South African Economy

Peter Brixen, Finn Tarp

• Publisher: Routledge
• Year: 2005
• Print ISBN:
  9780415142601
• Edition: 1
• eBook ISBN:
  9781134765416
• Topic: Business
• Subtopic: Business General
• Index: Business

What are the macroeconomic prospects for South Africa until the new millennium? Two methods of macroeconomic modelling, associated with the World Bank and IMF, are used here to generate three scenarios, based on moderately optimistic projections. The methodology used can be applied to other developing countries.

1

INTRODUCTION

The balance between the use of market forces and instruments of economic planning by the government in furthering the development process in third world countries has moved decisively in the direction of the former approach during the past 15–20 years. Yet the conditions under which unfettered market forces can provide for an optimal allocation of resources are exigent. This is the case when reference is made to a particular point in time, but even more so when alternative development paths for the future are considered. Furthermore, it is generally accepted that governments must assume responsibility for macroeconomic management and stability. Hence, governments continue to confront a series of routine and strategic planning issues, related, for example, to the need for systematic fiscal management and expenditure budgeting, on the one side, and the design of medium and longer-term production, investment and trade policies, on the other.

It is a basic theme of this study that formal macroeconomic models can play an important role in helping governments concerned with economic management and development planning. Economics is certainly not all that matters, when strategies for the future are analysed and designed in developing countries. Economic and non-economic factors are continuously interacting, at times reinforcing and at times contradicting each other. Nevertheless, economic analysis and modelling can inter alia help in (a) testing the internal consistency among established goals and policy instruments, (b) assessing trade-offs in light of underlying assumptions and value premises, and (c) defining a research agenda for the collection of further information on relations and mechanisms, which have so far been poorly understood. It can, moreover, be added that quantified modelling has taken a giant step forward due to the technological progress inherent in the computer revolution of the last 15 years. How such advances can be put into use in teaching and applied country economic work is another major concern of this volume.

The topic of the optimal size of the public sector is a difficult and complex one, as witnessed by the intense debate on development issues during the 1980s and early 1990s (Toye, 1993). Nevertheless, an attempt is made throughout this book to distinguish between this theme and questions related to the desirability of macroeconomic stability as measured by standard indicators such as the budget deficit, the inflation rate and the balance of payments. Public sectors can certainly over-expand, and in the past have frequently done so, causing economic instability. Yet, whether public expansion is possible without crowding out private activity and causing macroeconomic problems is an empirical question, which needs to be analysed rather than assumed. If space is, in reality, available, or can be created through government and private sector activity, it had better be used actively, provided of course that government is indeed committed to furthering the development process. The economic, social and political opportunity costs of not doing so may turn out to be very high in countries plagued by poverty and inequality. This thesis is particularly relevant to South Africa, which is the country case we are examining. As background, South Africa’s economic and social characteristics are reviewed in Chapter 2, which also sets out in some detail the competing strategic policy frameworks, which influence policy making.

It is an explicit aim of this study to provide a fully quantified macroeconomic input to the policy dialogue on the future prospects for the South African economy, including inter alia the important tasks of constructing a medium-term expenditure framework. Accordingly, the particular focus maintained in the research presented is macroeconomic, although the importance of sector and micro level policies is duly recognized. What we have done is to construct three fully quantified and documented six-year medium-term scenarios in Chapters 6 and 7. It is believed that they capture essential differences between the many possible directions which the South African economy may take, and they can each be seen as representative of a much larger sub-set of specific development paths. The need for developing such projections has been alluded to on many occasions by both South African policy makers and scholars, and outside agencies such as the International Monetary Fund (IMF) and the World Bank have also directed attention to the importance of this area of analysis in the context of policy formulation in South Africa.

While the South African case is in focus in Chapters 2, 6 and 7, the organization of the data and the methodologies followed are easily replicated in other country contexts. Moreover, the scenarios generated have not been constructed on the basis of entirely new theoretical constructs. The approach taken in the choice of modelling frameworks is the pragmatic one, that while aggregate models and quantified projections are indispensable, the theoretical frameworks do not have to be overly complicated.

In deciding which modelling frameworks to use, account was also taken of the fact that the South African government has to deal with a range of influential international agencies, including in particular the World Bank and the IMF. The two Bretton Woods institutions, which have a direct impact through the disbursement of their own financial resources as well as indirect power in affecting private and bilateral capital flows, use a variety of analytical tools in their policy-advisory work. Nonetheless, the so-called Financial Programming (FP) framework of the IMF and the Revised Minimum Standard Model (RMSM) of the World Bank have been important and widely used economy-wide tools in country economic analysis. Furthermore, they illustrate the type of economic reasoning and fundamental perceptions of macroeconomic causality upon which policy recommendations and conditionalities of the two institutions are often based. To master the analytical constructs mentioned is therefore an important precondition for making sure that the policy dialogue gets on a sounder footing than is often reported to have been the case elsewhere in Africa during the 1980s (Mosley et al., 1991).

