1. Real sector developments

After a cumulative fall of 3.5 percent between 1989 and 1992, GDP at market prices grew by 1.1 percent in 1993 (Appendix Table 1). The major contribution to growth came from the turnaround in the inventory cycle, with positive investment in inventories recorded for the first time since 1989. Private consumption expenditure remained subdued in 1993, rising by only 0.5 percent; by contrast, public consumption grew by 1.8 percent in 1993 (although at a slower pace than during the recession). Gross fixed investment fell in 1993 for the fourth year in succession, primarily as a consequence of lower parastatal investment (Appendix Table 2).

Net exports in 1993 were buoyed by continued strength in manufactured exports and recovery from the drought, and, in contrast to 1991 and 1992, made a positive (albeit small) contribution to GDP growth. Thus external transactions remained a use rather than a source of domestic savings and contributed negatively to the financing of domestic investment (Appendix Table 3). Government dissaving also increased in 1993, so that investment had to rely to an even greater extent on private domestic saving. ^1/ Higher private saving came primarily from the corporate sector, although personal saving also rose as a proportion of disposable income (Appendix Table 4). Net investment increased slightly, but remained barely positive in 1993. ^2/

In 1994, the economy was buffeted by political uncertainty related to the elections and by a spate of midyear strikes. Nonetheless, the latest available data suggest that the recovery has emerged intact: GDP (seasonally adjusted) is estimated to have grown at an annual rate of 2.6 percent in the third quarter. The prime contributor to growth in 1994 was fixed investment: the stimulus came from both private investment in large resource-based projects, ^3/ and a recovery in parastatal investment. However, with imports surging at an unprecedented rate and a weak performance by exports (see Section 4 below), net exports are likely to have made a negative contribution of 1½ percentage points to growth in 1994.

On the supply side of the economy, the major source of growth in 1993 was a revival of agricultural activity after the 1992 drought (Appendix Table 5). Despite constituting less than 5 percent of total value-added, the agricultural sector contributed half the growth in GDP at factor cost recorded in 1993. The recovery of nonagricultural activity--led by mining (Appendix Table 6), electricity and water, and transport and communications --was considerably less spectacular; nonagricultural GDP rose by only 0.6 percent in 1993. Moreover, with the construction sector contracting for the third year in a row and no growth recorded in manufacturing (Appendix Table 7), industry remained depressed.

Despite relatively normal weather conditions, agricultural activity remained unstable in 1994. Nonagricultural value added, by contrast, continued to record low but positive growth, with seasonally adjusted annualized rates of growth of 0.4 percent, 0.7 percent, and 1.8 percent in the first three quarters of 1994, respectively. Within the nonagricultural economy, mining activity was weak--one explanation is that lower grade gold and platinum deposits were worked on account of higher prices, but industrial action also seems to have reduced output.

Manufacturing activity initially remained subdued in 1994, although capacity utilization began to increase in the second quarter and by the third quarter, value added in manufacturing was growing by 1.8 percent (annualized). There was encouraging growth in both the construction and electricity sectors in 1994. Tertiary activity also grew strongly, principally on account of the financial services, and transport and communications sectors.

Employment losses, both during the recession and in the period since, have been enormous: 150,000 jobs disappeared in the gold mining sector between 1988 and 1993 (representing a 30 percent reduction) and a similar number of losses were recorded in manufacturing (a 10 percent reduction). Overall, nonagricultural employment fell by 6.7 percent between 1988 and 1993 (Appendix Table 8). If the public sector is excluded, ^1/ the picture is even bleaker: at the end of 1993, private nonagricultural employment was 9.4 percent lower than it had been five years previously.

The free fall in employment, which began with the recession, appears, at least initially, to have continued into 1994; the most recent data available show further job losses in the first quarter, amounting to 1.2 percent of end-1993 employment, with the mining sector especially hard hit. The only bright note in the first quarter of 1994 was that manufacturing employment (seasonally adjusted) rose by almost 2 percent over end-1993.

Unemployment data, which show that 44 percent of the labor force lacks formal employment should be interpreted with care, particularly as the same data indicate a 12 percentage-point rise in this rate over the past five years (and that less than 20 percent of the labor force was jobless as recently as 1980). There are a number of methodological issues concerning the unemployment data, including questions about why the participation rates used to estimate the economically active population have risen in recent years despite the dearth of employment opportunities. But it is clear that the major part of the jump in formal joblessness is genuine and is borne out by the trend in the nonemployment rate, which is the proportion of the population of working age without formal employment (see Table 2). The nonemployment rate--which is impervious to estimates of participation rates--has risen by 4-5 percentage points over the past five years (and by 11 percentage points between 1980 and 1993).

In essence, the unemployment data reflect some simple arithmetic about the South African economy: between 1988 and 1993, there was almost no growth in formal employment, whereas growth of the population (of working age) averaged 24 percent per annum. The same point can be made by examining GDP growth data: with the labor force rising by about 3 percent per annum and labor productivity increasing by up to 1 percent--and with unchanged relative factor prices--GDP growth has to amount to 4 percent to keep unemployment constant. However, GDP growth was 4 percent or more in only two of the last ten years (1984-93) and five of the last twenty (1974-93), whereas it was 4 percent or more in 12 out of 13 years between 1961 and 1973.

