ECONOMIC GROWTH CENTER
YALE UNIVERSITY
P.O. Box 208269
27 Hillhouse Avenue
New Haven, Connecticut 06520-8269
CENTER DISCUSSION PAPER NO. 796
Philip I. Levy
Yale University
February 1999
Note: Center Discussion Papers are preliminary materials circulated to stimulate discussions and critical
comments.
SANCTIONS ON SOUTH AFRICA: WHAT DID THEY DO?
Philip I. Levy
Yale University
February 1999
Abstract
This paper considers the economic sanctions that were applied in the mid-1980s to
pressure the South African government to end apartheid. It asks what role those sanctions
played in the eventual demise of the apartheid regime and concludes that the role was
probably very small. An alternative explanation for the regime change is offered: the
communist bloc combined to bring about the change. If one is to argue for the efficacy of
sanctions, two key obstacles are their limited economic impact and the substantial lag
between the imposition of sanctions and the political change. Since sanctions preceded
the change of government, it is impossible to rule them out as a determinant. However,
their principal effect was probably psychological. The implication is that the South
African case should not serve as the lone major instance of effective sanctions.
Keywords: Sanctions, South Africa, Political Economy, Trade
J.E.L. classification: F14
2
Sanctions on South Africa: What did they do?
by Philip I. Levy*
Although careful studies of economic sanctions have cast doubt on their
effectiveness,1 anecdotes can be powerful rhetorical tools. A single important case that
demonstrates sanctions’ potential allows advocates to argue that their cause is more akin
to the success than to the failures. Frequently, advocates point to the case of sanctions
applied in the mid-1980s against the apartheid regime in South Africa as just such a case.
On the face of it, South African sanctions appear to have been successful. In
response to the outrages of apartheid, many countries adopted trade and financial
sanctions and a significant amount of foreign investment was withdrawn from South
Africa. After the adoption of sanctions, South Africa experienced economic difficulty and
numerous domestic actors commented on how the economic situation was untenable and
required political change. By 1994, Nelson Mandela had been elected President of South
Africa. He and other black leaders attributed to economic sanctions a significant role in
bringing about the democratic transition.
This paper will also present an alternative view. One can tell a story of the change
in South Africa in which any positive contribution of trade sanctions was trivial and they
might even have deferred the achievement of the campaign’s objectives. The international
economic actions against South Africa that were most damaging were taken by private
actors, not governments. The actions taken by governments were not especially
economically damaging. To some extent, they caused the Nationalist Party government to
stiffen its repression. In this alternative story, the demise of apartheid, which followed
* Department of Economics, Yale University, P.O. Box 208269, New Haven, CT, 06520-
8269. E-mail: philip.levy@yale.edu. I am grateful to Stefan Krieger, T.N. Srinivasan and
William Foltz for helpful conversations.
1 Hufbauer, Schott and Elliott (1990) is the most thorough study of economic sanctions
and they find them to be effective roughly one third of the time. Morgan and Schwebach
(1997) use an expanded data set to argue that this rate significantly overstates the
effectiveness of sanctions.
3
sanctions with a substantial lag, can be attributed to three different factors: the
effectiveness of the political opposition of the black majority; the inefficiency and growing
economic cost of the apartheid system; and the fall of the Soviet Union.2
Alongside this alternative case the paper will try to illuminate the kind of
arguments that one must make if one is to argue that sanctions were effective. The
fundamental problem in assessing the role of sanctions is that the end of apartheid was
overdetermined. Given the sequence of events, it is impossible to prove that sanctions
were ineffective; they were among the many potential “causes” linked to the single
“effect.”3 In lieu of such a proof, the rest of the paper will describe the sequence of events
in South Africa, consider what it would mean for sanctions to be effective, and argue that
at best sanctions failed to interfere with the other forces that were bringing down the
apartheid regime.
I. The Sanctions Timeline
The apartheid regime against which sanctions were targeted had been in place
since the National Party took power in 1948. Prior to the episodes of the mid-1980s, the
United Nations adopted an arms embargo in the early 1960s and OPEC nations applied an
oil embargo starting in 1973.4 The focus of this paper will be the multilateral sanctions of
1985-87, particularly those dealing with trade.5
If those sanctions were to have played an important role in undermining the
apartheid economy, one would expect the imposition of sanctions to mark a break in
economic performance. In fact, the break for South Africa occurred much earlier. In the
2 Lowenberg (1997) offers a useful analysis of the reasons for the failure of the apartheid
economy.
