APPENDIX 1

A REVIEW OF ECONOMIC DEVELOPMENTS SINCE 1994

Recent Macroeconomic Performance

The growth performance of the South African economy has improved considerably over the past two years. Following three years of economic decline, real gross domestic product rose by 1.3 percent in 1993, 2.7 percent in 1994 and 3.3 percent in 1995. This stronger growth occurred in all secondary and tertiary sectors, although agriculture and mining experienced declines in 1995.

A marked increase in real fixed investment has been a welcome feature of the upswing. Private consumption expenditure also firmed as incomes rose and consumer confidence improved. In contrast, real government consumption expenditure, which increased sharply until 1994, levelled off in 1995. Export volumes stagnated in 1994 and then rose by an impressive 8 percent in 1995, while real imports have increased at an annual rate of 16 percent since 1993.

Inflation in 1995 declined to levels last seen in the early 1970s. The annual rate of increase in consumer prices has been below 10 percent for three years. Positive real interest rates and containment of money supply growth have been important factors in combating inflation.

Formal sector employment has responded sluggishly to the economic recovery. Formal non-agricultural employment declined throughout the period 1989 to 1994 and appears to have increased by little more than 50 000 during 1995. However, employment in unregulated sectors of the labour market appears to be on an upward trend.

The post-election period has seen considerable movement in the balance of payments. A net capital inflow amounting to R31 billion was recorded between mid-1994 and the end of 1995. The bulk of this inflow has been portfolio investment, which is comparatively mobile and sensitive to shifts in market perceptions. The balance on current account moved into deficit over this period as imports, particularly capital goods, rose strongly.

Since mid-February the foreign exchange market has been subjected to intense speculative pressure, causing a substantial real depreciation of the rand. This development to some extent reflects that the rand had become somewhat overvalued in response to a temporary capital surge, but was also the result of increased uncertainty regarding policy trends and economic prospects.

Fiscal Policy

In response to the unsustainable fiscal situation that had developed by 1992/93, when the overall deficit reached 9.0 percent of GDP, fiscal policy has been informed by the following goals:

• to cut the overall budget deficit and the level of government dissaving;
• to avoid permanent increases in the overall tax burden;
• to reduce general government consumption expenditure relative to GDP; and
• to strengthen the general government contribution to gross domestic fixed investment.

Although government dissaving has not yet been eliminated, progress has been made in this respect. The cash-flow adjusted exchequer deficit was reduced to an estimated 5.4 percent in 1995/96. Consolidated general government tax revenue has increased from 25.6 percent to 26.8 percent of GDP between 1992/93 and 1994/95, but still somewhat below the 1989/90 level of 27.1 percent. At the same time, significant shifts in the allocation of expenditure have been effected in accordance with reconstruction and development priorities.

The improved discipline in fiscal policy has contributed to greater financial stability and lower inflation. It has also enhanced local and international business confidence, contributing to capital inflows and investment. A gradual approach to deficit reduction has been necessary to prevent disruptions in the provision of essential public services, but has also had the advantage of avoiding a major contraction in aggregate demand.

It is widely acknowledged that the overall tax burden is high and that personal income tax rates, in particular, lead to excessive distortions in business and household behaviour. A range of tax reforms have been implemented over the past two years.

Tax issues and fiscal relations between the different tiers of government are subjects of major investigations, finalisation of which should lead to greater fiscal certainty and business confidence. The ongoing reform of the budgetary process introduced by the Department of State Expenditure will contribute to more effective planning and expenditure management.

Inflation and Monetary Policy

The basic objective of monetary policy has been to protect the value of the rand. The Reserve Bank operates within a framework in which effective control over the money supply serves as an intermediate objective. Annual guidelines for growth in broadly defined monetary aggregates are set, aimed at reducing the inflation rate.

Following a period of negative real interest rates in 1986/87, monetary policy has been kept relatively tight. A consistent anti-inflationary stance was maintained throughout the 1989-1993 cyclical downturn, bringing broad money supply growth down from 28 percent to single-digit levels. The lower rates of increase in monetary aggregates, with sustained surpluses on the current account, enabled nominal interest rates to decline. The Bank rate was reduced in steps from a maximum of 18 percent in 1989 to a low point of 12 percent in late 1993.

The acceleration in money supply and credit growth in 1994 called for tighter policies to prevent a resurgence of inflation. In September 1994 the Bank rate was raised by 1 percent, followed by three further increases of the same order to bring the rate to 16 percent by April 1996.

