Overcoming underdevelopment in South Africa's second economy

Abstract

This paper is a synthesis of the July 2005 Development Report published by the Development Bank of Southern Africa, Human Sciences Research Council and United Nations Development Programme (DBSA, HSRC and UNDP). The Report asks why, if the origins of economic dualism are rooted in the cheap, forced, migrant labour introduced by the mining industry and reinforced during apartheid, does dualism persist under democracy when all the relevant laws and many of the practices of the past have been abolished? The breakdown of apartheid did not immediately translate into improved material conditions for the majority of South Africans: 300 years of colonialism and 50 of internal colonialism had hard-wired a duality into the system. Two worlds, which may be conceptualised as the first and second economies, coexisted: a globally integrated world of production, exchange and consumption, and a constrained world of informality, poverty and marginalisation. This synthesis sheds light on the origin and nature of the ‘second economy’ metaphor, and suggests solutions.

This paper is a synthesis of the 2005 Development Report published by the DBSA in July 2005. It was prepared by Dr Michael Aliber (Human Sciences Research Council), assisted by Rasigan Maharajh (Institute for Economic Research), Marié Kirsten (Development Bank Southern Africa), Josephilda Nhlapo-Hlope (Development Bank Southern Africa) and Oupa Nkoane (Development Bank Southern Africa). Other authors of the development report upon which this article is based include Glynn Davies, Daniel Kekana, Professor Ben Turok, Nick Vink, Julian May, Anna McCord and Mark Oranje. It is available from the Knowledge Centre at the DBSA, www.dbsa.org/

1. Introduction

In 2003 President Thabo Mbeki refocused the debate around the stubborn persistence of poverty and underdevelopment in South Africa, when he reintroduced the concept of the two economies into the policy discussion. Since then, policy-makers and practitioners in the development community have debated the existence of two economies: is the concept an apt description of the South African reality or is it merely a metaphor and, if so, what is its value? Opponents argue that the two-economy conceptualisation runs the risk of distorting perceptions of the real problem, making solutions harder to find. Nevertheless, the introduction of the two-economy discourse has been remarkable for the extent to which it has drawn attention to the persistence of underdevelopment in South African society. And despite concerns that the metaphor may lend itself to simplistic prescriptions along the lines of ‘how to bridge the gap between the two economies’, it has in fact stimulated us to think more deeply about why poverty has turned out to be so difficult to defeat, and what an effective anti-poverty programme might look like.

The point of departure for the Development Report is a simple question. Why, if the origins of economic dualism are rooted in the system of cheap, forced, migrant labour introduced with the beginnings of the mining industry and reinforced during apartheid, does dualism persist under democracy when all the relevant laws and many of the practices of the past have been abolished?

In addressing this question, the Report has sought to provide a variety of perspectives on the question of underdevelopment's persistence. Briefly stated, it has considered the historical origins of underdevelopment in South Africa; the persistence of underdevelopment and poverty in the context of current policy; the current dimensions and trends of underdevelopment and poverty; and specific sectoral attempts to address underdevelopment. This paper offers a synthesis of these main themes, and ventures to suggest a few implications for government policy.

2. Origins of structural underdevelopment

South Africa's transition from apartheid oppression to democracy and freedom is widely regarded as a success. A racially based system of political power has been transformed into a non-racial democracy. The spatial planning system of ethnically separate homelands has given way to a united country in which power is decentralised to three tiers of governance. Gender equality is enshrined in the Constitution, and labour market and other sectoral interventions specifically seek to achieve equity and affirmative targets. Macroeconomic stability has been achieved through both monetary and fiscal austerity.

Yet, despite all this, the country still experiences high levels of poverty and extreme disparities in income, wealth and opportunity. This can best be understood as the result of the confluence of two distinct processes: first, South Africa's own particular history up to 1994, and secondly, the changes in the global economic environment in the second half of the 20th century, including the emergence of ‘ideological globalisation’.

South Africa's development, and in particular the development of its dual economy, is a mirror image of the evolution of the current world system. Our country has been subjected to conquest, dispossession, imperialism and colonialism, and more recently (and more happily) to national liberation and development. This has produced an economy categorised, on the basis of per capita GDP, as middle-income. However, this aggregate characterisation masks the hugely disparate circumstances under which South Africans experience their country.

The European presence in South Africa dates from 1652, when a port was established at the Cape to facilitate trade with the East. From the beginning, European mercantile interests took precedence. As the settlement expanded, primarily through agriculture, property rights at odds with existing customs were imposed on the indigenous people. The initial phase of colonialism was characterised by competition and conflict with local producers, especially of agricultural products. Subsequently, as the colonial power expanded to exploit South Africa's rich natural and mineral resources, racist ideology conspired with economic self-interest to turn the country's black majority into a pool of cheap labour. A massive apparatus of control and repression was created, attaining its highest form in the apartheid state.

From an economic point of view cheap labour practices were unsustainable, since the distorted labour market eventually placed fetters upon economic expansion. The evolution of the country's secondary and tertiary sectors was increasingly constrained, as industry began to demand higher levels of skill and greater mobility of human resources. The apartheid system could never increase the value of human capital on the scale demanded by a changing economy. Also, with over three-quarters of the population confined to a mere 13 per cent of the country's land, and the outlawing of sharecropping, blacks were prevented from advancing even in the agricultural sector; indeed, they were largely prevented from being self-sufficient in meeting their food requirements.

