to ‘issue orders that no buildings other than dwellings might be erected in any town or area where there was a scarcity …’. Control of commerce also entered the Bill: ‘A large quantity of cement was made in this country … but a considerable proportion of it was exported … and it was desirable that this should stop’. The Labour Party supported the Bill but wanted funding to be generous, its member JH Gow indicating that an initial half million pounds for a first year of housing subsidies was ‘very much too little’.

Opposition to the Bill arose on several issues, one being the central government role. Watt tried to deflect that: ‘it might appear at first sign that the question of housing ought to be left to the Provinces, but further consideration showed that the matter was one of widespread and national importance which could only be dealt with satisfactorily if parliament voted the necessary moneys …’. When the Bill reached its later stages in parliament, the Nationalists focused their opposition on the proposed subsidy being limited to urban areas. The future apartheid prime minister (1948–1954), Dr. DF Malan and Mr. APY Fourie, demanded ‘preference should be given to the rural areas, where houses must be built for people who would otherwise flock to the towns’. A concession made by the minister was to accept an amendment proposing that beneficiaries of funds under the future Act should not own a house—a critical element of ‘means testing’ that survived long after the 1920 Act.

On the critical race question, the minister argued ‘Since the want of shelter is particularly grievous among the location natives, any municipality that asks for a housing loan for Europeans will be required to allocate a percentage of the money for the construction of native dwelling should there be a deficiency of these within its area’ (RDM 18.06.1920: 4 ‘… Economic …’). The stipulation was included in the Act.

After discussion and minor amendment in a select committee of the House of Assembly (Union 1920e), the Bill passed successfully, just ahead of the end of the parliamentary session. The Governor-General signed on 16 August and the Housing Act was published on 19 August 1920 (Union 1920f). ‘The Union Government had thus taken a first positive step aimed at providing funds so that subsidised housing projects could be undertaken’ (Kagan 1978: 82). The Black press welcomed the first projects to be funded under the Act’s new system: thus Umteteli wa Bantu—not the most radical of such papers but the self-styled ‘mouthpiece of the native people’—gave muted praise on Christmas Day 1920. ‘We are pleased to learn that some hundreds of respectable houses will be built in the near future for respectable natives’ (25.12.1920 ‘Native Housing’ page 7).

The 1920 Housing Act, an indirect result of the flu pandemic but made urgent by a general housing crisis that went far beyond health matters, raised many points of debate and contest: public intervention, subsidy and roles of different spheres of government. Some issues were left unresolved including where to build subsidized housing—rural or urban, peripheral or central, and thus the unresolved issue linking housing provision to spatial planning. Major themes raised but not disposed of during enquiry, debate and legislation were rental versus ownership, and the underlying matter of incomes, touched on by the Housing Committee in 1919. Most obviously, the Act did not resolve the major question of racial segregation. What it did accomplish was to initiate a century of housing subsidy.

After the 1920 Act: Housing, Segregation and Apartheid

The Central Housing Board setup under the 1920 Act started work rapidly (cf. Union 1921). Annual reports of the Board reflect reluctance of municipalities to embark on schemes, perhaps mostly for fear of incurring losses; tensions between provinces and municipalities; and rejection on design and layout grounds by the Board. But shortage of funds formed the major limit, with the Board complaining in 1922 that the Housing Committee’s 1919 estimate that 7 million pounds would ‘make good the house shortage’ called for much more than the 1.75 million thus far made available by government (Union 1922b: 10).

In its first few months, the Board rejected proposals from several cities, in at least two cases because they addressed only their White populations. For example, in the case of East London the Board commented, after rejecting the city’s proposal, ‘it is understood that the preparation of a new and more comprehensive scheme is in hand, which will relieve not only the wants of the European population but will make some provision for the natives, who are in urgent need of ampler and better accommodation’ (Union 1921: 2). The Board provided funds for housing across the colour spectrum but it did so as Christopher (1988) suggested, for schemes almost without exception reserved for particular ‘races’ (a porous notion, 30 years before the Population Registration Act of 1950 tried to make definition systematic: cf. Breckenridge 2014).

