Empirical studies of economic sanctions long rested almost exclusively on databases of imposed sanctions, chiefly the Hufbauer–Schott–Elliott (HSE) compilation. Blake and Klemm (2006) used Boolean/qualitative-comparative analysis of 111 HSE enacted episodes to show that the enacted sample is depleted of episodes under conditions favourable to the sender, implying that imposed-only samples understate the effectiveness of economic coercion. They identified as the central unresolved task the construction of a replicable dataset of threatened-but-not-enacted sanctions (private and public) with operational criteria for credible threats and threat-induced compliance, so that the unconditional effect of threats plus enacted sanctions on overall effectiveness could be estimated. The dataset-construction program was subsequently carried out in substantial part by the Threat and Imposition of Economic Sanctions (TIES) project (publicly threatened and imposed cases, 1945–2005), and threat-stage analyses now routinely find that threats succeed at non-trivial rates. The open problem is therefore narrower: the systematic measurement of threatened-but-not-enacted sanctions whose threats were privately communicated (the public-threat component having been substantially realized by TIES), and a fully unconditional effectiveness estimate spanning the entire universe of coercion episodes — threatened-but-not-enacted (public and private) and imposed alike. The identification of the causal mechanisms that produce threat-stage compliance (existing bargaining-model tests have failed; Morgan and Kobayashi 2021) and the methodological questions the paper flags about dichotomized small-N designs and covariates unobserved at the decision moment are related later-literature context, not part of the source question.

Background

Economic sanctions are deliberate restrictions by one state (the sender) on trade, finance, or other economic relations with another state (the target), intended to change the target's policy. From the 1980s the empirical benchmark was the HSE dataset (Hufbauer, Schott and Elliott, "Economic Sanctions Reconsidered"), a case-level compilation of imposed sanctions episodes, on which the influential pessimistic finding rested that sanctions "work" in only about a third of cases — a conclusion echoed in Pape's (1997) critique that sanctions almost never achieve major goals. Game-theoretic models of coercion (Eaton and Engers 1992; Drezner's "hidden hand" argument, 2003) point out a logical flaw in such estimates: if the target would concede under sanctions, it can usually concede when sanctions are merely threatened, before any costs are paid. Threatened-but-successful episodes then leave the observable record of imposed sanctions, so imposed-only samples are selected on failure — sanctions are least likely to be observed exactly when they are most likely to be effective. This is the selection-bias (or selection-effect) problem in sanctions research.

Blake and Klemm (2006) contributed an empirical test of this logic using Boolean analysis, the core operation of Qualitative Comparative Analysis (QCA), a small-N method that dichotomizes case conditions and describes outcomes as Boolean functions of those conditions rather than as continuous regression coefficients. Modeling success as St=f(Mt,Ut,Pt,It,Tt)S_t = f(M_t, U_t, P_t, I_t, T_t) — where the five dichotomous conditions are a modest sender goal (M), target instability (U), sound prior relations (P), target isolation (I), and substantial trade linkage (T) — they found that only 16 of 111 HSE enacted episodes (14.4%), and 9 of 71 non-military episodes (12.7%), occurred under "clearly favourable" conditions (four or five of the five conditions present), significantly fewer than chance baselines, while such favourable cases succeeded in 62.5% of instances against roughly 33% overall. Enacted-sanction samples are therefore depleted of favourable configurations, and effectiveness estimates built on them are biased downward. The paper closes by stating what would be needed to correct the bias: an operational definition of a credible threatened sanction, a comprehensive dataset of threatened-but-not-enacted episodes (both public and private threats), and an estimate of the unconditional effectiveness of threats plus imposed sanctions. It also flags methodological loose ends: the arbitrariness of the dichotomization thresholds (e.g., the 33 percent trade-share threshold, the 4-of-5 favourable cutoff, and HSE coding cutoffs) and the loss of continuous variation; the exclusion of cost and duration variables that are unobserved at the decision moment; and the sensitivity of results to recoding borderline HSE cases such as human-rights episodes. The paper itself notes a funded data-collection effort (NSF grant, principal investigator T. Clifton Morgan) then under way as the pathway to the threatened-sanctions data — the effort that became the TIES dataset, which records cases in which sanctions were threatened and/or imposed and codes outcomes separately at the threat and imposition stages.

