Canada’s political conversation is increasingly being shaped by what happens south of the border. The latest Nanos weekly federal tracking, released September 22 and based on tracking ending September 18, puts Liberal support at 48.1% and Conservative support at 31.0%. At the same time, Trump and relations with the United States emerged as the most frequently named national concern, ahead of jobs and the economy. The combination captures an unusual moment in Canadian politics: foreign relations, trade uncertainty and domestic economic anxiety are increasingly overlapping rather than occupying separate political lanes. The numbers offer a snapshot of current public opinion, not a forecast of a future election, but they show how rapidly the Canada–U.S. relationship has moved up the public agenda.
Nanos Puts the Liberals at 48.1% and Conservatives at 31.0%
The headline finding from the latest Nanos federal ballot tracking is a 17.1-percentage-point difference between the Liberals and Conservatives. Liberal support measured 48.1%, Conservative support was 31.0%, and the NDP registered 11.2%. The Bloc Québécois stood at 4.8% nationally, the Greens at 3.3%, and the People’s Party at 1.0%. The ballot component of the tracking ending September 18 included 896 respondents, while the broader weekly tracking involved 1,026 Canadians.
That distinction matters because the numbers represent current voting preferences captured through Nanos’s rolling tracking model rather than an election result or projection. The pollster characterized federal preferences as broadly stable, and the previous release on September 15 had produced almost identical figures: 48.3% Liberal, 30.9% Conservative and 10.9% NDP. In other words, the newest release contains little week-to-week movement at the top even as the issues Canadians are discussing have changed considerably.
U.S. Relations Have Overtaken the Economy as the Most-Named Concern
The more striking movement is found outside the ballot question. Asked without prompting to identify the most important national issue of concern, 27.6% of respondents cited Trump or relations with the United States. Jobs and the economy came next at 21.3%, followed by inflation at 9.5%. Health care registered 5.8%, housing or the cost of housing 3.7%, the environment 3.6%, and immigration 3.3%.
The wording is important. Respondents were not handed a menu and asked to choose between Washington, inflation and health care. Nanos describes the question as unprompted, meaning people supplied the issue themselves. That makes the increase in Canada–U.S. concern particularly notable as an indicator of what was top of mind during the four-week tracking period. It does not mean 27.6% of Canadians consider every other issue unimportant, nor does it show what policies they favour. It measures which single national concern respondents mentioned first when asked an open-ended question.
The Rise in U.S. Concern Happened Quickly
Only four weeks earlier, on August 21, Trump and U.S. relations accounted for 14.9% of responses in Nanos’s issue tracking. By September 18, that figure had climbed to 27.6%—an increase of 12.7 percentage points. Over the same comparison, jobs and the economy moved only slightly, from 20.2% to 21.3%, while inflation declined from 11.3% to 9.5%.
The intervening weekly releases show how the shift developed. On September 1, Nanos described U.S. relations and the economy as statistically tied. By September 8, Trump and the Canada–U.S. relationship had become the leading national concern in its tracking for the first time in a year. On September 15, the U.S. figure reached 27.5%, compared with 21.1% for jobs and the economy. The September 18 tracking then recorded 27.6%. Because Nanos uses a four-week rolling average, these figures incorporate several weeks of interviewing and generally change more gradually than a single short field period might.
Leadership Numbers Tell a Separate Part of the Story
Nanos also measures who Canadians identify as their preferred prime minister, a question that should not be treated as interchangeable with party voting intention. In the tracking ending September 18, Mark Carney was selected by 58.6% of respondents, while Pierre Poilievre registered 21.3%. Another 10.9% were unsure. The preferred-prime-minister measure used the broader sample of 1,026 respondents.
Those figures have also moved over recent weeks. On September 8, Nanos measured Carney at 55.2% and Poilievre at 21.3%. A week later, the figures were 56.1% and 21.1%, respectively. The latest reading therefore showed further movement in Carney’s number while Poilievre’s figure remained around 21%. Still, leadership preference and voting intention measure different things: a respondent can have a preferred prime minister without necessarily intending to vote for that leader’s party. The results establish an opinion pattern in this tracking period; they do not establish why individual respondents hold those preferences.