Inflation reduction has been facilitated by other developments. Trade liberalisation has contributed significantly to the containment of domestic prices, while more moderate wage-setting and improved industrial relations have also played a role in holding cost increases in check.

The consistent application of monetary policy and the maintenance of positive real interest rates over the past few years have not only brought inflation down to single-digit levels, but have also, with the exception of recent developments, reduced the volatility of a number of financial variables including interest rates. This has facilitated business planning and reduced the intensity of cyclical movements in the economy.

Exchange Rate Developments

The current approach to exchange rate policy reflects the view that the exchange rate of the rand is best protected through consistent application of anti-inflation policies as this brings stability to the average value of the rand against the currencies of trading partners.

The monetary authorities have not attempted to fix a specific exchange rate of the rand. Because the variables which impact on the balance of payments are continuously in flux, the rate must be flexible. The Reserve Bank can to some extent smooth erratic movements in the exchange rate, but it cannot override fundamental market perceptions. Nevertheless, an unsustainable appreciation can be countered by appropriately relaxing exchange controls and accelerating the Reserve Bank's withdrawal from the forward exchange market.

Although the exchange rate is primarily market determined, its value at any moment cannot be considered a true reflection of the underlying value of the rand while exchange controls exist. The government has stated repeatedly that it is committed to phasing out controls in a prudent manner. In line with this commitment, the financial rand was abolished in 1995.

To maintain orderly conditions in the foreign exchange market, it is necessary to retain an adequate stock of foreign exchange reserves. Large capital inflows since mid-1994 enabled the Bank to increase the net gold and other foreign reserves by R9.1 billion during 1995, thus countering somewhat the tendency of the rand to appreciate in real terms.

In view of shortcomings in the forward exchange market, the Reserve Bank has been an active market-maker. While this has led to a large imbalance in the forward book, considerable progress has been made since 1994 in reducing the Bank's exposure. The combination of more market-related forward rates, interest rates and spot exchange rates has encouraged private parties to develop the forward cover market. This remains true although the Reserve Bank's involvement in that market has increased again since the depreciation of the rand.

Trade and Industrial Policies

The unreliability of raw materials exports in the 1980s persuaded policy-makers that the central thrust of trade and industrial policy had to be the pursuit of employment-creating international competitiveness. This entails a shift away from demand-side interventions, such as tariffs and subsidies, which raised prices received by producers, to supply-side measures designed to lower unit costs and expedite progress up the value chain.

While long-term survival strategies have had to be developed for certain sensitive sectors, general progress towards an outward-oriented stance is reflected in a number of achievements:

• substitution of former quantitative restrictions with tariffs;
• rationalisation of the tariff structure by almost halving the number of tariff lines;
• abolition of import surcharges, completed in October 1995;
• phasing down of tariffs, begun in 1995, by on average one-third over five years; and
• phasing out of the general export incentive scheme, to be completed by the end of 1997.

Another critical policy thrust has been the expansion of market access through preferential trade arrangements with industrial countries and pursuit of regional economic integration. The following developments deserve specific mention:

the signing of the Marrakesh Agreement has secured membership of the World Trade Organisation and continued "most favoured nation" access to member country markets;

the granting of Generalised System of Preferences status by various developed countries (the USA, the European Union, Japan, Canada, Norway and Hungary) has brought preferential access to their markets; and

the signing of bilateral trade treaties with several African countries and rapidly growing trade with neigbours has seen exports from within the Southern African Customs Union area to the rest of the Southern African region quadruple.

In the area of concessionary industrial finance, several schemes have been introduced by the Industrial Development Corporation (IDC), to meet strategic industrial objectives, including:

• the JOBS scheme for small enterprises satisfying certain employment criteria;
• the LIFE scheme for ventures which satisfy certain export performance criteria; and
• the Multi-Shift scheme for existing concerns which introduce an additional shift.

Although the Regional Industrial Development Programme (RIDP) is under review, it was expanded in 1993 to include a simplified scheme applicable to smaller enterprises. The phasing in of supply-side support measures has begun recently, with the introduction of the IDC's World Player scheme for the modernisation of plant and equipment in sensitive industries facing a significant decline in protection. The development of a full package of supply-side measures is the central focus of medium-term industrial and trade strategy.

