This article was downloaded by: [41.13.5.145]On: 14 May 2014, At: 06:09Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number:1072954 Registered office: Mortimer House, 37-41 Mortimer Street,London W1T 3JH, UK

Canadian Journal of AfricanStudies/La Revue canadiennedes études africainesPublication details, including instructions forauthors and subscription information:http://www.tandfonline.com/loi/rcas20

Political Change and theBlack Middle Class inDemocratic South AfricaRoger SouthallPublished online: 30 Oct 2013.

To cite this article: Roger Southall (2004) Political Change and the Black MiddleClass in Democratic South Africa, Canadian Journal of African Studies/La Revuecanadienne des études africaines, 38:3, 521-542

To link to this article: http://dx.doi.org/10.1080/00083968.2004.10751294

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of allthe information (the “Content”) contained in the publications on ourplatform. However, Taylor & Francis, our agents, and our licensorsmake no representations or warranties whatsoever as to the accuracy,completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views ofthe authors, and are not the views of or endorsed by Taylor & Francis.The accuracy of the Content should not be relied upon and should beindependently verified with primary sources of information. Taylor andFrancis shall not be liable for any losses, actions, claims, proceedings,demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, inrelation to or arising out of the use of the Content.

This article may be used for research, teaching, and private studypurposes. Any substantial or systematic reproduction, redistribution,reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of accessand use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

Political Change and the Black Middle Class inDemocratic South Africa

Roger Southall

RésuméLe développement de la classe moyenne noire a été délibérément retardé durant lerégime de la ségrégation et de l’apartheid. Économiquement, une couche de classemoyenne noire était perçue comme une concurrence potentielle à l’accumulationde richesses par les blancs. Politiquement, elle était perçue comme un groupe expri-mant ses exigences d’égalité raciale et d’inclusion. En conséquence, plutôt qued’être absorbée dans une démocratie capitaliste, la classe moyenne africaine noire(sauf les éléments “collaborationnistes” minoritaires) s’est précipitée dans les brasdu mouvement de libération, dont le but — ainsi que l’a exprimé l’AfricanNational Congress (ANC) — était de renverser le “capitalisme racial” et d’accom-plir une “révolution démocratique nationale” (National Democratic Revolution[NDR]) comme précurseur (au moins aux yeux de certains) de l’établissementéventuel du “socialisme.” Cependant, de manière presque inévitable, la nature dela classe moyenne noire après la libération est restée largement imprécise. Bien quel’ANC ait conservé un réel désir “d’offrir” aux pauvres des améliorations de base,ses buts essentiels étaient maintenant, en premier lieu, la consolidation de soncontrôle de l’état, et en second lieu, la “transformation” et des institutions socialeset de l’économie, transformation qui devait s’opérer d’abord grâce à des stratégiesd’”action affirmative” et de “puissance économique noire” (black economicempowerment [BEE]). Cet article examine l’impact de ces changements politiquessur la classe moyenne noire.

Introduction: Defining the “Black Middle Class”The very notion of “middle class” in industrialized, industrializing, andpostcolonial societies is problematic. The massive expansion of middle-strata has long brought into question the classic Marxian notion of capi-talist societies polarizing into two antagonistic classes. The “new middleclass,” which is typically in government or corporate employment, sharesmany of the characteristics of the classic proletariat, notably in the sensethat it has no direct ownership of the means of production and is in a subor-dinate relationship to capital-owning employers. Significantly, suchsalaried employees often enjoy a higher income than members of the “oldmiddle class,” which is typically composed of middle to small-scale

521

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

522 cjas / rcea 38: 3 2004

owners of capitalist enterprises and often employs others in production.Meanwhile, advanced capitalism is characterized by a significant divorcebetween ownership and control, which renders managers (often character-ized as “middle class”) rather than share-holders the typical decision-makers concerning the deployment of capital; this entire situation iscomplicated by the direct or indirect (via invested savings and pensions)ownership of capital by both the middle and working classes.

Competing neo-Marxist and neo-Weberian traditions have developedin response to these developments. The former has sought to explain whylarge sections of the populations of advanced capitalist countries havefailed to support working class parties. They have therefore been preoccu-pied with defining the boundary between the working class and the “newmiddle class,” assuming a conflict of interests between property-owningand non-property-owning classes. The second tradition, in contrast, allo-cates individuals to a variable number of classes according to their occu-pation, income, education, and other “life-chances,” not assuming anyinherent antagonism between them. I am going to duck these tricky issuesby adopting a definition that characterizes the middle class (or petty bour-geoisie) as drawing its primary income (directly or indirectly) from non-manual employment, as “white-collar employees,” managers,self-employed business persons, or professionals.

Given the ANC’s ideal of non-racialism, it may appear anachronisticto refer to the “black” middle-class. I argue, in contrast, that the racialconfiguration of South Africa renders this term meaningful, in the sensethat race “exists” as a social construct. Overall, whites are hugely advan-taged compared to blacks, amongst whom Africans (seventy-nine percentof the population) are relatively disadvantaged compared to the Indian (2.5percent) and Coloured (8.9 percent) minorities.1 Hence, historical racialgroup disadvantage is today central to politically driven attempts to renderthe political economy “representative.”

Before 1994, the separation of races meant that the historical trajecto-ries of middle class elements within these communities were different.While they all shared the fact of racial oppression, they were differentiallyoppressed. This allowed, for instance, for the development of an Indianmerchant class in Natal, which was strides ahead of an African tradingpetty-bourgeoisie, which was subject to much more crippling restrictions.This has meant that, in practice, any study of the disadvantaged middleclasses has replicated apartheid divisions, rendering treatment of the“black” middle class(es) difficult (see Kuper 1965; Southall 1980;Nzimande 1990; Hart and Padachayee 2000). Hence, although this articlewill refer to the “black middle class” as an emergent reality, it will, ofnecessity, draw primarily upon the literature that has dealt with the

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

Southall: Political Change and the Black Middle Class 523

“African” middle class.Although the present drive for “transformation” legally defines the

historically disadvantaged as “blacks,” the political practice is a moreAfricanist one, where Africans are most highly valued, and Indians(amongst whom the proportion of educated and professionally qualifiedindividuals is highest) are the least favoured. Hence the coherence of the“black middle class” cannot be taken for granted. Any tensions, however,are not the present object of inquiry.

