587. A major theme running throughout our Report and the ILO's Review is that voice regulation, namely, negotiation through democratic institutional structures, is both socially desirable and economically efficient. More co-operative relations at firm, industrial, regional and national level should encourage investment and productivity growth and allow for the balancing of business's need for competitiveness and adequate profitability, labour's need for secure and reasonably remunerated employment, and society's need for rapid employment creation. The case for a negotiated approach is all the stronger given the challenges posed by the increasingly globalised and competitive environment, by South Africa's high levels of unemployment, poverty and inequality, and by the uncertainties that accompany a period of fundamental economic restructuring.
588. In light of these challenges, the Commission argues that the practice of voice regulation at the national level must be enhanced by means of a nationally negotiated Accord for Employment and Growth, involving all of the social partners ­ business, labour and government. Such an Accord will provide the means by which macroeconomic policy and labour market policy may be co-ordinated, so that they work together to foster rapid employment growth. It entails deliberations between the social partners in respect of national policy and commitments around wages, prices and investment, along with a discussion of central government's contribution in terms of its expenditures on the RDP, the social wage, public works programmes and assistance from the Social Plan Fund, its tax policies, its industrial and trade policies, its competition policies, and its macroeconomic interventions. The Commission holds, and will argue below, that such an Accord is a crucial element of a successful growth strategy.
589. The Accord responds to the challenge posed in our terms of reference which require that we examine the "potential role of national tripartite institutions in income determination". In particular, as will be elaborated below, the formulation of the proposed Accord would entail both an expansion and a refocusing of the agenda of the National Economic Development and Labour Council (NEDLAC), and a renewed commitment by the social partners to the principle and practice of tripartite economic guidance.
590. The national Accord would also set a framework within which complementary multipartite regional, local, sectoral, and corporate initiatives could be negotiated. As will be described in greater detail below, these complementary accords will help provide the accountability and specificity needed to ensure that broad national agreements are implementable, and are refined to fit local conditions and opportunities. At these lower levels accords can be constructed around, inter alia, regional and local development policy and job creation initiatives, public service delivery and associated levels of taxation, specific investment projects, industrial policy and training agreements, the pooling and management of corporate social responsibility funds, payment of rates, and wages (to the extent that variability is allowed under national and sectoral agreements). The integration of regional, local, sectoral and corporate accords within the framework of the national Accord must balance the need for macroeconomic co-ordination with the advantages of allowing more detailed agreements to be struck at the appropriate lower levels.
591. It is vital that, where appropriate, these accord processes are inclusive, incorporating NGOs and other organisations of civil society, and that the parties to the discussions are properly representative of their respective constituencies. Furthermore, accords cannot be imposed by compulsion, but must rather be made self-enforcing by offering clear incentives to parties at all levels to participate. In the pages that follow we present a series of recommendations designed to address these twin requirements.
592. The Commission is under no illusions as to the difficulties involved in securing the required degree of trust, co-operation and mutual compromise among the social partners that will be required if the Accord for Employment and Growth is to be successful. We argue, however, that the economic stakes are sufficiently high as to warrant the investigation of all possible means by which co-operative solutions might be achieved. It is thus with deep concern that the Commission notes that recent economic and political developments have resulted in a counter-productive hardening of the rhetorical stances of both business and labour, which threatens to erode existing levels of confidence in the South African economy, both domestically and internationally. This observation forms the basis for our recommendation that a Presidential Jobs Summit, involving all of the social partners, be convened as a matter of urgent priority. The Summit should be used to urge both business and labour to renew their commitment to social partnership in economic affairs, and to make this commitment manifest by reaching agreement on the agenda, institutional arrangements and timetable for the negotiation of a comprehensive Accord for Employment and Growth.

The Economic Objectives of the Accord for Employment and Growth

593. The primary aim of co-ordinating economic policies and outcomes via an Accord for Employment and Growth is to allow for a more rapid expansion of employment than would be possible in an ad hoc environment. This is achieved through three inter-related mechanisms, which are listed here and elaborated on below.

• First, an Accord can enable the reconciliation of the twin objectives of steadily rising labour productivity and rapid employment creation. The Accord and the consequent accord processes can be used to ensure that the gains from productivity increases are shared between workers, investors and consumers in such a way that both domestic and international competitiveness are increased and employment growth is encouraged.
• Second, an Accord provides a more employment-friendly means of fighting inflation than do the relatively blunt instruments of fiscal and monetary austerity, both of which have dramatic and undesirable effects on employment in the short and medium term.
• Third, an Accord provides a forum in which the optimal sequencing of economic policy decisions can be debated, and co-ordinated responses to unforeseen macroeconomic shocks can be formulated.