Other Pollsters Show a Similar Broad Pattern, but Not Identical Numbers
Nanos is not the only national pollster to have measured a Liberal lead during September, although the precise figures differ. Leger’s September 5–7 online research placed the Liberals at 49% among decided voters and the Conservatives at 33%, with the NDP at 6%. Abacus Data, using research conducted September 4–9, reported 47% Liberal, 33% Conservative and 8% NDP after undecided respondents were removed.
A more recent Liaison Strategies tracker, conducted over September 6–19 and released September 21, measured the Liberals at 45%, Conservatives at 31% and NDP at 15% among decided and leaning voters. The variation illustrates why polling averages and methodology matter: firms use different interviewing modes, weighting systems, question wording, field dates and treatments of undecided respondents. The September polls broadly point in the same direction on which party had more stated support during their respective field periods, but they do not produce one uniform estimate of the size of the gap.
The Economic Weight of the U.S. Relationship Helps Explain Its Prominence
Canada’s economic connection to the United States remains unusually deep even as trade patterns diversify. Global Affairs Canada reports that the United States remained Canada’s largest trading partner in 2025. On a customs basis, 72.5% of Canadian merchandise exports went to the U.S. that year, although that share was down from 76.3% in 2024 and was the lowest since the early 1980s. Canadian merchandise exports to the United States declined by $31.3 billion, or 5.3%, in 2025.
Broader goods-and-services figures tell a similar story of continued dependence alongside diversification. Global Affairs reported that exports to non-U.S. markets increased 11.1% in 2025 and represented 32.8% of Canadian exports. By the first quarter of 2026, 64.1% of Canada’s goods-and-services exports were destined for the United States, the lowest share in that series. Those statistics do not prove that trade exposure caused the movement in political polling, but they demonstrate why changes in Canada–U.S. relations can have immediate relevance for businesses, workers and governments across the country.
Economic Anxiety Has Not Disappeared Behind the U.S. Issue
Although U.S. relations are currently the most frequently mentioned national concern in Nanos’s tracking, economic worries remain close behind. Jobs and the economy stood at 21.3%, only modestly above the 20.2% recorded four weeks earlier. Inflation was still the third-most-mentioned issue at 9.5%. Combined, the figures show that the public agenda has not simply shifted from economics to foreign affairs; instead, both are occupying substantial space at the same time.
A separate Bloomberg/Nanos measure adds another piece of economic context. The Canadian Confidence Index fell to 49.86 in the September 21 release, down from 52.81 four weeks earlier and below the neutral 50-point level for the first time in about five months. Nanos said weaker expectations concerning the economy and housing drove the decline, while assessments of personal finances and job security were comparatively steadier. That confidence index measures something different from voting intention, but its movement helps show why economic conditions remain politically relevant even while relations with Washington have become more prominent.
The Four-Week Rolling Method Is Crucial to Reading the Results
Nanos’s weekly tracker differs from a conventional poll conducted entirely over a few days. The September 18 release was based on 1,026 randomly selected Canadians aged 18 and older recruited through random-digit-dial landline and cellphone sampling. Roughly 250 new interviews are added each week while the oldest weekly group is removed, producing a four-week rolling average. Nanos reports a margin of error of plus or minus 3.1 percentage points, 19 times out of 20, for the overall 1,026-person random sample.
The company says results are weighted by age and gender using 2021 Census information and geographically stratified across Canada. Interviews were conducted in English and French, and the disclosed response rate was 6%. The ballot chart itself used 896 respondents, smaller than the full sample used for issues and preferred prime minister, so the overall 3.1-point margin should not automatically be treated as the precise uncertainty for every subgroup or measure. Most importantly, a poll records attitudes during its field period. It is not an electoral forecast and cannot by itself establish what caused respondents to change their views.
What the Latest Numbers Establish — and What They Do Not
The clearest conclusion supported by the September 18 Nanos tracking is narrow but significant: Trump and relations with the United States were the most frequently identified national concern at 27.6%, while the Liberals measured 48.1% on the federal ballot question and the Conservatives 31.0%. Similar September research from Leger, Abacus and Liaison also measured Liberal support above Conservative support, though by different margins and using different methods.
What the data cannot establish is just as important. The poll does not demonstrate that rising concern about Washington caused respondents to support one party over another, and it does not show how those same Canadians would react to future trade negotiations, economic changes or political events. Nanos itself described current political preferences as broadly stable even as the issue agenda shifted. For now, the measurable development is that Canada–U.S. relations have moved from a secondary worry to the top of Nanos’s national-issues tracker while economic concerns remain substantial.
