The international agenda of the second Trump administration has far-reaching economic consequences. Sweeping tariffs have disrupted longstanding trade norms and raised import prices, affecting consumers and firms in the US and abroad (Clausing 2025, Kawasaki 2025, Amiti et al. 2026), while sharp breaks in global alliances have cooled economic relationships between the US and its closest allies (Gensler et al. 2025). But beyond these well-documented effects on goods trade and supply chains, geopolitical tensions also deter foreign visitors; and when visitors stop coming, the consequences for employment in tourism-dependent communities can be swift and concentrated.

As a case in point, President Trump’s rhetoric about a possible acquisition of America’s northern neighbour, combined with escalating trade tensions, led to a 25% decline in Canadian visits to the US over the course of 2025 (Figure 1). This decline is economically meaningful. Tourism supports around 10 million US jobs and accounts for about 3% of GDP (US Department of Commerce 2026), while Canadians represented approximately 28% of the 72 million international visitors to the US in 2024. The episode echoes earlier instances in which geopolitical conflicts deterred foreign visitors with measurable consequences for host economies (Ahn et al. 2022, Greaney and Kiyota 2025).

Figure 1 Canadian resident visitors to the US

Notes: The figure shows the number of Canadian resident visitors returning to Canada from the US in each week of 2023, 2024, and 2025. Figures are in thousands of visitors. The percentage figure represents the change in 2025 relative to the average of 2023 and 2024.
Source: Statistics Canada, Frontier Counts.

In recent work (Kurmann et al. 2026), we present the first systematic evidence on how this negative demand shock affected US local labour markets. The shock is geographically concentrated in tourist-dependent destinations, sectorally confined to consumer-facing industries like leisure and hospitality and retail, and still actively developing. Accurately capturing its economic consequences therefore requires granular, high-frequency data not readily available from statistical agencies. To address this, we leverage two novel private-sector datasets.

Our first dataset is smartphone foot-traffic data from Advan, covering more than 8 million points of interest across the US. Because Canada and the US share a highly integrated mobile device ecosystem, the data provider can identify devices with home locations in Canada that visit US businesses. This allows us to construct a precise measure of Canadian visitor exposure, at the level of individual ZIP codes and industries, before the tourism decline began. Our second dataset is real-time establishment-level records on weekly employment, hours worked, and hourly wage rates from Homebase, a scheduling and payroll administration platform used by over 150,000 small and medium businesses in the US, primarily in food services, retail, and leisure.

Figure 2 illustrates the geographic distribution of our exposure measure, the share of foot traffic attributable to Canadian visitors in 2024, across US ZIP codes and service-sector industries. Canadian visitor shares are highest in border regions from Washington state to Maine, as well as major tourist hubs in California, Florida, and Nevada. While in most places, this share is small relative to the total number of visitors (which includes residents), some border communities derive a substantial fraction of their customer base from Canadian visitors – in some cases exceeding 10%.

Figure 2 Exposure to Canadian tourism

Notes: The figure shows the average weekly share of Canadian visitors relative to combined Canadian and US visitors across ZIP codes in 2024, for retail trade, and leisure and hospitality.
Source: Authors’ calculations using Advan data.

To infer the causal effect of the tourism decline on local labour markets, we compare employment, hours, and wages at establishments in highly exposed local markets, those where Canadian visitors account for roughly 1% or more of total foot traffic, to outcomes at less exposed establishments, before and after the tourism decline began in 2025. Our approach controls for local economic conditions, industry-wide shocks, and time-invariant establishment characteristics, so that differences in outcomes can be attributed to differential exposure to Canadian visitors.

We find significant employment losses in highly exposed local markets, as shown in Figure 3. Employment at highly exposed and less exposed establishments tracked each other almost perfectly through 2024. Beginning in spring 2025, a sharp divergence opens up: employment in the most exposed markets starts falling relative to less exposed ones in March-April, deepens through the summer, and then persists through the end of 2025. By mid-2025, establishments in the most exposed markets employed roughly 6% fewer workers relative to comparable establishments in less exposed markets.
Interestingly, we find no significant effects on average hours and hourly wages, indicating that the adjustment occurred almost entirely along the extensive margin, consistent with evidence from other demand shocks affecting service industries (Goolsbee et al. 2025, Kurmann et al. 2025).

Figure 3 Canadian tourism decline and US local business employment

Notes: The figure shows the event-study estimates of the effects of Canadian visits on weekly employment. 90% confidence intervals are in blue.
Source: Authors’ calculations based on Advan and Homebase data.

