The Effect of Socially Activist Investment Policies on the Financial Markets: Evidence from the South African Boycott¶
Siew Hong Teoh; Ivo Welch; C. Paul Wazzan (1999). The Effect of Socially Activist Investment Policies on the Financial Markets: Evidence from the South African Boycott. The Journal of Business, vol. 72, no. 1. University of Chicago Press.
| Type | journal article |
| Scope | Wider apartheid literature |
| Method | econometric |
| Period studied | 1973–1994 |
| Regions | United States, South Africa |
| Themes | sanctions divestment, macro fiscal finance debt, business and state, methods data sources, resistance social movements |
| Citation key | teoh1999 |
Argument¶
The authors use event-study methods to test whether the US legislative and shareholder boycott of apartheid South Africa hurt financial markets. They look at US banks with South African loans, US firms with large South African operations, firms that divested voluntarily, and South African financial series (Anglo American, JSE gold and industrial indices, the rand). They find that US involvement was small and that announcement effects were mostly statistically insignificant, or had the 'wrong' sign. They find only weak evidence that institutional shareholders returned to divesting firms. They conclude that the boycott did not visibly damage financial markets, that stock demand curves are highly elastic, and that financial markets avoided the brunt of the sanctions. A background section reviews South Africa's 1980s politics and macroeconomy, and argues that the recessions tracked the falling gold price rather than sanctions.
Evidence and method¶
Investor Responsibility Research Center data (Bowers and Cooper) on US firm and bank exposure; Dow Jones News Retrieval and newspaper dates for 10 legislative events leading to the 1986 Comprehensive Anti-Apartheid Act, 16 pension-fund divestments and 46 voluntary corporate divestments; CRSP returns; Compustat; Moody's data; S&P stock guide institutional holdings; Financial Times daily South African data (Anglo American, JSE indices, rand/dollar); IMF and US Bureau of Mines macro and gold series.
Place in the debates¶
- Sanctions effectiveness: did the financial boycott and divestment impose measurable economic costs on South Africa or on targeted firms? (see Sanctions, disinvestment and the anti-apartheid movement). Sceptical. Finds no discernible negative announcement effects on US banks, on exposed US firms, on divesting firms or on South African financial markets. It concedes that sanctions may have raised moral awareness.
- Cause of South Africa's 1980s economic decline: gold price and business cycle vs sanctions and unrest (see Did the apartheid economy fail? Growth, productivity, investment and the post-1970s crisis). Emphasises the falling gold price. It notes that recessions coincided with gold declines, that the JSE rallied during peak pressure, and that capital flight did not come with worsening terms of trade.
- Do stock demand curves slope down (price pressure from socially motivated investors)? (see Sanctions, disinvestment and the anti-apartheid movement). Finds little price effect from institutional withdrawal or return, consistent with highly elastic demand.
Findings¶
-
Overall conclusion: no visible effect of the boycott on the valuation of targeted firms or on South African financial markets. (p. 50)
Despite the prominence and publicity of the boycott and the multitude of divesting companies, the financial markets' valuations of targeted companies or even the South African financial markets themselves were not easily visibly affected.
-
US bank exposure to South Africa was too small for sanctions to register in bank stock prices. (p. 25)
The lack of significance can be rationalized ex post by the small size of the loan portfolios held by these banks, with the most involved bank holding only about 1.6% of all its loans in South Africa.
-
US bank lending was a minor share of South Africa's external debt. (p. 25)
Loans to South Africa by the U.S. banks in our sample were approximately $1.3 billion in 1985. This represented about 5.7% of South Africa's $23 billion external debt.
-
Across the legislative event dates, Anglo American and the JSE Gold Index reacted positively, the opposite of the expected sign. (p. 26)
Again, taking returns from all 10 event dates, Anglo-American Corporation and the Gold Index show an inconsistent statistically significant positive reaction to increases in the likelihood of sanctions.
