How much has Canadian travel to U.S. cities declined?

In response to increasingly strained political relations between Canada and the United States, many Canadians have reduced their travel to the U.S..

But how has this decline varied in terms of magnitude and geography?

We analyzed cell phone activity data, finding a year-over-year median decline of approximately 42% in Canadian visits to U.S. metropolitan areas between April 1, 2024 to March 31, 2025 (Year 1) and April 1, 2025 to March 31, 2026 (Year 2). This is significantly higher than the ~25% drop recorded by border crossings estimates. This means that a) border crossing data is not capturing the full drop in Canadian business and trade-related travel and b) when Canadians travel to the U.S., they are visiting fewer locations and staying for less time than they used to.

Consistent with media reporting, our data shows significant declines in snowbird destinations like Florida; border-region cities in states like New York, New Hampshire and Vermont; major tourist destinations like Las Vegas and Disney World; and winter recreation areas.

However, one of the most underreported findings is the marked decline in visits to large metropolitan economies. High-tech and financial centres like San Francisco and Houston appear to be experiencing reductions not only in tourists but also in business-related travel, reflecting changing travel preferences due to broader economic uncertainties on both sides of the border. As another example, Grand Rapids, which has close ties to the automotive industry in Ontario, has experienced the second largest drop in visitation, likely due to the tariffs.

Differences between cell phone-based estimates and border-crossing estimates likely reflect differences in measurement scope. Our cell phone data includes freight traffic, whereas border-crossing data does not. Notably, January and February 2025 were among the strongest months for Canadian exports to the U.S., likely driven by the anticipated tariff threats. The introduction of major tariffs, such as the 25% tariff in automotive parts, may explain the reduction in trade-related trips beginning in April 2025. In addition, our data measures not only Canadians crossing the border, but also Canadians living temporarily in the U.S., suggesting that the decrease in activity may reflect return migration to Canada. Finally, some Canadians have adopted burner phones when crossing the border; this could account in part for the decline we found in cell phone pings.

While Forbes estimates tourism-based revenue loss of US$4.5 billion from a 22% drop in Canadian visitation, this does not include the Canadians who are no longer living in the U.S., or the drivers moving Canadian goods. Therefore, these figures likely understate the total revenue lost from broader economic effects of changes in residency patterns and trade-related travel, as suggested from recent estimates by the State of Vermont.

A couple metros have seen an increase in activity by Canadian visitors. More research is needed to determine the factors behind this; however, it is worthy of note that Air Canada relaunched their flight service to Cleveland in May, 2025. This may have resulted in an increase in travel.

Data sources and methods

The data used to define Canadian devices traveling to U.S. metro areas is provided by Cuebiq. Cuebiq is a location intelligence platform that provides anonymized and aggregated cell phone data for academic research and humanitarian initiatives.

A stop is defined by Cuebiq as a point in space and time where the user spent a certain amount of dwelling time. A stop-detection algorithm by Cuebiq computes these stops. A "Canadian device" is defined as a unique device per day in the stops table with the country code set as "Canada" and the device type as "Home." Home devices are classified based on the duration and timing of the stops from an observation of the past 84 days, and each device is assigned a unique anonymized identifier. These Canadian devices have been recorded between April 1, 2024 - March 31, 2026 to measure daily trip occurrences.

Trips are deemed to occur when a device has a stop in Canada, followed by a stop in the U.S.. A subsequent trip begins when we see this pattern occur again. We selected the top 266 U.S. metropolitan statistical areas via their population from the U.S. Census Bureau. To determine which region gets a count on a specific day, the first stop that is in a U.S. metro at geohash level 6 is counted for that metro on that day. Following days within the same metro are not counted; however, if that device enters another metro, the first stop in that new metro is then recorded. In other words, this approach captures unique metro-device trip occurrences for Canadian travellers to the U.S..

All values are normalized by the total number of unique Canadian devices each day to account for daily variations in data coverage. The trend lines are fit via a LOESS curve.

You can download the selected normalized trip data from this link.