Economic sanctions rarely achieve their stated goals

The claim asks whether economic sanctions achieve publicly stated policy objectives only rarely, and its resolution turns on both observed outcomes and the rules used to define a sanctions episode, its objective, and its success. The central distinction is between sanctions' ability to inflict economic costs and their ability to cause the demanded political result; related distinctions separate imposed measures from threats, full and durable compliance from partial or temporary concessions, and coercion from constraint, deterrence, and signaling. Accordingly, the narrow proposition that imposed sanctions usually fail to secure major behavioral compliance can be assessed more firmly than the broader proposition that sanctions rarely accomplish any publicly stated purpose. The strongest support for the claim is that major policy compliance occurs in only a minority of imposed sanctions episodes, while stricter causal coding yields an even lower estimate. Pape's influential reassessment of the comparatively favorable Hufbauer case list concluded that only about five episodes were clear sanctions successes, rejecting many nominal successes because military force, covert action, or independently motivated policy changes better explained the outcome. Broader reviews and databases likewise report heterogeneous rather than generally successful political outcomes, even though their episode coverage and success coding differ (see Figure 1). Even research finding that imposed sanctions can work under identifiable conditions treats substantial target costs and limited demands as favorable conditions, not as features of every campaign. A second line of support is that economic injury does not reliably produce the specific political concession sought by the sender. A major review finds that sanctions measurably depress trade and economic activity but separates that established economic effect from the harder causal question whether the resulting pain induces compliance. Cross-national evidence also finds that sanctions, especially comprehensive ones, can reduce democratic freedoms as target governments repress opposition and control scarce resources, illustrating how political adaptation can frustrate broad policy aims despite substantial pressure. Research on targeted UN sanctions similarly finds coercing behavioral change less effective than signaling disapproval or constraining capabilities, which directly supports the claim when the stated objective is coercive compliance. Prominent success stories provide limited proof of sanctions acting independently because their outcomes have substantial competing or complementary explanations. Research on South Africa attributes limited direct economic effects to international trade and financial sanctions and places greater weight on domestic political change and longer-running internal and external pressures. Libya's concessions followed sanctions and diplomacy operating alongside security concerns, intelligence, negotiations, and the possibility of force, making the case evidence of a contribution within a package rather than sanctions-only causation. Sanctions also formed part of the pressure surrounding Iran's acceptance of the 2015 nuclear agreement, but Iran retained nuclear capabilities and the agreement later unraveled, so an interim negotiated concession did not amount to durable resolution of the underlying dispute. Low coercive reliability matters more because unsuccessful or prolonged campaigns can impose collateral costs rather than merely leave policy unchanged. A review of four decades of research associates sanctions, particularly broad measures, with worse health and humanitarian outcomes, while emphasizing causal-identification difficulties and wide variation by design and context. Those harms do not prove failure against a stated objective, but they raise the practical cost of treating a policy instrument with uncertain coercive effects as routine. The strongest challenge is that a minority success rate does not by itself establish that sanctions succeed only “rarely,” especially for a tool used in difficult disputes and evaluated under inconsistent standards. The earlier Hufbauer coding treated roughly one-third of episodes as at least partly successful, whereas Pape's much lower result followed unusually strict requirements for both success and causal attribution. Later global data and syntheses show meaningful variation across objectives, designs, institutions, and sender-target conditions rather than near-universal failure. Studies limited to sanctions actually imposed may systematically understate sanctions diplomacy because they omit concessions made after a credible threat. The TIES dataset records both threats and impositions and documents that some targets concede before sanctions are implemented. Bargaining research therefore characterizes imposed episodes as a selected set of disproportionately hard cases, because easier targets may yield during the threat stage. A low success rate among imposed sanctions consequently does not identify the effectiveness of the broader strategy of threatening and, if necessary, imposing economic penalties. Aggregate averages also obscure the materially better prospects of campaigns operating under favorable conditions. Comparative analysis and meta-analysis find that outcomes vary systematically with the objective, sanctions type, target costs, political and institutional context, international cooperation, duration, and sender-target conditions. Limited demands and substantial target costs are among the identifiable conditions associated with effectiveness, so an average that combines such cases with maximalist or poorly matched campaigns has limited predictive value for policy design. Case evidence further shows that sanctions can make a consequential contribution to negotiated concessions even when they are neither sufficient nor solely causal. The Libya analysis finds that sanctions and diplomacy contributed to concessions over Lockerbie and weapons programs in interaction with other pressures. In Iran, multilateral and national sanctions created substantial pressure and formed part of the context for the 2015 nuclear agreement. These cases do not establish sanctions-only success, but they refute the strongest reading that sanctions almost never matter to achieving a stated objective. The claim is substantially more defensible for direct coercion than for the full set of officially intended sanctions functions. Sanctions research distinguishes coercion from constraint, deterrence, signaling, stigmatization, containment, and domestic or normative purposes, so a compliance-only metric can omit goals that policymakers officially pursue. UN targeted-sanctions evidence reports comparatively weak coercion but better performance in signaling and constraining capabilities. Thus, counting every officially stated function broadens the denominator of possible achievements and weakens the categorical claim, whereas asking whether sanctions force a target to alter contested conduct strengthens it. Reported effectiveness also depends on whether partial, temporary, and jointly produced outcomes count as success. Iran illustrates the distinction: sanctions helped create pressure for an agreement containing nuclear limits, but retained capabilities and later unraveling prevent that episode from demonstrating permanent achievement of the ultimate objective. A coding rule accepting meaningful interim concessions will therefore produce higher success rates than one requiring full, lasting achievement attributable primarily to sanctions. No single percentage resolves the dispute because datasets differ in episode boundaries, demands, treatment of threats, causal attribution, and definitions of partial success. The large gap between favorable coding and Pape's strict reassessment reflects not only different observations but also different judgments about what qualifies as success and what caused it. The most defensible scope statement is therefore that major coercive goals are achieved in a minority of imposed cases, while the precise rate and the label “rarely” remain measurement- and context-dependent. The evidence base covers each expected section, but unresolved identification and classification problems limit how precisely it can answer the claim. There is no common counterfactual showing what each target would have done without sanctions, and outcomes produced by sanctions alongside diplomacy, security pressure, or domestic change resist clean causal allocation. Threat-stage concessions create selection problems, while episode boundaries, public versus inferred demands, partial success, durability, and multiple simultaneous objectives remain inconsistently classified. The bundle's dominant residual uncertainty is unresolved conflict-of-interest classification rather than an absence of pro, con, or conditional evidence. That uncertainty counsels caution in weighting individual sources, although the principal conclusions recur across peer-reviewed studies, data analyses, reviews, meta-analyses, institutional reporting, and government material. On balance, the evidence supports the claim with high confidence when “stated goals” means major behavioral compliance from imposed sanctions, but supports only a qualified version when threats, partial concessions, constraint, deterrence, and signaling are included. The central empirical finding is not that sanctions almost never matter, but that the ordinary imposed campaign does not reliably secure full and durable coercive success. The dominant substantive uncertainty is definitional and causal—whether to count threat-stage, partial, temporary, multi-instrument, constraining, and signaling outcomes—while unresolved conflict-of-interest classifications remain the principal source-weighting uncertainty.