3 Of course, with a larger set of countries, one could use econometrics to sort out the
determinants, but that would lead us back to the careful studies showing the
ineffectiveness of sanctions.
4 Up until the fall of the Shah in 1979, though, Iran continued to supply South Africa with oil. Lewis
(1990, p. 103).
5 There are a number of thorough tellings of the story of recent South African history. For
a general account, see Waldmeier (1997). For ones more focused on the effect of
sanctions, see Manby (1992) and Hufbauer, Schott and Elliott (1990).
4
decades leading up to 1974, real GDP in South Africa grew an average of 4.9 percent per
year. From 1974 to 1987 it averaged 1.8 percent per year.6 In the immediate aftermath of
the mid-1980s sanctions, GDP growth accelerated: it was 0.5 percent in 1986, 2.6 percent
in 1987, and 3.2 percent in 1988.7
The early-1970s break was contemporaneous with the oil embargo and the
worldwide oil crisis, which can certainly serve as one explanation for slower growth. In
addition, a number of factors beyond the oil shock combined to slow the expansion; two
are worth noting here.
First, labor market distortions were inherent to the apartheid system. Blacks
constituted the majority of the population but were restricted in their travel and in the jobs
that they could hold. One goal of the apartheid system was to keep blacks living in
separate areas from whites. As the South African economy developed, however, the
employment restrictions bound tightly. An expanding manufacturing sector created a
demand for additional workers in the cities at the same time that a central goal of
apartheid was to keep blacks out of cities and in separate homelands. In 1973, in response
to the resulting labor shortages, blacks were allowed to work at skilled jobs in white areas.
This was a dramatic undermining of the rationale of apartheid and was only the first in a
series of reforms in which the employment restrictions were relaxed.8 Nonetheless, even
when blacks were allowed to take jobs they had previously been denied, restrictions
remained and along with them an extensive bureaucracy to maintain apartheid.
The second notable factor in South Africa’s economic slowdown was the role of
external borrowing. South Africa pursued a policy of import substitution industrialization.
6 Jones and Muller (1992, p. 296).
7 Lipton (1989, p. 348). She also notes that there was increased activity in the informal
sector at that time so the figures may be underestimates. Note, though, that GDP growth
rates were lagging behind population growth.
8 Waldmeier (1997, p. 26). Stressing the importance of this change, she writes, “as the far
right warned at the time, jobs could not ultimately be shared with blacks without sharing
power as well.” Also see Lowenberg (1997) on the importance of labor market distortions
as a barrier to growth.
5
From the 1940s to 1984, South Africa ran current account deficits on the order of 2 to 3
percent of GDP.9 These were offset by substantial capital inflows. This dependence on
foreign capital left South Africa vulnerable to shifts in lending. One such shift occurred in
the period 1976-1980, when there was a net foreign capital outflow averaging 2.3 percent
of GDP.10 By the mid-1980s, South Africa’s external debt was roughly $24 billion, of
which two-thirds was short-term.11 This structure left South Africa dependent on the
willingness of foreign lenders to refinance.
Throughout this time, there was an anti-apartheid political movement that was
pressuring the government. A major turning point was the 1976 Soweto uprising. It began
as a revolt against a government plan to teach Afrikaans in black schools, drew
international attention and prompted the United Nations arms embargo as well as private
codes of conduct for foreign firm operating in South Africa. There was persistent
international condemnation of the apartheid regime, but no major economic sanctions
were adopted until later. Responding to reform pressures, in 1984 the South African
government adopted a new constitution that gave Indians and “coloreds” some right to
participate but continued to exclude blacks. Unrest in South Africa persisted and
intensified and the government responded with repression. In July of 1985, President P.W.
Botha declared a state of emergency. Shortly thereafter, Chase Manhattan Bank declared
it would not renew its short-term loans, touching off a liquidity crisis as other lenders
followed suit.
All of this preceded the multilateral economic sanctions. While foreign companies
doing business in South Africa experienced pressure in their home countries to disinvest, it
is difficult to distinguish the effects of this pressure from South Africa’s diminishing appeal
as a borrower. One Chase executive explained his company’s withdrawal by saying, “We
9 Jones and Müller (1990, p. 299).
10 Lewis (1990, p. 178).
11 Waldmeier (1997, p. 56). To put the number in perspective, South African GNP in 1984
was $70 billion (Hufbauer, Schott and Elliott, p. 246), so external debt was almost 35
percent of GNP.