The contradictions of segregation continued. After years of argument (cf. Union 1918, 1922a) and the eventual passage of the Natives (Urban Areas) Act in 1923, the Board reported that ‘The position at present is that there are no funds available under the Housing Act for Native, as distinct from Coloured, housing. The question of further funds for Native housing was advisedly held in abeyance pending the coming into operation of the … Urban Areas Act’ (Union 1925: 35). Having called repeatedly for action on housing and disappointed by the lack of progress, Umteteli wa Bantu editorialized in November 1921 ‘The sooner the Native Housing Problem is seriously taken in hand the better will it be not only for the Blacks but also for the Europeans’ (05.11.1921: 2). After the 1923 Act, the paper (07.07.1923: 2) regretted that freehold tenure had been denied by the Act—but indicated that ‘As a housing measure the Act has wonderful possibilities’. Funds did increase significantly over the next few years, partly because the Act compelled municipalities to address housing matters at least to some extent for their African populations, many more proposals were made and the government voted more funds. Approval of schemes became ‘a matter of mutual arrangement between the Native Affairs Department, Provincial Administrations and the Board (Union 1925: 36). As Parnell (2002) argued in relation to Johannesburg, deeper segregation of towns and cities accelerated in consequence.

Some municipalities prioritized projects intended for disenfranchised people. Butler (1985, 2017: 123–141) provided a rare account of how one smaller municipality acted in relation to housing schemes and then-new subsidies during the 1920s. Cradock, later the base of significant anti-apartheid politics, was also home to many liberals in the period. The Town Council’s first schemes were to build houses for African and coloured residents whom it did not always separate—but soon found itself forced to do so. Once national government shifted to the right when the National Party formed a ‘Pact’ government with the segregationist wing of the Labour Party in 1924, and especially after DF Malan became the minister responsible for housing, stricter segregation as well as emphasis on support for ‘poor Whites’ grew rapidly.

By 1940 after two decades of subsidies under the 1920 Act, over 25,000 units had been completed by municipalities with a further 14,000 under construction or approved: 28,000 intended for ‘non-European occupation’ and almost 11,000 ‘European’ (Union 1941: 54; cf. Nathan and Thornton 1935). Over decades that followed, municipalities supplied hundreds of thousands of houses on the basis of the system created by the 1920 Act. Relative to the fast-growing urban population, the pace of building was slow; in 1936 as South Africa emerged from the great depression of that time, negotiations between government and municipalities resulted in more favourable terms being offered to the latter, and that ‘gave a fillip to the provision of housing for the poor’ (Union 1944: 23). Work by Evans (1997), Hendler (1992), Huchzermeyer and Karam (2016), Mabin and Smit (1997), Morris (1981), Ndinda et al. (2011), Nkambule (2015), Okem et al. (2020), Parnell (1989, 1992, 19932002), Pigott (1985) and Rust and Rubenstein (1996) help to bring the account up to the recent past. These (and more) make required reading for anyone seeking to grasp how the shelter situation in the country came to take its present forms. At the same time, this literature leaves many gaps for further and deeper research.

Governments amended the 1920 Act many times over the years.Footnote 3 Events such as great depression, rural expulsion, massive growth of urban populations and South Africa’s participation in the Second World War over the period until 1945 radically increased the need for housing. The public housing system changed significantly in the 1940s and even more so after the advent of apartheid in the 1950s. Where the lines of apartheid would be drawn had to be worked out and the scale of division would be contested, there could be no doubt that the plan was to accomplish a strict and deeply oppressive division of society: comprehensive and centrally directed segregation (Mabin 1992), in the interests of maintaining, even if ‘modernizing’, racial domination (cf. Adam 1971).

