Lockdown,information quality,and political trust: An empirical study of the Shanghai lockdown under COVID-19

Lockdown and public political trust

A national crisis is usually accompanied by an initial ‘encirclement’ effect of public sentiment. With the development of the crisis and the emergence of response costs, political trust and confidence gradually decline. The lockdowns in European countries have been highly supported by residents and increased trust in the government (Broadbent, 2020). However, an increasing number of people in the United States have expressed opposition to the government’s lockdown measures, and the government has been severely questioned (Pressman and Choi-Fitzpatrick, 2021). Therefore, trust is to a large extent an ‘encapsulated interest’ and is based on the belief that others’ actions are good for us (rather than harmful). The trust of an individual in a political institution often indicates recognition of the performance of the institution and the political rules embedded in the institution. Therefore, the public’s trust in the government usually indicates the extent to which the actual performance of the trust object is in line with the subjective expectation of the trustee.

Citizens evaluate political institutions and policies in times of crisis and express a certain degree of gratitude for prompt relief (Bechtel and Hainmueller, 2011), especially when a high degree of uncertainty under COVID-19 has weakened the common cognitive processes used by the electorate to evaluate the political system. Therefore, the psychological response to the COVID-19 pandemic has reshaped political trust during the crisis. Shanghai residents were strictly prohibited from going out of their homes during the lockdown period. The longer the residents were isolated, the more negative their assessment of government performance was and the less confidence they had in the government’s competence to effectively control the pandemic. Therefore, the longer the lockdown, the stronger the level of the anxiety and depression of the residents due to the pandemic (Khubchandani et al., 2021).

Meanwhile, the longer that residents stay at home, the more difficulties they confront in securing medical care, food, living supplies and other necessities. According to Maslow’s hierarchy of needs theory (Maslow and Lewis, 1987), when the most basic needs for survival cannot be met, the benevolence of government policies will be questioned. Meanwhile, the protection of residents by the government has become a prerequisite for legitimacy. As part of the psychological contract, residents actually hope that the government can provide full support for life and therefore expect the government’s commitment to be further strengthened. However, as the form of the pandemic changed very rapidly, the government often found it difficult to meet the expectations and commitments of the residents, which reduces the residents' confidence on government. Considering residents’ psychological challenges in lockdown, we propose Hypothesis 1 as follows.

Hypothesis 1. Lockdown weakens residents’ political trust in term of competence, integrity and benevolence.

Information quality and political trust

In modern democracy, power comes from the people, and the citizens are the owners of the state, and so the government is the agent of the citizens, entrusted by the citizens to manage state affairs. According to principal–agent theory, citizens, as the principals of the governments, have the natural right to require the government as the trustee to perform with transparency (Schnell, 2018). The provision to the public of true and accurate information by the government can prompt greater public trust in the government (Grimmelikhuijsen et al., 2020). During the COVID-19 pandemic, the government released various pieces of information on pandemic responses through various channels, including pandemic information, residents’ action guidelines, and pandemic prevention policies and measures. Therefore, according to principal–agent theory, the public estimates the performance of the government through the information released by the government. The development of COVID-19 pandemics is highly unpredictable and is marked by great uncertainty, such that timely, accurate information can help residents understand the government’s efforts to respond to the pandemic and protect residents’ health, which helps to enhance political trust.

Information is the basis for people’s evaluations and decisions, and information quality can be measured by several characteristic dimensions (Levi and Stoker, 2000), such as timeliness, accuracy, competence and understandability. Different dimensions of information quality have differential effects on people’s psychological status and responses. When the accuracy of the information published by the government is called into question, residents probably believe that the government is not capable of scientifically conceiving and proposing pandemic prevention measures. At the same time, misinformation may also give rise to conspiracy theories and rumors. Therefore, misinformation is also a kind of invisible virus that weakens trust in the government. When the spread of the pandemic is rapid, residents need more accurate information to assess their risk, behavior and planning. Therefore, timeliness of information is critically important to residents. When the government provides information in a timely fashion, residents can feel the responsiveness and usefulness of the government (Houston and Harding, 2013), and the public ascribes greater legitimacy to the government.

