The Early Surveys

Clearly, this sort of work required black research assistants, to interpret, to explain to the households in question what was wanted of them, and in general to facilitate the process of collecting the data. With few exceptions, however, little is said in the reporting about these people. A typical example might well be Gertrude Kark’s unpublished but very extensive study of Alexandra, in northern Johannesburg, during the 1940s.^{Footnote 10} In her preface she profusely thanks Miss Ruth G. Cowles, the Nursing Superintendent of the Alexandra Health Centre, and other health workers in the township, and also Miss Rae Bernstein of the Fordsburg Community Centre for methodological help. She further notes that her assistant for half the time was Miss Hessie Margolis (later to become Mrs. J. Sachs) who “showed considerable initiative and enthusiasm in the work.” Thereafter, Kark writes: “My other co-workers were Mr. Paul Kotsi and Mr. Frans N’tsie.” No further discussion of their role is given, beyond the comment that “the work achieved was the outcome of joint endeavour and a harmonious team spirit.” Even this is better than the series of pamphlets on the cost of living of Africans on the Witwatersrand put out by the Institute of Race Relations in the 1950s.^{Footnote 11} These contain no acknowledgements of co-workers, who must have been employed in the work. As so often, the role of the assistants in the co-production of knowledge is shrouded in silence. Nor does the Bureau of Market Research provide any useful information on those who performed the surveys under their direction.^{Footnote 12}

To the best of my knowledge, the first systematic investigation of black household budgets was organized by the redoubtable Mabel Palmer in Durban in the 1920s.^{Footnote 13} This was as part of her work for the Joint Council of that city.^{Footnote 14} There were two investigations. One was done with the help of Mrs. Charles Dube, presumably the wife of the deputy principle of Ohlange Institute near Durban, and thus the sister-in-law of the first president of the ANC, John L. Dube.^{Footnote 15} This study was presented to the South African Association for the Advancement of Science in 1928. The second with Selby Ngcobo, at this stage a teacher at Adams College, who was later to take a PhD in economics at the University of London. In both cases, the material was also provided for the Native Economic Commission of 1930–1932.^{Footnote 16}

In the first set of budgets, Mrs. Palmer and Mrs. Dube gave information on three members of the “native intelligentsia” and a couple of artisans, as well as two unskilled laborers. In general, various individuals’ situations were dominated by the refusal of the Durban city council to allow the settlement of any families except for “a small number of educated detribalised Africans.” Many others were forced to live in what was described as “peri-Durban.” Palmer was using these budgets both to protest at the living conditions which the teachers (both male and female) and the clerks who were interviewed had to live in, and to bring forward some of their grievances, notably the fact that blacks had to pay for their children’s education, while whites, coloreds and Indians did not have this expense. Further of course she echoed the complaints of the amaKholwa elite, with whom she sat on the Joint Council, about the low salaries they received and the difficulty they had in making ends meet.

The studies she did with Ngcobo were more detailed, but they also showed how very few of the black inhabitants of Durban were able to live within their income. Only one family, that of a head teacher with accommodation provided, was able to achieve this. All the others were recorded as having shortfalls, which were primarily coped with by running up debts in the trading stores. Quite how this would have worked in the long term is difficult to understand. Presumably there were often sources of income, in cash or kind, that were not recorded, as the budgets themselves do not give evidence of much expenditure on items which could have been avoided. Indeed, four respondents were also asked what their expenses would have to be if they were to have what they “regarded (…) as a reasonable standard of comfort.” These men were considered, and considered themselves, “civilised natives.” The two garage attendants, a hotel cook, and an engineer’s mate, basically spent 68% of what they would have done in an ideal situation. The difference was primarily in the rent they would have expected to pay, in other words a less congested house, and in the amount they spent on furniture. It was a signal of the clear intent among the more prosperous black inhabitants of Durban to achieve a living standard consistent with their aspirations for respectability, and the difficulties they had in realising these hopes.

