Cover: 5 of 40 claimed sanctions successes survived a re-examination of the famous 34% success rate
|

Do Economic Sanctions Work? What the Evidence Shows

Key takeaways · 11 min read

  • Pape re-examined the same cases behind the best-known sanctions success rate, 34%, and cut it to about 4%.
  • Larger datasets that count threats put success between 37.5% and 56.3%, depending on the definition.
  • Sanctions that damage an economy tend to harm health, and one study estimated 1.2 to 1.4 years of life expectancy lost under UN sanctions.
  • Studies disagree on whether UN or US sanctions do more harm, and most have serious design limits.

When governments want to act against another country without going to war, they now reach for economic sanctions. The United States, the European Union, the United Kingdom and the United Nations together maintain sanctions on dozens of countries and thousands of people and companies. The measures freeze assets, block trade, cut banks out of the payment system and ban travel.

Two questions hang over every one of them. Do they change what the targeted government does? And who actually pays? The answers matter for anyone who lives in, trades with, sends money to or travels to a sanctioned country, and for voters in the countries that impose them.

Researchers have argued about the first question for forty years. The most famous success rate, 34%, fell to about 4% when another scholar re-examined the same cases. The second question has more consistent evidence: sanctions that bite hard tend to harm ordinary people’s health, though how much, and which sanctions do it, is still disputed. This article sets out what the research finds and where it divides.

The 34% figure and the 4% reply

The modern debate starts with Economic Sanctions Reconsidered, a book by Gary Clyde Hufbauer, Jeffrey Schott and Kimberly Ann Elliott, first published in 1985. They coded every sanctions episode they could find from 1914 onwards and judged whether each had achieved its goal and whether sanctions had contributed. In the edition covering cases to 1990, 40 of 115 cases, or 34%, counted as at least partly successful.

That figure helped move expert opinion from deep pessimism to cautious optimism. In 1997, the political scientist Robert Pape went back through the 40 successes in the journal International Security. He argued that most were not really sanctions successes at all: in many, military force did the work, the target made no real concession, or the outcome was not what the sanctions had demanded. By his reading, only 5 of the 40 held up, a success rate of about 4% across all 115 cases.

Elliott replied, Pape replied to the reply in 1998, and the disagreement was never settled. Reviewing the field in 2020, Lee Jones and Clara Portela noted that the dispute centred on how to prove that sanctions, and nothing else, caused a government to change course. That is still the hardest problem in the field.

How often do sanctions succeed?

Share of sanctions episodes judged successful, depending on the dataset and the definition of success.

34%of 115 cases judged at least partly successful
5 of 40claimed successes that survived Pape’s re-examination

Hufbauer GC, Schott JJ, Elliott KA, Economic Sanctions Reconsidered, 1990; Pape RA, International Security, 1997;22(2):90–136.

A pointillist illustration: a quiet container port at dusk with stacked containers and idle cranes, one crane light glowing amber over dark water.
An idle port at dusk. Trade sanctions are designed to squeeze an economy, not a single person.

Bigger datasets, wider answers

Later researchers built larger datasets. The best known is Threat and Imposition of Economic Sanctions, or TIES, run by T. Clifton Morgan, Navin Bapat and Yoshiharu Kobayashi. It covers more than a thousand cases from 1945 to 2005 and includes threats as well as sanctions actually imposed.

How successful sanctions look in TIES depends on how success is defined. Under the strictest definition, sanctions succeeded in 37.5% of the episodes with known outcomes; under the most relaxed, 56.3%. These figures are higher partly because threats which work are counted: a government that backs down before sanctions are imposed never appears in a list of imposed sanctions.

One question, four answers

Share of sanctions cases counted as successful.

Pape re-examination, 1997~4%
Hufbauer, Schott and Elliott, to 199034%
TIES, strictest definition37.5%
TIES, most relaxed definition56.3%

Pape RA, 1997; Hufbauer GC et al, 1990; TIES dataset as summarised by Onder M, Economies, 2020. TIES includes threats as well as imposed sanctions.

A few patterns recur across these studies. Sanctions tend to work better when the demand is modest, when the sender and target are close trading partners, and when the threat alone is enough. They rarely work when the aim is to overthrow a government or to force it to give up something it sees as vital to its survival. A 2025 meta-analysis of the determinants of success, by Binyam Demena and Peter van Bergeijk, found that many of these conclusions depend heavily on which dataset and model a study uses.

Who pays: the health evidence

The humanitarian cost of sanctions became a public issue in the 1990s, through Iraq, Haiti and the former Yugoslavia. We looked separately at the disputed estimates of child deaths in Iraq. Since then, researchers have tried to measure the effect across many countries at once.

