Changes in political trust in Britain during the COVID-19 pandemic in 2020: integrated public opinion evidence and implications

The study examines how political trust in the UK evolved over the first year of the COVID-19 pandemic and why these changes matter for crisis management. Against a backdrop of low pre-pandemic trust following years of Brexit-related division, the authors hypothesize a rally-round-the-flag increase in trust at the onset of lockdown, followed by decline as the crisis persisted and scrutiny of government performance intensified. They identify two key time points likely to affect trust trajectories: the introduction of the first national lockdown in March 2020 (expected to elevate trust) and the Dominic Cummings scandal in late May 2020 (expected to precipitate declines). The purpose is to integrate multiple survey sources to provide a granular, week-by-week picture of general political trust, trust in national leadership, and trust in the government’s handling of COVID-19, and to consider implications for public confidence, compliance, and future crisis response.

The paper reviews theory and evidence on political trust, defined as confidence in government’s trustworthiness, fairness, competence, and congruence with citizens’ interests. Trust is related to, but not reducible to, partisanship and has been declining in several democracies (e.g., UK, USA). During crises, psychological mechanisms such as needs for control, perceptions of common fate, and seeking support from leaders can temporarily boost trust, aligning with the rally-round-the-flag effect observed after dramatic national/international events. However, such boosts are typically short-lived and contingent on perceived performance and opposition/media responses. The review distinguishes trust, distrust, and mistrust, noting measurement complexities across surveys. It also summarizes evidence that higher political trust is associated with compliance with rules and public health measures (including during COVID-19), with potential moderating roles of concern and perceived risk. These literatures inform expectations of an initial trust surge at lockdown, followed by decline, and highlight why trust trajectories matter for behaviour (e.g., compliance, vaccination intentions).

Data collection: The authors contacted survey organizations and funders and scanned media to assemble UK survey sources measuring political trust from December 2019 to October 2020. Eighteen survey sources with relevant items were identified, encompassing nearly 240,517 respondents. Measures covered: (1) general political trust in government, (2) trust in national political leadership (Boris Johnson), and (3) COVID-19-related trust in the government's handling of the pandemic. A feasibility study with a new UK sample (N=400) established comparability of items and a scoring approach emphasizing proportions expressing directional trust or distrust (versus neutral). Items leaned toward a trust–distrust continuum; the middle category varied (neutral/unsure/neither) and was not interpreted as mistrust. Data handling: Where possible, data were logged by week of year; multiple sources in the same week were aggregated weighted by sample size. Weeks were included only if total N≥100 across sources. The primary outcomes were weekly percentages expressing trust and distrust for each domain. Statistical analysis: For each domain and outcome (trust, distrust), polynomial regression models were fit across Dec 2019–Oct 2020 to identify best-fitting temporal trends based on significant improvements in fit (testing linear, quadratic, cubic, quartic). Given expectations of a post–Dominic Cummings decline, additional models were fit for the post-May 2020 period to assess whether declines were linear or curvilinear. The analysis reports model forms, fit statistics, and interprets turning points in relation to key events.

