COMMENTARY FROM THE RDP TO GEAR: THE GRADUAL EMBRACING OF NEO-LIBERALISM IN ECONOMIC POLICY1 Asghar Adelzadeh Introduction Leading up to the election of 1994, the democratic movement in South Africa, headed by the ANC, released the Reconstruction and Development Programme (RDP) as its main policy platform. Politically, the document represented both a consensus across different interests and a compromise between competing objectives. Economically, the RDP was successful in articulating the main aspirations of the movement for post-apartheid South Africa, that is, growth, development, reconstruction and redistribution, in a consistent macroeconomic framework, using the Keynesian paradigm. It proposed growth and development through reconstruction and redistribution, sought a leading and enabling role for government in guiding the mixed economy through reconstruction and development, and it argued for a living wage as a pre-requisite for achieving the required level of economic growth. Socially, the systematically deprived majority supported the RDP as it promised a democratic society that will embark on unleashing the economic potential of the country in order to provide jobs, more equitable distribution of income and wealth, and provision of basic needs for all South Africans. The release of the RDP White Paper in September 1994 signified the first major point of departure from both the goals and the ethos of the initial RDP document. It unsuccessfully attempted to reconcile the original Keynesain approach to the RDP with a set of policy statements and recommendations that were inspired by the neo-liberal framework that had long been the alternative offered, even if in different variants, by big business, the International Monetary Fund (IMF) and World Bank, and, not least, the apartheid state itself in its twilight years in the form of the Normative Economic Model (NEM). In the departure from the Keynesian towards the neo-liberal framework, the White Paper transformed the role of fiscal prudence from a means to achieve RDP objectives to an objective TRANSFORMATION 31 (1996) ADELZADEH COMMENTARY of the RDP; the goal of redistribution was dropped as a main objective; and the government role in the economy was reduced to the task of managing the transformation. Since the publication of the RDP White Paper, the government has moved even further towards exclusive acceptance of the tenets of the neo-liberal framework and its associated policies. Despite continuing references to RDP objectives, the framework underlying government's fiscal, monetary and international economic policies has increasingly departed from the original thrust of the RDP framework. The draft National Growth and Development Strategy (NGDS), that was presented in February 1996, departed even more significantly from the RDP Base document by adopting a trickle-down approach to economic development. In June of this year, the Department of Finance released yet another document, Growth, Employment and Redistribution, a Macroeconomic Strategy (GEAR). As will be seen, this leaves little remaining of the RDP as it was first conceived, other than in name alone. The pace of delivery of new policy documents stands in sharp contrast to delivery of RDP objectives themselves. The main objective of this paper is to present a critique of this latest document's growth framework and policy recommendations. After a careful reading of the document, we have concluded that the proposed framework and policy scenarios represent an adoption of the essential tenets and policy recommendations of the neo-liberal framework advocated by the IMF in its structural adjustment programmes. This is all the more remarkable in view of the limited, even negative impact of such programmes, especially in southern Africa, the lack of any leverage that the international financial institutions such as the IMF and World Bank have over South African policymakers, the lack of any dramatic changes in the economic and political environment to warrant such major shifts in policy orientation, and the lack of a transparent and fully argued justification for the adoption of an entirely different policy framework. It is reasonable to conclude that the substantive abandonment of the RDP as originally formulated is indicative of a panic response to the recent exchange rate instability and a lame succumbing to the policy dictates and ideological pressures of the international financial institutions. This all represents a recourse to the policy goals and instruments of the past apartheid regime, and to the failed adjustment strategies being imposed on other countries under IMF and World Bank tutelage. This paper analyses how and why the proposed growth framework and policy scenarios (a) provide neither sound analytical arguments nor convincing empirical evidence for its main proposals; (b) are flawed and inconsistent in its conceptualisation, its policy proposals, and its empirical projections; (c) have not integrated some of the main issues of economic transformation in its modelling TRANSFORMATION 31 (1996) 67 COMMENTARY ADELZADEH exercise and proposed scenarios; and, (d) will fail in meaningfully addressing let alone transforming the inherited inequities of the apartheid system. On the Modelling Aspect of the GEAR Document The document states that detailed simulations were conducted with four models of the South Africa economy to assess the effects of the proposed strategy on growth and employment, (Section 2.3). The reported results in the document, however, are derived from a single model, that is, the Reserve Bank model (appendix 16). The document, though, is quick to argue that the projections described 'are broadly consistent with results obtained using models of the Development Bank of South Africa (DBSA), the Bureau of Economic Research, and the World Bank' (appendix 3). The document's report and analysis of the results of the underlying modelling exercise is incomplete and unsatisfactory for the following reasons: The main model used, the South African Reserve Bank (SARB) model, has never been made public and thus has never been the subject of independent and rigorous debate by professional economists; moreover, there is no evidence to believe that this same model which was used by the Reserve Bank during the previous regime has been restructured to take into account the new circumstances and policy goals as outlined in the RDP. For the public to engage meaningfully in debate on the proposed programme for growth and development of the country, it needs a clear understanding of the underlying economic framework of the Reserve Bank model and how the main behavioral relationships amongst macroeconomic variables are specified and justified in that model. No evidence is provided to substantiate the document's assertion that the reported results derived from the Reserve Bank model are 'broadly' consistent with the results obtained from using models as diverse as the World Bank's General Equilibrium model, the Keynesian-Neoclassical Econometric model of the World Bank, and the DBSA's Structuralist computable general equilibrium model. Given that the models are significantly different from each other in terms of their underlying framework, we are not convinced that these diverse models can yield similar results or even broadly consistent results. We are told that '[t]he key growth drivers, namely the investment and export projections were tested for feasibility and consistency on all the models and subjected to econometric analysis' (appendix 4:12). However, no results of these tests are presented nor analysed. There are clear and simple procedures for presenting comparative tables of models' results; we would have expected the government to provide all the results to the public and allow the public to draw TRANSFORMATION 31 (1996) ADELZADEH COMMENTARY its own conclusions as to whether these results are narrowly or broadly consistent with those derived from the Reserve Bank model. We are told that the expenditure implications of the fiscal strategy of the proposed package, namely a rapid reduction in the budget deficit towards the targeted 3 percent of gross domestic product (GDP), are analysed with the aid of the medium term expenditure model of the Department of Finance (appendix 9). From the document's overall proposals on the growth framework and its arguments in appendices 5 and 9, it is clear that the main concern of the Reserve Bank model is to capture the crowding-out effect of deficit spending by the government within the arbitrarily adopted target of the 3 percent budget deficit. The expenditure model of the Department of Finance is used to provide the distribution of funds amongst different government departments or economic functions so that the overall spending is consistent with the target budget deficit for each year. While some of the targets or results of the Reserve Bank model feed into the expenditure model, the restructuring and reprioritisation of the government expenditures for each economic category, if conducted in the expenditure model, does not feed into the macroeconomic model. This means that the modelling exercise and the results presented under the base and integrated scenarios have not incorporated the growth, employment and income distribution impacts of particular scenarios for the restructuring of government expenditures. This is a major shortcoming of the modelling exercise, since we believe that restructuring of government expenditures would have a specific quantifiable impact on growth, employment and incomes of different social groups. In appendix 15 we are told that '[a] simple model of the determinants of sectoral employment trends is used below to compute future employment on the basis of projections regarding sectoral output growth and real wage movements' (p34). The reported results are said to be 'broadly consistent with macroeconomic scenarios described elsewhere' (p34). Our problem with this section of the