At about the same time as Mabel Palmer was working in Durban, the Joint Council of Johannesburg issued a pamphlet with budgets from Africans both in that city and in Cape Town.^{Footnote 17} It differed from Palmer’s budgets primarily because it included lunch at work and transport in town. Like Palmer, the Johannesburgers saw the Africans as being forced to live beyond their means, or at least their salaries. A family would need from £6.1.2 to £6.11.8^{Footnote 18} to survive at a reasonable level, while the wages the father earned were only between £3 and £5. The difference was made up by the wife taking in washing, some juvenile labor, saving on food and on occasion illegal brewing. Similar work was conducted by the Joint Council in Bloemfontein, which is said to have been published in the Social and Industrial Review, for July 1928.^{Footnote 19} In Cape Town, the study, done in the black location of Ndabeni, pointed out that the poverty had led to an inordinately high infant death rate. Also, in 1930, evidence for the cost of living of Africans in Langa was collected by the Assistant town clerk of Wynberg, M.B. Williams, and presented in a court case, Rex v. James Bali, in the magistracy of that district on 27 August 1930.^{Footnote 20}

The major investigation into household budgets in the 1930s came from the American missionary social worker, Ray Phillips. It was initially presented as a Yale thesis in 1937, but was published in South Africa in the next year.^{Footnote 21} Phillips was nothing if not paternalist, and was hated by the more radical among the educated African elite for his attempts to control the expression of African opinion, as they held him responsible for the failure of Abantu Batho, the first paper under black control.^{Footnote 22} He was disdainful of African attempts to build up their own businesses. Perhaps realistically, he concentrated on those who worked for wages, and looked primarily at those Africans who had enjoyed a degree of European-style education, naturally enough the group to which a missionary had most access. From among his contacts, Phillips was able to collect the family budgets of forty-eight families. What is most remarkable in these budgets is the degree to which the various families relied on sources of income outside of wage labor. On average the monthly family income was £13.11.10 while the mean wage of the household head was just over £9, in itself more than double the median wage (£4.6.8) which Phillips reported for 3,654 residents of the Johannesburg locations. The earnings of wives, the renting out of rooms and various forms of petty trade were responsible for the difference.

Phillips was of course concerned to discover what sort of standard of living could be achieved with these sorts of incomes. He did this by making comparisons with the various budgets which had been worked out for the European inhabitants of the City, and also the judgements of various experts, particularly with regard to nutrition. He found that a budget in which only the essentials were covered was still £1.1.6 more than the median wage which the men received. This was without allowing anything for “fuel, lighting, clothing, medical fees, church fees, school fees, furniture, recreation, newspapers, supplementary food, nor anything for additional dependents.”^{Footnote 23} Even the educated, and relatively prosperous, to be found in his sample of forty-eight, were often in debt (though the total savings of these forty-eight more or less equalled their total indebtedness.) Often it was cost of hire purchase on furniture which drove these families into debt, which they could not cover on the basis of their wages. As one of Phillips’s informants told him:

Nonetheless, African wages were held down because they were based, not on the rates paid to the Whites, whose economic standard many were attempting to attain, but rather that of the migrant workers who “still have a rural home on which they and their families partially depend for subsistence.”^{Footnote 25} It was a plea for color-blind wage patterns and for succour to the emerging black middle class, respectable but nevertheless deferent.

In Phillips’s work, and in that of most other researchers on South African household economics, no matter of what sex, there was the presumption that a family should depend on the wages brought in by the pater familias. It was implicitly and unthinkingly considered improper that women’s work should be necessary to keep a family afloat.^{Footnote 26} That urbanization had the effect of re-ordering the gender relations of African society, was not considered, or rather was thought to be yet another of the deleterious effects of city life, and of the low wages paid to African men.

