Why Economic Sanctions Do Not Work¶

Literature › Works

Robert A. Pape (1997). Why Economic Sanctions Do Not Work. International Security, Vol. 22, No. 2 (Autumn/Fall 1997). MIT Press.


Type	journal article
Scope	Comparative and methods
Method	mixed
Period studied	1914–1990
Regions	International, Rhodesia, Lesotho, South Africa, Iraq
Themes	sanctions divestment, comparative theory, methods data sources
Citation key	`pape1997a`

Argument¶

Pape challenges the optimism about sanctions that rests on the Hufbauer-Schott-Elliot (HSE) database, which reports a 34 percent success rate. Recoding HSE's 40 claimed successes, he finds that most were settled by force, involved no real concessions, or were trade disputes rather than sanctions, leaving only 5 clear successes. He then argues deductively that modern states resist sanctions because of nationalism, administrative capacity to substitute, and the ability of elites to shift costs onto the weak. Statistical stratification finds no significant link between GNP loss and success. South Africa appears only peripherally: as coercer of Lesotho, and in a speculative remark that apartheid South Africa may have been unusually vulnerable because of highly unequal incomes.

Evidence and method¶

The HSE 115-case sanctions database (1985, 1990 editions), recoded case by case in the appendix using secondary literature, memoirs, archival references (British Treasury, Foreign Relations of the United States, Public Records Office), news reports and EIU reviews. Includes simple stratification by GNP loss and an OLS regression.

Findings¶

HSE's claimed 34% success rate is not supported on recoding; only 5 of 115 cases count as successes. (p. 5)

Practically none of the claimed 40 successes of economic sanctions stands up to examination. Eighteen were actually settled by direct or indirect use of force
Breakdown of HSE's errors among the claimed successes. (p. 11)

18 were determined by force, not economic sanctions; 8 are failures, in which the target state never conceded to the coercer’s demands; 6 are trade disputes, not instances of economic sanctions; and 3 are indeterminate.
Corrected overall success rate is far lower than HSE's. (p. 18)

This leaves a total of 85 valid cases, including 5 successes, or an overall success rate of 6 percent (9 percent if we include the overdetermined and indeterminate cases) versus the 34 percent HSE report.
Sanctions success is not significantly related to the economic harm inflicted on the target. (p. 20)

Ordinary least squares regression confirms that there is no statistically significant relationship between target state GNP loss and sanctions success.
Even extreme punishment (Iraq) has not produced capitulation; states are not fragile. (p. 18)

Iraq, for example, has been subjected to the most extreme sanctions in history—48 percent of its GNP has been eliminated by sanctions for over five years—and it has not buckled.
Regimes can shift sanctions' burden onto the powerless, as in Rhodesia, where black incomes fell while white living standards rose. (p. 19)

Thus it was possible for economic sanctions to cause the incomes of blacks in Rhodesia to fall while the standard of living for whites rose.
Rhodesia's white government fell to guerrilla war, not sanctions. (p. 30)

This outcome was determined by military force, not economic sanctions. By 1979 the government realized that it would be defeated by the guerrillas, and it surrendered.
Rhodesian growth was robust under sanctions, and decline after 1975 stemmed from the war. (p. 30)

while Rhodesia’s gross domestic product fell 5 percent from 1965 to 1966 as an immediate result of sanctions, the real annual growth rate from 1967 to 1974 averaged almost 8 percent
South Africa's coercion of Lesotho is coded overdetermined: economic and military pressure were each sufficient. (p. 47)

Although South Africa employed both eco-nomic sanctions and military pressure, the power imbalance between it and Lesotho was so great that either alone would almost certainly have been sufficient