Jump to ContentJump to Main Navigation
The Practice of Industrial PolicyGovernment—Business Coordination in Africa and East Asia$

John Page and Finn Tarp

Print publication date: 2017

Print ISBN-13: 9780198796954

Published to Oxford Scholarship Online: April 2017

DOI: 10.1093/acprof:oso/9780198796954.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (oxford.universitypressscholarship.com). (c) Copyright Oxford University Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 18 January 2022

Policy Coordination and Growth Traps in a Middle-Income Country Setting

Policy Coordination and Growth Traps in a Middle-Income Country Setting

The Case of South Africa

(p.211) 11 Policy Coordination and Growth Traps in a Middle-Income Country Setting
The Practice of Industrial Policy

Haroon Bhorat

Aalia Cassim

Alan Hirsch

Oxford University Press

Abstract and Keywords

South Africa has exhibited tepid economic growth over the past twenty years, as well as high levels of income inequality characteristic of a middle-income country growth trap. This chapter compares and contrasts South Africa’s growth trap relative to middle-income peer economies. In addition, it studies the policies and structures of the South African economy that have perpetuated the persistently low levels of growth observed. In particular, it considers the capital-intensive nature of manufacturing, regulation in the telecommunication and transport sector, and the inadequacies of Black Economic Empowerment (BEE) policies. The chapter concludes discussing the welfare outcomes on the vast majority of South Africans who are unable to participate in the economy.

Keywords:   South Africa, economic growth, growth trap, income inequality, middle-income country, capital-intensive manufacturing

11.1 Background

On a number of indicators and measures, South Africa’s post-apartheid performance has been both a significant improvement on the past, as well as representing a noticeable Pareto improvement in the welfare of its citizens. The economy witnessed the longest period of positive uninterrupted growth in real gross domestic product (GDP) since the 1960s. Welfare gains are apparent in access to social services, housing and infrastructure services, and in a significant reduction to extreme poverty.

Compared with many of its emerging market peers however, economic growth has been pedestrian. In terms of poverty, inequality, and unemployment the numbers are not encouraging. Using the national poverty line of US$43 per month (in current prices), 47 per cent of South Africans remain poor. In 1994, this figure was 45.6 per cent. More jarring, South Africa’s unemployment rate is an eye-watering 25 per cent, whilst the Gini coefficient at 0.69 renders the country one of the most unequal in the world (Bhorat and Tseng 2014).

Industrial policy was on the agenda of the African National Congress (the ruling party after 1994) and its allies before and during the transition to democracy, through the well-conceived Industrial Strategy Project for example (Joffe et al. 1995). However, the successful implementation of industrial policy is the exception rather than the rule in democratic South Africa. Apartheid South Africa built several industries, such as steel, automotive (p.212) assembly, and petrochemicals, through decisive interventions, and left the legacy of a well-endowed Industrial Development Corporation—a targeted development bank. Several factors have undermined the effectiveness of contemporary industrial policy initiatives; including poor coordination within government, and fractious relationships between government and the key social partners, business, and labour.

Persistently high levels of inequality are due, in part, to a number of policy coordination failures that have entrenched the deeply unequal structures of the pre-democracy era. Low levels of diversification in productive sectors, inadequate human capital accumulation, and inefficient state-run utilities have not promoted inclusive economic growth. In addition, an unintended consequence of the development of the South African economy has been to facilitate rent-seeking between key players in the economy. The structure of the economy has led to the outcomes observed today. This chapter examines South Africa’s economic trajectory and key policy decisions that have facilitated the country’s low growth trap.

11.2 South African Economic Overview

The South African economy grew at 3.2 per cent a year on average from 1994 to 2012. Potential growth is currently thought to be around 3.5 per cent, though it was estimated at around 4.5 per cent during the four-year period from 2004 to 2007, when growth averaged around 5.5 per cent. Economic growth however, has mainly been driven by domestic demand and financed through a persistent current account deficit. The current account balance was close to zero around 2003 but has subsequently increased, and regularly hovered at around 6 per cent of GDP.

11.2.1 The Structure of the Economy and Employment Growth

In post-apartheid South Africa, economic growth has been uneven: in contrast to the relatively dynamic sectors,1 such as financial services, transport, and communication, which grew annually at around 5 per cent between 2001 and 2012, sectors such as agriculture and manufacturing tended to grow slowly at around 2 per cent annually in the same period. In real terms, agricultural gross (p.213) value added rose by 1.4 per cent a year from 1994 to 2012, manufacturing by 2.7 per, and general government services by 1.9 per cent a year. As a result, the financial sector’s contribution to GDP was 5 percentage points higher than manufacturing in 2012. Furthermore, whilst primary sectors and manufacturing saw a decline in their GDP contribution in 2012, trade and transport were amongst the sectors that increased their share compared to 1994.2

Policy Coordination and Growth Traps in a Middle-Income Country SettingThe Case of South Africa

Figure 11.1. Gross value added and employment growth, by sector, 2001–12

Source: South African Reserve Bank (2014) and StatsSA (LFS 2001 and QLFS 2012), authors’ calculations.

Figure 11.1 explores the interaction between GDP and employment growth by sector between 2001 and 2012. Each of the bubbles represents a sector, while the size of the bubbles indicates the relative size of employment in that sector in the base year: 2001. The coordinates for the centre of each bubble are the relevant sector’s employment and gross value-added growth for the period. The 45-degree line divides the figure into two sections: bubbles below the line show sectors in which employment growth was lower than gross value-added growth, while those above the line show sectors in which employment growth exceeded output growth.

Figure 11.1 shows that primary sectors faired particularly badly in the period: output growth was negative for mining (−0.3 per cent) and lowest among positive growth sectors for agriculture (2.2 per cent). Furthermore, these are the only two sectors that experienced a contraction in employment in the period, with employment growth in agriculture and mining falling by 5.1 per cent and 4.1 per cent, respectively. The discrepancy between output and employment was the highest in the two primary sectors.

The tertiary sectors achieved relatively high output growth. Specifically, gross value-added growth for the finance and community services sectors stood at 5.4 and 3.1 per cent for the period, while employment growth was 5.3 and 3.0 per cent, respectively. These two tertiary sectors thus experienced labour-neutral growth, while all other sectors saw output growth that was faster than employment growth.