document stems from insufficient information regarding the Reserve Bank model. The above statement is puzzling because it is not clear why the Reserve Bank model provides projections for sectoral output growth and real wage movements, but it does not provide projections on sectoral employment effects of the policy proposals. It seems that for determining the sectoral employment effects of the policy proposals, the research team employed another model of which we have no knowledge. If we are to use macroeconomic models to give support to a given economic strategy, we need to be mindful of the fact that (a) none of the models used has been subject to independent rigorous debate or verification by professional TRANSFORMATION 31 (1996) 69 COMMENTARY ADELZADEH economists and (b) that most of the large macroeconomic models are designed for the short-run macroeconomic management of the economy, and not for economic restructuring of a society in transition over the medium term. Based on the document, the Reserve Bank model cannot provide information on the income distribution effects of policies (appendix 4:12), is unable to furnish sectoral employment effects (p34), and is unable to incorporate the effects of restructuring government expenditures on growth, employment and income distribution (appendix 9:21). Given these fundamental shortcomings of the Reserve Bank model, we do not think the government should have used this model for assessing the feasibility of policy packages that are expected to be concerned with employment creation, more equitable distribution of income and restructured government expenditure along the line of new priorities. However much the model may or may not have been appropriate for short-run macroeconomic policymaking during the apartheid era, and however much it may have been modified subsequently, as the Labour Market Commission (LMC) concluded in respect of employment, both the analytical framework and the policy orientation of the model are unduly negligent of RDP-type objectives and concerns. On Growth Issues Our questioning of the feasibility of the document's proposals stems from our main concerns as to whether the proposed growth framework and policy scenarios (a) lead to substantial increase in the growth rate of the economy, (b) reverse the unemployment crisis and reduce the unemployment rate significantly, and (c) yield sufficient progress towards an equitable distribution of income and wealth. Our detailed examination of the released document has revealed a number of problems regarding the overall and the specifics of the proposed framework and policies for growth. The document does not provide a clearly defined distinction between 'growth' and 'sustainable growth'. It is our contention that if the achievement of 'sustainable growth' is the goal of the framework, it is the responsibility of authors to clearly define the concept and provide analytical and quantitative criteria for the assessment of when and under what conditions we can claim that we have reached 'sustained growth on a higher plane' (pi). It is only in the context of such a detailed presentation that one can define the necessary and sufficient conditions for achieving sustainable growth. Without such an analysis the statement that the achievement of such a goal 'requires transformation towards a competitive outward-oriented economy,' (pi) is merely empty posturing and bears neither justification nor meaning to the conclusion. 70 TRANSFORMATION 31 (1996) ADELZADEH COMMENTARY The document's lack of rigour in explaining what it means by sustained growth as opposed to growth, and what the quantitative criteria are for examining sustainability, have led it to pronounce the simulation results for the year 2000 as results that will be sustained after the year 2000. The document states that '[t]he strategy developed below attains a growth rate of 6 percent per annum and job creation of 400 000 per annum by the year 2000 ...' (section 1.3:1, emphasis added; see also appendix 15:34). The phrase per annum is misleading. The provided data only shows that the growth rate for the year 2000 will be 6.1 percent and the job creation for the year 2000 will be 400 000. We are not presented with any evidence that after the year 2000, the growth rate will be maintained at 6 percent and jobs will be created at the rate of 400 000 per annum. In addition, the document does not provide any analytical nor empirical evidence that the results for the year 2000 correspond to 'growth on a higher plane'. Moreover, given the structural problems inherited from the apartheid economy, it is unrealistic to argue that we can achieve sustainable growth in less than five years if the concept of sustainability is linked to profound transformation in the economy. The claim that to achieve sustained growth on a higher plane requires establishment of an outward-oriented economy seems to be based on the perception that the South Africa economy has been relatively closed. Currently South Africa's non-gold export-GDP ratio is equal to 21 percent which is 5 percent above the average export-GDP ratios of the G7 countries for the period 1990-94; 2 percent above the average ratio for the OECD countries; 6 percent above Indonesia's export-GDP ratio; 4 percent above India's ratio, and 3 percent above Brazil's average export-GDP ratio. It is equivalent to the export-GDP ratios of Hong Kong and South Korea. If we include the export of gold, South Africa's exports-GDP ratio rises to more than 25 percent. This means that, currently one-fourth of total output of the economy is produced for foreign markets which is substantially higher than many of the OECD countries and the South East Asian tigers. We, therefore, believe that the GEAR document should have presented a clear analysis of why one of the two 'growth drivers' of its growth strategy should be the substantial expansion of exports when more than one-fourth of the country's output is currently exported. In all developed economies a much higher percentage of production is geared towards domestic use (consumption and investment). For example, USA and Japan's share of exports in their total output was only 8 and 10 percent on average between 1960 and 1994, respectively. This resulted despite their long-term growth and their vigorous promotion of exports for many years. Moreover, it has been empirically shown that production for the TRANSFORMATION 31 (1996) 71 COMMENTARY ADELZADEH domestic market is often significantly more job-creating than production for exports. Therefore, in considering the role of exports in economic growth, while we tend to support measures to expand the country's non-gold exports, a more balanced approach should be given to the role export growth can and should play in the overall growth of the economy. In this context, it would have been most welcome if we were also presented with an alternative growth scenario that was based on a more rapid expansion of production for the domestic market. Of course, outward-orientation might well be employed as an index of productivity and as a pressure to attain it through competition on world and domestic markets. But, even so, there is no reason for this to lead to a shift in the composition of output towards exports. If competitiveness increases, it is as likely to raise potential for production for all markets. The link, then, between outward-orientation and the share of exports in GDP is not established either as a policy objective or as a realistic empirical outcome. Are the policies proposed adequate to achieve the predicted increase in the GDP growth rater In standard, mainstream macroeconomics there are four sources of increased effective demand: household consumption expenditure, private investment expenditure, state expenditure, and net expenditure from abroad (exports minus imports). Since consumption is typically treated as determined by income itself, effective demand can be reduced to three elements: private investment, state expenditure, and net exports. We can now restate the question: Will the policies proposed in the document increase private investment, state expenditure, and net exports sufficiently to raise the GDP growth rate to an average of 4.2 percent per annum for the period 1996-2000? The overall policy stance of the document is succinctly summarised as follows: In brief, government consumption expenditure should be cut back, private and public sector wage increases kept in check, tariff reform accelerated to compensate for the depreciation and domestic savings performance improved. These measures will counteract the inflationary impact of the exchange rate adjustment, permit fiscal deficit targets to be reached, establish a climate for continued investor confidence and facilitate the financing of both private sector investment and accelerated development expenditure (sub-section 'Accelerated Growth ). In this context, the document recognises the need to programme a set of policies to generate the demand stimulus to bring about the projected rate of growth of 4.2 percent for the period 1996-2000. It argues that 'growth is mainly the result of steady increases in fixed investment and manufactured exports', with the former comprising an annual average rate of growth of private investment of 11.7 72 TRANSFORMATION 31 (1996) ADELZADEH COMMENTARY percent and public investment of slightly over 7 percent. These investment growth rates represent more than a doubling of the former and almost a trebling of the latter, compared to the 'base scenario' (with slower deficit reduction and less supply-side reform), with a growth rate of 3 percent. The increase in government investment can be directly implemented, while the increase in private investment would be the indirect effect of policies to 'establish a climate for continued investor confidence'. Without even considering whether the proposed policies would in fact simulate private investment to the necessary degree, we can point out that it is not