The COVID-19 pandemic has impacted and changed social norms and regulations. The public urgently desires to learn about the progress of pandemic responses and seeks to understand regulations of social affairs. The information on which regulations are based is usually fragmented and decentralized in different parallel government authorities due to professional bureaucracy. Meanwhile, government authorities often face great challenges in information collection due to a lack of resources and tremendous workloads in emergencies. Publication of timely and accurate information intimates the government’s competence, while distributing sufficient information in an understandable manner indicates benevolence to residents. Therefore, we have the following research hypotheses:

Hypothesis 2a. Government authorities’ provision of quality information positively impacts residents’ political trust.

The moderating effect of information quality

Publicity of governmental information helps residents understand the government’s actions and the government’s concern for residents’ interests (Alamsyah and Zhu, 2022). As mentioned above, effective information announcements can improve government credibility and political trust. The theory of emotional cognition indicates that individuals acutely perceive certain stimuli, such as information, in the environment to which they have an emotional response (Lazarus, 1991). When the government publishes valuable information, residents have a clear understanding of the government’s action plans. Political trust increases if the government’s actions are in line with the residents’ best interests and expectations (Wirtz et al., 2016; Zhai, 2022). With the increase in the duration of lockdown and the continuous increase in the number of people infected with the COVID-19 virus, the residents of Shanghai experienced a more obvious panic. Accordingly, the deputy mayor of Shanghai led other administrative departments to hold daily pandemic information and press release conferences. The information promulgated in the conferences normally included pandemic control progress, social life guidance and responses to public appeals. As a result, the residents knew the government regulations better and avoided rumors, which positively affected political trust.

Residents’ troubles and anxieties were increased by the lockdown. For more than 2 months, 25 million residents were asked to stay in their homes from time to time, which brought Shanghai to a standstill. A lack of survival supplies, reduced income and the uncertain risk of pandemic spread greatly aggravated the anxieties of many residents. The World Health Organization has reported a 25% rise in the prevalence of depression and anxiety around the world in the first year of the COVID-19 pandemic. As the lockdown lengthened, residents increasingly struggled with shortages, anxiety and uncertainty. Reliable timely information publicity is thus expected to relieve the anxiety of residents in general, especially those isolated at home for a long time. Thus, we have Hypothesis 3:

Hypothesis 3. The quality of official information has a positive moderating effect on the relations between lockdown and political trust.

Participants

As there was no way to complete face-to-face interviews during the Shanghai lockdown, we used a computer-assisted survey system to administer the survey in Shanghai. With reference to some existing studies (e.g. Cato et al., 2020; Homburg et al., 2022; Schernhammer et al., 2022), we managed to sample across various age groups and gender groups by quota sampling based on the age structure and gender structure of the Shanghai population. We sent out 1250 invitations in stages and discarded 17 surveys that took no more than 300 seconds, and finally obtained 1063 valid respondents (see Table 1 in the online supplemental material).

Measures

China’s ‘Zero-COVID’ policy was a nationally implemented policy that proposed a national unified diagnosis and treatment guide. China’s COVID-19 response committee, known as the joint prevention and control mechanism, led the consistent national response to the COVID-19 pandemic. As for large-scale outbreaks of COVID-19 in regions, the central government delegated high-level official medical experts to lead the governmental response on such occasions and dispatched resources nationwide. As many Chinese residents deem the government as monolithic in responding to COVID-19, we measured respondents’ political trust in the whole government system. Political trust is measured by three items used in prior studies (e.g. Adinugroho and Simanjuntak, 2021; Grimmelikhuijsen and Knies, 2017; Han and Yan, 2019), where the respondents were asked whether they were confident in the government’s competence, benevolence and integrity. Similar to the measure of information quality, responses were rated on a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree).

