This chapter presents the results of the OECD Survey on the Drivers of Trust in Public Institutions conducted in Brazil during April 2022. It provides a detailed analysis of the main drivers of trust in the federal government, local government and the civil service. Where possible, the chapter provides meaningful and insightful comparisons between trust levels in Brazil and other surveyed Latin American OECD countries, as well as OECD average, offering valuable context to the findings.
Drivers of Trust in Public Institutions in Brazil
Building Trust in Public Institutions
2. Trust in public institutions and its main drivers in Brazil
Abstract
This chapter presents the results of the OECD Survey on the Drivers of Trust in Public Institutions (OECD Trust Survey) carried out in Brazil, after considering the adjustments made to the original survey that was carried out in 22 OECD countries at the end of 2021 (OECD, 2022[1]). It then provides an overall analysis of the main drivers of trust in the federal government, the local government and the civil service, which is expanded in Chapters 3 and 4. Where possible, it compares results in Brazil with trust levels in other surveyed Latin American OECD countries, and the OECD average. While this chapter relies predominantly on data compiled through the OECD Trust Survey it also makes use of relevant secondary sources when discussing factors underpinning public trust in Brazil and in Latin America.
The analysis in the study relies mainly on the data collected through the OECD Trust Survey, carried out online in Brazil in April 2022. At the time of the survey, the country was still grappling with the aftermath of the COVID-19 pandemic, while experiencing rising inflation, a highly polarised political environment during the presidential campaign for national elections and increased tensions between branches of government.
2.1. Measuring drivers of trust and the OECD Survey
The OECD Trust Survey is a new measurement tool for democratic governments seeking to improve public governance and reinforce trust. It is the first cross-national investigation dedicated specifically to identifying the drivers of public trust across levels of government and across public institutions (OECD, 2022[1]). The results of the inaugural OECD Trust Survey, implemented in 2021 in 22 OECD countries, provide a cross-national stocktake of the complex relationship between public trust and democratic governance and constitute a baseline to inform policy making in various areas of public governance.
The 2021 OECD Trust Survey (Box 2.1) has been the subject of extensive reflections and testing since 2017. It has been revised and expanded based on methodological suggestions and empirical lessons reflected in the OECD Guidelines on Measuring Trust (OECD, 2017[2]); the TrustLab project (Murtin et al., 2018[3]), a consultative process “Building a New Paradigm for Public Trust” that took place through six workshops during 2020-2021; the updated conceptual Framework on Drivers of Trust in Public Institutions (Brezzi et al., 2021[4]); in-depth case studies conducted in Finland, Korea, New Zealand and Norway (OECD/KDI, 2018[5]; OECD, 2021[6]; OECD, 2022[7]; OECD, 2023[8]); and discussions held at the OECD Public Governance Committee in 2021 (GOV/PGC/RD(2021)) and at the OECD Committee for Statistics and Statistical Policy in 2020 (SDD/CSSP(2020)).
Box 2.1. The inaugural 2021 OECD Trust Survey
The 2021 OECD Trust Survey carried out by the OECD Directorate for Public Governance had significant country coverage. There were around 2 000 respondents per country in the 22 participating countries: Australia, Austria, Belgium, Canada, Colombia, Denmark, Estonia, Finland, France, Ireland, Iceland, Japan, Korea, Latvia, Luxembourg, Mexico, the Netherlands, New Zealand, Norway, Portugal, Sweden and the United Kingdom. These large samples facilitate subgroup analysis and help ensure the reliability of the results.
Most countries were surveyed in November-December 2021, with a few surveys taking place in January-February 2022.
The surveys were conducted online by YouGov, national statistical offices (in the cases of Finland, Ireland, Mexico and the United Kingdom), national research institutes (Iceland) or survey research firms (New Zealand and Norway). The survey uses a non-probability sampling approach, based on ex ante country-level quotas representative of the population by age, gender, level of education and region. This type of sample construction was the most feasible option for the OECD Trust Survey given its simplicity, timeliness and lower cost of implementing the survey, which took place in the same period in a large number of countries. The quotas were derived from national estimates of group prevalence based on probabilistic surveys, census data or administrative data.
The survey process and implementation were guided by an Advisory Group comprised of public officials from OECD member countries, representatives of National Statistical Offices and international experts.