The consequences for publicly subsidized housing were both radical and in part, perhaps, surprising: much more strict removal of areas seen by government as offensive, especially mixed ones, but also very much more housing, which in turn depended on continuing, expanding and adapting the subsidy system, as described in general by Pigott (1985). That produced public housing for all ‘groups’, in particular the Black townships, which still form such a major feature of the landscape today. Freund (2019a: 12) wrote that ‘their construction was a massive response to the needs of ‘modernisation’, of rapid economic growth in which heavy industry was taking an important place. In this sense, they could be aligned with what [took place] in Britain and particularly France simultaneously. For instance, they took advantage of the availability of cheap - often otherwise undesirable - land on the outskirts of cities as in France. However in several ways they were very distinctive. Thus the lack of integration into the broader urban integument was not just tolerated but a goal of policy’. From 1948 until 1990, the regime turned the housing subsidy system to the implementation of apartheid—funding perhaps a million houses over the period.

Recommodification, Markets and a Post-Apartheid Subsidy System

Struggle against the system accelerated from the strikes of 1973 and township revolts of 1976 and after. During the ‘late apartheid’ in the 1980s, the chief housing response of the state was to deregulate and privatize—involving the strange institutions of the time (Tricameral parliament, Black local authorities, etc.) as well as ‘recommodification’ of older rented units (Mabin and Parnell 1983) and an official policy of ‘self-help housing’ (Mabin 1993: 336; see also Republic of South Africa 1977). In contrast to earlier policy, the government now promoted its primary role as that of facilitator in the creation of favourable conditions and opportunities for private sector participation (Ruggunan 2016). New areas begun as site-and-service schemes expanded rapidly to their huge form today, such as Motherwell, Orange Farm, Khayelitsha and Diepsloot (Harber 2011). Meanwhile informal practices such as ‘backyard rental’ grew rapidly (cf. Gilbert et al. 1996).

The Independent Development Trust (IDT), set up by then president De Klerk’s government in the wake of his change of direction in February 1990, established a ‘capital subsidy scheme’ (CSS). The CSS aimed to create tens of thousands of serviced sites both in un-serviced informal areas and on greenfield sites (Nuttall 1997). The approach and limits of CSS fed into the National Housing Forum (NHF) as negotiations towards a democratic future began. There, the idea of subsidized individual title gained consensus across a wide range of interested parties (Rust and Rubenstein 1996). A new housing subsidy system added ‘top structure’ (a small house) and an income means test, at first made possible by interim Acts (Republic of South Africa 1993, 1996). The White Paper on housing published in 1994 ‘confirm(ed) Government’s commitment to the end user capital subsidy programme and to expanding this programme to cater for a wide variety of delivery approaches’ (Department of Housing 1994). In part a response to deprivation of property rights for the majority of the population in the apartheid past, public subsidy now supported a gift from the state to millions of individuals and households.

The 1997 Housing Act ‘abolishes all apartheid housing legislation and replaces it with a single Housing Act which reflects the priorities of the current government’ and remains the basis upon which publicly subsidized shelter continues to be built. Its aim is to ‘progressively realise the right to access adequate housing, as set out in section 26 of the Constitution, 1996’ (Republic of South Africa 1997). It obliged the Minister of Housing to phase out the housing subsidies of the previous dispensation within a year.Footnote 4

Responding to criticisms and implementation difficulties, the Department of Housing initiated a process that resulted in revised policy named ‘Breaking New Ground’ in 2004. That resulted in a multiplicity of programmes including supporting tens of thousands of rental units subsidized in various ways (Scholtz 2014), but according to Huchzermeyer and Karam (2016), fundamental direction did not change. Nor has government moved to protect tenants in the way that earlier rent control did, in any event difficult to apply under burgeoning informal circumstances. This is the framework that exists for public support for housing in 2020, a century after the first Housing Act.