strictly correct to say, even on the strategy's own figures, that growth would be mainly the result of increases in investment and manufactured exports. The target rate of growth is taken as 4.2 percent. If we consider the impact of exports, the fiscal effect of external trade is negative. This is because the integrated scenario projects an increase in the ratio of the external current-account deficit to GDP, from 1.4 percent to 2.4 percent. This increase, according to the document, is the result of increased GDP growth drawing in more imports compared to the base scenario's growth rate of 3 percent. Since expenditure on imports represents non-expenditure on domestic production, the result of an increase in the current-account deficit relative to GDP is a net depression of domestic demand of -0.2 percent (compared to the targeted stimulus of+4.2). Public investment, at the discretion of the central government, imparts a positive stimulus, but quite a small one of +0.5 percent, which when combined with the negative effect of net exports sums to a stimulus of +0.3 percent. Since public consumption (recurrent expenditure) is pegged at 19 percent of GDP for the period, its fiscal stimulus is derived from the rate of growth and the overall fiscal stimulus itself. This leaves private investment to fill the gap in the fiscal stimulus after accounting for public investment (positive) and net exports (negative). On the basis of the target rate of GDP growth of 4.2 percent, private investment must contribute 3.9 percent of the fiscal stimulus. This rather tedious calculation exercise yields the following conclusion: virtually all of the fiscal stimulus necessary to achieve the rate of GDP growth projected by the document must come from private investment (93 percent of the total stimulus). Thus, the realisation of the projected growth rate is almost completely dependent upon the rapid success of government policy in stimulating private investment. We can now ask the question, how credible is the projection of private investment in the document? The vehicle for stimulating private investment is the reduction of the fiscal deficit which is predicted to reduce interest rates, and in turn is expected tc provoke a rise in private investment. This logic is based upon what is called the 'crowding out' argument: when the state TRANSFORMATION 31 (1996) 73 COMMENTARY ADELZADEH borrows to finance a deficit, it competes with the private sector for funds, thus pushing up interest rates. There is no consensus among economists about the importance of 'crowding out', and no empirical evidence to suggest that it has ever occurred in South Africa. Thus, over 90 percent of the fiscal stimulus required to achieve the document's growth target derives from an unsubstantiated opinion about the operation of the South African capital market and private investor motivation. And, of course, the crowding-out of private investment through this mechanism has to be set against the stimulus that would be induced by government expenditure both directly and indirectly. Even if one accepts the assertion that 'crowding out' occurs in South Africa, the document's fiscal stance remains questionable. In the base scenario of the document, the average interest rate is projected to be 5.2 percent (adjusted for inflation), associated with a growth rate of private investment of 5.6 percent a year. In the integrated scenario with the 4.2 percent growth rate, the fiscal deficit is reduced from 4 percent to 3.7 percent of GDP, the predicted effect of which is to reduce the real interest rate to 4.4 percent. This lower interest rate is associated with a rate of growth of private investment of 11.7 percent. When investment growth is converted into levels of investment, the elasticity of investment with respect to the interest rate will be about 0.55, a rather high responsiveness. It is particularly high because over ten percent of private investment is in mining, a sector in which growth prospects are bleak. The dependence of the strategy on this interest-elasticity of investment is extremely insecure, particularly given the volatility of such investment functions and the absence of disaggregation across sectors so that the realism of the scenario can be more fully assessed. Further, there is a pressing problem in the proposed scenario. The key to the fiscal stimulus is a lower real interest rate, yet the model assumes that the external current account will deteriorate. Other things being equal, this would put pressure on the Rand, which could provoke the Reserve Bank to act independently to raise interest rates. The document seeks to square this circle by predicting an increase in foreign direct investment to cover the increased current-account deficit. Thus, the growth rate projected by the document is based on a number of questionable assumptions: 1) that 'crowding out' is an important phenomenon in South Africa, 2) that deficit reduction will result in a fall in the interest rate, 3) that an increased current-account deficit is consistent with a lower interest rate, and 4) that a lower interest rate will impart a strong stimulus to private investment. In other words, in the document, the government provides very little fiscal stimulus to reach the required growth target, and success is almost wholly dependent upon the response of the private sector. 74 TRANSFORMATION 31 (1996) ADELZADEH COMMENTARY On Growth and Fiscal Policy The document asserts that the current fiscal situation is unsustainable (section 3.1:7). Therefore, a key objective of the macroeconomic strategy is to reduce the fiscal deficit rapidly by cutting government expenditure. There are several problems associated with this argument: • The proposed programme fails to provide an analysis of why the current deficit-GDP ratio is unsustainable in a dynamic framework, nor does it provide any justification for why the public deficit should be lowered to precisely 3 percent of the GDP. Since the 3 percent debt-GDP ratio target has serious implications for the achievement of some of the RDP objectives, the government must provide a rationale for the choice of this particular target, specially since this target is not a model result but an exogenously set target. It is worth mentioning that the World B ank' s three policy scenarios for South Africa concluded that a rise in fiscal deficit to 12 percent of the GDP is sustainable because the higher growth pattern will gradually generate more public savings such that by the year 2000 the country will experience fiscal surplus (World Bank, 1993:5). • The issue of why the current deficit is not sustainable is neither analytically nor empirically explained. Any discussion of sustainability of a given government deficit-GDP ratio usually starts with the dynamic government budget constraint. In this sense, it is important that the assessment of the sustainability of fiscal policy is not static but forward looking, and sustainability should not be limited to the short term and/or the medium term, because government's commitments to specific programmes have implications for fiscal policy which reach far into the future. The formulation of spending and/or taxation policies must, therefore, take into consideration a number of factors. The document, while it imposes an arbitrary deficit-GDP ratio target and adopts spending cuts, provides no sign of having taken into consideration the complex factors involved in the dynamics of the debt-GDP ratio. For example, one important element in the overall assessment of sustainability is to compute the tax rate which would satisfy the conditions for sustainability given the forecasts of spending and transfers. The document does not present us with such an analysis. • The literature on sustainability clearly points out that if the difference between the real interest rate and the growth rate of the economy is TRANSFORMATION 31 (1996) 75 COMMENTARY ADELZADEH positive, the ratio of debt to GDP increases over time, unless the cuts in current spending, transfer and tax rules compensate for such an increase. In the integrated scenario, the real interest rate (real bank rate) is kept substantially higher than the expected growth rate of the GDP. This means between 1996 to 1998, there will be a tendency for the debt/GDP to increase by the amount equivalent to the current nominal debt-GDP ratio times the difference between the real interest and the GDP growth rate. This means that the only way for the model to impose a reduction rather than an increase in the debt/GDP ratio due to the adopted monetary policy is to reduce government spending to offset the above effect. In other words, the chosen contractionary monetary policy of high real interest rate, above the GDP growth rate, also dictates a contractionary fiscal policy in order to preserve the target deficit-GDP ratio. In the integrated scenario, government's expenditure policy is taken hostage by the contractionary monetary policy of the Reserve Bank that has been integrated in the overall growth strategy. As was pointed out earlier, one major tenet of the proposed growth framework argument is that increased government spending drives up interest rates which crowds out private investment, thus retarding the growth rate of the economy. Not only is the analytical and empirical applicability of this argument questionable, it also overlooks the recent international resurgence of focus on the role that public productive expenditures on infrastructure (such as investment on roads, transportation and housing) and social services (such as education, health care and welfare) play in promoting not only a country's economic well-being and growth, but also in encouraging private investment. In this case, public productive expenditure in fact 'crowds-in' private investment and sets the necessary foundation for long-term sustainable growth. In this context, the proposed policies overlook recent developments in endogenous growth theory which place much