The Black Middle Class before 1994Thinking about the black middle class in South Africa has drawn heavilyupon both neo-Marxist and neo-Weberian traditions, reflecting the “revi-sionist” and “liberal” approaches that did battle with each other from the1970s. Broadly speaking, we may characterize the neo-Marxist approach ashaving been primarily oriented towards determining the political orienta-tion of the black middle class, and the neo-Weberian approach as viewingthe black middle class as simultaneously an instrument and outcome ofthe modernization and growth of the economy. Both approaches haveoffered rich insights — and almost as many problems.

The Neo-Marxist Approach

The origins of the African middle class as mission-educated elites whoengaged in early constitutional protest politics have been well docu-mented. An important presentation is provided by Odendaal, who depictsthe members of the early political organizations that preceded the forma-tion of the South African National Native Congress (SANNC [forerunnerto the ANC]) in 1912 as a political class that was, “[b]orn into traditionalAfrican societies and equipped to fit into European society by virtue oftheir education, Christianity and economic assimilation.” This classcalled for African inclusion in national institutions (Odendaal 1984, 286).This moderate strategy was to be overtaken by the shift to industrial soci-ety and the establishment of the segregationist Union of South Africa in1910.

The focus upon the historical origins of the African middle class wasof particular importance to the revisionists, typical of whom was Bonner,who related how this early petty bourgeoisie was subject to contradictorypressures imposed by colonial racism. On the one hand, it identified withthe values of the liberal bourgeoisie against more immediate white oppres-sors; on the other, as was noted by the Godley report in 1921, the member-ship of the SANNC was comprised of “natives who have acquired a certainamount of education only to find the professional, clerical and skilledavenues of employment closed to them” (Bonner 1982, 272). The colonized

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

524 cjas / rcea 38: 3 2004

petty bourgeoisie was, therefore, unable to articulate with conviction themetropolitan values it espoused, although for every member of the pettybourgeoisie, there was “always a correspondingly great substratum amongthe upper level of the working class — generally described at the time asthe `educated’ or `civilised’ — who aspired to their position and struggledto get in” (Bonner 1982, 272).

Reflective of Bonner’s analysis, the revisionist literature that theo-rised the black petty bourgeoisie recognized the contrary pressures towhich it was subject and the differential responses they evoked. Nzimandehas identified two main approaches.2

The first was embodied in the theory of Colonialism of a Special Type(CST). This argued that South Africa was principally a class society char-acterized by capitalism, but its colonial nature meant that the classexploitation of Africans in particular, and of blacks in general, could not beseparated from national oppression. The petty bourgeoisie was, therefore,more immediately linked to the black working class than to their classcounterparts across the colour line. Even so, CST theorists recognized thatclass differentiation amongst the oppressed allowed for a petty bourgeois“project,” which might divide them from the black working class, andnoted, as well, that black middle and upper classes would be more likely toespouse reformist rather than revolutionary outcomes. Hence, the task ofthe liberation movement was to win these elements over rather than alien-ate them. Elaborations of CST were provided, inter alia, by Nolutshungu(1982), who stressed that the racially exclusive nature of the South Africanformation was likely to render the black petty bourgeoisie open toalliances with the rest of the oppressed, and Wolpe (1988), who proposedthat, in practice, the relationship between the petty bourgeoisie and theblack working class was fixed neither in favour of national liberation, noralways against an alliance.

The second approach dealt with the changing location of Africantraders and capitalists that was triggered by late-apartheid reforms involv-ing the lifting of restrictions upon them. This suggested that a qualitativechange was taking place in the position of the urban African petty bour-geoisie, which was to render unlikely their durable participation in a popu-lar alliance struggling for national liberation. African capitalists (who werelargely dependent for their existence upon the support of capital and thestate), as well as associated political segments of the petty bourgeoisie(notably homeland politicians and urban councillors), were overwhelm-ingly collaborationist and wedded to projects of apartheid-era reform. Thisperspective, however, is presented by Nzimande (1990, 173) as “reduc-tionist,” as failing to recognize that oppression affected all classes withinthe black community, albeit unevenly.3

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

Southall: Political Change and the Black Middle Class 525

Nzimande constructed his own approach upon the basis of debatesaround CST, arguing that the consolidation of monopoly capitalism andapartheid from around the early 1960s had drastically changed the compo-sition of the African petty bourgeoisie, which, by the late 1980s, wascomposed of four distinct fractions.

First, the bureaucratic petty bourgeoisie (BPB) constituted the mostfully coopted stratum of the petty bourgeoisie and was composed of rural;Bantustan; and urban, township-based strata. Whereas the former elementwas so closely tied to the apartheid state and monopoly capital that someof the most reliable allies of the white ruling bloc had been drawn from itsranks, the latter — while also closely tied to apartheid structures — wasseverely challenged by the militance of the urban masses.

Second, the trading petty African bourgeoisie (TAPB) was composed ofthree strata. The first consisted of traders in the bantustans, whosefortunes were closely bound up with those of the rural BPB. The secondwas made up of traders in the urban areas and was similarly associatedclosely with the urban BPB except in so far as the militancy of the youth(via boycotts, burning down of properties) forced it to relate to the MassDemocratic Movement (MDM). Finally, a third fraction, the“autonomous” TAPB, which was hegemonic within the National AfricanFederated Chamber of Commerce (NAFCOC) — the primary class organ ofthe TAPB — had openly sympathized with the MDM. Although undoubt-edly anti-apartheid, it was equally definitely pro-capitalist.

Third, the Civil Petty Bourgeoisie (CPB) consisted of the civil servantsand state employees in both the central state administration and thebantustans. This was the largest stratum of the African petty bourgeoisieand was composed largely of nurses, teachers, and clerks. It was from itsranks that independent professionals (who had provided the leadership ofthe national liberation movement) had been drawn. The CPB was poten-tially the most reliable ally of the working class and progressive forcesbecause its reproduction was dependent on neither the continued exis-tence of apartheid nor even capitalist structures. It also shared many of theconditions of the working class in relation to state and bantustan authori-ties (such as poor salaries and despotic working conditions).

Fourth, the Corporate Petty Bourgeoise (COPB) had emerged from theearly 1970s, since the time when corporate capital had sought to legitimizecapitalism through the discourse of the “free enterprise system.” Stronglysupportive of “deracialised capitalism,” the COPB was simultaneouslyhugely frustrated by both limited opportunity and white managerialracism, uneasily situated as it was between capital’s attempts to create ablack middle class and white management’s defence of its own class inter-ests.