The impact of trade restructuring is not easily measured. Many firms have been under intense pressure, compounded by the real appreciation of the exchange rate in 1995, and will require additional investment incentives to meet international competition. Nevertheless, manufacturing employment and exports have increased, taking advantage of the international cyclical upswing. The most positive sign, also evident in sectors sensitive to the lowering of trade barriers, has been the significant increase in new foreign and domestic fixed investment in the manufacturing sector.

Employment and Labour Trends

South Africa's labour market is extremely fragmented. Employment growth in the formal sectors of the economy has stagnated over the past decade and private sector employment has fallen. It is apparent that unregulated low wage employment has increased significantly since the 1970s, now accounting for perhaps a third of all job opportunities. In addition, a large pool of unemployed men and women who earn no income or derive sporadic earnings from informal self-employment make up about a third of the potential labour force.

Irregular, sub-contracted, out-sourced or part-time employment on semi-formal contractual terms is becoming the preferred source of labour for many employers. This is resulting in a growing gap between the wages and benefits in the regulated and unregulated parts of the labour market. Where regulations raise the costs of job creation, employers turn to unregulated forms of employment.

The following additional features of the labour market may be noted:

• spatial imbalances between supply and demand for labour;
• inordinate distances between places of residence and work;
• poor integration of education and training into job entry and career planning; and
• diversity in technology, skills and productivity across regions and sectors.

The major development in the primary segment of the labour market over the past two years has been the new Labour Relations Act. This has four key features. It establishes a single industrial relations system for all employees, promotes collective bargaining by providing certain organisational rights for trade unions, establishes new procedures and institutions for resolution of disputes and provides for workplace forums to facilitate a shift from conflictual employer-employee relations towards joint problem-solving with employee participation. The reduced incidence of industrial unrest in recent years attests to the considerable progress made in this regard.

What are the implications of labour market trends for income distribution? Over the past two decades, black wage income in the non-primary sectors of the economy has increased by 4.5 percent per year in real terms, mainly due to rising average wages. Though many households have been lifted above the poverty line, the trend is shifting in favour of higher income classes, where education, training and experience contribute to sustained improvements in productivity. Employment growth has contributed just 1.2 percent per year to the increase in black wage income over the past twenty years, and has been virtually zero over the past decade. The slow pace of regulated formal employment growth has resulted in limited benefits of growth reaching the poor. Though the unemployed are in some cases attached to households where other members have access to productive employment, in many of the poorest households the main breadwinner cannot find a job. Whereas the average household received R1 360 per month from wages in 1993, the poorest fifth received R47 per month from regular employment, mainly because so few of their work-seekers had jobs.

Restructuring the Public Sector

The process of administrative restructuring of the public service provided for in the Constitution gathered pace in 1995. With the first phase of the process, involving the integration of the public service at national and provincial levels, nearing completion, attention is shifting to the longer-term issue of creating a more cost-effective service.

A major step was taken in early 1996 with the devolution to line departments of all career-related personnel functions. The restructured Public Service Commission will retain a research and monitoring role, while the Department of the Public Service and Administration will be responsible for broad human resource policy, conditions of service and labour relations.

Agreement was recently reached on a three-year conditions of service adjustment package. Inter-related elements include the right-sizing of the public service, a voluntary severance package and simplification of occupational remuneration schedules. Substantial improvements in wages in 1996 are envisaged for certain categories of civil servants, redressing imbalances within the service and relative to the private sector. Improvements in subsequent years will depend on the overall reduction in personnel and affordability considerations.

Government has initiated a review of state assets, with a view to restructuring their ownership and governance. An agreement has been signed with organised labour which sets guidelines for further consultation. A process of enhancing the governance of public enterprises is under way, with the first step covering policy regarding dividend payments to the fiscus. Revised conditions for sovereign guarantees have also been issued which reduce fiscal exposure and extend the discipline of the market to public sector institutions.

Progress has been made over the past year regarding the restructuring of development finance institutions. A strong emphasis has been placed on ensuring financial sustainability. It is envisaged that grant components of project finance should be channeled through the fiscus in a transparent manner and that loan terms should be market-related.

Social Policy

The past two years have witnessed an energetic revision of social policies. Although implementation of these initiatives has been uneven, policy development is now sufficiently advanced to expect significant delivery across a broad front in the coming years. A few key linkages between growth and redistribution and new policy directions are highlighted below.