After the Second World War, while South Africa was pursuing its doomed policies, the United States emerged as a major world power. Increasingly, the United States would shape the new world order, culminating in the ‘American counter-revolution’ or the phase of ideological globalisation. During this period, the United States sought to deregulate its economy and liberalise financial flows to facilitate its external borrowing. This created the conditions for an expansion of its multinational corporations, while at the same time the indebtedness of developing countries rapidly increased.

South Africa entered the last three decades of the 20th century at a completely different level from other countries with similar economic profiles. The intensification of the liberation struggle, the emergence of an urban-based black working class with some degree of negotiating power, and general discontent with an increasingly militarised state formed the preconditions for a negotiated settlement. Under these circumstances, and amidst the global competition for capital, the apartheid regime's capacity to maintain the mechanisms of repression began to fail.

The breakdown of apartheid did not immediately translate into improved material conditions for the majority of South Africans. Three hundred years of colonialism, and fifty of internal colonialism, had hard-wired a duality into the system, whereby two domains coexisted: on the one hand, a globally integrated world of production, exchange and consumption and, on the other, a constrained world of informality, poverty and marginalisation. These two worlds may be conceptualised as the first and second economies.

3. Development policy since 1994: An overview of current debates

A convenient way of conceptualising the government's fight against poverty is the three-pillar formulation used by President Mbeki in a speech to parliament in May 2004:

At the core of our response to all these challenges is the struggle against poverty and underdevelopment, which rests on three pillars. These are: encouraging the growth and development of the First Economy, increasing its possibility to create jobs; implementing our programme to address the challenges of the Second Economy; and, building a social security net to meet the objective of poverty alleviation. (Mbeki, 2004)

What is the likely contribution of each of these pillars to the effort to overcome underdevelopment?

Regarding the ‘first economy’ pillar, there is endless debate as to whether South Africa's macroeconomic policy is the correct one. The introduction of growth, equity and redistribution programmes (GEAR) in 1996 was perhaps the most divisive event in the government's policy-making after 1994, and it has remained divisive. The present status of GEAR is somewhat ambiguous. To those not enamoured of the policy, the indications that the government increasingly wishes to embrace the idea of the ‘developmental state’ amount to a tacit admission that GEAR has failed. The government meanwhile asserts that it is the success of GEAR, and not least the stability brought about by fiscal and monetary discipline, that will make the developmental state a possibility. At present it is unclear what exactly the move towards a developmental state will entail, but it appears unlikely that it will involve a rejection of the mainstays of macroeconomic policy since GEAR.

In one respect, there is complete agreement on the proper goal of macroeconomic policy: faster growth. However, there is little consensus on how to achieve this faster growth and, as importantly, how to ensure that growth becomes more pro-poor. A key concern is that in the present macro environment the industries that appear to be thriving are those that are relatively capital-using and labour-saving. Many South Africans whose skills were appropriate to mining, agriculture and other primary industries are losing out as the nature of the economy changes. Some argue that the situation could and should be reversed with an industrial policy that aims to promote labour-intensity. Others argue that, in the absence of strong import controls, South Africa is structurally unsuited to competing in these industries, and must rather seek to become ever more competitive in high-tech industries and services. If the latter path is chosen, which seems to be the case, the challenge is all the greater, especially in creating the necessary human capital. One interpretation of persistent poverty over the past ten years is that it represents the short- and medium-term consequences of a strategy that, if followed successfully, can plausibly offer broad-based benefits only in the long term.

Whatever the future may hold as far as tweaking or overhauling macroeconomic policy is concerned, it is clear that the first pillar (‘first economy’) alone will not reduce poverty quickly on a massive scale. Thus, the limited short- and medium-term potential of first-economy growth to alleviate poverty is a principal argument in favour of robust interventions in what we are now calling the ‘second economy’.

Jumping for a moment to the third pillar, we observe that South Africa's social security net is justifiably regarded with both pride and concern. On one hand, it represents a principled undertaking on a massive scale to help the most vulnerable, which it turns out is not a small number of people. On the other, it is a fiscal burden which the government is struggling to contain. Beyond social grants such as the old-age pension, the child support grant and the disability grant, which are received by almost 30 per cent of households, there is even greater expenditure in the form of the ‘social wage’, including free basic water and electricity, free primary health care, subsidised education and so on. Both social grants and the social wage are distributed in a decidedly progressive manner.

Arguably the most important policy question in respect of the social security net is whether it induces dependency or rather acts as an economic stimulus. Happily, recent research seems to support the conclusion that, on balance, grants do not induce dependency, and may have beneficial effects such as enabling more and more effective job search, promoting school attendance, and shifting aggregate consumption in favour of goods with high domestic production content (EPRI, 2004). In other words, social grants are not ‘mere welfare’, but rather are developmental. None the less, at a time when the government is correctly seizing on the need to boost investment in the country's economic infrastructure, the steady increase in real expenditure and budget share on social protection is indeed reason for concern. The hope of some people for the introduction of a basic income grant – not least because the present social security system, immense though it is, omits direct support to huge numbers of people (e.g. single men) consigned to the economic margins – is understandable, but its realisation is hard to imagine at a time when the cost of the existing set of grants is rising too quickly for comfort. The hope that developmental interventions will help wean people off dependency on grants is not necessarily the main issue; the more critical question is whether second-economy interventions can assist the many people who, by the design of the system itself, are not eligible for grants in the first place.

