Last November, Nicholas Mulder sparked a debate about the place of sanctions in the toolbox for progressive foreign policy, writing in The Nation that progressives must “move beyond the dominant consensus on how to deal with foreign policy problems as framed by the establishment, in which there are only two flavors: the mild option of sanctions and the radical option of war, neither of which works particularly well.” The piece drew responses, notably from Neil Bhatiya on this blog and from Daniel Drezner in the Washington Post, both of whom made the case for sanctions as a flavor of policy worth preserving. The conversation continues here, with Mulder taking on the question of how we think about success in sanctions policy.

By Nicholas Mulder

If progressivism aims to create a better world, it should start by abandoning the tools of collective punishment. As I argued in The Nation, a whole suite of tools exist for directly going after specific individuals and companies who break or evade domestic and international laws. It is in this domain–by imposing sanctions on tax evaders, for example–that progressive foreign policy can regain some of the legitimacy that has been weakened by decades of excessive US interventionism and the Trump Administration’s antics.

Earlier this month Neil Bhatiya made a counter-case for economic sanctions as a tool of progressive foreign policy, arguing that “any measure that widens the distance between peace and war should be in the foreign policy toolbox.” If sanctions can effectively accomplish the same goals as military force, then they could be seen as a tool that forestalls war. Bhatiya and Drezner both argue that sanctions, if properly applied, can do just that: deter some state actions and compel others. Recent sanctions successes, they claim, justify keeping broad-based sanctions in the progressive foreign policy toolbox. The historical record, however, is not nearly that clear.

There are several reasons why the dominant discourse about sanctions is in need of serious revision. First, sanctions advocates usually fail to make a plausible case that war is really the only alternative to sanctions. Their reasoning often depends on mere assertion or unsupported counterfactuals. Second, many claims of policy successes through sanctions are empirically uncompelling–the process by which sanctions purportedly cause policy changes is murky at best. We need history and qualitative regional and area expertise, as well as a truly international viewpoint, to assess the record of sanctions. Last, given the moral risk of collateral damage that punitive economic measures entail, a high burden of proof that sanctions will achieve their stated aims is the least we can demand from policymakers.

Counterfactuals: What, exactly, did Iran sanctions deter?

If the Iraq sanctions from 1990 to 2003 failed because they were followed by war, then the threshold for the success of the Iran sanctions of the 2010s is presumably that they avoided war. Bhatiya claims that “war with Iran would have made the catastrophe of Iraq look humdrum by comparison.” He implies that in the absence of sanctions, a war between Iran and one of its neighbors or the United States was likely. What is the evidence for this? It is difficult to find any based on what we know or could observe about Iranian behavior. When the Obama Administration began its economic pressure campaign against Iran in 2009-2010, it narrowly wanted to counter Iran’s nuclear program. Economic pressure was a means to achieve an objective that the US had set for itself—preventing any Iranian nuclear capacity—rather than a way to stave off a likely hot war.

There is concrete evidence that Iran was prepared to talk even before sanctions were ramped up. Iran twice offered to seriously cooperate with the United States, first in the early 1990s when it invited US hydrocarbon firms back into the country; and then in 2003, when it offered ending support for Hezbollah and Hamas and helping Washington stabilize Iraq. Both times no sanctions were needed to produce an open door to negotiations and cooperation, and both times the United States rejected the offers.

All of this suggests that US sanctions on Iran are a recurring feature of American relations with Tehran rather than a pragmatic response to some imminent security threat. This week, the Islamic Republic celebrates the 40th anniversary of the revolution which created it. In that founding year, 1979, the Carter Administration first froze Iranian assets in response to the US embassy siege. Iran has been under some form of economic sanction from the US for four decades. Many people in Washington will argue that this is necessary due to Iranian behavior, but to the rest of the world, it appears that it is the United States which suffers from a peculiar Iran problem.

Despite US sanctions, the Islamic Republic has integrated itself into many of the networks of modern globalization, and the country has stable relations with many Asian, Latin American, African, and European countries. This is why there was, and is, such broad international support for upholding the Iran deal and the relaxation of sanctions that came with it.

Successes that aren’t

Sanctions advocates are quick to point out that sanctions only ever succeed in combination with other tools. If, however, sanctions only succeed as part of a bundle of carrots and sticks, that complicates claims that they are the cause of policy success. It is difficult to know if, in any given bundle, sanctions are a necessary ingredient for success. Statistical analyses often correlate sanctions use with positive outcomes, without showing qualitatively that the sanctions caused those outcomes. An example is the CNAS study “The New Tools of Economic Warfare” that Drezner co-authored with Elizabeth Rosenberg, Julia Solomon-Strauss, and Zachary Goldman. Drezner cited this study in his Washington Post article to counter my claim that sanctions have been failing much more than succeeding; the study claimed that sanctions had become more successful in the twenty-first century, citing “nine successful outcomes out of the 22 sanctions cases”—a 40.9% success rate.

The study’s technique for judging recent cases ‘successful’ or ‘unsuccessful’ was to poll thirty unnamed sanctions specialists. In other words, the CNAS test of success was whether other think tank experts found sanctions to have worked. This is a problematic way of using social science to inform policy. Instead of assessing individual cases based on qualitative and quantitative evidence, the CNAS report reflected discourse and received opinion among policymakers. It reproduced prevailing attitudes instead of critically interrogating them.