Among the secondary sectors, output growth in construction was high at 7.2 per cent, but employment growth in this sector was much lower, at 4.7 per cent. The construction boom can be attributed to, among other factors, infrastructure projects related to the 2010 FIFA World Cup, construction of the Gautrain rapid-rail system, and several other public and private sector investment initiatives including those undertaken by Eskom and Transnet (Hanival and Maia 2008). In contrast, neither the manufacturing nor utilities sectors saw a significant increase in employment over the period.

Post-apartheid South Africa has delivered an economy characterized by an eroding primary sector and an un-dynamic and un-diverse manufacturing sector. Yet the modest growth levels since 1994 have been marked by a rise (p.214) in financial and business services, and wholesale and retail trade. In short, this post-apartheid growth path has been built around South Africa’s sophisticated and globally competitive financial sector and its consumer-driven domestic aggregate demand. The weakness of mining, agriculture, and in particular manufacturing, endangers the longer-term ability of the South African economy to grow, generate employment, and improve living standards (Rodrik 2013). In Section 11.2.2, we assess whether the outcomes of the structure of the economy are unique to South Africa, or are a broader middle-income country experience.

11.2.2 Middle-Income Country Growth Traps: The Case of South Africa

Freed of the laws of apartheid for over twenty years, the new South Africa has existed in a global era dominated by fast-growing emerging markets such as China and India. These economies have regularly recorded growth rates in excess of 6 per cent over a sustained number of years. South Africa, despite its membership of the G20 and the BRICS (Brazil, Russia, India, China, and South Africa) group of economies, remains mired in a cycle of low, single-digit GDP growth. This experience can be located within a broader phenomenon first described by Gill and Kharas (2008) and Kharas and Kohli (2011), as representative of a middle-income country growth trap.3

So why, when South Africa shares so much in common with rapidly growing emerging markets, and has one of the largest and most diversified sub- (p.215) Saharan African economies, has the country failed to converge (Bhorat and Hirsch 2014)?

11.2.3 The Anatomy of a Growth Trap: A Cross-Country Comparison

Historically, a common trend for middle-income countries is that, as imported technology is employed, labour is switched from low productivity to high productivity sectors, increasing growth gains resulting in a rise in per capita GDP. The technologies are usually employed in labour-intensive sectors, absorbing the pool of underemployed or unemployed labour. Productivity growth from technological catch-up is then exhausted, wages rise, and labour-intensive exports become more expensive and less competitive in international markets. At this point growth no longer occurs from shifting workers from a low productivity sector to a high productivity sector, such as from agriculture to manufacturing, and thus the gains for importing foreign technology diminish (Eichengreen, Donghyun, and Kwanho 2011; World Bank 2011; Agénor, Canuto, and Jelenic 2012). Higher levels of growth then require exploitation of economies of scale through specialization and innovation, allowing a country to move up the value chain and integrate into international trade of goods, money, and ideas (Gill and Kharas 2008).

The middle-income country ‘growth trap’ is often illustrated by a cross-country comparison of GDP per capita over time. Many low income countries have moved up to middle-income country status by exploiting low labour costs. However, at some point, surplus labour is exhausted and wages start to increase (OECD 2014). Regulation, politics, and firm structure are amongst the factors that essentially block innovations required to sustain economic growth or to develop a better-skilled labour force that can produce higher value-added products, taking labour on to a better wage (OECD 2014). China is a key example of an innovator that managed to sustain high levels of growth and productivity, and as a result GDP per capita increased by 8 per cent annually between 1990 and 2013. Whilst not as high, India and Malaysia yielded annual GDP per capita growth of 5 and 4 per cent respectively, during the same period. On the lower end, Turkey, Brazil, and the Philippines yielded around 2 per cent annual increases of GDP per capita, whilst South Africa held the lowest increase in per capita GDP with a mere 1 per cent annual increase during the period.4

South Africa, Brazil, the Philippines, and Turkey would be viewed tentatively as representing the sample of emerging markets, which, over a period of thirteen years, have been unable to significantly increase per capita growth (p.216) rates. This may represent the underpinnings of a growth trap in these emerging economies. The lower growth path depicted for South Africa, can in part, be explained by policies of the past. The apartheid government attempted to exploit cheap labour for mining and agriculture for far longer than it actually served as a productive engine of growth. The model failed since the 1960s when the growth in agriculture and mining slowed (Levy, Hirsch, and Woolard 2014). The vast mass of the labour force was not moved to more productive sectors where they could be up-skilled. The entrenched racial politics of the time blocked the adjustment the economy needed for a better-educated and trained workforce, in the pursuit of higher growth levels.

Economic growth outside of the declining mining and agricultural sectors was insufficient to absorb the wave of new entrants into the labour market since the mid-1990s, and those who lost jobs in mining and agriculture. This resulted in unemployment rates persistently above 20 per cent.5 Structurally, the economy has grown in sectors such as the financial and services sector, demanding high skilled labour—not in primary sector employment, generally geared towards lower skilled workers. Low-skilled work seekers, however, make up the vast majority of the unemployed.

From the sample of upper middle-income countries mentioned, South Africa’s unemployment rate is extraordinarily high. This is even true amongst those economies that have experienced similarly low levels of growth and investment, such as Brazil and Turkey. Unlike these countries however, South Africa’s informal sector is unusually small and unable to absorb low-skilled workers, further exacerbating unemployment levels. Investment, Savings, and Economic Growth

Fixed investment and savings rates in South Africa are well below those of many other emerging markets. Between 1990 and 2012, countries experiencing consistently low levels of growth such as the Philippines, Brazil, and Turkey, had levels of investment and savings similar to South Africa. On the other hand, savings and investment in China, Malaysia, India, and Indonesia have experienced average investment levels at least 1.5 times greater than South Africa.6

Low investment rates are usually a result of low real returns to investment, but evidence suggests that real returns to capital in South Africa are highly favourable (World Bank 2011). Real returns averaged around 15 per cent between the 1994 and 2008 period, whilst nominal returns were 22 per cent in the 2005–8 period—which was the same as China, albeit over a longer period. Returns have been substantially higher than the prime lending rate, (p.217) which is particularly surprising given the modest growth experienced by the South African economy. However, investment does not respond to returns in the expected way. Domestic and foreign investors in South Africa seem inelastic with respect to the return on capital, but partially elastic to other factors such as perceived political risks, structural impediments, and low national savings rates. Structural barriers include low levels of industrial competition because of concentrated industries with high barriers to entry; and volatile labour relations that are essentially a tax on investment and the scarcity of skills.

