Picture a manufacturer deciding where to build a new factory abroad. A generation ago, the math was mostly about costs: cheap labor, accessible ports, low taxes. Today, executives are weighing something harder to put on a spreadsheet. Is the host country a political friend or a potential adversary? A new study suggests that this kind of thinking, often called “friendshoring,” has become a stronger force in global investment than it was a decade ago, though not everywhere and not for the reasons many assume.
The paper, published in the Journal of Comparative Economics, examines how political distance between countries shapes the flow of foreign direct investment, or FDI, which refers to companies building, buying, or expanding operations across borders. The authors find that political differences weigh on these decisions more heavily now than at any point since 2003.
The question behind the headlines
The recent years of trade tension, war, and diplomatic friction have produced plenty of talk about companies pulling production out of rival nations and relocating it to allies. But talk is not data. Arti Grover of the World Bank Group and Pierre-Louis Vézina of King’s College London set out to test whether the friendshoring story holds up across many countries, many sectors, and several different ways of measuring how “friendly” two nations are.
They also wanted to settle a debate within the research community. Some earlier work suggested that politics shaped investment most powerfully during the Cold War or in the early 2000s, raising the question of whether friendshoring is genuinely a new development or just an old pattern rebranded. To answer this, Grover and Vézina tracked the influence of political distance year by year rather than averaging it across long stretches of time.
Four ways to measure friendship
Measuring something as slippery as geopolitical alignment is not straightforward, so the researchers used four separate yardsticks. The first is how similarly two countries vote at the United Nations General Assembly, a widely used proxy for shared diplomatic positions. The second is public opinion, drawn from Pew Research Center surveys that ask people in one country how favorably they view another. The third measures differences in democratic governance using the liberal democracy index from the Varieties of Democracy project, which scores countries on voting rights, fair elections, civil liberties, and limits on executive power. The fourth sorts countries into five geopolitical blocs defined by Capital Economics, ranging from “US allies” to “China allies,” and measures the distance between them.
The authors explain that these measures capture different channels. UN voting reflects official foreign policy, public opinion captures popular sentiment that can shape the investment climate, the democracy index reflects trust in institutions and contract enforcement, and bloc alignment combines several signals into broad groupings. Using all four lets them check whether any finding holds up regardless of how political distance is defined.
On the investment side, they pulled together three types of data. Greenfield projects, which involve building new facilities from scratch, came from fDi Markets, a Financial Times subsidiary. Mergers and acquisitions, which involve buying existing assets, came from Refinitiv Eikon. And the accumulated stock of foreign affiliates, which reflects long-standing corporate presence, came from a multinational enterprise database. Each captures a different speed and time horizon: new factories take years to plan and are hard to reverse, acquisitions can happen quickly, and affiliate stocks shift only gradually.
How they crunched the numbers
To analyze all this, the researchers used a gravity model, a workhorse tool borrowed from the study of trade. The basic idea is that flows between two countries tend to shrink as the friction between them grows, like geographic distance, and tend to grow with shared traits, like a common language. Grover and Vézina added their measures of political distance to this framework and ran the model separately for each year from 2003 to 2022, controlling for things like distance, shared borders, common language, and colonial history. Their largest dataset covered 190 countries.
The results point in a consistent direction. Across all four measures, political distance has a larger negative association with investment today than it did ten years ago, with the sharpest effects appearing from 2018 onward. The impact of UN voting differences on new greenfield projects roughly doubled over the decade: a one standard deviation drop in voting similarity was linked to an 8 percent decline in greenfield investment in 2011, but a 16 percent decline by 2022. The public opinion measure showed an even steeper change, with an associated drop of 11 percent in 2011 ballooning to 38 percent in 2022. Being in a different geopolitical bloc was linked to roughly half as large a decline in 2011 as in 2021.
One pattern stands out in how investment responds. New projects and acquisitions have grown more sensitive to political distance, but the stock of existing affiliates has stayed relatively stable. The authors read this as a sign that companies are hesitating to make fresh commitments in politically distant places rather than dismantling operations they already have. In their words, there is “a lack of evidence of reshoring, while new projects are now more sensitive to geopolitical distance.”
A tale of two investors
The friendshoring story, it turns out, is mostly a Western one. When the researchers split investment by country of origin, they found that companies from the United States, Canada, and the United Kingdom were among the most sensitive to political distance, steering away from countries that vote differently at the UN. Firms from China, India, Singapore, Japan, and the United Arab Emirates showed the opposite tendency, investing more in politically distant places, not less.
The contrast between the United States and China was especially sharp. China fell from the second-largest destination for US greenfield investment in 2012 to fifteenth by 2022. Yet Chinese outward investment showed no sign of retreating from the United States, which remained China’s top destination. Chinese firms even increased their investment in countries closely aligned with the United States, such as Vietnam, Mexico, and Indonesia. The authors interpret this as evidence that fully untangling the two economies may be difficult, since the United States could stay indirectly connected to China through shared partners.
Not just about national security
Much of the policy conversation around friendshoring frames it as a matter of national security, protecting sensitive industries like semiconductors, aerospace, and defense. The researchers tested this by comparing strategic sectors against sectors deeply tied into global supply chains and sectors that rely on complex, relationship-specific contracts.
They found that the pull of political distance was roughly the same across all these groups, and close to the average across every sector. The authors argue that this offers evidence against the idea that friendshoring is driven purely by security or resilience concerns. If the trend reached only strategic industries, the security explanation would fit neatly. Because it spans the broader economy, they suggest something wider is at work.
What it may mean
The authors note that governments are increasingly nudging firms toward allies through policies like the US Inflation Reduction Act and the CHIPS Act in both the United States and Europe. They caution that if companies keep concentrating investment among politically aligned partners, the result could be hardening economic blocs that deepen geopolitical tension. Their analysis points to a particular risk for developing countries that have relied on Western investment for jobs, technology, and growth.
A few caveats are worth keeping in mind. The study documents associations rather than proving that political distance directly causes investment to fall, and its measures of friendship are imperfect proxies for a complicated reality. The public opinion data, in particular, covered a much smaller set of countries than the UN voting data. Still, the consistency of the pattern across four measures and three types of investment gives the central finding some weight: politics is shaping where the world’s capital goes more than it did before, even if the map of who is friendshoring looks different from the popular telling.
