Article High Productivity Now: a critical review of South Africa's growth strategy Nicoli Nattrass In order to make any dent on unemployment, South Africa's growth path needs to become significantly more labour-demanding. But what is the best way of achieving that end? On the face of it, the problem is simple: job creation depends crucially on investment growth, hence appropriate strategies are those which encourage capital accumulation and ensure that as many jobs as possible are created per unit of investment. But this apparently focussed answer raises two further fundamental questions: what drives investment; and what kind of investment is best suited to sustainable, labour-demanding growth? The first part of the paper considers some of the macroeconomic determinants of investment, and asks whether South Africa's macroeconomic growth strategy is' investor-friendly'. Although the jury is still out, it seems that the architects of the government's Growth, Employment and Redistribution (GEAR) framework underestimated the negative impact of constraining demand on private investment. But the determinants of growth extend beyond the macroeconomic policy stance and into the arena of the growth path itself. Part Two considers the question of what broader growth strategies are good for growth. In doing so, it considers some of the 'lessons' from international experience for growth. The South African experience is referred to where relevant. Part Three considers the narrower question as to what kind of investment is appropriate for a labour-demanding growth path. Is it better to go for maximising the number of jobs created now - in which case labour- intensive investment is preferable - or is there an argument in favour of capital- and skill-intensive investment as a catalyst for more dynamic and TRANSFORMATION 45 (2001) ISSN 0258-7696 Nicoli Nattrass labour-demanding growth later? 'High productivity now' (HPN) strategies (which appear to inform South Africa's industrial and labour policies) opt for the latter. 'Hard HPN' strategies seek to discourage (or even eliminate) low-wage, low-productivity activities, whereas 'Soft HPN' strategies seek merely to encourage greater productivity growth. It is argued that South Africa's industrial policy stance is consistent with Soft HPN, but that labour market policy towards wage-setting is more in line with Hard HPN thinking. Investment and the macroeconomic environment The question as to what drives investment is one of the most vexed in economics. Answers range from Keynes' s 'animal spirits', to more specific determinants such as savings, the interest rate, and current and expected profit rates etc. Given that the search for pro fit underpins private investment, all of the above probably have a role to play: higher expected returns attract higher investment; and the lower the cost of borrowing (and higher the level of savings), the greater the range of profitable investment projects. But these variables only capture part of the picture. The range of profitable investment projects is also a function of the available infrastructure, the existence of support services and the supply of available inputs, market size etc. In other words, less easy-to-measure factors such as the level and pattern of development, and the nature of government support, are also important. In his review of empirical literature on the determinants of investment, Chirinko found that the most significant determinant was (lagged) demand: investment rose when economic growth was strong, and fell when it weakened (1993:1883). This suggests that investors feel more confident about future streams of income when expenditure is rising, and hence are more prepared to make investments. But there is a clear limit on this relationship: if the rise in demand is so strong that it undermines macroeconomic balance (ie results in inflation and a depreciation of the exchange rate) then profitability may be threatened. The history of macroeconomic populism in Latin America shows very clearly that destabilising increases in demand undermine rather than induce investment. The message is thus that a stable, but growing, economy is probably the most suited to attracting high and sustained levels of investment. A critical review of SA's growth strategy In this regard, one of the lessons of the Asian growth path is that inflationary financing of budget deficits should be avoided. The eight highly performing Asian economies (HPAEs), namely Japan, Hong Kong, South Korea, Singapore, Taiwan, Indonesia, Malaysia and Thailand, chose either to keep budget deficits small (eg South Korea), or in the case of Malaysia and Thailand (which ran large deficits), took advantage of high domestic savings and the rapid increase in demand for financial assets which accompanied growth (Page 1997:25). As a result, inflation was lower, and real interest rates far more stable, than in Latin America. Such an environment was ultimately more conducive to investment (Birdsall and Jaspersen 1997). Differences in savings may be one of the reasons for cross-country differences in investment. Whereas the HPAEs were able to mobilise a huge pool of domestic savings to finance investment, the Latin American countries (and South Africa) were characterised by relatively low rates of saving and investment (Jaspersen 1997). However, as savings and investment tend to follow growth (Gavin et al 1997), these savings and investment rates are probably more a result, than a cause, of differences in growth performance. One of the benefits, however, of a large pool of domestic savings was that the HPAEs were able to avoid the serious debt crises which beset most of Latin America, were less prone to destabilising capital flight, and were better able to weather the external shocks of the early 1980s. Is South Africa's maCroeconomic policy stance conducive to investment? Yes and no. The GEAR strategy of reducing the fiscal deficit (and avoiding any inflationary financing thereof) ought to support investment in the sense that greater stability is being injected into the macroeconomic environment. But to the extent that this restrictive fiscal stance withdraws demand from the economy, investment demand is dampened accordingly. Thus, depending on which effect one imagines is more important, one can conclude that GEAR is either good or bad for investment. In deriving their predictions for future trends, the architects of GEAR (RSA 1996) assumed that investors would respond quickly and well to the lower deficit (and to the government's commitment to continued deficit reduction). The assumption was that the positive impact of increased investor confidence on investment would outweigh the negative impact of tighter fiscal policy on demand and investment (Nattrass 1996). Nicoli Nattrass Table 1. GEAR vs Reality PredictedResults cfCEAR Annual Growth Rates 1996 1997 1998 1999 Real GDP 3.5 2.9 3.8 4.9 Real Bank Rate 7 5 4 3 Private Investment 9.3 9.1 9.3 13.9 Employment (non-agricultural) 1.3 3.0 2.7 3.5 Inflation 8.4 10.9 9.6 7.7 Real Wage Growth (Private) -0.5 1.0 1.0 1.0 Change in Real Effective Exchange Rate -8.5 -0.3 0.0 0.0 Conventional Deficit/GDP -5.1 -4.0 -35 -3.0 AtudPerfonrmce Real GDP 4.2 25 0.6 1.2 Real Bank Rate* 8.8 7.4 8.1 7.3 Private Investment 6.1 4.7 -2.9 -4.4 Employment (non-agricultural) -0.7 -1.7 -3.7 -3.2 Inflation (CP1) 7.4 8.6 6.9 5.2 Real Wage Growth (Private) 1.7 23 8.6 3.0 Change in (average) Real Effective Exchange Rate -6.3 6.4 -92 _ Conventional deficit/GDP (December 31) •A.9 -4.6 -33 -2.6 Source: SARB Quarterly Bulletin, March 2000. T h i s is equal to the bank rate - CPI for 1996-1998, and the repo rate - CPI for 1999 Has it worked? On the 'positive' side, the fiscal deficit target was reached, and inflation fell to lower levels than predicted. However, the drop in the deficit was not accompanied by lower interest rates - indeed, the real bank rate rose sharply - and private investment has remained sluggish. As can be seen in Table 1, South Africa's investment, output and employment growth was way below target during the 1990s. This is consistent with the warnings of economists across the ideological spectrum that GEAR would reduce demand, and that private investment would follow demand downwards, rather than compensating for it.1 It is also consistent with evidence that aggressively anti-inflationary components of stabilisation packages have undermined, rather than supported, growth (Stiglitz 1998: 7-9). The GEAR modellers were almost certainly unrealistic in their assumption that business confidence would rise (drawing investment up with it) once the GEAR policies had been announced to the business community. A critical review of SA's growth strategy But not all of South Africa's poor growth performance can be laid at the door of GEAR. From early 1998, the economy suffered from the contagion effects of the Asian crisis that affected growth in ways that could not reasonably have been predicted. However, even as early as 1996 and 1997, differences emerged between the GEAR projections and actual economic trends. Furthermore, it appears that South Africa's growth performance was significantly worse than that of other middle-income countries, and worse than that which would have been predicted solely on the basis of changing world market conditions (Weeks 1999:804-8). This implies that the GEAR strategy exercised a major, and independently negative, impact on growth and investment. Another reason why the GEAR targets were not met may have been because the GEAR strategy was not implemented consistently, particularly with regard to labour-market policy. In contrast to the labour-market vision outlined in GEAR, the government passed various pieces of legislation that increased the costs of employing labour2 and extended minimum wage floors. Promised reforms to existing labour laws - such as amendments to the mandatory extension of collectively bargained agreements to non- parties - did not materialise.3 According to evidence from the OECD (1999:156-9), countries that have undergone macroeconomic stabilisation without addressing labour market rigidity have found the experience costly in terms of unemployment. South Africa is unlikely to be an exception. As can be seen in Table 1, instead of generating jobs, GEAR has presided over significant job losses in the formal (non-agricultural) sector. Although much of the decrease in employment (eg in mining) cannot be blamed, directly on GEAR, it is nevertheless feasible that co-ordination failures between fiscal, monetary and labour-market policy could be exacerbating unemployment (Nattrass 1998). And, even with regard to short-term poverty-relief employment programmes, the government appears not to be able to deliver on the jobs front. Although great plans were trumpeted about short-term job creation for the unemployed in public works programmes and the like (eg RSA 1998), the government (with the exception of the Department of Water Affairs) has had little success in deploying what little funds were actually allocated for such purposes.4 Note that there is some evidence of a marginal increase in employment in the iate 1990s, but this has not been enough to reverse the general downward trend of the decade. According to figures from the October Nicoli Nattrass Household Survey, total employment rose slightly between 1996 and 1998, but unemployment increased as the labour force grew even faster. According to these official statistics (which are questionable, particularly with regard to agriculture (see Simbi and Aliber 2000), total employment increased in these years because employment in agriculture and domestic services supposedly grew sufficiently to compensate for the decline in employment elsewhere. Even if employment did increase in this way, it is nevertheless worrying because informal employment tends to be of a survivalist nature, and agricultural employment tends to be seasonal and vulnerable to droughts and other disasters. There is also evidence of increased casualisation of the labour force (Kenny and Webster 1998), some of which is not picked up in official statistics. However, sustained growth in living standards is best based on a healthy growth in full-time formal employment - particularly in manufacturing and services. Hence trends in non-agricultural formal employment remain an important economic growth indicator. The macroeconomic policy stance is only one aspect of the growth path. Given that growth itself drives investment, it is thus necessary to consider the broader question of the growth path in order to gain a fuller picture of what drives investment. The determinants of growth: lessons from international experience? The international literature on economic growth is vast and varied. If there is a broad consensus on the 'lessons' of successful economic growth it is probably this: an export orientation and an educated labour force are good for growth. The relative importance of state versus the market in development remains a subject of great debate. • Export orientation and growth Perhaps the most oft-cited 'lesson' from international experience is that producing for the export market is good for growth. The HPAEs were able to expand output and employment dramatically (particularly in the manufacturing sector) through producing for the rapidly expanding external market.5 Between 1965 and 1990, per capita GDP averaged 5.6 per cent pa in the HPAEs, whereas per capita incomes grew by a mere 1.2 per cent pa in the major (inwardly-oriented) Latin American economies over the same period (Birdsall and Jaspersen 1997). South Africa followed the Latin American route, becoming one of the first developing countries to opt for A critical review of SA's growth strategy inward industrialisation (McCarthy 1998:66). And, having out-performed most similar developing countries during the first part of the twentieth century, the South African economy, like its Latin American counterparts, ran out of steam in the 1970s and 1980s. Whereas other countries benefited tremendously from the ideal external economic conditions of 1950-73, South Africa's 'post-war leap' was the 'lowest on record' (Moll 1993:5). This 'lesson' needs qualification in at least three important respects. Firstly, export-led growth has not been synonymous with trade liberalisation (as often assumed by proponents of neo-liberal economic reforms). Governments of the HPAEs provided a high level of temporary effective protection to firms and sectors conditional on their successful export performance. Secondly, an export orientation is only good for growth when external markets are expanding fast. The experience of Latin American countries that embarked on significant economic liberalisation in the 1980s and 1990s is instructive in this regard. Whereas the northern Latin American countries (which export mainly to the United States) experienced strong export growth in the mid- to late-1990s, this was not the case in the rest of the region. Panama, and