greater emphasis on the 'crowding-in' effects of public expenditure and dynamic economies of scale that would arise out of faster growing demand and the growth in skills of the workforce. These are all factors of considerable importance in the South African context. Another stated objective of the macroeconomic strategy is to reduce personal and corporate taxes if growth occurs (section 3.5:9). A few comments: • Studies of South Africa's tax effort indicate that the country is under-taxed by about 3 percent of GDP. There are two kinds of studies: ratio analysis and econometric models. The ratio analysis simply 76 TRANSFORMATION 31 (1996) ADELZADEH COMMENTARY compares various tax ratios for South Africa - individual income tax, corporate taxes, total taxes, etc as a percentage of national income to comparable figures for other countries - upper-middle-income countries. Organisation of European Co-operation and Development (OECD) countries, etc. With these studies, the conclusion depends on the reference group: if it is middle-income countries, South Africa is over-taxed according to some measures, but if the comparison group is OECD countries, South Africa is under-taxed. The more rigorous analyses, using econometric methods, control for country characteristics, such as income level, etc, and indicate that South Africa was unambiguously under-taxed by 3 percent of GDP in 1994. • The tax structure is particularly inefficient, with a very narrow tax base and a very heavy reliance on regressive taxes such as Value Added Tax (VAT). The absence of a capital gains tax undermines the progressivity of the tax structure. Tax reform aimed at increasing the progressivity of the tax system while promoting economic efficiency could probably yield an increase in revenue equal to 3 percent of GDP per year. For instance, the absence of a capital gains tax creates substantial economic distortions, as income is 'converted' from other forms into capital gains. Likewise, there is considerable scope for revenue raising through increased taxation of luxury consumption, and a tax on unproductive land could actually increase private investment, based on the real options theory of investment. • The tax advantages for contractual savings used to be offset by a compensating system of prescribed asset requirements. Financial liberalisation eliminated these requirements, but retained the tax advantages. A theoretical growth model based on the theory of the second best indicates that combining prescribed asset requirements with tax advantaged contractual savings might promote economic growth and improve the government's fiscal position. • The macroeconomic strategy proposes tax holidays for investments that meet certain criteria. International experience has been mixed. Fostering a tax bidding war among countries for external capital can turn into a negative sum game for the countries themselves, with mobile capital the TRANSFORMATION 31 (1996) 77 COMMENTARY ADELZADEH only winner. And entrepreneurs can be remarkably creative at milking tax advantages while minimising social contributions. It has also been argued in the document that national debt can be reduced through reducing government expenditure (section 1.4:2). This demands a few comments: • South Africa's national debt levels are not high by international standards. External indebtedness is particularly low by international standards. In terms of total public-sector debt as a percentage of GDP in 1995, South Africa stood at 56 percent compared to 95 percent for Japan, 63 percent for the USA, 60 percent for Germany, or an average of 72 percent for all OECD countries. External-debt ratios provide an even brighter picture, as do comparisons with middle-income countries. Fears of a debt trap are therefore unfounded. • Even if it were desirable to reduce debt levels, the question of timing is critical. Presumably debt should be increased when returns to social investment are high, and reduced when returns to social investment are lower than the interest rate. To reduce debt levels now would reverse this logic: debt was accumulated to finance apartheid (with its negative social returns), and would be paid off during the period of democratic consolidation, when resources could be allocated to long-neglected imperatives, with the associated high social returns. Given the country's need for radical improvements in social services and infrastructural development, the stated objective of the macroeconomic strategy to reduce government spending implies a shift to reliance on the private sector for the delivery of services. The experience of many countries indicates that in a market economy, the private sector generally fails to deliver public goods and social infrastructure adequately. Thus, reducing government expenditure diminishes the chances of RDP delivery, with adverse consequences for the long-term well-being and productivity of the majority of South Africans. The integrated scenario projections show '[sjtrong increases in investment by public authorities and public corporations, accelerating in real terms to 17 percent and 10 percent respectively, between 1999 and 2000' (appendix 4). This means the main increase in the public investment will not begin until later years with the major growth in government investment being projected only for the last year 78 TRANSFORMATION 31 (1996) ADELZADEH COMMENTARY of the programme. Obviously, the realisation of these investments depends on a significant increase in government revenue in previous years. If the forecast growth rate of the economy does not materialise, the whole promise of increase in public investment expenditure will also not materialise. Moreover, given government's emphasis on sharp reduction in tariffs, extension of tax holidays and plans to reduce taxes as growth takes off, it is quite possible that the forecast sharp increase in public investment during the last year of the programme will never occur. Furthermore, the forecast jump in the growth of real government investment in the last year of the programme from 7.5 percent in 1999 to 16.7 percent in the year 2000 seems totally out of line and is offered without explanation. If the forecast growth in government investment for the year 2000 is brought in line with the previous years' growth rates, the target 6 percent GDP growth rate by the year 2000 will not be feasible within the framework assumptions. On Growth and Monetary Policy The document states that the monetary authorities are to provide 'a consistent monetary policy to prevent the resurgence of inflation' (p3). While a consistent monetary policy is a desirable strategy, the second part of this statement seems irrelevant in the case of South Africa. The South Africa economy has not experienced any uncontrollable inflationary surge in the last 30 years. Therefore, the monetary authorities are overlooking the country's historical experience with regard to inflation. It exhibits a commitment to realise the RDP objectives by adopting a pillar of the neo-liberal framework advocated by the IMF - that inflation should be kept at a one digit level and possibly around five percent regardless of the country's circumstances. In this respect we find the second part of the following statement that defines the main objectives of monetary policy as 'the maintenance of financial stability and the reduction of the inflation rate' (section 4.1), to be damaging in its consequences for growth and employment. Even the World Bank acknowledges that the output and thus employment cost of reducing the traditionally moderate inflation rate in South Africa (between 10 and 20 percent) to a low inflation rate is significant (Fallon and da Silva, 1994:164). To set such an unconditional objective to guide the monetary policies of the Reserve Bank is also without precedent anywhere else in the world. The implication of this feature of the framework is that as soon as the economy starts to grow and the inflation rate begins to increase marginally, the Reserve Bank will immediately intervene by adopting contractionary measures. This built-in contractionary monetary policy within a growth framework goes against TRANSFORMATION 31 (1996) 79 COMMENTARY ADELZADEH the main objective of the document by effectively preventing the economy from attaining a significant growth rate. Also the document's forecasts of declines in the inflation rates in the last two years of the integrated scenario are puzzling because the inflation rate is expected to fall when there are simultaneous jumps in the growth rates of public and private investments from the sum of 31.4 percent to 43.7 percent and the GDP growth rate is forecasted to rise from 3.8 to 6.1 percent during the same period. If the Reserve Bank is given the task of ensuring the achievement of the forecast inflation rates, the forecast increase in investment and the target GDP growth rate for the last two years of the scenario will be unattainable. In the document the government commits itself to 'a gradual relaxation of exchange controls' (p3). However, no explanation is provided as to why and how the removal of exchange control on domestic residents needs to be part of a growth and development strategy. Obviously, foreign capital inflow is not dependent on exchange controls on domestic residents; if this was the case, no one could explain why China has been receiving such a high share of foreign investment despite its strict exchange controls on domestic residents. The mechanism through which removal of exchange controls on domestic residents are to enhance growth remains unexplained. Such measures will lead to the outflow of a significant amount of domestic savings, which is contrary to the document's objective of promoting savings. Furthermore, the recent and continuing exchange-rate crisis in South Africa has highlighted the shortsightedness and over-optimism of the Reserve Bank regarding the consequences of the removal of the Finrand and exchange controls on foreigners. In