Hence, despite their critical theoretical shortcomings, which are reviewed in detail in Tarp (1993a), there appears to be plenty of reason to become well acquainted with the FP and RMSM frameworks – even more so, because the two models can actually be used as macroeconomic accounting frameworks in a more flexible manner than often assumed. Finally, they certainly do meet the practical requirement of being relatively simple theoretical constructs. Yet, surprisingly little documentation is actually available in published sources on the IMF and World Bank models, and the same goes even more so for guidelines on how best to implement and use them in specific country cases. In addition, the FP approach and the RMSM are seldom – if ever – implemented simultaneously in a consistent manner even if this is, in principle, fairly easy to do.

To help fill the above gaps in a functional and reproducible manner, so that additional experiments and updating as well as training can be performed with relative ease and effectiveness, is another set of objectives of this volume. It therefore contains three methodological chapters in addition to the chapters dealing more directly with South Africa. Chapter 4 describes and discusses the modelling frameworks used, and the particular modifications made in order to ensure that relevant, compatible and consistent medium-term scenarios could be constructed are also explained. Three types of data were required for the simulation exercises, including a complete data set for a base year. Chapter 3 documents and reviews how these data were generated, and how they fit into a standard national accounting framework. Chapter 5, on the other hand, contains a number of policy experiments, which are performed in order to investigate the main economic mechanisms of the two models.

The detailed base run, which is subsequently developed in Chapter 6, is based on the assumption that it is justified to be moderately optimistic about the economic future of South Africa. The modelling frameworks presented in Chapter 4 are used to ensure consistency among the economic variables considered and the various qualitative assessments made. Since the base run represents what appears as a realistic prognosis, it is used as a benchmark in Chapter 7, where two alternative scenarios are developed. They are based respectively on a more optimistic and a more pessimistic assessment of future perspectives, and demonstrate how two different sets of changes in exogenous variables and parameters can affect the values of the endogenous variables.

Finally both the Financial Programming framework and the Revised Minimum Standard Model were programmed and solved using the General Algebraic Modeling System (GAMS) (Brooke et al., 1988). Appendix A contains the two GAMS programes used to generate the base run and the scenarios, as well as some methodological background. Appendices B, C and D provide detailed statistical tables on the base run and the optimistic and pessimistic scenarios, respectively.

2

GENERAL COUNTRY BACKGROUND

MACROECONOMIC FEATURES AND TRENDS

Based on average income measures, South Africa is a relatively rich country with a Gross National Product (GNP) per capita of close to US$ 3,000 (World Bank, 1995, p. 163). Thus, South Africa belongs to the World Bank category of upper-middle-income countries. Furthermore, South Africa is indeed a regional ‘giant’ with a GNP which is about three times as big as that of the Southern Africa Development Community (SADC) countries together. Nevertheless, if the average annual income of the black South Africans in the former so-called homelands is used as a measure, South Africa would be categorized among the poorest of the low-income countries. Moreover, while the GNP is large by African standards, total GNP is after all only equal to about 11 per cent of, for example, the GNP of the United Kingdom.

In terms of sectoral contributions to Gross Domestic Product (GDP), industry (including mining and quarrying) accounted for 39 per cent in 1993 according to World Bank data. Services such as finance, insurance, real estate and business services, public administration and defence as well as trade, etc. added up to a total of 56 per cent. Agriculture, forestry and fisheries made a relatively small contribution of around 5 per cent to GDP, but mining and agriculture are much more important to the economy than GDP figures suggest. These two sectors dominate the country’s exports and provide essential inputs for manufacturing. Close to 60 per cent of agricultural output is, for example, delivered to secondary industries for processing, and agriculture is also a major employment generator.

Exports of goods and non-factor services account for around one quarter of GNP, and gold is by far the most important export commodity, accounting in 1991 for almost one quarter of all exports. Other mineral products, base metals and chemicals amounted to more than 30 per cent, leaving about one-third for other commodities, including for example food, drink and tobacco as well as textiles. Machinery has been and remains the key import, followed by transport equipment and chemicals. Thus, the South African economy is open as well as highly dependent on fluctuating primary sector exports, and imported capital goods for the domestic industry are essential for the level of activity.

Traditionally, South Africa attracted substantial capital inflows, and international debt rose rapidly between 1970 and 1985. Yet, after 1985 the imposition of international financial sanctions forced the apartheid regime to run current account surpluses to finance external debt repayment and private capital outflows. Consequently, debt started falling to a historic low in the early 1990s.

The South African economy was dynamic during the 1950s and 1960s with annual growth rates in excess of 5 per cent. However, from 1973–4 stagnation and decline became characteristic. The growth rate started falling and by the end of the 1980s it was less than 2 per cent a year, and much more volatile than previously. In 1990 per capita output had actually fallen to only 85 per cent of the level reached in 1975, and GDP continued its downward trend until 1993. Per capita income in 1994 was therefore about the same as in the mid-1960s.