Sharing the Gains of Increased Labour Productivity

594. If rising labour productivity is to be sustained and to lead to increased employment, it is imperative that the gains from the productivity advance be properly distributed between lower prices, higher profits, and higher wages. While it is true that some markets expand rapidly due to shifts in demand, holding all else equal it is through price reductions, or slower rates of price increase, that firms secure increased demand, and are hence encouraged to expand output. However, the benefits of this increased demand can only be realised in the long run if accompanied by increases in the capital stock, i.e. increased investment. Investment will be encouraged if a share of the productivity increase is reflected in higher profits. Finally, it is no less important that part of the productivity gain accrue to workers in the form of a wage increase. At the macro level wages are an important component of aggregate demand; at the micro level it will be extremely difficult to bring about gains in productivity if workers do not share in these gains. Given the imperatives of investment, price competition and wage advance it is axiomatic that the increase in productivity will outstrip the share accruing to any one of the recipients of the advance. In essence accords are exercises in gain sharing and sacrifice. If either business or labour enter the Accord under the illusion that they will receive all of the gain, or simply share it between themselves (thereby excluding consumers from the benefits of lower prices) the Accord will collapse, or, at the very least, it will not result in employment creation.
595. We should reiterate the central place of productivity gains in this schema. In the absence of productivity advance investors, consumers and workers face a situation in which any one group's gain is another group's loss. Under these circumstances the difficulties of sustaining an Accord are multiplied. Hence the Accord that distributes the productivity gains will also have to contribute to making these gains. Techniques for increasing productivity have been elaborated in Chapters 3 and 5 of this Report. Suffice here to say that, while gains in dynamic efficiency are principally a workplace matter, our proposals to achieve these gains rest on participatory structures and practices at the workplace that mirror many of the principles and choices underlying the accords discussed here. We have also made proposals that refer to a role for national policy and multipartite institutions in promoting productivity gains.

Fighting inflation

596. Efforts in other countries to co-ordinate economic policy through the mechanism of an Accord have usually emerged in times when spiralling inflation had created a widely-shared sense of economic crisis. British incomes policies of the 'seventies, for instance, were rooted in inflation rates of the order of 25%. South Africa today enjoys much more moderate levels of inflation, and it is hence unlikely that the social partners would be willing to submit to the rigours of an Accord based on the threat of inflation alone. Instead, the South African Accord will only gain the widespread support upon which its success depends if it is seen to respond directly to the crisis of unemployment. This implies that the link between the wage- and price-moderating function of the Accord and its job-creating function must be explicit and credible.
597. Nevertheless, the ability to contain inflationary pressure remains one of the key benefits of an Accord. While it is true that inflation is currently at moderate levels, it is equally true that the rate of growth of output and employment is far from adequate. Faster economic growth is urgently needed, but faster growth will generate inflationary pressure. Indeed, it is largely the fear of inflation that has prevented government from attempting to stimulate the economy more aggressively. The rand's recent rapid depreciation will also intensify inflationary pressures, and has already prompted the Reserve Bank to raise the bank rate, a move which may be expected to retard the rate of growth of output and employment in the near term. In short, it is not the severity of current inflation that justifies the Accord; rather, it is the spectre of future inflation, as well as the negative employment effects of the measures that are now being used to contain inflation, that justifies seeking a better approach to the problem.
598. In particular, if the government's primary means of fighting inflation is through monetary contraction, employment may suffer considerably in the short run, and certain key interest-rate-sensitive sectors such as housing construction and consumer durables will be unduly punished. The Commission argues that the better strategy is for government to work with the social partners to moderate inflation via direct negotiations. Wages and prices are on the agenda for obvious reasons. However, encouraging investment in those industries and skills where there are supply constraints may also be important for containing inflation in the long run, and should be a goal of a national Accord and of industrial policy as well.
599. It is important to state from the outset that the establishment of an Accord cannot and must not impose bureaucratic control on the economy. Rather, the Commission holds that the Accord must be freely entered into and be largely self-policing; in the pages that follow we suggest various means by which this might be achieved.
600. This position is reflected in the Commission's support for a negotiated Accord as opposed to a statutory incomes policy. However, the minimum requirements for achieving a meaningful Accord must be appreciated. This strategy can only be successful if business, labour and government are all willing to enter into serious discussion of wages, prices and investment at the national level. This does not imply that individual wage bargains will be determined by the Accord: collective bargaining would still go on in the bargaining councils and at company or workplace level. Nor does it imply that individual product prices will be determined by negotiation. In the following discussion of the Accord's institutional mechanisms we explain how wages, prices and investment may be influenced in ways that fall well short of statutory ceilings or quotas, yet are nonetheless effective. However, if trade unions hold that under no circumstances can their independence in industry or company bargaining be constrained in any way, then there is no point in trying to establish an Accord. Similarly, if business insists that its pricing and investment decisions cannot be, respectively, constrained or guided in any way, then the Accord cannot go forward.