-
Voluntary divestment announcements had no significant abnormal return effect. (p. 44)
In sum, there is no evidence that firms were either negatively or positively affected by their divestment announcement
-
There is weak evidence that institutional shareholders returned to firms at divestment. (p. 40)
table 5 shows that the number of institutional shareholders increased by 1.2% (statistically significant at the 10% level) in the divestment announcement month, whereas control firms lost 0.4%.
-
Gold was central to South Africa's export earnings, so gold price movements dominated macroeconomic performance. (p. 10)
Together, gold accounted for 36% of South Africa's exports in 1979, rose to 46% in 1980, kept at about 40% until the mideighties
-
South African recessions tracked falling world gold prices. (p. 10)
The data indicate that world gold price decreases were associated with economic recessions in South Africa.
-
Capital flight in 1985 was large relative to gold export earnings. (p. 13)
Similarly, net identified private capital suddenly began to leave South Africa in 1985, hitting an all-time high in 1985 of 6.5 billion rands (compared, e.g., to 15.5 billion rands in gold exports).
-
Labour unrest surged in 1986-87, though it was not easily visible in employment and wage data. (p. 8)
which had averaged 200-400 per year from 1974 to 1985, tripled to 793 in 1986, and reached an all-time high of 1,148 in 1987.
-
The stock market rallied during the early period of most intense political pressure. (p. 16)
During most of the early period of intense pressure, both JSE indices actually rallied before dropping in the worldwide crash of 1987.
Evidence ledger¶
Sources used¶
| Source | Kind | Period | How used |
|---|---|---|---|
| Bowers and Cooper, U.S. and Canadian Investment in South Africa (1986, 1987, 1989 annual editions; IRRC) | secondary literature compilation | 1985-1989 | Supplies data on South African sales, assets, employees and bank loans, used to define the bank, high-exposure and divesting-firm portfolios and their weights. |
| Dow Jones News Retrieval service (broad tape) for event dates | newspapers periodicals | 1985-1986 legislative events; firm divestments through c.1990 | Source of voluntary divestment event dates and the 10 legislative event dates leading to the 1986 Act. |
| Wall Street Journal, New York Times, Los Angeles Times, Washington Post news reports | newspapers periodicals | 1985-1990 | Used to identify earliest disclosure of legislative events, event descriptions and pension fund divestment announcements. |
| CRSP daily and monthly return data | stock market financial data | c.1980s-1990 | Firm, market (equal-weighted) and 4-digit SIC industry returns for the event-study abnormal returns. |
| Compustat accounting data | company records or reports | 1980s | Worldwide sales, assets and employees, giving the denominators for the South African exposure weights. |
| Moody's Handbook of Common Stocks (1985-86) | company records or reports | 1985-86 | Hand-collected institutional ownership data. |
| Moody's Bank and Finance Manual (1984-87) and Moody's Bank Report 1985 | company records or reports | 1984-87 | Net loan totals used to compute each bank's South African loan share and portfolio weight. |
| Federal Reserve Historical Business Day Data (1-year Treasury bill yields) | official statistical series | 1980s | Daily interest rates for the fixed-income factor in the model-return regressions. |
| Financial Times daily South African financial data (Anglo-American, JSE Industrial and Gold Indices, rand/dollar commercial rate) | stock market financial data | 1980s-1990 | Event-study outcomes testing the reaction of South African financial markets to legislative and pension-fund events. |
| IMF International Financial Statistics 1979-92 | international trade or finance data | 1979-1992 | Background macro series (real GDP, openness, balance of payments, terms of trade, JSE indices, exchange rates) in figures 3-6. |
| U.S. Department of the Interior, Bureau of Mines, Minerals Yearbook 1988 | official statistical series | 1980-1988 | South African gold output (about 21.8m oz to 19.9m oz) in the macroeconomic background. |
| S&P monthly stock guide (institutional shareholder counts and shares held) | stock market financial data | c.1980s | Aggregate institutional shareholder numbers and shares held around divestment dates (tables 5-6, 8). |
| Moskowitz (1998) industry-momentum factor data | secondary literature compilation | 1980-1992 | Used to adjust returns for the real-time performance plot of the three portfolios (fig. 7). |
| Aggregate South African work stoppage series (fig. 1) | official statistical series | 1974-1987 | Documents unrest by number of work stoppages (200-400 per year, 793 in 1986, 1,148 in 1987). |
New data¶
Event-study dataset assembled by the authors: 10 legislative event dates, 9 banks, 17 high-exposure firms, 16 pension-fund divestments (25 events), 46 voluntary divesting firms with size/SIC-matched controls and hand-collected institutional ownership data, plus daily South African financial series.