Progress in education shows up consistently in comparative studies as a key determinant of long-run economic performance and income redistribution. Sustained improvements in the quality of public schooling available to the poor and greater equity in the flow of students through secondary and tertiary education are central to the Government's approach. Despite near-universal enrolment in primary education, only some 40 per cent of children currently complete secondary schooling successfully. Inadequate pass rates in science and mathematics are a cause for concern.

Reform initiatives underway aimed at qualitative improvements in the educational system include restructuring and decentralising of school governance and management, overhauling school curricula, establishing a national qualifications framework, addressing the culture of learning in schools, building and refurbishment of classrooms, rationalising and renewing teacher education, enhancing educational administration, and expanding further education. Suitable norms, together with quality enhancing rewards, are under review. With spending on education at nearly 7 percent of GDP there is a need to contain expenditure through reductions in subsidisation of more expensive parts of the system and greater private sector involvement in higher education. This will concentrate public resources on enhancing the educational opportunities of historically disadvantaged communities.

The systematic restructuring of health services, with strong emphasis on universal and free access to comprehensive primary care, represents a clear commitment to improving the health conditions of the poor. Within the public health system resources are shifting from tertiary services in metropolitan areas towards overcoming the inadequacies of hospitals and clinics in rural areas and townships.

Partnerships between the state and voluntary organisations centred on developmental welfare services will focus attention on the vulnerable, especially in under-serviced areas, while freeing resources from expensive institutionally based services. By far the greater part of welfare spending is devoted to social grants, which assist some 3 million elderly or disabled persons or needy children. These transfers play a vital role in poverty alleviation, especially in rural areas. Affordable alternatives to support families and children in need are being investigated.

Although the implementation of the housing and infrastructure programmes have been slow, there have been continuous refinements to the policy framework. Since late 1995, a marked acceleration in housing delivery has been evident. A continuation of this trend will see the provision of housing and related services on a substantial scale in subsequent years. This will have several beneficial distributional effects. Construction is largely labour intensive and provides jobs and training, while improvements in housing and infrastructure enhance the productivity of labour and the quality of urban life.

Improved water and sanitation is typically the first priority of rural communities. Some 500 projects costing R1.5 billion have been committed. Rapid progress with the supply of potable water to the 12 million people without adequate access will be a major contribution to poverty relief. These initiatives have been complemented by new policies regarding sanitation systems.

Enhanced crime prevention has been identified as one of the key pillars on which growth and development rests. The national crime prevention strategy, which will involve considerable coordination with other departments and includes several projects, such as community policing stations, will have a strong redistributive thrust, as the poor are hardest hit by violent and property-related crimes. The emphasis on community policing in the new approach recognises the crucial link between crime prevention and development at the local level.

The land reform programme, combining asset redistribution with enhancement of tenure, has an important role in improving the long-term prospects for employment and income generation in the rural economy. Progress has been made with finalising procedures for the rapid release of land and the introduction of a settlement grant. Complementary initiatives include emergent farmer support programmes. As these gain momentum, emphasis will shift to marketing support, appropriate technological interventions and streamlined extension services.

Growth, Employment and Redistribution: A Summary

As a result of political stability and sound policies, economic growth has revived, bringing to an end the stagnation that characterised the 1980s. Most commentators are confident of a return to a growth trend somewhat in excess of population growth, ensuring a gradual improvement in living standards. The higher growth path is linked to an improved balance of payments situation, which eases the major constraint on medium term economic growth. The last few years have also witnessed success in the fight against inflation as a result of effective monetary policies and more moderate wage settlements.

With regard to the challenge around redistribution, the past years have seen the introduction and extension of policies aimed at more equity in access to social services. These have important redistributive effects, and in the case of education and training, land reform and urban development, will also have favourable consequences for economic growth in the longer term. There has been some redistribution of wage income over the past two decades, partly as a result of union pressure. A continuation and possible acceleration of this process has been made possible by increases in productivity. Improvements in education, skills, labour effort and management are now crucial for wage increases and are increasingly being addressed in industrial negotiations.

Higher economic growth has not yet been translated into significant employment creation. This is to some extent to be expected, as many firms began the upswing with substantial unused capacity and economic restructuring aimed at improved international competitiveness has required cost reductions, including some shedding of labour. However, in the absence of rapid employment growth, unemployment has increased, so that the distribution of income exhibits an increasing dichotomy between those inside the formal economy and those on its edges. Accelerated economic growth associated with stronger employment creation is the key to continued progress towards an equitable distribution of income and improved standards of living for all.

