The United States wants to work with friendly and like-minded countries—so who are they?

On a host of issues ranging from green energy to security in the Asia/Pacific and cybersecurity, the Biden administration makes frequent mention of “friendly” and “like-minded” countries as preferred partners for international collaboration. The term “friend-shoring” was used by US Treasury Secretary Janet Yellen in April 2022 to describe re-orienting supply chains through countries “we know we can count on.” US Commerce Secretary Gina Raimondo has said “When you talk about friend-shoring, we have to find places that share our values, that play by the rules, that treat workers fairly, that respect the environment … those are friends that we’re talking about.”

But unlike “ally” or “free trade agreement partner,” terms like “friend” and “like-minded” have no established, agreed-upon definition—at least not in the domain of international affairs. This short piece outlines an approach to measuring friendliness and like-mindedness based on similarities in 1) political institutions and 2) voting patterns in the UN General Assembly.

Like obscenity, friendship and like-mindedness are hard to define but easy to recognize. Few Americans go to bed each night worried the nuclear-armed United Kingdom will launch a strike against Washington or that US Virginia-class nuclear submarines to be sold to Australia under the AUKUS partnership will wind up menacing the Port of Los Angeles. These scenarios read as too fanciful even for Hollywood. Why?

There are many reasons, ranging from power asymmetry—US forces would easily overwhelm Australian and UK defenses in response—and extensive trade and investment relationships to cultural similarity and shared liberal democratic values. The latter are probably more important than the former (power asymmetry). The United States doesn’t worry about Australian or UK aggression because it trusts those countries—they are governed (mostly) the same way and share similar visions of a liberal, rules-based international order. The political institutions and cultures that shape Australian, UK, and US national interests are more alike than different, so Americans broadly trust that Australia and the United Kingdom will use their capabilities in ways consistent with or even furthering of US interests. In common parlance, Australia and the United Kingdom are friendly and like-minded countries.

This like-seeks-like perspective emphasizes two related factors: 1) similarity of political institutions, or the rules that structure collective decision-making, govern competitive office-seeking, and place limits on the arbitrary use of state authority; and 2) similarity of perspectives on global affairs. The first factor is consistent with the literature on the democratic (and more recently “dictatorial”) peace, which demonstrates that pairs of democratic countries and specific types of authoritarian regimes (personalist dictators and military regimes) are substantially less likely to go to war with one another, even accounting for possible confounders like trading relationships and military power asymmetries. The second factor is consistent with the extensive use of state voting behavior in fora like the UN General Assembly (UNGA) as a proxy for interest alignment in global affairs: Countries that vote alike in the UNGA are inferred to have aligned interests.[1]

Figure 1 plots countries based on the absolute distance between a state’s polyarchy score and the state’s ideal point—the estimate of the state's UNGA voting position, a dimension that reflects a state's position toward the US-led liberal order—averaged over the three most recent years for which data are available (2020–22). The origin represents the US position during this time period. Building on Robert Dahl’s classic treatise on the United States, polyarchy is characterized by democratic institutions that invest power in and facilitate the participation of many social actors—civil society, business, political parties, etc.—and places limits on the exercise of government authority. Members of the North Atlantic Treaty Organization (NATO) are in red, other states in blue. Data are available for 171 countries.

Clearly, the two variables are moderately correlated (r = 0.66): Countries that are more democratic (or polyarchic, specifically) tend to be more aligned with the United States in the UNGA and vice versa. But they are not one and the same. On the polyarchy dimension, Argentina and Jamaica are very similar to the United States but are middle-of-the-pack when it comes to revealed preferences in the UNGA. Similarly, Hungary may be a NATO ally and vote like Australia—widely perceived as a strong US ally—but under Viktor Orbán, it has become more authoritarian and hasn’t governed itself like Australia.

These two dimensions can be further collapsed using factor analysis[2] to yield a single measure of friendliness and like-mindedness with the United States. Based on that measure, the 10 most and least friendly countries are listed in table 1.

On the face of it, the list of most friendly countries, with Israel at the top, is plausible. Presidents George W. Bush, Barack Obama, Donald Trump,[3] and Joseph R. Biden Jr. have all said some variant of “Israel has no better friend than the United States”—and the data would suggest the feeling is mutual. This is in part because the Israeli-Palestinian conflict is a major issue on the UNGA agenda and the United States is a staunch supporter of Israel in that forum: The United States and Israel standing together on this issue means standing apart from the rest of the United States’ closest partners. The top 10 includes also AUKUS partners: the United Kingdom (second place, a “special relationship” indeed) and Australia (sixth). Other than Australia and Israel, the top 10 is composed exclusively of NATO members. Among NATO members, the outlier is Turkey (not on this list), which is only slightly more friendly and like-minded than Serbia.

The bottom of the list is more controversial. It includes usual suspects like Iran, post–Hugo Chavez Venezuela, Cuba, and China—the most oft-mentioned country in the US Department of National Intelligence’s 2024 Annual Threat Assessment. But it also includes security partners like Egypt and Saudi Arabia—the latter of which is the United States’ largest foreign arms purchaser.

The adage “the enemy of my enemy is my friend” dates to at least the 4th century BCE, and it helps explain the depth of the US-Saudi strategic partnership, with the common enemy being Iran, as does Saudi Arabia’s pillar status in the global oil market. The US-Egypt relationship has several sources, but the most obvious is Egypt’s willingness to recognize and respect Israel’s sovereignty, as brokered at the 1978 Camp David Accords. But the analysis presented here helps more accurately categorize these relationships as strategic partnerships based on mutual interests rather than stemming from friendship or like-mindedness.

Of course, it behooves US presidents and diplomats to extend rhetorical olive branches and handshakes to a variety of trade, investment, and security partners, especially when hosting those partners for dinner. National interest, in the United States and elsewhere, often requires working and trading with partners that do not share “our” values. But if you accept the premise that state-to-state friendship and like-mindedness is premised in similar expressed preferences and ways of governing themselves, then the United States' closest friends are Israel, Australia, and its democratic allies in NATO.

Notes

1. Subject to some caveats; these caveats are what led the curators of the most widely used data on UNGA voting to focus less on agreement rates and more on using voting patterns to calculate ideal points, a technique used to infer underlying preferences from expressed voting behavior, pioneered by scholars studying the US Congress. See Erik Voeten, “Data and analysis of voting in the United Nations General Assembly,” in Routledge Handbook of International Organization (Bob Reinalda, ed.). Oxfordshire, UK: Routledge, pp. 54-66.

2. Factor analysis is a statistical method used to reduce some number of variables to fewer underlying “factors,” or dimensions that explain a large amount of the variation contained in the data. In cases where two variables are moderately or strongly correlated, the underlying factor will be strongly correlated with both. In this case, the underlying factor is correlated with both observed variables at r = 0.74.

3. While Donald Trump made this statement recently, it is harder to source evidence of him having said it during his presidency. This sentiment was expressed by Trump’s Secretary of State Mike Pompeo in 2019.