There were other investigations of the city’s poverty. The Johannesburg Political Economy Club launched an investigation into the wages paid to the employees of the Johannesburg Rotary Club, fifty-seven in all, with the exception of domestic servants.^{Footnote 27} In the course of their investigation, they found that they had to exclude two cases from their statistics, because the men in question earned £4 a week each. These two men, a bus driver and the works manager at a printing works, had acquired skills which put them out of the range of the rest of the group. For the others, the mean of those married workers whose families were living in town was £1.15.10 a week, and for those who were either single or whose families were in the country the wage was somewhat less. Clearly the permanently urbanized had been able in some way to get marginally better paid employment. Nevertheless, it was precisely this group which was in fact the poorest, because they were unable to benefit from the rural economy. On their average diet, the comment was made that “the inadequacy (…) will be apparent to any housewife, and it is important to remember that there are many below this average.”^{Footnote 28} The image of poverty and malnutrition was always maintained.

During and After World War II

These concerns were, in the 1940s, equally the concern of officials, as well as academics and campaigners. In 1941, Miriam Janisch published the results of her survey of Johannesburg African budgets, which she had done on behalf of the Municipal Non-European and Native Affairs Department.^{Footnote 29} Her sampling method was, to say the least, somewhat unusual. Her main source of informants for her study in the Western and Eastern townships and in Orlando were those who had fallen behind in their rent, though generally only for one month. She came into contact with them at the municipal offices. These methods clearly implied a downward pressure on her results, both because those who had not paid their rent would be more likely to be in debt than the others, and also because they would be less likely to mention the income they derived from subletting rooms in the location. Indeed only 1.7% of the families’ income came from sub-tenants, an improbably small proportion. Given these deficiencies, she found that the families had an average monthly income of £5.6.8, as against outgoings of £5.11.0, the difference being made up either by illicit means, such as the brewing of beer, or by going into debt. In addition, Janisch’s non-statistical comments provide a valuable addition to her numbers. She described how:

These households were frequently in debt, largely as a result of the hire-purchase of furniture. “It has become part of the urban tradition to possess cheap European furniture. A dining room suite, a double bed, and a wardrobe are almost essential outward and visible signs of an inward and urbanised culture.”^{Footnote 31} Far too often, families fell behind with the payments, and the furniture was repossessed. In addition most women had sewing machines, in part as a source of income. Wages, particularly male wages, were clearly too low to allow this style of life, to the detriment of the economy as a whole, as Janisch cited A.J. Norval, the Acting Chairman of the Board of Trade and Industry, to the effect that “before South Africa is able to progress industrially, it is essential that the native’s consumption of industrial products shall be increased. This can only be done by raising his standard of living.”^{Footnote 32}

Rather surprisingly, given the later history of South Africa, the figures collected by Janisch were used extensively by a Government Commission, that into the Bus Services of what was to become Gauteng.^{Footnote 33} This had been set up in reaction to the mass protests staged in Alexandra, in the northern suburbs of Johannesburg, as a result of the Bus companies raising the fares.^{Footnote 34} Part of the Commission’s remit was to investigate whether “the fares charged are within the means of the passengers using the service.” The Commissioners had Janisch’s information on wages, some of the early attempts by Edward Batson in Cape Town to develop a “Poverty Datum Line” to ascertain the income necessary to sustain a “minimum level of health and decency,”^{Footnote 35} and their own investigations into the changing cost of living on the Witwatersrand, which was steepling out of control as a result of the war. Essentially, the Parliamentary Commission concluded that:

Some increase in the wage rates had occurred, but this did not compensate for the rise in prices. Africans could only survive by eating too little and getting into debt. The Commission did what it could to freeze transport costs, but the report remains remarkable above all for the official recognition of the levels of poverty among urban Africans.