The OECD Trust Survey’s measurement approach focuses on situational questions (OECD, 2017[2]). Rather than typical behavioural questions, the survey does not ask respondents about individual behaviour but rather on the conduct they expect from a third party, in this case a public institution, a civil servant or a representative political figure. A battery of 15 core questions assess the trustworthiness of these third parties in alignment with the competence (responsiveness and reliability) and values (openness, integrity and fairness) expected of public institutions as included in the OECD Framework on Drivers of Trust in Public Institutions (Box 1.4 in Chapter 1).
The OECD Trust Survey uses an 11-point scale for the response choices on questions about levels of trust and drivers of trust, following reviewed best practice and applications in country studies. The questions on trust use a numerical 0-10 scale with verbal scale anchors, as this allows for variance in responses, increases overall data quality and complexity, and facilitates translatability across languages. The scale offers more nuanced analysis, allowing respondents to provide a “neutral” response that other surveys do not.
2.1.1. Ensuring the analytical and measurement tool is fit for purpose in Brazil
The OECD Trust Survey as implemented in Brazil (Box 2.2) includes the core questions on drivers, in order to ensure comparability of data, as well as additional questions introduced to capture some specific aspects of Brazil, and more generally of Latin American countries (find the complete questionnaire in Annex B).
To ensure that the analytical framework and the survey address the most relevant drivers of trust in public institutions in Brazil, the OECD set up an Advisory Group, comprising academics, policy makers, representatives from statistical offices, international organisations and experts, which discussed the context-specific factors that may have an impact on trust, and proposed adjustments to the survey questions and modalities.
Box 2.2. Methodological notes: The implementation of the survey in Brazil
The OECD Trust Survey in Brazil was implemented online by Netquest using a sample of 4 140 respondents between 7 April and 6 May 2022. The average response time of the questionnaire was 19 minutes.
Survey respondents were residents in Brazil aged 18 years and over. 3 723 respondents were part of Netquest’s own online panel; and 417 from a partner company. The partner company’s panel ensured that the survey was sent out to and answered by a sufficient number of people with low levels of education.
The survey sample is representative of Brazil’s population by education, age, gender, residency in the five main Brazilian regions, and socio-economic groups (Critério Brasil/ABEP),1 following the distributions of the most recent Instituto Brasileiro de Geografia e Estatística (IBGE) Census (Table 2.1). Post-stratification weights were used in the analysis to ensure the sample was fully representative.
Table 2.1. Composition of the weighted sample by gender, age, region, level of education and socio-economic category
1. The Brazil Economic Classification Criteria (Critério Brasil-ABEP) classifies households in five groups (A-E) estimating the purchasing power of urban individuals and families. It assesses access to public utility services, householder’s education and possession of several amenities (such as bathroom, dishwasher, freezer, etc.) (ABEP, 2021). For some analysis in this report and in order to aggregate results categories are grouped.
Note: The survey’s drop-out rate was 11%. One caveat to highlight is that the survey sample is not representative by state. Therefore, it does not present data or allow analysis at the state level, which is of utmost importance given regional disparities in the country.
Source: Netquest, 2022
After taking discussions held by the Advisory Group into consideration, and also making several methodological revisions (such as of the wording of questions), the trust survey questionnaire included six additional questions piloted in Brazil (Box 2.3).
Box 2.3. Questionnaire revisions
Country and context considerations underscored by the Advisory Group:
In Brazil, as is the case more broadly in Latin America “clientelist” relations could be widespread (Rocha, Souza and Araújo, 2019[10]; Koster and Eiró, 2021[11]). Clientelism involves a personal dependence that is based on the mutual exchange of favours between two people who control unequal resources (Medard, 1976[12]). It affects people’s perceptions about government, as they often tend to link the provision of public goods with social or political brokers.
Informality is a key challenge in Latin America which hinders tackling poverty and addressing inequality, thus undermining people’s trust. According to the latest Latin American Economic Outlook, on average, almost half (45.3%) of people in Latin American and the Caribbean (LAC) countries live in a household that depends solely on informal employment, and 21.7% live in households with formal and informal workers (OECD et al., 2022[13]).
During the early 1990s Brazil became a regional exemplar of social participation, but political participation has fallen significantly in the last decade (Avritzer, 2017[14]). Participation encourages the sense of having a stake in collective endeavours; civic-minded citizens participate more frequently and tend to have higher levels of trust than passive citizens (Almond and Verba, 1963[15]; Prats and Meunier, 2021[16]).