Much scholarly and professional critique of government policy and practice through subsidized housing developed since the Housing Act of 1997 (e.g. Bond and Tait 1997; Charlton and Kihato 2006; Huchzermeyer and Karam 2016; Myeni and Okem 2020a; Newton and Schuermans 2013; Turok 2016). Criticism often focuses on failure to fulfil the promise of the ‘right to housing’ inscribed in Section 26 of the constitution (Tissington 2010, 2011; Wilson et al. 2015; Dugard et al. 2016). The geographical position of most of the housing produced with subsidy over a quarter of a century also attracts concern (e.g. Myeni and Okem 2020b). Strains between different spheres or scales of government, national, provincial and local, often emerge (Khan and Thring 2003). Insistence on the power to expropriate land, present in the 1920 legislation, remains: but agitation for ‘expropriation without compensation’ has grown in force, partly because ‘if the priority was for cheap housing built quickly’, the use of cheap, peripheral land ‘was inevitable, despite the Reconstruction and Development Programme call (1994) for convenient location’. In addition, the system of housing as ‘gift’—a fairly extreme form of subsidy as opposed to an expectation of rentals—creates ‘opportunities for private sector benefits from the unending construction activity with its massive potential for state patronage’ (Freund 2019b: 18). Thus one significant shack dwellers’ organization suggests ‘The housing system is corrupted at every level from who gets the tender to build the houses to who provides security on site to how the houses are allocated’ (Abahlali base Mjondolo 2020).

Criticism is on occasion tempered by recognition of the enormity of the issue, the actual experiences of beneficiaries and the difficulties of implementation: for example, Charlton (2013: 377–8) ‘questions a view of these sorts of housing developments … that sees them in predominantly negative terms, without acknowledging what role they do play for households, despite their limitations’. A large-scale evaluation conducted by the Human Sciences Research Council in 2018 makes similar points (Department of Planning, Monitoring and Evaluation, and Department of Human Settlements 2018: 414). Some argue that ‘the government’s preoccupation with low-density ownership has been at the expense of expanding access to affordable rental accommodation’ (Dugard et al. 2016: 37). That reasoning chimes with calls for more attention to support for rental housing (cf. Turok and Borel-Saladin 2016, Turok and Scheba 2018, Scheba and Turok 2020, Department of Planning, Monitoring and Evaluation, and Department of Human Settlements 2018: 453).

Themes and contests observed around the 1920 Act and its genesis continue to circulate a century later. Policy and practice ‘still prioritises fully subsidised, low-density, detached, freehold family accommodation over other delivery modes, tenure systems and accommodation choices … this is not a justifiable response to South Africa’s diverse and changing demographic composition’ (Tissington 2010: 44), something that some argue reproduces historical exclusion. Planning and housing remain far from closely integrated (cf. Newton and Schuermans 2013), perpetuating something observed over a century. Whilst there are fears that ‘the current policy approach will continue to disguise capitalist agendas, which will benefit the rich suppliers of housing at the expense of the poor unhoused citizenry’, such concern coexists with anxiety that housing subsidy ‘simultaneously perpetuates a continued over-reliance on the welfarist developmental state – rather than alleviating poverty’ (Ojo-Aromokudu et al. 2020: 75, 61). There is a reproduction of anti-urbanism, seen when some claim ‘too many people are flocking into cities’ (Myeni and Okem 2020c: 48). The use of housing to control migration is still a spectre that haunts the country (Bank 2015; Bank and Hart 2019). A cabinet minister told the Human Settlements conference in 2020: ‘As we build new cities, we must focus on areas such as the OR Tambo district in the Transkei and locate opportunities and smart cities there. This will curb the rural to urban migration trends’ (Dlamini-Zuma 2020).

Solving the Housing Question?