In the Journal of Peace Research in 2013, Susan Hannah Allen and David Lektzian found that sanctions with a large economic effect on the target had public health consequences comparable to those of major armed conflicts. Sanctions with little economic effect had no measurable health effect.

Jerg Gutmann, Matthias Neuenkirch and Florian Neumeier studied 98 less developed and newly industrialised countries from 1977 to 2012, in the Journal of Development Studies in 2021. They estimated that an average episode of UN sanctions reduced life expectancy by about 1.2 to 1.4 years, and US sanctions by 0.4 to 0.5 years. Women were affected more than men, and the damage built up the longer sanctions lasted.

Life expectancy under sanctions

Estimated effect of an average sanctions episode, 98 countries, 1977–2012.

1.2–1.4years of life expectancy lost under UN sanctions
0.4–0.5years lost under US sanctions

Gutmann J, Neuenkirch M, Neumeier F, Journal of Development Studies, 2021;57(1):139–162. Women were affected more than men.

The studies disagree about which sanctions do the harm

A newer study reversed part of that picture. In The Lancet Global Health in 2025, Francisco Rodríguez, Sílvio Rendón and Mark Weisbrot analysed age-specific death rates in 152 countries from 1971 to 2021 using several methods designed to separate cause from correlation. They found the strongest effects for unilateral, economic and US sanctions, and no statistical evidence of an effect for UN sanctions. The effect was largest for children under five. As summarised later in the same journal, the estimate amounted to more than 560,000 deaths a year associated with unilateral sanctions.

A pointillist illustration: a pharmacy shelf with several empty gaps between a few boxes of medicine, a small amber price label on one box.
Gaps on a pharmacy shelf. Medicines are usually exempt, but supplies still fall.

Both studies are careful, yet they point in different directions on a key question. A 2023 systematic review in BMJ Global Health by Matteo Pinna Pintor, Marc Suhrcke and Christoph Hamelmann helps explain why. Of the 27 studies it synthesised, 21 reported consistent adverse effects of sanctions. Almost all had important limitations, though, and only one used a quasi-experimental design. Sanctioned countries also differ from others in ways that are hard to adjust for: they are often already in conflict, under autocratic rule or in economic decline. The review concluded that more evidence is needed on which sanctions cause how much harm.

Writing about Iran in 2026, Reza Majdzadeh added a further caution. Iran is one of the most heavily sanctioned countries, and life expectancy there kept rising through long periods of sanctions. He argued that life expectancy is a late and blunt signal, and that harm can show up first in medicine shortages, delayed treatment and household poverty long before it moves national averages.

Cross-country studies of sanctions and health

Main findings, in order of publication.

StudyCountries and yearsMain finding
Allen and Lektzian, 2013Cross-nationalHarm only when the economic effect is large
Gutmann and colleagues, 202198 countries, 1977–2012UN sanctions cost more years than US ones
Pinna Pintor and colleagues, 2023Review of 27 studies21 found harm; most had serious limitations
Rodríguez and colleagues, 2025152 countries, 1971–2021US and unilateral strongest; no effect found for UN

Allen SH, Lektzian DJ, J Peace Res, 2013; Gutmann J et al, J Dev Stud, 2021; Pinna Pintor M et al, BMJ Global Health, 2023; Rodríguez F et al, Lancet Glob Health, 2025.

Why exemptions do not fully protect people

Partly in response to Iraq, the UN and Western governments shifted from the late 1990s towards targeted or ‘smart’ sanctions: asset freezes and travel bans on named leaders, companies and sectors rather than whole economies. The idea is to put pressure on the people making decisions while sparing everyone else. Evidence that targeted sanctions work better is limited, and the line between targeted and comprehensive becomes thin when they hit a central bank, oil exports or access to international payments, as many now do.

Almost every modern sanctions programme exempts food, medicine and humanitarian aid. The problem is that trade in exempt goods still needs banks, shipping, insurance and paperwork, and each company in the chain has to decide whether the risk of a mistake is worth the business.

Many decide it is not. Researchers call this over-compliance, or de-risking: banks refuse transfers to anyone connected to a sanctioned country, and suppliers stop selling even permitted goods. Reviews of the Iranian experience describe medicines that were technically allowed becoming scarce and expensive because payments could not be processed. Migrants who send money home, charities working in sanctioned areas and ordinary people with the wrong country on their passport face the same pattern.

Sanctions can also strengthen the governments they target. In International Studies Quarterly in 2008, Reed Wood found that sanctions between 1976 and 2001 were followed by increased state repression, as leaders tightened their grip in response to the threat. A 2023 review of 31 quantitative studies by Francisco Rodríguez in the Journal of Economic Studies found that 30 reported negative effects on outcomes ranging from income and poverty to mortality and human rights.

What the quantitative studies report

Studies that measured the effect of sanctions on living conditions in target countries.