Moreover, during the 1970s and 1980s the composition of expenditure changed towards greater private consumption and less fixed investment. Hence, while fixed investment grew by 8–9 per cent a year during the 1960s, the increase was limited to only 2.2 per cent a year from 1972 to 1979, and during the 1980s investment fell by 1.5 per cent on an annual basis. Private consumption, on the other hand, maintained an annual rate of increase of 3 per cent during the 1980s. Thus, the capital stock became smaller and ‘older’, and consumption grew more than GDP, so the basis for future growth was gradually undermined.

All employment and labour statistics in South Africa must be interpreted with caution, but it is evident that South Africa is in the middle of a deepening and potentially explosive labour market crisis. The number of formal sector employees has stagnated since the mid-1970s, and while 80 per cent of new entrants on the labour market found formal sector employment at the end of the 1960s, the number is now less than 10 per cent (COSATU, 1992). The only formal sub-sectors in which employment increased from 1980 to 1988 were central government, local administration and homelands, and financial services. Thus, the build up of apartheid’s administrative control structures created more jobs, but this increase was cancelled out by less employment in other sectors, and total employment continued to fall steadily from 1989 to 1994 (SARB, September 1995). It is difficult to provide more precise estimates of unemployment, as it is unclear how many people have found at least part time employment in the informal sector. Yet, according to official statistics, 43 per cent of the economically active work force were either unemployed or occupied in the informal sector in 1994 as compared to 19 per cent in 1970 (SARB, September 1995). This obviously underlines the need not only for growth, but also for sector and micro level policies to increase labour absorption.

Public sector budgets have been in deficit ever since 1970. The increase in the number of public employees referred to above is one aspect hereof, but to this should be added increasing debt service obligations as well as military expenditures. The latter have fallen more recently, but nevertheless rose from 2 per cent of GDP in 1970 to 4 per cent during the 1980s. This, inter alia, led to a particularly sharp cut back in public investment. Hence, public expenditures, geared towards the realization of the national and international political and military objectives of the apartheid regime, contributed to undermining the basis for future growth and development. The levels of public expenditure and the tax burden are not by themselves worrisome, but the uncontrolled growth in the public sector and a public sector deficit, which grew rapidly to 8.9 per cent of GDP in 1992–3, certainly contributed to the relatively high inflation (Gelb, 1991). South Africa joined much of the rest of the world in experiencing a jump in the inflation rate from 4 per cent to around 15 per cent a year during the period 1960–73. Yet, contrary to what happened elsewhere, it remained at this level until very recently.

SOCIO-ECONOMIC CHARACTERISTICS

It is not only the macroeconomic situation and trends which have caused concern in South Africa (COSATU, 1992). Poverty is widespread, and South Africa has one of the most unequal distributions of income in the world with a gini-coefficient of around 0.65 (Eckert, 1991). Furthermore, the high-income group is relatively large compared to other developing countries (i.e. 20–25 per cent as compared to normally only a few per cent). Thus, much less is left for the poorest, and the number of absolute poor increased from around 15 million in 1980 to over 17 million in 1990 (Tarp, 1993b). Poverty is, in addition, overwhelmingly concentrated among African people and is both more widespread and more severe in rural areas.

It is generally held that inter-racial income inequalities, which are particularly pronounced in South Africa, were reduced as a consequence of deregulations in the labour market over the past decade. However, intra-racial inequalities grew, according to some sources, especially among Africans. It would appear that while modest income gains have been achieved among the middle 60 per cent of the black income scale, the top 20 per cent have experienced very rapid gains while the poorest 20 per cent have had no, or even negative, income growth. Thus, the poorest groups among the blacks have been marginalized, and widening income gaps between formal and informal workers constitute a growing distributional dilemma.

Income inequalities are partly a legacy of race-determined intervention in wage-formation, and the fact that many jobs have been exclusively reserved for whites. To this is added, however, the extremely unequal distribution of income generating assets. Thus, the ‘gini-coefficient’ for agricultural land ownership is over 0.80, and the white share of property income around 75 per cent. Also the extreme degree of concentration in the manufacturing industry is relevant here. Four private conglomerates (Anglo-American, Rembrandt, South African Mutual and Sanlam) control more than 80 per cent of the companies listed on the Johannesburg Stock Exchange, which is the financial and commercial power centre of the country.

Discriminator...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Table of Contents
• Figures
• Tables and boxes
• Preface
• Acknowledgements
• Abbreviations
• 1 INTRODUCTION
• 2 GENERAL COUNTRY BACKGROUND
• 3 ACCOUNTING FRAMEWORK AND BASE YEAR DATA
• 4 MODELLING FRAMEWORKS
• 5 MODEL EXPERIMENTS
• 6 BASE RUN
• 7 ALTERNATIVE SCENARIOS
• 8 CONCLUSION
• APPENDIX A: MODELLING FRAMEWORKS IN GAMS
• APPENDIX B: BASE RUN
• APPENDIX C: OPTIMISTIC SCENARIO
• APPENDIX D: PESSIMISTIC SCENARIO
• Bibliography
• Index