Is there a future for a labour accord in South Africa?

Introduction

Over the last 20 years, South Africa's economy has performed poorly in relation to the East Asian 'tiger' economies and itself in the 1960s and 1970s. Low gold and mineral prices, financial and trade sanctions in the late 1980s, and massive industrial unrest have all conspired to worsen South Africa's economic prospects (Wood and Harcourt 1998: 83). 'From 1980 to the end of 1993, per capita gross domestic product (GDP) rose in only four years, and only in one after 1984, for an annual average decline of almost 1% for the fourteen years' (Weeks 1999: 796). Modest growth temporarily resumed growing 1.3% in 1993, 2.7% in 1994 and 3.3% in 1995, but then fell back below 3% per year in the years following (Republic of South Africa, 1996; Weeks, 1999: 802). In contrast, the East Asian and Pacific economies grew at 7.8% per year from 1980 to 1993. Worldwide, 14 of the 18 non-Eastern Bloc economies the World Bank classified as having a similar level of economic development to South Africa in 1995 had higher growth rat es over the same fourteen-year period. Four of these had rates in excess of 5% (World Bank 1993: 164-5).

Unemployment has risen from approximately 20% in the early 1980s to over 35% in the late 1990s (Statssa.gov.za). Employment in the formal sector, where most of the best paying jobs are located, grew just 0.7% per year in the 1980s. Total employment expanded by 350,000 jobs over this period, less than 10% of the 4.5 million additional people in the labour force (Republic of South Africa 1997: 3). The 1990s have been even worse. Formal employment contracted 1.1% per year in the early 1990s (Republic of South Africa 1996: Appendix 13), recovering only slightly after that but remaining well below 1990 levels. Furthermore, many of the largest traditional employers of unskilled African labour have had mass redundancies. For instance, mining employment contracted from 769,000 workers in 1980 to 614,000 in 1994, with most of the decline occurring in the early 1990s. Employment in commercial agriculture also declined from more than a million in 1980 to 861,000 in 1994 (Republic of South Africa 1996: Appendix 13).

Declining average living standards and a shortage of jobs have exacerbated two of the main legacies of apartheid policies: extreme income inequality, virtually unrivalled in the World, and poverty. The poorest 40% of households continue to earn just 6% of national income, whereas the richest 10% earn more than 50% (Republic of South Africa, 1996: 2 [White paper on Social Welfare]). In addition, more than a third of all South Africans and more than half of South Africa's children live in poverty. Twenty percent of South Africa's population continues to rely on subsistence agriculture for a meagre living (Republic of South Africa, 1996: 2-3).

South Africa's economic trajectory is not simply the product of anonymous global economic forces; rather it reflects ongoing contestations between capital and labour. The South African labour movement has been remarkably successful in securing improved wages and working conditions in core areas of the economy such as manufacturing; it has been less successful in checking widespread redundancies, as employers seek to promote greater numerical flexibility. The latter represents an immediate challenge for unions--a shrinking employment base within the formal sector necessarily entails drops in membership (but not reduced levels of militancy); unions have to check stagnant or declining membership in key areas of the economy if long term aims of a more just society are to be realized. Neo-corporatism's track record would suggest that it represents a possible mechanism for alleviating unemployment, whilst making a material contribution to redressing widespread social inequality (Casey and Gold 2000).

But is neo-corporatism a worthy policy alternative to be touted by progressives in a country, where, until 1990, socialism seemed firmly on the agenda of the liberation movement? This reopens a long-standing debate within South Africa regarding the most feasible route for building a more equitable future. During its years in exile, the South African Communist Party adhered to what became known as the 'two stage theory' as a basis for its alliance with the ideologically heterogenous ANC: an interim 'national democratic phase' following the inevitable destruction of apartheid constituted a necessary prerequisite for socialism. This debate was somewhat recast following the collapse of Stalinist rule in Eastern Europe and the ending of apartheid in South Africa. Amongst progressives, three broad camps have emerged with opinions concerning the most desirable strategic options. The first, associated with the works of Richard Sklar, is loosely pragmatic. Sklar touts a mixed economic system that transcends both trad itional capitalism and socialism (Saul 1999:52). In practice, Sklar's prescriptions fall far closer to the IMF/World Bank viewpoint than a genuinely mixed system (ibid.). Nonetheless, this viewpoint is an extremely influential one, shared by a significant grouping of senior members of the South African Communist Party (SACP) who hold cabinet posts and have been outspoken in their defence of neo-liberal policy prescriptions.

Secondly, there is the ultra-left tradition. Associated with sections in the labour movement--above all, within the National Union of Metalworkers of South Africa (NUMSA)--and a number of Trotskyist splinter organizations, ultra-leftists argue that the highly visible nature of social inequality in South Africa, and a rich tradition of protest and activism, open the way for a genuine socialist revolution (Baskin 1991). Of necessity, this would fracture the Alliance between the ANC, the SACP, and the Congress of South African Trade Unions (COSATU) (Habib and Taylor 1999: 112-120). Given the viability of socialism, there is no need to compromise; indeed any short-term deals with capital may blur the starkness of the class struggle and defer the onset of socialism. However, given divisions within the ultra-left camp, and the lack of support for their ideals amongst COSATU's leadership and rank-and-file, this strategic option seems somewhat unworkable in the short- and medium-terms (Southall and Wood 1999).

Thirdly, there is a broad school of thought partly comprised of scholars loosely associated with COSATU (dubbed the 'left wing dogmatists' by the ultra-left) (Habib and Taylor 1999:112). This group argues that the negotiated nature of South Africa's transition has resulted in a very necessary compromise, the outcome of which has been in some respects conservative (Adler and Webster 1995). However, through strategic engagement, it is possible to ensure more radical reform. This viewpoint is partially informed by the contribution of Peter Waterman (1990). Drawing on both the Leninist tradition and classical theories of social movements, Waterman proposes an alternative strategic direction in the light of changes in global capitalism. Waterman argues that 'New Social Unions' should work within current institutional structures, but seek to promote more radical change. In pursuit of the latter, new strategies of 'radical engagement' represent a more workable option than those based on the 'sustained optimism' hel d by the left in the past (Waterman 1999:60).

Many of the premises of Waterman's arguments over the years have been echoed in the works of a number of South African scholars. Writing in 1994, Karl von Holdt (1994:306) suggested that the possibility of 'strategic unionism' had emerged, 'a labour-driven process of strategic change', building on the COSATU unions' early role as a social movement. Through close links with the ANC, the unions could steer the latter into a more leftward direction than would otherwise be the case. This would involve an ongoing commitment to radical reform 'based on independent labour and popular organization' (ibid.:306). These basic arguments are reiterated and further developed in an influential 1995 paper by Adler and Webster (1995).