Nevertheless, there is more to the slowdown in job creation than low GDP growth: indeed, capital intensity has risen appreciably since 1980. The obvious explanation for this trend is that it was dictated by rapidly rising labor costs, both in the direct sense, ^1/ and in the indirect sense, in that industrial relations were deteriorating markedly. ^2/ These developments are correlated in time with the growing strength of trade unions (see Chapter III). But other factors were also important. In the second half of the 1980s, the repeal of the apartheid labor laws reduced the cost of adjusting employment levels and made firms less inclined to hoard labor during a cyclical downturn. The Government also played a role in promoting capital intensity, both through investment in strategic industry (Sasol, Mossgas) and more recently through Section 37e tax concessions (see Chapter V, which compares effective tax rates on labor and capital).

The rise in formal joblessness described above would be less cause for concern if the informal sector had been rapidly absorbing labor. Indeed, apartheid implied major constraints on informal activity and some argue that its demise has allowed the informal sector to take off. The available data, however, are not supportive of the thesis that the informal sector has compensated for the decline in formal sector labor absorption. Certainly parts of the informal sector (such as the taxi industry) have flourished since the 1980s. But recent survey evidence suggests substantial unemployment even after account is taken of informal activity. ^1/ Survey data indicate that about 70 percent of economically active South Africans currently work in some capacity--whether it be formal or informal, regular or casual, full-time or part-time--and that one sixth of those working would like more work. This implies an aggregate unemployment rate (rather than a formal joblessness rate) of about 30 percent and an underemployment rate some 10 percentage points higher. ^2/ These statistics suggest that even if the informal sector has grown substantially over the past ten years, it cannot have compensated for the drop in the rate of formal sector labor absorption.

The downward path of inflation continued through 1993. Producer price inflation reached a historic low point of 5 percent in the year through the third quarter (Appendix Table 10). Consumer price inflation--which at the beginning of 1993 was at its lowest level in twenty years--was nudged upward in April by the effects of a 4 percentage point increase in the rate of value-added tax (VAT); by year-end it had returned to single digits with the underlying rate of inflation at about 8 percent.

Price developments in 1994 were less promising and both producer and consumer price inflation picked up during the year. Two influences on inflation are particularly noteworthy: a high rate of food price inflation during the year and a rapid nominal depreciation of the rand in the first half of the year. Food prices--and in particular, the price of meat, which rose as farmers reduced supply to the market in order to restock herds in the wake of the drought--boosted the level of inflation in 1994. But the consumer price index exclusive of food also began to accelerate during the year (Chart 2); by November, nonfood inflation had caught up with overall inflation.

View Full Size

SOUTH AFRICA PRODUCTION AND CONSUMER PRICES, 1990-94

(12 month percentage change)

Citation: IMF Staff Country Reports 1995, 021; 10.5089/9781451840919.002.A001

Source: South African Reserve Bank, Quarterly Bulletin.

• Download Figure
• Download figure as PowerPoint slide

The effect of the rand depreciation on inflation is difficult to discern, given that the rate of change of both the imported and domestically produced components of the producer price index picked up at the same time. This could be interpreted as indicating that the depreciation had an immediate impact throughout the economy, except that, as noted above, food price inflation was simultaneously increasing for other reasons.

2. Fiscal policies

The fiscal position of the Central Government, which worsened sharply between 1989/90 and 1992/93, has improved significantly over the last two years, despite continued expenditure pressures. ^1/ The deficit (excluding extraordinary revenue) rose from 2 percent of GDP in 1989/90 to 8½ percent in 1992/93 (Appendix Table 11; Chart 3). While the recession accounted for part of the deterioration of public finances, the cyclical deficit component in 1992/93 is estimated by staff at only about 1.5 percent of GDP, indicating that most of the rise in the deficit reflected structural factors. ^2/ The fiscal stance firmed in 1993/94, as reflected in a decline of the deficit (excluding extraordinary revenue) to 7.3 percent of GDP despite continued weak growth, and the 1994/95 budget targets a further decrease of the deficit to 6.6 percent of GDP.

View Full Size

SOUTH AFRICA GOVERNMENT FINANCES, 1989/90–1994/95

(In percent of GDP)

Citation: IMF Staff Country Reports 1995, 021; 10.5089/9781451840919.002.A001

Sources: South African Reserve Bank, Quarterly Bulletin; and Department of Finance.1/ Central Government.2/ Fiscal year beginning April 1; budget estimate for 1994/95.3/ General government.4/ End of fiscal year.

• Download Figure
• Download figure as PowerPoint slide

Central government revenue (excluding extraordinary revenue) declined by 1.5 percentage points of GDP between 1989/90 and 1993/94 (Appendix Tables 12 and 13). While overall income tax collections fell by 1 percentage points to 12 percent of GDP over the period, the buoyancy of corporate and individual income taxes differed significantly. Corporate taxes on mining fell sharply, declining from 1 percent of GDP in 1989/90 to 0.2 percent of GDP in 1993/94, reflecting, inter alia, lower gold mining profits as gold prices stagnated but costs of mining operations rose steadily. ^3/ A concomitant decline in corporate taxes on nonmining companies was due not only to the recession but also to a cut in the corporate tax rate from 48 percent to 40 percent in 1993/94, which was only partly compensated by the introduction of a new 15 percent tax on distributed dividends (Secondary Tax on Companies, or STC).