none of the documents provided by the Reserve Bank prior to or after the removal of the Finrand and exchange controls, did the country's monetary authorities predict or hint at a possible crisis. They forecast instead a sustained increase in the flow of foreign investment into the country as the result of the move. This prediction did not materialise. Instead the policies resulted in the increase in the inflow of short-term capital which has in turn resulted in the rise in the volatility of the country's financial market, placing severe upward pressure on interest rates in the absence of any other policy instrument. The preferred integrated scenario assumes that between 1988 and the year 2000 the real effective exchange rate remains unchanged despite the continuation of the current-account deficit during the same period. The key to this outcome is an assumed increase in foreign investment inflow such that it is precisely equal to the amount needed to offset the depreciating effect of the continued current-account deficit, not a Rand more or less. Given that the flow of foreign investment is determined by a complex set of economic, political and social 80 TRANSFORMATION 31 (1996) ADELZADEH COMMENTARY factors, the document provides no justification for its choice of the level of foreign investment into the country and does not entertain the possibility of not achieving or of over-achieving the targeted foreign investment during the projection period. Furthermore, we believe that it is quite possible that the modelling exercise has underestimated the impact on imports of final and intermediate goods as a result of the combination of increased trade liberalisation and the rise in the growth rate of the economy. Given the country's relatively high propensity to import (the real import penetration ratios of the 25 manufacturing sectors ranged from 3 percent to 58 percent in 1993, the average rate for the manufacturing was 17 percent), it is likely that the proposed speedy reduction in tariffs and the forecast growth rate, if they materialise, will lead to a larger increase in imports than predicted. At the same time, the expected sharp increase in exports above previous trends is unlikely to occur. We believe that the proposed scenario has underestimated the trend in the current-account deficit and thus the level of foreign investment needed to offset the more than expected pressure on the country's exchange rate. On Growth, Trade and Industrial Policy Section five of the document, on 'Trade, Industrial and Small Enterprise Policies', is thin on what the proposed trade policies, industrial policy and small enterprise policies are to be. On trade, the proposals appear to do away with any policies, as fast as possible. Some small enterprise policies are listed, and there is a complete absence of any proposal for an industrial policy or strategy. Trade Policy Issues of trade policy are high on the South African policy agenda. This is not surprising, given the emergence of the economy from the era of sanctions, with the need to build and develop new trading relations. However, this is not the spirit in which the issue is approached in the document. The objective should be how to use trade to assist in the reconstruction and development of the economy. Instead the objective appears to be a wholesale acceptance of the 'free-trade' dogma. It is important to point out that the free-trade consensus is based on a simplistic theory, and can, in policy terms, encourage a dangerous complacency as is certainly displayed in the document. There is a good case for protecting or supporting sectors that, for example, generate large positive externalities such as technological spillovers. Tariffs may also have large positive effects through increasing domestic demand, thus stimulating the introduction of new products. TRANSFORMATION 31 (1996) 81 COMMENTARY ADELZADEH By encouraging import-substitutes, protection can expand the domestic traded-goods sector. The means of expansion operates through reducing the propensity to import and thus reducing the leakages from the domestic economy. The objective of protection in an underemployed economy is to reduce the propensity to import competitive goods, not to reduce the actual volume of imports. If the policy is successful, the rise in domestic incomes should encourage more imports of complementary and subsequently competitive goods. The document presents the proposed acceleration in tariff reductions to compensate for the depreciation of the currency. Instead, an active industrial policy should be used to expand domestic production of those goods on which tariffs apply. This is the opportunity which the currency depreciation allows - an opportunity which the document proposes to squander by accelerated tariff reductions to benefit imports. The document suggests that allowing in cheap imports will reduce prices for domestic consumers. But the aim should be to expand domestic production, allowing the spreading of overheads and investment in new technologies to lower unit costs and hence reduce prices. This is the sustainable way to lower prices and increase South Africa's competitiveness. Furthermore, government's revenues from maintained tariffs could be used in a range of ways to further reduce domestic prices, including subsidies or tax cuts. The document underestimates the positive impact of depreciation on export prices and overestimates its negative impacts on imports. Historical experience of many countries demonstrate that (a) trade liberalisation leads to a significant increase in imports of final consumption goods, and (b) international companies that are eager to capture shares of newly emerging markets are more than willing to absorb some of the short-term effects of depreciation, by lowering their mark-ups, in order to ensure a long-term position in the domestic market. Therefore, the document is too optimistic regarding the effects of depreciation on the current account Given the country's current export-GDP ratio and taking into consideration the last 35 years experience of all the developed and newly developed economies, we think the document's target of an average yearly growth rate of 8.4 percent for the non-gold export sector is unrealistic and unattainable. This is because the expected growth rate of exports implies a 22.7 percent increase in the export-GDP ratio within the next four years, from 20.9 percent of GDP in 1995 to 25.64 percent of the GDP by the year 2000. Such a high increase in the growth rate of the export-GDP ratio in such a short period has not been experienced by any of the European, American or Asian countries. For example, in 25 years between 1970 and 1994, the average export-GDP ratio of the OECD countries 82 TRANSFORMATION 31 (1996) ADELZADEH COMMENTARY increased only 6 percent. During the same period the average export-GDP ratios of Brazil, India, Indonesia, and South Korea increased 5 percent, 1 percent, 2 percent and 4 percent, respectively. The average export-GDP ratio of Hong Kong remained unchanged between 1970 and 1994. A faster rate of growth of non-gold exports is clearly desirable. Unfortunately, the Government has not developed a carefully formulated and precisely targeted industrial strategy geared to those sub-sectors that have potential for export growth. Worse, it makes unrealistic assumptions about the probability of success of the policies it proposes for export growth. Small and Medium-Sized Enterprise Development (SME) The paper refers to the policies outlined in the White Paper on small business promotion but adds nothing new. The emphasis on cutting the fiscal deficit, maintaining interest rates above the rate of inflation, etc, suggests an orthodox economic approach which will squeeze small and medium-sized enterprise. The prospects for small and medium-sized enterprise development will remain poor in the absence of an industrial policy committed to the creation and support of such enterprises, if necessary through the expansion of public expenditures on construction and infrastructure to be carried out by SMEs. Yet there is no mention of any such industrial policy. There is a lack of integration of the SMEs development strategy into the proposed integrated macroeconomic scenario. For example, in chapter 4 (on Monetary and Exchange Rate Policy) in one sentence it is stated that high interest rates might limit SME access to finance. Clearly, the envisioned high interest rate in the document will dampen demand for loans; how the integrated scenario intends to ensure SMEs access to finance is not spelled out. Industrial Policy It is perhaps symptomatic that while the reference in the table of contents is to 'Trade, Industrial and Small Enterprise Policies', the corresponding sub-heading within the body of the paper is not to industrial policy but to 'industrial support measures'. And the section does little more than list policies that have failed so conspicuously in the past. It is said that the review of competition policy will 'open up new opportunities for investment'. But there is no indication of what will be done to ensure that the opportunities are met by investment rather than imports. High real interest rates do not help. There is no commitment to public enterprises in any form - such as, the creation of new public-private ventures, encouraging local enterprise initiatives, etc. The only reference in the document to public-private ventures is TRANSFORMATION 31 (1996) 83 COMMENTARY ADELZADEH to transform existing public ventures into public-private ones; yet there are many areas of the economy where private enterprise is failing to establish new ventures but where a partnership with the public sector might facilitate their entry. A case in point would be in housing. A housing parastatal would ensure that more houses are delivered as opposed to current policies which rely primarily on the goodwill of the private sector. It is said that industrial innovation support programmes