Co-ordinating Macroeconomic Policy Responses

601. The rand's recent slide has illustrated that macroeconomic conditions can change without warning. It has equally underlined the value of an institutional forum in which a coherent response to macroeconomic shocks could be formulated. At present, the Reserve Bank's call for wage restraint on the part of unions (in order that the inflationary impact of the depreciation not be extreme) is likely to fall on unsympathetic ears. Unions may understandably wonder what is being offered in return for their forbearance.
602. An Accord would facilitate the desired wage outcomes by crafting a bargain in which the size of wage increases is placed on the national bargaining agenda alongside tangible commitments around prices (which affect both workers' cost of living and the competitiveness of South African goods and services), tariffs, investment, training, the social wage, supply-side measures to assist exporters in expanding capacity, and other matters. An Accord could thus offer a means for ensuring that the nominal depreciation just witnessed is not undone by rising prices, and that employment in export-oriented sectors is maximised.

Why Bipartite Collective Bargaining is not Enough

603. It might be objected that existing collective bargaining arrangements are sufficient to achieve the goals outlined above and that the additional institutional mechanism of a national Accord is unnecessary. This objection, however, is misguided; there are very real advantages to broadening the agenda of national multipartite discussions to include the matter of wage determination. Foremost among these is that multipartite negotiations can address the question of the real wage in ways that bipartite negotiations cannot. Trade unions and employers in bipartite collective bargaining can negotiate money wages, but they cannot negotiate real wages because not even industry-level bipartite collective agreements can control the prices that determine the workers' cost of living. If trade unions want to negotiate over the real living standards of their members they need assurances about future price movements which they cannot get from negotiating with the employers in their industry alone. Indexation of wage contracts is at best a partial solution: it may defend real wages of some workers, but it does little to contain inflation, and may even exacerbate it. The only sure way to negotiate changes in real wages is thus to involve government in a comprehensive, economy-wide, multipartite accord.
604. The alternative strategy, that of pushing up money wages as rapidly as possible, can be self-defeating. If unions succeed in raising money wages faster than the increase in labour productivity, and if employers defend their profit margins, then the result will likely be rising product prices in the industry in question or a decline in employment. If unions in other sectors of the economy are doing the same, the result will be an increase in the general rate of inflation which will erode some or all of the hard-fought money wage increases, and which may also deter new investment. Moreover, if a determined central bank steps in to quell inflation via restrictive monetary policies, employment is likely to suffer. An Accord offers a way out of this trap, one that should allow for faster real wage gains and higher employment than would be possible in an uncoordinated economic environment.
605. Government's participation in an Accord need not be limited to its role in influencing wages and prices. By bringing taxation and social welfare spending policies to the bargaining table, government may facilitate more complex three-way deals than may be achieved in bipartite negotiations. Workers may, for instance, moderate their wage demands in return for public commitments to increase spending on various components of the social wage, public works programmes, or assistance from the Social Plan Fund, or in return for government support for training at the workplace. Business may be willing to commit to increased expenditure on investment, research and development or workplace health and safety if government agrees to defray part of the cost via reduced corporate taxes. Government may also structure taxes to reward investment in return for business agreeing to moderate the rate of price increases.
606. The Accord thus facilitates the simultaneous discussion of major components of workers' standard of living, with the potential for negotiated trade-offs between the various goals of business, labour, and government. This stands in contrast to bipartite bargaining between unions and firms, which must always take place against a backdrop of uncertainty as to what government, and the rest of the economy, will provide. This uncertainty ties the hands of negotiators on both sides of the traditional collective bargaining table, and encourages a hard-line stance. The greater co-ordination afforded by a comprehensive national Accord should reduce uncertainty and increase the parties' willingness to negotiate flexibly. For its part, government gains a degree of breathing room for its macroeconomic policy. It can then pursue a more aggressive growth strategy, knowing that the odds of an inflationary response are much reduced.