Inference¶
Quantification: Econometric (causal design).
Standard event-study methodology (market/industry/risk-free model returns with equal and exposure-weighted portfolios) tests for abnormal stock returns around legislative, pension-fund and corporate divestment announcements. This is supplemented by matched-control comparisons of institutional ownership changes, cross-sectional regressions, and descriptive macro and real-time return plots.
Key claims¶
| Claim | Basis | Strength |
|---|---|---|
| Legislative events leading to the 1986 Comprehensive Anti-Apartheid Act had no significant adverse effect on U.S. banks with South African loans. (p. 58) | Event study on 10 dates for nine banks (equal and loan-weighted) using CRSP returns and IRRC loan data. | Systematic data |
| Banks' South African exposure was too small for the sanctions to matter, with the most involved bank holding only about 1.6% of loans in South Africa. (p. 58) | Bank loan data from IRRC (Bowers and Cooper) and Moody's net loans (table A1). | Systematic data |
| South African financial markets (Anglo-American, JSE indices, rand) were not adversely affected by U.S. legislative events; some reactions were perversely positive. (p. 62) | Event study using hand-collected Financial Times daily data on 10 legislative dates. | Systematic data |
| Pension fund divestment announcements did not significantly hurt exposed U.S. firms or South African financial markets. (p. 68) | Event study on 16 pension fund divestments (25 events, 21 dates) for a 17-firm portfolio and the South African series. | Systematic data |
| Voluntary divestment announcements by 46 U.S. firms produced no significant abnormal stock returns, and the returns were unrelated to South African exposure. (p. 77) | Event study (table 7) and cross-sectional regressions (table 8) using CRSP, Compustat and IRRC data. | Systematic data |
| There is weak evidence that institutional shareholdings rose around corporate divestment, relative to size/SIC-matched controls. (p. 74) | S&P stock guide and Moody's institutional data versus matched control firms (tables 5-6). | Partial data |
| Macroeconomic evidence cannot cleanly attribute South African recession or capital flight to sanctions; gold price declines and the JSE's strength confound it. (p. 49) | Descriptive IMF International Financial Statistics series, gold price and output figures, and JSE index plots (figs. 2-6). | Illustrative or anecdotal |
| U.S. portfolios exposed to South Africa did not perform unusually poorly in real time during the 1980s. (p. 79) | Cumulative abnormal returns (Moskowitz-adjusted) of the bank, exposed-firm and divestor portfolios (fig. 7). | Partial data |
The author's caveats¶
The authors note that classifying legislative events required judgment and that prior anticipation of legislation may leak information. The S&P stock guide is not completely reliable, so clear outliers were coded as missing. Matching of divesting firms was difficult, ten of 27 candidate firms were dropped for lack of CRSP data, and pension fund holdings data were unavailable. They also say daily real macroeconomic data are not available.
Gaps in the evidence¶
The study measures only stock-market reactions of U.S. firms and the JSE, not the real economy (output, employment, trade, credit costs), so it cannot show sanctions had little real effect on South Africa. The macro section is descriptive plots without a counterfactual, and no South African firm-level or Chamber of Mines/company data or primary South African archives are used. Fund-level holdings data, loan terms and spreads, and South African capital-flow microdata would test the argument better; some of these exist now in archives and datasets.
Replicability: Data described but not public.