APPENDIX 2

AN INTERPRETATION OF THE RECENT DEPRECIATION

Since mid-February the foreign exchange market has been the focus of intense speculative pressure, leading to a significant real depreciation of the rand. These developments were to some extent the result of unfounded rumours and increased uncertainty over the direction of Government economic policy, but also reflected the fact that the exchange rate had become overvalued and that an adjustment was overdue.

The table and graph below place the depreciation in perspective. The February adjustment appears to have been a purchasing power parity correction which brought the real exchange rate back to its 1994 level. It is also clear that this level is not as low as those that prevailed in 1985.

During 1995 South Africa experienced a falling rate of inflation, which, coupled with continuing tight monetary policy, resulted in real interest rates that were high by international standards (8 percent vs 2 percent). These high rates helped to attract a net capital inflow in excess of R30bn. This inflow dominated the growing current account deficit, thus placing upward pressure on the nominal exchange rate. The Reserve Bank responded by preventing the nominal appreciation. In the course of 1995, gross reserves increased from R11 to R16 billion and net reserves from R6 to R16 billion.

The stable nominal exchange rate meant that the real rate was appreciating over this period. Such a strong rand did not, however, reflect the underlying fundamentals as the economic recovery was not sufficiently strong to justify a real appreciation and an inflation differential still persisted between South Africa and its major trading partners.

Market participants, lulled into accepting that the nominal rate would remain stable, maintained uncovered positions in the belief that the real interest rate differential would be enough to compensate holders of rand assets for any expected depreciation. The markets also displayed a false sense of complacency about the stability of the foreign exchange market. The fact that the rand market is relatively thin and therefore subject to a high degree of volatility appears to have been ignored. Moreover, the financial rand, which previously insulated the commercial rand from such volatile capital movements, no longer existed. This underlying reality was hidden during 1995 since it is relatively easy to contain fluctuations in the exchange rate when upward pressure on the currency coincides with the need to accumulate foreign reserves.

The February depreciation was finally triggered by unsubstantiated rumours concerning President Mandela's health and possible exchange control relaxation, as well as a rise in US bond rates. During the subsequent two weeks intervention in the spot market amounted to just over R1 billion. The Reserve Bank's efforts were, however, concentrated in the forward market where the oversold position increased by R5 billion. Although this eased the pressure on the spot market and the reserve position, it could create a fiscal burden in future.

The depreciation in April, which resulted in a further decline of the real exchange rate of approximately 9 percent, can be attributed to the following:

• rumours of Minister Liebenberg's resignation;
• rumours of an impending removal of exchange controls;
• perceptions of divergent views between business and labour; and
• perceptions of policy indecision within government.

The initial policy reaction to the April crisis included further intervention on the spot and forward markets as well as an increase in the Bank rate by one percentage point. Although the interest rate increase was not a welcome development from a growth perspective, it had the effect of assisting in stabilising the foreign exchange market.

How the exchange rate crisis is managed from here onwards is critical to South Africa's economic fortunes. The commercial banks have already reacted by increasing their lending rates by a further one percentage point. This will clearly have a more depressing effect on growth and employment in the remainder of 1996 and beyond. In the absence of a clear and credible macroeconomic policy framework the currency weakness will continue with further financial market reactions, raising the spectre of a fully-fledged recession with inflation accelerating and escalated capital flight.

However, the depreciation of the Rand also represents a significant window of opportunity. The integrated strategy is designed to take advantage of the implied shift in the prices of tradables and non-tradables, and to lay the basis for an export-led revival of the economy. But this will only occur if the depreciation is not dissipated through increased wages and prices and if it is supported by appropriate monetary, fiscal and trade policies - in short the integrated macroeconomic policy package.

APPENDIX 3

THE MACROECONOMIC OUTLOOK OF CURRENT POLICIES

In this appendix results are reported of simulation exercises based on current fiscal and sectoral policies utilising inputs from four macroeconomic models of the economy. The projections described here are derived from an adaptation of the Reserve Bank's econometric model, but are broadly consistent with results obtained using models of the Development Bank of Southern Africa, the Bureau for Economic Research and the World Bank.

