Skip to content

Effect of Announcements of Withdrawal from South Africa on Stockholder Wealth

Literature › Works

Martin B. Meznar; Douglas Nigh; Chuck C. Y. Kwok (1994). Effect of Announcements of Withdrawal from South Africa on Stockholder Wealth. Academy of Management Journal, Vol. 37, No. 6. Academy of Management.
Type journal article
Scope Comparative and methods
Method econometric
Period studied 1985–1989
Regions United States, South Africa
Themes sanctions divestment, business and state, comparative theory, methods data sources
Citation key meznar1994

Argument

The authors use an event study to test how the US stock market reacted to 39 US multinationals' announcements that they were leaving South Africa. They set the 'dominant benefits' hypothesis (new business opportunities, lower issue-management costs) against the 'dominant costs' hypothesis (lost cash flows, poor sale proceeds). They find significant negative cumulative abnormal returns, which supports the dominant costs view. Early withdrawers (1986 or before) lost much more value than late withdrawers (1987 or after), because pressure and sanctions grew over time. The authors conclude that managers served stakeholders other than stockholders, which supports stakeholder theory, and they reject Innes's claim that disinvestment was a profitable ploy.

Evidence and method

Event-study data on 39 US publicly traded firms, drawn from 207 US firms that left South Africa from the early 1970s to January 1991 (lists from UN 1990/1991 and the Investor Responsibility Research Center). Announcement dates come from seven major US newspapers. Stock returns are tested with cumulative abnormal returns and Z-statistics following Mikkelson and Partch, and the two timing groups are compared following Miles and Rosenfeld. Background comes from press reports, UN hearings and IRRC data.

Findings

  • Announcements of withdrawal were associated with significantly negative abnormal returns, supporting the dominant costs hypothesis. (p. 10)

    Thus, results support Hypothesis lb, the dominant costs hypothesis. Announcements of withdrawal from South Africa are associated with significant decreases in stock value.

  • The cumulative abnormal return was -5.46% over days -30 to +10 (Z = -2.58, p < .01). (p. 10)

    Days -30 to 10 -0.0546 -2.58**

  • The negative effect appears before the press announcement, as information leaked, and persists after it. (p. 10)

    In the South African case, the negative effects of the information are evident well before the announcement makes it into the press, and the negative impact on share value continues for several days after publication.

  • By day 10 the stocks of the studied firms were about 5.5 percent lower than they would otherwise have been. (p. 10)

    By day 10, the stocks of the studied firms were, on the average, almost 5.5 percent lower than they would have been had the firms not withdrawn from South Africa.

  • Early withdrawers lost far more value than late withdrawers, and the difference is significant at p < .05. (p. 11)

    However, it decreased much more for the early withdrawers, falling over 8 percent for them and slightly over 2 percent for the late group.

  • Investors viewed later withdrawal announcements less unfavourably, which supports the timing hypothesis. (p. 11)

    Investors saw later announcements of withdrawal from South Africa less unfavorably than earlier ones.

  • The authors read the result as managers sacrificing stockholder interests to other stakeholders. (p. 13)

    The evidence of this study is that this change was, in general, contrary to the economic interests of the stockholders.

  • The sample may understate the true negative effect through self-selection of announcers. (p. 13)

    If the firms that most expected negative market reactions avoided publicly announcing their withdrawal intentions, the results of this study may be biased and actually understate the full magnitude of the negative impacts.

  • The results contradict Innes's claim that disinvestment let firms profit even more from South African business. (p. 14)

    Our results do not support this conclusion. As far as the stock market was concerned, disinvestment was a net loss to the exiting firms.

  • The 1986 Comprehensive Anti-Apartheid Act and the Rangel Amendment (1987) formed the US legal context of disinvestment. (p. 3)

    In 1986, the U.S. Congress overrode a presidential veto and passed the Comprehensive Anti-Apartheid Act, "imposing one of the most far-reaching packages of [economic] sanctions to date"