These sorts of comments could not be expected after the National Party’s victory in the 1948 elections. As the apartheid policy began to be implemented, political discourse would no longer allow the sort of comment which the Bus Commission had propagated.^{Footnote 36} Countervailing arguments had to come from non-governmental organizations, which indeed had to be very careful that they did not run foul of the Nationalists’ ideological positions. The most notable of these was the South African Institute for Race Relations, under the chairmanship, until his death in 1953, of J.D. Rheinallt Jones. The Institute was always careful, perhaps too careful, to preserve a reputation for non-partisan factuality, even when its own material made this a difficult path to take.

The first study specifically commissioned by the SAIRR on the standard of living was conducted, by A. Lynn Saffery, in Kroonstad, in the Free State, in 1940. It formed part of the discussions about the introduction of a minimum wage, but was never published (although it does seem to have had a considerable circulation).^{Footnote 37} A more significant report was compiled in Alexandra by Gertrude Kark, working for the Alexandra Health Committee, on the basis of research carried out from 1944 to 1946.^{Footnote 38} This report, too, was not published, which is unfortunate as it is the most detailed and self-conscious of the studies of South African urban conditions. There are two reasons for this. First, there is the question of the accuracy of the work. Mrs. Kark herself wrote that “the difficulties she encountered in securing detailed information, on both income and expenditure” led to the conclusion that family budgets which were compiled in the course of the survey are to a large extent hypothetical and so not permit of deduction or generalization.” On one family she commented that “no expenditure on alcohol recorded, but the family head was habitually drunk,” and spent all the money he could spare from his addiction on Fah-fee gambling.^{Footnote 39} One wonders whether this level of candidness led to this remarkable reports not being published; it might have been seen as undermining the reputation of objectivity which the Institute for Race Relations was trying to build up. Against this, there were comments from two African members of the Board of Management of the Health centre “that the publication of a report of this nature would be harmful to the population of the Township” because “the conditions described are not peculiar to Alexandra Township (which in itself may be true enough) and would simply be used as a political weapon against the Township.”^{Footnote 40} There had in previous years been those who suggested that Alexandra be raised to the ground, and these men did not want to fuel the potential agitation. Mrs. Kark herself disagreed with these comments, but nonetheless, the report languished in the Institute’s files and had little immediate effect on public debate.^{Footnote 41}

Under Apartheid

Through the 1950s the Institute of Race Relations continued to organize and to publish reports on the cost of living of the African inhabitants of Johannesburg. On the basis of these studies, the Institute made regular submissions to Government Commissions and so forth. The message remained the same. Working men’s wages were not sufficient to allow a decent life to be led by the majority of the city’s black families. To cite their conclusions, Ethel Wix in 1950 wrote that:

There had been no significant increase in wages over the past six years, and the rise in the cost-of-living allowance was insufficient to meet the sharp rise in the cost of both food and clothing, especially women’s clothing over the same period.^{Footnote 42} In subsequent years, things seemed to get worse. Olive Gibson, in a follow up study, wrote that: “In 1950, family income in the largest wage-earning classes was estimated at 72.4 per cent of the cost of the minimum essentials. (…) Today on the same basis it is 63.4 per cent.” In particular the rise in the cost of clothing and food was responsible for this. Nevertheless, African homes were always spick and span. “The great majority of Africans have no desire to be tied down to this minimum standard of decency, and are struggling hard to escape its degrading confines. Nothing perhaps more clearly shows this striving than these figures of the consumption of cleaning materials, which is reflected in the cleanly state of nearly all houses in which we unexpectedly intruded to find crisp, fresh linen, shining floors and stoves in the most unpromising areas.” This was a community which had not fallen into poverty, but was in the “upward stage of urban development.” But in order for this to continue, there would have to be an increase in wages, accompanied by systematic efforts to raise productivity – not in itself so difficult, as a healthy and better fed labor force could be expected to work harder.^{Footnote 43}

