I. The Number the Diplomats Will Not Say
In July 2025, a peer-reviewed study published in The Lancet Global Health concluded that unilateral economic sanctions are associated with an annual death toll of 564,258 people, with a 95 per cent confidence interval of 367,838 to 760,677 (Rodriguez, Rendon, and Weisbrot 2025, e1363). Children under the age of five account for 51 per cent of those deaths. Children under fifteen and people over sixty together account for more than three-quarters of the total. These are not combatants. They are not political actors. They are dying at a rate that matches or exceeds the annual toll of most of the world’s armed conflicts, and they are dying from a weapon whose name appears in no prohibited-weapons treaty and whose use is described in press releases as “diplomatic pressure.”
The study, conducted by Francisco Rodriguez of the University of Denver and Silvio Rendon and Mark Weisbrot of the Centre for Economic and Policy Research, is methodologically significant because it is the first to establish causation rather than correlation. Using data from 152 countries spanning fifty years, the authors applied entropy balancing, Granger causality tests, and instrumental variables to isolate the effect of sanctions on mortality from pre-existing trends. The lead author has served as an expert witness in sanctions-related litigation involving Venezuela, and the research was CEPR-funded; both facts are disclosed in the paper. The paper was peer-reviewed and published in one of the world’s most rigorous medical journals. Its methodology survived that scrutiny. Its findings did not make the front pages of the newspapers that faithfully report each year’s figures for battlefield deaths.
That silence is not an accident. It is the first thing that needs to be explained.
II. How the Weapon Works
The Rodriguez study’s most forensically important finding is the gap between two kinds of sanctions regime. Unilateral sanctions, particularly those imposed by the United States and the European Union, show a statistically robust causal association with increased mortality. United Nations sanctions show no such effect. The researchers found no statistically significant mortality impact from multilateral, Security Council-mandated sanctions (Rodriguez, Rendon, and Weisbrot 2025, e1362).
This distinction exposes the specific architecture of the harm. UN sanctions, subject to broader international deliberation, tend to be narrowly targeted at individuals and specific entities. Unilateral sectoral sanctions operate differently. They are designed to destroy macro-economic capacity: to cut off energy revenues, freeze central bank reserves, and sever trade links at the systemic level. The stated goal is to create conditions of economic deterioration so severe that a government changes course. The population is the instrument through which that pressure is applied.
Most sanctions regimes formally exempt humanitarian goods. Medicine is permitted. Food is permitted. Surgical equipment is permitted. The exemptions are real, and they are also, in practice, almost entirely irrelevant. International banks, facing the secondary reach of US and EU sanctions enforcement, refuse to process any transaction involving a sanctioned state regardless of what the transaction is for. The legal and reputational risk of being associated with a sanctioned country outweighs the commercial value of any single transfer. The result is what researchers and practitioners call “over-compliance.”
The exemptions are real, and they are also, in practice, almost entirely irrelevant. International banks refuse to process any transaction involving a sanctioned state regardless of what the transaction is for. The result is what researchers and practitioners call “over-compliance.”
Human Rights Watch documented this mechanism in detail in its 2019 report on Iran. Broad sanctions against Iranian banks, combined with aggressive rhetoric from US officials, had deterred international banks and firms from participating in any commercial or financial transactions with Iran, including for exempted humanitarian transactions, due to the fear of triggering US secondary sanctions on themselves. The report documented shortages of epilepsy medication, chemotherapy drugs, and basic surgical supplies (Human Rights Watch 2019). When Secretary of State Mike Pompeo was asked about the impact on ordinary Iranians in February 2019, his response was direct: the situation was getting much worse for the Iranian people, and the administration was convinced that this would lead them to rise up and change the behaviour of the regime. Collective punishment, administered at scale, described as strategy.
The Rodriguez data show that the mortality impact grows with each successive year of sanctions, rising from 5.8 log points in the first three years after adoption to 10.0 log points after seven or more years (Rodriguez, Rendon, and Weisbrot 2025, e1362). The longer a siege runs, the more it kills. Iran’s sanctions have now run for over forty years.