Several studies have documented how the use of social media has a key impact on people’s political attitudes and behaviour in Brazil (Duque and Smith, 2019[17]). The lack of regulation and the speed with which information spreads increase the risks of mis- and disinformation. For instance, in 2018, 32% of Brazilians reported trusting social media for news1and around half of voters reported being informed by social media during 2018 election.2 In a survey of 41 countries, Brazilian Internet users aged 14-64 ranked second only to Filipinos in the amount of time spent online daily (We Are Social, 2019[18]).
Latin America is often portrayed as the world’s most violent region. Brazil, along with three other LAC countries (Colombia, Mexico and Venezuela), together account for one-quarter of all the murders in the world (Muggah and Aguirre Tobón, 2018[19]). Exposure to criminal violence is likely to erode both interpersonal and institutional trust. Trusting others entails taking a risk that might be increased or reduced depending on certain individual and societal-level characteristics. Thus, the sense of insecurity that prevails in countries like Brazil, where violence and crime abound, represents an important impediment to the development of trust (Delhey and Newton, 2003[20]).
Table 2.2. New questions included in the OECD Trust Survey in Brazil
Some lessons learned:
The responses to the question addressing economic informality had contrasting results to other questions on reliability. One possible reason is that people are asked about themselves instead of their perceptions about public institutions and they tend to be more positive and less likely to report that they would be willing to break the law. In future survey questionnaires, this question would need further revision and reformulation.
The question addressing perceptions of (in)security was found to be significant for trust in local government. This is an important question to keep in the questionnaire, although it would be worth exploring if the wording of the question, i.e. referring to people’s neighbourhoods as opposed to their country, may have affected the results.
The perceived importance of social contacts for accessing services has a strong and negative association with trust. Respondents who attach more importance to contacts for accessing goods and services have lower trust in all institutions. Similarly, the more willing someone is to register a business, the higher their trust in all institutions (strong positive relationship).
Feelings about the likelihood of success of collective action in the form of a petition in respondents’ neighbourhoods is also strongly associated with trust. Generally, the results on perceptions of clientelism and perceptions of collective action are closely aligned with the academic literature and previous research. However, the question used different response options and different response scales compared to the main set of questions of the OECD Trust Survey, which uses a 0-10 response scale, and it would be relevant to test whether this had an impact on the results.
The survey questionnaire also includes questions about respondents’ access to services and the quality of their housing’s environment to infer respondents’ living situation. Water access matters for trust in political parties and news media. However, the quality of the road at the respondents’ home is not associated with perceptions of trust in institutions, except for trust in international organisations, which is likely to be driven by other factors.
1. Reuters Institute and University of Oxford, Digital News Report, https://www.digitalnewsreport.org/interactive-2018/.
2. Datafolha Institute, https://datafolha.folha.uol.com.br/
The revised version of the questionnaire could serve as a first step, paving the way to extending the trust survey to other Latin American countries. Since the 2008 global financial crisis, public trust has continued to fall in Latin America, and the political scene in the region has been marked by numerous demonstrations and social unrest. In 2022, less than four in ten (39%) Latin Americans trusted their government and close to eight in ten (77%) believed that a few powerful groups governed their country for their own benefit (World Gallup Poll 2022; Latinobarometer 2020). Extending the OECD Trust Survey in the region would provide actionable evidence to inform and sharpen national and regional dialogues about how to enhance trust and advance on the reform agenda for Latin America.
2.2. Results of the OECD Trust Survey in Brazil
2.2.1. Trust in government in Brazil aligns with other Latin American countries but is below the OECD average
Unlike in many other countries, more respondents in Brazil trust the federal than trust the local government and the civil service. According to the OECD Trust Survey, one-quarter of Brazilian respondents (25.9%) reported having high or moderately high trust in their government, while 57.8% reported low or no trust and 11.5% were neutral (Figure 2.1).1 The proportions of people with high or moderately high trust or neutral trust is slightly higher than in Colombia, but below OECD averages. Fewer respondents reported high or moderately high trust in the local government (19.6%) and the civil service (23.6%) in Brazil, both below the corresponding OECD averages. In many OECD countries, local governments and the civil service inspire more confidence than national governments, probably because they are perceived as closer to people. In fact, civil servants are usually characterised as the human face of public institutions (OECD, 2021[21]). At the same time, public opinion experts interviewed for this study highlight that in countries with presidential systems like Brazil, people tend to associate the government with the president. This can mean that he or she may serve as the reference point for people’s evaluations of public institutions and explain the higher levels of trust in the federal government. The comparatively low trust in civil servants may reflect cultural bias and negative prejudices –burophobia — towards public officials (Güemes, 2016[22]), as well as perceptions of a lack of integrity of civil servants, corroborated by some cases of corruption (IPEA, 2019[23]).
