Skip to content

On the Determinants of the Success of Economic Sanctions: An Empirical Analysis

Literature › Works

Jaleh Dashti-Gibson; Patricia Davis; Benjamin Radcliff (1997). On the Determinants of the Success of Economic Sanctions: An Empirical Analysis. American Journal of Political Science, Vol. 41, No. 2. Midwest Political Science Association.
Type journal article
Scope Comparative and methods
Method econometric
Period studied 1914–1989
Regions International (cross-national sanction episodes), predominantly US-imposed sanctions
Themes sanctions divestment, comparative theory, methods data sources
Citation key dashtigibson1997

Argument

The paper tests the determinants of sanctions success on the Hufbauer, Schott and Elliott (1990) universe of sanction episodes from 1914 to 1989, using logistic regression on a dichotomised policy-result outcome. Success depends on the sender's goal. For destabilization, the target's initial stability matters most, while cost and duration matter little. For other goals (e.g. compelling policy change such as ending apartheid), financial sanctions raise the probability of success, and there is a modest downward trend over time. The authors infer that sanctions work by hitting policy-making elites directly, not by crippling the whole economy.

Evidence and method

Hufbauer, Schott and Elliott (1990) dataset of sanction episodes, 1914-1989: 29 destabilization cases and 80-81 other-goal cases. Variables: cost to target as % of GNP, trade linkage, target health/stability (1-3), episode length, financial-sanction dummy and year. Diagnostics use Pregibon's dfbetas.

Findings

  • Success depends on the sender's goals, with different determinants for destabilization versus policy-change sanctions. (p. 1)

    The factors affecting success depend upon the goals of the sending nations. When that goal is simply destabilization, the principal determinant of success is the initial stability of the target.

  • The authors criticise the Hufbauer et al. success score as unreliable and conceptually unsuited to statistical estimation, and use the policy result alone as a dichotomy. (p. 5)

    Defining the success of a sanctions episode as the product of the two indices may be perfectly appropriate for their purposes, but it makes little sense from the perspective of quantitative analysis.

  • For destabilization, higher cost, shorter episodes and weaker targets are associated with success. (p. 7)

    Substantively, the model suggests that sanctions are likely to succeed the greater the costs imposed, the shorter the sanction's episode, and the weaker or more unstable the target.

  • Target health has an enormous effect on destabilization probability, while changes in cost or duration have only marginal effects. (p. 8)

    If instead of being generally stable the target has ''significant problems'' (health = 2), the probability of a successful outcome increases to .61; if the target is ''distressed'' (health = 1), the probability soars to .80.

  • For non-destabilization goals, financial sanctions almost double the probability of success from a reference case of .37. (p. 8)

    The use of financial sanctions has a much stronger impact, increasing the probability of compliance almost twofold to .64.

  • Weak targets are easily destabilized, but economy-crippling sanctions are not productive; stronger states resist high costs. (p. 9)

    In neither case does the ''common sense'' approach of crippling a nation's economy seem to be especially productive.

  • Policy-change sanctions show a modest downward trend in success over time, possibly due to declining credibility or symbolic use, mainly by the US. (p. 10)

    for goals other than destabilization, there is a modest but real downward trend over time in the likelihood that sanctions will succeed.

  • The authors argue that direct costs to elites matter more than aggregate costs, and that efficacy and ethics may coincide. (p. 9)

    If our conclusions are correct, questions of efficacy and morality may dovetail nicely: financial sanctions are not only more ethical, they are also more effective.

  • The theoretical rationale is that policymakers yield to pressure only when they or their elite supporters suffer severe losses. (p. 8)