Governments may not always succeed in communicating information to residents (Li, 2020), and effective communication depends in part on information credibility and information quality. Information credibility represents residents’ perceived truthfulness of information, which depends on the perceived believability of the information source (James and Van Ryzin, 2017). When residents distrust governments, residents do not believe governmental information publicity and the government would have to rely on credible information intermediaries (e.g. experts, volunteers, non-governmental organizations (NGOs), frontline staff, etc.) to access information (Chen and Liu, 2020; Zha et al., 2020). The items quantifying information quality measured the government’s provision of quality information with an instrument from Mansoor (2021) and Arshad and Khurram (2020) by asking respondents whether the government authorities provided sufficient, accurate, timely and understandable information regarding the progress of and solutions to the COVID-19 pandemic and responses to residents’ major concerns. Responses were rated on a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree). The survey asked the respondents to report the number of days since they started isolation at home. The demographic variables (age, gender, education, family income and number of underage children in the household) were measured in our survey and controlled in the regression models. The main measurement items are illustrated in Table A1 in Appendix 1.

Political trust during lockdown

Models 1, 2 and 3 presented in Table 2 in the online supplemental material examine the effects of political trust on competence, benevolence and integrity, respectively. The variance inflation factor (VIF) detects the severity of multicollinearity in our regression models, and statistical experience suggests that the VIF values of explanatory variables should be no more than 10 (Shrestha, 2020). The regression results of models 1–3 suggest that the VIF value of each variable is far below the threshold value of 10, which proves that there is no multicollinearity problem among the explanatory variables. It can be found that age and family have a negative impact on political trust. It is a common research finding that age has an impact on people’s psychological status, which is not supported by the results of analysis in Table 2. Depression information has a greater negative psychological impact on residents. Elderly people have experienced and/or might have suffered from many historical governmental affairs, and thus they report lower political trust than younger people. During the pandemic, adults were more concerned about physical health problems to avoid being infected, and had struggled to secure sufficient supplies of necessities for their families. Thus, the respondents with underage children present, as is reasonable, a low level of trust in the government in terms of competence, integrity and benevolence. When we add more explanatory variables to the regression (models 4–9 in Tables 3 and 4 in the online supplemental material), the residents with underage children do not show a significant difference in political trust from those without. Because the impact of the COVID-19 pandemic is practically the same on everyone, all families faced the same urgent difficult situations. Therefore, the effect of concerns about underage children is thereby weakened. From this perspective, it is more feasible that the presence of underage children does not significantly affect residents’ attitude of policy trust.

In many studies, gender, education and income structure are taken as key demographic variables that affect attitudes. The results of the study in Table 1 show that the social class of residents does not affect political trust in the COVID-19 pandemic. Women are usually more psychologically vulnerable and psychologically sensitive than men, so they have different psychological attitudes on many issues. However, although COVID-19 is also unfamiliar to men, both men and women share the same pressures and debilitating anxieties. This may explain the consistency in political trust between male and female respondents.

The impact of the main explanatory variable – that is, lockdown – on political trust can be observed in models 1, 2 and 3. Unsurprisingly, it was found that lockdown had a negative impact on residents’ policy trust. When we add more explanatory variables, the effect of lockdown on benevolence trust is no longer significant (Tables 3 and 4). Because the priority target of China’s zero-COVID policy is to reduce deaths, the benevolence of government is a widely publicized concept of governance among residents. Therefore, it is credible that lockdown has no effect on the benevolence aspect of political trust.