will be enhanced but again few details are elaborated. This is an area where public sector involvement is vital - without it the private sector is extremely likely to underinvest and not capitalise. We therefore believe that the document is weak in all three areas. The lack of any serious trade, industrial or small-enterprise policies is worrying, committed as the government is to a rather orthodox economic agenda. For example, 'greater flexibility in the labour market' may actually undermine the incentive for firms to train their workforce if it increases labour turnover and makes it more likely that the benefits from such training will be enjoyed by rival firms poaching the skilled workers. There is also a remarkable lack of imagination when it comes to public sector initiatives. Where private enterprise is failing to invest, and the 'free market' is failing to respond to new opportunities by launching new enterprises, the public sector should show the way. This could be done by government investing in productive infrastructure; by supporting start-ups and co-ops; by establishing new joint ventures with the private sector; and by establishing new public sector firms where necessary. Free enterprise is all too often neither free nor enterprising. It is the government which has a duty and a mandate to reconstruct and develop the economy. On Growth and Income Distribution We all agree on the need to seek 'a redistribution of income and opportunities in favour of the poor' (pi). However, we fundamentally question the underlying premise of the document which focuses on job creation as the main source of income redistribution, and focuses on achieving a desirable level of growth as an automatic mechanism towards an equitable distribution of income and wealth (see pi, and p21). While there is a common agreement that the unemployment rate must significantly decline as a result of social and economic restructuring, we need to pay due attention to the recent FLO and UN reports that argue that wage reductions that are designed to increase employment can increase poverty at the same time. In fact, a careful reading of the reported forecasts for the proposed growth framework indicates that income distribution will deteriorate 84 TRANSFORMATION 31 (1996) ADELZADEH COMMENTARY during the course of the programme. This is contrary to the RDP's statement that 'our growth path must ensure more equitable distribution of incomes' (ANC, 1994:8). For 1996-2000, the integrated programme calls for real wage growth of about 1 percent per annum, with an associated rate of growth of employment of 2.9 percent. This results in a growth of the total wage bill of 3.9 percent per annum. Since this is below the rate of growth of national income as a whole, the share of income going to employees will of necessity fall. Further, the logic of the document seems necessarily to imply that much of the employment growth would be at wages below the prevailing average for all employees (see appendix 15 on employment elasticity of output in different sectors). This implies a further depressing effect on real wage growth and, thus, on the wage share. While there are countries in Africa for which a decline in the wage share might not imply an increase in inequality (because average wage incomes are above average incomes), South Africa is not among them. The expected increase in the income inequality will not be surprising given the proposed policies of budget cuts, trade liberalisation, deregulation of the labour market and privatisation. More fundamentally, as stated earlier, we believe that there is a problem with the framework's focus on the growth of exports and investment as the only two drivers of economic growth. In the South African context, we think that income distribution policies that lead to a more equitable society also promote economic growth. The document clearly looks at the issue of redistribution of income and wealth as an outcome of growth. In this perspective, growth implies redistribution. The linkages between a better distribution of income and growth is completely unexplored. We think that policies aimed at achieving a more equal distribution of income will promote growth while reconditioning the economy towards a long-term sustainable system. Recent United Nations findings regarding the relationship between equity and growth are worth summarising: It previously was thought that a trade-off existed between growth and equity - that distributing income too equally would undermine incentives and thus lower everyone's income. The assumption was that the rich needed special encouragement to save and invest more. Recent evidence suggests that this conventional wisdom is wrong. Many economies in Asia, Hong Kong, Indonesia, Malaysia, the Republic of Korea, Singapore, Taiwan and Thailand have had both rapid growth and relatively low inequality. Between 1960 and 1993 the East Asian economies, excluding China, had annual per capita growth of 7.6 percent, while income inequality remained TRANSFORMATION 31 (1996) 85 COMMENTARY ADELZADEH stable or declined. Japan and Sweden have also combined rapid growth with low inequality. These are important findings, since they contradict the conventional view that it is better to channel income to the rich, who tend to save and invest more. {Human Development Report, 1996:52)... Whether international or national, increasing income inequality is a major constraint to sustaining both economic growth and human development. Intra-generational equity is as important as inter-generational equity. (Human Development Report, 1996:13). On Growth, the Labour Market and Employment3 The government's macroeconomic strategy will lead, at best, to limited employment gains. Its own figures show unemployment almost unchanged by the year 2000. While government's intention to focus on job creation is welcome, the strategy is dependent on wage moderation, foreign investment and a host of incentives for the private sector. Given the last two years' dismal record of the private sector in employment creation, this strategy, in which the projected growth rate is almost completely dependent on government policy stimulating private investment, is overly optimistic regarding the private sector's employment creation. The document is too quick to downplay the leading role that the government should play in this regard. Moreover, the document's underlying estimation of the demand for labour is based on the assumption that the level of employment is only affected by the real wage rate; it follows thatflexibilityin the labour market, ie flexibility of the wage rate, is the only factor that can ensure a rise in employment. However, the already flexible labour market in South Africa (see below) did not yield significant increase in formal employment whether induced by the growth in the GDP or by triggering substitution of capital for labour in the production process. The document overlooks the crucial role that the overall level of demand in the economy plays in determining employment An increase in the aggregate demand (stemming from one or more than one component of the overall demand in the economy) will raise the demand for labour through its direct and positive effect on production. Government's new strategy seeks to generate 1.35 million new jobs over five years. This represents an average employment growth rate of 2,9 percent up to 86 TRANSFORMATION 31 (1996) ADELZADEH COMMENTARY the year 2000. Given the labour force growth rate of 2,5 percent, such a strategy will have little impact on the unemployment rate. According to this strategy, government's contribution to the rise in employment will flow from increases in its capital expenditure budget. However, government's planned infrastructure spending will kick in only at the back-end of the five year period and is thus conditional on the success of the market-driven strategy during the first three years of the programme. This issue combined with those that we have raised in this paper regarding the feasibility and soundness of the overall growth framework, raises serious doubts regarding government's ability to fulfil its minimal role in job creation in the proposed growth framework. The proposed macroeconomic strategy assumes that the labour market is inflexible and calls for a more flexible labour market. Since the government document does not provide analytical or empirical evidence to support its assumption regarding the inflexibility of the labour market in South Africa, we raise the issue: is the labour market inflexible in South Africa? If not, why raise it? The recent ILO report, the most comprehensive study done on South Africa's labour market, argues that the labour market is generally flexible, and in some cases, it is too flexible. In terms of employment flexibility, the ILO report on South Africa observes that many workers have minimal employment protection, and are easily retrenched. Furthermore notice periods are short or non-existent and firms can resort to temporary or casual labour or even to contracting (Standing et al, 1996). Clearly, there is ample employment flexibility in South Africa, and in fact, more regulation is needed. In terms of wage flexibility, the ILO report and the LRS concur that there is a significant wage flexibility in South Africa's labour market. An example of wage flexibility, according to the LRS, is the motor industry. In this industry, wages have tended to go up when GDP increased, and down when the economy goes into recession. Moreover, wages as low as under R100 per week (half the poverty line) have in fact suffered a fall by 13 to 16 percent in real terms since 1992. The ILO concurs with this by pointing out that industrial councils cover a small number of workers, and often only partially. Moreover, exemptions from these agreements are widespread (about 17 percent of firms apply and get exemptions easily, according to the ILO). On the issue of functional flexibility in South Africa's labour