Model characteristics

1. The base scenario assumes a decline of almost 10 percent in the real effective exchange rate in 1996. This implies a partial reversal of the exchange rate movements which have occurred in April 1996, with the rate averaging about R4.30 to the US$ during 1996. For the rest of the period 1996-2000 an almost constant real exchange rate is assumed.
2. The main fiscal assumptions in the base model are:

• A reduction in the budget deficit to 3 percent of GDP by the 2000/01 fiscal year, broadly consistent with current fiscal targets;
• A decline in real government consumption expenditure as a proportion of GDP to about 18 percent in the year 2000;
• A decline in tariff revenue relative to dutiable imports from 10 percent to 8 percent, broadly in keeping with the present commitments to the World Trade Organisation;
• An average increase in the government real wage of about 1 percent per year, based on the agreement reached in the Central Bargaining Chamber and projected inflation;
• Average real increases in general government investment and capital formation by public corporations of between 2 percent and 3 percent per year.

3. Real interest rates remain positive through the simulation period, but fall somewhat with a projected lower inflation environment in the second half of the period.
4. Modest increases in real wages in the private sector are anticipated throughout 1996-2000, associated with productivity improvements and relative stability in labour relations.

Key results

1. As is widely expected, GDP growth reaches a cyclical peak in 1996 of about 3.3 percent and then falls somewhat before recovering to about 3 percent again in 1999 and 2000. Growth is mainly the result of steady increases in fixed investment and manufactured exports.
2. The current account of the balance of payments remains in deficit throughout, but improves to around 1 percent of GDP as a result of relatively weak growth. Without any gains in competitiveness, substantial foreign direct investments are not likely and the current account deficit is financed largely by public borrowing.
3. Government consumption expenditure in the year 2000 is about 2 percent higher than in 1996 in real terms.
4. Inflation rises to nearly 11 percent in 1997 as a result of the exchange rate depreciation and falls slowly thereafter, consistent with adherence to fiscal and monetary policy guidelines.
5. Private saving remains fairly stable, averaging just under 21 percent of GDP, while government dissaving falls to below 1 percent of GDP in 2000.
6. With an estimated output elasticity of 0.4, formal non-agricultural employment growth averages 1 percent per year. Allowing for somewhat faster growth of informal and semi-formal employment (elasticity = 0.7), which accounts for 50 percent of total employment, annual employment creation averages 100,000 between 1996 and 2000 and unemployment rises from 35 percent to 38 percent. Nearly two-thirds of new jobs are non-formal.

In sum, the base scenario represents a stable macroeconomic trajectory in the presence of adherence to deficit reduction guidelines, but export and output growth are insufficient to reverse the worsening unemployment trend. The base model assumes that the deficit reduction programme will be adhered to, and envisages a reduction in real government consumption expenditure relative to GDP. The resulting reduction in government borrowing for consumption purposes represents an enhancement of the capacity of the domestic economy to finance private sector investment and somewhat improves long-run growth potential.

Base Scenario Projections: 1996-2000

APPENDIX 4

THE MACROECONOMIC OUTLOOK OF THE INTEGRATED STRATEGY

The simulation results reported here introduce several policy adjustments to the base scenario sketched above, illustrating the possible macroeconomic consequences of the proposed integrated strategy. The key growth drivers, namely the investment and export projections were tested for feasibility and consistency on all the models and subjected to econometric analysis. There is a strong redistributive thrust to the policy shifts envisaged which is not reflected here due to the macroeconomic nature of the modelling framework. Model characteristics

1. A real effective exchange rate depreciation of 8.5 percent, averaging R4.25 to the US$ for the full 1996 calendar year, and maintenance of this real rate over the 1996-2000 period.
2. A tighter fiscal stance, reducing the deficit to 4 percent of GDP in 1997/98 and falling to 3 percent by 1999/00, resulting in the virtual elimination of government dissaving by 2000 and a deficit equivalent to 3 percent of GDP by the 1999/00 fiscal year.
3. Accelerated tariff reform, bringing forward by two years the scheduled adjustments on clothing, textiles and vehicles and reducing all other lines by 5 percentage points on average in 1997. This improves competitiveness and dampens inflation.
4. A reduction in the average annual increase in the private formal sector real wage from 1.4 percent to 0.7 percent, brought about by wage moderation within the collective bargaining system and sectoral shifts in favour of more labour-intensive industry.
5. Strong increases in investment by public authorities and public corporations, accelerating in real terms to 17 percent and 10 percent, respectively, between 1999 and 2000.
6. Additional growth in real non-gold exports, particularly manufacturing, rising to over 10 percent per annum in 2000, driven by improved competitiveness.
7. Additional foreign direct investment averaging US$509 million over the period, induced by the more favourable investment environment.

