That Russia interfered in the 2016 election on behalf of the Donald J. Trump campaign is no longer disputed. To varying degrees, Trump administration officials and Democratic presidential candidates have warned against such interference in 2020. For his part, President Trump has downplayed both the effect of Russian interference and its impropriety, telling Russian officials in a 2017 meeting that the United States interfered in elections in other countries.
Many on the left also hold this view, arguing that turnabout is fair play given past interference by the U.S. intelligence community in foreign elections. In October 2019, then Democratic presidential candidate Andrew Yang said in a debate that he would tell Russia, “We’ve tampered with other elections. You’ve tampered with our elections. And now it has to stop.”
Despite these claims, recent evidence of covert U.S. interference in foreign elections is scant. The Snowden leaks of classified data have revealed extensive details about the activities of the intelligence community but did not reveal any programs for election interference. Moreover, during the eight years of the Barack Obama administration, it is difficult to identify democratic elections in which the U.S. government could have covertly interfered and succeeded in changing the outcome. The global slide toward authoritarianism has started at the ballot box.
If the U.S. intelligence community is in fact interfering in elections, it should stop. Interfering in democratic processes undermines lasting interests in democracy promotion and weakens America’s soft power. While some could question the value of claiming the moral high ground on the international stage, from a realist perspective, banning covert election interference is in the United States’ interest because it is one of the countries that is more susceptible to it.
Given the relative low cost of carrying out clandestine election interference in the digital age, most countries could muster the resources to conduct such campaigns at multiple levels, from municipal to congressional, if they deemed it in their interest. Thus, the United States should work to create a global norm against covert election interference by unilaterally banning the U.S. intelligence community from such activity, enshrining global respect for democratic processes. It should then assemble a coalition of democratic nations that pledge not to covertly interfere in elections, which would better position the United States to pressure Russia and other nondemocratic states to cease their interference activities or face sanctions by coalition members.
Background: The History of U.S. Election Interference
During the Cold War, the U.S. government routinely interfered in foreign elections with the goal of preventing the spread of communism through the ballot box. Dov H. Levin, a professor at the University of Hong Kong, has identified at least eighty-one cases of U.S. interference in foreign elections between 1946 and 2000, including fifty-three cases involving covert interference. While Levin concludes that electoral interventions can “significantly increase the electoral chances of the supported candidate or party,” by an average of 3 percent. Contrary to popular wisdom, Levin’s model shows that overt interventions are typically more effective than covert interventions. Levin’s model suggests that overt interference, on average, increased the vote share for the supported candidate by 5.4 percent, compared to 2.2 percent for covert interference.
Looking at Levin’s data, instances in which the increase in vote share caused by a covert intervention could have changed the outcome of the election are difficult to identify. Out of the fifty-three known incidents of covert U.S. election interference, in only four cases did the U.S. government provide covert support to a successful candidate in an election in which the ultimate outcome was a vote difference of less than 5 percent (Malaysia in 1959, Peru in 1962, Panama in 1984, and Bulgaria in 1991).
Following the end of the Cold War, U.S. activity in this area appears to have subsided, if not disappeared altogether. Skeptics point to the Clinton administration’s support for the reelection of Boris Yeltsin in 1996 and efforts to defeat Slobodan Milosevic in 2000. The George W. Bush administration also scuttled plans for clandestine efforts to change the outcome of the first democratic election in Iraq. During the Obama administration, Hamid Karzai was reportedly angered by the efforts of Richard Holbrooke, U.S. special representative for Afghanistan and Pakistan, to support the opposition running against him.
As Peter Beinart notes in the Atlantic, most post–Cold War election interference by the United States has been overt not covert, including supporting democratic processes and aiding governments in the hopes of supporting their reelection. As a global power, the United States has interests in the outcomes of elections in other countries and has a variety of tools it can use to influence those outcomes. Through the National Endowment for Democracy and other organizations, the U.S. government supports efforts to ensure that elections are free and fair. Through trade policy and foreign aid, it can help a friendly government gain and maintain popular support. What it should not do is attempt to alter votes, manipulate foreign media, or attempt to manipulate voters through social media.
The Challenge: Providing Clarity on U.S. Policy
Countering the narrative that the United States covertly interferes in elections and creating a global norm against such interference requires the U.S. government to explicitly ban it. While the United States is party to a number of international instruments that directly or indirectly prohibit international interference, none of them have required the United States to modify its own laws to prohibit the activity. For instance, while the charter of the Organization of American States specifically prohibits intervention in any form, since the charter has been in effect, the United States is known to have covertly interfered in seventeen elections, fully one-third of all known cases of U.S. covert interventions.
On a similar challenge, combatting Chinese economic espionage, the Obama administration made public a long-standing policy barring the intelligence community from conducting industrial espionage to benefit U.S. companies. The policy was formalized in Presidential Policy Directive 28, in order to help create a global norm against the kind of economic espionage that the Chinese intelligence community was flagrantly committing. This unilateral declaration led to a 2015 agreement with the Chinese government to halt the practice, which held until the breakdown of trade relations during the Trump administration.
On political assassinations, the U.S. government has also taken unilateral action in order to shape global norms. First introduced in an executive order by President Ford, the current ban is contained in a Ronald Reagan–era executive order (EO 12333) on “United States Intelligence Activities” that includes an explicit ban on the U.S. intelligence community carrying out assassinations. That executive order also provides many of the authorities that govern the intelligence community today. Amending EO 12333 would therefore be the most expeditious way to ban the U.S. intelligence community from participating in foreign election interference.
However, clarifying what kind of activities should be considered permissible and what the United States is trying to create a norm against is not simple.