So the need for so-called second-economy interventions is doubly important, on one hand because of the short- and medium-term limitations of what the first economy can offer, and on the other hand because of the necessarily large gaps in the social safety net. In addition, it has been mooted that second-economy interventions are not mere palliatives, but can and should be designed to equip people to either link to, or eventually be absorbed in, the first economy.

However, as the Report illustrates, despite some progress in identifying and launching second-economy interventions, the general picture is that it is here that the government has the furthest to go in improving its performance. It is noted with approval that the government is candid about the need for improvements as expressed, for example, in the President's 2005 State of the Nation address (Mbeki, 2005). But it is doubtful whether the government in general fully appreciates how inadequate its second-economy interventions are. Shifting attention away from ‘welfare’ and towards development is no doubt very difficult, but it is unclear whether the magnitude of the challenge is sufficiently appreciated. The absence of a more coherent, scale-appropriate second-economy strategy is a major impediment.

4. Getting the measure of underdevelopment

South Africa's per capita GDP, corrected for purchasing power parity (PPP), was $11 240 per annum in 2001, making the country one of the 50 wealthiest in the world. However, strikingly poor social indicators resulted in a ranking of 111 of 175 countries in terms of the Human Development Index (HDI) for that year. In fact, South Africa's HDI ranking declined from 93 in 1992 to 115 in 2003 (UNDP, 2002, 2003), and it is one of only a handful of countries that has experienced a decline since 1995. Despite being among the 50 wealthiest nations in the world, the country now has a life expectancy that is among the 30 worst, while projections of mortality suggest that this measure will deteriorate further as the number of deaths from the HIV/Aids pandemic increases (UNDP, 2003, 2004; Dorrington et al., 2001: 25). The racial disparities in South Africa's development indicators have also remained largely unchanged.

Money-metric measures of poverty and inequality largely support the same conclusions, notwithstanding debates about data quality and methodology. Generally speaking, the statistics show that both poverty and inequality have increased since 1993, accompanied by divergence rather than convergence between the poor and non-poor. The data also suggest the presence of poverty traps that may hinder mobility and prevent individuals from getting ahead, despite the dramatic economic and political reforms since 1994.

The trend of increasing poverty is all the more alarming in the light of the massive increase in government spending on the poor. Van der Berg & Burger (2002: 10–11) show that social spending on the African population increased from 51 per cent of the total in the immediate transition period to 80 per cent in 1997. After taking account of cash and non-cash transfers to the poor from the government in the form of pensions and the child support grant, water and electricity, health care, housing, sanitation, education and transport, Meth (2004, Table 6a) amends his estimate of the increase in the numbers of poor between 1997 and 2002 from 3.5 million to 2.7 million people. Of course, the aim of much of this spending is to improve the quality of life by providing better services and infrastructure, which does not necessarily affect people's incomes.

It is well known that a central contributor to poverty and inequality is the high level of unemployment. Between 1995 and 2002, the working-age population of South Africa grew by about 2.2 per cent per year, while the economically active population increased by about 4.8 per cent per year, implying a massive increase in labour force participation. Interestingly, three fifths of the new labour market entrants over this period were women. Various factors appear to account for this increase, and to some extent it can be interpreted as a positive reaction to greater freedom – of mobility, for instance – and perhaps higher aspirations. Be that as it may, over this period the number of employed people increased by only one-third as much as the economically active population, with the result that an ever larger number of people and an ever greater share of the labour force were unemployed.

More telling is the nature of labour market participation (and non-participation) of the poor. First, poor households tend to be disconnected from the labour market: only 36 per cent of adults in the bottom decile of the income distribution are economically active and three-quarters of these are unemployed, implying that only 9 per cent of working-age people in this decile are actually working. This compares with the top decile, in which 68 per cent of adults are working.

Secondly, there has been a rapid increase in the number of households, accompanied by a decrease in average household size and increased in-migration and inter-provincial migration. Given that employment has been growing very slowly, this increase in the number of households has been accompanied by a sharp increase in the number of households in which no one is employed. According to Pirouz (2004), the proportion of households containing unemployed people rose from 13 per cent in 1995 to 27 per cent in 2002. Over the same period, the proportion of workerless households containing unemployed people grew from 4.7 per cent to 11.6 per cent. Many of these households do not appear to receive any remittances. Households of this type seem to have increased in number from 1.3 million in 1997 to about 2 million in 2002. They were home to about 5.8 million people in 1997 and roughly 7.4 million in 2002 (Meth, 2004).

Thirdly, there has been a marked increase in the number of ‘working poor’: that is, those who are employed, even in the formal sector, but remain in poverty. For example, using the PPP$2 per day poverty line, Casale et al. (2004: 11) show that there has been an increase in the number of the employed living in poverty from just over 900 000 in 1995 to some 2 million in 2003. Of the latter, only a quarter are self-employed in the informal sector.

In the face of rapidly expanding labour market participation, unemployment has grown despite the increasing numbers of people actually finding work. In some ways this might be taken as a positive sign, signalling that the livelihood options have increased for people formerly excluded from the labour market by restrictive apartheid legislation. However, the disproportionate job losses in unskilled work and the increasing presence of workerless households suggest that, for most people, better opportunities to seek employment have not been matched by improved prospects of actually finding it.