More recently, it has been argued that the deeply seated loyalty of COSATU's rank and file to the ANC--reflected in surveys of COSATU's membership-makes it infeasible for COSATU to 'go it alone', or to form a new workers' party with the SACP in the short term (Southall and Wood 1999: 74-7). Rather, it is necessary to work within the Alliance to mitigate some of the worst excesses of neo-liberalism, aiming to bring a real improvement in the material conditions of the working class. At the same time, the establishment of a more radical workers party at a more favourable occasion in the future should not be ruled out. However, it is not desirable to defer meaningful social redress until a hypothetical future stage when progressives may be in a position to set the entire agenda (c.f. Webster and Adler 1995; Southall and Wood 1999). Indeed, COSATU has been extremely influential in ensuring that the bulk of labour-friendly employment legislation is retained and that the delivery of basic social services is priorit ized (Southall and Wood 1999:77). In other words, COSATU's capacity to secure real gains in an increasingly adverse environment should not be discounted (Southall and Wood 1999). It is from this standpoint that we explore the possibilities of neo-corporatism as a possible step towards a progressive future. It is not enough to highlight the failures of neo-liberalism, but also to have in place alternatives that demonstrate the viability of policies that accord a greater primacy to the role and effects of institutions.

Orthodox economic strategies: the case of South Africa

Many poorer countries like South Africa have been pressured by transnational financial institutions and Western governments to pursue an orthodox or neo-liberal economic strategy as the solution to their economic problems, despite mounting evidence of its failings. In general, this strategy entails macroeconomic policies to curb inflation and restore a balance of payments in the national income accounts, and microeconomic policies to free markets of any regulatory restrictions on prices or quantities. The ANC did not have a well-articulated set of economic policies prior to its 'un-banning' in 1990, although its general philosophy was vaguely leftist. The ANC's accession to political power was marked by pragmatism and a willingness to enter into compromises with other social groups from former homeland elites to union leaders (Wood and Psoulis 2001).

At first, the ANC's emerging interest in orthodox policies was not readily apparent to the electorate, given the ANC's ostensible commitment to the Reconstruction and Development Program (RDP) in 1994. This program focused on production for basic needs, chiefly through infrastructure investment and education in the rural hinterland, and expansion of the manufacturing sector's exporting capacity, but provided few specifics on how this might be achieved in terms of the requisite organizational, financial, and human resources (Fine and Van Wyk 1996; Saul 1999). For this reason, some criticized the RDP as little more than a well-meaning, wish list (Von Holdt 1993), to be used to attract voters during the 1994 general election and quietly ignored once in office. Two years later, the ANC effectively abandoned the RDP when it endorsed a more explicitly market-driven policy, the Growth, Employment, and Redistribution (GEAR) Macroeconomic Strategy (Republic of South Africa 1996; Saul 1999). GEAR does maintain some el ements of a non-orthodox program, such as infrastructure expansion, a compulsory payroll levy to finance vocational training, and tax incentives for labour-intensive, business start-ups in the poor hinterland (Republic of South Africa 1996). However, expenditures on these programs are limited to what can be afforded, given an orthodox strategy to reduce the budget deficit, stabilize the exchange rate, tighten monetary policy to curb inflation, privatize state assets, relax foreign exchange controls, and reduce and homogenize tariffs (Republic of South Africa 1996). GEAR does not envision a Keynesian role for the state in promoting aggregate demand or a social democratic role in using extensive social welfare policies to redistribute income. Instead, the state's role is to facilitate growth by providing a stable economic environment to encourage private-sector investment and the creation of a globally competitive, export-oriented economy (Republic of South Africa, 1996). Nevertheless, GEAR's forecasts for 1997 to 2000 were for low Gross Domestic Product growth of less than 3.5% per year and low employment growth of less than 1.5% per year, rates that are insufficient to raise living standards or lower the unemployment rate. The only improvements forecast were for the budget deficit, expected to decline from 5% to 3%, and inflation, expected to decline from 10% to 7% (Republic of South Africa 1996: 8).

However, the neo-liberal GEAR is by no means hegemonic: an alternative economic and social development strategy remains viable. Most notably, in June 2000, the Millennium Council was launched at the initiative of Charles Nupen, the International Labour Organization's representative in South Africa. The Council is a bipartite chamber of representatives from business and organized labour, dedicated to reaching an Irish-style labour accord. Any agreement negotiated by this Council shall be presented as a fait accompli for approval by the statutory National Economic, Development and Labour Advisory Council (NEDLAC), a tripartite policy-formulating body (Daily Mail and Guardian 14/6/2000). Some feel that the Millennium Council could represent a fresh start, given that NEDLAC has gained an unfortunate reputation in recent years for being fractious and ineffective (Daily Mail and Guardian 14/6/2000). In the following sections, we examine labour accords in more detail, and the possibilities for their implementation i n South Africa, with or without the Millennium Council.

Labour Accords

A labour accord or social contract, social compact, or concerted action, as it is also called, is a negotiated agreement between one or more union federations, sometimes one or more employer federations, and the state. Unions typically offer industrial peace, political support for the government, and some degree of wage restraint in exchange for more pro-labour, public policies regarding training, unemployment insurance, social welfare, workers' compensation, or macroeconomic management (Wiarda 1997).

Labour accords are not always, or even usually, explicit agreements between the main labour market actors. They can emerge as the implied duties and obligations of a neo-corporatist relationship between employers, unions, and the state. Neo-corporatism, as practised in northern Europe since the War, emphasizes consensual decision-making, independent interest group representation, their democratic control by members, and voluntary participation in the organs of the state (Wiarda 1997). Representatives from the employer and union federations meet with government officials on a regular basis to try to develop consensual approaches to labour market problems. Employment policies are a key focus for discussion, but the contract outcomes negotiated by employers and unions are as well (Corvers and van Veen 1995).

A labour accord that provides wage restraint and industrial peace can furnish substantial advantages. First, a union commitment to wage restraint can substantially reduce cost increases and, through a flow-on effect, price inflation (AlMarhubi and Willett 1995; Newell and Symons 1987), without the need for restrictive monetary and fiscal policies that depress output and raise unemployment. On the contrary, wage restraint allows the government to pursue stimulative, but non-inflationary, macroeconomic policies that create jobs, expand output, and encourage investment (Broersma 1996; Bruno and Sachs 1985; Hicks and Kenworthy 1998). Second, a government commitment to more generous and accessible social welfare programs compresses the income distribution and reduces poverty, given that the primary beneficiaries of these programs, the working poor, the unemployed, the sick, the disabled, and the elderly, tend to have few, if any, alternative sources of income (Corvers and van Veen 1995; Goodin and LeGrand 1987; Gr een, Henley, and Tsakalotos 1994; Hicks and Kenworthy 1998). Moreover, active labour market programs for vocational training, mobility assistance, job placement, and recruitment subsidies increase employment (Kraft 1998), particularly for the most disadvantaged workers, the least skilled and the least experienced (Harcourt 1996). Structural Adjustment Programs, by contrast, normally entail the reduction or elimination of such programs in an effort to balance the government's budget (Taylor 1997). Third, a union commitment to industrial peace provides a more cooperative context for negotiating the restructuring of work and the introduction of new technologies to increase productivity. On the other hand, labour market deregulation and macroeconomic austerity engender the labour movement's mistrust and hostility (Tidjani 1998).