Government and public enterprise expenditure tends to be fairly low, and even in the World Cup boom period did not reach levels comparable with public investment in the boom of the 1960s and 1970s. Private investment has made up between 65 and 75 per cent of total investment, and has been facilitated through large-scale capital expenditure projects in mining, platinum, automotive, chemical manufacturing, retail, real estate, and tourism. Industries that attracted investment were invariably capital-intensive in nature, so reinforcing the economy’s growth trajectory.7 Overall gross fixed investment in recent years has been very low at around 19 per cent of GDP, though it is slightly better than during a long period between the 1980s and the mid-2000s when it hovered around 15 per cent.

The low savings rate, relative to comparator countries, has meant that financing investment has necessitated a fairly large current account deficit. Financing of the deficit has largely been dependent on portfolio investment rather than foreign direct investment (FDI), the latter being the less volatile capital flow option for emerging markets.

The dependence on short- to medium-term capital inflows tends to perpetuate dependence on the resource sector, processers of resources, and powerful, publically quoted oligopolies in the services sector. The market power of these companies produces the generous margins that portfolio investors seek (Bhorat et al. 2014). Is South Africa Competing in a Global Value Chain?

South African export performance has been poor, even in a period of relatively high commodity prices. Exports grew by just 3.5 per cent per annum over the period 1990–2012, less than half the rate of Malaysia, which reflects not only its poor response to the commodity boom due to regulatory and infrastructure complexities and deficiencies, but also the poor degree of integration of South Africa into global value chains.8

(p.218) Extensions to trade theory hold that international competitiveness is a function of micro-level innovations in technology and increasing technological sophistication (Dosi and Soete 1983). This suggests that the method of production and the basket of exported goods are critical to growth, development and international competitiveness (Gallagher, Moreno-Brid, and Porzecanski 2008).

A number of East Asian countries performed well in this respect, taking advantage of lower transport costs as well as international trade barriers (Agénor, Canuto, and Jelenic 2012). The Asian region also benefits from regional integration that had been established to a lesser extent in Latin America and Africa. China, for example, is currently one of the single largest high technology exporters in the world, with average annual growth of 7 per cent between 1993 and 2012, as well as growing exports in high skilled services. Malaysia and the Philippines have, however, seen a decline in high technology exports as China becomes more competitive. Amongst its middle-income peers, South Africa relies relatively more than others on exports of ores and metals, insurance, and financial services, but falls short in innovative or value-added exports.9

After the advent of democracy in 1994, South Africa re-entered the global economy through a rapid process of trade liberalization. While imports and exports increased relatively sharply, South Africa lagged behind its emerging market peers who had better production methods and a more diversified basket of goods. Though sanctions had encouraged domestic diversification of production in South Africa, management, skill, and technology levels were not comparable to developing country peers. In addition, liberalization led to greater import penetration than export diversification, though in the earlier years of liberalization it appeared that South African producers were able to respond to opportunities. South African competitiveness was also hampered most notably by labour market constraints and the volatile exchange rate. In addition, investment has been biased towards processing and heavy manufacturing as opposed to light manufacturing industries. As a result, the composition of exports is still made up of a large share of commodity type exports (ores and metal exports). Furthermore, manufactured exports still rely heavily on primary commodity inputs. As such, South Africa’s export profile continues to be capital-intensive in nature and driven by natural resources (Bhorat et al. 2014). The only evident and significant diversification in exports other than the heavily subsidized automobile sector is through insurance and financial services (4 per cent average annual growth), and tourism.

(p.219) 11.2.4 Human Capital Accumulation

It is well known that human capital accumulation will raise earnings levels, both in terms of private and social returns. This suggests that educational attainment has a key role to play in reducing income inequality through improved labour market opportunities. The literature on what differentiates high growth economies from stagnating ones notes that it is often the accumulation of technical knowledge through high-quality education, thus allowing a society to assimilate foreign technology and improve levels of productivity (Nelson and Phelps 1966; Abramovitz 1986; Engelbrecht 1997; Gill and Kharas 2008).

An outcome of the apartheid government’s policy was first a highly unequal schooling system, and second, a tertiary education system that was not accessible to those with poor levels of schooling. Data on school scores in 2012 suggest that firstly, in every grade and for every subject other than Grade 3 Language, pupils in the South Africa schooling are, on average, failing standardized tests (Presidency 2012). Second, a pupil’s progress through the schooling system falls, as shown by declining average pass rates.10

Unsurprisingly then, when examining cross-country results for standardized mathematics and reading scores, South Africa falls below a number of African countries including Tanzania, Swaziland, Kenya, Botswana, and Zimbabwe, as well as below the global average scores, in both subjects.11

Using a standard Cobb–Douglas production function approach, we find that amongst all education cohorts (schooling, schooling with matric, vocational training, and university) a statistically significant impact on economic growth returns is only derived for the employed with university degrees (Bhorat, Cassim, and Tseng 2014). Whilst the number of enrolments into tertiary institutions has more than doubled since 1994, the proportion of graduates from the population of high school leavers varies between 15 and 17 per cent. Of those enrolled, only between 4 and 6 per cent graduated with a Science, Engineering, and Technology degree.12 Whilst Brazil had a similar proportion of engineering graduates to South Africa, Turkey had just over double the proportion of graduates, but Malaysia had just under five times the proportion of graduates in this field. In comparison, 40 per cent of the students in China are enrolled in mathematics, science, and engineering fields, constituting around 6.7 times the proportion of graduates that South Africa has in this field. Needless to say, significant investment in technical human capital accumulation is essential for developing a domestic absorption capability of global knowledge (Yao et al. 2008).

(p.220) This lack of technical knowledge accumulation has, in part, prevented the diversification of exports and the absorption of the labour in the pursuit of a more diversified growth path. Ultimately though, the quantity and quality of human capital held by individuals remains poorly matched to meet the conditions required for this more diversified growth path. This structural mismatch may be a key constraint in the economy’s attempt to break out of its current low growth trap.