30 of 31studies found negative effects on living standards, health or rights
21 of 27studies in a health review reported consistent harm

Rodríguez F, Journal of Economic Studies, 2023; Pinna Pintor M, Suhrcke M, Hamelmann C, BMJ Global Health, 2023.

A pointillist illustration: a bank counter with closed grilles and a small handwritten notice, a queue barrier in front and a single amber lamp above.
A closed bank counter. Much of the harm travels through payments, not through bans on goods.

Recent cases: Russia and Syria

The sanctions imposed on Russia after its full-scale invasion of Ukraine in February 2022 are the largest ever placed on a major economy. They froze a large part of the central bank’s reserves, cut major banks from the SWIFT messaging system, banned many exports and later capped the price of Russian oil sold with Western shipping and insurance.

The economic effect was real but smaller than many predicted. Official Russian figures showed a contraction of about 1% in 2022 and growth of more than 3% in 2023, driven by military spending and oil sold through other routes. By 2024 the sanctions had not ended the war, which was their central aim. Supporters argue that the goal also included raising the cost of the war and degrading Russia’s military supply chains over time, which is harder to measure.

Syria poses the reverse question. After the fall of the Assad government in December 2024, the European Union and the United States moved in 2025 to lift most economic sanctions, arguing that they now blocked recovery more than they pressured anyone. Researchers will be able to measure what happens to health and incomes when sanctions end, which is rarer and in some ways more informative than studying when they begin.

What this means for individuals

For most readers, sanctions are something that happens to other people, yet they reach ordinary life in practical ways. Money transfers to relatives in some countries can be blocked or delayed. Travel insurance may exclude sanctioned destinations. Buying from or selling to a sanctioned person, even unknowingly, can freeze a payment or an account.

Official sanctions lists are public, and governments publish guidance on humanitarian exemptions and licences. If you send money to family, work for a charity or run a small business with links to a sanctioned country, check the rules that apply to your bank and your country before a payment is stuck.

What the official pages say about sanctions and civilians

This section collects what the UN, US, EU and UK pages say about humanitarian exemptions from sanctions, and what the US audit agency says about whether sanctions work. It is general information, not legal advice.

Source and what it says
SourceWhat it says
UN Security Council, press release on Resolution 2664 (meeting 9214)Creates a standing humanitarian exemption to UN asset freezes, so funds, goods and services needed for timely humanitarian aid do not breach them. Adopted 14 to 0 with one abstention (India). For the ISIL and Al-Qaida regime it runs two years before a decision on extension, and it applies to future asset freezes unless a decision says otherwise.
US Treasury OFAC, supplemental humanitarian guidance (Feb. 27, 2023)Says the December 2022 general licences implement Resolution 2664 and are written by user: US government, international organisations, NGOs, and agricultural and medical goods. They amended more than 30 programmes but not the comprehensively sanctioned jurisdictions. General licences need no pre-approval, and the fact sheet is informational and not law.
US Treasury, Afghanistan general licences 17 to 19 (Dec. 22, 2021)Implements the UN carve-out in Resolution 2615. It does not authorise payments to the Taliban or Haqqani Network except taxes, fees and similar charges.
European Commission, humanitarian aid guidance note (June 30, 2022)An exemption means the restriction does not apply to humanitarian actions, with no need to notify an authority, but the operator bears responsibility for confirming the purpose. Who may use exemptions varies by sanctions regime, and derogations need prior authorisation. It predates Resolution 2664.
UK OFSI, Syria humanitarian general licence INT/2025/5810196 (Feb. 12, 2025)Allows payments for relevant humanitarian assistance activities under the Syria Regulations 2019. The licence terms sit in a linked PDF that we did not read.
US Government Accountability Office, GAO-20-145 (Oct. 2, 2019)Says agencies have not determined whether sanctions work and find their effects hard to isolate. Studies it reviewed say sanctions may have unintended consequences, including for human rights and public health.
GAO, GAO-25-107079 (Sept. 8, 2025)Agencies have not set measurable objectives for Russia sanctions and export controls. The page does not discuss humanitarian effects.

What we could not confirm: no official page we read measures civilian harm caused by sanctions, and GAO-20-145 only notes unintended consequences from academic studies without quantifying them. No official page gives a verdict on effectiveness. Who can use the exemptions differs: OFAC writes them by user type, the EU by regime, and the UN carve-out by regime and renewal. The UN press page we read carries no calendar date, so we give none. We could not confirm whether OFAC’s 2023 fact sheet has since been replaced, or the expiry dates of OFSI’s licences. UN Special Rapporteur reports and a Congressional Research Service primer were blocked, so we report nothing from them. Compliance questions need a qualified lawyer.