Over the past century, the notion that with some adjustment to practice the housing question will be soluble has proved resilient—but results have always proved to fall far short of the hope. Regulations proclaimed under the 1920 Act anticipated ending the subsidy system when the ‘purposes for which the Housing Loans Funds were created, have been fulfilled’ so that the government could ‘fix a date, after which no further advances shall be made … to local authorities’ (Union 1920g). In the 1930s, the Durban ‘City Council and other departments persisted in their belief that all shackdwellers could eventually be housed in formal, controlled accommodation. In 1964 they began to admit their failure to achieve this goal’ (Maylam 1983: 425). And much more recently, a critic has written ‘Arguably, most housing practitioners, NGOs, academics, lawyers and activists would agree that the housing policy in South Africa could deliver on the constitutional vision of housing provision if interpreted and implemented correctly’ (Tissington 2010: 6). The housing subsidy system invented in 1920—which improved shelter for some—failed to solve the housing question and largely funded segregation of South Africa’s urban populations.

Yet such outcomes can unfortunately be globalized. Potts (2020) suggests that the housing question is insoluble when separated from inequality and poverty. She noted that the gap between housing costs and incomes is worsening, probably worldwide: ‘OECD data also show that in 34 of 39 countries average (share of) expenditure by all households on all housing costs increased between 2000 and 2013.’ She goes on to say that ‘By (re)presenting the issue as one of incomes, the nature of the struggle and possible solutions come into sharper focus. The primacy of market forces is not denied – indeed, it is emphasised – but in relation to the determination of incomes. It is these which drive the sharp end of problems of housing affordability in all cities. … This point … has important implications for our understanding of the nature of the contemporary urban housing dilemma’ (Potts 2020: 249).

In the conditions of the 2020 virus pandemic, histories of the influenza (flu) pandemic that devastated many parts of the world a century ago resurfaced (cf. Gevisser 2020). ‘There are critical themes … that cut across each of the historical moments … they may help shape our attention in productive ways (such as) how the pandemic … intersected with local governance, politics, economy, environment and social organisation’ (Harrison 2020: 80 ts.). Certainly there were then and are now optimistic hopes for radical changes: thus ‘The epidemic has done more than years of agitation could have accomplished. Never had there been a greater certainty of securing popular approval and sanction for practicable improvements …’ (editorial in Diamond Fields Advertiser 28.10.1918, cited in Phillips 1990: 215). The 2020 pandemic also appears to provide opportunities to alter arrangements of local governance and other forms of organization. For example, a new ‘district development model’ was reported as ‘Covid-19 could be used to change local governance’ (Department of Cooperative Governance and Traditional Affairs 2020). What will come of such initiatives remains to be seen.

Many voices, pens and keyboards call in 2020 for fundamental changes in all sorts of features of contemporary society, internationally as well as locally (captured e.g. in Therborn 2020). No doubt the 2020 pandemic will stimulate more discussion and perhaps policy on ‘the right to decent income’ in addition to ‘basic income’ (Wispelaere and Morales 2016; for an early example during the pandemic see Dawson and Fouksman 2020). Perhaps this is where there is a ‘need for more realistic debate’ (Royston and Wilson 2014). As speculation and contest about post-pandemic futures rages, such a shift seems possible to many, welcome to some and threatening to others. Yet the role of the state in housing that emerges in the short term seems unlikely to mean ‘the end of the housing question’ whilst the fundamentals of capitalist market society continue (cf. Hodkinson 2012: 441).

The Urban Forum has carried many articles on housing subjects since it began over 30 years ago (cf. Visser and Rogerson 2014: 2, 3, 5 and citations in the present article). Most focused on immediate circumstances, and few reported historical research. By contrast, this article concentrates on events, actors, ideas and politics one hundred years ago and suggests that whilst there is no simple ‘learning from history’, ‘judicious’ contemplation of the past (Harrison 2020) can stimulate thinking about the housing question today. Myeni and Okem (2020b): 7 write ‘it is of utmost importance to reflect on the past … in order for the democratic state to creatively engage in the restructuring of human settlement patterns in the post-apartheid era’. Jacob Dlamini demonstrates that artefacts, ideas and archives from the past can be reinterpreted to show the unexpected, challenging what conventional thinking suggests (Dlamini 2020a). He argues that ‘we need new ways to think about the past’ (Dlamini 2020b). This article seeks to open some approaches to doing that, at least in relation to the housing question.