The data presented thus far clearly illustrates that South Africa falls below peer emerging economies in terms of key economic variables such as investment, savings, trade volume and diversification, and human capital accumulation. In Section 11.3 we consider the interaction between policy coordination and key economic players that has led to certain industrial biases that may have inhibited inclusive economic growth and perpetuated unemployment.

11.3 Policy Coordination and Economic Growth Traps in South Africa

In the early 1990s, before and during the transition to democracy, engagement between business, labour and the old and new governments was extraordinary in its depth and breadth. The last minister of finance in the old regime was also briefly the first minister of finance in the democratic government, and in 1992, he sanctioned the establishment of a National Economic Forum where the three social partners gathered, informed by high quality research by the Industrial Development Corporation and other government agencies (Hirsch 2005).

One of the first laws to pass through the democratic parliament led to the establishment of a tripartite National Economic Development and Labour Council of South Africa (NEDLAC), designed to negotiate labour laws and to coordinate potentially binding agreements on social and economic policies. Labour, business, and government met, considered challenges, jointly commissioned policy research, and agreed to recommendations on the basis of the findings. This generally led to good faith action by government and the other social partners.

Nevertheless, signs of poor economic policy coordination emerged as early as 1996, two years after democracy, when three different and conflicting economic policy frameworks were developed and published by different parts of government: the Reconstruction and Development Planning (RDP) office in the Presidency produced the National Growth Path Framework; the Labour Department’s Presidential Labour Market Commission produced an approach (p.221) towards a social plan; and the Treasury produced the influential Growth, Employment and Redistribution (GEAR) strategy.

GEAR included a commitment to the National Treasury offering to convene a process of economic policy coordination within government. However, coordination hosted by Treasury lasted a few months before it was abandoned. From this uncertain point of departure, economic coordination proceeded on an erratic downward path.

While the World Trade Organization (WTO)-linked trade liberalization programme forged steadily ahead on the principle of reducing oligopoly power to encourage new investment, the policy around network industries seemed to move in the opposite direction. Electricity provision remained a monopoly of Eskom—the very large state-owned corporation—and while the national legacy fixed-line telecommunications provider Telkom was partially privatized, its monopoly over fixed line provision was extended by the state to a five-year period and beyond, in the misguided belief that it would follow through on its commitment to the extension of services to unserved customers in poor rural areas, as well as to raise the privatization sale value of the company. And while the export strategy of GEAR rested explicitly on a competitive exchange rate resting on an accommodative monetary policy, the Reserve Bank followed its own course and raised interest rates sharply in two volatile episodes.

11.3.1 Industrial Policy

In the area of industrial policy, the Department of Trade and Industry (DTI) pushed a sector, or cluster-based, development strategy which would combine various incentives. But the National Treasury opposed what it called ‘picking winners’, and blocked programme implementation, weakening the already shaky relationship between domestic producers and the government.

When a targeted tax holiday to encourage investment was introduced following the adoption of the GEAR policy in 1996, its design was compromised by the inability of its architects to fend off entrenched interests in the labour and business communities. This rendered it difficult for anyone to benefit from the tax holiday, and it was soon withdrawn.

Historically, industrial policy incentives introduced under apartheid, such as the Regional Industrial Development Plan (RIDP) in 1960 and the Simplified Regional Industrial Development Plan (SRIDP) in 1993, were essentially spatial tools to create industrial zones away from city centres, to restrict migration of black people into urban areas. RIDP support was biased in favour of labour-intensive sectors and evidence suggests that increased investment incentives in the 1990s and early 2000s led to increased employment (Kaplan 2003). However, given the priorities of the incentives, there was little impact on skills development, technology upgrades, or attracting foreign investment. (p.222) The Small and Medium Manufacturing Development Programme (SMMDP) was introduced in 1998 and was biased towards capital intensive industry, despite the fact that the SMMDP was designed ‘to encourage new investments in small and medium sized manufacturing companies…’ (Kaplan 2003). The Spatial Development Initiative was also introduced, targeting smaller projects that would be managed within a unit of the DTI.

Several studies have shown that although the subsidies tended to favour the formal sector and some had a bias towards capital investment, the industrial incentives introduced by the DTI in the post-1994 period were reasonably well targeted and appeared to have some impact (Rustomjee 2006; World Bank 2006). However, the 1994–2005/6 period certainly saw a decline in the overall value of the budget for DTI programmes (Rustomjee and Hanival 2008). What the same studies tend to point to though, was the lack of coordination between DTI programmes and the orientation of other influential government departments. For example, incentives for small businesses were rendered relatively impotent by the very slow movement towards improving the regulatory environment for small businesses. Indeed, a series of reports prepared for the Presidency in 2005 showed how municipal laws and labour laws, amongst others, stood in the way of the expansion of the small business sector. The Mineral-Energy Complex

What Fine and Rustomjee (1996) named ‘the mineral-energy complex’ (MEC), elicits significant government support and attention. In most cases government has sought, in the words of one senior DTI official, ‘quick wins’—such as the Maputo Corridor and other large projects. These projects invariably serve to support the mining industry and are capital-intensive.

Historically, economic policy interests were aligned to the MEC. In recent decades, macroeconomic stability was biased towards ensuring suitable exchange rates for commodities exports such as gold and minerals; and state-owned enterprises (SOEs)—largely the minerals and energy complex (Fine and Rustomjee 1996; Takala-Greenish 2008). Capital within the MEC was also highly concentrated and strongly linked to the financial sector, which maintained control and economic power, and MEC owners and managers had a huge influence on policy.

11.4 Big Business, Big Government, and Big Unions: A Policy Coordination Paradigm in South Africa?

An unintended consequence of the structural development of the South African economy has not only been to perpetuate low levels of growth, but (p.223) also to facilitate rent-seeking between key players in the economy. Empirical evidence suggests a link between rent-seeking, poor economic performance, and high levels of income inequality (Olson 1971; Chakraborty and Norris 2005). Rent-seeking activities are often driven by the relatively elite or wealthy in a society through social pacts, who choose rent-seeking over engaging in productive activities. In this section we assess the role of the political economy, policy coordination, institutional factors, and market structure, in reinforcing rent-seeking and poor economic performance.

