A Meta‐Analysis of the Government Performance—Trust Link: Taking Cultural and Methodological Factors into Account

• Published date: 01 January 2022
• Author: Jiasheng Zhang,Hui Li,Kaifeng Yang
• Date: 01 January 2022
• DOI: http://doi.org/10.1111/puar.13439

Research Article

A Meta-Analysis of the Government Performance—Trust Link: Taking Cultural and Methodological Factors intoAccount 39

Abstract: The performance-trust link constitutes an important issue, but the public administration and political

science literature has been equivocal regarding how it is influenced by cultural and methodological factors. Meta-

analyzing 72 empirical studies, this article finds that the link is stronger in low power distance countries, when

outputs are used to measure government performance, or when the focus is on local government. In addition, the

performance-trust link holds true regardless of whether performance data are subjective or objective, or whether

the studies focus on performance of government as a whole or specific agencies. The results imply that in order to

nurture and sustain trust in government, we should pay more attention to societal cultures and the way government

performance information is provided. Trust in government studies should become both more scientific and more

culturally sensitive.

Evidence for Practice

• The performance-trust relationship is stronger in low power distance countries. Government officials need to

better understand how cultural values shape citizens’ expectations and trust opinions.

• The performance-trust relationship holds true regardless of the use of objective or subjective performance

data. We should adopt a more balanced view and recognize the pros and cons of both subjective and

objective performance data.

• The performance-trust relationship is strongest when outputs are used to measure performance. Most

citizens react more strongly to what government does, such as the accessibility of test centers, the availability

of vaccines, and the quality of health services.

• The performance-trust link holds tr ue regardless of whether the studies focus on government as a whole or

specific government agencies. When citizens assess government performance and form trust opinions, they

often blur the distinction between the whole and the parts.

• The relationship between performance and trust is stronger at local levels. To increase trust in government,

national governments should consider granting local governments more discretionary power and provide

them with more resources and support.

Trust in government is essential in modern

governance, significantly influencing a

government’s legitimacy and citizens’

willingness to follow government rules and support

public initiatives (Van Ryzin 2011; Warren 2018).

The value of trust in government has become

particularly salient in the still ongoing global

pandemic of COVID-19, where governments with

higher levels of trust are seen to be better positioned

to convince citizens to comply with anticoronavirus

measures such as social distancing and wearing masks

(Bargain and Aminjonov 2020; Fukuyama 2020;

Han et al. 2020). That is, trust in government seems

to influence government performance. On the other

hand, the social science literature has extensively

studied the determinants of trust in government

and the past performance of governments has been

considered by many scholars as a crucial determinant

(e.g., Christensen and Lægreid 2005; Kumlin 2004;

Van Ryzin 2007; Vigoda-Gadot and Yuval 2003).

Taking together, there seems to involve mutual

causality between performance and trust. As Yang

and Holzer (2006) argue, countries need to avoid

falling into a vicious spiral: Low trust leads to poor

performance, which in turn leads to even lower trust.

When we zoom in on one part of the mutually

causal relationship (i.e., how past performance affects

trust)—as social scientists often do, many issues

remain to be clarified. For one thing, some studies

find no relationship between government performance

and citizen trust (Nye, Zelikow and King 1997; Van

der Meer 2010). For another thing, the empirical

studies supporting the relationship vary not only in

A Meta-Analysis of the Government Performance—Trust

Link: Taking Cultural and Methodological Factors

into Account

Jiasheng Zhang

Hui Li

Kaifeng Yang

City University of Hong Kong

The University of Hong Kong

Renmin University of China

Florida State University

Kaifeng Yang is a professor in the School

of Public Administration and Policy at

Renmin University of China and in the

Askew School of Public Administration and

Policy at Florida State University. He has

published widely on public and performance

management and citizen participation.

Email: yangkaifeng@ruc.edu.cn

Hui Li is an assistant professor in

the Department of Politics and Public

Administration at the University of Hong

Kong. Her research focuses on public and

nonprofit management, collaboration,

civil society and governance, and civic

engagement—issues that link nonprofits

and nongovernmental organizations (NGOs)

to the policy process.

Email: lihuipa@hku.hk

Jiasheng Zhang is a postdoctoral fellow

in the Department of Public Policy at City

University of Hong Kong. He is interested

in applying quantitative methods, such as

experimental design and computational

social science methods, to examine issues

related to citizen-state interactions,

performance information, and collaborative

governance.

Email: jzhan39@cityu.edu.hk

Public Administration Review,

Vol. 82, Iss. 1, pp. 39–58. © 2021 by

The American Society for Public Administration.

DOI: 10.1111/puar.13439.

40 Public Administration Review • Janu ary | Feb ruary 2022

the effect size but also in national contexts and how they measure

performance, leaving open several questions. First, existing studies

tend to anchor at a particular country (e.g., Grimmelikhuijsen

et al. 2013; Kim 2010; Van de Walle and Bouckaert 2003), so it

is unclear how national differences affect the results. Again, take

COVID-19 as an example, in the U.S., the dismal performance

of the Trump administration did not lead to a dramatic decline

but a 3% increase in citizen trust in government; despite its more

successful management of COVID-19, South Korea experienced a

slight decrease in trust levels (Edelman Trust Barometer 2021,

p. 44). Why do citizens in different countries associate government

performance and trust differently? While international comparative

studies have been done on trust in government (e.g., Grönlund

and Setälä 2012; Kim 2010) and some even emphasized the role

of culture (e.g., Mishler and Rose 2001; Wong, Wan, and Hsiao

2011), few have assessed how national cultures moderate the

performance-trust link.