In contrast to corporate taxes, individual income tax collections rose by some 14 percentage points of GDP between 1989/90 and 1993/94, largely because of inflation-induced bracket creep. As regards indirect taxes, the replacement of the general sales tax with the VAT at a rate of 10 percent in September 1991 caused a significant decline in revenue, which, however, was fully reversed by the increase in the VAT rate to 14 percent in the 1993/94 budget. Extraordinary revenue accrued unevenly over the period, reflecting privatization revenue of 1 percent of GDP in 1989/90 and, in more recent years, the proceeds from sales of strategic oil reserves. ^1/

Central government expenditure rose by 4.5 percentage points of GDP between 1989/90 and 1992/93. However, the sharp upward trend in expenditure was arrested in 1993/94 when expenditure fell by more than a half percentage point to 31 percent of GDP (Appendix Table 11 and Chart 3; also see Appendix Table 14). Of the expenditure increase between 1989/90 and 1993/94, 1½ percentage points reflected higher interest payments, while the remainder was due mainly to higher spending on education, health, welfare, and other social services. A breakdown of general government expenditure by function indicates an increase of spending on social services of some 4½ percentage points of GDP over this period (Appendix Tables 15 and 16).

The significant increase in social spending between 1989/90 and 1993/94 reflects chiefly a shift in policy under the previous Government toward providing black South Africans with more equal access to services: growth of appropriations for nontertiary education and primary health care outpaced growth of the other main spending categories in the health and education budgets, and racial disparities in payments of social pensions and other social assistance payments were abolished on September 1, 1993. ^2/ Defense expenditure has fallen by almost 1½ percentage points of GDP since 1989/90. However, most of the defense savings have been diverted to internal protection services, mainly the police, leaving the costs of overall protection services unchanged at 6 percent of GDP.

The Central Government’s debt-GDP ratio, which stood at 38.1 percent at the beginning of 1989/90, rose to 52.5 percent at the end of 1993/94. However, the increase in the debt ratio can be fully accounted for by fiscal transactions that were not reflected in above-the-line deficits. ^1/ Two major categories of transactions that increased outstanding debt, without being reflected in above-line deficits, can be identified from financing data compiled by the Reserve Bank (Appendix Table 17). First, the provision of forward cover for foreign borrowing through the Reserve Bank has resulted in significant losses. Upon realization, these losses are recorded in the Reserve Bank’s Gold and Foreign Exchange Reserve Contingency Account but have no immediate effect on the Government’s fiscal accounts. ^2/ Second, actuarial underfunding of state pension funds led to special below-the-line transfers totaling R 11.3 billion between 1989/90 and 1993/94. ^3/ Two other types of transactions also added to the stock of debt without being reflected in above-the-line deficits. Some government stock was issued at coupon rates below market interest rates, resulting in a discount on government paper. As debt statistics are compiled at book values, the discounts add to the stock of outstanding debt although they are not reflected as part of below-the-line financing. Finally, debts of the former homelands, which resulted from overdraft facilities and were guaranteed by the Central Government, amounted to R 15.4 billion by the end of 1993/94 and were incorporated into the central government debt register.

The first budget of the new Government, which was presented in June 1994, ^4/ projects a deficit of 6.6 percent of GDP for 1994/95. Total expenditure growth is projected to remain below the growth rate of nominal GDP, despite transition cost overruns, related to the April election, of 0.5 percent of GDP. Expenditure savings were anticipated primarily in the areas of economic services (transport and communications, and export promotion). On the revenue side, the 5 percent surcharge on imports of capital and intermediate goods was removed effective June 1994, which is expected to reduce revenue by about 0.3 percent of GDP on a full-year basis. The corporate tax rate on undistributed profits was cut from 40 percent to 35 percent, while the STC rate was increased from 15 percent to 25 percent. The resulting reduction in net revenue is estimated at 0.1 percent of GDP. However, the cut in the corporate tax rate also widened the gap between the top marginal income tax rate (43 percent) and the tax rate on net income from incorporated activities (now 35 percent) and has the potential to encourage tax arbitrage. The budget also made inflation-related adjustments to specific excise duties and allowed for a modest amount of fiscal drag. Finally, to finance transition costs, the budget introduced a temporary surcharge of 5 percent on annual personal and corporate incomes exceeding R 50,000, projected to yield 0.6 percent of GDP in 1994/95 and 0.2 percent of GDP in 1995/96.

The budget provides for an allocation of R 2.5 billion (0.6 percent of GDP) to the newly established Reconstruction and Development Program (RDP) fund, which was financed by an equivalent reduction of planned allocations to spending departments. The RDP fund allows rapid mobilization of resources for special initiatives to redress inherited inequities. It is envisaged that the amounts to be made available to the RDP fund will increase by R 2.5 billion in each budget over the next five years. More broadly, a White Paper on the RDP released in November 1994 proposed a fundamental reorientation of present expenditure priorities; it is expected that zero-based budgeting will be used as a lever to ensure that all expenditure allocations are consistent with RDP objectives.

Programs in each of the critical social sectors--health, education, housing, and land reform--are currently being formulated. Plans appear to be most advanced in housing, where an impressive consensus has been forged: a housing accord was signed in October 1994 by all players in the housing market including the Government, the construction industry, the trade unions, the financial institutions, and even the homeless. The housing program, as currently envisaged, will consist of a one-time subsidy scheme--eventually benefiting up to 300,000 families a year--and mortgage indemnities against political risk to facilitate private sector financing of low cost housing. The electricity parastatal, Eskom, has also embarked on an ambitious electrification program.

The issue of restitution for those dispossessed of their land under apartheid is being addressed with the establishment of a Lands Court, but the potential budgetary cost is small compared with that of achieving the ANC’s original land redistribution objectives; land redistribution, however, does not appear to be currently prominent on the political agenda. ^1/ Progress in health and education is difficult to judge. The Government has introduced free health care for pregnant women and for children under six years, but this initiative was not part of a fully developed health strategy and the cost implications are largely unknown. Education reform is a political imperative, but little concrete has yet been achieved.