6
felt that the risk attached to political unrest and economic instability became too high for
our investors. We decided to withdraw. It was never the intention to facilitate change in
South Africa, the decision was taken purely on account of what was in the interest of
Chase and its assets.”12
It is also important to distinguish between private and public acts. The financial
crisis was brought on by the decisions of private lenders who saw a deteriorating political
and economic situation and doubted the country’s creditworthiness. For the purpose of
public policy discussions, this privately induced financial crisis — the repercussions of
which were substantially greater than any of the public sanctions that ensued — cannot be
used as evidence of sanctions’ effectiveness.
In August of 1985, President Botha gave a speech intended to introduce further
reforms to apartheid. The reforms, however, were substantially less than what the
international community had been expecting. The crisis intensified, the South African
Rand fell further and in late August the government temporarily closed the stock exchange
and foreign-exchange markets and suspended interest payments on its debt.
II. Sanctions deployed
The mounting frustration with the persistence of apartheid, coupled with South
Africa’s apparent economic vulnerability, prompted a round of multilateral economic
sanctions. In September 1985 the European Community imposed a set of very limited
trade and financial sanctions on South Africa, and the Commonwealth countries adopted
similar measures in October. In the United States, the Reagan administration was opposed
to South African sanctions but imposed a limited export ban to head off stronger action in
Congress.
The political and financial crises in South Africa continued as the government
failed to reach agreement with its creditors. In December, it extended its freeze on debt
12 Quoted in Waldmeier (1997, p. 56). I draw on Hufbauer, Schott and Elliott (1990, pp.
221-235) for their detailed timeline of events surrounding the imposition of sanctions.
7
repayments until March 1986. In February 1986, an interim agreement with creditors was
reached and in March Botha lifted the state of emergency. The state of emergency was
reimposed in June, however, as the tenth anniversary of the Soweto uprising approached.
In the fall of 1986, the second and more significant round of sanctions ensued. In
September, the EC banned imports of iron, steel, gold coins from and new investments in
South Africa. They did not, however, extend this ban to cover the most important South
African exports, such as coal, diamonds or other forms of gold. Japan passed similar
sanctions shortly thereafter, although omitting iron ore. In the United States, Congress
passed the Comprehensive Anti-Apartheid Act (CAAA).13 President Reagan vetoed the
legislation, but his veto was overridden in October. The CAAA severely restricted lending
to South Africa (although not loan rescheduling) and imposed import bans on iron, steel,
coal, uranium, textiles, and agricultural goods. Strategic materials, diamonds, and most
forms of gold were omitted.14
The direct impact of the trade sanctions was limited. South Africa developed
extensive measures to circumvent the sanctions, although these sometimes involved costly
import-substitution. South Africans also were able to transship through countries that
were not participating in the embargoes. In sum, from 1985 to 1989, export volumes rose
by 26 percent, although terms of trade suffered.15 One estimate of the marginal cost to
South Africa of the mid-1980s trade sanctions was $354 million annually, or 0.5 percent
of GNP.16
13 Other countries imposing financial or trade sanctions included the Nordic countries,
Australia and Hong Kong (Hufbauer, Schott and Elliott, 1990, pp. 238-239).
14 Jones and Müller (1992, p. 344). To put this in perspective, gold was easily South
Africa’s largest export. In 1985 it accounted for 42.6 percent of the value of merchandise
exports. Manufactured exports accounted for 33.1 percent, mineral exports for 18.2
percent, and agricultural exports for 2.9 percent.
15 Waldmeier (1997, p. 133).
16 Hufbauer, Schott and Elliott (1990, p. 246). They calculate the losses based on
reduction in trade in the targeted goods relative to the early 1980s and estimating what
fraction of the lost exports constitute a welfare loss (3 percent for gold coins, 30 to 40
percent for other goods). Higher estimates of sanctions’ cost sometimes include the oil
embargo of the 1970s and financial sanctions.
8
There is little doubt that the reversal of the flow of capital into South Africa was
far more painful economically than trade sanctions. The doubt concerns how much these
effects can be attributed to public rather than private measures. The effort by the banks to
withdraw preceded the imposition of sanctions and seemed to show that South Africa’s
faltering economy, market uncertainties and political turmoil had made it an unattractive
site for investment. In fact, the outflow of dividends and profits from existing investments
in South Africa exceeded new FDI,17 thus presenting default as an attractive option for the
country and an important concern for sanctioning governments. Thus, it is not clear how
binding restrictions on further lending were.