Honouring the Accord: Obstacles and Opportunities

607. The Commission does not wish to suggest that the outcomes described above will flow automatically from the signing of a national Accord. For the Accord to be effective its terms must be honoured by all parties to the negotiations, and this raises very real institutional and political challenges. First, there is the question as to whether business in general will feel compelled to respect the requirements of an Accord negotiated by its national representatives, such as Business South Africa. Some firms, particularly smaller businesses, may argue that they are not well represented in such national organisations, and hence that they ought not be bound by its decisions. It may be further doubted whether the resources of these national organisations are adequate to the task at hand. Second, there is the parallel problem of the degree to which individual unions will feel bound by the terms of a national agreement. South Africa's system of sectoral bargaining would not appear to lend itself to national wage determination.
608. While these difficulties are real they are not insurmountable. In the sections that follow we address the institutional requirements of a meaningful Accord, and argue that the problems raised above may be addressed in various ways. First, resources should be devoted to increasing the representivity and strengthening the leadership and negotiating capacities of both business and labour federations. Second, complementary local and regional accords should be constructed wherever possible that mirror the requirements of the national Accord, but that involve individual unions, firms, and local or regional government officials who are more directly accountable for promises made. The third requirement has already been mentioned, but is of such fundamental importance that it bears repeating here. It is that business, labour and government take bold steps to foster an atmosphere of mutual trust.

Building Trust Among the Social Partners

609. In the final analysis the key to the Accord is the will of the parties involved. The Commission is of the view that South Africa's recent history of accord-making is evidence of the considerable political maturity that the social partners bring to the debate. The negotiation of the Peace Accord, the transformation of the National Manpower Commission, the formation of the National Economic Forum, and the success of the national elections are ample testimony to the power of negotiation in the face of crisis. Likewise the inclusion of NEDLAC in the Interim Constitution was an impressive victory for the principle of participatory economic democracy. South Africa in 1996 is at an economic cross-roads no less decisive than the political cross-roads of the period 1990-94. Recent economic and political events, including the fall of the rand and the divisions that were generated around aspects of the new Constitution, make it absolutely imperative that business and labour both respond in constructive ways if confidence in the South African economy is to be maintained. The Commission holds that there is nothing to be gained, and much to be lost, by retreating from the principle of co-operation in economic affairs. An Accord for Employment and Growth is conditioned upon this co-operation. It does not seek to deny the existence of conflict between business and labour, but rather to contain and manage this conflict through negotiation and compromise.

