Language policy is not merely symbolic – it shapes the direction of trade, the allocation of subsidies, and a country’s internal cohesion. Using new data on linguistic diversity across and within the US and Canada, this column examines how language affects trade patterns and economic welfare. The authors find that enforcing a single language without supporting bilingualism risks fragmenting the very markets such policies aim to unify. Economic gains arise not from linguistic uniformity, but from enabling shared fluency. A more bilingual society indicates a more integrated, productive, and resilient economy.
Across the world, governments continue to revisit the role of language in national identity and cohesion. Moves to designate or reinforce a national language may seem symbolic, but symbols matter – especially when they embed structural incentives and frictions. A century ago, US President Theodore Roosevelt made a similar declaration: “We only have room for but one language here and that is the English language, for we intend to see that the crucible turns our people out as Americans, of American nationality, and not as dwellers in a polyglot boarding-house” (Roosevelt 1919).
History tells a more layered story. As Lozano (2018) shows, Spanish was once a language of governance in the American Southwest, embedded in civic institutions, not confined to immigrant households. Its marginalisation was not cultural drift; it was policy. Recent efforts to formalise one language as the national standard extend that longstanding trajectory – now with modern economic consequences.
Such measures are often justified as promoting cohesion and streamlining communication. But cohesion without inclusion is an illusion. Unless accompanied by robust bilingual education, they risk deepening what Ginsburgh and Weber (2005, 2025) call linguistic disenfranchisement: the systematic exclusion of minorities from public life and economic opportunity.
This column shows that language policy does not just shape identity; it also reallocates trade, alters domestic integration, and affects national income. Using new data on language proximity (Gurevich et al. 2025a) and a structural general equilibrium model for both Canada (Gurevich et al. 2025b) and the US (Gurevich et al. 2025c), we find that bilingualism delivers strong economic gains. These gains arise whether majority-language speakers acquire minority languages or the reverse, but are largest when speakers of minority languages become bilingual. In contrast, increasing linguistic segregation leads to sharp welfare losses. Policies that prioritise a single official language may aim to foster unity, but without support for bilingualism, they risk undermining integration. Effective economic inclusion requires fluency on both sides of the linguistic divide.
Mapping the language landscape
Language is not just a marker of identity. It is economic infrastructure that governs how easily people communicate, how efficiently firms operate across space, and how goods flow within and between countries. Ignoring it means overlooking a key source of frictions in both domestic and international trade.
To quantify these effects, we developed the Domestic and International Common Language (DICL) database (Gurevich et al. 2025a). Drawing on detailed linguistic trees and data from over 6,500 native language communities worldwide, we construct eight language indexes of which the following are central:
- The Domestic Common Language (DCL) index measures the degree of linguistic overlap within a country.
- The International Common Language (ICL) index captures the proximity between languages spoken across countries.
These measures allow us to go beyond the binary shared-language indicators standardly used in the literature by accounting for the closeness of languages on a continuous scale. In our gravity model for the US, we find that a one standard deviation increase in DCL raises domestic trade by 17.9%, a magnitude comparable to major reductions in internal transport costs. For international trade, a one standard deviation increase in ICL raises bilateral trade by 7.8%, and introducing a shared official language (COL) increases trade by 29.1%.
The message is simple: language lowers trade frictions. When people understand each other, markets work better at home and abroad. Our results show that boosting linguistic proximity, especially through bilingualism, pays off in higher welfare. Treating linguistic diversity as a problem – something to be flattened rather than built upon – may come at a cost.
How language policy rewires trade
Language shapes trade through three tightly linked mechanisms.
- First, greater linguistic similarity within a country reduces internal trade costs. When people share a language, it becomes easier to communicate, do business, coordinate supply chains, and match workers to jobs. Markets function more efficiently. This is why countries with high domestic language cohesion trade more within their borders. Language, in this sense, is an economic glue.