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

526 cjas / rcea 38: 3 2004

The Neo-Weberian Approach

From the 1960s, liberal scholars began to interpret the efforts by employ-ers to respond to a shortage of white artisans by promoting African work-ers as evidence that the colour bar was eroding. Whereas revisionistscholars argued that this led to previously “white” work being done byAfricans at cheaper wages, liberals responded that the wage gap betweenwhite and African workers was narrowing. However, by the late 1970s,both sides of the debate were able to agree upon the growing presence ofAfrican middle strata, with Wolpe (1977) pointing to an “enormousincrease in the African middle class between 1960 and 1970,” and Simkinsand Hindson (1979) indicating the growing upward mobility of blacks intoclerical, technical, and non-manual jobs, and of Africans into skilledemployment.

Crankshaw (1996, 635-36) observes that despite agreement about thegrowth of the African middle class, there were wildly varying estimates ofits size, ranging from Nolutshungu’s (1983) low estimate of 121 948 in1970 to Wolpe’s figure of 1 315 800 for 1974. This, he argues, was the resultof scholars using different definitions of the “middle class” and the factthat official employment statistics did not readily translate into class cate-gories. Hence, for instance, it would be crucial to know whether or not an“increase of the African middle class was due ... to the expansion of Africanemployment in teaching or nursing jobs or in middle management” (Wolpe1977, 638).

Crankshaw builds upon both neo-Weberian class schemes (based onoccupation) and labour process theory (which offers explanations aboutwhy changes in the racial division of labour occur). He demonstrates diffi-culties presented by census and other data used for disaggregating occupa-tional categories into class categories, arguing the superiority ofManpower Surveys (undertaken from 1965 to 1985 by the Department ofManpower, and from 1987 by the Central Statistical Service), whichrecorded employment by race and sex for some six hundred occupations.These allow him to develop a classification that ranges from seniormanagement at the top to unskilled manual labourers at the bottom.Thereafter, his findings include the following:• “The almost insignificant extent of African advancement into

managerial and professional occupations (at a mere 3 and 11 percentrespectively in 1990) contrasts with the extensive advancement ofAfricans into semi-professional occupations” (Crankshaw 1996, 643).This is mainly a function of the high number of African nurses andteachers.

• “The African proportion of routine white-collar employment doubledbetween 1965 and 1990, rising from 15 percent to 31 percent ... from

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

Southall: Political Change and the Black Middle Class 527

89 425 in 1965 to about 300 906 in 1990” (Crankshaw 1996, 645-46).Africans occupy a substantial proportion of the most common routinewhite-collar jobs in shops, offices, and transport establishments.

• “The decline in the employment levels of white routine white-collarand artisanal workers has been a result of their upward occupationalmobility, of either an inter- or intra-general character, into manager-ial, professional, semi-professional and supervisory jobs” (Crankshaw1996, 652). There is, thus, evidence of a “floating colour bar,”“whereby African advancement only takes place when whites moveupwards into more skilled and better paid jobs” (Crankshaw 1996,656).

Crankshaw has added major substance to the debate about black advance-ment. Yet how can his insights be combined with those drawn from theneo-Marxist tradition to assist us in understanding the impact of politicalchanges after 1994 upon the black middle class?

Re-thinking the Black Middle Class Post-ApartheidRivero, du Toit, and Kotze have recently described the South Africanmiddle class by contrasting responses to questions about occupation askedin its February Omnibus survey in 1994 and 2000 by the Human SciencesResearch Council. Their major findings are:• The middle class as a whole increased considerably in size, from 8.8

percent of the population (around 3 571 350) in 1994 to 11.9 percent(around 5 333 550). Their calculations suggest that the black middleclass numbered 3 616 504 in 2000 (Rivero, du Toit, and Kotze 2003,17).4

• The major factor responsible for the enlargement of the middle class isthe increase in the proportion of Africans who belong to the newmiddle class, from 3.3 percent in 1994 to 7.8 percent (2 762 460) in2000.

• African advance has been swiftest at the highest — namely the“professional” — level (up from 1.7 percent of Africans in 1994 to 4.9percent in 2000), and much slower in the “manager” (0.1 percent to 0.4percent) and “clerk” (1.3 percent to 2.0 percent) categories, largely asa result of affirmative action within public institutions (Rivero, duToit, and Kotze 2003, 18).

• Whereas twenty-nine percent of the middle class was African in 1994,the corresponding figure for 2000 is around fifty percent (Rivero, duToit, and Kotze 2003, 19).

Having proposed that a well-developed middle class plays a positive role indemocratization, Rivero, du Toit, and Kotze (2003) note that although ithas increased since 1994, the middle class in South Africa remains small.

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

528 cjas / rcea 38: 3 2004

They, therefore, remain uncertain whether its enlargement will consoli-date democracy, and whether class will come to overtake race as the majorcleavage within the political system.

Rivero, du Toit, and Kotze (2003) provide a broad picture of the SouthAfrican middle class under democracy, yet their interpretation remainsproblematic. For instance, they ignore many of the difficulties raised byCrankshaw, most notably in their failure to interrogate the “professional”category and their depiction of this as belonging to the upper middle class.Most particularly, although they problematise their enquiry around thepotential contribution an enlarged middle class may make to democrati-zation, they ultimately fail to tell us anything about that issue at all! That,I would argue, is because in their haste to impose a “democratization”perspective upon post-apartheid South Africa, they have dismissed the richbody of neo-Marxist thinking that has developed around the struggle forliberation.

The Black Middle Class and the State

Any analysis of the contemporary black middle class necessarily beginsfrom the fact that the national liberation movement has captured controlof the machinery of state and, in so doing, has been able to make substan-tial changes to the political and economic terrain. During late apartheid,Nzimande and others could refer to the “ruling white bloc.” Today, inanswer to the question “Who Rules South Africa?” Tom Lodge answerspragmatically, “those people who constitute the `governing class’: presi-dents, premiers, members of the cabinet and executive (provincial) coun-cils as well as the heads of civil service departments” (2002, 22), notingfurther that, “command of the state invests politicians and administratorswith considerably social autonomy, but an autonomy that is constrainedby social actors, by ideology and moral beliefs, and by perceptions of thepossible” (2002, 25).

Where I depart from Lodge is with regard to his failure to engage withthe pre-1994 notion of the “ruling bloc,” which suggested the close entan-glements of the then rulers with white monopoly capital. While Lodge iscorrect to emphasise that the state possesses considerable autonomy (andhence, for instance, had considerable differences with large scale capitalduring the 1980s), it does not follow that the political influence of whitecapital has, post-1994, suddenly melted away!