Whatever the desirability of a labour accord, none of the parties may be either able or willing to strike a comprehensive deal over social and economic policies. In other words, centralized bodies representing the interests of labour, employers, and the state must be able to strike a deal, and have the moral and/or legal authority to carry their constituents with them. Moreover, there must be strong incentives for them to do so. For organized labour, it may be greater job security and a 'better world' in improving the least endowed; for business, stability and greater productivity; and for the state, order and general social progress.

Institutional prerequisites of a Labour Accord: The ANC

The ANC is in an extremely strong position to close a tripartite deal without having to satisfy powerful opposition lobbies. The 1999 elections have confirmed that South Africa has reemerged as a dominant party system, with the ANC superseding the National Party as the 'natural party of government' (c.f. Southall 1993). In a dominant party system, a single political party repeatedly wins elections by a wide margin, with no registered political party proving capable of mounting a credible electoral challenge (Southall 1993). In South Africa's second democratic election held in 1999, the ANC increased its slice of the vote to 66%, up from 62% in the 1994 'liberation election' (Independent Electoral Commission (IEC) 1999). At the same time, the votes garnered by the former ruling party, now known as the New National Party, dropped from 20% to a dismal 7% (see IEC 1999). The Official Opposition, the neo-liberal Democratic Alliance (DA) (a merger of the formerly liberal Democratic Party, and the New National Party ) received less than 20% of all votes cast in 1999, but has yet to gain a significant following amongst the African community. Meanwhile, organizations with impeccable liberation credentials, such as the Azanian People's Organization and the Pan Africanist Congress, struggled to gain the 0.8% of the vote that is, in practice, required for parliamentary representation. Whilst both these organizations have, at various stages, committed themselves to 'African socialism', the PAC's policy trajectory has been most erratic, and incorporates extremely conservative elements, most notably with regard to law and order. The below table depicts the performance of the main parties in the 1994 and 1998 elections. The Democratic Party (DP), the Federal Alliance, and the (New) National Party (NNP) have since merged to form the Democratic Alliance, under the leadership of the DP.

A significant component of the electorate is dissatisfied with the government's performance to date. However, almost all the anti-ANC votes cast in 1994 and 1998 were drawn from the ranks of ethnic minorities (from which the DP, Freedom Front, Federal Alliance and NNP drew almost all their support), and from those located on the periphery of society. The latter would include those resident in the former homeland capitals of Mmabatho and Umtata, who opted for political parties headed by their former dictators, the United Christian Democratic Party and the United Democratic Movement, respectively (see IEC 1999). The predominantly Zulu Inkhatha Freedom Party gained 9% of the votes in the 1999 elections, and has re-entered into a close coalition with the ANC; Inkhatha leaders have been willing to lend their support to the government in return for a proportion of the spoils of governing; Inkhatha leader, Chief Buthelezi, is now a senior cabinet minister.

South Africa's governmental structures prevent internal dissension within the governing party from inhibiting its capacity to pass laws, irrespective of its level of internal diversity. Legislative authority is invested in just one house of parliament. The executive and legislature have no separate, constitutionally defined powers. The constitution bars MPS from switching political allegiance without forfeiting their seats. Power is centralized; the presidency is an indirectly elected office that devolves on the leader of the largest party in parliament. Whilst limited power is devolved to provincial level, the ANC is the majority party in all but two of South Africa's nine provinces, and a coalition partner in one other.

This leads us to the contentious question as to where a credible opposition party could come from. By far the largest national movement outside of the ANC has been the labour movement. This has led those within the ultra-left camp to argue that the unions--given the ANC's espousal of neoliberalism in the face of union opposition--should form an independent 'workers' party' (see Habib and Taylor 1999: 114). It is argued that-given the persistence of widespread inequality--such a party would be likely to gain mass support in subsequent elections, particularly if it campaigned from an explicitly socialist platform. However, both the 1994 and 1998 nationwide surveys of COSATU members, conducted, inter alia, by one of the author's of this paper, revealed a deeply seated loyalty towards the ANC (see Ginsburg et al 1995; Southall and Wood 1999; Wood and Psoulis 2001). Over 70% of respondents to the 1998 survey believed that the Alliance remained the most effective mechanism for representing workers in parliament; tw o-thirds believed that the Alliance should continue and contest the scheduled 2004 elections (Southall and Wood 1999: 76). Only 10% of respondents favoured the formation of an independent workers party and 4% an Alliance between the South African Communist Party and COSATU alone.

However, a strategic retention of the Alliance need not entail the emasculation of the union movement, and its becoming a hapless hostage to broader economic forces. The unions have managed to ensure the ANC stays on track in terms of the delivery of basic social services, and helped secure the retention of a large body of labour legislation--centring on the 1995 Labour Relations Act--that is broadly union--friendly, despite external pressures (c.f. Southall and Wood 1999). We would not reject the formation of an independent workers party at some future date; however, the unions have much to lose by abandoning the Alliance for an uncertain electoral future at this time, a point recognized by both union leadership and rank-and-file (Southall and Wood 1999). The continued support for the Alliance is likely to secure the ANC's dominant position until the 2004 elections at least, and probably until 2009.

The incentives for a Labour Accord: The ANC

Even though the ANC very likely has the capacity to implement union- and worker-oriented policies, it may not have the desire to do so. The ANC does have close connections to the trade union movement to the extent that 20 former COSATU officials sit as ANC members of parliament (Boroughs 1996: 40) whilst considerably more hold senior positions in the ANC government (Barrett 1996: 37). However, the 20 'labour' MPS, internally selected by COSATU, have had to formally sever their links with the union federation prior to entering parliament.

The lack of any direct union influence on ANC policy would not be a problem if other ANC-affiliated groups pushed labour's concerns. However, the ANC is not committed to socialism or any specifically labour-oriented philosophy, despite the pro-union sympathies of some of its leaders. The ANC represents an extremely broad coalition of interests, and has subsumed much political competition into internal debate (see Southall and Wood 1999). On the one hand, the Alliance has, as noted earlier, resulted in a continuing emphasis being placed on service delivery and on the retention of a broad body of labour friendly legislation. On the other hand, the unions have been patently unable to check the ANC's conservative drift in other areas.