In June 1994, an independent tax commission was appointed (the Katz Commission) to conduct a review of selected aspects of South Africa’s tax structure and, in particular, to provide recommendations for tax changes to be included in the 1995/96 budget. The Commission’s interim report was submitted to the Government in late November 1994. As regards the personal income tax system, the report recommends, inter alia, elimination of discriminatory tax treatments based on gender and marital status, elimination of child rebates, reducing the number of brackets of the standard tax schedule from 10 to 5, a flat rate of 9 percent income tax on annual taxable income up to R 30,000 and an increase of the top marginal rate from 43 percent to 45 percent, tighter caps on tax-exempt pension fund contributions, and streamlining of the tax collection system for low-income earners. The Commission estimates the annual budgetary costs of the proposed personal income tax reform package at about R 3 billion or 0.7 percent of GDP.

On company taxation, the report advises, inter alia, investigation of alternative methods of dividend taxation with a view to eventually replace STC by an imputation system, the elimination of non-resident shareholders’ tax for foreign investors controlling a minimum of 25 percent of a company, and the introduction of a presumptive tax on company gross assets, with an estimated yield per year of about R 500 million or 0.2 percent of GDP. The report strongly advocates preserving the present VAT system, in particular it argues against extending the present list of zero-rated goods and against introducing a multiple VAT rate system. Finally, the report identifies considerable scope for strengthening tax administration and tax collection. Recommended measures to this effect include endowing revenue authorities with more flexibility regarding their budget and wage structure, modernization of computer facilities, improved tax assessment and audit procedures, and enhanced taxpayer education. The report also recommends reducing compliance costs of small and medium-sized enterprises but advises against special tax incentives for these enterprises. The potential revenue gains from administrative reforms are gauged by the Commission at about R 5 billion or 1 percent of GDP.

Data for the first half of 1994/95 suggest that the deficit target for 1994/95 is likely to be met. Expenditure overruns of about R 1.5 billion or 0.3 percent of GDP, are more than balanced by a projected overperformance of revenue, currently estimated to amount to about 2.3 billion, or 0.5 percent of GDP. However, the sharp rise in interest rates has led to higher-than-projected discounts on government stock, which at R 7 billion for the first six months of 1994/95 already exceed the level budgeted for the whole year by R 3 billion. Of the R 2.5 billion allocated to the RDP fund, only about R 1.9 billion is projected to be actually spent in 1994/95.

3. Monetary policies

After several years of sustained and successful effort to reduce inflation, monetary policy through 1994 was buffeted by a series of shocks, including uncertainty ahead of the elections in April and speculation of an imminent liberalization of foreign exchange controls from August onward. These shocks greatly complicated the task of distinguishing noise from underlying evidence of the development of the stance of monetary policy. But as 1994 progressed, the signs that the monetary stance had eased, and that inflation was likely to accelerate, became clearer.

Monetary policy was tightened substantially during the period 1988-89 (Chart 4), when the discount rate was increased progressively from 9.5 percent to 17 percent, where it remained until March of 1992). The guidelines set for the growth of M3 ^1/ were progressively reduced, from 14-18 percent in 1989, to 6-9 percent in 1993. M3 growth remained within its guideline ranges--after exceeding them in 1989--with its growth rate trending downward (see Chart 5). ^2/ Inflation, however, did not respond until 1992, when it fell from over 16 percent to 10 percent.

View Full Size

SOUTH AFRICA INTEREST RATE DEVELOPMENTS, 1987–94

(In percent)

Citation: IMF Staff Country Reports 1995, 021; 10.5089/9781451840919.002.A001

Sources: South African Reserve Bank, Quarterly Bulletin: and IMF, International Financial Statistics.1/ Deflated by the 12 month CPI.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

SOUTH AFRICA SELECTED MONETARY INDICATORS, 1989–94

Citation: IMF Staff Country Reports 1995, 021; 10.5089/9781451840919.002.A001

Source: South African Reserve Bank, Quarterly Bulletin.1/ Base of guidelines is average of the fourth quarter (seasonally adjusted).2/ Total credit is inflated in March 1994 by a shift of accrued forward losses from other items net into net credit to Government.

• Download Figure
• Download figure as PowerPoint slide

From mid-1993 to early 1994, monetary developments were dominated by uncertainty arising from the political transition. There were two monetary consequences of this uncertainty. First, the narrow monetary aggregates (see Chart 5 and Appendix Table 19) grew rapidly both in absolute terms and in relation to the broader aggregates, and second, there were large unrecorded net outflows in the balance of payments.

Reserve money ^3/ grew at an annualized rate of over 45 percent in the first quarter of 1994, increasing the ratio of cash in circulation outside banks to bank deposits from 5.2 percent to 5.5 percent. Despite the increased demand for cash, banks did not raise or reduce their holdings of cash relative to deposits, indicating that the increased demand for cash on the part of the general public was handled smoothly. M1 also grew rapidly, in the last quarter of 1993 and in the first of quarter of 1994, after having remained broadly unchanged since the third quarter of 1992. In large part, these developments reflected a “liquidity preference” on the part of bank depositors in the context of the uncertainty ahead of the political transition, but other factors may also have been at work. The increased demand for cash could be attributed to ongoing adjustment to the continuous fall in nominal interest rates on bank deposits since 1990, and the rapid growth of the deposits component of M1 may also have been encouraged by the reduction in reserve requirements on short-term relative to those on long-term deposits that occurred progressively over the period after May 1993.