III. The Ecosystem That Makes It Acceptable
Half a million deaths a year would ordinarily generate sustained international pressure. It does not, because a coordinated ecosystem of institutions actively frames those deaths as either invisible, deserved, or a lesser evil than the alternative. That ecosystem consists of three interlocking parts: the major media, the Western-headquartered human rights industry, and the academic and policy institutes that supply the intellectual scaffolding for both.
Consider how the New York Times, the BBC, and Reuters typically cover a sanctioned country. The format is consistent. The story opens with the political repression: detained journalists, fraudulent elections, imprisoned dissidents. The sanctions appear in the third or fourth paragraph as a response to that repression: the international community has “imposed measures,” the government has “faced pressure,” the West has “demanded accountability.” The deaths in the pharmacy queue, the child who cannot receive insulin, the woman with cancer who cannot afford the medication that is technically available but practically inaccessible because no bank will process the payment: these appear, if at all, as a subordinate clause. “Critics say sanctions have hurt ordinary people,” the article will note, then pivot back to the political prisoners who give the story its moral clarity. The structure of the story trains the reader to understand the suffering of the many as the regrettable cost of standing up for the few.
Human rights organisations have been essential to constructing this framework. Human Rights Watch openly lobbies for sanctions against specific governments as a tool of rights enforcement. It called on the international community to impose targeted sanctions and an arms embargo on the Burmese military following the Rohingya genocide (Human Rights Watch 2017). Amnesty International made equivalent calls in the same situation (Amnesty International 2017). Neither position is without moral logic: targeted sanctions against named perpetrators of mass atrocities are a different instrument from the sectoral sanctions that the Rodriguez study measures. The problem is that the legitimacy of targeted Magnitsky-style designations has been systematically used to provide cover for the sweeping economic warfare that operates beneath that designation. When Human Rights Watch calls for sanctions on a military junta, and the US Treasury responds with sanctions on the energy sector of the same country, the political architecture of advocacy has been converted into a licence for something qualitatively different. Human Rights Watch itself documented the humanitarian harm of Iran’s sectoral sanctions in 2019. It continued to advocate for targeted Iranian sanctions in the same period. The two positions are not logically incompatible. But in practice, each round of targeted advocacy provides new political legitimacy for the broader sanctions infrastructure to remain in place.
The Fordham International Law Journal’s 2025 symposium on economic sanctions is instructive here. Alexandra Hofer’s contribution argues that the sustained violence of unilateral sectoral sanctions contradicts the self-image of the sanctioning states as “normative actors” committed to human rights (Hofer 2025, 1095). The same states that invoke human rights to justify sanctions are, on the best available epidemiological evidence, imposing a form of coercive harm that kills more than half a million people a year. The human rights framework is not protecting those people. It is providing the rhetorical architecture that makes their deaths politically palatable.
John Mueller and Karl Mueller identified this mechanism as far back as 1999, when they argued in Foreign Affairs that the sanctions then imposed on Iraq had produced a death toll larger than all weapons of mass destruction in history combined, and called for sanctions to be treated with the same humanitarian scrutiny applied to military action (Mueller and Mueller 1999, 49). The argument was made. It did not penetrate the dominant framing. The dominant framing is too useful to too many actors to be dislodged by data alone.
IV. The Energy Architecture Behind the Rhetoric
Understanding who benefits from sanctions requires following the energy market data rather than the press releases.
The three most comprehensively sanctioned states in the world are Iran, Russia, and Venezuela. These three countries together hold a substantial share of global proven hydrocarbon reserves. Iran sits on the world’s second-largest natural gas reserves and the fourth-largest proven crude reserves. Venezuela holds the world’s largest proven oil reserves. Russia is the world’s largest oil exporter, contributing roughly eleven per cent of global crude production. All three have been subject to escalating US sanctions over the past decade. All three have actively sought to operate their energy sectors outside the dollar-denominated trading system that underpins US financial hegemony.