Problem Statement

The source question (Blake and Klemm 2006): correct the selection bias in sanctions-effectiveness research — the bias arising because samples of imposed sanctions are selected on failure, since episodes in which the target concedes to a threat alone never enter the imposed-sanctions record — by (a) developing an operational definition of threatened economic sanctions — covering both privately communicated and publicly announced threats that did not result in imposition — with explicit criteria for what counts as a credible threat and for threat-induced compliance (a target policy concession caused by the threat rather than occurring independently of it); (b) constructing a comprehensive, replicable dataset of threatened-but-not-enacted sanctions episodes; and (c) using it to estimate the unconditional effect of threats plus enacted sanctions on the overall effectiveness of economic coercion — the success rate measured over the entire universe of coercion episodes defined by (a)–(b) (threatened-but-not-enacted episodes with publicly announced as well as privately communicated threats, plus imposed episodes), threat-stage and imposed alike, rather than over imposed episodes only — in place of the downward-biased estimates obtained from imposed-only samples. The TIES dataset (1945–2005), which operationalizes threats as officials' verbal statements, drafted legislation, or conditional laws collected from public news indexes and codes outcomes separately at the threat and imposition stages, substantially realizes the public-threat component of (a)–(b), and joint threat-and-imposition analyses of TIES substantially advance (c) for observable threats. Privately communicated threats remain unmeasured in TIES and other existing data resources, so what remains open is the private-threat component of (a)–(b) — extending the measurement of threatened-but-not-enacted sanctions to privately communicated threats, with defensible operational criteria for credibility and threat-induced compliance, in a replicable episode record that implements that definition, states the sender–target population frame, the time period covered, and the coverage criterion, and documents, for each included episode, the source establishing the private threat and its outcome — and, under (c), an unconditional effectiveness estimate spanning the full universe of episodes defined by (a)–(b), including the private-threat episodes that record leaves unobserved. Because privately communicated threats are absent from existing data, re-estimation over public-threat data alone cannot satisfy (c); existing stage-aware joint estimates are substantial partial progress, not satisfaction. Later literature additionally raises a causal-mechanism question — direct tests of the standard bargaining account of threat-stage compliance have failed (Morgan and Kobayashi 2021) — and the source flags methodological caveats about dichotomization-threshold sensitivity and the treatment of covariates unobserved at the decision moment in small-NN (small-sample) designs; these are later-literature context and flagged caveats, not part of the source question.

