South Africa: Selected Economic Issues

This Selected Economic Issues paper examines economic development in South Africa during 1995–96. The paper highlights that in 1995, the economy of South Africa grew by 3.3 percent, the third consecutive year of economic growth, and it is expected to grow between 3½ and 4 percent in 1996. Some aspects of the unemployment problem are addressed in this paper. The paper also focuses on the implications for policy of the steps taken in 1994 and 1995 to establish an outward-oriented economy, after many years of effective autarky.

Abstract

This Selected Economic Issues paper examines economic development in South Africa during 1995–96. The paper highlights that in 1995, the economy of South Africa grew by 3.3 percent, the third consecutive year of economic growth, and it is expected to grow between 3½ and 4 percent in 1996. Some aspects of the unemployment problem are addressed in this paper. The paper also focuses on the implications for policy of the steps taken in 1994 and 1995 to establish an outward-oriented economy, after many years of effective autarky.

II. Capital Flows: Recent Developments and Policy Challenges

1. Introduction

One of the most striking changes in the economic situation of South Africa has been a dramatic shift in capital flows between mid-1994 and early 1996. Following the reduction of political uncertainty with the election of the Government of National Unity in April 1994, and the continuation of the economic recovery that began in 1993, confidence in South Africa’s prospects strengthened. By the end of 1994, the capital account registered a net surplus for the first time since 1984. After a temporary slowdown in early 1995, following the Mexican crisis, capital flows resumed and total inflows for the year reached nearly R 22 billion (US$6 billion).

Substantial capital inflows are likely to have a major impact on South Africa’s growth potential as they raise the availability of saving and hence investment and growth. But such flows may be volatile, as has been evidenced in a number of emerging markets in the recent past, and in South Africa in early 1996. In mid-February, the rand came under significant pressure as a result of unfounded rumors about the President’s health and an imminent removal of capital controls on residents. There was also uncertainty regarding the future course of policies following the resignation of the Minister of Finance in late March 1996. As a result, by end-April the rand had depreciated by about 16 percent since mid-February.

This chapter examines the factors that have contributed to the surge in capital inflows, and describes how the nature of these flows has evolved. The chapter then briefly discusses the factors that contributed to the currency attacks in early 1996 and examines the broad implications of the capital inflows for macroeconomic policy in the specific context of South Africa, including the effects of volatility of these flows.

2. capital flows: recent developments

a. Developments prior to 1994

During the period 1970–85, South Africa relied heavily on foreign saving, equivalent to an average of about 2 percent of GDP a year (Table 1 and Chart 8). However, with the intensification of political problems, the imposition of financial sanctions, and the debt standstill in 1985, 1/ the situation was reversed, and South Africa became a net exporter of capital. This involved a substantial depreciation of the currency in the period through 1989; the maintenance of restrained demand-management policies, with growth of real domestic demand during 1985–92 limited to 1 percent a year; the intensification of trade protection; and a tightening of capital controls. In response to these policies, the current account balance shifted to a surplus that averaged about 2.5 percent of GDP a year during 1985–93.

Table 1.

South Africa: Capital Inflows and Macroeconomic Balances

(In percent of GDP)

article image
Source: South African Reserve Bank, Quarterly Bulletin, and the International Monetary Fund, International Financial Statistic.

Fiscal year ending March 31.

A negative sign indicates a depreciation.

1979–84.

Including debt denominated in rand.

Chart 8
Chart 8

South Africa: CURRENT AND CAPITAL ACCOUNT BALANCES

(In percent of GDP)

Citation: IMF Staff Country Reports 1996, 064; 10.5089/9781451840926.002.A002

Source: South African Reserve Bank.

Notwithstanding the moratorium on bank lending to South Africa, some private borrowers regained access to international capital markets after 1990. In 1991–92, the Government and major public enterprises also returned to the capital markets with the placement of bonds (Table 2). However, these flows were offset by pronounced private outflows as political uncertainty intensified in the wake of the breakdown in constitutional talks in May 1992. Moreover, debt under the first interim arrangement following the debt standstill began to be amortized and margins offered to South African borrowers doubled to 3 percentage points. As a result, many borrowers elected not to roll over their debts, and thus led to net outflows of capital of R 3.7 billion in 1992 and R 15 billion in 1993 (Tables 3 and 4).

Table 2.

