South Africa’s ‘new growth path’ (NGP) plan aims to create 5 million jobs by 2020 and bring about a new more inclusive, labour-absorbing and efficient economy.¹ The proposal is to achieve this through a mix of direct government job creation, social-democratic consensus building and macroeconomic, labour and industrial policies. It is a vision rather than a plan or a projection. Although a few specific objectives are listed (e.g. creating youth brigades in which a million young adults are to be provided with training and work experience), the NGP mainly takes the form of tabling a broad set of objectives (e.g. a more competitive, labour-absorbing, fast-growing, knowledge-intensive green economy which produces millions of decent jobs while reducing poverty and inequality) alongside an ‘if then’ commentary: if this vision is to be achieved then we need a stronger more efficient state, appropriately targeted policies, co-ordinated monetary and fiscal policies, an improved supply of skilled labour, co-operative partnerships with labour and capital, national will to accept some wage restraint and government mobilisation of savings and direction of investment etc. Precisely how important the various elements are is never spelled out.

What is the appropriate response to this document? One option is to ask for more details and greater clarification. But as the drafters do not appear to have a detailed background plan (the contingent nature of economic policy formulation is noted at various points), a more productive approach is probably to concentrate on the logic of the vision, pointing to some of the key problems, trade-offs and requirements which have not been specified, and to ask questions about how realistic the objectives are given our current position.

Figure 1 presents total employment trends during the 2000s using two different data sources: the Quarterly Employment Survey (QES) of business enterprises and the household Labour Force Survey (LFS) data which was produced bi-annually up until 2007 and subsequently on a quarterly basis (the QLFS). The QES tracks formal sector jobs but is limited by its sampling frame (existing businesses); whereas the QLFS picks up the full range of full-time and part-time income-earning activities, but has been plagued by definitional changes and inconsistencies. Fortunately, Stats SA has produced an adjusted LFS series for the period from 2000 to 2007 that is consistent with subsequent QLFS data.

Using this data series (the most reliable of the existing sources), employment grew by 17% during the economic upswing of 2001 – 2008, whereas real output grew by just over twice that (35.7%). At this rate, South Africa would need a real economic growth rate of over 6% per annum to create 5 million jobs by 2020^1(p.9) – a growth rate which is most unlikely to materialise in the current global economic slowdown.

The NGP does not, however, rely on the existing growth path to generate jobs. The objective is to create a ‘new’ (more labour-absorbing) growth path through the judicious use of government policy. This entails reversing some of the structural shifts of the 2000s. As shown in Table 1, employment contracted in the agriculture; mining; manufacturing; and private household sectors and expanded in the finance, real estate and business services; community, social and personal services (i.e. mainly government); transport, storage and communication; and construction sectors.

The NGP, however, worries that growth was debt financed and not ‘underpinned by a strong production base’.^1(p.4) While South Africa was, indeed, spending more than it earned (facilitated by strong capital inflows – see below), the problem is not the lack of production per se, but rather a shortfall in exports and in this respect, tourism is just as suitable a contributor to economic growth as manufacturing. The focus on production is thus somewhat overstated. In any event, the NGP proposes a ‘developmental state’ to support new productive activities, especially in light manufacturing, the mining and agricultural supply chains and in a range of knowledge-intensive and skill-intensive activities (e.g. biotechnology) and in green technology (notably solar power). Resources are to come from progressive taxation and by boosting and ‘mobilising’ domestic savings, presumably by requiring pension funds to purchase ‘development bonds’.^1(p.27)

The term ‘developmental state’ is evocative of the East Asian model of development (notably South Korea) where an interventionist state provided protection and selective incentives to targeted firms and sectors. But whereas export subsidies and tariffs were tolerated in the 1960s and 1970s when East Asian economic development took off, the World Trade Organization now frowns upon the use of such policy instruments. The NGP’s response is to observe that South Africa should continue to ‘resist rigid formulation-driven reductions in industrial and agricultural tariffs that would undermine employment and growth’.^1(p.24)