Institutional Mechanisms for Formulating the Accord

610. There are two broad choices of institutional mechanisms by which an Accord may be implemented. The first is a statutory incomes policy where government specifies, through statute, limits on price and wage movements. This approach ­ adopted by some countries in the high inflation era of the 'seventies ­ has proved highly unstable. Essentially, bureaucratic mechanisms are simply substituted for the market and collective bargaining. This approach is rejected in favour of an accord negotiated between the social partners. In the Commission's view co-ordination involving wages, prices and investment will be more durable if effected through a negotiated mechanism. We further argue that NEDLAC is the appropriate forum in which to formulate the Accord.
611. NEDLAC, the National Economic Development and Labour Council, was created as one of the first acts of the democratic government. It assumed the functions of its predecessors, the National Economic Forum (NEF) and the National Manpower Commission (NMC), which were both tripartite forums on which organised business, organised labour and government were represented. The NEF was a voluntary (that is, non-statutory) negotiating body, concerned with the formulation of economic and social policy. The NMC was a statutory advisory body concerned with industrial relations and labour market policy. Initially shunned by organised labour, the structure and functioning of the NMC was reworked in the early 'nineties thus securing the participation of the unions. The NEF, in particular, posted some significant successes, including the negotiation of South Africa's offer to GATT.
612. NEDLAC functions through four chambers: the public finance and monetary policy chamber (in which the Reserve Bank participates), the trade and industry chamber, the labour market chamber, and, finally, the chamber dealing with community development issues. NEDLAC differs from its progenitors in the inclusion in its development chamber and Executive Council of a community constituency comprising representatives of women's groups, civics, the youth, rural communities and the disabled. Policy discussions are conducted through the four chambers and submitted to the decision-making structures of NEDLAC for formal acceptance. The Minister of Labour is responsible to the Cabinet and Parliament for the functioning of NEDLAC.
613. The founding of NEDLAC is rooted in a purpose larger than the co-ordination of economic policy. It represents one of the clearest expressions of the nation's commitment to a comprehensive and participatory democracy, to social partnership. Its manifest purpose is to accord civil society the institutionalised opportunity to participate in the formulation of social and economic policy. Moreover, the form of participatory democracy that NEDLAC embodies is not only an expression of high principle; it is also a representation of a view that holds that economic prosperity, especially in time of deep structural change, requires maximum "buy-in" from the key economic stakeholders. By extending to civil society the opportunity to participate as active partners in the process of structural change, the likelihood of buy-in is increased.
614. The Commission fully endorses the principle of participatory democracy that NEDLAC embodies. We endorse the view that anchors sustained economic prosperity in social cohesion and reject the notion that the measure of a strong and determined government is its ability to "stand up to" legitimate constituent elements of the society. NEDLAC's very existence represents the rejection of that viewpoint and for this reason it deserves the unequivocal support of government.
615. However, the mere fact of participation is insufficient. The participatory process, if it is to provide the glue required to cohere the society, must highlight the difficult trade-offs central to economic policy formation and then seek to resolve them through the structures of the institution. An attempt to influence the interaction between wages, prices and investment would provide the participatory process with one of its sternest tests to date, but the potential rewards are considerable: interaction of this nature between public and private decision-makers puts in place the pillars of rapid economic growth.
616. NEDLAC appears well placed to contribute to co-ordination of economic policy, that is to identify and resolve the trade-offs between diverse private interests and between those interests and the public interest. In contrast with most other attempts to formulate incomes policies, NEDLAC does not originate in a focused attempt to confront a particular symptom ­ inevitably spiralling inflation ­ of poor co-ordination. These efforts ­ the traditional European and Latin American incomes accords of the 'sixties and early 'seventies ­ frequently floundered because of their inability to extend the coverage of the accord beyond the narrow range of problems that they had been set up to resolve. Although South Africa has no previous experience of an incomes accord, our recent political history has propelled NEDLAC into existence with a mandate that traverses an unusually comprehensive range of public policy matters, thus, on the face of it, overcoming some of the key shortcomings that limited the impact of the more narrowly circumscribed incomes accords.
617. However, the effective exclusion of the discussion of wages, prices and the level and composition of investment from NEDLAC's agenda severely limits its role. In this sense then, NEDLAC has both more and less on its plate than traditional incomes accords and it has, concomitantly, both a greater and lesser capacity to provide the forum for the co-ordination of economic outcomes.
618. Current practice, if anything, loads the dice against government in NEDLAC. Government's key macro-policy instruments ­ fiscal policy and trade policy ­ are on the NEDLAC agenda, but other key macro instruments remain within the exclusive competence of an independent Reserve Bank (which is, however, represented in NEDLAC's public finance and monetary policy chamber) or else remain in the private sphere of business and labour. As noted, government powerfully influences the environment within which business and labour (and the central bank) take their decisions, but current NEDLAC practice does not incorporate these decisions, whether arrived at unilaterally or through collective bargaining, in the agreements reached. Effectively, NEDLAC is a mechanism that affords business and labour a powerful institutional capacity to directly influence public policy but does not afford government the parallel opportunity to influence wages, prices, and investment, despite their powerful impact on macroeconomic performance.
619. While there is no formal or necessary reason for the exclusion of these matters from the NEDLAC agenda, their incorporation would cut across well-established practices and institutions. Their omission from the public policy agenda was therefore entirely predictable. However, as we have stated, in the absence of willingness on the part of the unions and employers to bring wages, prices and investment to the table, there is no possibility of co-ordination. Certainly it is not possible to sustain the present situation where government commits the deployment of its key policy instruments to negotiation with the social partners, in the absence of reasonable expectations regarding wage and price movements and investment behaviour.
620. The extent of the attitudinal change and the degree of institutional transformation that comprehensive co-ordination along these lines presupposes makes it clear that an Accord will not be achieved without considerable difficulty. Some members of the Commission expressed concern that an expansion of NEDLAC's agenda would weaken the fledgling organisation; others felt that the procedures and structures that governed an inevitably complex statutory body like NEDLAC were too formalistic and cumbersome to initiate as sensitive a process as that proposed here. A growing concern on the part of business and labour over aspects of NEDLAC's performance to date was also noted by some Commissioners.
621. The Commission holds that many of NEDLAC's difficulties relate to its newness, and to the uncertain economic and political environment in which it is situated. We argue that the single most important factor that could contribute to NEDLAC's success would be the expression by business and labour of their continued, and strengthened, commitment to the principle of tripartism upon which NEDLAC rests. While the Commission does not wish to make detailed recommendations as to the organisational structure and operational procedures that NEDLAC should adopt, we have identified a number of reforms that would strengthen NEDLAC's prospects and help rebuild support for it among the social partners:

• Creative means must be found for enabling NEDLAC to accommodate regional concerns, and to support regional accords of the kind mentioned above and explored in more detail in the pages that follow.
• Government should encourage the regular participation in NEDLAC of even more senior members of organised labour and of the business community.
• NEDLAC's role in relation to Parliament and the Executive should be clarified. NEDLAC is neither a law-making nor an executive body. Rather, its contribution should be to provide a supportive framework for the formulation and implementation of national policies and, through the participatory process, to foster broader social support for these policies. This is precisely the aim of the proposed Accord for Employment and Growth.

622. While the Commission holds that NEDLAC is the appropriate forum for the negotiation of the Accord, it argues that the initial impetus, and final accountability, for the Accord must come from the highest level of government. The object of the proposed Jobs Summit is to draw on the authority of the President to firmly establish the urgency of formulating a coherent tripartite response to the current economic challenge. The Commission further recommends that government place before this Summit a clearly formulated proposal regarding the process by which an Accord for Employment and Growth might be achieved and an initial suggestion as to what government's contribution to the Accord might entail.

Business's Contribution: Commitments Regarding Pricing and Investment

623. There is a strong view that business's contribution cannot be determined ex ante but that it will be manifest in the investment that flows automatically from the successful conclusion of an agreement that moderates wage increases. The Commission argues, however, that this will not be an adequate basis for the construction of an Accord: labour and government cannot be expected to make concrete commitments if business's contribution is not just as clearly specified. It is therefore incumbent upon business to establish its bona fides with respect to participation. The following mechanisms should be considered, although this list is by no means exhaustive:

• Large producers could pre-announce the conditions for a major, specified investment programme and the social partners could attempt to negotiate and secure these conditions through accord processes at the appropriate levels;

• Major producers could follow the example of several of South Africa's larger corporations in agreeing to hold price increases to below the rate of inflation;

• Business could commit to negotiating credible profit-sharing arrangements and corporate governance mechanisms that provide labour with the opportunity to participate in decisions regarding the deployment of those profits;

• Business could commit to exercising moderation in setting managerial salaries;

• Business could make a stated commitment to a minimum expenditure on training at lower levels of the workforce.
624. The Commission wishes to reiterate that we do not recommend that prices of individual products produced in the private sector be set through the Accord mechanism. However, in certain instances, notably the publicly-owned "natural monopolies", the terms of the Accord may incorporate agreement on the actual price of the product or service. Indeed, one of government's contributions to an overall Accord may be commitments around the pricing of particular key products ­ fuel, public transport and electricity are obvious examples.
625. For the most part, however, price restraint will rely on competitive market pressures. Hence low tariff barriers and a much strengthened competition policy are essential pillars of an Accord that seeks to incorporate prices within its ambit. A strengthened competition authority should include enhanced capacity to monitor product price movements. Unusual price movements should trigger investigations by the competition authorities in respect of underlying market conditions.
626. A final mechanism by which prices may be influenced is by making access to public resources of various kinds (e.g. public tenders, concessionary tax treatment, assistance with finance, or complementary infrastructural investment) contingent on firms' conforming to any price-related agreements reached in the Accord.

Qualifying Comment: Commissioner Segal

I endorse the concept of a high-level process that would seek to establish a common understanding among government, labour and business of, and create a framework for, a strategy for employment and growth. The country's recent political history shows emphatically that none of the "social partners" has a model that is comprehensively acceptable to the other two, and nor are there directly appropriate models available from elsewhere in the world. It shows too how valuable it will be to raise the level of public debate about these matters so that, for instance, the full import of South Africa's now being part of the global economy can be appreciated.

I endorse too the proposals that the process take material form in an accord which will need revision from time to time, and that the national process/accord could be supplemented by regional and sectoral processes/accords that would be more specific and practical.