all countries to the South, experienced a decline in exports in the mid-1990s because of the recession in their major export market (Asia and Europe). Morley describes the phenomenon as 'export- led decline' (2000:26-27). Thirdly, it is important to realise that successful export-led growth depends on controlling costs, and keeping wage growth in line with productivity. In the 1960s and 1970s, the HPAEs dominated the low-wage, labour-intensive end of international trade. As manufacturing exports expanded rapidly, labour was increasingly drawn out of lower value-added sectors and into manufacturing. Over time, the manufacturing sector itself started moving into higher value-added product lines - a process aided by targeted industrial policies (Weiss 1998) and the fact that most workers had a basic education (Wood 1994). Wage flexibility appears to have kept unemployment low and steady, and a more rapid rise in productivity than wages (up until at least the mid-1980s) helped fuel investment and growth (World Bank 1993). Strong growth in output and employment ensured that wages and living standards rose. Unfortunately, by the time that South Africa and Latin America started liberalising their trade regimes, low-wage countries like Bangladesh, Indonesia, Pakistan and China had moved into the lower value-added end of the international market. This meant that those middle-income countries Nicoli Nattrass with developed industrial structures and intermediate ratios of skilled to unskilled labour, experienced a fall in demand for unskilled labour as competition from low-wage economies increased at the lower end of the value-chain (Wood 1997, OECD 1997, Dirwan and Walton 1997).6 In Latin America, the result has been falling demand for unskilled labour and an increasingly skill-intensive growth and inegalitarian growth path (Stallings 2000, Morley 2000). Similar trends are evident in South Africa. Manufacturing exports have grown, but mostly in capital- and skill- intensive product lines (Bell and Cattaneo 1997:6-7, ILO 1999:14-16, Edwards 2000), and increased import penetration has had a significantly negative effect on employment in ultra-labour-intensive sectors.7 There is evidence that South Africa is moving towards more (intermediate) skill- intensive product lines, with technological change in manufacturing resulting in an increased demand for skilled labour (Standing 1997). The demand for skilled workers (particularly in the information technology sector) has been rising, and the general shift towards capital intensity has further increased the demand for skilled workers, and reduced the demand for unskilled workers (Bhorat and Hodge 1999). This suggests that rising wage inequality and persistent unemployment will characterise the South African growth path for some time. • State versus market There is general agreement that competition is necessary for sustained growth. However, the exact role that the state should play in the development process remains the subject of much debate. Except for Hong Kong (and to a lesser extent Singapore), the state played a highly interventionist role in the HPAEs with regard to the direction of investment and industrial policy. The developmental states of East Asia mobilised resources (in Korea's case by nationalising the financial sector), acted entrepreneurially where necessary (in Taiwan's case through state-owned enterprises) and otherwise coerced and disciplined capital (see, for example, White 1988). Indeed, the coexistence of rapid economic growth, selective protection and targeted support of key industries in South Korea, Japan and Taiwan lead some analysts to conclude that industrial policies were central to the growth of total factor productivity and technological catch-up (eg Amsden 1989, Wade 1990, Weiss 1998). A corollary of this argument is that efficient bureaucracies and performance-related support measures enabled East Asian industrial policies A critical review of SA's growth strategy to avoid the pitfalls of rent-seeking, corruption etc, that characterised Latin American support of industry with tariff barriers. Any policy of strategic intervention is risky in the sense that state officials have to make educated guesses about which industries or firms will turn out to be winners. Tying support to clear performance criteria - most notably measurable trends such as export performance - and being willing to drop support in cases of clear failure - helps limit the potential for wasting resources. According to Evans (1995), the success of interventionist states like Japan lay in the fact that they were both 'embedded', in the sense that information flowed easily and well between bureaucrats and firms, and 'autonomous', in that they were able to withstand pressure from firms to maintain support where it was not efficient to do so. This helped facilitate the development of specific state capacities geared around industrial policy (Weiss 1998). Other analyses (eg World Bank 1993, Page 1997), however, play down the role of strategic intervention and attribute the success of the HPAEs to the business environment common to all of them, ie stable macroeconomic policies, stable and competitive exchange rates and high savings rates. Page, for example, points out that productivity growth in Korea and Taiwan was high by international standards, but not higher in sectors targeted for special attention by such industrial policies (1997:46-47). He thus concludes that export orientation — rather than industrial policy - was the main reason behind productivity growth (1997:48). However, aggregate industrial statistics can easily hide the impact of industrial policy targeted at firm (rather than industry) level; so, to an important extent, this debate remains moot. The debate over state intervention and market signals recently resurfaced again as a result of the 1998 Asian crisis that started in Thailand and reverberated across the globe. The IMF adjustment packages for Thailand, South Korea and Japan (which placed strong emphasis on liberalisation and deregulation) implicitly blamed government intervention in large part for the crisis. Others, however, attributed the crisis to the financial deregulation that occurred in East Asia during the 1990s, and predict that IMF policies are pushing Asian economies in precisely the wrong direction (see, for example, Chang 1998, Stiglitz 1998, Jomo 2000). The debate about the role of state and market in East Asia is clearly far from over. The history of government intervention in industry in South Africa is not an edifying one. Although massive government investments in electricity (ESCOM) and steel (ISCOR) arguably helped create a basis for the Nicoli Nattrass expansion of the manufacturing sector, the focus on mega projects and armaments production tended to inject a strong bias towards capital- intensity in industry (Kaplinsky 1995). So too did restrictions on the employment of African labour in metropolitan areas, subsidies to producers, accelerated depreciation allowances, negative real interest rates and racial discrimination in skills acquisition. In addition, the state wasted vast sums of money in attracting industry to the 'border' areas of homelands and to low-wage 'decentralised' areas (Haines 1996). Incentives of 100 per cent of the wage bill were offered along with transport and rent subsidies. While labour-intensive forms of production were indeed attracted to poor areas, many of the industries (particularly in clothing) were