Key Results

1. GDP growth improves markedly relative to the base scenario, increasing to just over 6 percent in 2000 and averaging 4.2 percent over the period. Much of this expansion is associated with manufacturing export growth averaging over 10 percent per year accompanied by increased public and private sector investment activity.
2. Gross private savings rises somewhat as a percentage of GDP, due to fiscal discipline, increasing the availability of domestic financing for the accelerated investment.
3. Inflation is lower due to the tight fiscal and monetary stance and the containment of wage increases.
4. The current account deficit increases due to the higher growth, partly financed by foreign direct investment inflows.
5. Employment creation in the formal sector increases from 1.0 percent to 2.9 percent on average per year, brought about in part by an increase in the employment elasticity relative to output growth. Reinforced by increased non-formal employment induced by income growth, job creation increases to an average of some 270,000 per year and just over 400 000 in the year 2000.

The main strength of the integrated strategy lies, however, in the growth prospects after 2000. Annual GDP growth averaging 5 percent or higher, fuelled by strong private sector capital formation, export growth and improvements in labour productivity, leads to employment growth of some 500,000 annually, increasingly dominated by higher formal sector wage growth.

Integrated Scenario Projections: 1996-2000

APPENDIX 5

THE MACROECONOMIC IMPLICATIONS OF A FISCAL EXPANSION

As part of the analysis of the strategic options for South Africa, the possibility of utilising a fiscal expansion as the primary basis for a growth strategy was explored. Such a strategy would normally be aimed at stimulating aggregate demand with the aid of large-scale increases in government spending. The evident attraction of this option lies in the fact that it complements the urgent need for accelerated provision of services and infrastructure.

An area of concern immediately arises. The experience of the past two years casts doubt on the capacity of the public sector to absorb significant additional resources and on the prospects for reprioritisation in the absence of budget compression. A fiscal expansion may therefore exacerbate the unacceptable level of wastage in the public sector inherited from the past.

To investigate the macroeconomic feasibility of a public expenditure driven growth strategy, the effects of an increase of 10 percent per annum in real general government expenditure and in parastatal fixed investment for the period 1997 to 2000 were simulated on the econometric model of the Reserve Bank, and compared to the baseline growth scenario.

The results of the simulations are reported in the table below. Increased spending drives up interest rates which crowds out private investment. The growth rate, after an initial surge therefore declines and falls below the baseline. There is an improvement in employment as a result of government employment growth, however real wages are eroded by inflation. A growing proportion of the increased expenditure goes towards debt servicing. The major problem areas are a current account deficit that increases to well over 4 percent of GDP, a fiscal deficit which rises to 16 percent of GDP and an inflation rate that reaches 15 percent in the year 2000.

Furthermore, it should be noted that these models do not provide for realistic expectations. Modern financial markets are aware of the fact that unsustainable increases in government expenditure cause balance of payments and debt servicing problems. These expectation effects, given South Africa's relatively large public debt and limited foreign reserves, would precipitate capital flight at an early stage of the expansion. The outcome would be further currency instability and a possible monetisation of the budget deficit.

These conclusions are not surprising. The case for a public expenditure driven growth strategy rests on a version of the Keynesian model developed for a closed economy with constant prices prevalent in the 1950s and 1960s. In a modern open economy with mobile capital it is not possible to expand government expenditure without a deterioration in the budget and foreign balances and an increased exposure to a loss of confidence.

This analysis is confirmed by international experience. During the 1970s and 1980s several South American countries opted for growth and redistribution strategies characterised by large-scale government spending. The effects in all of these cases were similar: an initial period of sharp increases in output and employment, followed by severe macroeconomic bottlenecks in the forms of balance of payments crises, massive currency depreciation, large government deficits, high inflation and sharply declining real wages and growth rates.

In conclusion, large-scale increases in government spending as a macroeconomic strategy will create major macroeconomic imbalances in the form of high inflation, serious balance of payments difficulties and poor long term growth and employment prospects. This does not imply that increases in government spending aimed at addressing backlogs should not be considered under current circumstances - the macroeconomic strategy presented in this document does in fact provide for substantial spending in this respect. It does suggest, however, that increased government spending of the magnitudes required for it to serve as the foundation of growth is likely to defeat the very purposes of the strategy.

APPENDIX 6

A REVIEW OF RECENT FISCAL DEVELOPMENTS

Since 1993, the main goal of fiscal policy has been to achieve an annual reduction in the budget deficit of about 0.5 percent of GDP, together with a reduction in government consumption expenditure and avoidance of permanent increases in the tax burden. It is also envisaged that there should be an increase in public sector investment spending.