The theory of the NDR acknowledges that the defeat of nationaloppression will lead to the development of a black capitalist class and themajor growth of black middle strata. This, it stresses, would be a welcomedevelopment, although it also recognizes the dangers attached to thecreation of a crassly consumerist, capitalist-oriented black bourgeoisie

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

Southall: Political Change and the Black Middle Class 529

amidst a South Africa that retains the major contours of mass poverty andsocial inequality. The way forward, it therefore argues, is for the masses tomaintain vigilance to prevent the new bourgeoisie from becoming the toolsof monopoly interests, or parasites who thrive on privilege and corruption.In this context, the ANC should engage with the black bourgeoisie to securetheir adherence to codes of conduct that are compatible with the postponedgoals of the liberation movement. The new bourgeoisie should be “patri-otic” rather than parasitic (Jordan 1997; Netshitenzhe 1996).

Much is left unsaid by the theory of the NDR, not least of which is theassumption that a watchdog ANC will retain its “historically progressiverole,” when an alternative scenario might see it becoming rather more thevehicle of middle class interests. Other difficulties are raised by Adam,Slabbert, and Moodley (1997). A “patriotic” bourgeoisie is expected toremember the African communitarian tradition, yet moral obligationstowards the poor are likely to be undermined by rapid enrichment.Egalitarian reasoning can easily be overwhelmed by selfishness and greed,whilst ethics of communal solidarity will impede the accumulation offortunes. And celebration of black capitalism can invoke class warfare andencourage a destabilizing inequality.

This focus upon black capitalists is understandable if only because thesudden appearance of black business magnates is one of the most dramaticdevelopments of the democratic era. Yet such individuals constitute onlya tiny, albeit highly visible, elite. In contrast, we gain a deeper under-standing of contemporary South Africa by looking at the wider middleclass, as well as its changing relationship to the state and “white,” multi-national capital. It will therefore be useful to return to Nzimande’s cate-gorization of the petty bourgeoisie although prior to that, it will be usefulto revisit the overall size and income of the black middle class.

The Black Middle Class: Size, Growth, and Income

Given Crankshaw’s strictures regarding the difficulties of using officialoccupational data to determine class categories, it may seem contrary torefer to Table 1 to provide an indication of the size and rate of growth of theblack middle class from 1991 to 1999.5 However, cautiously utilized, thedata are instructive.

As Crankshaw would immediately point out, the data for 1991 and1999 are not strictly comparable, as all the listed occupational categorieshave somewhat changed. Furthermore, the categories of “Sales” (1991),and especially of “Service, shop and market sales workers” (1999), willinclude large numbers of employees better identified as “working” ratherthan “middle” class. Nonetheless, these categories would seem to accountfor the large body of middle class occupations.

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

530 cjas / rcea 38: 3 2004

Overall, the numbers of Africans and blacks employed in these cate-gories have grown dramatically. According to Statistics South Africa(2002c, 86), only thirty-six percent of employees in services in 1999 occupymanagerial, professional, or technical positions, in addition to which cler-ical workers in the sector should be included as “new middle class.” Takentogether, therefore, if we arbitrarily define one half of “Service, shop &market sales workers” as white-collar employees rather than workingclass, then we may “guestimate” that by 2004, around 1.8 million Africanemployees, or around twenty-seven percent of formally employedAfricans, can be broadly defined as “middle class,” the equivalent figuresfor blacks as a whole being around 2.5 million and thirty percent.

Table 1Selected Employment by Occupation, Race and Sector

Occupational Year African Coloured Asian White TotalCategory

Professional, Semi-professional, and Technical 1991 287 503 72 448 38 739 442 058 838 748

34.3% 8.6% 4.6% 52.7%Managerial, Executive &Administrative 1991 33 320 15 303 23 707 275 827 348 157

9.57% 4.4% 6.8% 79.2%Clerical & Sales Occupations 1991 519 490 183 718 122 691 777 860 1 603 759

32.4% 11.5% 7.7% 48.5%Total EconomicallyActive 1991 7 497 041 1 359 215 379 702 2 388 410 11 624 368

Professionals, Associate Professionals,& Technicians 1999 744 670 158 650 80 900 628 000 1 596 000

46.7% 9.9% 5.1% 39.3%Legislators, Senior Officials, & Managers 1999 199 050 60 190 47 200 374 830 684 000

29.1% 8.8% 6.9% 54.8%Clerks 1999 450 890 153 150 81 400 383 420 107 000

42.1% 14.3% 7.6% 35.8%Service, Shop & Market Sales Workers 1999 839 130 133 530 42 330 204 580 1 225 000

68.5% 10.9% 3.7% 16.7%Total Formally Employed 1999 6 667 270 1 285 760 394 020 2 001 220 10 369 000

Sources: Adapted from Sikhosana (1996, 71) and SAIRR (2000/01, 346), citing StatsSA, October Household Survey 1999. Figures for 1999 are rounded.D

ownl

oade

d by

[41

.13.

5.14

5] a

t 06:

09 1

4 M

ay 2

014

Southall: Political Change and the Black Middle Class 531

This “guestimated” figure of 2.5 million is considerably below the 3.6million, which Rivero, du Toit, and Kotze (2003) provide for the size of theblack middle class. As neither of our methodologies is rigorous, our differ-ence merely goes to endorse Crankshaw’s insistence upon extremely care-ful disaggregation of official data to arrive at anything approachingaccuracy regarding the allocation of the population into class categories.Meanwhile, if we conceive of professionals and managers as providing thecore of an “upper middle class” and clerks and sales persons as “lowermiddle class,” we arrive at approximate totals of 950 000 and 1.5 millionAfricans and blacks, as belonging to the upper middle class respectively,contrasting with around 870 000 and 1.2 million Africans and blacksbelonging to the lower middle class.

The distinction between an upper and lower middle class is importantas it would seem that the former is the principal beneficiary of a changingdistribution of income. Three basic points need to be made in this regard.• First, long-term trends point to a substantial growth in the distribu-

tion of national income accruing to blacks: the African and black shareof total income increased from 19.8 percent and 28.9 percent in 1970to 35.7 percent and 48.1 percent in 1996 (SAIRR 2000/01, 376).

• Income accruing to the upper middle class is increasing, while theshare of the lower middle class is decreasing. Hence, average annualincome of heads of household from managerial, professional, techni-cal, and administrative grades increased from R116 000 to R150 000(29.3 percent) from 1995 to 2000 (at year 2000 market prices value),while decreasing from R79 000 to R59 000 for those from Clerical andSales grades (SAIRR 2002/03, 179).

• While increasing numbers of blacks are joining whites amongst theupper middle class, South Africa is seeing a widening income gapbetween the rich and the poor. The country’s gini co-efficient rosefrom 0.73 in 1995 to 0.80 in 1998, that for whites increasing from 0.55to 0.67, and for Africans from 0.70 to 0.81 (SAIRR 2000/01, 374).