market, which includes working time and working week arrangements, it is clear that South African workers work too many hours and shifts are underutilised; the opportunities for beneficial and innovative flexibility is greater here. The ILO, finding that 82 percent of firms use temporary labour and 45 percent contract 87 TRANSFORMATION 31 (1996) COMMENTARY ADELZADEH labour, concluded that collective bargaining is needed to ensure 'equitable' employment conditions for these workers. Government proposes the 'pursuit of regulated flexibility aimed in part at extending the protection and stability afforded by this regulatory framework to an increasing number of workers' (pi 7). Firstly, by simply proposing extending regulation, government's strategy stops short of the Labour Market Commission (LMC) proposal that the regulatory net should coverall workers and sectors. The LMC had recognised the existence of a segmented labour market, but tried to overcome it by proposing regulated flexibility. This entails covering all workers but allowing for regulated differences, as not all workers would be able to enjoy the best conditions and agreements without having a negative impact on profitability in some sectors. The second part of government's proposal is that it wants to exempt certain categories of workers and the economy from regulation. These measures would include lower wages for young workers, exempting smaller businesses from industrial agreements, and giving even wider ministerial discretion as to whether industrial agreements should be extended to non-parties. We are concerned that (a) lower rates for youth employment may result in youths being substituted for other workers (for example, women). In any event, according to the ILO, youth are often already employed at lower rates; (b) wages are not the only constraint facing small, micro and medium enterprises - access to cheap credit, technical skills and marketing count more; and, (c) the right of the Minister to prevent agreements being extended would undermine collective bargaining, and should be challenged. An alternative would be to explore an LMC option of having smaller qualifying firms being subjected to a lower schedule in industrial agreements. However, we would add the caveat that smaller firms must apply to the council, not the Minister (nor get automatic exemption), to be exempted from the higher schedule. Related to increased flexibility is variation, referred to as 'variable application of employment standards'(pl7). In terms of the document's proposal, variation envisages collective bargaining, individual contracts of employment and the Employment Standards Commission being mechanisms which could lower employment standards. However, we think the government should adopt labour's model of a basic floor of rights, which can only be varied upwards or improved upon. This will ensure that even the most vulnerable workers are protected. According to the government's strategy, there is likely to be a further proliferation of 'atypical' employment (through domestic service and tourism). These workers already rank among the most unorganised and exploited workers. Both variation and voice regulation, which refers to participation in public policy 88 TRANSFORMATION 31 (1996) ADELZADEH COMMENTARY making, are important when considering atypical (mostly temporary and casual) employment. The document does not provide answers to questions such as what form of voice regulation will the atypical workers have, how will they be represented, and what will be their employment conditions? Government's strategy emphasises the need for productivity enhancement and stresses that wage increases should not exceed productivity gains. To enhance productivity, the document highlights the importance of a stronger focus on training with a possible mandatory levy on payroll for funding training. While such a proposal is a step in the right direction, any discussion on productivity should also consider the following issues: Productivity Issues Theoretical and empirical work on the labour market indicate that in many instances low wages translate into low labour productivity; and although insufficiently skilled workers also contribute to low productivity, the LMC and ILO point to poor management as a key source for low productivity in South Africa. Moreover, international studies points to an empirical link between productivity and capacity utilisation. Local evidence shows thatunderutilisation of capacity (caused by low demand) accounts for much of the decline in productivity between 1980 and 1990. Therefore higher levels of demand will contribute to higher productivity. Expansionary, but focused and prudent, fiscal policies could therefore improve productivity. Clearly, attempts to improve productivity must go beyond wages. This leads us to the important point that a large number of issues associated with greater 'productivity' are not in the hands of workers. These include investment decisions, volumes of output within a factory, the type of product produced, the type of technology employed and importantly the efficiency and conduct of management. The ILO also includes the need for skill formation, social equity, work security and industrial democracy. According to the ILO and the LMC, encouraging multi-shifts improves capacity utilisation and productivity. However, there must be serious safeguards against the negative aspects of shift work including tighter health and safety regulations, and longer leave for shift workers (they usually die at a younger age). Also linked to shifts is the need to increase the efficiency of the public transport system. Similarly, a shorter working week (40 hours) could have a significant impact on productivity. TRANSFORMATION 31 (1996) 89 COMMENTARY ADELZADEH Wage Moderation Versus Closing the Wage Gap Government wants wage moderation (including wages and salaries) to keep wage increases within productivity gains. However, government's strategy explicitly avoids dealing with the existing wage gaps in formal employment (between skilled/ unskilled or white/ black), instead focusing on the wage gap between regulated and unregulated employment, (pi7). The apartheid wage gap has resulted in managers earning about fifteen times the wages of the unskilled worker. This compares negatively with the East Asian tigers, where the differential is about three times. Moreover, the BLO report provides evidence that countries with large wage differentials are less productive than those with lower differentials. Clearly, then, the formal sector wage gap must be addressed. Government's strategy of wage moderation allows for a one percent real increase every year. However, this is an aggregate figure, and government implies that skilled wages/ salaries (spurred on by alleged skill shortages) would probably account for that increase. So, unskilled workers are likely to see real wage cuts. This version of wage moderation will not only perpetuate an apartheid legacy, but also contribute towards lower productivity. Such widening of differentials will also be unsustainable, even within a national social agreement. As the LMC notes, income inequality places upwards pressure on wages as workers fight for fair pay. Moreover, the weight given to alleged skills shortages may be excessive. Firstly, the BLO Report finds little real evidence of real skills shortages; like the inflexible labour markets it could be a self-sustaining myth. Secondly, local management is often inefficient and therefore cannot justify higher wage increases than to ordinary workers. Thirdly, if there is a real shortage of some skills which cannot be met in the short-term by more training, a solidaristic solution would be importing skilled workers - the LMC suggests Cuba and India. Rather than wage moderation, private sector provision of non-wages benefits could be made an issue. The private sector has fought hard for the privatisation of wafer benefits (retirement, medical aids, etc), the result being duplicated and expensive systems that do not cover many workers. For example, there are about 16 000 private retirement funds, with high administration costs and accordingly low benefits. Centralising and rationalising these funds could see wider coverage and better benefits at lower cost. This would be in line with the RDP's objectives. Both government's strategy document and the LMC Report propose the creation of a national agreement/ accord, complemented by sectoral and regional agreements. It is not this paper's purpose to assess whether a social agreement is generally good or bad, but only what and how these major issues are dealt with. 90 TRANSFORMATION 31 (1996) ADELZADEH COMMENTARY Government has, essentially, two accords in mind. In the longer-term, it proposes a 'broad national agreement' (p20) that looks at a wide range of issues, including investment, delivery of services and equity. However, its short-term or 'immediate objective' is to focus narrowly on wages, ostensibly because of the rapid depreciation of the Rand. The inter-relationship between the two is not clearly explained. Moreover, as a policy instrument, accords, even in the short-term, are unacceptable if they focus on wage restraint or 'moderation' for the following reasons: (a) an income policy advocating wage restraint (for lower paid workers) could merely freeze or perpetuate inequality and poverty; (b) conventionally the trade-off for wage moderation has been an increase in the social wage. In South Africa, employers rather than the state have been the main contributors to the social wage. However, this has not been widespread and has not covered most workers. And while the government's market-driven strategy ostensibly brings an increase in social and community living standards (although at a minimalist level) into a possible accord/incomes policy, if foreign and local investment is not forthcoming, this will be a lame-duck strategy. According to various reports, government intends