5. Short-term interventions to address underdevelopment

We have argued that second-economy initiatives in South Africa are generally not guided by an overall, coherent framework that explains the respective roles of each main state agency or the scale that the initiatives should seek to achieve. Paradoxically, the government is certainly not stingy when it comes to spending money on the poor, yet appears incapable of effecting the shift that it has often declared so essential, namely away from welfare and towards development. If anything, the stark asymmetry between levels of spending on welfare and on developmental projects to reduce poverty has been growing rather than shrinking in recent years. One of the reasons – apart from the manifest difficulty of spending money developmentally – is the apparent lack of strategic coherence when it comes to second-economy initiatives. This is all the more peculiar in light of the fact that one of the departments chiefly responsible for such initiatives, the Department of Social Development, successfully manages the much larger social security programme. The contrast between the comprehensive social security framework, which looks to formulate a meaningful, interdepartmental system for dealing with vulnerability, and the much more disparate collection of second-economy initiatives is worth pondering.

Reflecting the 2004 Pretoria conference on which it is based, the Report examines four sectors in which the government has sought to intervene in the second economy. The intention is not to provide an exhaustive account of what the government is doing; rather, it is to examine key examples that may be taken as illustrative of the government's developmental programmes, with the aim of showing what is and is not going well, discerning common patterns among diverse initiatives, if any, and offering constructive criticism. The four sectors examined are the Department of Provincial and Local Government's nodal and spatial development programmes (the Integrated Sustainable Rural Development Programme and the Urban Renewal Programme); the public works programmes, with a particular focus on the Expanded Public Works Programme; the agricultural sector, including but not limited to land reform; and small business development initiatives.

5.1 Nodal and spatial development

President Mbeki announced the Integrated Sustainable Rural Development Strategy in February 2001 during his State of the Nation address. Since then the strategy has been transformed into an implementable programme. The Integrated Sustainable Rural Development Programme (ISRDP) is designed to build socially cohesive and stable rural communities with viable institutions and sustainable economies, offering universal access to social amenities and able to attract and retain skilled and knowledgeable people who can contribute to growth and development. The impetus to create the ISRDP was the observation that investment in rural areas did not have the envisaged effect of alleviating poverty and strengthening local institutions. While the range and quality of existing development and anti-poverty programmes appeared to be appropriate, their ability to provide an integrated basket of services that matched local priorities was weakened by poor coordination and integration. The ISRDP aimed to correct these institutional failures. Currently the programme is running in the District Municipality ‘nodes’ of Zululand, Ugu, OR Tambo (named after Oliver Reginald Tambo), Alfred Nzo, Umzinyathi, Chris Hani, Umkhanyakude, Central Karoo, Maluti a Phofung, Kgalagadi, Sekhukhune and Bohlabela.

The Urban Renewal Programme (URP), which was also announced in the 2001 State of the Nation address, is based on a similar premise, but obviously focuses on urban areas, generally townships. Currently the programme is running in six nodes, representing the largest, most extreme concentrations of urban poverty and neglect in South Africa. The nodal points are Galeshewe (Northern Cape), Alexandra (Gauteng), Khayelitsha and Mitchell's Plain (Western Cape), KwaMashu and Inanda (KwaZulu-Natal), and Mdantsane and Motherwell (Eastern Cape).

Even though the ISRDP and the URP differ in focus and strategy, among other things, they have some key similarities. First, they both take development nodes as their point of departure – whole district municipalities (with a single exception) for the ISRDP, and segments of urban areas (i.e. townships) for the URP. Secondly, both programmes were conceived as pilot projects to test intergovernmental attempts at renewal, reconstruction and poverty alleviation. Over time, however, the focus increasingly shifted away from being pilot projects, and the call for ‘more nodes’, which was heard as early as 2003, suggested that for many both programmes had acquired the status of ‘normal government action’. Thirdly, both programmes sought alignment with integrated development plans, and viewed these plans as the primary vehicle for mainstreaming and giving effect to the nodal projects. Fourthly, the idea was not to start a new fund but to ‘do things differently’, to package an area's existing services, programmes and budgets in a more integrated, coordinated and programmed manner. Fifthly, to lend the process the necessary urgency, senior politicians were named as champions of the nodes. Finally, the programmes were built not on a welfarist premise but rather on the idea of ‘helping the poor to get into jobs and out of welfare’.

Four years into their lives, these programmes have had mixed results. Gains have been made in redirecting budgets towards the nodes; fast-tracking infrastructure investment and service delivery in nodes; establishing shopping centres or precincts with a strong BEE component; creating temporary jobs; setting up multi-purpose community centres; stimulating intergovernmental dialogue around infrastructure investment and development spending in the nodes; and identifying and channelling investment by provincial governments to ‘provincial nodes’. But progress has been uneven, and failure has frequently been a result of poor coordination, the main problem the programmes were meant to address. The reasons for this include confusion about the roles of actors in the various spheres of government, and lack of skills, especially in the area of urban renewal.

The biggest problem facing both the ISRDP and the URP may have to do with conceptualisation rather than implementation. Both programmes are premised on the idea that problems or backlogs must be addressed in the spaces where they occur. In contrast to the National Spatial Development Perspective, there is insufficient recognition that many people find themselves in zones which have little or no potential, largely because of the designs of colonial powers or the apartheid state. Freezing the country in this apartheid space economy would be not only economically unwise, but morally perverse. An allied concern is the failure to recognise the limitations of ‘developmental local government’, which for many means a municipality that kick-starts and supports local economies, ensuring that people can find employment, regardless of the fact that some areas have extremely limited development potential. It is counterproductive to argue for ‘getting people out of welfare’ in areas where the best that can be hoped for is actually more and better focused welfare.

