(Credit: © motortion - stock.adobe.com)
Voters Who Backed a Losing Candidate Are Less Trusting of Strangers Afterward
In A Nutshell
- Voters who backed losing parties in Canadian federal elections reported less trust in other people afterward, a gap of roughly 4 to 6 percentage points on average.
- The more a voter disliked the opposing party, the wider that trust gap grew, reaching about 10 percentage points among the most polarized voters.
- Losers who were confident their party would win took the hardest hit, with trust dropping nearly 10 percentage points below winners.
- The same pattern barely showed up in provincial elections, suggesting national results send a stronger signal about the country as a whole.
When a favored candidate loses, the letdown apparently runs deeper than politics. A new study finds that voters who backed losing parties report less trust in other people after federal elections, not just less faith in politicians or government. Among those with the strongest dislike of the opposing party, that gap grows even wider.
Published in the journal Electoral Studies, the study drew on survey data from six Canadian federal elections spanning 2004 to 2021, plus eleven provincial elections through 2024. After a national election, voters who backed the losing party were measurably less likely to say that most people can be trusted. That trust gap between winners and losers averaged 4 to 6 percentage points, growing notably larger in the most recent federal elections, which also coincided with higher partisan hostility in the dataset.
What stands out is that this trust has nothing to do with politicians or government. Social trust, as the researchers describe it, is the gut-level belief that fellow citizens are basically decent and honest and will play by the rules. It is the feeling that lets people cooperate and hold society together. When elections start eroding that belief, something deeper than partisan frustration may be at stake.
Partisan Hostility Widens the Post-Election Trust Gap
Before voting day, most people hold fuzzy ideas about what others in society are like. Are they honest? Do they share the same values? When results come in, voters gain concrete information: a large share of fellow citizens just backed the other side. For someone who deeply distrusts that side, the results are consistent with the idea that this lands hard.
For winners, the outcome functions as validation, since many others apparently see the world the same way. For losers, particularly those who dislike the winning party’s supporters most intensely, the researchers argue the election works as a signal that society contains more untrustworthy people than assumed. Under their theory, that signal helps explain the trust gap seen in post-election surveys, and it should matter most for people who feel the strongest pull between liking their own side and disliking the other one.
Not every loser reacts the same way. Among voters with mild or moderate feelings about the opposing party, losing made little difference to trust. Among voters with the strongest negative feelings toward rival supporters, though, the gap between winners and losers ballooned to roughly 10 percentage points, while the least polarized voters barely showed a gap at all. That distinction matters because partisan hostility in Canada has been climbing for two decades: the average distance between how Canadians rated their own party versus rival parties grew from 28 in 2004 to 43 in 2021. The trust gap tracked that rise almost exactly, staying flat in earlier elections before showing up loud and clear in 2019 and 2021, the two most polarized contests in the dataset. Researchers say the pattern raises concerns about the relationship between political competition and social cohesion, even in an established democracy like Canada.
When Losers Didn’t See It Coming
One sharper finding involved voter expectations. Researchers split losers into those who expected to win and those who already anticipated defeat. Among confident losers, the post-election drop in social trust was especially steep, nearly 10 percentage points below winners and more than twice the average gap.
That fits the researchers’ broader theory: if an election reveals something about society, the revelation should hit hardest when least expected. Voters who braced for a loss had already adjusted their view of fellow citizens well before the votes were counted. Those blindsided faced a sharper, more sudden update.
National Elections Trigger the Trust Gap, Provincial Ones Don’t
Canada offers a useful test case. It’s a country where people generally trust one another more than most, and its multiparty system should, in theory, make election results a fuzzier signal about who is in the majority. Even so, the gap between winners and losers showed up clearly at the national level.
A telling wrinkle appeared at the provincial level. Researchers drew on ten provincial election studies from 2020 to 2023, plus an original panel survey from the 2024 British Columbia election, for eleven subnational contests total. At that smaller scale, the winner-loser trust gap largely disappeared, shrinking to around one percentage point, even though the federal results had led researchers to expect a gap in British Columbia too. Smaller elections may simply send a weaker signal: a provincial result reveals something about people in one province, while a federal result speaks to the entire country, carrying more weight for a form of trust that concerns society broadly.
A broader pattern still holds. As political hostility rises across established democracies, election outcomes may connect to how voters view strangers around them, from the person bagging groceries to the neighbor two doors down. Partisan conflict’s consequences may extend beyond politics, though more research is needed to establish how election results shape individual attitudes over time.
Disclaimer: This article summarizes findings from a peer-reviewed academic study and reflects the researchers’ own framing of their results. It is intended for general informational purposes and should not be read as a definitive statement about the causes of any individual’s attitudes or beliefs.
Paper Notes
Limitations
Researchers note several constraints on their findings. The primary federal survey data captures social trust only after elections, not before, making it difficult to confirm within-person changes caused directly by the electoral outcome. Pre-election survey data from separate cross-sectional samples before the 2019 and 2021 elections allowed some before-and-after comparison at the population level, but because those samples were not composed of the same individuals, true within-person tracking was not possible at the federal level. An original panel study during the 2024 British Columbia provincial election did measure the same individuals before and after, but produced null results that researchers suggest may reflect features specific to that contest, limited statistical power, or the possibility that provincial elections are simply less relevant to voters’ social trust assessments. Researchers also cannot identify whether observed trust differences are driven more by winners updating positively or losers updating negatively, since no true neutral control group exists. Finally, the study focuses on Canadian voters who expressed an intention to vote, leaving out abstainers, which limits how broadly the conclusions can be applied to the general population.
Funding and Disclosures
The authors declare no known competing financial interests or personal relationships that could have influenced the work. Amber Hye-Yon Lee is noted to have participated in this research during her time as a post-doctoral fellow affiliated with the Consortium on Electoral Democracy (C-Dem).
Publication Details
Authors: Mark Williamson (McGill University), Amber Hye-Yon Lee, David Sumantry, Jonah Davids, Annika Maulucci, and Kristina Kisin (Toronto Metropolitan University), Daniel Rubenson (University of Toronto), and Iva Srbinovska (University of Zurich). | Journal: Electoral Studies, Volume 101 (2026), Article 103082 | Paper Title: ‘Election outcomes and social trust in polarized times’ | DOI: https://doi.org/10.1016/j.electstud.2026.103082 | Published online: May 12, 2026 | Replication data and code are available at https://doi.org/10.7910/DVN/RQMMGT.