decentralising anyway because of wage pressure in metropolitan areas (Bell 1997). Many of these firms stayed on after the withdrawal of incentives (Sharp and Spiegel 1996) and are now being threatened by the minimum wage determinations - from which they had previously been exempt (Nattrass 2000a). During the 1990s, the decentralisation strategy was restructured and simplified and now consists of a tax holiday scheme, spatial development initiatives and industrial development zones (DTI 1998:23-36). The idea is to create industrial clusters in key areas in order to promote agglomeration economies, co-operative relations between firms, and backward and forward linkages. It is too early to tell if this approach will in fact deliver dynamic, labour-demanding growth. So far, the investment costs have been very high in relation to the number of jobs created. The spatial development initiative (which the DTI regards as 'one of its most successful programmes' has so far created 10,000 jobs at an investment cost of R17 billion and a further 400 programmes are in the pipeline promising to create 6800 jobs at a cost of R83 billion (DTI 1998:28). At between R1.2 million and R1.7 million a job, this is a very capital intensive outcome! Industrial policy likewise nods in the direction of supporting labour-intensive sectors, but in practice has failed to do so. Even the tax holiday scheme, which was aimed in part at encouraging labour-intensive manufacturing, has succeeded only in creating jobs at a cost of R237.000 each (DTI 1998:26). The record relating to the Industrial Development Zones is similarly uninspiring. Rather than initiate large-scale job creation, the DTI has succeeded mainly in attracting large capital-intensive mega projects (1LO 1999:25) - which is not surprising given that existing labour regulations and standards are enforced in these areas. 10 A critical review of SA's growth strategy • Education, growth and inequality As emphasised by the new 'endogenous growth' theory, higher levels of accumulation of both physical and human capital are vital contributing factors to per capita income growth as it is the interaction between ideas and accumulation which results in increasing returns to scale of physical and human capital. This literature suggests that if human to physical capital ratios are initially high, then a country's subsequent economic performance will be characterised by faster investment and per capita income growth. Empirical work by Barro (1991) - which shows that for a given quantity of initial human capital, a poor country tends to grow faster than a rich country, so that incomes converge over the period among countries with similar levels of education - supports this hypothesis. This is suggestive of the costs, in terms of the waste of human potential and lost growth, of apartheid education and training policies on South African development. It also supports the emphasis that South Africa is placing on improving education and skills development. An increase in the supply of educated labour is only beneficial for growth if it is matched by a similar increase in demand - as occurred in the HPAEs (Birdsall et al 1997:103-4). Increasing education on its own, is not a sufficient condition for growth, and rapid growth is required to finance education and to boost the creation of both skilled and relatively unskilled jobs. The provision of health and educational services is a necessary condition for the creation of a 'virtuous cycle' between human development and growth (Ramirez et al 1997). But it is not a sufficient condition. Note also that when growth is particularly demanding of workers with a basic education relative to those with more skills, then wage inequality will fall - thus contributing towards an egalitarian growth path. This was the case in the HPAEs during the early growth phase. However, where growth is skill-intensive - and particularly when skilled labour is in short supply - then wage inequality is likely to widen. This outcome was typical in countries pursuing ISI strategies, including Latin America and South Africa. Latin America has subsequently shifted away from ISI and inequality has widened further- largely as a result of increased inequality within the skilled section of the workforce (Morley 2000). In South Africa's case, two forces appear to be acting on the wage distribution. On the one hand, the demand for skilled labour is putting upward pressure on the top end of the wage distribution. On the other hand, trade union pressure appears to be favouring lower paid workers (Schultz 11 Nicoli Nattrass and Mbawu 1998), thus contributing to a narrowing of wage inequality. But wage inequality is not the only factor driving inequality in South Africa. The gap between employed and unemployed households is also a significant driving factor (Leibbrandt et al 2000, Seekings 2000). To the extent that higher unskilled wages restrains job creation, a narrower wage distribution may be associated with higher overall inequality. In general, improving the productivity of the poor through education and training should help reduce inequality, as it allows them to participate more fruitfully in the economy and obtain a bigger share of the growth dividend. However, this must be matched by an increase in income-earning opportunities -particularly jobs - if inequality is to fall. Where large pools of surplus-labour exist, creating jobs (even low-wage, low-productivity jobs) is almost certainly a necessary first step. In South Africa's case, where unemployment is a major determinant of inequality, job creation for the unskilled is a necessary condition for any significant and sustained reduction in inequality. This means that either capital accumulation has to be so rapid that employment and incomes rise along with an increase in capital intensity, orthatmore labour-intensive sectors expand significantly alongside higher productivity activities. Dilemmas of the High Productivity Now (HPN) growth path This brings us to the question of a potential trade-off between short-term job creation, and laying the basis for sustainable growth and higher future job creation. Is it better to go for maximising the number of jobs created now - in which case labour-intensive investment is preferable - or is there an argument in favour of capital- and skill-intensive investment as a catalyst for more dynamic and labour-demanding growth later? In the case of the HPAEs, there was no tension between these objectives because the growth path was labour-demanding, although not, in the main, labour-intensive (Birdsall 1997:101-2, Jaspersen 1997:75-82). To be sure, workers were initially drawn into labour-intensive sectors and activities, but as economic growth gathered pace, the production structure shifted steadily towards higher productivity activities. Labour productivity rose, but so did employment because of the even faster (export-driven) increase in output (Galenson 1992). There appears to be little chance of replicating this growth path. Firstly, the international conditions facing South Africa are very different to those enjoyed by the HPAEs in the post-war boom. Secondly, the South African growth strategy is not premised on the notion that labour should first be 12 A critical review of SA's growth strategy drawn into low-wage, low-productivity activities, and then only subsequently be drawn up the value chain. Indeed, the very opposite notion appears to be driving labour and industrial