The United States, during this same period, became the world’s largest oil producer and, in 2024, the world’s largest exporter of liquefied natural gas (US Energy Information Administration 2025). The Atlantic Council’s Energy Sanctions Dashboard notes that US sanctions on Iran were designed with explicit coordination with alternative oil suppliers to ensure that targeting Iranian oil exports would not trigger immediate spikes in global crude prices (Atlantic Council 2025). In plain terms: before restricting a competitor’s ability to sell oil, the sanctioning state secured alternative supply from its own producers and allies. The sanctions did not simply punish a government. They restructured the global energy market in the sanctioner’s favour.
Before restricting a competitor’s ability to sell oil, the sanctioning state secured alternative supply from its own producers and allies. The sanctions did not simply punish a government. They restructured the global energy market in the sanctioner’s favour.
The Venezuela case makes the mechanics visible in unusually clear detail. The Trump administration imposed comprehensive oil sanctions on Venezuela’s state company PDVSA in 2019, citing democratic backsliding under Nicolas Maduro. By 2022, with global energy prices rising after Russia’s invasion of Ukraine, the Biden administration quietly stopped enforcing Iranian oil sanctions to allow Iranian production to rise and stabilise global prices (Washington Post, October 2023). In October 2023, it formally eased Venezuela’s oil sanctions under the Barbados agreement, exchanging sanctions relief for electoral guarantees from Maduro. Venezuelan crude exports to the United States reached 250,000 barrels per day in January 2025, their highest level since the sanctions were imposed, while Chevron, which had maintained its Venezuelan investments throughout the sanctions period, increased production and exports (Columbia University CGEP 2025).
The sanctions had not been removed because the human rights situation improved. Venezuela’s 2024 elections were widely condemned as fraudulent by international observers. The sanctions were eased because the Biden administration needed to contain gasoline prices in an election year. When they were subsequently reimposed, Venezuela’s foreign minister called the measure a crime of aggression. He was using language that, according to the Rodriguez data, is medically accurate.
The pattern is consistent across cases. Sanctions on the energy sectors of targeted states are imposed when those states challenge dollar hegemony, nationalise foreign-owned assets, or align with geopolitical competitors of the United States. They are eased when the sanctioner needs the oil. The human rights justification is present throughout, as scaffolding for the policy’s political viability, but the human rights situation does not determine when sanctions are applied or removed. The energy market does.
Alexandra Hofer’s analysis of unilateral sectoral sanctions in the Fordham International Law Journal describes this as a contradiction at the heart of the normative project: states that impose economic coercion causing mass civilian mortality cannot, without circularity, justify that coercion in the name of protecting civilians (Hofer 2025, 1096). The contradiction is not rhetorical. It is structural, and it has been maintained for decades precisely because the institutions best positioned to name it, the major human rights organisations, the leading media, the policy institutes, are based in the countries imposing the sanctions and are funded, in significant part, by the foundations and governments of those same countries.
V. The Compounding Instrument: Aid Withdrawal
Economic sanctions are one mechanism by which populations in the Global South are being structurally impoverished. A second mechanism, distinct in its instrument and perpetrators but compounding in its effect, is now operating simultaneously: the withdrawal of Official Development Assistance by the world’s largest donor states.
These must be named separately. Sanctions actively degrade a target economy through financial isolation. Aid withdrawal removes the compensatory resource transfers that have been partially offsetting decades of structural disadvantage in the world’s most vulnerable countries. Both harm the same populations. Both are decisions made in Washington, London, Berlin, and Paris. And both are accelerating at the same time.
A study led by the Barcelona Institute for Global Health, published in The Lancet Global Health in February 2026, provides the most comprehensive available quantification of the mortality consequences of development defunding. Drawing on longitudinal panel data from 93 low- and middle-income countries across two decades, the researchers found that Official Development Assistance had contributed to a 39 per cent reduction in child mortality, a 70 per cent reduction in HIV/AIDS deaths, and a 56 per cent reduction in deaths from malaria and nutritional deficiencies across recipient countries between 2002 and 2021 (Rasella et al. 2026, e240).