The table below indicates that the budget deficit represented 7.9 percent of GDP in 1992/93, and is budgeted to be 5.1 percent in 1996/97. For the general government as a whole and measured on a national accounts basis, current dissaving peaked at 5.9 percent of GDP in 1993 and fell to an estimated 4.0 percent in 1995. Although these trends represent a significant reduction in the public sector borrowing requirement, this has partly been the result of lags in spending and a robust recovery of revenue during the economic upswing. The underlying fiscal stance has probably remained fairly stable. Given the rising trend of debt service payments, however, the underlying deficit remains a cause for concern.

The actual spending outcome in recent years has been lower than budgeted mainly due to greater end-of-year rollovers, which can to a large extent be carried forward. Some rollovers reflect accrued commitments, including substantial unspent RDP funds. Accelerated delivery and tighter budgetary allocations must in due course lead to a reversal of the trend.

The conventional deficit excludes extraordinary transfers to offset liabilities incurred over a longer period by the government pension fund on early retirement benefits and by the Reserve Bank on the foreign exchange forward book. There are several further aspects to note:

• Debt cost is recorded on a cash flow basis and does not include provision for the discount at which stock is issued, estimated at about 0.5 percent of GDP.
• Inflation-adjusted accounting, incorporating the decrease in the real value of the debt associated with inflation, would reduce debt service charges by about 4 percent of GDP.
• Annual contributions to the government pension fund to improve the pension funding level over-state the level of remuneration by about 0.2 percent of GDP.

Non-interest current expenditure on the national budget has fallen from 25.5 percent of GDP to an estimated 22.3 percent in 1996/97, although outlays in the earlier years included some once-off expenditures. Increases in expenditure associated with the democratic transition were matched by a temporary tax surcharge on income which took effect in 1994/95 and 1995/96. Expenditure containment has been most effective in respect of transfers to local authorities and to businesses, partially offset by an increase in transfers to households (mainly social grants). As remuneration comprises nearly three-quarters of government consumption expenditure, containment of public sector employment growth and/or real wages are key elements in managing the consumption spending aggregate in future.

Although capital expenditure on the national budget has been increasing, actual spending has lagged, largely due to unspent RDP funds. Local authorities have also had some difficulty in maintaining capital expenditure. For the public sector as a whole, this has been compensated for by a stronger recovery in investment by public corporations. The considerable work on programme design done since 1993 will shortly be reflected in increased capital spending.

The higher tax-GDP ratio in 1994 and 1995 reflects, in addition to the cyclical upswing, the effect of the transition levy, phased out with effect from the 1996/97 year. Sales of oil reserves have also contributed to revenue in recent years. Tax reforms since 1993 include:

• consolidation and simplification of the personal income tax schedules;
• reduction in average income tax rates while raising the maximum marginal rate to 45 percent;
• elimination of discrimination between men and women in the tax code;
• lowering of the corporate tax rate to 35 percent; and the secondary tax on companies to 12.5 percent;
• abolition of import surcharges;
• withdrawal of the non-resident shareholders' tax;
• reduction of the marketable securities tax and stamp duty from 1 percent to 0.5 percent;
• increase in the rate of estate duty and donations tax from 15 percent to 25 percent;
• increases in excise duties, including significantly higher duties on tobacco products;
• extension of VAT to certain financial services and abolition of the financial services levy;
• introduction of a 17 percent tax on the gross interest and net rental income of retirement funds.

The combined effect of recent tax reforms has probably been roughly neutral with respect to the overall burden. Several measures have had a favourable impact on the distortionary effects of the tax system, while the overall tax incidence has remained progressive. However, annual adjustments to the personal tax schedule have not fully compensated for the effects of inflation. There has been a strong focus on improving the capacity of the tax authorities and an important step was the establishment of a more independent revenue service.

Total government debt at the end of 1995 amounted to 57 percent of GDP. Foreign debt is less than 5 percent of the total, although a proportion of domestic debt is also held by non-residents. The debt-GDP ratio has risen markedly in recent years, reflecting historically high budget deficits, lower inflation, high real interest rates and the issue of stock to cover liabilities of the former regional governments, forward cover losses of the Reserve Bank, accelerated pre-funding of pensions and provision for early retirement of civil servants. Improved debt management has been identified as a priority of the Department of Finance.