| 8 mins read
Over the course of the Covid-19 pandemic, trust in the government became crucial to managing the crisis. Research undertaken during the pandemic showed a consistent, positive correlation between the extent individuals followed public health policies and their level of political trust in politicians generally as well as the government of the day.
However, while research to date has been informative, previous studies have also been limited by the quality of longitudinal data that is available (often drawing on individuals who participated in different surveys and answer different questions) and most have settled for providing only snapshots of trust levels at specific points of time during the pandemic. To advance this existing pool of research, I undertook a new longitudinal study on political trust that avoided these issues by sampling the same individuals and using the same measures at multiple intervals across the crisis.
A 20 month study into levels of political trust
The longitudinal study took place between 2-4 April 2020 and 10-13 December 2021 and measured changes in participants’ trust in politicians (specifically Members of Parliament (MPs) using a thorough 24-item inventory of statements about their trustworthy, and untrustworthy, characteristics. A nationally representative sample of 1200 UK citizens expressed their agreement with each statement and 700+ participants completed an identical exercise in another survey fielded in December 2021. The first and second surveys also included Covid-related questions about participants’ satisfaction with the government’s pandemic policy choices and performance as well as their own adherence to a range of Covid-19 public health measures.
Acknowledging the complexity of political trust as a concept, I adopt a definition comprised of three key features. Firstly, trust needs to be understood as a relationship between two or more people or a person and an institution (i.e. between a ‘trustor’ and a ‘trustee’). Secondly, it involves a series of multi-faceted judgements about a trustee’s trustworthiness rather than a general notion or feeling. In particular, trust arises from positive evaluations of a trustee’s competence, integrity and benevolence. Finally, trust needs to be differentiated from its brethren concepts of distrust and mistrust, which are instead are based on cynical and sceptical evaluations respectively, and which invoke expectations of either harm or the need to scrutinise the trustee more intently.
Findings from the study, a loss of trust in the government
On average, levels of overall political trust dropped by 13 per cent between April 2020 and December 2021, whilst average levels of distrust increased by 7 per cent. Mistrust did not alter. In other words, over the course of the pandemic the UK public became less positive, more cynical, and equally sceptical about their elected representatives.
Though the study was aimed at measuring trust in MPs generally, participants were also asked to self-report the ‘target’ of trust that they had invoked when providing their answers to each trust judgment about ‘MPs’. Four dominant categories of trustee emerged: ‘all MPs’, ‘the UK government’, ‘the Prime Minister’, and ‘my local MP’ (in that order).
Accounting for these new categories, the study found that levels of trust in individual local MPs actually remained broadly stable over the period. However, trust dropped meaningfully where participants thought about any of other targets. Where participants had been thinking about the Prime Minister, trust fell by 19 per cent, the largest decrease of any target. Interestingly, negative changes occurred to a broadly equal magnitude across both Conservative and opposition voter groups.
This change in trust matters. Research prior to the pandemic has shown that people withdraw their support for policies—especially those entailing risk or sacrifice—when they do not trust those in power and previous cross-sectional evidence from the pandemic suggests that a loss in trust can reduce compliance with lockdown measures.
I retest these assumptions by analysing the relationship between individual-level changes in political trust over time and participants’ compliance with two key public health measures: getting a vaccination and mask wearing. Here, longitudinal data allows me to better probe the causal inferences of previous studies. As anticipated, the data suggest a relationship between trust and compliance with pandemic-related health measures. For example, unvaccinated participants had lost nearly twice as much trust in UK politicians during the pandemic than those who vaccinated. Similarly, levels of political trust were strongly correlated with everyday compliance with obligations to wear a face mask among the survey’s participants. Participants who and lost the most trust in politicians over the pandemic were least likely to be wearing a mask ‘all of the time’ where it was either requested or mandatory by law.
Explaining why trust levels changed
Two hypotheses stand out. The first is that trust was artificially inflated at the start of the pandemic and returned to pre-pandemic levels. The second is that an initial upturn in political trust was lost because the government managed the crisis poorly - a so-called ‘trust-as-evaluation’ thesis that is current in social science.
In line with the trust-as-evaluation theory, I find that public satisfaction with the government’s pandemic performance was extremely negative (six out of eight criteria received approval ratings of less than 50%) and that aggregate approval ratings correlated strongly with changes in individual levels of political trust. Put simply, participants who lost the most trust in politicians during the pandemic were also significantly less satisfied with public health policies and outcomes, government communication and decision-making.
Whilst Conservative voters were more likely to have a positive bias favouring government performance – in line with classic theories of partisan motivated reasoning – the relationship between satisfaction and trust described above was seen to an equal or greater extent among these voters. A plausible explanation could be that, having placed their trust in the Conservative Party in 2019, these voters felt a greater sense of betrayal over any admission of policy failure by the government. This is an important break with academic expectations, and shows that trust is far from fixed among voters and can change swiftly in response to political decisions taken in a crisis scenario.
This longitudinal study provides important new evidence supporting the linear relationship between perceptions of government performance and levels of political trust in politicians. As one decreases, so does the other. It also shows that as levels of trust declined during the pandemic, so too did the likelihood of people following the government’s health measures. This replicates findings on trust and behaviour found in other non-pandemic contexts. Looking ahead, current and future UK governments will need to invest heavily in rebuilding the political trust lost during the pandemic in order to govern effectively during the next ‘unknown unknown’.