My concerns, which lead me to submit this qualifying comment, are two-fold:

• the Report may over-estimate what commitments can meaningfully be entered into by national-level business representatives, especially in respect of such items as investment and prices which in the final analysis are commercial decisions by individual companies;
• the Report may under-estimate the complexity and dynamics of a modern, internationally open economy, which can lead to an over-emphasis on the possibilities for co-ordination that can be achieved through central negotiations and committee structures.

This does not detract from the proposal to formulate an Accord in NEDLAC. Rather it means that the process consequent upon acceptance of the Commission's proposal must be designed bearing in mind what as a practical matter can be "delivered" at the various levels. Commissioner Lamprecht associates himself with this caution.

Labour's Contribution: Commitments Around Wage Determination

627. Existing wage determination mechanisms are thought to constitute a significant obstacle to co-ordination via a national Accord. One may envisage an agreement at sector (or potentially even enterprise or corporate) level that effectively sees wage moderation being traded for investment (job creation) guarantees, job security, profit sharing or training. The wide-ranging nature of the agendas of the statutory and bargaining councils will facilitate such agreements, and they are to be encouraged. However, whilst such wide-ranging bipartite bargains can enable industry-specific wage/productivity/profitability trade-offs, on their own they do little for macroeconomic co-ordination. Obtaining macroeconomically efficient outcomes is difficult in industrial level wage bargaining systems because there is no compulsion for the negotiating partners in industrial bargaining systems to assume responsibility for the overall consequences of their claims. Because each industrial level agreement is concluded in isolation, no single set of negotiating partners is in a position to consider (or feels they have to be concerned about) the macroeconomic effects of their bargaining outcomes. The result is a failure to co-ordinate wage claims in ways which control inflation and ensure continued competitiveness.
628. We should emphasise that the national Accord would not be a substitute for collective bargaining. Sectoral negotiations through the bargaining councils and enterprise-level agreements would continue to set minimum and actual wages, respectively. However, the Accord could set upper and lower limits within which wage bargaining would then take place. There are a variety of ways in which the upper limit might be formulated which may grant a greater or lesser degree of independence to the bargaining councils. Limits may be specified in absolute or relative terms, with or without reference to the rate of productivity advance. For example, the upper limit may be specified as a percentage increase in the firm's wage and salary bill per employee. The distribution of the increase between different job categories would remain a matter for collective bargaining.
629. This mechanism of wage determination would limit aggregate wage and salary drift, while incorporating sectoral and enterprise-level specificities into a co-ordinated bargaining arrangement. This flexible formulation will also ease the ability of individual unions and employer associations to honour the terms of the Accord. Their collective bargaining functions will not be usurped by central business and union representatives.
630. We have emphasised the importance of voice regulation as opposed to legal or bureaucratic enforcement. An Accord will not be sustained if it relies on narrowly specified regulation. With respect to product price movements we have outlined how, without resorting to price setting, corporate commitments and the actions of the competition authorities may act to ensure price moderation. Similarly, without resorting to punitive legal enforcement, the government has an important role to play in ensuring that collectively bargained outcomes are sensitive to macroeconomic requirements. There are three mechanisms through which government could influence wage outcomes. These are:

• Through outcomes achieved at public sector wage negotiations. While the public sector is undergoing the convulsion of major restructuring, it will be extremely difficult for public sector wage negotiations to play a role in influencing the outcomes of private sector negotiations. In the medium term, however, the government should promote its view on wage levels and structures through its participation as a bargaining partner in the public sector. It is important that public sector wage bargaining outcomes be subject to negotiation as part of the Accord. Public sector compliance with the resulting guidelines will send an important signal to employers and unions in the private sector.

• Through Ministerial responsibility for gazetting and extending collective bargaining agreements. Before gazetting and extending agreements, the Minister should require explanation of wage settlements that vary considerably in either direction from the average of settlements achieved in other negotiating forums or from the terms agreed upon in a national Accord.

• Through the activities of the competition authorities. Where a bargained wage settlement considerably exceeds the national average, the competition authorities should be alerted to exercise particular vigilance over product price movements in the market in question.