After the establishment of NEDLAC in 1994, it seemed that the foundations were being put down for a system of social pacts that would overcome the divisions within South Africa’s economic society. Policy coordination would ostensibly be pursued through this tripartite structure—much in the spirit of the country’s political negotiations which shaped the peaceful transition to democracy. However, cooperation and trust between the three social partners weakened after the GEAR was adopted in mid-1996, and lost further credibility with the failure of a Jobs Summit in October 1998.

One consequence of the weakening of NEDLAC has been that the corporate sector and trade unions settled in an uneasy, but stable, political economy equilibrium defined by high margins, or rents, distributed between organized labour and big business. In particular, for sectors where the cost of not complying with big government and big labour was high, big business’ growth and development trajectory has been invariably shaped by the implicit contract between these three actors. Recent labour market disputes in the mining industry are strongly indicative of this growth narrative. It is also within this environment of a strong alliance with the union movement that the ruling party, whilst publically pushing for employment-friendly labour market policies, has found it difficult to materially counter trade union interests in terms of labour market reform.

The wage premia associated with militant trade unions have, to some extent, worked as a disincentive to employment creation. Overall mean monthly earnings, and specifically in agriculture, mining, manufacturing and for private households, grew at a faster rate than employment between 1997 and 2012.13 In manufacturing and mining in particular, employment declined steadily. What this suggests is that the distribution of rents for organized labour may have been counter to employment-generating policies.

Mahajan (2012) describes the firm owners, government, and organized labour as ‘locked in a continual, rambunctious public tussle over the distribution of the high rents being generated under the system’. This triad is constrained by the bounds of labour regulation, competition policy, tax policy (p.224) and equity considerations. These constraints have not always been productivity enhancing. This process has created a barrier to entry for new firms that could create a more competitive environment and normalize returns. Within this, the unemployed are left out, as are those who would like to see rents being transferred into higher productivity, higher investment, growth-enhancing actions. The outcome is an economy that performs below its potential. Some form of modification of this unintended agency triumvirate then, it is argued, may be essential for placing South Africa on a more inclusive growth path.14

Perhaps the overriding characteristic of government in South Africa today is fragmentation, particularly in economic policy formulation and implementation. So, while the DTI steadily drives forward the fifth iteration of its Industrial Policy Action Plan with nine cross-cutting interventions and over thirty sectoral strategies, this is generally not seen, or acted on, as if it were a strategy followed by all arms of government. Although political support for industrial policy is stronger than before, at the level of policy, the ability of government to implement and monitor programmes systematically is limited.

Government, despite its initiatives towards developmental partnerships, has not proved strong or effective enough to lead a process of breaking free from the path dependent pattern of low value-added exports, and oligopolistic market structures with high margins. Such a strong, efficient state is also essential for managing a change from a growth path financed by short-term capital flows and dependent on domestic consumption levels.

11.5 Testing the Paradigm

We hope here, to describe representative examples of policy outcomes or sectoral growth experiences which have in effect excluded the unemployed and those in the informal sector—thus generating this ongoing problem of a low-level growth trap for South Africa.

11.5.1 Capital Investment Has Not Favoured Labour-Intensive Sectors

The notion of investment bias towards MEC subsectors can be clearly seen when annual growth in capital is measured at a subsectoral level. There was an annual decline, between 1990 and 2012, of capital investment into labour-intensive sectors such as clothing (−4.9 per cent); electrical machinery and apparatus (−4.4 per cent); and textiles (−3.9 per cent) amongst others.15

(p.225) Those sub-sectors that have seen increasing annual capital investment within the period include medical and dental services (8.5 per cent); civil engineering and construction (6.5 per cent); coal mining (5.1 per cent); and motor vehicles (4.1 per cent). At an industry level, annual growth in capital investment in community services as well as finance and business services has in fact out-weighed investment growth in manufacturing, and particularly light manufacturing.

Policy Coordination and Growth Traps in a Middle-Income Country SettingThe Case of South Africa

Figure 11.2. Growth of employment and capital by sector, 1997–2012

Source: LFS and QLFS sources from Statistics South Africa (2014a, 2014b), Quantec (2014), and author’s calculations.

Figure 11.2 examines the interaction between capital and labour at an aggregate sectoral level between 1997 and 2012. The horizontal axis measures annual growth in capital intensity whilst the vertical access measures annual growth in employment.16 The interaction of capital intensity and employment first speaks to whether mechanization through capital deepening has been used to displace labour, and, second, the level of investment in capital formation associated with increasing labour in a particular sector.

Rodrik (2006) found that between 1980 and 2005 capital deepening resulted in a higher demand for skilled as opposed to low-skilled workers, which is amongst the reasons for the wage push in skill-intensive occupations. Sectors (p.226) such as construction, manufacturing, mining, and electricity have increased capital intensity at a rate greater than employment. Whilst mining increased capital intensity by just under 1 per cent per annum, gross value added declined for the sector. Construction increased annual capital formation by 7.54 per cent at a rate greater than employment (4.84 per cent) and gross value-added growth (7.2 per cent), indicating that generating both employment and value added in construction came with a cost of high levels of investment in capital deepening.

Manufacturing has seen higher levels of capital formation than growth in employment, hence reinforcing a capital intensive growth trajectory. Annual capital formation was around 3.15 per cent whilst employment growth was less than 1 per cent. This indicates increasing mechanization in sectors that may have previously been more labour-intensive, as well as increased investment in heavy manufacturing.

Where employment has grown it has followed a high-skilled labour demand trajectory such as that found in the finance and business and community services sector (Bhorat, Goga, and Stanwix 2013). Employment has exceeded growth in capital in finance (6.36 compared to 2.05 per cent), community services (3.15 compared to 2.44 per cent) and wholesale and retail (3.28 compared to 2.74 per cent). These sectors have therefore had lower levels of investment in the form of gross capital formation, yet increased employment suggesting that the cost of creating employment was lower. Driving these trends in wholesale and retail for example, was in part an increase in informal sector employment. The apparent labour intensity in finance and business services was driven by the rapid rise of temporary employment service providers—a relatively new phenomenon in South Africa.

Sectors that yield the lowest levels of capital accumulation, such as agriculture, increased employment by 1 per cent annually. This sector could have had the potential to absorb low-skilled labour yet is stagnating; in part because of poor levels of investment, and the imposition of the minimum wage (Bhorat, Kanbur, and Stanwix 2014).