Bonds Issued by the Central Government and Public Companies From 1/01/90 to 2/29/96

(In millions of U.S. dollar.)

article image
Source: International Financing Review. Euroweek, Financial Times, and staff estimates.
Table 3.

South Africa: Net Capital Movements, 1993-95

(In millions of rand)

article image
Source: South African Reserve Bank and staff estimates.
Table 4.

South Africa: Net Capital Movements, 1990–95

(In millions of dollars)

article image
Source: South African Reserve Bank and staff calculations.

b. Developments in 1994–95

International confidence in South Africa’s prospects gradually improved during 1994 with the decline of political uncertainty following the elections and the continued strengthening in domestic market conditions—real economic growth doubled to 2.7 percent, investment expanded by nearly 9 percent in real terms, and inflation decelerated to 9 percent. Two other factors also facilitated South Africa’s access to the international capital markets. First, the conclusion of the 1994 Debt Arrangements normalizing the debt to bank creditors which remained under the standstill. Second, South Africa’s sovereign debt was rated investment grade by Moody’s (Baa3) in October 1994 and by the Nippon Investor Services (BBB); Standard and Poor’s rated this debt at below investment grade, BB, but with a positive outlook.

Net inflows amounted to R 5.2 billion (US$1.5 billion) in 1994, despite the continuation of substantial unrecorded outflows in the first half of the year (estimated at about R 3.1 billion) and the repayment of external long-term debt of R 5.9 billion. The inflows were primarily short-term, with the major recipient being the financial sector, as banks took advantage of the renewed access and favorable cost of borrowing to contract significant sums in the form of short-term lines of credit. This general trend was mitigated somewhat by the practice of some foreign banks of reducing their outstanding claims on South Africa at the end of their financial year. Other short-term inflows reflected increased trade finance available directly to firms, largely mirroring the strong growth of imports.

Long-term flows were dominated by the public sector with the successful placement of a global bond issue equivalent to US$750 million (R 2.7 billion) in December 1994, as well as net borrowing by public corporations. 1/ In addition, nonresidents extended loans to the large capital projects in progress, 2/ and purchased securities on the Johannesburg Stock Exchange (JSE) equivalent to R 1.2 billion on a net basis. These purchases were made through the financial-rand mechanism 3/ and therefore did not affect the foreign reserves of the country, but served to reduce the discount on the financial rand.

Despite a slowdown associated with the Mexican crisis in early 1995, capital inflows rose to R 21.7 billion (US$6 billion) in 1995, reflecting continued growing confidence in the rand. This strengthening in sentiment was particularly evident in the narrowing of the spread between yields on South Africa’s global bond and the equivalent U.S. Treasury paper, and in the spread between the long bond yield with the U.S. (Charts 9 and 10). The former—having increased following the Mexican crisis to about 250 basis points by mid-year—declined to 210-220 basis points in the latter part of 1995, 4/ and fell below the spread at the time of issue—193 basis points—in early January 1996. In addition, South Africa’s weight in emerging market indexes increased: in June 1995, it achieved the largest weighing in the IFC’s index and the second largest in Morgan Stanley’s index.

Chart 9
Chart 9

South Africa: GLOBAL BOND AND U.S. TREASURY RATES

Citation: IMF Staff Country Reports 1996, 064; 10.5089/9781451840926.002.A002

Source: Bloomsberg (U.S. Treasury: six percent maturing 10/15/99).
Chart 10
Chart 10

South Africa: LONG BOND YIELD SPREAD OVER U.S. RATE

(through April 4, 1996)

Citation: IMF Staff Country Reports 1996, 064; 10.5089/9781451840926.002.A002

Source: Financial Times and Routers.

The improved sentiment was also reflected in the development of a Euro-rand market beginning in September 1995 (Box 2), which indirectly contributed to the inflow of capital. There was also a further upgrading of South Africa’s credit rating in November 1995 by Standard and Poor’s and Moody’s. However, the firmest evidence of the turnaround in sentiment toward the rand was the sharp decline in the net open forward position of the South Africa Reserve Bank (SARB), which fell to less than US$7 billion in mid-February 1996 without any substantive disturbances to exchange or interest rates.

Long-term borrowing by the public sector continued at about the same magnitude as in 1994, about R 4 billion, reflecting the Government’s placement of bonds equivalent to US$346 million in the Japanese Samurai market in May 1995, and two subsequent placements by parastatals also in the Samurai market amounting to the equivalent of about US$400 million. The parastatals also made increased use of syndicated bank loans throughout the year, at maturities ranging between 3 years and 5 years and interest margins generally less than 100 basis points.