The disinvestment campaign, which largely consisted of private pressure but also
included some government action, may have been most costly to the foreign firms that
withdrew. In the words of one observer, “Many disinvesting companies simply sold their
assets cheaply to local white businessmen, but maintained non-equity links such as
franchise, licensing, and technology agreements that permitted them to keep operating.”18
Further, in September 1985 South Africa introduced a dual exchange rate regime to
discourage disinvestment, whereby those firms that wished to repatriate their holdings did
so at the “financial rand” rate, which was frequently at a 40 percent discount to the
commercial rate.
Finally, there were non-economic sanctions, such as the prohibition against South
African participation in international rugby and cricket competitions.
Although these sanctions were in place, the repression of the black majority
continued and at times intensified. In early 1988, for example, the government banned all
major nonwhite opposition groups and prohibited political activity by trade unions.19
During the late 1980s, however, secret negotiations were underway between
representatives of the government and leaders of the African National Congress (ANC).
17 Lewis (1990, p. 110).
18 Waldmeier (1997, p. 133).
19 Hufbauer, Schott and Elliott (1990, p. 229).
9
Eventually, in February 1990 Nelson Mandela was released from prison. In April 1994,
with a new South African constitution negotiated, Nelson Mandela won the first all-race
elections to become President.
III. Judging the Effectiveness of Sanctions
One can ask what criteria one should apply to determine effectiveness. Hufbauer,
Schott and Elliott (1990) rank on a scale of 1 to 4 the extent to which the goals of
sanctions were achieved and the extent to which sanctions contributed to the achievement.
They multiply the scores and if the result is 9 or greater, they deem sanctions successful.
At the time of publication, they assigned the mid-1980s sanctions against South Africa a 2
for effectiveness and a 3 for sanctions’ contribution — unsuccessful. Presumably now,
since all the stated goals of the CAAA have been met, they would deem the sanctions
successful. Their score for sanctions’ contribution, however, is open to debate.
One measure of sanctions’ effectiveness could be the extent of economic damage
they inflict on the target country. Although the trade sanction cost of 0.5 percent of GNP
is not trivial, neither is it very large. Kaempfer and Lowenberg (1988) argue that targeted
sanctions can impose minimal economic hardship and still effect policy change. However,
the South African sanctions were not tightly targeted; whites often benefited from the firesale
disinvestment and blacks were often hurt by the loss of jobs.
Perhaps a clearer standard for the effectiveness of sanctions, particularly in a case
in which the goals were so well defined, is whether the goals would have been achieved in
the absence of the sanctions. In the South African case, the goals were clearly achieved,
albeit after a lengthy interval. The question is whether sanctions played an integral role.
Below the case is made for and against.
IV. An “effective sanctions” story
Although the direct economic impact of governmentally imposed sanctions was
quite small, one could argue that they were the “final straw” that made economic
10
conditions intolerable and forced political change. There were numerous contemporaneous
statements from South Africans that drew such a link between the economic climate and
political change, although they were usually accompanied by denials of sanctions’ role.
However, one might discount such denials as an effort to dissuade foreign governments
from further sanctions.
Certainly an important set of evidence in favor of sanctions’ effectiveness is the
statements by black leaders expounding upon their importance. Bishop Desmond Tutu,
Rev. Allan Boesak, and Nelson Mandela all made such statements. Even upon Mandela’s
release from prison in 1990, he said, “To lift sanctions now would be to run the risk of
aborting the process toward ending apartheid.”20
Their arguments may have less to do with tipping an economic balance than with
sanctions’ psychological impact. The sanctions signaled the extent to which South Africa
was isolated in the international community. While cultural sanctions played the same role,
perhaps the greater cost of imposing trade sanctions sent a stronger signal. Although
sanctions of one form or another had been in place for a long time, the ones of the mid-
1980s included key South African allies such as the United States in a non-trivial way.
While there was a substantial lag between the imposition of sanctions and the change of
regime, it may have been unrealistic to expect instantaneous change.21
V. An “ineffective sanctions” story
An alternative interpretation of events can be offered in which sanctions played no
important role. Instead, one can identify three factors that led to the downfall of apartheid.
First, the extensive labor market distortions of apartheid proved increasingly costly and
unacceptable. The South Africans linking economic conditions with the need for political
change could well have been sincere in dismissing sanctions; the economic forces
20 Quoted in Hufbauer, Schott and Elliott (1990, p. 233).
21 However, since asymmetric information seemed to play no role here, instant capitulation
would seem to have been more rational than prolonged endurance of sanctions, followed
by capitulation.