Ultimately, capital investment in a number of South African main sectors has not yielded high growth returns for employment. Sectors which have gained employment on the back of capital investment have either been in the public sector CSP (Community, Social, and Public Services), reinforced the consumption-based growth trajectory of the economy (wholesale and retail trade), or indeed have been labour outsourcing sectors. In particular, in the growth engine of many fast-growing emerging markets in the world—manufacturing—investment is low and it remains unable to convert capital expenditure into sufficiently high levels of employment generation.

(p.227) 11.5.2 Transport Tariffs Have Not Facilitated Trade

The bulk of South African export and import trade is through its seaports,17 yet tariffs have been inhibitive of trade in certain sectors, whilst favouring others. An SOE, Transnet, owns all South African ports through the Transnet National Ports Authority (TNPA) and controls the bulk of operations through the Transnet Ports Terminal. Monopoly state ownership means that there is little room for competition between or within ports and therefore little incentive to improve efficiency or productivity levels, which could reduce costs by generating economies of scale.

An outcome of limited competition was that the South African marine tariff structure (port costs to shipping lines) was recognized to be amongst the highest amongst emerging markets (Ports Regulator South Africa 2012). A comparison of terminal handling charges (THC) suggests that Durban—South Africa’s major port—has very high THCs compared to the benchmarked ports.18 In 2012, charges in Durban were found to be in line with European ports, which have higher labour costs than South Africa does. Behind these charges is the differential pricing that operates for bulk and container transport. The Ports Regulator South Africa (2012) roughly estimated that a Durban cargo owner faced between 364 per cent to in excess of over 800 per cent higher charges than the global average for containers. Not all industries are subject to uncompetitive tariffs and bulk commodities (e.g., coal and iron, mild steel, raw timber, etc.) are charged much lower total port costs than the global averages.19

The relatively lower charges for low value add and larger firms reinforce biased policy coordination in favour of particular industries to the detriment of other economic activities. Indeed, this underpins South Africa’s pattern of low economic growth, shaped by large, capital intensive firms within the mineral energy complex. An outcome of the MEC in particular, has been to offer support to various industries through incentives, and of particular interest here, lower transport costs. It is clear that these tariff structures serve to reinforce a growth trajectory favouring large, capital-intensive, and resource-based firms and sectors. In so doing the economy’s port tariff regime is indirectly a constraint on labour-intensive growth.

11.5.3 Regulation in the Communications Sector Has Perpetuated Anti-Competitive Conduct

A number of East Asian countries managed to escape potential growth traps through developing advanced infrastructure in the form of high-speed (p.228) communications networks and broadband technology (Gill and Kharas 2008). Information and communication technology (ICT) manufacturing and implementation was formally embedded in the national development strategy of Korea and the result was a far more developed ICT system than a number of other emerging markets, manifest in part in an internet usage rate in excess of eighty users per hundred people.20

South Africa’s ICT sector in 2000 lagged significantly behind comparator countries: in 2009, there were only nine internet users per hundred people. However, in the past five years internet access has increased significantly, with forty-one users per hundred people in 2012. Access to mobile phones has been central to this shift. Pricing, however, remains a barrier to access and usage of both fixed-line and mobile phone services. In particular, the cost of equipment such as internet-enabled mobile phones and personal computers is high—as is the cost of accessing services, which has limited growth in the uptake of data services. Growth and development gains that could potentially be achieved through this sector have only recently been realized.

Historically, the telecommunications market in South Africa has been highly concentrated and dominated by the incumbent state-owned operator Telkom: 67 per cent government owned, 30 per cent held by strategic equity partners. For much of the last twenty years, Telkom has had a monopoly over local and long-distance calls, backbone infrastructure for internet provision, international services, and public pay phones (Hodge 1999). The Telecommunications Act (Act 103 of 1996) allowed Telkom exclusivity, while it met its obligation to roll out 2.81 million new lines. After its period of exclusivity, a second network operator was to be licensed in 2003. However, this was severely delayed due to legal and regulatory bottlenecks and the second operator only became operational in 2007. Telkom has been charged by the competition regulators for anticompetitive behaviour—notably in terms of pricing—which has had an overall impact on service providers and ultimately consumer access.

The evidence is thus of another majority state-owned entity where regulation has been used to distort markets to extract rents in favour of certain stakeholders. This has perpetuated a poor quality, inefficient system that has hampered knowledge transfers and increased the cost of doing business in South Africa that, essentially, lowers the potential levels of output and economic growth.

11.5.4 Black Economic Empowerment for the Minority

The premise of Black Economic Empowerment (BEE) was to redress inequalities of the past through a redistribution of ownership rights to those who (p.229) were previously economically disadvantaged. Essentially, BEE was introduced (in 2001) as a tool to support greater economic equality, as it was unlikely that this would be a natural market outcome of the then new political environment (Acemoglu, Gelb, and Robinson 2007).

BEE in practice focused on transferring ownership and control to the politically networked elite. This meant BEE had limited impact and largely excluded low-skilled labour, the unemployed, and those in the informal sector. As a tool for redistribution, the policy has been a failure. Instead it has become, apparently unintentionally, a tool for redistributing rents amongst the elites of the society—so reproducing the pattern of highly unequal economic growth. As Manning (2014) suggests, it has created clusters of rent-seeking behaviour by elites and thus has had a limited impact in empowering the broad majority of black South Africans.

Acemoglu, Gelb, and Robinson (2007) presented evidence of the prevalence of narrow-based BEE. Information was collected on board members of companies listed on the Johannesburg Stock Exchange (JSE), as well as the African National Conference (ANC) executives and elected officials since 1994.21 Names of the Board members and ANC members were then matched up. Fifty-six ANC politicians were found to be on the board of directors of JSE-listed firms. The result essentially reinforced the strong interrelationship between political power and large corporates.22 In particular, it is suggested that previously white-owned firms have been trying to match politically connected people to secure their property rights, or to influence government policy.