The Growth of the Euror and Market

Beginning in September 1995, a series of rand-denominated bonds were issued in European market by non South African entities, including Merrill Lynch, EBRD, and the Government of Sweden. Total issues amounted to US$1.8 billion at mid-February 1996 (Table 5). The coupon rate fell from 15 percent on the initial issues to as low as 13 percent on issues in late January and early February, reflecting the popularity of the issues, which were clearly supported by the continued stability of the rand. There has been only one issue since the pressures on the rand emerged in mid-February.

Table 5.Rand Bonds Issued in the Euromarket(In millions of U.S. dollars)
article image
Source: Euromoney Bondware (c) Eurononey Publications plc (Data) - Computasoft Ltd. (Software) 1983–94.

The market developed in response to demand by European retail investors for rand-denominated paper, who were either unwilling or unable to invest in the South African market. Some of the reasons for this retail demand for rand-denominated Eurobonds are the long settlement periods in South Africa; South African bonds are registered and not bearer bonds; and that bonds are not traded on the international exchanges such as Euroclear. These factors established arbitrage opportunities between the yield differentials on the Eurorand bonds and returns on rand paper in South Africa, and between yields on paper in major currencies, the Euro-rand yields and the forward market prices. And the rand-denominated Eurobond issues reflected both sets of arbitrage opportunities.

According to various sources, some of the issuers covered their rand exposure in private forward rand markets. The primary counterparts to these transactions were major South African entities which were covering new dollar-denominated liabilities. Other issuers used the proceeds to invest directly in South Africa’s securities market, often simply using the proceeds of the issuer to purchase gifts, contributing directly to capital inflows.

Long-term flows to the private sector shifted sharply from a net outflow of R 600 million in 1994 to a net inflow of R 8.5 billion in 1995, reflecting a substantial increase in the participation of nonresidents in the securities market. Nonresident activity began to grow following the abolition of the financial rand in response to the relatively high yields. In the nine months to December 1995, net purchases of bonds on the JSE by nonresidents amounted to R 2.2 billion, compared with a net outflow in the first three months of the year of R 0.3 billion and net inflows of R 1.1 billion in 1994 (Table 5). Nonresident participation also grew rapidly in the secondary equity market, as total purchases rose from R 185 million in 1994 to R 4.8 billion in 1995.

Another important channel of long-term flows to the private sector was a significant expansion in syndicated bank borrowing by the South African banks and corporations. Whereas in 1994, only one commercial bank and one private corporation accessed this source (for short-term financing), in 1995 15 such loans were made to commercial banks—with more than half of the loans at a 3-year maturity—and another 7 loans were made to corporations, the majority with medium-term maturities. At the same time, domestic banks started to issue Euro commercial paper, while South African corporations increased their placement of international equity and bond issues.

Commercial banks substantially raised their access to short-term capital, in large part to finance domestic credit expansion. Total short-term flows to the banking sector increased from R 3.4 billion in 1994 to R 9 billion in 1995. In contrast, the public sector and the private nonbank sector made net repayments equivalent to R 2.6 billion, largely reflecting a decline in trade-related liabilities, as importers made significantly less use of forward cover from abroad owing in part to the continued stability and growing confidence in the rand.

In its continuing effort to reestablish South Africa in the international capital markets the Government, in late January 1996, placed bonds amounting to £ 100 million (R 560 million), with a 10-year maturity. At the time of issue, the spread above the equivalent British government bond was 190 basis points, providing further evidence of improved investor rating of South Africa. This increase in confidence was also reflected by the fall in the spread between the yields on the South African global bond and the U.S. Treasury paper from 200-220 basis points to 160 basis points by end-January 1996 and to 140 basis points by mid-February; the spreads on long bond yields also fell during this period by approximately the same amount (Charts 9 and 10).

3. The currency attack

After nearly ten months of stability in its external value, fluctuating narrowly around R 3.64 per U.S. dollar, the rand came under substantial pressure in mid-February 1996, and fell during the subsequent weeks to R 4.32 per U.S. dollar by end-April (Charts 11 and 12), depreciating by 16 percent. The SARB intervened to “lean against the wind” and ensure orderly market conditions, selling the equivalent of R 5 billion in foreign exchange during that period.