policies, namely, that high value-added, high (labour) productivity8 - and hence high-wage - activities should be promoted wow in order to transform South African manufacturing into a dynamic and competitive lead sector. On one level, it seems absurd given South Africa's massive labour surplus, to be engaging in any form of increase in labour productivity. Indeed, in a relatively capital-scarce and labour-abundant economy, there is a strong argument to be made that it is capital productivity that should be maximised, even if this implies falling labour productivity.9 However the implication of a low wage labour path is unpalatable to organised labour, and to proponents of HPN who populate the corridors of the Ministry of Labour and the Department of Trade and Industry (DTI). This 'high productivity now' (HPN) strategy assumes that even in a labour surplus middle-income economy like South Africa, it is necessary to increase productivity today in order to project the economy onto a more dynamic (and ultimately more labour-demanding) growth path tomorrow. It belongs in the class of growth strategies which assume that competitiveness rests primarily (if not solely) on the adoption of cutting-edge technologies, on nurturing firms and segments of the economy that demonstrate competitive advantage,10 and otherwise generally encouraging structural change in favour of higher value-added activities..Proponents of such views point to the success of government industrial policies in the HPAEs, and locate themselves within the theoretical ambit of endogenous growth theory, highlighting the dynamic benefits of human capital, learning-by- doing, the spill-over effects of technology etc. South Africa's Industrial Strategy Project (ISP) is an example of this genre (Joffe et al 1995), and much of this thinking is evident in DTI documents (DTI 1995, 1998). The message is attractive: invest in people, technology, infrastructure, work-place re-organisation, inter-firm co-operation etc, and achieve the win-win scenario of greater competitiveness, a better-paid workforce, and faster, sustainable growth. But behind the lure and glamour lurks the cost: significant employment creation is relegated to second- or third-round effects. HPN is, in other words, a reincarnation of the old 'trickle-down' story: increases in productivity drive the rising tide of growth; the unemployed must go to night school and training programmes whilst waiting for the employment waters to rise. 13 Nicoli Nattrass HPN comes in 'soft' and 'hard' versions. The soft version argues that active support should be given to higher productivity firms and sectors, and productivity improvements should be sought wherever possible. Such a strategy only harms low-wage, low-productivity activities in the sense that resources are not channelled in their direction. DTI policies appear to be supportive of this soft HPN position. Hard HPN, by contrast, not only promotes high productivity through supportive measures, but engineers an economy-wide shift in favour of higher productivity firms and sectors by actively undermining (even destroying) low-wage, low-productivity sectors. Using wage increases to force low-productivity activities out of business would constitute part of a hard HPN strategy. The ISP, for example, comments favourably on the German system of wage determination which 'keeps wages higher, and differentials lower than would be determined by a freely-functioning labour market' and thus 'encourages investment in training and retraining as a mechanism to enhance productivity to match these wages'. A highly regulated labour market, argues the ISP, will 'encourage restructuring up the value chain rather than restructuring towards low-wage, low-productivity forms of production' (Joffe et al 1995:213). The ISP situates COSATU's approach to industrial restructuring within this particular logic, pointing out that it is 'premised on the need to move South African firms out of their low-wage, low-skill, low-productivity vicious circle in which they are out-competed by the second-tier Newly Industrialising countries' (1995:214). The presentation of binary opposites - a low-wage, low-productivity 'vicious circle' versus a dynamic high-wage, high-productivity nirvana —. is typical of hard HPN. The notion that one can have high-wage, high- productivity firms and sectors alongside lower-wage, lower-productivity activities is an anathema to hard HPN thinking. The ruling assumption appears to be that low-wage, low-productivity firms will necessarily undermine high-wage, high productivity activities. Talk of'vicious circles' implies that there will be a race to the bottom of the value-chain unless a policy of zero tolerance is adopted towards low-wage, low-productivity activities. Using the wage to drive out low-wage, low-productivity sectors is part of such a strategy. • Using the wage to force restructuring up the value-chain (hard HFN) Increasing the costs of employing labour can be a deliberate part of a high productivity growth strategy (see for example Lee 1996:495). It was, for example, an integral part of the philosophy behind the post-war 14 A critical review of SA's growth strategy 'Scandinavian model'. Centralised wage bargaining set wages across all firms, regardless of productivity performance, thus putting pressure on low-productivity firms and sectors (Henley and Tsakalotos 1993). Workers who lost their jobs as a result of such restructuring received generous welfare benefits whilst being retrained for employment. The strategy worked well in Sweden until it fell apart under the combined impact of productivity slowdown, unemployment, and fiscal crisis (Lindbeck 1997). The apartheid wage-setting machinery was used to similar effect in South Africa, except with the more limited aim of providing full employment for white workers (Nattrass and Seekings 1997). Once African workers were incorporated into the wage-setting machinery in a context of rising unemployment and inadequate labour market welfare provision, the system manifestly became incapable of supporting full employment for any group, and instead probably contributed to falling labour absorption. As Moll has argued (1996), larger firms use more capital-intensive techniques, achieve higher labour productivity, and hence pay higher wages than smaller firms. Data from the South African Manufacturing Survey supports this supposition: value added per worker rises with size of firm, thus indicating that larger firms tend to manifest higher labour productivity; and larger firms tend to be more capital-intensive and pay higher wages." As bargaining councils are dominated by large firms - where it is easier to organise workers12 - the bargained wage is more likely to suit capital- rather than labour-intensive activities. Compulsory extensions of collectively bargained agreements to non-parties thus probably harm labour-intensive firms and activities, and hence reduce labour absorption in manufacturing (see also Nattrass 2000b). As noted earlier, the ministry of Labour has not yet acted to amend the provision pertaining to compulsory extensions of collective bargains to non-parties - this, despite various government policy documents and statements in support of reform. Perhaps the ministry of Labour buys into a hard HPN view of restructuring, and is thus resisting such changes. The fact that the current minister of Labour was drawn from tiade union ranks lends credibility to this speculation. And, despite calls for greater lenience