Second, existing studies have paid insufficient attention to the way

government performance information is provided. One issue is

whether objective or subjective performance information should

be provided. For COVID-19, citizens can see daily numbers

of infected cases and death cases for all countries and localities,

which enables an objective evaluation of government performance

globally in fighting the coronavirus. At the same time, citizens form

their judgment about governments’ anticoronavirus performance

via watching videos, reading stories, and reflecting on personal

experiences about the pandemic. Which type of data is more

important for trust formation? Another issue is what performance

contents should be provided. Are citizens more sensitive to output

information such as how many patients are treated or how many

medical masks are distributed? Do citizens react more to outcome

information such as the recovery rate? Do they care more about

democratic processes such as following the due process and ensuring

liberty and transparency in fighting coronavirus? With several

exceptions (e.g., Mishler and Rose 2001; Morgeson and Petrescu

2011; Van Ryzin 2011), the literature is in its infancy in uncovering

such complex dynamics. Moreover, existing studies tend to focus

on the performance of a particular level of analysis (e.g., agency

versus jurisdiction; local versus national government); few have

compared across levels of analysis. Are citizens more attentive to

performance information of a whole jurisdiction or specific agencies

such as public health or emergency management departments?

Do they react more strongly to performance information of local

governments or national governments?

This study attempts to shed light on the above issues by meta-

analyzing over 30 years of empirical research on the performance-

trust link in the public administration and political science

literature. Capitalizing on the fact that the existing studies cover

100 countries, we examine how a particular culture dimension—

power distance—affects the performance-trust link. This study

contributes to a more nuanced international comparative view of

trust in government, as public administration scholars have called

for a renewed interest in comparative studies (Fitzpatrick et al.

2011; Moynihan et al. 2011). Clarifying whether the performance-

trust link is affected by how performance is measured and at what

level of analysis, we map the boundary conditions of the link.

Such knowledge will help governments decide what performance

information should be provided to citizens in order to improve

trust in government. It will contribute to not only the trust in

government literature but also the literatures on government

communication and performance management.

The Performance-Trust Link: Potential Sources of

Variation

Much has been studied on the performance-trust link, and it is

reasonable to expect that the variation of results is caused by the

studies’ design features, such as context, sampling, measure, and

level of analysis. We think the effects of such factors are not random

but systematic or patterned. Following the two broad theoretical

approaches in the trust in government literature—cultural and

institutional (e.g., Mishler and Rose, 2001; Wong, Hsiao, and

Wan 2009), we focus on two sets of factors. The cultural approach

posits that trust in government is culturally rooted, originating from

citizens’ deeply seeded beliefs about social relations (Almond and

Verba, 2015; Inglehart, 1997). Thus, citizens’ trust in government

is exogenous to the political sphere and shaped by cultural values,

beliefs, and norms in a given society (Mishler and Rose 2001).

National culture would be a critical factor. The institutional

approach, however, views trust as endogenous and a consequence

of government performance. It assumes that trust in government is

rationally based and depends on citizens’ assessment of government

performance, which has been examined differently in terms of types

of measurement and levels of analysis (e.g., Mizrahi, Vigoda-Gadot,

and Cohen, 2009; Van Ryzin, 2011).

The Cultural Approach: National Culture

Citizens’ evaluations of government performance and their trust in

government can be shaped by cultural norms and values (Almond

and Verba 2015; Mishler and Rose 2001). That is, whether and how

trust is established depends on a society’s culture, “the collective

programming of the mind which distinguishes the members of

one group from another” (Hofstede 2001, p. 21). Hofstede (2001)

identifies five dimensions of cultural values, including power

distance, individualism, masculinity, uncertainty avoidance, and

long-term orientation. Although Hofstede’s work suffers from some

limitations, it provides a useful and well-established framework for

comparing cultures across countries1 (Doney, Cannon, and Mullen

1998; Grimmelikhuijsen et al. 2013). Scholars have examined how

national cultures affect public management processes and outcomes,

such as administrative reforms and quality of government (e.g.,

An et al., 2021; Meyer and Hammerschmid 2010; Porcher 2021).

For example, Grimmelikhuijsen et al. (2013) find that the negative

effect of transparency on trust in government is stronger in high

power distance or long-term oriented cultures such as South Korea.

Porcher’s (2021) cross-countr y analysis shows that the impact

of national culture (especially individualism) on the quality of

government is significant and more important than the impact of

formal and political institutions.

In this article, we focus on power distance, which concerns the

extent to which power inequalities in society are accepted and seen

as normal.2 In contexts with high power distance (e.g., Malaysia,

Mexico, South Korea), citizens are more accepting of a hierarchical

order in which power is unequally distributed, and positions

of authority need no further justification. Citizens’ respect,

defer to, and depend on the government, which is expected to