Modeling Selection Bias in Studies of Sanctions Efficacy

Abstract

Sanctions rarely work but they continue to be used frequently by policymakers. I argue that previous studies of sanctions ignore the problem of strategic censoring by focusing only on cases of observed sanctions. In this paper, I develop a unified model of sanction imposition and success and test it using a simultaneous equation censored probit model. This selection-corrected sanction model finds that the process by which sanctions are imposed is linked to the process by which some succeed while others fail, and that the unmeasured factors that lead to sanction imposition are negatively related to their success.

Determinants of Sanctions Effectiveness: Sensitivity Analysis Using New Data

“…Positive: Morgan and Schwebach (1997), Krustev (2007) McLean and Whang (2010) [+], Nooruddin (2002) Drury (1998) [0], Jing, Kaempfer and Lowenberg (2003) [−]…”

“…1 For example, Drezner (1999) and Allen (2005Allen ( , 2008 find that sanctions against allies are more likely to succeed, but Drury (1998) and Krustev and Morgan (2011) find no support for this claim, and Nooruddin (2002) and Early (2011) find that sanctions against allies are less likely to succeed for US sanctions. Even when we consider the cost of sanctions, which many see as the most important predictor of sanctions success, the empirical findings are not conclusive (Bonetti 1998;Jing, Kaempfer, and Lowenberg 2003;Nooruddin 2002).…”

“…First, most findings reported in any one article result from one regression model (Kaempfer and Lowenberg 2007). Even though many scholars conduct tests to ensure the robustness of their findings, these attempts are often Lam (1990) [+], Drezner (1999) [+], Allen (2005) [+], Krustev and Morgan (2011) [0], Early (2011) [−], Nooruddin (2002) [−]…”

Not all dictators are equal

“…However, authoritarian backsliding can take many forms and different signs of democratic retreat are likely to cause varying levels of public and political attention. Surprisingly, existing research analyzing the imposition of sanctions has almost exclusively focused on structural variables -be they related to the sender or to the target, or to the dyadic relationship between the two 3 -and has neglected the actual autocratic behavior of the target (see for instance Cox & Drury, 2006;Drezner, 1998;Hafner-Burton & Montgomery, 2008;Lektzian & Souva, 2007;Nooruddin, 2002;Whang, 2010). 4 In addition, there has been puzzlingly little discussion on whether sanctions are mostly used as a reaction to authoritarian stability or democratic decline (Escribà-Folch & Wright, 2010;Escribà-Folch, 2012).…”

“…Moreover, the discussion about 'selection effects' has often been associated with the study of sanction effectiveness, arguing that strategic target selection has skewed the empirical results of studies focusing on the general effectiveness of these coercive measures (e.g. Drezner, 1999;Lacy & Niou, 2004;Morgan, Bapat & Krustev, 2009;Nooruddin, 2002). Lack of systematic research on the selective use of democratic sanctions is troubling, especially given new contributions to comparative democratization research suggesting that Western powers have had a tendency to promote stability rather than democracy in strategically and economically important countries (Brownlee, 2012;Donno, 2013).…”