for small business with regard to minimum wages and labour standards, very little concrete has been done. Exemptions to bargained agreements remain largely in the hands of the parties to the agreement and wage exemptions are rarely given (see review in Nattrass 2000b). All of this is consistent with a hard HPN strategy which rests on the assumption that 15 Nicoii Nattrass growth has to be based on increases in average productivity, and that this has to be bought at any price - including the destruction of low-wage jobs. A central problem with this position is that there is no necessary reason why low-productivity, low-wage firms should out-compete high-wage, high-productivity firms. Firstly, they are often producing different quality products and hence are competing in different market niches. There is no reason why a range of firms, using a range of technologies, cannot exist side by side. Secondly, it is the ratio of wages to productivity that is important to competitiveness - not the absolute level of the wage. As long as high wages are justified by high productivity, there will be no 'race to the bottom'. A further difficulty arises with regard to the notion that economic growth has to be based on rising average productivity. Rising labour productivity can be achieved through the destruction of jobs (as has been happening in South Africa) or it can be achieved through upgrading technology, changing employment practices, shop-floor organisation etc. Either way, rising average productivity implies that less labour is being used per unit of output, and hence that the employment elasticity of output is falling. An alternative strategy is one which tolerates the expansion of low-wage, low-productivity jobs - even though this would slow down the rate of growth of average productivity. Indeed, given South Africa's unemployment crisis, it would be desirable to have a strategy which encourages rising productivity in the high-wage, high-tech, skill-intensive sectors, but which is characterised by falling average productivity as new low-wage, low-productivity jobs are created for the unemployed at an even faster rate. • The distributional dilemma of HPN The HPN strategy essentially asks the currently unemployed generation to make a sacrifice for the sake of the employed and better skilled amongst them, and for the next generation which will (supposedly) enjoy the fruits of a more dynamic economy. In the case of soft HPN, the sacrifice being asked of the less-skilled is time, and in the case of hard HPN it may be loss of income (as low-productivity jobs are destroyed) and time. One way of addressing this distributional dilemma would be if those who gained from HPN (either soft or hard) - ie high-tech industry and the better educated, skilled and currently employed workforce — were taxed sufficiently to finance adequate welfare grants and training programs for 16 A critical review of SA's growth strategy the unemployed. This was the implicit 'social contract' behind the Scandinavian model - but it only held together while growth was rapid and unemployment relatively low. In a labour-surplus middle-income country like South Africa, such a scenario seems highly unlikely. Although there is talk about a 'basic income grant' the numbers being bandied about are tiny (R100 a month), and it is questionable whether the political will exists to pay the associated tax bill.13 Those favouring HPN strategies have to make an ethical judgement about the costs involved. Not only do they have to be sure that a HPN strategy will actually deliver more labour-demanding growth in the future, but they have to be sure that the benefits to future generations are worth the costs (most notably unemployment) borne by many in the present. As Sen has pointed out, it is also worth considering that the cost of unemployment reaches beyond welfare payments: "The penalties of unemployment include not only income loss, but also far-reaching effects on self-confidence, work motivation, basic competence, social integration, racial harmony, gender justice and the appreciation and use of individual freedom and responsibility' (1997:169). Ultimately the decision as to which strategy is optimal depends on the prevailing social ethics and on the level of development. For a middle- income economy like South Africa with high unemployment and limited fiscal resources for providing adequate welfare, opting for a hard HPN strategy is fraught with danger. Firstly, it assumes that an aggressive high productivity strategy will indeed deliver the promised growth benefits. This is a leap of faith, because it depends on how well South Africa achieves that goal relative to similar strategies in other middle-income economies in East Asia and Latin America. Secondly, even if boosting average productivity does generate growth, it will exclude too many poor people for too long from the fruits of that growth. Notes 1. See, for example, Gibson and van Seventer 1995, Nattrass 1996, NIEP 1996, Standing et al 1996:33, NLC 1996, Samson 1997. 2. The Basic Conditions of Employment Act provided for longer annual and family leave (thus increasing the indirect cost of employing labour) and reduced hours of work (thus increasing hourly fixed costs). The over-time premium was also increased, with the result that over-time labour is now paid over two and a half times that of workers in comparable middle-income countries (Barker 1999:19). 17 Nicoli Nattrass 3. Both the ministry of Finance's GEAR strategy (RSA 1996) and the ministry of Labour's Employment Strategy Framework (RSA 1998:44) recommend that amendments be made to the extension of collectively bargained wage agreements to non-parties. Soon after the Labour Market Commission (LMC 1996) presented its report, the Minister of Labour announced that changes to the mandatory extension provision were immanent - yet nothing came of it. The State President made a similar announcement in early 2000, but this too appears to have been empty. 4. South Africa's new democratic institutions have been plagued with capacity problems which have undermined various development and poverty alleviation initiatives (Nattrass and Seekings 1998). The Welfare Department and the Department of Public Works appear to be particularly incapable of spending money on job creation and poverty alleviation programmes (see 'Now public works betrays the poor', in Sunday Independent June 11, 2000). 5. The export-led growth model applies more to Malaysia, Indonesia, Thailand (and Hong Kong and Singapore) than it does to Korea, Taiwan and Japan (Jomo 2000:4). 6. Although there is still controversy regarding how much of the decline in demand can be placed at the door of international trade, there is widespread agreement that trade with low wage countries is at least one contributing factor. As Wood observes, 'the debate is now over the magnitude of the effects, with their direction- adverse to unskilled workers -being largely agreed' (1997:33). 