Revisiting NzimandeThis brief excursion provides an important backdrop for applyingNzimande’s categorization of the black petty bourgeoisie. Yet post-1994political changes demand some adjustments.

Bureaucratic Bourgeoisie or “State Managers”?

An immediate problem is posed by the overlap between Nzimande’sbureaucratic and civil petty bourgeoisies. This distinction is consonantwith a differentiation drawn between the bantustan political leaderships,chiefs, and senior bureaucrats, who “ruled” and directly benefited from the

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

532 cjas / rcea 38: 3 2004

bantustan state apparatus, and the bulk of homeland civil servants, whomerely “administered” and shared many of the disadvantageous condi-tions of the working class in relation to the state as an employer (Graaff1990).

Clearly, Nzimande’s categorization of the BPB as ruling on behalf ofapartheid no longer stands, given the post-1994 absorption of the home-lands and the unification of local government. Another problem is that theterm “bureaucratic bourgeoisie” has perjorative overtones that implyauthoritarian class rule, repression, and undue privilege, which cannotsimply be assumed under present conditions. Nonetheless, intuitively, thedistinction between “ruling” and “administering” retains value: the func-tions clearly overlap, yet it is not difficult to identify a hierarchy of seniorpoliticians and civil servants, who are responsible for the major decisionsthat determine policy, and the far more plentiful middle-level and juniorstate civil servants and state employees, who are largely responsible forimplementation of policy. The term “bureaucratic bourgeoisie” is, there-fore, better replaced by the more neutral term “state managers,” a termborrowed from Graaff (1990), who, ironically, used it to depict the rulers ofthe bantustans.

State Managers and the Civil Petty Bourgeoisie.

The fraction of state managers is composed of those elements that Lodgehas depicted as “the governing class” — the president, premiers, membersof the cabinet and provincial governments, as well as senior civil servants,plus (I would add) the senior executives of parastatals. As well as being thekey political decision-makers, this small, relatively tight group is bondedtogether by an ideology of public service and, for the majority, by loyaltyto the ANC. In contrast, the civil petty bourgeoisie is a much larger andmore heterogeneous fraction, composed overwhelmingly of thoseemployed in white-collar and service occupations. As before 1994, the CPBmay be said to include many independent professionals outside the publicservice, yet is overwhelmingly composed of the mass of civil servantsbelow state manager level, as well as professionals inside the wider publicservice (notably teachers and nurses). The two fractions are closely inter-linked, not least through the transformation of the public sector.

Following the ANC’s “capture of state power,” the reformulation ofthe public service has been extensive, entailing both the merging offormerly racially segmented administrations and major efforts to renderthis restructured public service “representative.” First, in pursuit of whatthe ANC-in-government considered the need to rationalise the publicservice, the number of employees within the “government sector,” whichwas extensively defined was considerably reduced after 1994 (see Table 2).

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

Southall: Political Change and the Black Middle Class 533

Second, the representativeness of the government sector hasmassively improved. Although not strictly comparable to Table 1, Table 3is illuminating. It is immediately evident that the African and Colouredgroups have been the principal beneficiaries of civil service “restructur-ing.” This has been brought about through strategies of affirmative action,spelt out in the Employment Equity Act of 1998 and the White Paper onthe Public Service of 1995.

Only a small minority of civil servants belong to the senior echelonsand, therefore, qualify as state managers. In 2000, the Department ofPublic Services indicated that there were 1 766 “senior managers” innational departments, and 1 175 in the provinces, of whom around sixtypercent were black,6 (compared with a black composition of the “manage-

Table 2Employment in Government Sector 1994-2001

Year National Provinces Local “Other” Govt Total %Govt Govt Business Govt Formal

Enterprises Employees Employment

1994 549 336 701 632 238 247 100 011 207 890 1 797 117 33.72001 323 152 802 587 220 759 88 449 147 372 1 582 319 34.0% change- 41.2 14.4 -7.3 -11.6 -29.1 -12.0 -

Source: SAIRR (2002/03, 150).

Table 3Racial Composition of the Civil Service, 1993 and 2003

National Non- Independent National Provinces1993 Homelands 2003 2003

1993

African 316 929 241 335 195 238 563 30041.3% 62.8% 78.6%

Asian 30 453 16 641 21 4933.97% 5.4% 30.0%

Coloured 124 711 91 994 65 15216.25% 29.6% 9.1%

White 295 429 73 319 66 79738.50% 23.6% 9.34%

Total 767 521 310 907 716 742

Sources: SAIRR (1994/95, 476) and Public Service Commission (2004).

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

534 cjas / rcea 38: 3 2004

ment echelon” of only six percent in 1994). This development has beenreflected in the government sector more widely. For instance, a majority ofthe directorships of the key parastatals are now black (Southall 2005), asincreasingly are the senior positions in universities and research councils.

Alongside their “blackening,” the state managers are leaving behindthe ANC’s apartheid-era rhetoric of class and national struggle in favour ofa more ideologically neutral language of representativeness, accountabil-ity, “delivery,” management, and human resource development. There isno question that this shift is necessary to cope with the highly complexchallenges posed by running a modern state and that it is attuned to anideology of professionalism and public service, which may, at times,conflict with direct class interest. Nonetheless, the management ethos isitself likely to encourage state managers to define the public good in a waythat favours middle class interests.

The CPB is much larger and more varied than the tightly knit fractionof state managers. Its core continues to be rooted in the public service, yetNzimande’s pre-1994 categorization should be qualified in at least tworegards. First, as we have seen, there has been a continuing expansion ofAfrican white-collar employment by the private sector, rendering thiselement an increasingly important segment of the CPB. Second, blackprofessionals who are privately employed are better included amongst theranks of the corporate bourgeoisie, not least because there is an increasingtendency for them to hire out their services either to the corporate sectordirectly, to the corporate bourgeoisie indirectly, or to public sector organi-zations.7 However, the principal point is that the black CPB is extensivelyunionized.

During the late “struggle years,” trade unions were able to dictate thedirection of change in industrial relations. In contrast, the concurrence ofdemocratization and globalization during the 1990s has thrown them ontothe defensive. Although, overall, the membership of registered unions hasincreased (from 2.9 million in 1992 to 3.9 million in 2001), union member-ship as a proportion of the economically active population has declined(from just under twenty-five percent to around twenty-one percent)(SAIRR 2002/03, 185-86). Within this context, the most rapid growth ofunions has taken place within the public service, matching a shift fromrepressive labour relations to a system of representative collective bargain-ing (Macun and Posel 2000, 101). The outcome of this rapid unionizationin income terms is difficult to unravel. Woolard (no date) demonstratesthat union membership enhances earnings by as much as thirty-twopercent, although this effect is considerably more pronounced in theprivate than the public sector.