Representivity & Capacity of Union Federations and Business Associations

631. Compliance with the Accord will, in large measure, rely directly and indirectly on the capacities and standing of the representative bodies involved. This poses special challenges to both business and labour. The intrinsic diversity of the business community as well as the uneven strength and sometimes divergent purposes of its representative bodies may make it difficult for business to participate effectively in the proposed Accord process. There is also the question of securing the meaningful participation of small- and medium-sized enterprises. The Commission holds that despite these difficulties, the business associations' successful involvement in other forums in recent years suggests that they should be able to address these issues so that their social partners can be confident that they are dealing with a properly representative and authoritative national body. The Commission also wishes to re-emphasise the urgency of the task at hand: without well-resourced and genuinely representative business associations, negotiated solutions to the looming economic stand-off between business and labour cannot be achieved. To facilitate the increased representivity and negotiating capacity of the business associations, the Commission recommends that government, wherever practicable, should insist on speaking to business on matters of national policy through a single representative organisation, and that SMMEs participate meaningfully in this organisation.
632. The unions appear somewhat better prepared for the role demanded by a national Accord. For example, in contrast with their employer counterparts, they do have some experience of co-ordinated national wage initiatives. However, the unions are divided ­ the three union centres are jointly represented in the labour constituency in NEDLAC; and these national centres are federal in nature with extensive powers in the hands of their industrial affiliates. Moreover, they have suffered a substantial loss of senior personnel, particularly to the public sector. In an effort to strengthen labour's representational capacities:

• The Commission emphasises a recommendation made earlier that government provide resources for the establishment of an institution for the further training of union leadership; and

• Government should, as with business, wherever practicable insist on speaking to labour on matters of national economic policy through a single representative voice.

Complementary Regional and Sectoral Accords

633. The final means by which the difficulties inherent in securing compliance with the Accord for Employment and Growth may be addressed is through the construction of complementary regional and sectoral multipartite agreements. The advantage of such accords, which may be forged at local, district or provincial level, is that the issues under discussion are more tangible and of greater direct relevance to the parties to the negotiation. Hence the incentive to participate is greater and self-enforcement is more likely. Furthermore, social pressure (from the press, trade unions, local government and business people) is likely to be greater, better targeted, and more effective at the local or regional level. This is all the more true if the accord-making process actively solicits the involvement of representatives of local communities.
634. One of the difficulties faced by organised labour when considering the developmental needs of a particular region is the tension which arises between bipartite bargaining at an industrial level and multipartite bargaining at the regional level. Industrial trade unions are generally committed to reducing wage differentials between regions. However, if there is to exist sufficient space to make meaningful multipartite accords at the regional level, then some tension may arise between this process and the goal of reducing inter-regional wage differentials through sectoral agreements forged in those bargaining councils that are national in scope. The Commission argues, however, that the recommendations made in Chapter 4 of this report will allow for sufficient flexibility in this regard.
635. Given the geographically concentrated nature of the effects of major retrenchments, it may also make sense to include negotiations over the distribution of Social Plan funding as part of a regional accord.
636. One disadvantage of regional accords, however, is that in attempting to create favourable conditions for attracting investment, one region's gain might be another's loss. By embedding regional agreements within a national framework, however, competition between the regions can be managed so as to avoid counter-productive bidding wars.
637. Multipartite sectoral agreements may also be useful complements to the national Accord. In particular we draw attention to our proposal in Chapter 7 for an accord between government, unions and employers in the construction sector. The purpose of this accord is to create jobs through the adoption of labour-intensive construction methods. The mechanism of this sectoral accord would be to have employers commit to labour-intensive techniques and desired industrial relations and employment practices, to have unions agree to task-based remuneration, and to have government agree, through public sector contracts, to smooth the boom-bust demand cycles that characterise the construction industry.

Conclusion

638. An Accord brings the social partners together to try to agree on a means to increase employment, raise real wage growth by reducing inflation, and increase investment in human and physical capital. No one can guarantee these outcomes, because no one can control all the factors that determine employment, inflation and growth. However, the importance of an Accord must be assessed in reference to what is likely to happen to the rate of unemployment if there is no Accord. The poor record of job creation in the past year of economic growth strongly suggests that a business-as-usual approach will be an utterly inadequate response to South Africa's employment problems.
639. The jobs crisis is undeniably severe, but events of the recent past demonstrate a remarkable political will and institutional capacity to forge consensus out of crisis. Just as the imperatives of political transformation challenged the social partners to achieve, through negotiation and compromise, what many international observers said could not be achieved, so too does the imperative of the rapid creation of secure, reasonably remunerated employment challenge the social partners to enter into a new round of path-breaking negotiations in order to forge a national Accord for Employment and Growth.

---