5.2 The expanded public works programme

The Expanded Public Works Programme (EPWP) was launched in 2004 with the aim of increasing employment and supporting skills development. As an umbrella programme, it subsumed initiatives that were previously entirely independent, such as the Community Based Public Works Programme and Working for Water. Their activities continue, and generally under the same government departments that originated them, but now within the EPWP's comprehensive framework. In addition, the EPWP subsumes some new public works activities, such as ‘social sector public works’, and promotes the idea of labour-intensive activities wherever government investment in infrastructure is involved.

The EPWP is led by a sector coordinating department in each of its main sectors. The Department of Public Works is both the overall coordinator of the programme and the sector coordinator for the infrastructure sector. The Department of Labour funds programmes to meet the training entitlement for workers employed on EPWP projects, and coordinates an EPWP training committee with representatives of all the sector coordinating departments.

The EPWP is supported by funds earmarked in the budgets of line-function departments, provinces and municipalities. In the infrastructure sector, R15 billion of the conditional infrastructure grants which will be allocated to provinces and municipalities over the next five years has been earmarked for the EPWP. In addition, R4 billion has been earmarked in the environmental sector departments' budgets for environmental EPWP programmes over the next five years, and at least R600 million will be allocated to social sector EPWP programmes over the same period.

There are two main points to make about the EPWP. First, it occupies a place in the popular discourse, and the public's expectations of the programme are out of proportion to its actual potential. The contribution of the EPWP to job creation (‘one million jobs in five years’), as occasionally touted by politicians, is perhaps unintentionally misleading, given that most of these jobs are very temporary, and that at any given time there are likely to be no more than 200 000 of them. An interesting example of the gulf between the rhetoric and the reality is the suggestion that the EPWP will make a meaningful contribution to moving people off social grants and into public employment, which is frequently cited as a key (ideological) benefit of the EPWP over alternative interventions. This is problematic, as the working-age poor who are to participate in the programme are not eligible for grants under the existing social protection regime, hence public works employment will not help them to ‘move out of dependence on social grants’. Moreover, even if this were not the case, the scale of the EPWP is in fact very modest, as already suggested, and certainly too modest in relation to the number of people receiving grants. A counterclaim might be that the EPWP is the most significant transfer targeted at those who do not receive grants, but unfortunately this is not saying a great deal given the relatively small numbers of potential beneficiaries.

The second main point to make about the EPWP is that, irrespective of its inadequate scale, its design does not appear to correlate well with the main problem it is trying to address, namely unemployment. In so far as public works programmes are used internationally to deal with this problem, their design depends on whether they are intended to respond to unemployment as a transient labour market problem or, more rarely, as a chronic problem. In the former case, the policy instrument of choice would be temporary employment programmes to address cyclical unemployment, for instance, or unemployment resulting from an acute economic or climatic shock. In the latter case, the policy instrument would aim to create demand through the implementation of large-scale public sector employment programmes, offering employment guarantees and sustained employment.

Unemployment in South Africa is clearly chronic rather than acute or cyclical, yet the EPWP offers short-term, temporary employment only, and characterises the unemployment problem as transient, pending the rising tide of employment resulting from economic growth – this despite the fact that the structure of the South African labour market is expected to keep unemployment among unskilled workers above 30 per cent even in a positive growth scenario (Lewis, 2001). In short, the EPWP offers an acute response to a chronic problem. It is worth mentioning, too, that prior to the establishment of the EPWP the various public works initiatives in the country operated according to different norms, in which for example some offered longer-term employment than others. Despite a body of research showing the greater appropriateness of this approach to the South African situation, the EPWP established a short-term employment norm. Either this was a failure to learn the lessons of the past; or it was an attempt to boost the appearance of efficacy, by proclaiming a programme which would reach a large number of individuals relative to budget, even if only briefly and to little effect.

5.3 Agriculture

The role of agriculture in the economies of developing countries is generally acknowledged. However, there is no consensus on whether agriculture is the most appropriate way to fight rural poverty. One school argues that since most people in developing countries live in rural areas and are engaged in agriculture or agriculture-related activities, agriculture is the most effective way to reduce poverty. A second school recognises the contribution of agriculture to poverty alleviation, but attaches more importance to non-agricultural activities, e.g. rural non-farm enterprises and social services.

In any event, there are various reasons why the experience with agricultural development in other parts of the developing world does not apply without qualification to South Africa. In the first place, there is the dualism of the agricultural sector. Dualism is a key theme of the Report, but it is especially relevant to the agricultural sector, which on the one hand comprises some 46 000 largely white commercial farmers who occupy 87 per cent of the total agricultural land, and on the other more than two million black households who farm on the remaining 13 per cent of this land. Whereas the commercial sector is highly capitalised and increasingly integrated into global markets, the black agriculturalists produce mainly for their own consumption.

A second consideration is that in South Africa, in contrast to most of the other economies of sub-Saharan Africa, agriculture accounts for a small share of GDP (about 4 per cent at present) and a modest share of total formal employment. Even in terms of informal employment – that is, including self-employment in the former homelands – it is only a fraction of those who practise agriculture who do so on a full-time basis or regard it as their main source of income or even food.