Chart 11
Chart 11

South Africa: EXCHANGE RATES, JANUARY 1990-JANUARY 1996

Citation: IMF Staff Country Reports 1996, 064; 10.5089/9781451840926.002.A002

Sources: IMF’s Informations Notices System and International Financial Statistics.
Chart 12
Chart 12

South Africa: SOUTH AFRICAN RAND PER U.S. DOLLAR RATE

January 1-April 30, 1996

Citation: IMF Staff Country Reports 1996, 064; 10.5089/9781451840926.002.A002

Source: Financial Times and Routers.

Market reports indicated that the initial attack on the rand was sparked by rumors about the health of the President and about the imminent removal of all capital controls on residents. The rand fell further as a result of uncertainty prior to the announcement of the budget on March 13, 1996, and subsequently the resignation of the Minister of Finance at end-March. The rand declined even further through April due to uncertainty about the course of future policy, particularly regarding the removal of capital controls.

Another factor that may have contributed to the initial attack was the real appreciation of the rand in the preceding months (of about 7.7 percent in real effective terms between June 1995 and January 1996). Other explanations include the view that some investors interpreted the stability in the nominal exchange rate to mean that the SARB was targeting the rate which made it vulnerable to an attack. For instance, the Union Bank of Switzerland (UBS) published a report several days before the attack indicating that “the rand’s trading range bears a strong resemblance to a currency moving within an exchange rate band,” given that the fluctuation margin was about 1 percent in the past year. Others suggest the sharp increase in international long-term rates in February and the strengthening of the U.S. dollar vis-a-vis other currencies played a greater role; however, the impact of these elements on other emerging markets was only marginal and therefore cannot explain the extent of the attack on the rand (Chart 13).

Chart 13
Chart 13

South Africa: DEVELOPMENT IN SOUTH AFRICA AND WORLD EMERGING MARKETS

(Jon 5=100)

Citation: IMF Staff Country Reports 1996, 064; 10.5089/9781451840926.002.A002

Source: IFC’s emerging markets database.

While all the factors that triggered the attacks cannot be identified with certainty, few observers have argued that the attacks were based on fundamental macro-imbalances such as those that triggered the Mexican crisis. 1/ South Africa’s external current account position had deteriorated to 2.6 percent of GDP in 1995, but a continuation of deficits of this order of magnitude was regarded by most observers to be readily financeable; in contrast, Mexico’s external current account deficit was nearly 8 percent of GDP in 1994. Moreover, South Africa’s external debt is rather modest by world standards, and even though a significant proportion of the flows were short-term, there was a determined effort (particularly if the SARB is included) to reduce short-term exposure. But perhaps more importantly, all other macro variables in South Africa were moving in the right direction.

Other recent episodes of sudden reversals in capital flows were the result of the contagion effect of the Mexico crisis. They mainly reflected investors’ uncertainty regarding the authorities’ commitment to an officially fixed exchange rate. However, it is worth noting that the degree to which countries were affected and how fast the flows resumed, reflected how far advanced they were in their adjustment efforts to begin with, notwithstanding the substantial tightening of financial policies by the majority of the affected countries. 1/

4. Policy Challenges

The renewed access for South African borrowers to international capital markets and the abolition of exchange controls on nonresidents in March 1995 described above are substantive structural changes to the economy.

The volume of foreign saving available has increased—accommodating larger current account deficits—and the way in which external developments impinge on the domestic macroeconomic balance has changed. In particular, these flows now affect the domestic balance directly through the capital account of the balance of payments, and are no longer merely reflected in movements in the financial rand discount. Moreover, with the removal of exchange controls on nonresidents, domestic real interest rates are no longer independent of foreign rates.

The additional access to foreign saving has relieved one of the constraints on higher growth, namely the difficulties of financing the associated current account deficits. This may alter the “business cycle” behavior of the economy, as the downturns in the past decade have often been directly related to the early emergence of a binding external financing constraint. However, the external financing constraint has not vanished; rather the challenge is to ensure that the greater inflows are absorbed efficiently, with a key test being the extent to which they finance additional private fixed investment. On that measure, South Africa performed well in 1995, with a large portion of the current account deficit induced by rapid growth in private investment.