On one end of the spectrum, the ban could be so narrow as to allow any activity that falls short of actually changing votes. On the other end, authoritarian regimes would favor a policy of total noninterference—no democracy promotion, no public commentary, no aid to opposition groups. However, a broader policy of noninterference in foreign affairs is not a realistic or desirable position for the U.S. government to take. Overt election interference may be as soft as President Obama’s statement on the Scottish referendum on independence or as hard as Vice President Joe Biden’s threat to end military aid to Lebanon if Hezbollah-backed parties took power. In both cases, the interference, at a minimum, did not backfire as the supported side came out on top.
As a world leader, the United States believes there are circumstances in which helping an allied government win reelection or supporting opposition groups in their efforts to oust an unfavorable leader is in its interest. Particularly given Russia’s broad efforts to interfere in elections as well as Iran’s growing efforts, the United States will, if nothing else, want to combat these efforts by supporting the other side. In those cases, overt actions are both in keeping with democratic principles and more effective than covert interference.
There is some danger that the global rise in authoritarianism may provide an argument in favor of restarting election interference. The “one person, one vote, one time” phenomenon regularly occurs with authoritarian leaders initially elected in a free and fair election who then quickly move to consolidate power, eliminate or constrain the opposition, and then conduct sham elections henceforward. Yet, combatting undemocratic regimes with actions that run counter to U.S. democratic values is likely to do more harm than good in the long run. And while the United States had little to fear from reciprocation in the Cold War, today the internet allows almost any adversary of the United States to interfere. Thus, if clandestine interference in foreign elections was ever in the national interest, forgoing it as a tool of foreign policy in order to create a strong norm against it is in the U.S. interest today.
Based on the history of the assassination ban contained in EO 12333, the ban on clandestine election interference will need to be constructed carefully to avoid ambiguity about the president’s intention. Following 9/11, the U.S. intelligence community and the U.S. military carried out targeted drone strikes and other operations to eliminate the leadership of al-Qaeda and other terrorist organizations. Because EO 12333 provides no definition of the term “assassination,” government legal advisors concluded that the prohibition on assassination only applied to heads of state or other foreign government officials. Government lawyers have also concluded that assassination requires the act to be political in nature and that different standards apply in wartime and peacetime. Thus, the killing of Major General Qasem Soleimani would not qualify as an assassination under EO 12333. Based on this experience, it would be wise to provide a definition for clandestine election interference in the revised executive order to avoid ambiguity on what constitutes such interference.
U.S. interests lie in creating a strong norm against clandestine election interference. The clearest path to the creation of that norm is for the United States to, of its own accord, prohibit the intelligence community from participating in such activity. Critics will contend that an explicit presidential ban can be undone with the stroke of a pen and that the U.S. government has many other channels to influence the outcome of foreign elections. Others will argue against restraining from election interference when deemed in the national interest. Yet, in an era in which election interference tools are not held in a Cold War duopoly but are globally available, creating a strong norm against clandestine interference in democratic processes is in the national security interest of the United States. The best way to do that is to begin with a unilateral ban on such activity with the goal of claiming the moral high ground and combating the perception among U.S. government leaders and in the media that the United States does conduct such activity. Then, the United States can begin to press other democracies to make such commitments. With a norm against clandestine interference accepted by other democracies, the United States will be better positioned to create a coalition to punish Russia and other nondemocratic states when they interfere in democratic processes.
The president should amend EO 12333 to expressly prohibit the U.S. intelligence community from interfering in foreign elections. Similar to the prohibition on assassination contained in the executive order, the prohibition can be short and clear; however, based on the lessons from the assassination ban, the amendment should include direction to the director of national intelligence to promulgate more specific guidance on the subject and to define a process for adjudicating activities in gray areas.
The president should issue a national strategy on democracy promotion. Issued to accompany the release of the revised executive order, the national strategy should lay out the U.S. government’s position on providing support to democratic processes. The strategy should make clear that the U.S. intelligence community does not conduct clandestine activities intended to alter the outcome of democratic elections. It should set out policy for how the U.S. government provides support to build democratic processes and the programs through which this activity takes place. It should also make clear that it will overtly intervene to combat efforts by Russia, Iran, or other authoritarian regimes to undermine democratic processes abroad.
Congress should pass legislation banning the intelligence community from expending any funds on election interference. While the president’s executive order has the weight of law, it can be undone with the stroke of the president’s pen. There are also open questions about whether a president could secretly contravene an executive order through a finding or other presidential memorandum. Therefore, a congressional prohibition on the expenditure of funds will provide stronger protections.
The State Department should begin talks with Brazil, India, Indonesia, European allies, and other democracies on agreeing to a norm against covert election interference. Agreement could be obtained through unilateral declarations by these countries or through a more formal process of attachment to existing treaty instruments such as the Budapest Convention on Cybercrime. Creating this coalition would lay the groundwork for sanctioning countries that continue to covertly interfere in elections.
With a strong coalition in place, the United States and its allies should open discussions with Russia on its election interference. While the Russian government may not be an honest negotiating partner, building a norm against covert election interference will require a direct effort to curb Russian activity. As with the agreement with China on economic espionage, the United States and allies could need to agree to abstain from covert election interference even if they are already not doing so in order to allow the Russian government sufficient cover to present any agreement to its citizens as a triumph for Russia.
This Cyber Brief is part of the Digital and Cyberspace Policy program. The Council on Foreign Relations takes no institutional positions on policy issues and has no affiliation with the U.S. government. All views expressed in its publications and on its website are the sole responsibility of the author or authors.