The impact of information quality on political trust

Table 3 suggests that different characteristic dimensions of information have different effects on different components of trust. Since the VIF values among the explanatory variables of regression models 4–6 are far below the threshold value of 10, there is no multicollinearity problem among the explanatory variables. The regression results suggest that the timeliness and sufficiency of information have a significant impact on all dimensions of political trust. This result clearly points to one conclusion: if the published information is not timely or accurate, or even if there is an error, then it will definitely reduce residents’ political trust. Although it is very difficult to quickly collect and process information and accurately send information under the chaotic situation of a crisis, it is indeed a very obvious tendency. Models 4–6 show that information sufficiency and understandability negatively affect political trust in integrity and benevolence (p < .01). Model 4 demonstrates that information sufficiency does not affect residents’ attitudes toward the government’s political competence (p > .05), and that the information quality of understandability affects political trust with slight significance (p < .05), which is opposite to the results for model 7 when more explanatory variables are included. As presented above, political trust is a kind of resident perception of government performance, and the timeliness and accuracy of information actually confirm the government’s competence in obtaining information. Correspondingly, information understandability and sufficiency does not affect people's perception about government performance. Thus, it is reasonable that information understandability does not significantly affect political trust.

Two enlightening results about the determinants of political trust are also shown in Table 3. Political trust in integrity and benevolence are affected by all dimensions of information quality. Meanwhile, political trust in competency is affected by information accuracy and timeliness, but not by information sufficiency and understandability.

Moderating effects on political trust

In Table 4, we note significant interactions among the three groups. First, information timeliness and information understandability do not show significant interaction effects with lockdown. However, both information sufficiency and information accuracy have significant interactive effects with lockdown. Second, information sufficiency and information understandability neither directly affect competency trust nor moderate the relationship between lockdown time and political trust in competency. At the same time, political trust in integrity and in benevolence are simultaneously directly affected and moderated by various dimensions of information quality. Therefore, political trust in integrity and in benevolence are more sensitive to information quality in many ways.

Models 7 and 8 suggest that information accuracy weakens the negative impact of lockdown on political trust in competence. Meanwhile, information accuracy also weakens the negative impact on integrity (see Figures 1(a) and 1(b) in the online supplemental material). In model 9, it was found that lockdown time moderates the relationship between information sufficiency and benevolence trust (see Figure 1(c) in the online supplemental material).

Discussions

Strict lockdown under the COVID-19 pandemic is a remarkable experience of struggle that has not been borne before by this generation. This paper studies the impact of lockdown on political trust and examines the potential moderating effects of information quality. Unlike previous studies that explored the determinants of general political trust, this paper considers three categories of political trust. Meanwhile, some prior studies have taken information quality as a general determinant, while this study classifies information quality along four dimensions and explores their individual effects on political trust. The empirical results indicate that lockdown has a negative impact on political trust in integrity and competence, but it does not affect residents’ political trust in benevolence. Therefore, the impact of lockdown measures on political trust is heterogenous and niche-targeted. This empirical study has found that information quality has an impact on political trust in integrity and benevolence, while information sufficiency and understandability affect political trust in competence. At the same time, it was found that information accuracy significantly weakens the negative influence of lockdown on political trust in competence and integrity. Information sufficiency was also found to moderate the relationship between lockdown and political trust in benevolence. Therefore, the empirical results hold some management implications.

First, the lockdown time does not affect residents’ political trust in benevolence. COVID-19 brought chaos to society and fundamentally affected the lives of people all over the world. The COVID-19 pandemic also exhausted the service capacities of the people. The gap between the supply capacity and residents' demand widened with the prolonged lockdown.

Unsatisfied physiological needs trigger strong emotional responses from residents and cause them to blame their government for the missing supplies. As a result, lockdown weakened the residents’ confidence in the government’s competence and service for the people through its political announcements. This explains the negative impact of lockdown on political trust in competence and integrity.

Although lockdown weakened some categories of residents’ political trust, it did not have a significant impact on residents’ political trust in benevolence. In contemporary governance, residents always ask the government to take full responsibility for each activity and ensure that the public interest has greatest priority (Farazmand and Carter, 2004). Therefore, the interests of the majority have come to be the main concern of democratic governments (Turcott Cervantes et al., 2021). Since the beginning of the COVID-19 pandemic, China has employed a sweeping draconian measure, namely the zero-COVID policy, to control infections. The government promoted the success of the policy in keeping infections and deaths much lower than in western countries. Although residents were frustrated and depressive in lockdown, they still believe in the widely broadcast message that the government is taking all measures to protect people’s lives, which is why lockdown did not significantly affect residents’ political trust in benevolence.