These findings are consonant with the comment concerning growing

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

Southall: Political Change and the Black Middle Class 535

inequality, more particularly a widening gap between the formally and theinformally unemployed. Yet there is the further complication that in thepublic sector, collective bargaining has to accommodate the interests ofthe public, as well as those of the state and its employees. Seekings (2004)suggests that this has not always been achieved. While the trade unionmovement has espoused increased public spending on “pro-poor” reforms,these calls have, in part at least, been driven by the self-interest of sectionsof the non-poor. He argues this by reference to teachers (located in thehighest decile of income-earners), who have been represented largely bythe South African Democratic Teachers’ Union. Per capita expenditure onpoor children has increased, but the large proportion of this improvementhas gone to paying higher teacher salaries, without resulting in superioreducational outcomes.

In summary, it may be argued that the black CPB is Janus-faced. It hasbeen a major beneficiary of ANC rule, affirmative action, and long-termtrends working towards the deracialisation of the economy. On the otherhand, while ideologically sympathetic to the poor, its self-interest appearsto be widening the gap between at least the upper sections of the blackmiddle class and the poor. In other words, the progressive role that it hasplayed historically may be at increasing variance in the future because ofaspirations to achieve greater social equality between the classes.

The Black Trading Petit-Bourgeoisie

African business under apartheid was stunted by official licensing, whichrestricted African entrepreneurs to operating in racially demarcated areas.To be sure, from the 1960s, these restrictions were complemented by theencouragement of African business in the homelands via the developmentcorporations, this leading to the emergence of a stratum of businessmenthat was closely tied to bantustan structures. Subsequently, too, after the1976 Soweto uprising, Pretoria hastened to promote African business inthe urban areas to develop a black middle class (Southall 1980). Both thesestrata were to be identified by the mass movements as beneficiaries ofapartheid, although — as Nzimande argues — the urban element wasrendered ambivalent by political pressure. As Nzimande (1990, 200-03)also demonstrates, African business operated under the financial shadowof monopoly capitalism, which provided it with the overwhelming supplyof its finance, credit, and supplies. Nonetheless, he also asserts the emer-gence of an “autonomous” stratum of the trading bourgeoisie, which heimplies developed late, became hegemonic within NAFCOC, and aligneditself politically with the MDM. These latter were individuals who werecapitalists, yet they were too few and weak to constitute an African capi-talist class as such.

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

536 cjas / rcea 38: 3 2004

Today, there are quite striking continuities and discontinuities withthe apartheid era. The major discontinuity is the collapse of the bantus-tan/urban African business divide into a more integrated, small andmedium business sector, which overlaps heavily with the informal econ-omy. However, as under apartheid, African enterprises remain undevel-oped, and Africans, as a whole, continue to lack relevant experience,traditions, skills, and capital (even if this generalization applies consider-ably less to the Indian community).

Some indication of the character of small enterprise is offered byNtsika Enterprise Promotion Agency (see Table 4). The businesses it iden-tifies are distinguished as “survivalist” (hawking etc), “micro” (informal,with a maximum of five employees), “very small” (self-employed individ-uals or small enterprises operating in the formal sector, with up to nine-teen employees), “small” (more complex operations with up to forty-nineemployees), and “medium” (still owner / manager controlled, yet with upto two hundred employees in more advanced sectors such as manufactur-ing and mining).

Overall, it is estimated that survivalist enterprises provided 2.2percent, micro businesses 10 percent, very small 13 percent, small 15.7percent, medium 13.01 percent and large-scale businesses 46.1 percent oftotal employment in 1999, with micro enterprises contributing 5.8percent, small businesses 13.9 percent, medium 15.1 percent and largebusinesses 65.2 percent of GDP in the year 2000 (SAIRR 2002/03, 194-95).

It is immediately evident that there is a strongly inverse relationshipbetween size and racial ownership, with blacks far more likely to own thesmallest enterprises. Only around one-fifth of “very small” businesses areblack-owned. The proportion of “small” and “medium”-sized ones thatare black-owned would be even less. This implies the tiny size of the blackTPB nationally, as intuitively it is difficult to conceive of characterizingowners of “micro” and even many “very small” businesses as “middle-

Table 4Distribution of Small Business Sector by Race and Size, 1996 and 1999

Survivalist Micro Very Small Unspecified

Year Black White Black White Black White Black White1996 163 268 24 125 271 150 200 700 15 875 84 411 16 625 29 745

87% 13% 57% 43% 22% 78% 36% 64%1999 88% 12% 63% 37% 12% 88% 32% 68%

Sources: SAIRR (2000/01, 412; 2002/03, 200).

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

Southall: Political Change and the Black Middle Class 537

class.” Not surprisingly, therefore, given that the small business sector isregularly hailed as the key to the promotion of both employment and BEE,it is the target of official assistance. Hence, while the homeland develop-ment corporations are long gone, the state continues to provide support tosmall black enterprises through ventures such as the Black BusinessSupplier Programme and the South African Women Entrepreneurs’Network (both operated by the Department of Trade and Industry), as wellas through affirmative procurement. However, the impact of these initia-tives remains uncertain, especially given that the potential of the informalsector for generating growth may be reaching its limit (Simkins 2004).

Its small size suggests that the political weight of the trading pettybourgeoisie is minimal, at least in comparison with that of the increasinglyinfluential black corporate bourgeoisie. This has recently been recorded bybitter struggles within NAFCOC. By 2003, the presidency of the organisa-tion had been assumed by one of most prominent black industrialmagnates, Patrice Motsepe, who proceeded to steer towards the merger ofNAFCOC with the white-dominated South African Chamber of Business(SACOB), much to the dismay of small black business.

Nzimande noted the heterogeneity with NAFCOC, yet argued that bythe mid-1980s, the “autonomous” TAPB had become hegemonic within it,and was supportive of the MDM. He is less clear on the basis for the“autonomy” of this element, but by implication, he is referring to emer-gent, large-scale African capitalists. Today, these latter have joined thecorporate bourgeoisie and are amongst the primary beneficiaries of BEE.