There is, however, an even bigger difference between the East Asian developmental state model and that being proposed by the NGP, namely its approach to labour. Whereas wages and working conditions improved in East Asia only once the labour-market tightened after decades of strong growth, the NGP proposes to advance the South African trade union agenda of ‘decent work’ as an integral part of the growth strategy. Indeed, the document strongly suggests that some kinds of employment, notably the flexible forms of employment available to largely unskilled workers, such as those provided through labour-brokers and by firms hiring workers on a part-time, non-contractual basis, will be restricted if not eliminated. The strategy thus appears to be to create a new growth path which favours the employment of full-time skilled workers. Making this a labour-absorbing growth path is the task of industrial policy which supposedly will identify and support economic activities which have the potential to create large numbers of well-paid formal jobs in which workers have contracts, access to benefits, training and a career path. This is an even more daunting challenge than achieving a 6% annual real growth rate.

An obvious precondition for South Africa becoming a ‘developmental state’ is to improve government efficiency and reduce corruption and rent-seeking. The NGP acknowledges the many weaknesses which need to be addressed but gives little clear direction how this, along with better co-ordination of government policy, is to be achieved. The NGP talks about creating ‘delivery forums’ and enhanced mechanisms for monitoring and accountability, but the devil, of course, will be in the details – presumably still to be negotiated. That the NGP proposes, as a ‘core’ intervention to support small businesses, a ‘name and shame’ campaign to get government departments to pay suppliers within 30 days – rather than just demanding that governments act according to existing policy – speaks volumes about the challenges involved in transforming the state into anything approximating a developmental one.

Also in contrast to the Asian developmental state model, the NGP is replete with social democratic discourse about engaging constructively with stakeholder representatives, partnering with capital and giving workers greater voice. This is evocative of the Nordic social democratic model of co-ordinated wage bargaining, where attempts were made (albeit with declining success over time) to keep wage growth in line with productivity in order to avoid squeezing profits or employment. The NGP appears to be suggesting a move in this direction, although the institutional mechanisms needed to affect it remain unclear. Even so, the mention of ‘wage and productivity gain-sharing policies’^1(p.8) and the possibility of constraining real wage growth for all but the lowest paid workers^1(p.26) is an interesting and potentially game-changing development. The crucial issue, of course, is whether organised labour and business will buy into it. South African trade unions are typically hostile to social democratic compromises and quick to evoke a discourse of class struggle. Business, in turn, is also likely to bristle at the idea of restraining executive pay. However, some restraint at the top end will be necessary – if only for symbolic reasons – if this is to work. Excessive executive earnings have long been grist for COSATU’s mill. Cutting them is probably a precondition for any commitment from COSATU to broader wage moderation (although it is possible that nothing will be able to convince COSATU to contemplate wage moderation).

The NGP does not provide an economic model for how future job creation is to be achieved. It assumes that ‘most of the projected new jobs will come from the private sector’^1(p.8) and that these will be leveraged through targeted assistance for five identified ‘job drivers’ (Table 2). Many of the job growth estimates appear to come from the Industrial Policy Action Plan 2. It is unclear what the status of these job estimates is because the NGP itself says that they are ‘not set in concrete’ and that a ‘mapping process’ is being used ‘to think innovatively about new opportunities for job creation’.^1(p.10) In any event, the five job drivers provide some indication of the strategic direction of government policy – at least in the areas of industrial and agricultural or rural development policy, regional integration and with regard to non-governmental organisations and co-operatives. Note that these interventions are in addition to government proposals to expand the community-based public works programme and introduce youth brigades.

The NGP acknowledges the key economic problems facing South Africa: high unemployment, low levels of domestic saving and investment, persistent balance of payments deficits, an overvalued exchange rate, skilled labour shortages, energy and infrastructural bottlenecks, economic concentration (and related anti-competitive behaviour), government inefficiency, rent-seeking and regulatory burdens on business. Many of these constraints have been highlighted by external reviews of South Africa’s economic challenges, such as those by Spence⁴, Hausmann ⁵ and most recently, the Organisation for Economic Co-operation and Development⁶. But whereas the rigidity of the labour market, especially with regard to the employment of unskilled low-wage workers, was highlighted in all these external assessments, it is notably absent in the NGP. Rather, the ‘decent work’ agenda of South Africa’s trade union movement is taken as a given.