A third such study, by Joy de Gruchy, was published in 1960. It too recorded that minimum expenditure, which was calculated at £24.3.2. a month over the summer 1958–1959 exceeded the average minimum wage by £9.3.2 and the average family income by £4.13.6, a shortfall in the first instance of 37% and in the latter of 19%. This, moreover, was based on a standard of consumption, particularly of food, which was at minimum or subsistence levels, and could not be considered “adequate.” As she wrote, “the income of at least 50 per cent and possibly 75 per cent of African families is inadequate to meet minimum necessary expenditure. (…) They manage to survive for the reason that they maintain a standard of living far below the theoretical minimum based on western standards: in fact very few indeed, relatively speaking, are able to maintain a standard of living which would be tolerated by the average white South African.” This, De Gruchy commented, was “an indictment of our social and economic system” and also one which at least the Johannesburg Chamber of Commerce deplored, as they hoped that Africans could play their part “as large scale consumers of the products of primary and secondary industry.”^{Footnote 44} It should however also be noted that De Gruchy had depressed the figures for wages by reporting those paid to workers beginning in any industry, and thus did not take into account the increase which a man would receive after working some time for the same employer, a fact which she admitted.

It is notable that a large proportion of these studies were produced by white women. Clearly they were highly educated and dedicated to the work on which they were engaged. Of the main characters who have so far been discussed, Mabel Palmer had had a long career as lecturer in economic history at what was to become Natal University, and after her retirement she remained intellectually and politically active, as a life-long Fabian. Miriam Janisch, a Cambridge graduate, became Assistant Director of Education in Kenya in the early 1950s, a change of career that was not a particularly great step.^{Footnote 45} Ethel Wix too returned to London, where she had grown up, and became a leading City solicitor, and activist on behalf of Palestinian Israelis.^{Footnote 46} Olive Gibson also returned to the UK, and indeed had apparently only been in South Africa while conducting the survey for the SAIRR. Joy de Gruchy, finally, seems to have moved to Brisbane, Australia, with her architect husband, and established herself as a craftswoman and designer.^{Footnote 47} So far as can ascertained, then, they were all professional women who worked for the University, the Johannesburg City Council, or the SAIRR, as part of a developing career. Nevertheless, it may well be that the analysis of household budgets was seen in some circles as a particularly appropriate activity for female researchers, who might have been assumed to be at some stage concern to run their own households.

There was one other location for budget surveys and related work which flourished briefly in the early 1950s. At the time, there was a substantial increase in the black population of, in particular, the East Rand, as a result of the industrialization from the earl y 1930s onwards. This process was speeded up by the Second World War.^{Footnote 48} These workers needed housing. Those town planners, in the broadest sense of the word – administrators, architects, engineers, and so forth – who had a concern with the building of the new Black townships in the East Rand, and elsewhere, wanted to know what number of dwellings in which category were needed. As a result, in the process of planning the building of KwaThema, to the south-west of the town of Springs, the National Building Research Institute of the South African Council for Scientific and Industrial research commissioned a survey among those who would be moved to the new location. This was organized by a “European sociologist,” H.J.J. van Beinum, and the questioners were administered by “four specially trained qualified Native social workers. Again, their names are not given in the report.^{Footnote 49}

The town planners did not acquire as much data on the precise income and expenditure as did those working for the SAIRR. Certain forms of income they ignored. Both the criminal earnings, for instance from illegally brewed beer, and the proceeds from subletting room space, were not taken into account, the latter presumably because such practices would not be allowed in the new township. The main concern was with the structure of the households. They commented that over three-quarters of Payneville households consisted of a nuclear family, perhaps with (illegitimate) grandchildren as well, which was considered to be evidence that the population was fully urbanized. They needed to know how many would be able to pay rent at what level. They therefore divided the households into three categories: the sub-sub-economic, whose income was below a calculated Poverty Datum Line, even if rent was excluded, about 46% of the total; a sub-economic group, who would be below the PDL if it were to include rent (13%); and finally those who could afford to pay rent (40%).^{Footnote 50} There was no clear correlation between the size of the household and its economic categorization, at least given the assumptions and methodology of the study.