Second, different dimensions of information quality have differential effects on political trust. Researchers have shown that the high-quality information published by government agencies on social media affects the public’s trust in the government (Schraff, 2021). This study further subdivides information quality into three dimensions for more detailed explorations. This study suggests that the information qualities of timeliness and accuracy have significant impacts on all categories of political trust, while information sufficiency and understandability did not increase political trust in competence.

According to the zero-COVID policy, the travel trajectory of each infected resident in the past one week is tracked back and all corresponding high-risk close contacts to the infected cases are also identified. Both infected patients and close contacts are transferred to special cavernous quarantine centers for treatments, and risky places are strictly disinfected to clean up their environments. The actions and progress, as well as responses to residents’ appeals, are officially announced in daily press conference. Accordingly, official reports and briefing are available to residents by rolling media broadcasts. Therefore, residents can accurately and in timely fashion learn about the actions, measures and progress under the fluid and rapidly changing situation. As most of their fears and anxieties come from unknown measures and uncertain risk, information timeliness and accuracy enhance residents’ confidence and certainty. Equally, this explains how information timeliness and accuracy have a significant influence on political trust of competence.

Indeed, there were many obscure terms in the announcements of the governments. For example, dynamic clearing (i.e. zero virus in communities), silent management (i.e. strict lockdown), daily clean (i.e. plan and finish the daily workloads) and trio individual measures (i.e. wear masks, hold social distance, keep personal hygiene) are high-frequency terms in official announcements and news. Since they represent technical strategies in pandemic prevention and control, knowing such vocabulary supplies additional knowledge about the progress of virus control measures. As presented in model 4, information sufficiency and understandability do not show an influence on political trust in competence. However, the residents recognize the persistent efforts of governments through sufficient and understandable information. The political trust in benevolence and in integrity are thereby affected by information sufficiency and understandability.

Third, the interaction between information quality and lockdown with political trust enlightens policy improvements. First, information accuracy moderates the influence of information accuracy on political trust in competence and integrity. As mentioned above, the uncertainty and anxiety of individuals in terms of income, career opportunities and social service provision increase with lockdown. All of the shortages and depressions agglomerate into political distrust in competence and integrity. In the United States and the United Kingdom, the governments are sometimes blamed because of ambiguous and inconsistent information, which leads to unreliability and anxiety (Bearman et al., 2020). Due to the rapid spread of the pandemic in the community, the lockdown has been continuously extended many times in Shanghai. Many residents started to panic after running out of necessities at the beginning of lockdown. These residents were eager for reliable information to estimate the situation, organize their on-hand supplies, and prepare for the worst. This explains the empirical finding that information accuracy mediates the relation between lockdown and political trust.

Cognitive dissonance theory posits that people experience psychological conflict between new and old cognition when they deal with new situations. The inconsistency of cognition normally aggravates tension and discomfort in human beings, in which information and knowledge help to restore emotional and psychological balance. Sufficient information allows individuals to update their knowledge of changing social norms, provides guidance regarding the adjusted regulations and earns political trust in benevolence in return. In the context of the Shanghai lockdown, residents were not accustomed to go from unrestrained social life to isolation with many shortages. People’s emotional discomfort and desire to return to normal life increase with the time they are isolated at home, so information sufficiency becomes increasingly helpful in enhancing political trust in benevolence over time.