Figure 1: Employment and output growth over the past decade.

The very idea of low-wage, relatively unskilled employment appears to be an anathema to the NGP. It mentions decent work regularly throughout the document (it is almost as if someone has edited the document inserting ‘and create decent work’, or ‘which is consistent with decent work’ at every point where the writing might imply that relatively low-paying, unskilled employment is on the cards). In addition, calls are made for every worker to be trained and provided access to benefits and a ‘career path’. The assumption appears to be that the private sector will respond to legislative efforts in support of decent work by providing new and better jobs. But will it deliver the ‘inclusive’ growth path the NGP hopes for?

Inclusive growth requires strong employment growth, especially for the many unskilled unemployed who have never worked before. To maximise employment growth, it makes sense to expand activities and sectors which require relatively low amounts of capital investment and output expansion for each additional job created. As can be seen in Table 3, these are the construction sector and the commerce, catering and accommodation sectors which have capital to labour and output to capital ratios orders of magnitude lower than capital-intensive sectors like electricity and mining. But note that these sectors also have the lowest average earnings per employee and higher proportions of what the NGP calls ‘precarious’ employment (i.e. jobs without access to benefits or a contract) than other sectors. Presumably such precarious employment does not count as ‘decent’ work and thus will be targeted for contraction rather than expansion. If so, this will probably (but not necessarily) improve the quality of existing jobs – but it certainly will be at the cost of rapid employment growth for the less skilled.

The NGP clearly regards low-wage, ‘precarious’ work as a problem. The section on ‘core challenges’ argues (incorrectly) that most of the jobs created in the 2000s were poorly paid ‘dead-end’ jobs and links this to the fall in the wage share (and rise in the profit share) of value added from 1994. This analysis, however, is misleading. As shown in Figures 2, 3 and 4, the profit share was indeed on an upward trend, but this was not because of restrained wage growth. Average real earnings rose over the period and it was only because labour productivity rose that the profit share rose.

Figure 2: Trends in profitability: 1989-2009.

Figure 3: Trends in labour productivity, real remuneration, capital-intensity, employment and the profit share.

Figure 4: Deconstructing trends in profitability.

The share of wages falls (and the profit share rises) when output per worker (labour productivity) rises faster than average real wages. In other words if employers can restructure production such that the average output produced by workers rises faster than their real wages, then both employed workers and capitalists benefit from the growth path. But rising labour productivity is a double-edged sword for the working class when it is generated by stagnant or falling employment. Rising labour productivity is necessary for the sustained growth of real incomes for the employed, but when it comes at the cost of jobs for others, there is a clash between the objective of high-wage, high-productivity growth and that of a more inclusive growth path involving job creation for the poor. The central weakness of the NGP is that it never confronts this problem and its associated trade-offs.

Aggregate labour productivity can rise in three ways: existing production processes can be made more efficient with the same number of workers; low-skilled workers can be laid off (and the production process made more skill-intensive and capital-intensive); or structural changes towards more skill-intensive and capital-intensive activities can be made. Figure 3 shows that average real remuneration rose steadily (except for a few years in the mid-2000s), but because labour productivity rose faster across the period, the share of profits rose. Why did labour productivity rise? Figure 3 suggests that this was a combination of shedding workers (employment contracted over the period) and rising capital-intensity (the capital to labour ratio rose over the period). Put simply, the story is that those workers who retained or gained employment typically enjoyed rising real wages, and because capitalists were able to respond by driving up labour productivity, the profit share rose. In other words, employed workers and capitalists benefitted at the cost of those who lost their jobs, or who were hoping to obtain work.