Institutional prerequisites for a Labour Accord: South African employers

South Africa's unemployment rate is above 30%, and so there are presently no employer institutional prerequisites for a labour accord. However, if unemployment were to fall dramatically, many individual employers would eventually compete against each other by offering more than the wage norms established under a labour accord, particularly to skilled workers in short supply. In these circumstances, strong employers' federations would need to assume central control over wage-setting, a role which present federations could not fulfil. The South African business community has traditionally been divided along ethnic and racial lines between the dominant English-speaking community, the formerly politically influential Afrikaans community, and a rapidly expanding African business community that became a major player in the 1990s. There are three separate umbrella employers associations: the South African Chamber of Business (SACOB), the Afrikaanse Handelsinstituut (AHI), and the National African Federated Chambers of Commerce (NAFCOC), each serving one of these three distinct communities, respectively. None of these bodies has any formal executive powers, either in terms of their own constitution, or under law. Although major employers in particular industries have joined together in industry-wide employer associations for collective bargaining purposes, it has only been in the 1990s that attempts have been made to establish a united national employers' body that transcends racial and ethnic barriers, but these have been largely unsuccessful. In July 2000, NAFCOC and SACOB announced their intention to merge; however, this aim has yet to be realized. Moreover, the bulk of South African employers, especially small and medium-sized enterprises, are not affiliated to any of the major associations. Recently, SACOB's key Johannesburg region has disaffiliated from the national federation, further fragmenting employer associations. Meanwhile, a new employer association, Confederation of Employers of South Africa (COFESA), has centred its activities on promoting and facilitating greater deregulation and individualization of employment contracts. COFESA claims to have 120,000 member companies, but these are mostly small or medium-sized enterprises; COFESA lacks political clout and has been shunned by South Africa's largest employers.

Perhaps the greater problem is that the 1995 South African Labour Relations Act allows many employers to avoid dealing with unions, thereby enabling them to escape the constraints of any labour accord. Unions that organize more than 50% of a plant's workforce are automatically entitled to employer recognition, and a range of organizational rights pertaining, for example, to shop steward structures and union access to the workplace (Wood 2000). Nevertheless, South African employers have been able to gradually de-unionize and individualize their workplaces for three reasons. First, newer unions, and older unions operating in newly organized and growing sectors, enjoy relatively little legal protection compared to older unions in more traditionally organized but shrinking sectors (Wood 2000). There is no obligation in the 1995 Labour Relations Act for employers to bargain in good faith. Other provisions in the Act are widely flouted in more conservative rural areas and in the informal sector, where local law enf orcement agencies are unlikely to be sympathetic to the cause of organized labour.

Second, some employers have exercised their right to withdraw from their bargaining council, leading in some cases to the council's demise and the collapse of collective bargaining in some sectors. The law does not require every industrial sector to have a council, and so the withdrawal of the main trade union or employer association automatically leads to a council's collapse (Wood 2000). However, councils continue to operate in key areas such as the metal industry, and a new council was established in the chemical industry in the mid-1990s.

Incentives for a Labour Accord: South African employers

South African employers may choose one of two strategies to secure a more cost-effective workforce. First, they can try to work more cooperatively with unions, seeking broad wage agreements at national level, and productivity-based deals at plant level. Alternatively, they can exploit loopholes in the 1995 Labour Relations Act to promote greater individualization. Their decision either way is likely to be prompted by the extent of perceptions of crisis; closer relations may expedite the normalization of employment relations. Webster (1998: 6) notes that white racism led, in the 1980s, to the emergence of 'militant abstentionism, a refusal to identify with any of the goals of the enterprise, or the concerns of management. This culture of resistance merged with the insurrectionary political climate of the time creating conditions of ungovernability at the workplace.' This deeply rooted ethos of non-collaboration persists in high levels of absenteeism, a general unwillingness to make life too easy for management or their representatives, and in periodic outbreaks of violent 'wildcat' protests at the workplace (Southern African Report 1998). The unions continue to represent a stabilizing influence in many South African workplaces, and so many larger South African employers remain committed to a long-term partnership with organized labour.

The relative position of employers has improved greatly under the new order. The incidence and duration of strikes has gradually decreased through the 1990s. Reflecting a growing reluctance by union officials to make use of the strike weapon, wage settlements over the past few years have been modest and more closely linked to the inflation rate (see Webster and Psoulis 1999). However, the fragility of the situation has been underscored by the 1998 'winter of discontent,' when a number of relatively large strikes took place, the more violent instances at odds with formal union policy (Webster 1999: 9). In 1999, the volume of strike action declined again, but future outbreaks remain likely. Even amongst the ranks of those employers who would prefer radical deregulation as their ideal, many believe that, as long as the bargaining council system remains government policy, they should use it to their advantage (Webster 1999: 8).

Institutional prerequisites of a Labour Accord: the South African unions

It is evident from the above that there are a number of bureaucratic barriers to a labour accord in South Africa. In addition, there are possible contradictions concerning the role of unions. On the one hand, successful accords depend on the parties having a broad social base. On the other hand, the viability of any accord depends on partial demobilization (unions have to deliver their supporters in underwriting a deal, effectively keeping the rank-and-file in line). This contradiction has sometimes been referred to as the 'Olsonian trap' (c.f. Kelly 1998).

The union federations are unlikely to obtain ANC agreement to even consider a labour-oriented development strategy, without first offering labour peace and wage restraint. The ANC needs labour peace to facilitate and speed the restructuring of the economy and thereby improve international competitiveness and efficiency. It needs wage restraint to encourage investment, especially in the labour-intensive sectors, and to ensure that tight monetary and fiscal policies lower the inflation rate rather than raise the unemployment rate (Calmfors and Driffil 1988; Sachs and Bruno 1985). However, the union federations are generally unable to make such concessions. They have no right to bargain collectively, veto the negotiated settlements of their affiliates, or control access to strike funds (Macun 1995). Neither do union leaders have the formal power to stop union members from striking or to enforce guidelines for maximum wage increases.

In many sectors, collective bargaining in South Africa takes place at the industry level through multi-union, multi-employer Bargaining Councils. The Minister of Labour then typically extends council agreements to non-signatory unions and employers, irrespective of whether or not they are unionized, in the same industry. However, the Council agreements only establish the minimum terms and conditions, which leaves the parties free to negotiate additional terms and conditions through plant-level bargaining.