Net short-term outflows of the nonmonetary private sector, including unrecorded net capital flows, were running at somewhat over R 1 billion a month in the second half of 1993 (see Section 4 below). This reduced substantially the net foreign reserves of the Reserve Bank.

The Reserve Bank appeared to interpret both sets of events as transitory factors related to the political transition, rather than as signs of a weakening of the stance of monetary policy. On this basis, the monetary authorities acted to support the rand through the second half of 1993. When the paucity of reserves made this no longer feasible, rather than endanger the nascent recovery by raising the discount rate, the authorities allowed a 7-percent depreciation against the U.S. dollar between the end of 1993 and May 1994.

In April 1994, it appeared that these monetary developments were indeed transitory responses to political uncertainty. Once the participation of all parties in the election was secured, yields on ten-year government paper fell by a full percentage point, remaining at that level until mid-May, and sentiment in the foreign exchange markets improved. But the improvement in market sentiment proved short-lived; by early June, long yields were back at pre-election levels, the stock of reserve money continued to grow after the first quarter--albeit at much slower rates--rather than falling back, and the strengthening of sentiment in the foreign exchange markets proved fragile. The rand appreciated somewhat, but this only partly reversed the earlier depreciation. In hindsight, it seems that the first quarter monetary accommodation of pressures ahead of the election ultimately represented a relaxation of the underlying monetary stance.

There were signs, however, from developments in the broad monetary and credit aggregates that the monetary stance was easing even before the relaxation in the first quarter of 1994. Stimulated in part by developments in the real economy--real GDP in the fourth quarter of 1993 was 4.8 percent above the fourth quarter of 1992--M3 was rising rapidly through 1993. However, this underlying development was obscured by overfunding of the budget that occurred in mid-1993, which artificially depressed broad money and raised its velocity in the latter part of the year. ^1/ As surplus government deposits were run down in the third and fourth quarters, M3 began to Increase rapidly, though this acceleration only became visible in the 12-month growth rates in the first quarter of 1994. Thereafter, the 12-month rates rose substantially above the 1994 guidelines for M3, of 6-9 percent, peaking at 16.8 percent in July 1994, before falling to 14.4 percent in November.

These developments in broad money had their counterparts in falling net foreign assets and rapid growth in credit to nongovernment (See Appendix Tables 20 and 21). Net credit to nongovernment grew at annualized rates of 12 percent and 17 percent in the third and fourth quarters of 1993, respectively, and by a further 13 percent in the first quarter of 1994, after a period of being unchanged.

Net credit to government from the monetary sector increased by nearly R 14 billion in the first quarter of 1994. The data are inflated by R 7.5 billion of government securities issued to the Reserve Bank in compensation for realized losses on forward currency contracts. ^1/ The first quarter data also reflect the continued reduction in surplus government deposits with banks--reflecting the earlier overfunding described above--and a switch in the term structure of government funding. The Government switched to funding with short-term paper, anticipating that long-term rates would fall from the levels implicit in the yield curve at the time. Banks, the main purchasers of short-term government paper, took up most of the new issues. Government net borrowing from nonbank domestic sources dropped correspondingly.

By midyear, the signs of nascent inflationary pressures had become clear: broad money was growing well in excess of its guidelines, the 12-month growth rates of bank credit to the private sector were accelerating toward 15 percent, both consumer and producer price inflation rose from their low points in April, and imports were rising sharply. Special factors partly account for the magnitudes of these measures. The 12-month growth of M3 exaggerated underlying growth owing to the depression of the base in mid-1993, which was due to the overfunding that had occurred at that time, and to the subsequent reacquisition of claims on banks by the nonbank sector. ^2/ As noted in Section 1 above, the inflation indices partly reflected the behavior of food--and especially meat--prices. And the growth of imports reflected a number of unusual influences described in Section 4 below. But even allowing for these special factors, the evidence of increased inflationary pressures was clear.

As the signs of the underlying strength of the impulse from monetary policy became better appreciated, yields on long-term government paper began rising--from their mid-May trough of a little over 12 percent to over 15 percent by mid-August (Appendix Table 22). Though this paralleled rising yields worldwide over this period, the spread of long South African paper over equal term U.S. Government paper widened steadily.

In mid-August, the markets anticipated an imminent abolition of the financial rand, driving yields on ten-year government paper up from 15 percent to 17 percent virtually overnight, with similar increases in other long-term paper. This reaction effectively compensated foreign investors for the loss of the implicit subsidy they received under the Financial Rand system, and it reflected the expectation that abolition would weaken the commercial rand. Yields have remained at these elevated levels despite the favorable ratings obtained from the international rating agencies in September, ^1/ despite the successful global bond issue in December, and despite the efforts of the Reserve Bank to counter speculation of imminent abolition of the Financial Rand. The latter included setting four preconditions for abolition (see Section 4 below).

Following publication of the August consumer price index--which showed inflation exceeding 10 percent--and the sharp deterioration in the August trade balance, the Reserve Bank raised the discount rate from 12 percent to 13 percent. This was the first increase in the rate since 1989. It had been widely anticipated and thus did not substantially influence long-term rates.