Figure 4 looks more closely at the trends in profitability, i.e. looking at the factors which drove trends in the profit share and the profit rate (i.e. the rate of return on capital) over the recent period. As trends in the profit share vary according to capital utilisation, such profit rate deconstructions are best made across peak-to-peak periods (i.e. 1989–1996, and 1996–2007). Data are also provided for the period from 2007 to 2009 to bring the analysis up to date (but should be treated with caution as these are recessionary conditions rather than reflective of long-term trends).

As expected, Figure 4 shows that the profit share rose because labour productivity growth exceeded that of real wages. This, in turn, boosted the profit rate. Broadly speaking, the growth in the profit rate is equal to the growth in the profit share plus growth in capital productivity. Figure 4 shows that rising capital productivity also contributed to rising profit rates. In other words, the growth path was characterised by a more efficient use of capital and a more efficient aggregate use of labour (brought about in large part by contracting employment). Since 2007, however, the profit share has been squeezed by real remuneration rising substantially faster than labour productivity and by sharply falling capital productivity (owing to recessionary conditions). Restoring profitability by boosting growth and ensuring a better match between wage and productivity growth must be a priority for the future if private sector investment is to recover and act as the prime source of job creation.

To expect the economy as a whole to generate large-scale new employment without making it substantially easier (and more cost-effective) to hire large numbers of lower-skilled workers, is unrealistic. Such growth as will occur in the current high real average wage environment will favour small numbers of highly skilled workers and will rapidly come up against the skilled labour constraints. Even assuming that the government is able to improve South Africa’s educational and training institutions, it seems most unlikely that this will occur fast enough to fuel anything like a 6% growth rate.

The vision of a high-wage, high-productivity growth path has enjoyed strong staying power in the post-apartheid period. The central idea, evident in the old Reconstruction and Development Programme and in the trade union aligned Industrial Strategy Project, is that South Africa should seek to move up the value chain (if necessary using the wage as a lever to force employers to become more efficient) and compete in skill-intensive and knowledge-intensive industries. The NGP is the latest manifestation of this vision. The hope is that it will create such a growth dynamic that jobs will grow elsewhere through the multiplier effect and through active government programmes, such as to promote agricultural co-operatives, public works programmes and the like.

Given the limited size of South Africa’s domestic market, this of course requires strong export growth. The NGP talks about increasing trade with Africa, China, India and Brazil and developing stronger exports of value-added manufactured goods (rather than commodities). But where South Africa is assumed to have a competitive advantage is left unspecified. Furthermore, the status of imports of low-wage goods from China and Africa is left unexamined. Are they to be allowed, or blocked by tariffs to protect South African jobs? This is just one of many questions which are likely to arise when selective and protective trade policy measures have to confront the trade-offs involved in growing international trade.

With regard to macroeconomic policy, the NGP proposes:

a careful balancing of more active monetary policy interventions to achieve growth and jobs targets, inter alia through a more competitive exchange rate and a lower cost of capital, with a more restrained fiscal stance and reprioritisation of public spending to ensure sustainability over time^1(p.15).

This proposal, however, involves a very tricky balancing act indeed.

Consider the issue of the exchange rate. South Africa has been experiencing sustained inflows of capital which not only cause the rand to appreciate, but are increasingly volatile and potentially destabilising (Figure 5). One of the proposed policies in this regard is for the Reserve Bank to accumulate more foreign exchange reserves to be used as a buffer and as a source of capital for a ‘sovereign wealth fund’ which ‘invests accumulated foreign reserves in productive projects with a higher yield than investment in developed-country bonds’^1(p.15) both in South Africa and the broader region. Such foreign exchange purchases could be financed by borrowing (selling bonds) or by printing money. As bond sales will put upward pressure on yields, thereby increasing rather than reducing the cost of capital, it seems more likely that the NGP envisages that foreign exchange is to be purchased by printing money. This, in turn, will be inflationary (and thus will cause the real exchange rate to appreciate) unless demand can be withdrawn from the economy in some other way – notably by a restrictive fiscal policy.

Now consider the issue of lowering the real cost of capital, i.e. lowering interest rates. This will reduce the cost of borrowing, thereby stimulating demand and further contributing to inflationary pressures. Lower real interest rates will also dampen the flow of foreign capital into the bond market, thereby depreciating the currency and further boosting inflationary pressures.