The federations could try to use a membership referendum to promote the voluntary adoption of national guidelines on wage restraint in industry- and enterprise-level bargaining. The success of this strategy would very much hinge on winning such a vote. However, recent South African experience suggests that such a vote would be difficult to win. Years of intense conflict, precipitated, at least in part, by a system of 'racial Fordism' at the workplace, have radicalized workers, so that calls for wage moderation and cooperation with management would probably be perceived as signs of co-optation. Indeed, there have been periodic rumblings from radical grassroots elements against what has been perceived as a too conciliatory stance by centralized leadership. This took the form of highly public criticism of COSATU's leadership by the Witwatersrand Stewards' Council in 1995. The Council called for an end to the Alliance between COSATU and the ANC (Von Holdt 1995: 21), and attacked what it saw as excessively close t ies between the unions and the ANC-led government and increasing levels of bureaucratization within the unions (Von Holdt 1995: 21). Since then, there has been a proliferation of what are often neo-Trotskyite breakaway unions, such as the colourfully named 'Mouthpiece' and 'Turning 'Wheels' unions.

Union members expect shop stewards to represent their interests rather than serve as mediators between the membership and management. In the 1998 survey of COSATU members, 59% of respondents felt that shop stewards should consult with their members every time the stewards acted on their members' behalf, and 93% felt that their shop steward should be dismissed, if he or she failed to represent the members adequately (see Wood and Psoulis 2001). A range of procedures is used to keep leaders in step with their members, including: annual or biennial elections, weekly or monthly meetings, mandates to perform specific duties, and 'reportback' sessions to inform members of important decisions made on their behalf. Seventy one percent of respondents admitted to attending and participating in union meetings at least once a month, with the express purpose of holding union leaders to account. More than a third even admitted to having had a personal experience of removing an errant shop steward from office (Wood 1999).

The 1998 Employment Equity Act may have worsened membership suspicions of their leaders. The Act requires employers of 50 or more employees to develop an Employment Equity Plan to attract, develop, and advance Africans, women, and other 'designated groups'. Employers have increased African representation in management, in compliance with the Act, by hiring shop stewards, who are often seen as more articulate and informed than their peers. These promotions have undermined the confidence of members in their leaders, and have produced an increase in the incidence of shop steward ejections from office (Wood 1999). As suspicions have intensified, stewards have been placed under even greater pressure to articulate the demands of the shop floor, even if this is at odds with formal union policy. For instance, shop stewards sided with a large group of disaffected workers against management and senior union officials in the violent 2000 strike at Volkswagen's Uitenhage plant. Although the dispute was terminated with th e dismissal of the strikers, it did highlight some of the underlying tensions within the labour movement.

Disaffected members can also retaliate against their leaders by easily switching their membership to another union. South African unions actively compete against each other for members in areas such as chemicals, construction, and, above all, the public sector, which became one of the main areas of union growth in the 1990s. It is therefore unlikely that any union federation would unilaterally agree to limit wage increases, as part of its obligations under an accord, for fear that its members would desert to other better paid unions (Albertyn and Rycroft 1995). This is particularly true of the affiliates of CASATU and the Federation of Unions of South Africa (FEDUSA), which have recently clashed in their rush to organize the same groups in the public sector.

However, the above discussion should be followed on by a caveat. We reject the line peddled by right wing South African newspaper columnists--but often echoed by the ultra-left--that the that the black working class constitutes an insatiable rabble, whose demands will preclude any compromise deal (Southall and Wood 1999). The 1998 survey of COSATU members revealed not only a deeply entrenched culture of accountability--which, in effect, would limit union leadership's room for manoeuvre at any neo-corporatist bargaining table-but also a deep-seated loyalty toward the ANC (Southall and Wood 1999). This loyalty is not just a legacy of the liberation struggle. It also reflects a recognition that there has been material delivery in a range of areas since 1994. Union members specifically highlighted improved access to healthcare and telecommunications, and to domestic electricity and water, all areas which the ANC has prioritised (ibid.). In short, despite the ANC's patchy record-and many of its actions have indeed been somewhat inimical to the labour movement--union members were willing to give credit where it was due. Their attitudes toward any labour accord are likely to be similarly pragmatic; if a deal delivers material benefits, it is likely to gain the backing of union members. At the same time, there is little doubt that union members expect continuous improvements. The 1998 survey also revealed that union members were willing to engage in mass action should the ANC fail to deliver in the future (Southall and Wood 1999).

Thus, the viability of a labour accord would depend on tangible benefits provided to union members. In turn, this would depend on concessions which the state and capital may be unwilling to make on an ongoing basis. Ironically, even though the state and capital may have the most to gain from an accord--the possibility of greater stability and (if demobilization takes place) the indefinite deferment of more radical change--they are, in the end, the groups most likely to hasten their own destruction. However, sections of capital may at some stage be willing to make some concessions to lure labour to a national bargaining table; strategically, unions could make much by grasping such concessions, whilst keeping their long-term options open.

Incentives for a Labour Accord: The South African unions

South African unions would be reluctant to sacrifice the possibility of securing further wage increases via conventional collective bargaining for improvements in social programs. First, the ANC's past record suggests that major reforms and expenditure increases would be unlikely, irrespective of what it might be prepared to promise and what it is institutionally capable of delivering. The government has whole-heartedly embraced an orthodox economic program through the GEAR strategy, which it has single-mindedly and unwaveringly pursued since 1996 to the complete exclusion of the RDP and more labour-oriented policies advocated by unions (Saul 1999). The ANC also remains committed to a contractionary and anti-inflationary, monetarist role for monetary policy rather than an expansionary and reflationary, Keynesian one. Furthermore, the Party's continuing success at the polls is not forcing any re-think of policy choices. Even if the Party were prepared to enact policies to please the labour movement, large ideo logical differences between the union federations make it unlikely that they would agree to the same terms and conditions in any labour accord (Macun 1995).

Second, since the early 1980s, union members have achieved large wage gains through aggressive bargaining and strike action, despite high and rising unemployment. Trade union members earn almost 25 percent more than comparable non-union workers in the formal sector (Moll 1993: 260), whilst their lowest paid members typically earn two to three times more than average workers in the informal sector (Moll 1996: 328). As a result, the perceived benefits of pursuing union goals via the labour market are considerable, especially when compared to the political arena (Wood 1995).

Whither a South African Labour Accord?

It may seem that neo-corporatism represents a policy option of little relevance to a semi-industrialized economy in the developing world. However, the desire to promote a neocorporatist option has featured strongly in South African policy debates, and is reflected in the formation of NEDLAC, and, more recently, the Millennium Council. Nonetheless, the process of liberalization in South Africa has gathered a momentum that may appear to be too difficult to check in the short- or medium- terms. The ANC government has recently indicated its willingness to facilitate greater labour market deregulation through amendments to the 1995 Labour Relations Act. The 1999 elections proved that the ANC remains extremely popular despite the sacrifice of 'any serious project of political mobilization...to the primacy of the marketplace' (Saul 1999:62). However, the limited dividends accrued by the orthodox GEAR strategy could well put a labour accord firmly back on the policy agenda. In particular, the recent creation of the M illennium Council could open up exciting new possibilities for a labour accord. However, a successful agreement hinges on two factors. First, as noted earlier, the major parties to an accord must be motivated by a clear sense of crisis. In other words, the costs of not reaching an accord must greatly outweigh any benefits that might accrue from retaining the status quo. Major unemployment and income inequality may seem to be crisis enough, given that their persistence could provide the basis for future political instability. However, South Africa's economic performance in the 1990s, although lacklustre by international standards, still represented a considerable improvement on the recession of the 1980s. Moreover, the political stability that has emerged since the first free election in 1994 has meant that the principal employer associations are far less committed to reaching compromises with organized labour than in 1995 when they endorsed the relatively labour-friendly Labour Relations Act.