On the basis of this sort of data, it was proposed, the city could make informed decisions as to the number and size of the houses it had to build in the new townships. It could also have some idea if, and how quickly, it could recoup its investment. It was not so much social engineering in the normal sense of the word as engineering the environment to fit the society, though with a hefty dose of control involved.^{Footnote 51} It was not a process with much future, however. The Minister for Native Affairs, Hendrik Verwoerd, was not prepared to allow municipalities to make the expenditures that such distinctions in housing quality would entail. This meant that continuing such studies became pointless.^{Footnote 52}

There can be no doubt of the political message of these enquiries, and, at least for those of the SAIRR, their heart was also clearly in the right place, struggling as they were for the uplift of the very many, very poor families on the Witwatersrand, and for some degree of equalization in South African society. However, by concentrating on the average standard of living, they had little eye for the distribution of the statistic.^{Footnote 53} If the incomers of between half and three-quarters of the African families were below some measure of minimum necessary expenditure, then that of between a quarter and a half were above it. The comments of the Johannesburg Chamber of Commerce reflected a growing awareness in the South African business community of the possibilities of the African market – by no means of the same size as the white market, but growing quickly, particularly in the cities of the Transvaal. Drum, for instance, the iconic publication of the short-lived Sophiatown Renaissance, could only exist because the advertisers, who bought about 60% of its page space, could expect it to be bought by people wanting to purchase the medicines, food, clothing and furniture which they propagated.^{Footnote 54} Probably Drum was outsold by Zonk!, a publication with much the same commercial intent, but much less political – it was after all published by the Afrikaanse Pers, who remarkably had seen the market possibilities for such a journal as early as 1949. Indeed, the 1959 Convention of the Society of Advertisers was primarily concerned with the African market in South Africa, and heard speeches by W.W.M. Eiselen, Secretary of the Department of Bantu Administration and Development (and one of the major ideologues of apartheid) and by Nimrod Mkele, an old student of Monica Wilson at the University of Fort Hare who by then had acquired an MA in Psychology and was employed by the advertisers J. Walter Thompson.^{Footnote 55} Mkele indeed quoted from Selling Age, the main journal of the advertising industry, to the effect that:

This was not an argument to the effect that each of the millions of Africans would be spending a little, and that together these very many very littles would add up to a substantial amount.^{Footnote 57} Rather, it was an argument about black prosperity, not about black poverty.

Professionalization of Market Research and the Rise of the Black Middle Class

It was out of this that the study of black incomes became totally institutionalized, and professionalized. This came from the development of market research during the 1950s and 1960s, as advertisers, and behind them of course manufacturers and importers, wanted to know how to penetrate the African market. In 1960, the black share of the country’s income was around 20% of the total, and this was to rise to between 25 and 30% from 1970 onwards. Part of this rise was the consequence of the growing African proportion of the total population, but equally the expansion of the African Middle Class was beginning to matter. By 1975, indeed it was estimated that just over half of the 8^th decile of incomes in the country (and much more for all lower deciles) consisted of African households.^{Footnote 58} From 1960, the reports of the Bureau of Market Research, attached to the University of South Africa, but funded primarily by commercial firms,^{Footnote 59} began to appear. These reports, which initially concerned the population of the major urban centres of the country, were much more systematic in their sampling procedures than those of the SAIRR had ever been, principally because the funds available for the administration of the surveys – and probably also for the compilation of the relevant tables – was much greater. They came to lie at the basis of the global statistical analysis of the incomes of South Africans, leading to the construction its Gini coefficient of income inequality. The information was admittedly not complete. When Charles Simkins attempted to make the first calculations of this Gini coefficient, he was forced to use data from Botswana to estimate the incomes of those living in South Africa’s rural reserves, for want of the relevant South African information.^{Footnote 60}