In short, the policies proposed for the Reserve Bank are likely to be inflationary, yet the NGP also states that the monetary policy stance will ‘continue to target low and stable inflation’^1(p.16). How are we to make sense of this? One possible interpretation is that the NGP is suggesting that the Reserve Bank purchases foreign currency by printing money and by lowering the interest rate – but if fiscal policy does not succeed in withdrawing demand out of the economy and inflation starts to rise above targeted levels, then the Reserve Bank will allow interest rates to rise. In other words, a lot depends on the government being able to run a budget surplus and otherwise reduce inflationary pressures to ensure that the currency and interest rates fall in real terms.

The NGP accepts that fiscal policy should be tight – though it also rather confusingly states that government should ‘enhance domestic demand’^1(p.17). It is thus rather unclear precisely how government expenditure is to be managed or how the long list of new policies and interventions mentioned in the NGP can be made consistent with an expanding social wage and tight fiscal policy. It is only when the Treasury becomes involved in the discussion of the NGP that it will become clear just how hard the ‘hard choices’ the NGP alludes to are going to be.

The NGP does not only propose a tight fiscal policy to control inflation. Other anti-inflationary mechanisms include a more effective competition policy, increasing the supply of skilled labour and encouraging wage moderation. As noted earlier, wage moderation is a central feature of social democratic compromises, and it is interesting that the NGP has put this on the table. Perhaps the excessive public sector wage growth over the past few years and the challenge of controlling inflation into the future have concentrated the minds of economic planners and steeled their spines with regard to confronting organised labour on this issue.

Two challenges are involved here: to get trade unions to accept the principle of wage moderation and to develop mechanisms which enable it to be put into practice. South Africa’s trade unions typically resist suggestions to link their demands to economic trends, preferring instead to fight for absolutist demands such as ‘living’ wages and ‘decent’ work. Achieving a shift in trade union thinking such that wage growth and working conditions be linked to issues of productivity and employment growth is a major ideological challenge. This is why some constraints on the obscene levels of some executive pay packages may well be necessary. An executive pay cap on salaries over R550 000 seems rather low (pay caps on salaries of over R5 million would target obscene salaries more clearly), but even so, there is some value in the suggestion. One of the problems the South African economy routinely faces in periods of economic upswing is inflationary pressure arising from skilled labour shortages. A cap on all salaries (not just on managers and executives) of over R550 000 would have the benefit of reducing this wage pressure substantially because it would make it much more difficult for firms to poach skilled labour from each other using inflated salaries as an inducement. But for this to work, one needs very strong buy-in from both labour and capital. Even in the highly co-ordinated Nordic social democracies, wage drift was a persistent problem in the post-war period, especially for skilled labour, and was a key reason for the eventual breakdown of the system.

If wage growth can be moderated in real terms, and if government’s proposed attempts to boost efficiency in both the public and private sectors succeed, then rising labour productivity and rising employment will be possible in a context of stable (if not slowly increasing) profitability. This, in turn, should be good for investment and growth over the longer term. This early attempt to put wage moderation on the table is thus to be welcomed. It can potentially facilitate productive discussion in the National Economic Development and Labour Council (Nedlac) and other forums about the linkages between growth in wages, employment, productivity and profitability.

There are, however, real question marks over how this wage moderation is to be affected. Wage moderation requires either national framework agreements by powerful employer and labour unions with the power and will to discipline their members to stick to it (the old Nordic model) or it requires agreement in a leading sector (e.g. the metals industry in Germany) which then sets the parameters for wage growth elsewhere. South Africa has neither. The 2010 OECD economic assessment of South Africa report suggests that national-level agreement on the path of wage growth should be backed up by labour-market reform, notably the removal of the extension of collective bargains to non-parties. The reasoning being that if unionised workers demand too high a wage increase, they would know that their jobs could be undercut by workers outside of the bargaining council. The report states:

Currently South Africa is characterized by an intermediate level of wage co-ordination, which is found elsewhere to be associated with poor employment outcomes. Increased co-ordination could be achieved by bringing social partners together at the beginning of each annual wage negotiation round and getting agreement on guidelines for increases that year. Actual bargaining would continue to take place in the same way as it does at present, but against the background of such guidelines. Government involvement in the process could help to make the trade-offs between wages, employment and unemployment clearer to the social partners. Weakening the extension of sectoral bargains would likely help with wage moderation, since social partners would know that agreed wage levels could be undercut by other firms^6(p.18).