Those gains are now at acute risk. In 2025, least-developed countries were projected to face a 13 to 25 per cent reduction in net bilateral assistance. Preliminary projections as of late 2025 indicated reductions by 2026, relative to 2023 levels, of 56 per cent for the United States, 39 per cent for the United Kingdom, 36 per cent for Germany, and 19 per cent for France (Rasella et al. 2026, e236). Under the study’s severe defunding scenario, projected additional mortality reaches 22.6 million deaths by 2030, including 5.4 million children under the age of five.
These are projections, not measurements. They carry the uncertainty that all modelling does. The direction of the finding is not contested. The same donor states that impose sanctions on countries whose governments they oppose are simultaneously withdrawing the development assistance that keeps people in unrelated countries alive. The populations of Iran, Sudan, and Venezuela live under sanctions. The populations of Mozambique, Mali, and Nepal live under aid withdrawal. Different policies, same political economy, same structural outcome: the transfer of resources from the Global South to the Global North, enforced through financial instruments rather than military ones.
VI. Reclassification as the Only Coherent Response
The argument that has been made in policy circles for thirty years, that if Western governments simply understood the mortality data they would moderate their use of economic coercion, has been tested by thirty years of evidence. The sanctions on Iraq in the 1990s produced documented mass child mortality. The studies were published. The policy continued. The studies on Iran, on Venezuela, on Sudan accumulated. The Rodriguez study in 2025 provided the most methodologically rigorous causal evidence yet. On the available evidence, the data does not drive policy change when the policy serves the interests of those receiving the data.
What the evidence does support is a legal argument that has been building in the international law literature. Unilateral sectoral sanctions that are deliberately designed to degrade the economic capacity of an entire state, and that produce mortality at the scale the Rodriguez study documents, satisfy the definitional criteria for prohibited coercive measures under international law. Joy Gordon, whose 1999 analysis of the ethics of economic sanctions remains foundational, described this as a “peaceful, silent, deadly remedy” and argued that sanctions imposed with knowledge of their humanitarian consequences are not morally distinguishable from the military force they are positioned as an alternative to (Gordon 1999, 123).
Alexandra Hofer’s 2025 analysis takes the argument further into positive law. Unilateral sectoral sanctions that the imposing states are aware will cause civilian mass mortality, and that are structured primarily to achieve geopolitical rather than humanitarian objectives, undermine the normative authority of the sanctioning state in ways that should trigger accountability under the same legal frameworks that govern the use of military force (Hofer 2025, 1095). The framework exists. The application of it has been blocked by the same structural interest that has kept the mortality data off the front pages.
The practical implication of that reclassification is not an abolition of sanctions as a policy tool. Targeted Magnitsky-style designations of named perpetrators of human rights abuses are a different instrument, and the Rodriguez study finds no mortality effect from UN-mandated multilateral sanctions. The implication is that unilateral sectoral sanctions must be subject to independent humanitarian impact assessment before imposition and at regular intervals thereafter, with binding authority to require modification or termination when the mortality evidence meets the threshold that would, in any other context, constitute a prohibited act of war.
VII. The Baseline Worth Reconsidering
Five hundred and sixty-four thousand, two hundred and fifty-eight. That is the annual figure the Rodriguez study puts on the death toll from unilateral economic sanctions. It is a more tentative figure than it appears: it is a statistical estimate with a confidence interval, produced by researchers who have disclosed their funding sources and their financial relationships with sanctioned countries. Take the most conservative end of the confidence interval: 367,838 deaths per year. That is still more than the average annual battle-related casualties from all the world’s armed conflicts combined over the same period.