A third consideration – related to both of the above – is that in South Africa rural people tend to migrate to urban areas in search of incomes and jobs, even when the chances of landing a permanent job and receiving a predictable income are minimal. Under these conditions, households often migrate as a second-best option in search of superior infrastructure such as better school and health facilities, improved housing, water and sanitation, electricity, and better transport. The historical and demographic consequences of apartheid have led to high levels of artificial rather than spontaneous densification for the poor African rural population, and to acute demand for living sites in areas with access to the developed economy. In effect, South Africa's rural population is becoming more concentrated in dense rural informal settlements, where cultivation and stock farming are disappearing. Poor rural people are moving away from a collapsing land economy towards the nearest location of the developed cash economy (Bekker, 2004).

None the less, agriculture forms a small but important buffer against poverty and hunger for many households in the former homeland areas, while also serving as part of a strategy to create wealth for a relatively small number of typically wealthier households. What is the government's vision for the agricultural sector, and how is it seeking to give effect to this vision? The key theme in most official statements coming out of the sector is one of improving equity by means of improving access, whether this is access to land, markets or credit. It is also clear that, on some level at least, the objective is to address the dualism that so starkly characterises the agricultural sector: ‘It is government's policy to facilitate a comprehensive and structural transformation in the agriculture economy in order to achieve a United and Prosperous Agricultural Sector in partnership with the other stakeholders’ (Department of Agriculture, 2004: 5).

It is tempting to interpret this and other statements as suggesting an excessive focus on the commercialisation of agriculture, with the attendant risk that benefits will go mainly to those who are better off to begin with (the uneven benefits to small and large farmers have always been a major theme of evaluations of Asia's Green Revolution). Further, it has been credibly argued that the policy shift in the government's approach to land redistribution in 2001 put an emphasis, albeit a non-exclusive one, on providing access to land for commercial purposes. However, there are also numerous initiatives that stress the importance of agriculture for purposes of food security, among which those of the various provincial agriculture departments may be singled out.

Indeed, to the credit of the Departments of Agriculture and Land Affairs, initiatives in the agricultural sector touch on a wide range of issues. In addition to the two most visible initiatives, that is, the Land Reform Programme and the more recently launched Comprehensive Agricultural Support Programme, institutional restructuring and programme interventions have also addressed marketing policy, the consolidation of what had been racially distinct agriculture departments, water law reform, labour market policy, trade policy, credit infrastructure, and so on. However, some of the specific initiatives are either not being implemented on a sufficient scale to have a meaningful impact (e.g. land redistribution), or are not well conceptualised or targeted. A concern in this second respect is the soon-to-be-launched agricultural credit scheme, Mafisa. While it is true that credit access is lacking in most former homeland areas, among other places, it is well known that small-scale producers struggle to service debts out of agricultural incomes, and thus are often further impoverished by debt rather than empowered by it.

Certainly, debt in the absence of comprehensive farmer support is a doubtful proposition for many recipients. In this respect, there is reason for concern. Although in theory the Comprehensive Agricultural Support Programme is aimed at enhancing farmer support services, in fact at present the policy is that 70 per cent of the R200 million budget is being directed to beneficiaries of land reform, generally for capital improvements on their land. Apart from the fact that these individuals have already benefited from relatively large amounts of government support to acquire land in the first place, they represent a small fraction (less than 2 per cent) of the number of blacks involved in agriculture. It is difficult to imagine that the balance of the budget will make an appreciable difference to the millions of households involved in agriculture, not to mention the hundreds of thousands of others who have access to land but are not using it. Moreover, there is little evidence that the capacity of agricultural extension services has undergone the necessary transformation to make a meaningful difference on the ground.

5.4 Small enterprise

Small enterprise policy has been dominated by the view that small and medium-sized enterprises (SMMEs) are smaller versions of formal sector enterprises within the first economy. This is reflected, for example, in President Mbeki's opening address to Parliament in February 2005, in which he attended to small enterprise issues under the banner of ‘interventions to grow the first economy’ (Mbeki, 2005: 7). More crucially, this underlying perception of SMMEs informs the expectation that policies that shed first-economy jobs in the short term, however painful, will lead to a net gain of jobs in the long term, as ‘downstream’ demand from modernising big firms boosts small business.

While this is no doubt partially true, it obscures the fact that in reality small enterprises are heterogeneous. A distinction should be drawn between ‘opportunity entrepreneurship’, which indeed may be stimulated by a growing economy, and ‘necessity entrepreneurship’, which is the province of smaller, survivalist enterprises that are more likely to emerge counter-cyclically. The argument of the Report is that both types of enterprise are important and merit support, but that government policy tends to acknowledge and cater to the former at the expense of the latter. The suggestion is that government initiatives must be appropriately differentiated in order to reach each group effectively.

The main issue is the type of support services offered. Here there are parallels with the agricultural sector, in which there is a growing emphasis on ensuring access to credit regardless of whether that is the most pressing need. In an integrated economy, in which many firms produce outputs as inputs for other firms, potential micro-entrepreneurs will want to take loans to enable them to cater to an identified market. ‘Opportunity entrepreneurs’ would fit this description. In a dual economy, however, those in the second economy are less likely to be able to identify, let alone take advantage of, such opportunities, nor are they likely to be sought out by potential first-economy firms, who are not accustomed to doing business with the informal sector. As a result, such small enterprises as exist in the second economy tend to be ‘necessity entrepreneurs’, or what are sometimes called the ‘non-entrepreneurial self-employed’, whose demand for credit (never mind eligibility) is limited by the fact that they are unable to form contractual relationships with buyers in order to ensure a reasonably constant cash flow.