General political trust: • Immediately post–2019 General Election (Dec 2019/Feb 2020), trust was low: about 20% trusted and about 60% distrusted the government. • Around the March 23–26, 2020 lockdown, trust rose and distrust fell, with trust fluctuating 25–35% during early lockdown; at the start of May 2020, trust briefly exceeded distrust for the only time in 2020. • Following media reports of the Dominic Cummings rule breach (from May 22), distrust sharply increased, reaching about 65% by end of June. Thereafter, trust remained relatively stable while more respondents shifted from neutral to distrust; by September 2020 only 23% trusted the government, approaching pre-pandemic levels. • Best-fit models (Dec 2019–Oct 2020): trust and distrust followed quartic trends with three turning points, capturing the rise to March, reversal after late May, and slight improvement around late September. Post-May 2020: trust declined linearly (R2≈.47), while distrust followed a quadratic pattern (R2≈.78), rising to August then stabilizing/slightly decreasing by September. Trust in national leadership (Boris Johnson): • Late December 2019: 28% trusted Johnson; 57% distrusted him. • After lockdown, trust rose to 49% and distrust fell to 14%—the only point when trust exceeded distrust, evidencing a rally effect. • From April onward, trust declined and distrust increased; by October 2020, trust was 24% and distrust 57%, near initial levels. • Models: trust best fit by quartic (R2≈.56) across the full period; distrust by quadratic (R2≈.35) with a turning point between April–May indicating reversal from falling to rising distrust. Post-May: trust declined linearly (R2≈.53); distrust followed a quadratic with stabilization/decrease by September–October (R2≈.66). COVID-19-related trust (handling of the pandemic): • First measures (March 2020) showed high trust (≈48–60%) during early lockdown (April peak ≈60%). • From May to October 2020, trust declined linearly to ≈25% by October. • Distrust was lowest in March, then rose from May, leveling from September and reaching ≈68% by October. • Models: trust showed a strong linear decline across March–October (R2≈.82) and from May onward (R2≈.69). Distrust followed a quadratic across March–October (R2≈.72) with stabilization by July/October; from May onward, distrust increased linearly (R2≈.38). Overall pattern: A short-lived rally-around-the-flag increase in trust across all domains around the first lockdown, followed by persistent declines, with the Dominic Cummings scandal coinciding with a marked inflection toward increased distrust. By autumn 2020, trust metrics reverted to pre-pandemic levels while distrust remained elevated.

The integrated week-by-week evidence confirms a rally-round-the-flag effect at the onset of the UK’s first lockdown, particularly pronounced for trust in the national leadership, which then appeared to spill over to general political trust and COVID-19-specific trust. However, this boost was short-lived. From late May 2020, coincident with the Dominic Cummings scandal and growing scrutiny of government performance, trust declined and distrust rose, returning to or exceeding pre-pandemic levels by autumn. These trends align with theory that crisis-induced trust increases are contingent on perceived competence, fairness, and norm adherence by leaders. The observed spillover among trust domains supports the view that trust in specific actors and actions feeds into broader system trust. The erosion of trust likely increased challenges for government in securing compliance and cooperation for ongoing and subsequent pandemic measures, with implications for public health behaviors (including vaccination and support for measures such as vaccine passports). The findings highlight the need for strategies to maintain trust during prolonged crises, including transparent communication, consistent rule adherence by leaders, and timely, effective policy actions.

The study integrates 18 UK survey sources to provide a granular picture of political trust dynamics in 2020. It documents an initial trust surge around the first lockdown across general, leadership, and COVID-19-specific trust, followed by sustained decline to pre-pandemic levels by October 2020, with a notable inflection around the Dominic Cummings scandal. These results underscore that while crises can temporarily elevate trust, sustaining it requires consistent, credible leadership and effective policy. The work contributes an integrated, multi-time-point account of trust during a national crisis and offers practical implications for managing public confidence and behavior in future emergencies. Future research should examine how to leverage narrow windows of elevated trust, assess differential trust at local versus national levels and across institutions, and explore cross-national parallels and divergences in trust trajectories.

- Heterogeneous data sources: Surveys varied in sampling, timing, wording, and scale ranges; although a feasibility study supported comparability and analyses focused on directional trust/distrust, residual heterogeneity may increase unexplained variance. - Exclusion of neutral responses: Variability in the presence/labeling of neutral categories meant analyses focused on trust and distrust proportions; shifts between neutral and directional categories cannot be precisely inferred. - Limited scope of trust objects: Only three consistently measured forms (general government, prime ministerial leadership, COVID-19 handling) were analyzed; trust in specific institutions (e.g., health authorities, local leadership) could not be examined comprehensively. - UK-specific timeline: Findings reflect UK events and context (e.g., Brexit legacy, specific scandals); generalization to other countries requires caution, though similar psychological processes may apply. - Data availability: Not all underlying datasets are publicly available due to consent restrictions; aggregation by week may mask within-week dynamics.