4. The external sector

The external current account surplus increased from 1.2 percent of GDP in 1992 to 1.5 percent of GDP in 1993 (Appendix Tables 23-25 and Chart 6). Imports, especially of machinery and capital equipment, grew strongly during 1993, and import volumes increased by almost 7 percent (Appendix Table 26). The rise in imports was more than compensated for, however, by a robust export performance, a reflection primarily of the higher price of gold, but also of the end of the drought and increased exports of manufactures. The volume of exports of goods and nonfactor services rose by 6 percent in 1993, well above the growth of both domestic demand and merchandise imports in partner countries.

View Full Size

SOUTH AFRICA SELECTED BALANCE OF PAYMENTS INDICATORS, 1980–93

Citation: IMF Staff Country Reports 1995, 021; 10.5089/9781451840919.002.A001

Source: South African Reserve Bank, Quarterly Bulletin.

• Download Figure
• Download figure as PowerPoint slide

In the services accounts, the most striking developments in 1993 were a large jump in travel payments and a fall in dividend and profit payments to nonresidents (Appendix Table 27). Travel payments climbed as South Africans began to reintegrate themselves internationally; this was partly offset by a rise in travel receipts. The factors underlying the fall in payments of dividends and profits to nonresidents in 1993 are less clear, although the introduction of the tax on dividend distributions in March (the STC; see Section 2 above) may have played a role; it did not reflect a switch in Financial Rand portfolios since nonresidents were net purchasers of equities in 1993. Nonresidents were also net purchasers of bonds in 1993: nevertheless, and despite considerable short-term borrowing by the Reserve Bank, interest payments to nonresidents were almost unchanged in U.S. dollar terms over 1992.

The outlook for the external current account in 1994 appears to have been almost universally misread. Whereas most observers had projected a surplus of around 2 percent of GDP for 1994, the latest available data indicate that the current account will certainly be in deficit. The dramatic turnaround in the current account in 1994 reflects both a flood of imports--which in nominal terms were 31 percent higher in the third quarter of 1994 than in the third quarter of 1993--and sluggish exports. It is possible that August and September imports were artificially high because importers had speculated that there would be a step depreciation of the rand following exchange rate unification. Alternatively, imports could have been distorted by large capital imports associated with the Section 37e mineral-beneficiation projects. Indeed, the October statistics showed more modest import growth; however, the import surge continued apace in November and it now seems likely that import volume growth for 1994 as a whole will be about 10 percent.

Export performance has been disappointing in 1994, particularly in light of the termination of trade sanctions and the significant real effective depreciation of the rand during the year (see discussion below). Notable areas of weakness included diamonds and manufactured goods such as transport equipment. In addition, gold export volumes fell sharply in 1994. Nongold export volume growth will fall short of the growth of both world demand and trade in 1994: with the economy recovering, this poor export performance is consistent with the hypothesis that South African manufacturers treat foreign sales as a residual after satisfying domestic demand.

The capital account of the balance of payments worsened dramatically in 1993, with the deficit increasing by $3.3 billion over 1992 (Appendix Table 28). Isolating the components of the capital account that contributed most to this deterioration is complicated by the fundamental shifts in Financial Rand portfolios that have occurred over the past two years. Thus, for example, the balance on the short-term capital account of the monetary sector switched from a net inflow of $1.1 billion in 1992 to a net outflow of $1.0 billion in 1993, but this can be almost entirely accounted for by movements in Financial Rand deposits over this period (which increased by R 3.7 billion in 1992 and decreased by R 3.6 billion in 1993). ^1/

Financial rand transactions aside, the dominant influences on the capital account in 1993 appear to have been further reductions in South Africa’s foreign debt--the debt remaining within the standstill arrangement ^2/ fell by $1.1 billion during the year--and illegal capital flight. The latter appears to have been predominant in the second half of the year when the average monthly balance on the nonmonetary private short-term capital account (which includes errors and omissions) was an outflow of R 1 billion.

The evidence suggests that the illegal capital outflow continued into 1994, but began to moderate after the election. By the second half of the year, the capital account had turned around, recording a net inflow of $1.5 billion in the third quarter, which exceeded the outflow in the first half of the year; for 1994 as a whole the capital account is likely to be in surplus for the first time since 1984. Capital inflows in 1994 have been predominantly short term. The major recipient has been the monetary sector, which had recorded a net inflow of short-term capital of almost R 7 billion by end-September. Part of this is simply a function of a further change in Financial Rand deposits--which increased by R 3 billion in the first three quarters--but the remainder represents short-term borrowing by South African banks overseas. Renewed access after the final debt agreement and favorable costs of borrowing induced the commercial banks to seek finance abroad: nonofficial short-term foreign liabilities of the monetary sector rose by R 7 billion in the first nine months of 1994.

In 1994, there was also a sharp pickup in short-term inflows to the private nonmonetary sector. This arose partly from the apparent diminution of illegal capital flight. The remainder of the increase probably reflects increased trade finance available directly to firms and simply mirrors the growth in imports. Another positive development affecting the capital account in 1994 was the rolling over of about 50 percent of the repayments of standstill debt scheduled for 1994. Finally, long-term inflows in 1994 were boosted by the $750 million sovereign debt issue in December.

The foreign debt of South Africa fell further in 1993, to $16.7 billion by year end (Appendix Table 29), compared with $17.3 billion at end-1992 and $23.7 billion at end-1985 (the year in which the debt standstill was declared). Rand-denominated debt held by foreigners, the only category of foreign debt that has increased in recent years, is estimated to have amounted to a further $9 billion at end-1993.