The experience of Khula Enterprise Finance is instructive. Launched in 1995, Khula's strategy was based from the beginning on the assumption that there are enough linkages between the first and second economies to generate demand for credit for micro-enterprises that would act as suppliers to the first economy. As it turned out, however, these linkages were not there in sufficient quantity. This resulted in a dramatic reduction in the number of credit retailers through which Khula could extend its capital, as well as an ‘upgrading’ in the nature of the client base of those retailers who managed to remain, because in fact the viable demand base turned out to be more limited than originally thought. As a consequence, Khula eventually opened ‘micro-credit outlets’ in order to cater to the smaller enterprise clientele it was not able to reach through its intermediaries, but ultimately it was trying to encourage credit uptake among a category of entrepreneurs for whom credit is often not the most immediate need.

The government's decision to introduce two new small business credit schemes, the new apex institution and Mafisa, is a reaction to the perception that the existing institutions do not cater adequately to poorer entrepreneurs. While they are not unwelcome, what is revealing is the continuing emphasis on micro-finance – indeed, on micro-credit – rather than other types of initiatives which in other developing countries are found to be especially potent, not least important being encouragement, monitoring and advice. Why the neglect of non-financial support? The answer would seem to be the tendency in South African policy-making circles to regard support mechanisms for non-entrepreneurial self-employment as ‘welfare’. Micro-credit remains the domain of the Department of Trade and Industry, or the Department of Agriculture in the case of small farmers, where the entrepreneurial enterprise vision predominates. Meanwhile, the few South African micro-finance institutions that adopt a genuinely pro-poor approach, working to support non-entrepreneurial self-employment – providing opportunities for ‘graduation’ to entrepreneurial small business where appropriate – remain outside the mainstream despite remarkable impact and even financial success.

6. What to do about the second economy?

There are two levels to the discussion about how to address the challenges of the second economy. At one level, it is important to draw lessons from the concrete experience of current and past initiatives. Four of these initiatives have been briefly summarised in Section 5, above. Some preliminary observations are offered below, not for the sake of communicating suggestions for these specific sectoral initiatives, but to illustrate principles that are presumably more broadly relevant to second-economy initiatives. At another, higher, level it is critical to consider second-economy interventions as a whole and what they accomplish in relation to the broader, national objective of overcoming underdevelopment and poverty.

Taking the four initiatives as a whole, we observe a number of common patterns that help to explain the relative ineffectiveness of interventions in the second economy.

•	Failure to disaggregate and quantify: there is a failure to recognise that most sectors are diverse. For example, black farmers include those who produce as entrepreneurs, those who produce casually to reduce the grocery bill at the margin, and those whose lack of other options is such that home production, however modest in scale, is vital to the household's food security. Similarly, small and micro-entrepreneurs differ significantly in their tolerance of risk, their ability and willingness to bear debt, and their capacity to network their way into the first economy. The failure to disaggregate tends to result in the design of interventions that may be appropriate to one segment of the target group but less so to the others. Moreover, there appears to be a pattern whereby interventions tend to be suited to the relatively able, well-off segment, but less so to everyone else. Typically, there are more people in the latter category. Thus it is necessary not only to identify the different segments but also to quantify them, failing which appropriate interventions cannot be scaled. The fact that 70 per cent of the budget of the Comprehensive Agricultural Support Programme is earmarked to benefit land reform beneficiaries, who themselves represent not more than 2 per cent of all blacks involved in agriculture (excluding farm workers), is a case in point.
•	Neglect of what is there: a related tendency is for developmental interventions to ignore people's existing skills and experience. For example, there is a tendency in poverty reduction projects to introduce people to types of productive activities with which they are not familiar, rather than supporting them in further developing activities with which they are. Agricultural interventions often neglect people's existing farming activities in favour of what is more marketable, which may make sense for some people, but for many others does not.
•	Failure to understand the real constraints: it is not always easy to determine what it is marginalised people actually need in order to better their circumstances, and in particular to become more economically active. One reason for this is that, when undertaking consultations with target groups, a wish-list emerges that might not give an accurate idea of priorities, or might be more accurate for some people (e.g. those who are most vocal) than for others. It may also be the case that the government tends to introduce the programmes it believes it is capable of implementing, rather than those that are most needed. For whatever reason, there is a tendency to misidentify what interventions are most pressing to help people realise their economic potential. Such is almost certainly the case with the new R1 billion agricultural credit scheme, which has little prospect of making a difference, never mind becoming sustainable, largely because it is seeking to alleviate the wrong problem, or at any rate relieving it in the wrong way. Similarly, support to the small enterprise sector continues to be fixated on improving the flow of micro-credit, which might not be bad in itself, but is not the priority for most micro-entrepreneurs.
•	Neglect of lessons from the past: for reasons that are not clear, the government sometimes fails to learn from its own valuable experience. A case in point is the EPWP. Prior to the introduction of the EPWP, there was a diverse group of public works initiatives, which provided a valuable opportunity to identify how such interventions can be designed to have maximum impact on the poor in the particular context of South Africa. However, the design of the EPWP does not consider these lessons sufficiently. A second example is the Land Redistribution for Agricultural Development programme, which sought to fix the deficiencies of the previous approach to land redistribution without making a rigorous effort to understand what the actual problems were.

None of this is to suggest that these sectoral initiatives are not worthwhile or accomplishments in their own right, but that they tend to fall well short of what might have been hoped of them.

