Do Economic Sanctions Destabilize Country Leaders?

Literature › Works

Nikolay Marinov (2005). Do Economic Sanctions Destabilize Country Leaders?. American Journal of Political Science, Vol. 49, No. 3. Midwest Political Science Association.

Type	journal article
Scope	Comparative and methods
Method	econometric
Period studied	1947–1999
Regions	Global cross-national panel (136 countries)
Themes	sanctions divestment, comparative theory, politics ideology elections, methods data sources
Citation key	marinov2005

Argument

Marinov argues that sanctions work in at least one respect: they destabilize the leaders they target. He develops a simple theory in which destabilization is a necessary condition for coercion: a leader concedes only if sanctions threaten survival in office more than conceding does. Using a country-year panel of 136 countries (1947-1999) with conditional fixed-effects logit, he finds that sanctions in one year raise the chance of losing office in the next year, by an average 28%. He argues that earlier pessimism rests on selecting on the dependent variable, since studies compare only cases where sanctions were imposed. Sanctions may therefore be more effective than commonly thought, though their costs for target populations remain serious.

Evidence and method

Hufbauer, Schott and Elliott sanctions dataset (updated IIE 2002 chronology; 169 episodes, 1,181 sanctioned country-years); Archigos leader data (Chiozza and Goemans/Goemans et al.); Polity IV regime codes; World Bank income and growth data; Militarized Interstate Disputes data; Fearon and Hassan data on leadership changes in autocracies. Panel has 5,295 usable observations after fixed effects.

Findings

• Sanctions raise the next-year risk of leader loss, using panel data on 136 countries. (p. 3)

  Specifically, sanc- tions increase the baseline risk of losing power by 28%.

• The sanctions coefficient in the main fixed-effects logit is positive and significant, with lower growth also hurting leaders. (p. 8)

  A leader who is subject to economic sanctions in a given year is, on average, more likely to lose office in the following year.

• Existing studies compare only cases of coercion, so the question of whether sanctions work has not been properly tested. (p. 3)

  If the question is "Does coercion work?," we need to compare cases in which pressure was applied, to those in which it was not.

• Only regime type clearly conditions the effect; sanctions against democracies destabilize much more. (p. 11)

  Pressure directed at democracies causes much more gov- ernment instability.

• Sanctions episodes with foreign force are few; most leadership changes under sanctions in autocracies occurred without military intervention. (p. 11)

  First, there are 70 lead- ership changes under the influence of sanctions and out of the shadow of force.

• Selection effects probably bias the estimates against the hypothesis, making the test hard for the theory. (p. 12)

  Selection effects, to the extent they are present, are likely prejudicing the findings against the hypothesized relationship.

• Sanctions are more likely to be lifted after leadership change in the target, supporting the policy-change mechanism. (p. 12)

  Survival analysis on the data at hand confirms that sanctions are, in fact, more likely to be lifted after a change of power in target.

• The data show the sanctions-sample size and leader tenure used in the analysis. (p. 7)

  there are 1,089 leaders being removed from office for the 1947- 1999 period, and leaders stay in power for an average of 4.24 years.