A child in Tehran cannot get epilepsy medication because no bank in Europe will process the payment required to import it. A woman in Caracas with cancer cannot access chemotherapy because the pharmaceutical supply chain has been severed by secondary sanctions on the companies that would otherwise supply it. A man in Khartoum goes without antiretroviral drugs because the financial infrastructure that would transfer payment for them has been frozen. None of these deaths appear in the casualty counts that define our understanding of global violence. They are not counted as war deaths. They are not counted as sanctions deaths. They are counted, if at all, as excess mortality: a category so passive in its grammar that it contains no agent, no perpetrator, no decision.
None of these deaths appear in the casualty counts that define our understanding of global violence. They are counted, if at all, as excess mortality: a category so passive in its grammar that it contains no agent, no perpetrator, no decision.
Behind each of those deaths is a decision. The decision to impose sectoral sanctions was made in Washington or Brussels. The decision to coordinate with alternative oil suppliers before imposing those sanctions, ensuring that the energy market benefit to domestic producers was secured, was made in the US Treasury. The decision to lift or ease those sanctions when the domestic political calculation changed, with no reference to the humanitarian situation that had supposedly justified them in the first place, was made in the same building. The decision to cover the story as a human rights intervention rather than an energy market intervention was made in editorial meetings at newspapers whose governments imposed the sanctions. The decision to call for more targeted pressure rather than the abolition of the sectoral framework was made in the offices of human rights organisations based in the same cities as the governments writing the sanctions legislation.
Peace is the name we have given to a world in which those decisions continue to be made, and the deaths they produce continue to accumulate, invisibly, in the statistics that nobody reads aloud.
Bibliography
Amnesty International. 2017. “Myanmar: Security Forces Target Rohingya during Vicious Rakhine Scorched-Earth Campaign.” September 2. https://www.amnesty.org/en/latest/news/2017/09/myanmar-security-forces-target-rohingya/.
Atlantic Council. 2025. Energy Sanctions Dashboard. Washington, DC: Atlantic Council. https://www.atlanticcouncil.org/energy-sanctions-dashboard/.
Columbia University Centre on Global Energy Policy (CGEP). 2025. “The Impact of the New US Oil Tariffs on Venezuela.” April. https://www.energypolicy.columbia.edu/the-impact-of-the-new-us-oil-tariffs-on-venezuela/.
Gordon, Joy. 1999. “A Peaceful, Silent, Deadly Remedy: The Ethics of Economic Sanctions.” Ethics and International Affairs 13: 123-142.
Hofer, Alexandra. 2025. “Our Humanity at Stake: The Human Costs of Economic Sanctions for the Sanctioner.” Fordham International Law Journal 48, no. 5: 1087-1126.
Human Rights Watch. 2017. “Burma: UN Security Council Should Act on Rohingya Crisis.” September 13. https://www.hrw.org/news/2017/09/13/burma-un-security-council-should-act-rohingya-crisis.
Human Rights Watch. 2019. “‘Maximum Pressure’: US Economic Sanctions Harm Iranians’ Right to Health.” October 29. https://www.hrw.org/report/2019/10/29/maximum-pressure/us-economic-sanctions-harm-iranians-right-health.
Mueller, John, and Karl Mueller. 1999. “Sanctions of Mass Destruction.” Foreign Affairs 78, no. 3: 43-53.
Rasella, Davide, Andrea Ferreira da Silva, Davide Cavalcanti, et al. 2026. “The Impact of Two Decades of Humanitarian and Development Assistance and the Projected Mortality Consequences of Current Defunding to 2030: Retrospective Evaluation and Forecasting Analysis.” The Lancet Global Health 14, no. 3: e233-e249.
Rodriguez, Francisco, Silvio Rendon, and Mark Weisbrot. 2025. “Effects of International Sanctions on Age-Specific Mortality: A Cross-National Panel Data Analysis.” The Lancet Global Health 13, no. 8: e1358-e1366.
US Energy Information Administration. 2025. “The United States Remained the Largest Liquefied Natural Gas Exporter in 2024.” Today in Energy, March 27.
Washington Post. 2023. “In the World of Oil Sanctions, Venezuela Is No Iran.” October 19.