The large capital account deficit recorded in 1993 translated into a sizable loss of reserves during the year and the Reserve Bank shored up gross reserves by drawing on its credit lines; by year-end net reserves had dropped to $1.7 billion, compared with $3.4 billion at end-1992 (Appendix Table 30). ^1/ In 1994, net reserves continued to worsen until the election and were negative in May. Subsequently, however, the capital account has improved more rapidly than the current account has deteriorated, permitting the Reserve Bank to repay some of its borrowing. Net reserves had risen to about $1 billion by end-September.

The nominal effective exchange rate depreciated by 8.9 percent in 1993, leading to a real depreciation of 2.7 percent for the year (Appendix Table 31 and Chart 7). In 1994, the rand at first fell rapidly against the U.S. dollar and by July, the nominal effective exchange rate was 13 percent more depreciated than at end-1993. In the second half of 1994, however, the combination of a rekindling of inflation and a constant rand-dollar exchange rate led to a partial erosion of the real depreciation recorded by July. The real effective exchange rate for 1994 is likely to be 3-4 percent more depreciated than in 1993.

View Full Size

SOUTH AFRICA COMPETITIVENESS, 1984–94

(Indices January 1984 = 100)

Citation: IMF Staff Country Reports 1995, 021; 10.5089/9781451840919.002.A001

Sources: South African Reserve Bank, Quarterly Bulletin: and IMF International Financial Statistics.1/ The prices of domestically produced goods divided by the prices of imported goods in the PPI.2/ Based on relative consumer prices using the Commercial Rand exchange rate.3/ Unit labor costs in the non agricultural sector divided by nongold export prices.4/ Unit labor costs in the manufacturing sector divided by manufactures prices.

• Download Figure
• Download figure as PowerPoint slide

The financial rand appreciated against the U.S. dollar--from R 4.86 at end-1992 to R 4.39 at end-1993--but continued to be highly volatile. The financial rand discount decreased from 37 percent to 21 percent over the same period. In 1994, the financial rand continued to strengthen and by the fourth quarter the discount was in the 10-15 percent range. The authorities have been quite explicit about the conditions necessary for the abolition of the financial rand. One of these preconditions--a successful return to international bond markets--was met in December with the sovereign debt issue. However, as is clear from the discussion above, the other three preconditions--a narrowing of the financial rand discount to below 10 percent, a comfortable reserve position, and a decrease in Financial Rand deposits--have yet to be completely satisfied.

5. Trade reform

For at least the last seventy years, the South African trade regime has been inward oriented. Beginning in the 1920s, import substitution was seen as an instrument of industrial development and protective barriers were erected to that end. The instruments of the protection system included quantitative restrictions, custom duties, formula duties, ^1/ and the institutionalization of mechanisms which enabled a multitude of interest groups to lobby for special treatment as and when the need arose. The cumulative result has been an extraordinarily complex and fluid tariff structure. ^2/ In addition to serving industrial policy, import controls were also seen as safeguarding the external position of the country.

Over time, the import-inhibiting bias of the trade policy regime came to be appreciated and in the early 1970s, there was a shift toward some promotion of exports. ^3/ However, the restrictive import regime was not relaxed and, in 1985, following the widespread implementation of sanctions against South Africa, import controls were reinforced. ^4/

The basic design for a more outward-oriented trade policy was first formulated by the Industrial Development Corporation (IDC) in 1990. After a comprehensive review of the trade system, the IDC recommended a package of trade reform measures to be implemented within a coherent macroeconomic framework--including prudent fiscal policies and a realistic and stable real exchange rate. The main elements of the IDC’s recommendations included the simplification and the gradual reduction of tariff protection, the replacement of formula duties by effective anti dumping measures, the elimination of import surcharges, and the improvement of the duty drawback and rebate system for exports--so as to provide exporters access to inputs at world market prices, while at the same time phasing out the General Export Incentive Scheme (GEIS). ^5/

Even before the formation of the Government of national unity and the lifting of economic sanctions, a change had begun with respect to South Africa’s views about international economic relations. This change, and the perception that the past inward-oriented trade policy had been inimical to growth, have resulted in the authorities’ acceptance of the notion that growth--and in particular growth of employment--will require trade in manufactures and services and an outward-oriented trading regime. The authorities used the opportunity of the Uruguay Round of the GATT to pursue their trade liberalization agenda.

On the basis of the IDC’s recommendations, and consultation within the National Economic Forum, South Africa submitted offers for agricultural and industrial products to the Uruguay Round of the GATT. After bilateral negotiations, its offers were accepted by other participating countries in December 1993. The agricultural offer consisted of the tariffication of all quantitative restrictions on agricultural products and the reduction of the base rate tariff equivalent by a minimum of 15 percent individually (and by 36 percent on average over the next six years as required by the GATT). ^1/ In addition, agricultural subsidies, which in any case have declined sharply in recent years, are to be reduced further. The new agricultural tariff schedule is expected to be completed by April 1995.

The industrial products offer provides for a substantial increase in the degree to which tariff lines are to be bound under the GATT agreement (from 55 percent of tariff lines to 98 percent), conversion of all outstanding quantitative restrictions on imports to equivalent ad valorem rates, and reduction of all bound tariff rates by an average of one third over five years. The authorities are also unilaterally planning a complete overhaul of what had been an exceptionally complex system: there will be a sharp reduction in the number of tariff lines, and tariff rate ceilings will be standardized at bands of 0, 5, 10, 15, 20, and 30 percent. In addition, tariff rates will escalate with the level of processing, with lower rates applied to primary products and inputs for manufactures and higher rates applied to manufactured and consumer goods. Furthermore, under the trade reform program, formula duties are to be replaced by measures consistent with the anti-dumping and subsidies codes of the GATT agreement. The new industrial tariff schedule is expected to be completed by November 1995.