Questions people ask

Do economic sanctions work?

Sometimes, for modest goals. Estimates of success range from about 4% to 56%, depending on the dataset and how success is defined.

What is the 34% sanctions success rate?

It comes from Hufbauer, Schott and Elliott, who judged 40 of 115 cases to 1990 as at least partly successful. Pape argued only 5 of those 40 held up.

Do sanctions harm ordinary people?

Most studies find that sanctions with a large economic effect worsen health and living standards, especially for children and women.

Are UN or US sanctions more harmful?

Studies disagree. One found larger effects for UN sanctions; a 2025 study found the strongest effects for US and unilateral sanctions and none for UN ones.

Are medicines exempt from sanctions?

Usually, but banks and suppliers often avoid sanctioned countries altogether, so exempt medicines can still become scarce.

The short version

  • Pape re-examined the same cases behind the best-known sanctions success rate, 34%, and cut it to about 4%.
  • Larger datasets that count threats put success between 37.5% and 56.3%, depending on the definition.
  • Sanctions that damage an economy tend to harm health, and one study estimated 1.2 to 1.4 years of life expectancy lost under UN sanctions.
  • Studies disagree on whether UN or US sanctions do more harm, and most have serious design limits.
  • Exemptions for food and medicine often fail in practice because banks and suppliers avoid the risk.

This article summarises published research on economic sanctions. It does not take a position on any particular sanctions programme or the conflicts behind them, and it is not legal or financial advice. If you need to send money or goods to a sanctioned country, check the official guidance for your country or speak to your bank.

Further reading. Pape, ‘Why Economic Sanctions Do Not Work’, International Security, 1997, and Pinna Pintor and colleagues, ‘The impact of economic sanctions on health and health systems in low-income and middle-income countries’, BMJ Global Health, 2023, are good starting points.

Sources

  1. Hufbauer GC, Schott JJ, Elliott KA. Economic Sanctions Reconsidered: History and Current Policy. 2nd ed. Institute for International Economics, 1990.
  2. Pape RA. Why economic sanctions do not work. International Security, 1997;22(2):90–136. doi:10.1162/isec.22.2.90.
  3. Pape RA. Why economic sanctions still do not work. International Security, 1998;23(1):66–77. doi:10.1162/isec.23.1.66.
  4. Morgan TC, Bapat N, Kobayashi Y. Threat and imposition of economic sanctions 1945–2005: updating the TIES dataset. Conflict Management and Peace Science, 2014;31(5):541–558. doi:10.1177/0738894213520379.
  5. Onder M. Regime type, issue type and economic sanctions: the role of domestic players. Economies, 2020;8(1):2. doi:10.3390/economies8010002.
  6. Jones L, Portela C. Evaluating the success of international sanctions: a new research agenda. Revista CIDOB d’Afers Internacionals, 2020;125:39–60. doi:10.24241/rcai.2020.125.2.39.
  7. Demena BA, van Bergeijk PAG. A meta-analysis of determinants of success and failure of economic sanctions. Econometrics, 2025;13(2):16. doi:10.3390/econometrics13020016.
  8. Allen SH, Lektzian DJ. Economic sanctions: a blunt instrument? Journal of Peace Research, 2013;50(1):121–135. doi:10.1177/0022343312456224.
  9. Gutmann J, Neuenkirch M, Neumeier F. Sanctioned to death? The impact of economic sanctions on life expectancy and its gender gap. Journal of Development Studies, 2021;57(1):139–162. doi:10.1080/00220388.2020.1746277.
  10. Rodríguez F, Rendón S, Weisbrot M. Effects of international sanctions on age-specific mortality: a cross-national panel data analysis. The Lancet Global Health, 2025;13(8):e1358–e1366. doi:10.1016/S2214-109X(25)00189-5.
  11. Majdzadeh R. Life expectancy is a late signal of sanctions-related harm. The Lancet Global Health, 2026. doi:10.1016/j.langlo.2026.104046.
  12. Pinna Pintor M, Suhrcke M, Hamelmann C. The impact of economic sanctions on health and health systems in low-income and middle-income countries: a systematic review and narrative synthesis. BMJ Global Health, 2023;8(2):e010968. doi:10.1136/bmjgh-2022-010968.
  13. Rodríguez F. The human consequences of economic sanctions. Journal of Economic Studies, 2024;51(4):942–963. doi:10.1108/JES-06-2023-0299.
  14. Wood RM. ‘A hand upon the throat of the nation’: economic sanctions and state repression, 1976–2001. International Studies Quarterly, 2008;52(3):489–513. doi:10.1111/j.1468-2478.2008.00512.x.
  15. UN Security Council; OFAC; US Treasury; European Commission; OFSI; GAO-20-145; GAO-25-107079.

Similar Posts