At the same time, the recent experience In South Africa has underscored the importance of appropriate policies to deal with capital inflows as well as with the volatility of these flows. In response to the capital inflows in the second half of 1994 and in 1995, the authorities remained broadly on track in their efforts to reduce the fiscal deficit, allowed a limited appreciation of the rand, implemented a sterilized intervention in the spot market, and reduced the net oversold forward position of the SARB. As a result the rand appreciated 7.7 percent in real effective terms from May 1995 through January 1996, net reserves increased from about zero in mid-1994 to US$3.5 billion by end-1995, and the net open forward position declined rapidly, as noted above. At the same time, the authorities removed capital controls on nonresidents and relaxed controls on residents. The sterilization of the intervention on the spot market appears to have been effective as the rate of inflation decelerated to below 7 percent by end-1995. In sum, the direction of policies was appropriate and their achievement significant.

However, these policies were not sufficient to avert the deterioration in the sentiment toward the rand—and the associated capital outflows in recent months. This rapid swing underscores that one of the primary features of recent developments is the volatility of such sentiment. Thus, an identification of the source of the volatility in sentiment is necessary. In part, it derives from the fact that the foreign exchange market in South Africa is—by international standards—relatively small. International developments also have played some part in it—notably the marked recovery in market sentiment toward emerging markets in mid-1995, and the impact of the swings in U.S. dollar interest rates in early 1996—but domestic developments have been the prime cause of the volatility.

Even though the swings in market perceptions concerning South Africa appear overdone in hindsight, they would seem to be directly traced to changing perceptions of the prospective stance of policies:

  • pessimism ahead of the 1994 election primarily reflected concerns that the transition might fail and concerns with the stance of policies likely to be adopted by the Government of National Unity following the election;

  • the turn in sentiment after the election and the associated capital inflows primarily reflected the political consensus that emerged, and the elaboration and pursuit by the Government of cautious financial policies, backed by structural reform, including the announcement of medium-term fiscal targets and the commitment to phased trade and exchange control liberalization;

  • but sentiment turned again after mid-February 1996 as uncertainty over the policy stance reemerged—notably concerning the authorities’ stance on the abolition of exchange controls on residents, and toward the underlying structural problems of slow growth and unemployment. These uncertainties were compounded by the resignation of the Minister of Finance.

The recent volatility in market perceptions of South Africa suggests that in a context where the domestic macroeconomic balance is directly linked to external developments—particularly on the capital account of the balance of payments—the authorities should by guided by three concerns; the need to implement financial and structural policies that strengthen market confidence; the need to ensure that their commitment to persist with such policies does not come to be called into question; and the need to ensure that markets are confident that the policies will be adjusted appropriately in the face of unexpected developments.

If uncertainties in these areas are not addressed, the benefits for growth of an easing of the external financing constraint since the early 1990s would be lost as a result of the associated increase in the country risk premium, and thus higher domestic interest rates.

1/

In September 1985, South Africa imposed a moratorium (“standstill”) on repayments of private debt to foreign commercial banks and, subsequently, negotiated with creditor banks a series of interim agreements providing for repayment of principal.

1/

The global bond issue was intended to establish a benchmark for other South African borrowers.

2/

These were primarily mineral-processing projects, including Alusaf (aluminum) and Columbus (stainless-steel).

3/

The financial rand was a separate currency maintained for nonresident capital transactions which was freely traded in South Africa, generally at a substantial discount from the regular rand.

4/

Trading of the global bond during the first half of 1995 was sparse.

1/

A discussion of this episode can be found in the “Factors Behind the Financial Crisis in Mexico,” Annex I, May 1995, World Economic Outlook, IMF.

1/

See Chapter III, May 1995, World Economic Outlook; and Chapter I, October 1995, World Economic Outlook, IMF.

South Africa: Selected Economic Issues
Author: International Monetary Fund
  • View in gallery

    South Africa: CURRENT AND CAPITAL ACCOUNT BALANCES

    (In percent of GDP)

  • View in gallery

    South Africa: GLOBAL BOND AND U.S. TREASURY RATES

  • View in gallery

    South Africa: LONG BOND YIELD SPREAD OVER U.S. RATE

    (through April 4, 1996)

  • View in gallery

    South Africa: EXCHANGE RATES, JANUARY 1990-JANUARY 1996

  • View in gallery

    South Africa: SOUTH AFRICAN RAND PER U.S. DOLLAR RATE

    January 1-April 30, 1996

  • View in gallery

    South Africa: DEVELOPMENT IN SOUTH AFRICA AND WORLD EMERGING MARKETS

    (Jon 5=100)