Second, the chances of an accord succeeding would be greater if the 1995 Labour Relations Act were revisited to discourage second or third tier 'top up bargaining' and contracting out of all of the Act's provisions. However, the institutional foundations for an accord would have to be more solid than simply a body of legislation, particularly given pressures from potential overseas investors and sections of domestic business for greater diversity in employment practices in a range of areas. The European experience highlights the institutional depth needed for an accord to succeed: accords are reconstituted when possible alternatives seem remote ones; when union memories of effective action at plant level are widespread and recent, the prospects for an accord are necessarily weaker. Unfortunately, this may make any accord rather less than durable in the South African case.

Finally, we return to a question raised earlier in the paper: is neo-corporatism a worthy option for progressives to pursue in a society marked by both gross social inequality, and a rich tradition of activism and protest? Here a distinction should be drawn between what is desirable in the long term and what is feasible in the short term; the ideological purity of the South Africa ultra-left is matched by its political weakness and internal divisions. However, the politics of the feasible should not be dominated by the essentially liberal pragmatism of writers such as Sklar (Saul 1999). Rather, what are necessary are policy alternatives that highlight the ameliorating effects of institutions. Indisputably, the currency of neo-corporatism is that of messy compromise. Nonetheless, neo-corporatism's track record underscores the role institutions can play in redressing social inequality, yet creating the conditions for economic growth; a successful socialist alternative would itself have to depend on effective an d ultimately inclusive institutional configurations to secure more radical transformation, subordinating the needs of capital to the good of society as a whole. It is unlikely that a labour accord in South Africa would result in a reconciliation between the desire for capital accumulation and the pressures for social justice. However, unions could use an accord as a means of extracting vital concessions, and ensuring that the social question stays firmly on the national policy agenda. A culture of deeply entrenched internal democracy within South Africa's major unions would preclude the mass demobilization necessary to make an accord sustainable indefinitely; at the same time, surveys of union members indicate that the rank and file would be willing to give their leadership some room for manoeuvre in return for material gains. Through a process of strategic engagement, unions could make worthwhile advances, whilst not abandoning a more radical long-term agenda (Waterman 1999); we would reject the determinist viewpoint that individual interventions lack the power to impact on broader economic trends. Ironically, whilst employers could--and, in other national cases, have--use an accord to politically incorporate the unions, this is unlikely to occur in South Africa. This would reflect both the rich culture of mass mobilization and internal democracy within the labour movement, and the deeply-divided nature of capital in South Africa.

Table 1

National election results 1994 and 1999

Party                               Percentage  Percentage  Seats  Seats
                                    Vote 1994   Vote 1998    1994   1994

ANC                                   63.12       66.35       252    266
Democratic Party                       1.75        9.56         7     38
Inkhatha Freedom Party                10.62        8.58        43     34
New National Party                    20.55        6.87        82     28
United Democratic Movement              N/A        3.42       N/A     14
African Christian Democratic Party     0.45        1.43         2      6
Pan Africanist Congress                1.26        0.71         5      3
United Christian Democratic Party       N/A        0.78       N/A      3
Freedom Front                          2.19        0.8          9      3
Federal Alliance                        N/A        0.54       N/A      2
Afrikaner Eenheids Beweging             N/A        0.29       N/A      1
Azanian Peoples Org.                    N/A        0.17       N/A      1
Minority Front                         0.07        0.3          0      1
Others                                             0.19         0      0

Total                                   100         100       400    400

(Independent Electoral Commission 1999).

References

Adler, Glen and Eddie Webster (1995). 'Challenging Transition Theory: The Labour Movement, Radical Reform and the Transition to Democracy in South Africa'. Paper presented at Albert Einstein Initiative Workshop, University of Witwatersrand, Johannesburg.

AI-Marhubi, Fahim, and Thomas D. Willett (1995). 'The anti inflationary influence of corporatist structures and central bank independence: The importance of the hump shaped hypothesis'. Public Choice, Vol. 84, pp. 153-162.

Albertyn, Chris, and Alan Rycroft (1995). 'Multi-union Collective Bargaining Structures in South Africa: Case Studies'. British Journal of Industrial Relations, Vol. 33, pp. 329-43.

Barrett, J. (1996). 'Trade unions in South Africa: dramatic change after apartheid ends'. Monthly Labour Review, Vol. 119:37-41.

Baskin, Jeremy (1991). Striking Back: A History of COSATU. Johannesburg, South Africa: Ravan.

Bauer, Peter (1984). Reality and Rhetorie: Studies in the Economics of Development. Cambridge, Massachusetts: Harvard University Press.

Bocarro, Lucio (2000). 'Centralized Collective Bargaining and the Problem of "Compliance": Lessons from the Italian Experience'. Industrial and Labour Relations Review, Vol. 53, No. 4.

Boroughts, D. (1996). 'First Freedom, then jobs; South Africa has staggering unemployment that threatens the country's political compact'. US News and World Report, August 19, p. 40.

Bray, Mark and PatWalsh (1995). 'Accord or Discord: The Differing Faces of Corporatism under Labo(u)r Governments in Australia and New Zealand'. Labour and Industry, Vol. 6, No. 3, pp. I-26.

Broersma, Lourens. (1996). 'The effect of wage restraint on labour market flows'. Applied Economics, Vol. 28, pp. 999-1007.

Calmfors, Lars, and John Driffil (1988). 'Centralization of wage bargaining'. Economic Policy, Vol. 6, pp. 14-61.

Casey, Bernard and Michael Gold (2000). Social Partnerships and Economic Performance. Cheltenham, United Kingdom: Edward Elgar.

Central Statistics Services (1996). RSA Statistics in Brief. Pretoria, South Africa: Central Statistics Services.

Collier, Ruth and David Collier (1991). Shaping the Political Arena: Critical Functures, the Labour Movement, and Regime Dynamies in Latin America. Princeton, New Jersey: Princeton University Press.

Constitution of the Republic of South Africa (1996). Government Gazette-Act 108 of 1996. Pretoria, South Africa: South Africa Government Printer.

Corvers, Frank and Tom van Veen (1995). 'On the Measurement of Corporatism'. Labour, Vol. 9, No. 3, pp. 423-442.