What the analysis of these programmes as a group tells us is in fact quite different. Arguably the most sweeping conclusion is that there is a need to have realistic expectations relative to the resources and efforts that are brought to bear. The EPWP provides a very specific instance of this, as mentioned above, but here we mean rather that the objective of transforming the second economy (as mooted by President Mbeki in the 2005 State of the Nation Address) can only meaningfully be understood as a hugely ambitious, long-term enterprise. The point can be argued in various ways. First, as far as the pattern of government expenditure is concerned, the balance between spending on welfare, the social wage and infrastructure and spending on second-economy interventions is so vastly in favour of the former that, whatever the underlying reason, it is unimaginable that it can be significantly altered quickly. Secondly, with at least one notable exception, the present array of second-economy interventions does not appear to form a solid basis from which to plan a massive increase in such interventions. The exception is public works, as we have seen, but only if they are reoriented to respond to the nature of South Africa's unemployment problem.

Thirdly, arguably the most important constraint on transforming the second economy is that the vast majority of those who inhabit this economy are manifestly unprepared to make the transition to the first economy or to link to it somehow. This, after all, is why a number of the government's second-economy interventions have not had more impact – explicitly or otherwise, they targeted those in the second economy who are relatively entrepreneurial, or bankable, or experienced in commercial agriculture, and sought to help them grow. While these individuals are certainly deserving of support, they are few in number compared to all the other micro-entrepreneurs and small-scale farmers whose prospects of growing are slim. (Nor does the idea of two million households becoming full-time commercial farmers make much sense in a country that at present can support only 40 000 to 50 000 of these.)

The point is not that second-economy interventions are not worthwhile or should not be attempted on a much larger scale. On the contrary, the point is very nearly the opposite. However, to the extent that the state can make an impact on the second economy in the short and medium term, it is not in terms of transforming the second economy, that is, making it disappear, but in making it somewhat more vibrant and broad-based. Another way of saying this is to caution against taking too much to heart the parallel that has been drawn in policy circles between the second economy on the one hand, and a war-damaged country standing to benefit from massive foreign assistance, e.g. via the Marshall Plan, on the other. As stated by the US Marshall Foundation (quoted in ANC Today, 22 October, 2004), ‘The [Marshall Plan] … worked because it was aimed at aiding a well-educated, industrialised people temporarily down but not out’. The same cannot be said of efforts to boost South Africa's second economy – the damage done to it was over such a long period of time, and so far-reaching, that there is effectively no better, prior state to which it can be restored. Restoration is relatively easy; transformation is far more difficult. This is not a simple conceptual hurdle, especially given that it could easily be caricatured as accepting that the poor must remain poor, albeit marginally less so. However, the evidence points to the conclusion that long-term transformation of the second economy must begin with much more aggressive interventions aimed at assisting large numbers of people to perform better, however modestly, within their existing economic potential.

7. Conclusion

This synthesis of the Report has sought to understand what accounts for underdevelopment in South Africa's second economy, to reflect on current efforts to intervene in this economy, especially those of the government, and to comment on what, if anything, might be done differently in order to meet this challenge.

The historical explanation for South Africa's particularly vicious case of dualism is well understood. What may have been underestimated, however, is how difficult it is to undo this legacy. The government has managed to steer a course that can reasonably be expected to lead to future, broadly shared prosperity. However, it might well take decades to get there. In the meantime, the state has put in place a social security system that can only be the envy of the poor living in many other developing countries. Yet this, too, has its limitations. There is, therefore, a clear rationale for interventions in what is now called the second economy. The government has taken positive steps to launch a number of such interventions, and it has correctly recognised that, although one might wish it to be otherwise, the burden is indeed upon the government.

Part of the purpose of the 2005 Report is to reflect on how well these interventions are going and to suggest a way forward. A sober examination of the government's efforts leads to the conclusion that, despite some successes and useful experiences, it has not yet achieved impact, especially if the increased number of poor is considered. Apart from the apparent absence of a coherent, scale-appropriate strategy for the second economy, a fundamental shortcoming of current efforts is a tendency to design assistance in a way that does not suit the ordinary poor person.

Indications that the government is contemplating a more interventionist role in the economy, including the second economy, are therefore very welcome. But they are also cause for concern. Is there an adequate awareness of the challenge ahead? What are the short- and long-term expectations of interventions in the second economy?

The key conclusion of this Report is that actual transformation of the second economy can only reasonably be understood as a long-term goal. The much more immediate task is to intervene broadly in the second economy in order to help as many people as possible to further realise, incrementally, their economic potential. This will not result in a quick, dramatic alleviation of underdevelopment and poverty. However, it will make an important difference, and it will lay the basis for the longer-term project of eliminating the second economy altogether.

Notes

This paper is a synthesis of the 2005 Development Report published by the DBSA in July 2005. It was prepared by Dr Michael Aliber (Human Sciences Research Council), assisted by Rasigan Maharajh (Institute for Economic Research), Marié Kirsten (Development Bank Southern Africa), Josephilda Nhlapo-Hlope (Development Bank Southern Africa) and Oupa Nkoane (Development Bank Southern Africa). Other authors of the development report upon which this article is based include Glynn Davies, Daniel Kekana, Professor Ben Turok, Nick Vink, Julian May, Anna McCord and Mark Oranje. It is available from the Knowledge Centre at the DBSA, www.dbsa.org/