This would probably help moderate wage increases and boost employment growth. But it, of course, will be seen as undermining the agenda of decent work…

Figure 5: The balance of payments in South Africa, 1990-2010.

The NGP is a frustrating document in that it sets out a vision which is possible, but so conditional on a range of structural, organisational and ideological changes that it veers towards the utopian. The failure to confront the trade-offs between wages, employment, productivity and profitability is particularly disappointing. It is also frustratingly short of detail and long on wishful thinking. My copy is covered in annotations such as: ‘what does this mean, exactly’ and ‘how is this to be achieved?’ And, lurking in the back of my mind are questions like: ‘I can see that mobilising savings to fund productive infrastructural investment is potentially a good idea, but will the infrastructural investment be done effectively and in areas which maximise growth?’ and ‘Do I really want my retirement funds tied up in development bonds to finance government projects?’

But it is also a potentially game-changing proposal. Government inefficiency and excessive wage growth have at least been tabled as serious constraints on the vision of a new, inclusive, economic growth path. There is some worrying dirigiste discourse, but this is accompanied by a social democratic discourse running through the document in a different register. Negotiation and partnerships with business and labour is on the cards. The challenge is to rise to this opportunity and make the best of it.

1. South African government. The new growth path: Executive summary [homepage on the Internet]. No date [cited 2011 Mar 02]. Available from: http:// www.agbiz.co.za/LinkClick.aspx?fileticket=NiqkybzD%2FeA%3D&tabid=351

2. Statistics South Africa. Labour Force Survey: Historical revision September series 2000 to 2007. Statistical release P0210 [document on the Internet]. c2009 [updated 2009 Mar 23; cited 2011 Mar 02]. Available from: http://www.statssa.gov.za/publications/P0210/ P0210September2000,2001,2002,2003,2004,2005,2006,2007.pdf

3. Africa. Quarterly Labour Force Survey, Quarter 3, 2010. Statistical release P0211 [document on the Internet]. c2010 [update 2010 Oct 26; cited 2011 Mar 02]. Available from: http://www.statssa.gov.za/publications/P0211/P02113rdQuarter2010.pdf

4. Spence Commission. The growth report: Strategies for sustained growth and inclusive development [homepage on the Internet]. c2008 [cited 2011 Mar 02]. Available from: http://www.growthcommission.org/index.php?Itemid=169&id=96&option=com_content&task=view

5. Hausmann R. Final recommendations of the International Panel on ASGISA [homepage on the Internet]. c2008 [cited 2011 Mar 02]. Available from: http://www.hks.harvard.edu/centers/cid/publications/faculty-working-papers/cid-working-paper-no.-161

6. Organisation for Economic Co-operation and Development (OECD). Economic survey of South Africa [homepage on the Internet]. c2010 [cited 2011 Mar 02]. Available from: http://www.oecd.org/document/20/0,3343,en_2649_33733_45637781_1_1_1_1,00.html

7. Statistics South Africa. Quarterly Employment Statistics, September 2010. Statistical release P0277 [document on the Internet]. c2010 [updated 2010 Sept; cited 2011 Mar 02]. Available from: ( http://www.statssa.gov.za/publications/P0277/P0277September2010.pdf)

8. South African Reserve Bank. Quarterly Bulletin no. 258 [document on the Internet]. c2010 [updated 2010 Dec; cited 2011 Mar 02]. Available from: http:// www.reservebank.co.za/internet/Publication.nsf/LADV/ECCBDD0C0B337943422577F3003ECF02/$File/QBDec2010.pdf