7. According to Edwards (2000:6), import penetration reduced employment by 9.1 per cent in ultra-labour intensive industries, and by 2.4 per cent overall between 1993 and 1997. However, once the position impact of export expansion on employment (in export and related industries) is taken into account, he estimates that the net impact of international trade on employment was to raise employment by one percent over the period (2000:13). Fedderke et al (2000) also found that the net measurable impact of trade was to increase employment rather than reduce it. These results are in line with most decomposition analyses in the international literature which estimate the impact of international trade on employment to be small relative to that of technology (see overview in OECD 1997:112-119). However, one's choice of model and decomposition technique profoundly influences the results. Furthermore, where firms undertake 'defensive innovation' in the face of increased international competition, it is impossible to disentangle meaningfully the relative impacts of trade and technology. A recent ILO report points to the joint impact of trade and technology in South Africa by attributing employment losses to 'a process of rationalisation or downsizing which might occur as a reaction to increased international competition'(1999:21). 8. Labour productivity refers to the ratio of employment to total output. For convenience, labour productivity is simply referred to as 'productivity'. 18 A critical review of SA's growth strategy 9. This argument was made by Sam Bowles in the Labour Market Commission, but to no effect. 10. Michael Porter's The Competitive Advantage of Nations (1990) is a popular bible for this set of beliefs. Monitor, an international consultancy which operates in a Porter-framework, has had a lot of influence on the formulation of industrial policy thinking in South Africa (Haines 1996:14-16). 11. See the 1993 Manufacturing Survey (Report no 30-01-01 1993). 12. Survey evidence indicates that the average party firm is between two and four times larger than non-party firms (Boccara and Moll 1997, Du Toit et al 1995). In the largest national bargaining council (for iron, steel, engineering and metallurgical industries), less than one third of the firms (employing 65 per cent of the workers) set wages for the entire industry (Standing et al 1996:143). 13. According to estimates by Samson et al (2000), it would cost R52 billion to provide every person in South Africa with a grant of Rl 00 a month. This comes to about 21 per cent of total government spending - ie well over twice the amount currently spent on welfare. They argue that R24 billion can be reclaimed from higher earners through the income tax system through adjustments to marginal tax rates and income thresholds. This leaves the net cost of transfer at R28 billion to be financed through further increases in taxation. Given that South Africa's top income earners are already highly burdened, it is thus likely that significant and sustainable increases in tax revenues could only come about through either adding to the income tax burden of households in, especially, the seventh and eighth income deciles (ie the labour movement's chief constituency) or increasing value-added tax. If the burden falls mainly on value-added tax, then the redistributive impact of the basic income grant will be reduced, as the poor will be paying for part of it every time they spend it on goods and services. At present both the trade union movement and the Democratic Party (representing most of the rich) profess support for a basic income grant, but both seem to anticipate that their constituencies will not shoulder the financial burden. References Amsden, A (1980) Asia's Next Giant: South Korea and late industrialisation. New York: Oxford University Press. Barker, F (1999) 'On South Africa's labour policies', South African Journal of Economics 67(1). Barro R (1991) 'Economic growth in a cross-section of countries', Quarterly Journal of Economics (May). Bell, T (1997) 'South African regional industrial development policy: critical issues', Transformation 32. 19 Nicoli Nattrass Bell, T and N Cattaneo (1997) Foreign Trade and Employment in South African Manufacturing Industry. Occasional Report no 4, Employment and Training Department, International Labour Office, Geneva. Bhorat, H and J Hodge (1999) 'Decomposing shifts in labour demand in South Africa', South African Journal of Economics 67(3). Birdsall, N and F Jaspersen (eds) (1997) Pathways to Growth: comparing East Asia and Latin America. Washington, DC: Inter-American Development Bank. Birdsall, N, D Ross and R Sabot (1997) 'Education, growth and inequality', in N Birdsall and F Jaspersen (eds) Pathways to Growth: comparing East Asia and Latin America. Washington, DC: Inter-American Development Bank. Boccara, B and P Moll (1997) 'Labour market flexibility and industrial councils in South Africa'. Unpublished paper, World Bank, (May). Chang, H (1998) 'Korea: the misunderstood crisis', World Development 26(8). Chirinko, R (1993) 'Business fixed investment spending: modeling strategies, empirical results and policy implications', Journal ofEconomic Literature 31. Dirwan, I and M Walton (1997) 'How international exchange, technology and institutions affect workers: an introduction', World Bank Economic Review 11(1). Department of Trade and Industry (DTI) (1995) Support Measures for the Enhancement ofthe International Competitiveness of South Africa's Industrial Sector. Submission by Government to the Trade and Industry Chamber of Nedlac, (November). (1998) Industrial Policy and Programmes in South Africa: Discussion Document, (April). Du Toit, D, S Godfrey, M Goldberg, J Maree and J Theron (1995) Industrial Councils and Small Enterprises: an investigation into the operation of industrial councils with particular reference to their accommodation of small enterprises. Cape Town: University of Cape Town, Sociology Department. Edwards, L (2000) 'Globalisation and the skill bias of occupational employment in South Africa'. Unpublished paper, School of Economics, University of Cape Town. Evans, P (1995) Embedded Autonomy: states and industrial transformation. Princeton, New Jersey: Princeton University Press. Fedderke, J, Y Shin and P Vaze (2000) Trade and Labour Usage: an examination of the South African manufacturing industry, ERSA Working Paper no 15. Department of Economics, University of the Witwatersrand, Johannesburg. Galenson, W (1992) Labor and Economic Growth in Five Asian Countries: South 20 A critical review of SA's growth strategy Korea, Malaysia, Taiwan and the Philippines. New York: Praeger. Gavin, M, R Hausmann and E Talvi (1997) 'Savings, growth and macroeconomic vulnerability', in N Birdsall and F Jaspersen (eds) Pathways to Growth: comparing East Asia and Latin America. Washington, DC: Inter-American Development Bank. Gibson, B and D van Seventer (1995) 'Restructuring public sector expenditure in the South African economy'. Unpublished paper, Development Bank of Southern Africa. Haines, R(1996) 'Industrial policy in South Africa: acritique'. Paper presented to the South African Sociological Association Congress, University of Natal, Durban. Henley, A and E Tsakalotos (1993) Corporatism and Economic Performance: a comparative analysis of market economies. Vermont: Edward Elgar. Internation Labour Organisation (ILO) (1999) (drafted by S Hayter, G Reinecke and R Torres). South Africa: studies on the social dimensions of globalisation. Final Draft, February 1999, Task Force on Country Studies on Globalisation, International Labour Office, Geneva. Jaspersen, F (1997) 'Growth of the Latin American and East Asian economies', in N Birdsall and F Jaspersen (eds) Pathways to Growth: comparing East Asia and Latin America. Washington, DC: Inter-American Development Bank. Joffe, A, D Kaplan, R Kaplinsky and D Lewis (1995) Improving Manufacturing Performance in South Africa: report of the industrial strategy project. Cape Town: University of Cape Town Press. Jomo, K (2000) 'Globalisation, liberalisation, poverty and income inequality in South East Asia'. Paper presented to the Conference on Poverty and Income Inequality in Developing Countries hosted by the OECD Development Centre, in Marseille (November). Kaplinsky, R (1995) 'Capital intensity in South African manufacturing and unemployment, 1972-90', World Development 23(2). Kenny, B and E Webster (1998) 'Erodingthecore: flexibility and the resegmentation of the South African labour market', Critical Sociology 24(3). Lee, E (1995) 'Globalisation and employment: is anxiety justified?', International Labour Review 35(5). Leibbrandt, M, H Bhorat and I Woolard (2000) 'Understanding contemporary household inequality in South Africa', Studies in Economics and Econometrics 24(3). Lindbeck, A (1997) "The Swedish experiment', Journal of'Economic Literature 35 (September). 