Scaling up from establishment-level estimates to national totals requires assumptions about the broader population of affected businesses. Under conservative assumptions – focusing on the small and medium establishments covered by Homebase and assuming no demand spillovers onto non-exposed local businesses – we estimate that the Canadian tourism decline cost between 14,000 and 42,000 jobs in exposed US markets. The wide range reflects different choices about which markets to count as exposed and whether to focus on the average effect for 2025 or on the period after the full effects were realised.

While these numbers may be modest relative to total US employment, their geographic concentration is striking. The most exposed markets span only about 1,500 to 3,300 ZIP codes, home to between 9 and 26 million residents. In those communities, our estimates imply a persistent contraction of around 4% to 6% in retail and leisure-sector employment at small establishments, a substantial local shock.

Several factors suggest the true employment losses are larger. As mentioned, we abstract from demand spillover effects onto neighbouring businesses, we do not track large establishments that are important in the hotel industry, and we cannot fully capture establishment closures. A growing literature shows that tourism generates significant local economic gains through demand spillovers onto non-tourism sectors (Faber and Gaubert 2019, Allen et al. 2020), suggesting that the indirect effects of the Canadian tourism decline may be non-negligible.

The political geography of the shock is also noteworthy. ZIP codes most exposed to Canadian tourism are disproportionately likely to have voted Democratic in the 2024 presidential election and tend to have somewhat lower household incomes. This contrasts with the pattern documented for tariff retaliation in the US-China trade war, which primarily affected Republican-leaning counties (Fetzer and Schwarz 2019).

The US-Canada case illustrates a channel of harm that trade policy debates often overlook. Beyond goods trade and supply-chain disruptions, geopolitical tensions can deter foreign visitors and hit the economy fast. Such shocks strike specific places and non-tradable sectors that may find it difficult to absorb the blow. Whereas major tourist destinations like Las Vegas or Miami may be able to offset the shock by attracting more domestic visitors, less prominent places such as communities along the US-Canada border may have less capacity to do so.

More broadly, our analysis demonstrates that combining smartphone-based visitor origin data with high-frequency payroll records can deliver rapid, granular assessments of tourism disruptions as they unfold – well before traditional government statistics become available. This methodology provides a template applicable to future shocks, whether arising from geopolitical tensions, public health crises, or macroeconomic downturns.

References

Ahn, J, T M Greaney, and K Kiyota (2022), “Political conflict and angry consumers: Evaluating the regional impacts of a consumer boycott on travel services trade”, Journal of the Japanese and International Economies 65: 101216.

Allen, T, S Fuchs, S Ganapati, A Graziano, R Madera, and J Montoriol-Garriga (2020), “Is tourism good for locals? Evidence from Barcelona”, Dartmouth College, mimeograph.

Amiti, M, C Flanagan, S Heise, and D E Weinstein (2026), “Who is paying for the 2025 US tariffs?”, Liberty Street Economics, Federal Reserve Bank of New York, 12 February.

Clausing, K (2025), “The aftermath of tariffs”, VoxEU.org, 29 August.

Faber, B, and C Gaubert (2019), “Tourism and economic development: Evidence from Mexico’s coastline”, American Economic Review 109(6): 2245–93.

Fetzer, T, and C Schwarz (2025), “Tariffs and politics: Evidence from Trump’s trade wars”, VoxEU.org, 23 April.

Gensler, G, S Johnson, U Panizza, and B Weder di Mauro (2025), “The second Trump administration: Consequences for the ‘rest of us’”, VoxEU.org, 8 December.

Goolsbee, A, C Syverson, R Goldgof, and J Tatarka (2025), “The curious surge of productivity in US restaurants”, NBER Working Paper 33555.

Greaney, T M, and K Kiyota (2025), “Regional impacts of international tourism boycott: A China – Japan conflict”, Economic Inquiry, first published 3 December.

Kawasaki, K (2025), “Economic impact of US tariff hikes: Significance of trade diversion effects”, VoxEU.org, 15 September.

Kurmann, A, E Lalé, and L Ta (2025), “Measuring small business dynamics and employment with private-sector real-time data”, Journal of Public Economics 250: 105477.

Kurmann, A, E Lalé, and J Martin (2026), “When neighbors stop knocking: The impact of Canada’s 2025 tourism decline on US local businesses”, CEPR Discussion Paper 21187.

Moder, I, and T Spital (2026), “The risk of tariffs as a tool to attract manufacturing investment”, VoxEU.org, 8 January.

US Department of Commerce (2026), “US travel and tourism fast facts”, International Trade Administration, National Travel and Tourism Office.