It is envisaged that certain sensitive industries--motor vehicles and components, textiles and clothing, and electrical products--will receive special treatment in the trade reform process. While the reduction in bound tariff levels for these industries will be more than the average one third reduction for other industries, the phase-down period is eight years for motor vehicle and electrical products and ten years for textiles and clothing. The maximum ad valorem rate at the end of the phase-down period will be 50 percent for motor vehicle products and 45 percent for textiles and clothing. The longer phase-down period is expected to provide enough time for these industries to adjust from very high levels of protection to international competition.

The South African authorities began to reform the trade system prior to the GATT agreement, which came into effect on January 1, 1995. As noted in Section 2 above, in June 1994, the import surcharge of 5 percent on capital and intermediate goods was removed. In September 1994, custom duties on imports of fully assembled motor vehicles were reduced to 80 percent from 100 percent and the 15 percent import surcharge on motor vehicles was abolished. Furthermore, in September 1994, it was announced that GEIS, which contravenes GATT, would be revised.

The revised GEIS, which will come into effect on April 1, 1995, will have the following characteristics: (i) the guaranteed minimum net benefit level for Category 2 of 2.5 percent will be abolished immediately (see Table 3); ^2/ (ii) the maximum net benefit levels for categories 3 and 4 will be phased out over two years and nine months; (iii) certain products falling under Category 3 will be shifted to Category 2, owing to the primary nature of these products; (iv) the cutoff point for qualifying for the full incentive under each category will be reduced from 75 percent to 60 percent local content in order to increase competitiveness by enabling exporters to purchase a larger amount of imported inputs and components duty free ; (v) the exchange factor will remain in the formula, but, if it reduces the net benefit to a lower level than is recommended, the lower level of net benefit will apply; and (vi) GEIS subsidies will become taxable as income of the recipient as of March 1, 1995.

Table 3:

Current and Future Net Payout Levels under GEIS

Source: Data provided by South African authorities.

Table 3:

Current and Future Net Payout Levels under GEIS

Period	Category 1/
	2	3	4
	(In percent)
Current	2.5	3.0	15.5
4/1/95-3/31/96	--	3.0	14.0
4/1/96-3/31/97	--	2.0	12.0
4/1/97-12/31/97	--	--	10.0
After 1997	--	--	--

Source: Data provided by South African authorities.

View Table

Table 3:

Current and Future Net Payout Levels under GEIS

Period	Category 1/
	2	3	4
	(In percent)
Current	2.5	3.0	15.5
4/1/95-3/31/96	--	3.0	14.0
4/1/96-3/31/97	--	2.0	12.0
4/1/97-12/31/97	--	--	10.0
After 1997	--	--	--

Source: Data provided by South African authorities.

With the gradual phasing out of the GEIS, the authorities are considering introducing alternative export support measures consistent with the GATT agreement. These include the provision of financing to exporting firms and the improvement of the duty drawback and rebate scheme so that the tariff system does not disadvantage South Africa’s exporters in their ability to purchase material inputs and components at world market prices. To this end, the authorities have recently introduced an export guarantee scheme amounting to R 20 million for exporters who have limited access to financing; also the pre- and post-shipment export financing scheme of the IDC--which provides soft loans to small- and medium-sized export firms--has been extended to all exporters. Furthermore, the authorities are considering export measures for the sensitive industries--textiles and clothing, and motor vehicles and components--discussed above.

In response to the formation of a Government of a national unity in May 1994, South Africa has received Trading Preferences from some developed countries. In May 1994, the United States granted South Africa GSP (Generalized System of Preferences) status. In this program approximately 4,300 South African agricultural and semifinished products will be allowed to enter the U.S. duty free. The European Community (EU) granted South Africa GSP-facilities for the period covering August 1, 1994 to the end of 1994. The details of a new GSP that will come into effect in January 1995 are currently being worked out; some EU members including France would like to see a reduction of the number of South African products eligible for GSP status. On October 5, 1994 Canada granted South Africa Generalized Preferential Tariff facilities that are applicable to about 5,000 products. These products can enter Canada at most-favored-nation rates combined with certain additional tariff reductions. On May 6, 1994 Norway granted South Africa “Receiver Country” status under the Norwegian Customs Preference Arrangement for developing countries. This program will exempt South African exports, with the exception of certain agricultural and textile products, from import tariffs. Japan has also announced willingness to provide trading preferences to South African exports.

Members of the Lome convention have agreed formally to set up a Task Force to study the implication of South Africa membership: if admitted, all South African exports would, in principle, be allowed to enter the European Union duty-free. The Task Force is expected to submit its report to the Joint Council of Ministers in February 1995.

South Africa joined the South African Development Community (SADC) on August 29, 1994 and is a party to the recent SADC proposal to establish a permanent secretariat charged with the synchronization of industrial and trade policies of member countries. ^1/ Recently, Ministers from Botswana, Lesotho, Namibia, Swaziland, and South Africa met to launch a comprehensive review of the current Southern African Customs Union (SACU) agreement, which was drawn up in 1969 and amended in 1977; a Task Force was set up to review the agreement. Its terms of reference are wide and it is expected that all aspects of the agreement--including the revenue sharing formula--will come under scrutiny. The Task Force is expected to report to ministers in March 1995.