The demographic distribution of economic assets through the process of BEE was skewed, and redistribution benefits are ostensibly shared by a small section of the previously disadvantaged. Black ownership on the Johannesburg Stock Exchange is estimated to be about 15 per cent, up from about 4 per cent in the mid-1990s (National Planning Commission 2011). Whilst firms are actively responding to BEE and in its newer guise—Broad-Based Black Economic Empowerment (BBBEE)—it is suggested that this is done in a static structural context where firms look to established networks for participation in the economy, limiting the number of beneficiaries of BEE (Andrews 2008).

Ultimately this suggests that BEE and BBBEE, have served as instruments of political and policy access for large corporate South Africa, reinforcing existing economic structures with the consequence of neither redistributing wealth nor serving as an engine for a more dynamic inclusive growth path for South Africa.

(p.230) 11.6 Conclusions

We have described South Africa’s economic development trajectory as being starkly representative of a low level economic growth trap. An undiversified export profile, a low-quality schooling system, and insufficient savings and investment levels, all serve to perpetuate this growth malaise. Our analysis points to incoherence in policy coordination and a lack of clear politically supported industrial policy framework. In turn, the default social contract between big business, government and labour is readily evident in key areas of the economy ranging from port tariffs to telecommunications. We argue that those instances of a sewn-up social contract serve to perpetuate a growth path which favours: capital intensity over labour intensity; the currently endowed to the marginalized; oligopoly to competition; importation ahead of innovation; and heavy manufacturing over light manufacturing. These are the ingredients we identify, for a low growth performance and the perpetuation of a growth path unable to engender strong redistributive outcomes and employment gains.

Arguably, the pattern and level of economic growth lies at the core of persistently high levels of inequality, which has ensured that a minority of high-end, well-educated households at the top of the distribution have gained relative to those at the bottom. This lack of pro-poor growth in South Africa is a direct manifestation of a path dependency in growth which consistently favours modes of economic activity which inhibit innovation and diversification, reproduce patterns of inequality, and marginalize employment creation.


Bibliography references:

Abramovitz, M. (1986). ‘Catching Up, Forging Ahead, and Falling Behind’. Journal of Economic History, 46(2): 385–406.

Acemoglu, D., S. Gelb, and J. Robinson (2007). ‘Black Economic Empowerment and Economic Performance in South Africa’. Non-Technical Policy Brief, Harvard University Center for International Development, Cambridge, MA.

Agénor, P., O. Canuto, and M. Jelenic (2012). Avoiding Middle Income Growth Traps. Washington, DC: World Bank.

Andrews, M. (2008). Is Black Economic Empowerment South Africa’s Growth Catalyst? A Study of the Policy’s Impact on Relational Structures in the Economy, Working Paper RWP08-033. Cambridge, MA: Harvard University Center for International Development.

Bhorat, H., and Hirsch, A. (2014). ‘South Africa: Perspectives on Divergence and Convergence’. Paper presented at Think Tank 20, Growth Convergence, and Income Distribution: The Road from the Brisbane G-20 Summit, Brookings, November.

(p.231) Bhorat, H., and D. Tseng (2014). Poverty and Inequality in South Africa: A Consideration of Trends: 2005–2010. Cape Town: University of Cape Town.

Bhorat, H., A. Cassim, and A. Hirsch (2014). Policy Co-ordination and Growth Traps in a Middle-Income Country Setting: The Case of South Africa, WIDER Working Paper WP2014/155. Helsinki: UNU-WIDER. Available at: <https://www.wider.unu.edu/publication/policy-co-ordination-and-growth-traps-middle-income-country-setting> (accessed 28 April 2016).

Bhorat, H., A. Cassim, and D. Tseng (2014). ‘Higher Education, Employment and Economic Growth: Exploring the Interaction’. LMIP Report 5, Human Sciences Research Council, Pretoria.

Bhorat, H., S. Goga, and B. Stanwix (2013). ‘Occupational Shifts and Shortages: Skills Challenges Facing the South African Economy’. LMIP Report 1, Human Sciences Research Council, Pretoria.

Bhorat, H., R. Kanbur, and B. Stanwix (2014). ‘Estimating the Impact of Minimum Wages on Employment, Wages, and Non-Wage Benefits: The Case of Agriculture in South Africa’. American Journal of Agricultural Economics, 96(5): 1402–19.

Bhorat, H., A. Hirsch, R. Kanbur, and M. Ncube (2014). ‘Economic Policy in South Africa Past, Present, and Future’. In H. Bhorat., A. Hirsch, R. Kanbur, and M. Ncube (eds), The Oxford Companion to the Economics of South Africa. New York: Oxford University Press.

Chakraborty, S., and E. Norris (2005). Rent Seeking, IMF Working Paper WP/05/43. Washington, DC: IMF.

Dosi, G., and L. Soete (1983). ‘Technology and Cost-Based Adjustments: Some Explorations on the Determinants of International Competitiveness’. Metroeconomica, 35(3): 197–222.

Eichengreen, B., P. Donghyun, and S. Kwanho (2011). When Fast Economies Slow Down: International Evidence and Implications on China, Working Paper 16919. Cambridge, MA: National Bureau of Economic Research.

Engelbrecht, H. J. (1997). ‘International R&D Spillovers, Human Capital and Productivity in OECD Economies: An Empirical Investigation’. European Economic Review, 41(8): 1479–88.

Fine, B., and Z. Rustomjee (1996). The Political Economy of South Africa: From Minerals-Energy Complex to Industrialisation. London: Hurst, and Wits University Press.

Gallagher, K. P., J. C. Moreno-Brid, and R. Porzecanski (2008). ‘The Dynamism of Mexican Exports: Lost in (Chinese) Translation?’ World Development, 36(8): 1365–80.

Gill, I., and H. Kharas (2008). An East Asian Renaissance: Ideas for Economic Growth. Washington, DC: World Bank.

Hanival, S., and J. Maia (2008). An Overview of the Performance of the South African Economy since 1994. Pretoria: The Presidency, Government of South Africa.

Hirsch, A. (2005). Season of Hope: Economic Reform under Mandela and Mbeki. Pietermaritzburg: University of KwaZulu-Natal Press and IDRC.

Hodge, J. (1999). Liberalising Communications Services in South Africa. Washington, DC: World 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: UCT Press, IDRC.

(p.232) Kaplan, D. (2003). ‘Impact of Government Programmes using Administrative Data Sets, Investment Incentives’. Project 6.2 of the Ten-Year Review Research Programme, Interim draft.