References

  1. Blake, Charles H., and Noah Klemm (2006). "Reconsidering the Effectiveness of International Economic Sanctions: An Examination of Selection Bias." International Politics 43(1):133–149. DOI: 10.1057/palgrave.ip.8800137. https://doi.org/10.1057/palgrave.ip.8800137
  2. Morgan, T. Clifton, Navin Bapat, and Valentin L. Krustev (2009). "The Threat and Imposition of Economic Sanctions, 1971–2000." Conflict Management and Peace Science 26(1):92–110. DOI: 10.1177/0738894208097668. https://doi.org/10.1177/0738894208097668
  3. Morgan, T. Clifton, Navin Bapat, and Yoshiharu Kobayashi (2014). "Threat and imposition of economic sanctions 1945–2005: Updating the TIES dataset." Conflict Management and Peace Science 31(5):541–558. DOI: 10.1177/0738894213520379. https://doi.org/10.1177/0738894213520379
  4. Bapat, Navin A., Tobias Heinrich, Yoshiharu Kobayashi, and T. Clifton Morgan (2013). "Determinants of Sanctions Effectiveness: Sensitivity Analysis Using New Data." International Interactions 39(1):79–98. DOI: 10.1080/03050629.2013.751298. https://doi.org/10.1080/03050629.2013.751298
  5. Whang, Taehee, Elena V. McLean, and Douglas W. Kuberski (2013). "Coercion, Information, and the Success of Sanction Threats." American Journal of Political Science 57(1):65–81. DOI: 10.1111/j.1540-5907.2012.00629.x. https://doi.org/10.1111/j.1540-5907.2012.00629.x
  6. Krustev, Valentin L. (2010). "Strategic Demands, Credible Threats, and Economic Coercion Outcomes." International Studies Quarterly 54(1):147–174. DOI: 10.1111/j.1468-2478.2009.00581.x. https://doi.org/10.1111/j.1468-2478.2009.00581.x
  7. Nooruddin, Irfan (2002). "Modeling Selection Bias in Studies of Sanctions Efficacy." International Interactions 28(1):59–75. DOI: 10.1080/03050620210394. https://doi.org/10.1080/03050620210394
  8. Peterson, Timothy M. (2013). "Sending a Message: The Reputation Effect of US Sanction Threat Behavior." International Studies Quarterly 57(4):672–682. DOI: 10.1111/isqu.12017. https://doi.org/10.1111/isqu.12017
  9. Miller, Nicholas L. (2014). "The Secret Success of Nonproliferation Sanctions." International Organization 68(4):913–944. DOI: 10.1017/s0020818314000216. https://doi.org/10.1017/s0020818314000216
  10. Drezner, Daniel W. (2003). "The Hidden Hand of Economic Coercion." International Organization 57(3):643–659. DOI: 10.1017/s0020818303573052. https://doi.org/10.1017/s0020818303573052
  11. Eaton, Jonathan, and Maxim Engers (1992). "Sanctions." Journal of Political Economy 100(5):899–928. DOI: 10.1086/261845. https://doi.org/10.1086/261845
  12. Pape, Robert A. (1997). "Why Economic Sanctions Do Not Work." International Security 22(2):90–136. DOI: 10.1162/isec.22.2.90. https://doi.org/10.1162/isec.22.2.90
  13. Morgan, T. Clifton, and Yoshiharu Kobayashi (2021). "Talking to the hand: Bargaining, strategic interaction, and economic sanctions." European Economic Review 134:103685. DOI: 10.1016/j.euroecorev.2021.103685. https://doi.org/10.1016/j.euroecorev.2021.103685
  14. Weber, Patrick, and Gerald Schneider (2020). "How many hands to make sanctions work? Comparing EU and US sanctioning efforts." European Economic Review 130:103595. DOI: 10.1016/j.euroecorev.2020.103595. https://doi.org/10.1016/j.euroecorev.2020.103595
  15. Walentek, Dawid, Joris Broere, Matteo Cinelli, Mark M. Dekker, and Jonas M. B. Haslbeck (2021). "Success of economic sanctions threats: coercion, information and commitment." International Interactions 47(3):417–448. DOI: 10.1080/03050629.2021.1860034. https://doi.org/10.1080/03050629.2021.1860034
  16. Felbermayr, Gabriel, T. Clifton Morgan, Constantinos Syropoulos, and Yoto V. Yotov (2021). "Understanding economic sanctions: Interdisciplinary perspectives on theory and evidence." European Economic Review 135:103720. DOI: 10.1016/j.euroecorev.2021.103720. https://doi.org/10.1016/j.euroecorev.2021.103720
  17. van Bergeijk, Peter A. G., and Muhammad S. H. Siddiquee (2017). "Biased Sanctions? Methodological Change in Economic Sanctions Reconsidered and Its Implications." International Interactions 43(5):879–893. DOI: 10.1080/03050629.2016.1242584. https://doi.org/10.1080/03050629.2016.1242584
  18. Demena, Binyam A., Alemayehu S. Reta, Gabriela Benalcazar Jativa, Patrick B. Kimararungu, and Peter A. G. van Bergeijk (2021). "Publication bias of economic sanctions research: a meta-analysis of the impact of trade linkage, duration and prior relations on sanctions success." In Peter A. G. van Bergeijk (ed.), Research Handbook on Economic Sanctions. Cheltenham: Edward Elgar. DOI: 10.4337/9781839102721.00012. https://doi.org/10.4337/9781839102721.00012
  19. Morgan, T. Clifton, Navin A. Bapat, and Yoshiharu Kobayashi (2021). "The Threat and Imposition of Economic Sanctions data project: a retrospective." In Peter A. G. van Bergeijk (ed.), Research Handbook on Economic Sanctions. Cheltenham: Edward Elgar. DOI: 10.4337/9781839102721.00009. https://doi.org/10.4337/9781839102721.00009