- Second, when internal trade becomes cheaper, international trade becomes relatively less attractive. This is the multilateral resistance effect: as the domestic market integrates, foreign exporters face stiffer competition. The result is trade diversion, more internal exchange, and less reliance on imports.
- Third, language policy doesn’t just shift the balance between domestic and foreign trade; it also reshapes who one trades with. More English speakers in the US means deeper ties with Anglophone partners like the UK, Canada, or Australia, and weaker links with non-Anglophone countries like Mexico or France. Trade follows linguistic familiarity. The Canadian case makes this clear. Quebec’s French-speaking identity affects both its trade with English-speaking provinces and its international flows, tilting them toward French-speaking countries. Language acts as a filter, lowering frictions with some partners and raising them with others. US language policy works the same way. More English tightens bonds with English-speaking markets and raises the cost of engaging everyone else.
In short, language policy reallocates trade. It affects not just how much countries trade, but with whom. Even marginal changes in language use – through education, immigration, or legal status – can reshape both domestic integration and global engagement.
From rhetoric to reality: What happens when domestic language policy changes?
Language policy is often framed as a tool for promoting cohesion, streamlining communication, and reinforcing shared national values. It is frequently justified as a means of supporting economic inclusion and civic participation. But when not accompanied by robust bilingual education, such efforts may have the opposite effect. Rather than unifying, they risk fragmenting societies by disenfranchising linguistic minorities and reinforcing internal economic divisions. Our findings show that bilingualism is not an optional complement to integration; it is a necessary condition for reducing internal trade frictions, enhancing economic participation, and strengthening national cohesion.
To move from rhetoric to evidence, we use our estimates of the impact of language on international and domestic trade to simulate three policy-relevant shifts in the linguistic composition of the US population. In each, 10% of the total US population changes language status, roughly the magnitude of a major education reform or sustained generational shift. The results are large and telling. In both the US and Canada, changes in language use generate measurable shifts in trade patterns and welfare. In the US, increased bilingualism enhances domestic integration but diverts trade away from close international partners.
Scenario 1: More Spanish-only speakers
Imagine a rollback of bilingual education. In this scenario, 10% of the US population shifts from English-only to Spanish-only. The result: internal frictions rise, domestic market integration weakens, and US welfare falls by 1.1%, a loss comparable to imposing a uniform national tariff.
Mexico and other Spanish-speaking countries gain, as linguistic proximity with the US improves. But this is classic trade diversion: gains abroad come at the cost of economic efficiency at home.
Scenario 2: English speakers become bilingual
Now flip the approach. Suppose 10% of the population moves from English-only to bilingual status. The result is modest but positive. Domestic frictions fall, and both the US and Spanish-speaking partners benefit. Welfare in the US rises by 0.3%, while Mexico and others gain from tighter linguistic ties.
This is a trade-creating policy, at home and abroad. It raises total global welfare and strengthens economic links with partners in Spanish-speaking countries.
Scenario 3: Spanish speakers become bilingual
This is the most economically powerful scenario: If 10% of the population transitions from Spanish-only to bilingual, the US sees a 1.6% welfare gain, larger than most regional trade agreements. Domestic integration surges. Internal trade expands. But there’s a twist: Canada, Mexico, and Panama all lose. As the US turns inward, trade is diverted away from neighbours and toward itself.
The policy implication is clear. English proficiency among minorities strengthens internal markets but also changes America’s external posture. It boosts welfare but also redistributes it.
Conclusion
Language policy is not just symbolic. It is a structural lever that shapes the direction of trade, the allocation of welfare, and the degree of internal cohesion. Enforcing a single language without supporting bilingualism risks fragmenting the very markets such policies aim to unify.
Real economic gains come not from linguistic uniformity, but from enabling shared fluency. A more bilingual society indicates a more integrated, productive, and resilient economy. And unlike tariffs or trade agreements, the tools for linguistic integration – education, local policy, institutional design – are already in place. If the goal is national cohesion and economic opportunity, then bilingualism is not a concession. It’s the strategy.
Source : VOXeu