Contributions

Public relations scholars emphasize the importance of strategic risk communication during a public health crisis for organizations to build quality relationships with the public and influence their behaviors in a way that is desired by the organizations (e.g. Guidry et al., 2017). In contrast, the lack of risk communications through information sharing provokes controversy, universal anxiety or even social conflict in society. For example, US government organizations (e.g. the Centers for Disease Control and Prevention (CDC)) have consistently publicized that social distancing is effective in slowing the spread of COVID-19 across communities. However, it has been found that practicing social distancing is very difficult for the public (Lee and Li, 2021). Strict political measures lacking broad understanding and support (e.g. lockdown measures, social policies, compulsory vaccination during the pandemic, etc.) fueled public demonstrations and violent protests during the COVID-19 pandemic in many countries between January 2020 and January 2021 (Bartusevičius et al., 2021; Filsinger and Freitag, 2022). Thus, some literature explores the effectiveness of the infodemic of misinformation (Lockyer et al., 2021), information transparency (Petersen et al., 2021) and fake news (Hartley and Vu, 2020), among others. Since government information provides vital knowledge, reports the measures taken against the pandemic and ultimately enhances political trust in return (Chryssochoidis, 2009), good-quality information plays a positive role in enhancing political trust.

A number of studies have explored and emphasized the influence of information quality on residents’ trust in government and information sources (Lu et al., 2021). However, information quality is a broad concept that comprises several quantitative dimensions. To the best of our knowledge, little theoretical research has examined the specific influences of each dimension of information quality on political trust. This study classifies information quality into three different dimensions, and political trust into three components. The COVID-19 outbreak provides a particular crisis scenario to investigate how the different dimensions of information quality affect each component of political trust. The study presents a full map of the relationships among the dimensions of information quality and components of political trust.

As the public increasingly expect transparency and accountability from organizations, information sharing is regarded as a key element in contemporary strategic communication to foster public confidence and trust in organizations, such as the government (Kim et al., 2022). The findings of this study further suggest ways in which political trust can be enhanced that shed light on strategical information sharing for government authorities. Given that people’s intentions and behaviors in a crisis are largely determined by the contextual environments (Jong, 2020), governments’ communicative efforts during a pandemic are essential and crucial. However, the crisis itself challenges the government significantly in many ways. Historical experience suggests that lack of resources, shortage of knowledge about the crisis, poor social management and overloaded emergency services have often emerged in times of crisis. Consequently, strategic responses and reactions during a crisis are helpful and critical for government authorities. As people’s attitudes are largely influenced by the information they receive, government information disclosure is a predictor of their political trust and behaviors. This study provides a comprehensive analysis of relationships between information quality and political trust, which are useful for optimally imposing implications on information gathering, processing and publicity.

Limitations and future directions

Although this study yielded significant results, there are still some limitations, and more extensive studies in the future may offer additional insights and policy implications.

First, more interesting explanatory factors can be taken into consideration and enhance the findings. For example, the Chinese Communist Party (CCP), which has 96 million members, is the ruling party in mainland China. The CCP central committee and its local party organizations appeal to CCP members to serve the people all the time; further analyses may explore whether and how party affiliations affect people’s political trust. At the same time, this study was conducted during a city-wide lockdown, where the lockdown time was included as an important explanatory variable. Future studies can conduct more research on different crisis scenarios, including natural disasters, technical crashes, and political and economic crises.

Second, the sampling methods could be more diverse and include a greater variety of respondents. Because of the government regulations on home isolation, the current study only collected information via an online survey due to the strict lockdown in Shanghai, which limits the generalizability of the results. We managed to sample across various age groups and gender groups, but high-educated and middle-aged residents seemed to be over-represented. To include a more varied of population, future researchers can conduct a parallel offline survey to provide comparisons in future research.

Third, the measurements of political trust can be improved by performing more extensive studies. Although some Chinese residents deem the government as monolithic in response to the COVID-19 pandemic, more residents perceive substantial differences between local government and central governments in their daily lives. Thus, the causal linkages among the explanatory factors of political trust in different levels of government afford extra room to perform more studies. As residents’ behaviors and attitudes are affected by regional factors, future studies sampling different regions may yield new findings on the relationship between information quality and political trust. In any case, it is clear that information quality in a crisis presents a variety of research directions.