A closer look at the cases in the report’s appendix shows the insufficiency of its assessments. Nine cases in the report were labelled successes: the restoration of democracy in the Central African Republic (2003-2005) and in Guinea-Bissau (2004); the response to Ivory Coast’s violation of a ceasefire agreement in 2004; punishment of the 2009 military coup in Honduras; making Iran agree to the Joint Comprehensive Plan of Action (JCPOA); stopping Libya’s armed suppression of protests in 2011; fostering democratic elections and the release of Aung San Suu Kyi in Myanmar; convincing Nigeria to extradite Liberian warlord Charles Taylor; and punishing the refusal of the Uzbek government to allow an international investigation into the 2005 Andijan massacre, in which hundreds of civilians were killed.

Which of these very different events can actually be ascribed to sanctions? Arguably, economic pressure helped to achieve the Iran deal, but only in combination with bona fide negotiations and military threats. Perhaps they also helped secure Taylor’s extradition and the stabilization of Ivory Coast, where an arms embargo and asset freezes were used from 2004 to 2016. Beyond those two cases, many of the other supposed successes are post hoc ergo propter hoc claims, which simply claim success because a good outcome followed the imposition of sanctions. What is the evidence, for example, that sanctions aided the restoration of democracy in the Central African Republic, where François Bozizé, who committed a coup in 2003, saw sanctions against his government lifted as soon as he was democratically elected in 2005? Did US measures have any appreciable effect on outcomes on the ground? Bozizé convened a National Transitional Council soon after his coup, issuing in a new constitution two years later. Were sanctions necessary for this or was he seeking to shore up his rule democratically in any case? What we do know is that since 2005, civil war and human rights abuses in the Central African Republic have continued, even prompting renewed UN Security Council and American sanctions against Bozizé in 2014. Even if it was a success–itself a highly questionable claim in need of substantiation–then sanctions have not durably improved the CAR’s political situation.

The same difficulty of discerning whether sanctions changed anything about domestic political dynamics applies to Myanmar: Aung San Suu Kyi was released from house arrest in 2010 and entered government in 2016. Yet since then, her acquiescence in the ongoing ethnic cleansing of the Muslim Rohinga has placed the circumstances of her release in a different light: did the generals allow her into power because they realized she would not fundamentally contest their control?

Uzbekistan, under US sanctions from 2005 to 2009, saw even less change: a decade after the bloody Andijan massacre, Islam Karimov remained in power, and no international investigation took place. So why does the CNAS study classify the sanctions against Tashkent a success?

Claims for sanctions success in this context are really assertions that the West did something in the name of a good cause, but they do not amount to convincing cases that long-distance economic pressure changed the course of history in other countries.

The case of Honduras in 2009 is even more problematic. The military’s coup against elected president Manuel Zelaya prompted widespread condemnation and sanctions, but the US then lifted these restrictions after it judged an election reached in November 2009 to be sufficient progress, even though this plebiscite left the usurper presidency of Lobo Sosa in power and no other Latin American government recognized the results. So sanctions were ‘successful’ only in expressing superficial US disapproval of a coup for five months and then letting the plotters get away with it. This is no success, least of all for Latin American democracy.

At a time when the Trump Administration is imposing major economic sanctions against Maduro’s government in Venezuela (running the risk of a repetition of the 1990s Iraq fiasco) we need to be realistic and honest about their generally weak ability to spread democracy.

Then there are the competing narratives of the effect of sanctions on Libya. The CNAS report considers sanctions to have been successful in stopping the Gaddafi government’s repression of protest—a strange claim given that it was clearly the civil war and NATO intervention that unseated the Libyan dictator. Bhatiya, by contrast, argues that ‘the 2011 overthrow of Qaddafi undid the positive effects of the US sanctions effort’, which he claims forced the surrender of his weapons of mass destruction in 2003. This, too, is not borne out by the evidence. In his assessment of this episode, Libya expert Ronald Bruce St John credits ‘a policy of engagement, supported by persistent, patient diplomacy,’ with success in re-establishing relations with Gaddafi. St John draws attention to the longstanding preparedness of the Libyans to surrender WMDs in return for normalizing relations. They first made such an offer in 1992 (before major sanctions) and repeated in 1999 (after the US imposed sanctions) and then in March 2003, as the Bush Administration was invading Iraq. So whose behavior had to change for a deal to happen?

Like the JCPOA with Iran in 2015, the rapprochement with Gaddafi in 2003 was achieved not by the application of sanctions to coerce an otherwise recalcitrant target into agreement. Instead, the process was one of cycling through various rounds of ineffective economic pressure before breaking the domestic American political taboo on engaging seriously with weaker adversaries that had shown themselves willing to strike a deal several times before. Achieving better outcomes after sanctions may reflect the US political elite’s struggle to escape the myopia of rogue state doctrine—a worldview that has colored foreign policy since the late Reagan era—more than the crumbling of autocracy abroad.

The moral stakes of the evidence

Both the Libyan and Iranian cases raise the question of whether the use of sanctions was even necessary to begin with. Sanctions always run the danger of causing collateral damage. But what this means is that evaluating them as a policy must also be morally conscious. This is why the weak empirical basis of the CNAS report is a problem. If a policy instrument is morally fraught because it has the potential to severely affect the lives of innocent people, we should require higher evidentiary standards than usual to evaluate its success. But since the advocates of sanctions often invoke war as the only real alternative, they downplay these consequences.

At the very least, we should demand clear evidence that sanctions are more likely than not to succeed. Since there is little such evidence, and the history of sanctions is mostly the history of failure interspersed with a few limited successes, I believe that history, empirical utility and morality point in the same direction: we should demote sanctions in our toolbox, at least as ways to put pressure on entire countries or on governments able to devolve their negative consequences onto civilians.

---

Nicholas Mulder is a doctoral candidate in history at Columbia University working on the origins of economic sanctions in the first half of the twentieth century.