Crouch, Colin Industrial Relations and European State Traditions. Oxford, United Kingdom: Clarendon.

Dolvik, Jan and Stokke, Torgeir (1998). 'Norway: The Revival of Centralized Concertation'. Ferner, Anthony and Hyman, Richard (eds.), Changing Industrial Relations in Europe. Oxford: Blackwell.

Douwes Dekker, Louet (1995). 'Labour Market Flexibility in South Africa: A Preliminary Investigation'. South African Journal of Labour Relations, Vol. 19, pp. 26-48.

Fine, R., and G. Van Wyk (1996). 'South Africa: State, Labour, and the Politics of Reconstruction'. Capital & Class, Vol. 58, pp. 19-31.

Ginsburg, D. et al. (1995). Taking Democracy Seriously. Durban: IPSA.

Gordon, Colin (1998). 'Why no Corporatism in the United States: Business Disorganization and its Consequences'. Business and Economic History, Vol. 27, No. 1, pp. 29-46.

Goodin, Robert, and Julian LeGrand (1987). Not Only the Poor: The Middle Classes and the Welfare State, London: Allen and Unwin.

Green, Francis, Henley, Andrew, and Euclid Tsakalotos (1994). 'Income Inequality in Corporatist and Liberal Economies: A Comparison of Trends Within OECD Countries'. International Review of Economics, Vol. 8, No. 3, 303-331.

Gunnigle, Patrick, MacCurtain, Sarah, and Michael Morley (2000). 'Pluralist in Theory, Unitarist in Practice: Industrial Relations Management in Irish Greenfields Sites'. Personnel Review. In print.

Habib, Adnan and Rupert Taylor (1999). 'Daring to Question the Alliance: A Response to Southall and Wood'. Transformation, 40: 112-120.

Habib, Adnan and Vishnu Padyachee (2000). 'Economic Policy and Power Relations in South Africa's Transition to Democracy'. World Development, Vol. 28, No. 2, pp. 245-263.

Hampson, Ian and David Morgan (1998). 'Continuity and Change in Australian Industrial Relations'. Relations Industrielles, Vol. 53, No. 3, pp. 564-589.

Harcourt, Mark (1996). 'Unemployment Reduction in Canada: Lessons from Japan and Sweden'. Ralations Industrielles, Vol. 51, No. 1, pp. 177-202.

Hicks, Alexander, and Lane Kenworthy (1998). 'Cooperation and Political Economic Performance in Affluent Democratic Capitalism'. American Journal of Sociology, Vol. 103, No.6, pp. 1631-1672.

Holden, Steinar (1998). 'Wage Drift and the Relevance of Centralised Wage Setting'. Scandinavian Journal of Economics, Vol. 100, No. 4, pp. 711-731.

Hyman, Richard (1975). Industrial Relations: A Marxist Introduction. London: Macmillan.

----- (1986). 'British Industrial Relations: The Limits of Corporatism'. Otto Jacobi, Bob Jessop, Hans Kastendiek and Marino Regini (eds), Economic Crisis, Trade Unions and the State. London: Croom Helm.

Independent Electoral Commission (1999). 'Election Results'. Pretoria, South Africa: South Africa Government Primer.

Johnston, Alexander (1994). 'South Africa:The election and the transition process-five contradictions in search of a resolution'. Third World Quarterly, Vol. 15, pp. 187-204.

Johnston, Paul (1994). Success While Others Fail:Social Movement Unionism and the Public Workplace. Ithaca, New York: Industrial and Labour Relations Press.

Kelly, John (1998). Rethinking Industrial Relations: Mobilizarion, Collectivism and Long Waves. London: Routledge.

Korpi, Walter (1983). The Democratic Class Struggle. Boston: Routledge and Kegan Paul.

Kraft, Kornelius (1998). 'An evaluation of active and passive labour market policy'. Applied Economics, Vol. 30, pp. 783-93.

Krueger, Anne (1988). Trade and Employment in Developing Countries. London: University of Chicago Press.

Layard, Richard, Nickell, Stephen, and Richard Jackman (1994). Unemployment: Macroeconomic Performance and the Labour Market. Oxford, United Kingdom: Oxford University Press.

Lazar, David (1996). 'Competing Economic Ideologies in South Africa's Economic Debate'. British Journal of Sociology, Vol.47, No. 4, pp. 599-626.

Levitan, Sar (1982). Programs in Aid of the Poor. Baltimore:John Hopkins Press.

Lind, Jens (1999). 'Tripartism and Collective Bargaining. The Social Pact on Employment in Denmark'. International Journal of Employment Studies, Vol. 7, No. 1, pp. 33-51.

Lipset, Seymour Martin (1990). Continental Divide: The Values and Institutions of the United States and Canada. New York: Routledge, Chapman, and Hall, Inc.

MacDonald, M., and Wilmott James (1993). 'The Hand on the Tiller: the Politics of State and Glass in South Africa'. Journal of Modern African Studies, Vol. 31, pp. 387-405.

Macun, Ian (1995). 'Growth, Structure and Power in the South African Union Movement'. Paper presented at the 2nd International Conference on Emerging Union Structures. The Swedish Institute for Work Life Research, Stockholm, 11-14 June.

Macun, Ian and Eddie Webster (1998). 'Recent Developments in South African Industrial Relations and Collective Bargaining: Continuity and Change'. South African Journal of Labour Relations, Vol. 22, No. 1, pp.35-50.

Mare, Gere (1995). 'Ethnicity, Regionalism and Conflict in a Changing South Africa'. South African Journal of International Affaris, 3,1: 1-21.

Maree, Johann (1993). 'Trade Unions and Corporatism in South Africa'. Transformation, Vol. 23, pp.24-54.

_____(1995). 'The Changing Role and Perception of Members as Unions Grow: The Congress of South African Trade Unions.' Paper presented at the 2nd International Conference on Emerging Union Structures. The Swedish Institute for Work Life Research, Stockholm, 11-14 June.

Mboweni, Tito (1995). 'Interview'. South African Labour Bulletin, Vol. 19, No. 1, pp. 23-30.

Moll, Peter (1993). 'Black South African Unions: Relative Wage Effects in International Perspective'. Industrial and Labour Relations Review, Vol. 46, pp. 245-261.

______ (1996). 'Compulsory Centralization of Collective Bargaining in South Africa' American Economics Association Papers and Proceedings, Vol. 86: 326-329.

Murray, Alan, and Yonatan Reshef (1988). 'American Manufacturing Unions' Stasis: A Paradigmatic Perspective'. Academy of Management Review, Vol. 13, No. 4, pp. 639-652.

Nattrass, Nicola (1998). 'A Social Accord in South Africa? Lessons from Sweden and Australia'. Unpublished working paper, School of Economics, University of CapeTown, CapeTown.