The Corporate Black Bourgeoisie

Nzimande’s COPB was composed of Africans employed at the manageriallevel by corporate capital. This trend has developed considerably since1994, promoted latterly by the Employment Equity Act of 1998 and thePromotion of Equality and Prevention of Unfair Discrimination Act of2000. Whereas the former required all medium and large-scale employersto submit once-off equity reports, profiling their employment demograph-ics and plans to achieve representativeness, the latter imposed wide-rang-ing prohibitions of unfair (racial and other) discrimination upon all mannerof institutions (including companies). Legislative impetus is supported bythe pronounced skill shortage amongst blacks, with employers reportedlypaying premiums in excess of thirty percent to attract and retain appropri-ately qualified black staff (SAIRR 2002/03, 207).

Increasingly, this stratum of black corporate employees overlapsextensively with an identifiably black capitalist class, whose formationhas been hugely assisted by BEE. We may characterize this development asfollows.

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

538 cjas / rcea 38: 3 2004

First, black penetration of corporate ownership is growing, albeitslowly. (Southall 2004) In broad terms, an early flurry of empowermentdeals that saw black business secure up to ten percent of shares on theJohannesburg Stock Exchange (JSE) by 1997 fell victim to the latter’s catch-ing a heavy dose of Asian flu, resulting in a sharp fall-back in black owner-ship to somewhere from one to three percent by 2000. Basically,shareholdings purchased by black consortia through finance capital hadunderperformed, with their value further undermined by the stock marketcrash of 1997. White institutions, therefore, repossessed the monies theyhad lent to black consortia, which increasingly thereafter disposed of theirassets on the JSE and focused on unlisted ones.

Careful distinction needs to be made between black majority-ownedcompanies and total black investments. The number of black controlledcompanies on the JSE increased from seventeen in October 1997, with amarket capitalization of R37 billion, to thirty-eight in August 1999 (R98billion), before dropping to twenty-two in September 2002 (R33 billion).However, according to another calculation, black investors currently indi-rectly own twenty-two percent of shares on the JSE, which are managed forthem by institutional investors (SAIRR 2002/03, 209). These contribute toan appreciably changing backdrop, against which a number of blackcontrolled conglomerates (such as Johnnic, MTN, Mvelaphanda Holdings,and NAIL) are increasingly teaming up with established corporate capitalvia joint ventures and holdings. However, black ownership at this level ishighly concentrated, symbolized by the massive wealth accumulated byindividuals such as Patrice Motsepe, Cyril Ramaphosa, and TokyoSexwale, leading to widespread accusations that empowerment deals havefavoured a small black elite rather than facilitating a wider reallocation ofwealth.

Second, to promote a more determined assault upon white economicpower, the government has recently adopted a more assertive strategy torender BEE more “broad-based.” Most particularly, this has taken the formof official prodding of the different sectors of industry to draw upEmpowerment Charters. These establish targets to be met regardingownership, employment equity and training, worker conditions, procure-ment from black-owned firms, and community development over a giventime span (usually from ten to fifteen years). Although not backed bylegislative sanction, realization of targets will be spurred by the govern-ment’s deployment of its spending power only to those companies that candemonstrate achievement of targets.

Finally, the drive towards the blackening of capitalism is politicallydriven. This has several interrelated aspects. The first is quite simply thatthe government is tilting the playing fields in favour of blacks. Second, BEE

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

Southall: Political Change and the Black Middle Class 539

has become tainted with suggestions that it is promoting “crony capital-ism,” with certain individuals and companies closely associated with theANC having unduly benefited from the awarding of state contracts. Thismeans, thirdly, that the route into high-level business is often throughpolitics or the state, there being an increasing trend for high-level politi-cians and civil servants to secure highly paid positions in the private sectorafter acquiring skills and information within the public service. In otherwords, there is a very close relationship between “state managers” and theemergent black corporate bourgeoisie.

Conclusion: The Black Middle Class and DemocracyMuch more extensive debate between the neo-Marxist and neo-Weberiantraditions is required before we have an adequately nuanced understandingof the black middle class and its impact upon the character of SouthAfrican democracy. Nevertheless, the following broad propositions can beventured.

The most obvious point is that a fairly rapidly growing black (new)middle class is the prime beneficiary of ANC rule. The liberation struggleagainst apartheid had as its prime objective the capture of state power.Although the negotiated settlement placed constraints on character ofpolitical change, the first ten years of the ANC in power have registered aremarkable transition away from white minority rule. Today, high politi-cal and state office is truly reflective of the demographic composition ofthe population, while the public service as a whole has made massive stepstowards “representativeness.” This transformation is the single mostimportant factor accounting for the present rate of growth of the blackmiddle class and is clearly instrumental in consolidating the growth in thedistribution of national income accruing to blacks. Even so, various frac-tions of the black middle class have shared these benefits differentially.

One of the great ironies of the arrival of the democratic era is thewidening inequality of incomes. Within this context, it is “statemanagers” and the corporate fractions of the black middle class, along withthe more advantaged sections of the civil petty bourgeoisie, that are join-ing the majority of whites on higher income levels. Most particularly, thetwo prior fractions can be depicted as working most closely with the wider“ruling bloc.” Blacks may have taken over the state, and may now largelycontrol the parastatals, yet whites continue to dominate the economy.However, what is increasingly clear is that large corporations are energet-ically courting the new rulers and the new breed of black magnates.

The emergent alliance between large-scale capital, state managers,and the COPB implies the embrace of capitalism by blacks, which wassuggested by the “economistic” critics of the theory of CST. This blunt

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

540 cjas / rcea 38: 3 2004

conclusion, however, needs to be qualified, for while both governmentpolicy and the aspirations of the COPB are directed unambiguously at the“blackening” of capitalism, there remain major counter-pressures withinthe ANC, which impose ideological and policy constraints upon the move-ment’s explicit adoption of a pro-capitalist path. Nonetheless, the tradi-tion of liberation struggle and of “freedom for all,” along with the pursuitof equalitarian goals, continues to impose constraints upon the ANC’sadoption of a raw capitalism in favour of a more redistributionist vision,entailing the “delivery” of services to the poor. Importantly, too, asoutlined by the theory of the NDR, it remains the role of popular forces toensure that black capitalists remain “patriotic”: that is, that they combinetheir pursuit of personal gain with reinvestment of profits within the coun-try and their recognition of obligations to the black communities fromwhich they have sprung.