From the early 1970s onwards, the Bureau of Market Research’s work was complemented by that of the South African Advertising Research Foundation,^{Footnote 61} which was set up by the industry to engage in the sort of market research which was beyond the scope of any individual agency or company. It is well funded, receiving 1% of the advertising industry’s turnover to conduct its surveys, and of course to analyse the results and distribute their findings. Its major concern is to investigate the proportion of the readers of South Africa’s magazines and newspapers, and listeners and watchers of the radio and television outlets who make use of particular types of products, thus enabling advertisers to decide where best to place their messages. In the course of this work, though, a mass of other data is produced much of which is of clear relevance to the estimation of income and expenditure among the various classes, however defined, of the South African population.

In the nature of things, such institutions are probably inherently conservative. Their brief is to provide what they see as neutral, empirical data, and within the limits of that claim they do as well as can be expected. This does not mean, though, that their results are necessarily as neutral as it might appear. In 1963, The Bureau of Market Research published the results of their first survey in Soweto. In its introduction the director, Fritz Rädel, wrote that: “Of the many changes that have occurred in the past few decades, two of the most important are probably the rapid growth of the urban population, especially the urban Bantu population, and the considerable rise in their per capita income.”^{Footnote 62} This was probably an accurate description of the state of affairs, but against this it was not welcome news, either for the apartheid government, committed as it was to restrict black urbanization outside the reserves, or for the nationalist or liberal opposition, which would not have appreciated the comment that black incomes were rising. At the high point of apartheid, in the years immediately after Sharpeville, the world of commerce was pointing out clearly that at least one of the prime goals of government policy was steadily failing, and, at the same time, that apartheid did not inexorably lead to the continued and growing impoverishment of the black population of the country. It is hard to think of a more political statement, nor one that was to be more systematically ignored.

At the same time, the activist concern with low wages and African poverty continued. From the 1940s onwards, Edward Batson in Cape Town had been attempting to calculate what he termed the Poverty Datum Line for many poor communities. This was calculated as the bare minimum income necessary for a family to survive at an acceptable level of decency, and even then only short term expenditure was taken into account. The cost of food, housing, transportation, clothing, electricity (where available), fuel and cleaning materials were ascertained, but such matters as the hire purchase on furniture, medicines, alcohol, newspapers, and any form of entertainment were ignored, or at least held out of the calculation. Initially, this was a liberating way to combat white prejudice about black poverty, as being caused by extravagance and an unaffordable longing for European goods, but eventually the PDL turned into a mechanical instrument of measurement.^{Footnote 63} Nevertheless, in the early 1970s, it could be used as an “objective” means to back up union claims for higher wages. This was particularly the case in Durban, where the Durban Student Wages Commission used the PDL to provide the dockworkers (and some others) with the weapons they needed in the process of union mobilization. This led in time to the Durban strikes of 1973, generally seen as the moment at which black political activity re-emerged from the relative torpor of the post-Sharpeville era. Once again, even statistics could be liberating.^{Footnote 64} They could also be dangerous. Their work for the UCT Student Wages Commission led to a number of students, most notably Debbie Budlender, being banned.

And so it went on, in the latter days of apartheid, as the two streams of research continued, not always fully in contact. The Bureau of Market Research continued to produce its reports, slowly documenting the rise of South Africa’s black middle class, and the advertisers knew more and more about its consumption habits. It remained a line of argument to which not a great deal of attention was paid before 1994. As against this, most academic investigations stressed the poverty of far too many South Africans, and saw this, surely accurately, as primarily the consequence of government policy.^{Footnote 65} After 1994, in contrast, the two lines of inquiry came together, as the level of poverty, mainly among blacks, has remained incredibly high, while the black bourgeoisie has been growing in numbers and wealth at a great rate. Though less contested, the studies of household income, many of which now come from the Government, remain at the centre of the country’s politics.^{Footnote 66}