Exchange controls and exchange rate policy in the South African economy¶
Brian Kahn (1987). Exchange controls and exchange rate policy in the South African economy. Social Dynamics 13(1). Routledge (Taylor & Francis).
| Type | journal article |
| Scope | Economics and economic history of apartheid |
| Method | policy analysis |
| Period studied | 1960–1986 |
| Regions | South Africa |
| Themes | macro fiscal finance debt, business and state, labour markets wages, sanctions divestment, manufacturing industrial policy |
| Citation key | kahn1987a |
Argument¶
Kahn argues that South Africa's balance of payments is exposed to gold-price shocks on the current account and to political shocks on the capital account. Exchange rate policy has had a downward (undervaluation) bias that protects mining and agriculture and shifts income from wages to profits. The liberalisation of exchange controls under the De Kock Commission was inappropriate for a country whose access to foreign capital closes in political crises. It encouraged short-term foreign borrowing and capital flight, and controls had to be reimposed in 1985. He builds the case from balance-of-payments data, debt tables, the Commission's own arguments, and comparison with Southern Cone liberalisation failures.
Evidence and method¶
South African Reserve Bank Quarterly Bulletins (balance of payments, reserves, debt, exchange rates), De Kock Commission interim (1978) and final (1984) reports, Kleu Committee report (1983), Euromoney interview material (Grant 1985), Nedbank (1983), and secondary literature on devaluation and capital flight (Diaz-Alejandro, Krugman and Taylor, Khan and Ul Haque).
Place in the debates¶
- Exchange control liberalisation vs retention (De Kock Commission vs critics) (see Finance, capital flight, the 1985 debt crisis and banking). Kahn opposes liberalisation. He holds that controls give needed protection to reserves in crises, that market-based depreciation cannot be relied on to deter capital flight, and that reimposition in 1985 discredited the policy.
- Distributional effects of exchange rate policy (wages vs profits) (see Finance, capital flight, the 1985 debt crisis and banking). Kahn argues that a persistently depreciated rand redistributes income from wages to profits and favours mining and agriculture over import-dependent manufacturing.
- Causes of the 1985 debt crisis (see Finance, capital flight, the 1985 debt crisis and banking). He attributes it to the lax 1981-82 attitude to private foreign borrowing, the shift to short-term debt and the political loss of confidence, not to gold prices alone.
Findings¶
-
Gold dominates exports and its unpredictable price shocks the current account. (p. 3)
During the 1980's, approximately 70% of exports (excluding gold) have been agricultural and mineral, with gold alone averaging 47% of total exports.
-
State-led infrastructure borrowing shifted foreign investment toward loans and increased vulnerability to capital-account shocks. (p. 4)
increase in the proportion of indirect investment to total foreign investment from 32 percent in 1970 to 61 percent by 1984.
-
Capital account shocks followed major political crises (Sharpeville, Soweto, 1985) and shut off new loans. (p. 4)
Such shocks occurred in the aftermath of Sharpeville (1960), Soweto (1976) and again after the recent debacle in 1985.
-
The exchange rate has a downward bias that redistributes from wages to profits. (p. 6)
Given the bias towards depreciation rather than appreciation in the conduct of South African exchange rate policy, such a stance tends to discriminate against wages in favour of profits.
-
Depreciation sharply inflated the rand value of foreign debt in 1983-84. (p. 6)
Between 1983 and 1984 debt rose by 6,6 percent in dollars, but because of the currency depreciation the rand value of this debt increased by 65,6 percent
-
Mining and agriculture are the main beneficiaries of depreciation, and policy implicitly protects them. (p. 7)
Clearly mining and agriculture benefit most from a depreciation.
-
Lax oversight of private foreign borrowing and bunching of short-term maturities worsened the 1985 crisis when US banks recalled loans. (p. 8)
between 1980 and 1985 total foreign debt increased from $ 16,7b to $27b and the proportion of short term debt to total debt increased from 49,1 percent to 72 percent.
-
Evidence of exchange control evasion after 1976 appears in the 'errors and unrecorded transactions' item. (p. 9)
the rise in the 'error and unrecorded transactions' item of the balance of payments from R77m in 1975 to R585m in 1977.
-
Liberalisation relied on market access that disappears in crises, and reimposition destroyed credibility. (p. 9)
It is precisely at times of crisis that these markets are closed to South Africa, as occurred in 1976 and 1985.
-
Liberalisation mainly serves asset-holders wishing to move funds abroad and has failed elsewhere. (p. 10)
It is clear that full liberalisation primarily serves the interests of those who wish to hold their assets outside the country.