21 Nicoli Nattrass LMC (Labour Market Commission) (1996) Restructuring the South African Labour Market. Report of the Presidential Commission of Inquiry into Labour Market Policy, Cape Town. McCarthy, C (1998) 'South African trade and industrial policy in a regional context', in L Petersson (ed) Post-Apartheid Southern Africa: economic challenges and policies for the future. London: Routledge Studies in Development Economics. Moll, P (1996) 'Compulsory centralisation of collective bargaining in South Africa', American Economic Review: papers and proceedings 82(2). Moll, T(1993) 'Once a dog, always a dog?: the economic performance of modern South Africa'. Paper prepared for the EBM Research Conference '94, Rand Afrikaans University, Johannesburg, (November). Morley, S (1995) Poverty and Inequality in Latin America: the impact of adjustment and recovery in the 1980s, Baltimore and London: Johns Hopkins University Press. (2000) 'Distribution and growth in Latin America in an era of structural reform'. Paper presented to the Conference on Poverty and Income Inequality in Developing Countries: A Policy Dialogue on the Effects of Globalisation, hosted by the OECD Development Centre, Paris. Nattrass, N (1996) 'Gambling on investment: competing economic strategies in South Africa', Transformation 31. (1998) 'Growth, employment and economic policy in South Africa: a critical review'. Paper published by the Centre for Development and Enterprise, (September). (2000a) 'Wage-strategies and minimum wages in decentralised regions: the case of the clothing industry in Phuthaditjhaba, South Africa', International Journal of Urban and Regional Research, (December). •— (2000b) 'Inequality, unemployment and wage-setting institutions in South Africa', Studies in Economics and Econometrics 24(3). Nattrass, N and J Seekings (1997) 'Citizenship and welfare in South Africa: deracialisation and inequality in a labour-surplus economy', Canadian Journal of African Studies 31 (3). (1998) 'Democratic institutions and development in post-apartheid South Africa", in M Robinson and G White (eds) The Democratic Developmental State: political and institutional design. Oxford: OUP Studies in Democratisation. National Institute for Economic Policy (NIEP) (1996) From the RDP to GEAR: the gradual embracing of neo-liberalism in economic policy. Johannesburg: National Institute for Economic Policy. 22 A critical review of SA's growth strategy NEDLAC Labour Caucus (NLC) (1996) Social Equity and Job Creation. Johannesburg: NLC. Page, J (1997) 'The East Asian miracle and the Latin American consensus: can the twain ever meet?' in N Birdsall and F Jaspersen (eds) Pathways to Growth: comparing East Asia and Latin America. Washington, DC: Inter-American Development Bank. OECD (1997) 'Trade, earnings and employment: assessing the impact of trade with emerging economies on OECD labour markets', OECD Employment Outlook (July). (1999) 'Labour market performance and the OECD jobs strategy', Economic Outlook 65 (June). Porter, M (1990) The Competitive Advantage of Nations. New York: The Free Press. Ramirez, A, G Ranis and F Stewart (1997) Economic Growth and Human Development. Centre Discussion Paper No.787, Economic Growth Centre, Yale University. Samson, M (1997) Fiscal Policyfor Job Creating Growth: a NIEP report prepared for COSATU. Johannesburg: National Institute for Economic Policy. Samson, M, O Babson, K MacQuene, I Van Niekerk and R van Niekerk (2000) 'The macroeconomic implications of poverty-reducing income transfers'. Paper presented to the conference: Towards a Sustainable and Comprehensive Social Security System, hosted by the Institute for Social Development, UWC. Schultz, T P and G Mwabu (1998) 'Labour unions and the distribution of wages and employment in South Africa', Industrial and Labour Relations Review 51(4). Seekings, J (2000) 'Visions of society: peasants workers, and the unemployed in a changing South Africa', Studies in Economics and Econometrics 24(3). Sen, A (1997) 'Inequality, unemployment and contemporary Europe', International Labour Review 136(2). Sharp, J and A Spiegel (1996) Report on an Evaluation of the Regional Industrial Development Programme and Simplified Regional Industrial Development Programme in the Free State IDPsofBotshabelo/Selosesha and Phuthaditjhaba/ Industriqua. Social Analysis Network, Department of Social Anthropology, University of Cape Town. Simbi, T and M Aliber (2000) 'The agricultural employment crisis in South Africa'. Paper presented at the TIPS Policy Forum, Muldersdrift, September. South Africa (1996) Growth, Employment and Redistribution: a macroeconomic strategy. Department of Finance, South Africa. 23 Nicoli Nattrass (1998) 'Creating jobs, fighting poverty: an employment strategy framework', Department of Labour, published in Government Gazette, 397(19040), July 3. Stallings, B (2000) 'Growth, employment and equity: the impact of economic reforms in Latin America and the Caribbean'. Paper presented to the Conference on Poverty and Income Inequality in Developing Countries hosted by the OECD Development Centre, in Marseille, November. Standing, G (1997) 'Labour market dynamics in South African industrial firms: the South African labour flexibility survey'. Paper presented to the Workshop on Labour Markets and Enterprise Performance in South Africa, Johannesburg, January 30. Standing, G, J Sender and J Weeks (1996) The South African Challenge: restructuring the South African labour market. Geneva: International Labour Organisation. Stiglitz, J (1998) 'More instruments and broader goals: moving towards the post- Washington consensus'. WIDER Annual Lectures 2, UNU World Institute for Development Economics Research, Helsinki. Wade, R (1990) Governing the Market: economic theory and the role of government in East Asian industrialisation. Princeton: Princeton University Press. Weeks, J (1999) 'Stuck in low GEAR? macroeconomic policy in South Africa, 1996-98', Cambridge Journal of Economics 23(4). Weiss, L (1998) The Myth of the Powerless State: governing the economy in a global era. Cambridge, UK: Polity Press. White, G (ed) (1988) Development States in East Asia. London: Macmillan. Wood, A (1994) North South Trade, Employment and Inequality. Oxford: Oxford University Press. (1995) 'How trade hurt unskilled workers', Journal of Economic Perspectives 9(3). (1997) 'Openness and wage inequality in developing countries: the Latin American challenge to East Asian conventional wisdom', World Bank Economic Review 11(1) (January). World Bank(1993) The East Asian Miracle: economic growth and public policy. New York: Oxford University Press. 24