11
mandating change were internal. This is supported by the steady relaxation of apartheid
labor laws, beginning in the 1970s, well before serious sanctions.
Second, there was an effective political movement against the apartheid
government. While this movement did not have sufficient power to overthrow the
government by force, it was able to create substantial internal disruption. This political
instability combined with the economic flaws of apartheid (and of the accompanying
program of import-substitution industrialization) to make South Africa unattractive to
foreign investors, both directly and indirectly. The direct effect was the increase in risk.
Indirectly, the measures taken by the government to combat the turmoil raised spending
and taxes, thus further diminishing South Africa’s attractiveness as a host for FDI.
Despite these two effects, which were in full play by the mid-1980s, the apartheid
regime remained in power. The final key ingredient in the change was the fall of
communism in Eastern Europe and Mikhail Gorbachev’s forswearing of regional proxy
wars. The Afrikaner government of South Africa saw the ANC as a party of godless
communists, ready to take their country into the communist bloc. Once the communist
bloc fell apart and a withdrawal of Cuban troops from Angola was negotiated, a political
deal with the ANC became conceivable.22
The timing of this story works better than that of the “effective sanctions” version.
Shortly after the imposition of the mid-1980s sanctions, repression intensified. Observers
have attributed this, at least in part, to President Botha’s strong reaction against foreign
interference. Other sanctions had been signaling the world’s disapproval since the early
1960s without prompting any change. Yet once the perceived communist threat
diminished, the release of Mandela and negotiations on a transition followed fairly
quickly.23
22 Waldmeier (1997) describes the anti-communism of the Afrikaners.
23 To be fair, this time period also coincided with F.W. de Klerk’s succession to the South
African Presidency. One could argue whether this change was essential.
12
While it is clear that sanctions had a psychological impact, this was nowhere near
enough to swing the balance against apartheid. As long as the staunchly nationalist
Afrikaners perceived themselves as facing complete disaster should they succumb to a
communist foe, a small decrease in the payoff to persisting in apartheid did nothing to alter
the government’s course of action. Once it became clear that a negotiated solution might
be palatable, the public sanctions were an unnecessary addition to the substantial
incentives to talk.
VI. Conclusion
It is impossible to argue conclusively that trade sanctions failed in the South
African case. Given the small economic effects of trade sanctions, an argument for their
effectiveness ends up hinging on their psychological impact on the governing party. Most
South African leaders claimed the impact was minimal, but one can choose not to believe
them. Such issues are very difficult to resolve; the best one can do is argue that it is
implausible that trade sanctions played a significant and positive role. This paper has
attempted to do so by offering an alternative interpretation that seems to fit the events
more closely and that offers no role for governmentally imposed sanctions. At the very
least, this should cast serious doubt on the applicability of the South African case as a
model for further trade sanctions.
13
References
Hufbauer, Gary Clyde, Jeffrey J. Schott, and Kimberly Ann Elliott. Economic
Sanctions Reconsidered: History and Current Policy, 2nd Edition, Washington
D.C.: Institute for International Economics, 1990.
Jones, Stuart and Andre Müller. The South African Economy, 1910-90, New York: St.
Martin’s Press, 1992.
Kaempfer, William H., and Anton D. Lowenberg. “The Theory of International
Economic Sanctions: A Public Choice Approach,” American Economic Review,
September 1988, 78(4), pp. 786-93.
Lewis, Stephen. The Economics of Apartheid, New York: Council on Foreign Relations,
1990.
Lipton, Merle. “The Challenge of Sanctions,” South African Journal of Economics,
December 1989, 57(4), pp. 336-61.
Lowenberg, Anton D. “Why South Africa’s Apartheid Economy Failed,” Contemporary
Economic Policy, July 1997, XV, pp. 62-72.
Manby, Bronwen,. “South Africa: The Impact of Sanctions,” Journal of International
Affairs, Summer 1992, 46(1), pp. 193-217.
Morgan, T. Clifton and Valerie L. Schwebach, “Fools Suffer Gladly: The Use of
Economic Sanctions in International Crises,” International Studies Quarterly,
March 1997, 41(4), pp. 27-50.
Waldmeir, Patti. Anatomy of a Miracle: the End of Apartheid, New York: W.W.
Norton, 1997.
http://www.econ.yale.edu/growth_pdf/cdp796.pdf