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| South African Reserve Bank Quarterly Bulletins (current and capital accounts of the balance of payments, 1967-1986) | official statistical series | 1967-1986 | Supplies the series plotted in Figure 1 to show the current and capital accounts moving in opposite directions. |
| South African Reserve Bank Quarterly Bulletins (net and gross gold and foreign exchange reserves) | official statistical series | c.1967-1986 | Supplies Figure 2, showing the substantial falls in net reserves in 1976, 1981 and 1985. |
| South African Reserve Bank Quarterly Bulletins (South Africa's foreign debt, Table 1) | official statistical series | 1980-1985 | Documents the rise in dollar and rand foreign debt, debt/GDP and the short-term debt share, used to show the wealth effect of depreciation and the bunching of maturities. |
| Real and nominal trade-weighted rand effective exchange rate (Figure 3) | official statistical series | 1975-1986 | Shows that the real and nominal rates appreciated with the gold price in 1980 and 1982 and depreciated after the 1984 political events. |
| Balance of payments item 'error and unrecorded transactions', 1975-1977 | official statistical series | 1975-1977 | Its jump from R77m to R585m, compared with gross reserves, is used as an indicator of exchange control evasion after the 1976 unrest. |
| De Kock Commission Interim Report, 'Exchange Rates in South Africa' (RP 112/1978) | commission or official report | 1978 | Marks the shift to variable dollar pegging and the start of the case for relaxing exchange controls. |
| De Kock Commission Final Report, 'The Monetary System and Monetary Policy in South Africa' (RP 70/1984) | commission or official report | 1984 | Supplies the Commission's arguments and quotations on exchange control, capital flight and depreciation, which the paper critiques. |
| Franzsen Commission Report (1970) | commission or official report | 1970 | Cited as an earlier official position strongly favouring the maintenance of exchange controls because of capital flight risk. |
| Kleu Study Group (Committee) Report on Industrial Development Strategy, 1983 | commission or official report | 1983 | Used for the finding that manufacturing is the highest net user of foreign exchange and has the highest import content in its exports. |
| Grant, C. (1985), 'The Banks Abandon South Africa', Euromoney, December | newspapers periodicals | 1985 | Supplies quotes from a bank director and Gerhard de Kock on the Reserve Bank's lack of information on corporate foreign debt and the failure of direct controls. |
| Nedbank Group Economic Unit (1983), South Africa: an Appraisal | company records or reports | 1983 | Source of Chris Stals' statement that abolition of the Financial Rand was permanent, later contradicted by the 1985 reimposition. |
| Kahn (1987), 'Import Penetration and Import Demands in the South African Economy' (unpublished mimeograph, UCT) | secondary literature compilation | c.1970s-1980s | Source for import penetration ratios rising in machinery and motor vehicles/transport equipment. |
| Kaplan (1983), 'South Africa's Changing Place in the World Economy', South African Review One | secondary literature compilation | to 1980 | Source for the statistic that by 1980 South African investment abroad exceeded one third of foreign investment into South Africa. |
| Black and Stanwix (1986) and Kleu (1983) on capital intensity of manufacturing | secondary literature compilation | c.1960s-1980s | Support the claim that capital intensity rose, especially in manufacturing, driving the import dependence of output expansion. |
Inference¶
Quantification: Descriptive tables.
Analytical, policy-focused argument that combines descriptive time-series charts and one table of balance of payments, reserve and debt data with macroeconomic theory (real vs nominal exchange rates, distributional effects of devaluation) and a critical reading of the De Kock Commission reports. Episodes (Sharpeville, Soweto, 1985) and comparison with Southern Cone liberalisations carry the inference. There is no econometric testing.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Exchange rate policy has a downward (undervaluation) bias: the rand is allowed to fall sharply after gold price falls but is held back from appreciating when the gold price rises, protecting mining and agriculture. (p. 5) | Inference from the narrative of policy changes since 1971 and Figure 3 on real and nominal effective rates, with no formal test of the bias. | Illustrative or anecdotal |
| Currency depreciation sharply raised the rand burden of South Africa's foreign debt between 1983 and 1984, and the debt became increasingly short-term. (p. 6) | Table 1, built from SARB Quarterly Bulletins, giving dollar and rand debt, debt/GDP and the short-term share for 1980-85. | Systematic data |
| Political shocks (Sharpeville, Soweto, 1985) cut off access to international capital markets, forcing deflation or exchange rate adjustment. (p. 4) | SARB current/capital account and reserve series (Figs 1-2) plus narrative of the 1976-77 and 1985 episodes. | Partial data |
| Exchange controls leaked, especially in crises; the rise in 'errors and unrecorded transactions' after 1976 suggests substantial evasion. (p. 9) | SARB balance of payments item compared with gross reserves of R837m in September 1977, treated as an indicator rather than a measure. | Partial data |
| Liberalisation of exchange controls in the early 1980s was inappropriate for a country subject to political crises: it facilitated capital outflow and short-term foreign borrowing, and the controls had to be reimposed in 1985. (p. 10) | Critical reading of the De Kock reports, the Grant (Euromoney) quotes, Table 1 debt data and comparison with Latin American and Israeli experience. | Partial data |
| Manufacturing is the highest net user of foreign exchange and has the highest import content in its exports, so mining and agriculture benefit most from depreciation. (p. 7) | Kleu Committee Report (1983: 74) for manufacturing; mining and agriculture benefits are inferred. | Partial data |
| South Africa's growing reliance on foreign loans rather than direct investment raised its vulnerability: indirect investment rose from 32 percent of total foreign investment in 1970 to 61 percent by 1984. (p. 4) | Statistics given without a stated source, presumably from SARB data. | Partial data |
The author's caveats¶
Notes that the Reserve Bank stopped reporting 'errors and unrecorded transactions' separately in 1978, so the item cannot be tracked through the recent crisis (fn 7). Concedes that not all unrecorded transactions were illegal outflows, and that it is impossible to say what outflows would have been without controls.
Gaps in the evidence¶
The central distributional claim (undervaluation shifts income from wages to profits) is not tested with South African wage share, profit share or sectoral profitability data, which were available from national accounts and census sources. The downward-bias claim needs a counterfactual or an estimated equilibrium exchange rate model against gold prices and inflation differentials. Capital flight is evidenced by a single, ambiguous balance of payments residual; fuller Reserve Bank data on corporate foreign debt and outward investment, partly unavailable at the time, would test the argument better.
Replicability: Data described but not public.