The pressures imposed upon the ANC to realize the fruits of racialequality are imposed to a significant extent by the civil petty bourgeoisie.This is the largest fraction of the black middle class and is one that hasrecorded substantial gains since 1994. Most particularly, it has benefitedfrom affirmative action and equity legislation that have brought about, formany, increased incomes, secure employment, and improved workingconditions, even if — especially in the private sector — its advance remainsconstrained by an upward floating colour bar. While macro-data maysuggest that the lower middle class is less well placed than thirty or fortyyears ago, that lower middle class was largely white and the present one isincreasingly black. As Seekings (2004) describes them, they are the “non-poor” in a country in which the proportion of people in informal employ-ment, or who are unemployed, is inexorably growing. Nzimande definedthe COPB before 1994 as the most reliable ally of the working class andprogressive forces. However, their relative advantage today is likely tohave rendered them more ambivalent to the pursuit of greater socialequity. It would seem, in contrast, that it is the black trading petty bour-geoisie, largely locked into the informal economy, which is objectively theclosest to the working class, even if its aspirations towards entrepreneur-ship may compromise its alignment with progressive goals.

The growth of the black middle class suggests that the egalitarian aspi-rations of the NDR are now at risk and that differences of interest betweenthe middle class and the poorer mass of the population are increasinglylikely to arise.

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

Southall: Political Change and the Black Middle Class 541

Notes1 The 2001 Census gave the South African population as Black Africans: 34 416 166;Coloured 3 994 505; Indian or Asian 1 115 467; and Whites 4 293 640 for a total of44 819 778 (SSA 2003, vii).2 Actually, he laid out four approaches, but I have condensed those founded upon thetheory of the CST.3 The principal targets of Nzimande’s criticism were Hudson and Sarakinsky (1986)and, “to a certain extent,” Southall (1980)!4 The numerical estimates of the size of the middle class are my own calculations,based upon the size of the population as indicated by the 1996 and 2001 censuses,respectively Statistics South Africa (2003).5 Statistics South Africa formerly the Central Statistical Services replaced itsOctober Household Survey by the Labour Force Survey from the year 2000. Thelatter does not list occupation by race.6 I cannot give exact figures, as I have had access only to bar charts.7 For instance, the implementation of equity legislation has significantly increasedthe demand for the services of black labour lawyers and human resource consul-tants especially.

BibliographyAdam, H, F. Van Zyl Slabbert and K. Moodley. 2002. Comrades in Business: Post-

Liberation Politics in South Africa. Cape Town: Tafelberg.Bonner, P. 1982. “The Transvaal Native Congress 1917-1920: The Radicalization

of the Black Petty Bourgeoisie on the Rand.” In Industrialisation and SocialChange in South Africa: African Class Formation, Culture andConsciousness, 1870-1930, edited by Shula Marks and Richard Rathbone,270-313. New York: Longman.

Graaff, J. 1990. “Towards an Understanding of Bantustan Politics.” In ThePolitical Economy of South Africa, edited by N. Nattrass and E. Ardington.Cape Town: Oxford University Press.

Hart, K. and V. Padayachee. 2000. “Indian Business in South Africa afterApartheid: New and Old Trajectories.” Comparative Studies in Society andHistory 44, no.4: 683-712.

Hudson, P. and M. Sarakinsky. 1986. “Class Interests and Politics: The Case of theUrban African Bourgeoisie.” In South African Research Services, SouthAfrican Review 3 (Johannesburg: Ravan Press): 169-85.

Jordan. P. 1997. “The National Question in post 1994 South Africa: A DiscussionPaper in Preparation for the ANC’s 50th National Conference.”www.anc.org.za/ancdocs/discussion/natquestion.html

Kuper, L. 1965. An African Bourgeoisie: Race, Class and Politics in South Africa.New Haven: Yale University Press.

Lodge, R. 2002. Politics in South Africa: From Mandela to Mbeki. Cape Town:David Philip; Oxford: James Currey.

Macun I. and C. Psoulis. 2000. “Unions inside the State: The Development ofUnionism in the South African Public Service.” In Public Service Labour

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4

542 cjas / rcea 38: 3 2004

Relations in a Democratic South Africa, edited by Glenn Adler, 91-109.Johannesburg: Witwatersrand University Press.

Netshitenzhe, J. 1996. “The National Democratic Revolution: Is it Still onTrack?” Umrabulo 4.

Nolutshungu, S. 1982. Changing South Africa: Political Considerations.Manchester: Manchester University Press.

Nzimande, B. 1990. “Class, National Oppression and the African PettyBourgeoisie: The Case of African Traders.” In Repression and Resistance:Insider Accounts of Apartheid, edited by Robin Cohen, Yvonne Muthien andAbebe Zegeye, 165-210.. London, Melbourne, Munich, New York: Hans ZellPublishers.

Odendaal, A. 1984. Vukani Bantu! The Beginnings of Black Protest Politics inSouth Africa to 1912. Claremont: David Philip.

Public Service Commission. 2004. State of the Public Service Report 2004.Rivero, C, P. du Toit and H. Kotze. 2003. “Tracking the Development of the

Middle Class in Democratic South Africa.” Politeia 22, no.3: 6-29.SAIRR. South African Institute of Race Relations. 1994/95; 2000/01; 2002/03.

Race Relations Survey. Johannesburg: South African Institute of RaceRelations.

Seekings, J. 2004. “ Trade Unions, Social Policy and the Political Economy ofClass Compromise in Post-Apartheid South Africa.” Review of AfricanPolitical Economy 31, no.100: 299-312.

Sikhosana, M. 1996. “Affirmative Action: Possibilities and Limitations.” InAffirmative Action and Transformation, edited by Blade Nzimande andMpumelelo Sikhosana, 65-79. Natal: Indicator Press.

Simkins, C. and D. Hindson. 1979. “The Division of Labour in South Africa, 1969-1977.” Social Dynamics 5, no.2: 1-12.

Southall, R. 1980. “African Capitalism in Contemporary South Africa.” Journal ofSouthern African Studies 7, no.1: 38-70.

——. 2004. “The ANC and Black Capitalism in South Africa.” Review of AfricanPolitical Economy 31, no.100: 313-28.

——. 2005. “Black Empowerment and Corporate Capital.” In The State of theNation: South Africa 2004-05, edited by J. Daniel, J. Lutchman and R.Southall, 455-78. Cape Town: HSRC Press and Michigan: Michigan StateUniversity Press.

Statistics South Africa SSA. 2003 Census 2001: Census in Brief. Pretoria.Wolpe, H. 1977. “The Changing Class Structure of South Africa: The African

Petit-Bourgeoisie.” In Research in Political Economy, edited by P. Zarembka,volume 1. Greenwich, Connecticut.

——. 1988. Race, Class and the Apartheid State. London: James Currey.Woolard, I. No date. “An Investigation of wage levels and wage inequality in the

South African Public Sector.” Human Sciences Research Council.

Dow

nloa

ded

by [

41.1

3.5.

145]

at 0

6:09

14

May

201

4