Newell, A. and Symons, J.S.V. (1987). 'Corporatism, Laissez-Faire and the Rise in Unemployment'. European Economic Review, Vol. 31, pp. 567-614.

Olson, Mancur (1982). The Rise and Decline of Nations: Economic Growth, Stagflation and Social Rigidities. New Haven:Yale University Press.

Piore, Michael (1991). 'The Future of Unions'. In Strauss, George, Gallagher, Daniel, and Jack Fiorito, eds., Madison, Wisconsin: Industrial Relations Research Association.

Republic of South Africa (1996). Growth, Employment and Distribution. Pretoria, South Africa: Republic of South Africa.

_____ (1997). White Paper for Social Welfare. Pretoria, South Africa: Republic of South Africa.

Rigby, Mike and Teresa Lawlor (1994). 'Spanish trade unions 1986-1994; life after national agreements'. Industrial Relations Journal, Vol. 25, No. 4, pp. 258-271.

Roberts, Michael (1999). 'The Future of Labour Unions: A Review'. Monthly Labour Review, Vol. 122, No. 10, pp. 38-39.

Ross, David (1981). The Working Poor: Wage Earners and the Failure of Income Security Policies. Toronto: James Lorimer and Company.

Sachs, Jeffrey, and Michael Bruno (1985). Economics of Worldwide Stagflation. Oxford, United Kingdom: Basil Blackwell Limited.

Saul, John (1999). 'Magical Market Realism'. Transformation, Vol. 38, pp. 49-67.

Shubane, Kehle and Louise Stack (1999). 'Election 1999: Prospects for a Consolidation of Democracy'. Issue, Vol. 27, No. 2, pp. 2-8.

Sisson, Keith (1999). 'The "New" European Social Model: The End of the Search for Orthodoxy or Another False Dawn'. Employee Relations, Vol. 21, No. 5, pp. I-54 (web copy).

South African Institute of Race Relations (1997-8). Southern Africa Survey. Johannesburg, South Africa: South African Institute of Race Relations.

Southall, Roger (1993). 'Constructing a Culture of Consent: Dilemmas in the Making of a South African Democracy'. IAAS, Vol. 28, Nos. 3-4, pp: 259-272.

Southall, Roger and Geoffrey Wood (1999). 'COSATU and the ANC: Whither the Alliance?' Transformation, No. 38: pp. 68-83.

Suttner, Ray and Jeremy Cronin (1986). Thirty Years of the Freedom Charter. Johannesburg, South Africa: Ravan Press.

Tarantelli, Ezio (1986). 'The Regulation of Inflation and Unemployment'. Industrial Relations, Vol. 25, No. "pp. 1-15.

Taylor, Lance. (1997). 'Editorial: The Revival of the Liberal Creed--The IMF and the World Bank in a Globalized Economy'. World Development, Vol. 25, No. 2, pp. 145-152.

Tidjani, Bassirou (1998). 'African Unions Under Structural Adjustment Programs'. Relations Industrielles, Vol. 53, No. 2, 278-299.

Visser, Jelle (1998). 'Two Cheers for Corporatism, One for the Market: Industrial Relations, Wage Moderation and Job Growth in the Netherlands'. British Journal of Industrial Relations, Vol. 36, No. 2: 269-292.

Von Holdt, Karl 'Cosatu's special Congress: Uncertain New Era'. South African Labour Bulletin Vol. 17, pp. 18-26.

_____ (1994). 'What in the Future of Labour?' Eddie Webster et al (eds.), Work and Industrialization in South Africa. Johannesburg: Ravan.

_____ (1995). 'The LRA Agreement'. South African Labour Bulletin, Vol. 19, No. 4, pp. 16-26.

Wachter, H. (1997). 'German Co-Determination: Quo Vadis?' Employee Relations, Vol. 19, No. 1, pp. 1-32 (Web copy).

Waterman, Peter (1999). "The New Social Unionism: A New Model for a New World Order'. Rinaldo Munck and Peter Waterman (eds.), Labor Worldwide in the Era of Globalization. London: Macmillan.

Webster, Eddie (1998). 'Race, labour Process and Transition: The Sociology of Work in South Africa'. Unpublished paper presented at the Annual Conference of the South African Sociological Association, Rand Afrikaans University, Johannesburg, July 1998.

_____ (1999). 'ILO Research Project on the Contribution of Collective Bargaining to Employment Protection or Creation and to Competitiveness: The Case of South Africa'. Unpublished report, Sociology of Work Programme, University of Witwatersrand Johannesburg, South Africa.

Webster, Eddie and Christene Psoulis (1999). 'Innovations in Collective Bargaining: An Overview of the Evolving labour Relations System during 1999'. Paper presented at Industrial Relations Association of SouthAfrica Annual Conference, Cape Town, South Africa.

Weeks, John (1999). 'Stuck in Low GEAR? Macroeconomic policy in South Africa, 1996-98'. Cambridge Journal of Economics, Vol. 23, pp. 795-811.

Wiarda, Howard (1997). Corporatism and Comparative Politics: The Other Great 'Ism'. Armonk, NewYork: M. E. Sharpe.

Williamson, John (1989). What Washington Means by Policy Reform'. In John Williamson, ed., Latin American Adjustment.. How Much Has Happened? Washington D.C.: Institute for International Economics.

Wood, Geoffrey and Mark Harcourt (1998). 'The Rise of South African Trade Unions'. Labour Studies Journal, Vol. 23, No. 1, pp. 74-92.

Wood, Geoffrey (1995). 'Comprehending South African Strike Action 1950-c.1995'. PhD Thesis. Rhodes University, Grahamstown, South Africa.

_____ (1999). 'Shop floor Democracy in the Congress of South African Trade Unions in the 1990s'. Unpublished working paper, Centre for European Human Resource Management, Cranfleld University, Cranfield, United Kingdom.

_____ (2000). 'South African Trade Unions in a Time of Adjustment'. Labour/Le Travail, No. 133-150.

Wood, Geoffrey and Psoulis, Christene (2000). 'Globalization, Democratization, and Organized labor in Transitional Economies. Work and Occupations, Vol. 28, no. 3, pp. 293-314.

World Bank (1995). World Bank Development Report 1995. Oxford, United Kingdom: Oxford University Press.

Mark Harcourt is an associate professor in the Department of Strategic Management and Leadership at the Waikato Management School, University of Waikato in Hamilton, New Zealand. He has published work on unions, discrimination, health and safety, and corporatism.

Geoffrey Wood is Professor of Comparative HRM at Middlesex University, and Overseas Research Associate of the Sociology of Work Programme, University of Witwatersrand. He has published widely on union renewal, the political economy of lusophone Africa, contemporary institutionalist theory, and workplace participation.