Kharas, H., and H. Kohli (2011). ‘What is the Middle Income Country Trap, Why Do Countries Fall into It and How Can It Be Avoided?’ Global Journal of Emerging Market Economies, 3(3): 281–9.

Levy, B., A. Hirsch, and I. Woolard (2014). South Africa’s Evolving Political Settlement in Comparative Perspective. Cape Town: SALDRU.

Mahajan, S. (2012). When are Macroeconomic Stability and Exceptionally High Returns Not Enough for Private Investors?. Washington, DC: World Bank.

Manning, C. (2014). ‘Origins, Trends, and Debates in Black Economic Empowerment: The Rationale for Black Economic Empowerment’. In H. Bhorat, A. Hirsch, R. Kanbur, and M. Ncube (eds), The Oxford Companion to Economics of South Africa. New York: Oxford University Press.

National Planning Commission (2011). Economic Diagnostic (June). Pretoria: Government of South Africa.

Nelson, R. R., and E. S. Phelps (1966). ‘Investment in Humans, Technological Diffusion, and Economic Growth’. American Economic Review, 56(1/2): 69–75.

OECD (Organisation for Economic Co-operation and Development) (2014). Perspectives on Global Development 2014: Boosting Productivity to Meet the Middle-Income Challenge. Paris: OECD Development Centre.

Olson, M. (1971). The Logic of Collective Action; Public Goods and the Theory of Groups. Cambridge, MA: Harvard University Press.

Presidency (2012). Development Indicators. Pretoria: Government of South Africa.

Ports Regulator South Africa (2012). Global Port Pricing Comparator Study: 01/04/2012 Research Summary. Durban: Ports Regulator South Africa.

Quantec (2014). Capital Intensity dataset (various years). Pretoria: Quantec.

Rodrik, D. (2006). ‘Understanding South Africa’s Economic Puzzles’. CEPR Discussion Paper No. 5907, CEPR Discussion Papers.

Rodrik, D. (2013). The Past, Present and Future of Economic Growth, Working Paper 1. London: Global Citizen Foundation.

Rustomjee, Z. (2006). ‘DTI Incentives Review Process Phase 1 Incentives Policy Framework, Final Report’, 16 November. A study commissioned by the Department of Trade and Industry, Pretoria.

Rustomjee, Z., and S. Hanival (2008). A Review of Industry Policy, Instruments and Support Programmes 1994–2008. Pretoria: TIPS.

Statistics South Africa (2014a). ‘Labour Force Survey (LFS) (various years)’, Statistics South Africa, Pretoria.

Statistics South Africa (2014b). ‘Quarterly Labour Force Surveys (QLFS) (various years)’, Statistics South Africa, Pretoria.

Takala-Greenish, H. L. (2008). ‘The Role of Industrial Policy and the Minerals-Energy Complex in the Decline of South African Textiles and Clothing’. TIPS Annual Forum Paper: mimeograph. Available at: <http://www.tips.org.za/research-archive/annual-forum-papers/2008/item/1580-the-role-of-industrial-policy-and-the-minerals-energy- (p.233) complex-in-the-decline-of-south-african-textiles-and-clothing> (accessed 15 September 2016).

Trade & Industrial Policy Strategies (TIPS) (2014). Review of Regulation in the Ports Sector. Pretoria: TIPS.

World Bank (2006). Enhancing the Effectiveness of Government in Promoting Micro, Small and Medium Enterprise, AFTPS Consultants. Washington, DC: World Bank.

World Bank (2011). South Africa Economic Updates, issue 1. Washington, DC: World Bank.

Yao L., J. Whalley, S. Zhang, and X. Zhao (2008). The Higher Educational Transformation of China and its Global Implications, NBER Working Paper 13849. Cambridge, MA: National Bureau of Economic Research.


(1) In this chapter, shortened names are used; the sectors’ full names are as follows: Agriculture, Hunting, Forestry and Fishing; Mining and Quarrying; Manufacturing; Electricity, Gas and Water Supply; Construction; Wholesale and Retail Trade; Transport, Storage and Communication; Financial Intermediation, Insurance, Real Estate and Business Services; Community, Social and Personal Services; and Private Households, Exterritorial Organizations, Representatives of Foreign Governments and other Activities not adequately defined.

(2) See figure 1 in Bhorat, Cassim, and Hirsch (2014), available at: <http://www.wider.unu.edu/publications/working-papers/2014/en_GB/wp2014-155/> (accessed January 2015).

(3) See figure 3 in Bhorat, Cassim, and Hirsch (2014).

(4) See figure 4 in Bhorat, Cassim, and Hirsch (2014).

(5) See figure 5 in Bhorat, Cassim, and Hirsch (2014).

(6) See figure 6 in Bhorat, Cassim, and Hirsch (2014).

(7) See figure 7 in Bhorat, Cassim, and Hirsch (2014).

(8) See figure 9 in Bhorat, Cassim, and Hirsch (2014).

(9) See table 2 in Bhorat, Cassim, and Hirsch (2014).

(10) See table 3 in Bhorat, Cassim, and Hirsch (2014).

(11) See figure 10 in Bhorat, Cassim, and Hirsch (2014).

(12) See figure 11 in Bhorat, Cassim, and Hirsch (2014).

(13) See figure 12 in Bhorat, Cassim, and Hirsch (2014).

(14) See figure 13 in Bhorat, Cassim, and Hirsch (2014).

(15) See figure 14 in Bhorat, Cassim, and Hirsch (2014).

(16) Annual employment growth differs in Figures 11.1 and 11.2 as the versions of the Labour Force Surveys (LFS) and Quarterly Labour Force Surveys (QLFS) used were not the same; as such, different quarters or annuals were used and may yield different growth rates.

(17) Ninety-six per cent of export volumes are estimated to be transported by the sea (TIPS 2014).

(18) See figure 17 in Bhorat, Cassim, and Hirsch (2014).

(19) A few examples of the disparity in pricing are worthy of noting: see section 6.2 in Bhorat, Cassim, and Hirsch (2014: 24).

(20) See figure 18 in Bhorat, Cassim, and Hirsch (2014).

(21) The ANC has been the dominant political party since 1994.

(22) See figure 19 in Bhorat, Cassim, and Hirsch (2014).