eventually to do away with exchange controls altogether. Therefore it will be extremely hard to hold business to any social agreement - after all, unlike government and labour (which are tied to South Africa) capital will be mobile and able to withdraw easily. The lessening of exchange controls therefore not only deprives South Africa of scarce investment capital, but also strengthens the already strong hand of business in any social agreement. Any social agreement would need to ensure that all partners (especially business) can be held within the agreement. In short, an agreement or accord must not be confined to an income policy only. After all, both the government strategy and the LMC report identify the importance of policy co-ordination in achieving the goal of employment creation. These issues include determining an appropriate role for government in reconstruction and development, fiscal policy, private sector investment, role of foreign direct investment, industrial policies to encourage SMEs, the extension of basic services, targets for redistribution and economic transformation, and labour market policies. All of these issues should be blended into a broad social agreement. Summary and Conclusions In this paper we have been concerned with the problems associated with the government's growth strategy and policy document from analytical and empirical points of view. Our criticisms of the government document stemmed 91 TRANSFORMATION 31 (1996) COMMENTARY ADELZADEH from our main concerns as to whether the proposed growth framework and policy scenarios (a) lead to substantial increase in the growth rate of the economy, (b) reverse the unemployment crisis in this country and reduce the unemployment rate significantly, and (c) yield sufficient progress towards an equitable distribution of income and wealth. Some of our findings are: • The document fails to present an analytically sound and empirically justified strategy and policy scenario that integrates the main RDP objectives. The document's primary concern is to boost investor confidence by adopting the main tenets of neo-liberal strategy and policies. This unduly conservative macroeconomic framework far from achieving the RDP objectives, will constrain growth, employment, and redistribution. • Despite the document's claim that the proposed scenario is integrated, one striking feature of the document is the lack of integration between various issues under discussion and how they might impact upon one another. For example, the framework treats redistribution as an outcome of growth and does not consider the direct effects of redistribution on growth and employment; it does not integrate into the proposed scenario the growth, employment and income distribution effects of restructuring government expenditures; the document has not established enough linkages with the various White Papers that have been issued by different Government departments over the past two years. Moreover, the proposed integrated scenario suffers from policies that tend to work antagonistically (ie the promotion of a fast-growing economy is accompanied by tight monetary policy that has the effect of constricting economic activity; decreasing public consumption and increasing the provision of social services). • Since the Reserve Bank model cannot provide information on the income distribution effects of policies, is unable to furnish sectoral employment effects, and is unable to incorporate the effects of restructuring government expenditures on growth, employment and income distribution, we do not think the government should have used this model for assessing the feasibility of policy packages that are expected to be concerned with employment creation, more equitable distribution of 92 TRANSFORMATION 31 (1996) ADELZADEH COMMENTARY income and restructured government's expenditures along the line of new priorities. The growth rate projected by the strategy is based on a number of questionable assumptions: 1) that 'crowding out' is an important phenomenon in South Africa, 2) that deficit reduction will result in a fall in the interest rate, 3) that an increased current-account deficit is consistent with a lower interest rate, and 4) that a lower interest rate will impart a strong stimulus to private investment. In other words, in the document the government provides very little fiscal stimulus to reach the required growth target, and success is almost wholly dependent upon the response of the private sector. The adopted contractionary monetary policy of a high real interest rate, above the GDP growth rate, also dictates a contractionary fiscal policy in order to preserve the target deficit-GDP ratio; that is, in the integrated scenario, government's expenditure policy is taken hostage by the contractionary monetary policy of the Reserve Bank. Moreover, building a contractionary monetary policy into the growth framework goes against the main objective of the document by deterring the economy from attaining a significant growth rate. A careful reading of the reported forecasts for the proposed growth framework indicates that income distribution will deteriorate during the course of the programme. This is contrary to the RDP's statement that 'our growth path must ensure more equitable distribution of incomes'. The document is weak in the area of trade, small and medium size enterprise and there is a complete absence of any proposal for an industrial policy or strategy. Furthermore, we think the document's target of an average yearly growth rate of 8.4 percent for the non-gold export sector is unrealistic. The document clearly looks at the issue of redistribution of income and wealth as an outcome of growth. In this perspective, growth implies redistribution. The linkages between a better distribution of income and growth is completely absent. We think that policies towards a more equal TRANSFORMATION 31 (1996) 93 COMMENTARY ADELZADEH distribution of income will promote growth while positioning the economy towards a long-term sustainable system. Overall, the proposed growth framework and policy scenarios are analytically flawed, empirically unsupportable, historically unsuitable for this country, and if implemented will lead to disappointment and failures in achieving the RDP objectives of fundamentally transforming the inherited patterns of inequality. For a macroeconomic framework and policy scenarios to be viable and consistent with the RDP, they should be guided by and incorporate the following issues: • The framework must explicitly include the links between income distribution and growth. In South Africa policies aimed at achieving a more equal distribution of income also promote economic growth. At the same time, the framework must incorporate the impact of various policies (fiscal, monetary, private and public investment, trade and industrial) on income distribution, to assist in guiding policy towards the correction of South Africa's highly skewed distribution of wealth. • The impact of government expenditure (as it is restructured and reprioritised) on growth, employment levels and income distribution should be incorporated into the framework. If the present modelling assumptions persist, all government action will be deemed, by definition, to be ineffective. This is obviously a highly inappropriate modelling style for use by a developmental government guided by the RDP. • The employment effects of various policies should be properly incorporated into the framework as the present model does not do this. In addition, an employment-level target should be set as an overall policy guide. • Monetary policy should be integrated with other aspects of macroeconomic policy. The various consequences of reduced real interest rates should be incorporated, as should the use of differential interest rates as a way of achieving the priorities of an industrial strategy through appropriate incentives and disincentives. • The macroeconomic framework should incorporate the impact on capital outflow of easing existing exchange controls, particularly as such 94 TRANSFORMATION 31 (1996) ADELZADEH COMMENTARY controls presently only apply to South Africans and place no restriction on foreign investors. • The overall economic framework should include an explicit rejection of any form of two-tier (or multiple-tier) labour market. • Lastly, a social contract involving labour, business and government should be conditional on an agreement on the overall growth and development framework, including government's macroeconomic framework. NOTES 1 In July 1996, the National Institute for Economic Policy (ND3P) was commissioned by COSATU to write an assessment of GEAR. Based on the work done for COSATU, the NIEP later on released its critical review of GEAR which is reprinted here. I am grateful for comments from NALEDI's research team and other colleagues notably, B Fine, J Michie, J Newton, M Samson, J Sender, J Weeks and H Zarenda. The NEP and myself, however, take the final responsibility for the views expressed in this paper. 2 The rest of this section is based on J Weeks' article on 'Macroeconomic strategy: implications for the North West province', July 19%. 3 This section has drawn substantially from NALEDI's policy memo, 'Government's macroeconomic strategy: labour market issues', July 1996. REFERENCES ANC (1994) Reconstruction and Development Programme, Johannesburg: Umyanyano Publication. Fallon, P and Luis Perriera da Silva (1994) South Africa: economic performance and policy, World Bank Discussion Paper, No 7, Johannesburg: World Bank South African Department. NALEDI (1996) Government's macro economic stragegy: labour market issues, Johannesburg. Standing, G, J Sender and J Weekes (1996) Restructuring the Labour Market: the South African challenge, ILO Country Review, Geneva: ILO Publication. United Nations Development Project (1996) Human Development Report, New York. World Bank (1993) Paths to Economic Growth, Washington D.C. TRANSFORMATION 31 (1996) 95