Figure 2.1. Unusually, Brazilian respondents are more likely to trust the federal government than local government or the civil service
Note: Figure presents responses to the question “On a scale of 0 to 10, where 0 is not at all and 10 is completely, how much do you trust the national government, local government, civil service?”. “High or moderately high trust” corresponds to responses of 6-10; “neutral” to a response of 5; “low or no trust” to responses of 0-4; “don't know” was a separate answer choice. OECD (22) refers to the unweighted average across 22 OECD countries.
Source: OECD Trust Survey (http://oe.cd/trust).
Trust is influenced by a variety of factors, some linked to the context of when the survey took place, others reflecting economic, social and political outcomes of governance, as well as underlying cultural or societal factors. The OECD Trust Survey was implemented in April 2022 in Brazil, at a time when the COVID-19 pandemic was relatively controlled and vaccination rates in the country were comparatively high. However, people’s perceptions of public institutions at the time were likely to be affected by rising inflation, as well as by national elections, and the presidential campaign and high levels of political polarisation which resulted from it. In particular, electoral alliances were being formalised, one of the candidates with the highest popular support among voters had decided not to compete and tensions were running high between the judiciary and executive branches over electoral rules and political campaigns.
Historical data beyond the immediate time when the OECD Trust Survey took place show that levels of trust in the federal government in Brazil have fluctuated widely in the past three decades (Figure 2.2). Patterns of public trust in Brazil overall coincide with those in Latin America, but with peaks at election times (e.g. 2003, 2006 and 2010). In general, trust in government in Brazil increased between 2003 and 2010, during the period of economic growth, the commodity boom, and widespread expansion of social protection programmes. A sharp decline between 2010 and 2018 coincided with regional changes in the political orientation of governments (the end of the “pink tide”, as the period of prevalence of left-wing governments was known) and also a period of political instability in Brazil, including corruption scandals, the Lava Jato investigation (see Chapter 4) and Dilma Roussef’s impeachment in 2016, coupled with the end of the commodity boom. Trust levels increased in 2020, although they have not yet recovered to the levels of previous decades.
Figure 2.2. Trust in government in Brazil has mostly tracked regional levels, with deviations coinciding with key political events
Note: Figure presents responses to the survey question ‘’ Please look at this card and tell me how much trust you have in each of the following groups/institutions. Would you say you have a lot, some, a little or no trust in? The National Government’’. Percentages show the aggregated responses of “a lot’’ and ‘’some’’ trust in the national government.
Source: Latinobarometer, 1995-2020.
2.2.2. Levels of trust vary across institutions but are generally low in Brazil
Although in general Brazilian respondents have low confidence in their institutions and other people, their responses vary widely when asked how much they trust different public institutions. On a scale of 0-10, with zero meaning no trust at all and ten complete trust, average trust levels are below three for all the institutions considered, except the police and other people.
In line with patterns of trust observed in other countries surveyed through the OECD Trust Survey, the police are the most trusted institution in Brazil (39.9%) and political parties (7.6%) the least trusted (Figure 2.3). Political institutions, such as the Congress, and government ones, such as the civil service, tend to be less trusted than law and order institutions (OECD, 2022[1]). Although the aggregated concept of “police” ensures cross-national comparability, it may not fully capture the complexity of law enforcement institutions in Brazil, with multiple police forces with diverse mandates and cultures (see Box 2.4).
Figure 2.3. Even the most trusted institution, the police, are trusted by less than half of Brazilian respondents
Note: Figure presents responses to the question “On a scale of 0 to 10, where 0 is not at all and 10 is completely, how much do you trust the institutions?” “High or moderately high trust” corresponds to responses of 6-10, “neutral” to a response of 5 and “low or no trust” to responses of 0-4; “don't know” was a separate answer choice. OECD (22) refers to the unweighted average across 22 OECD countries.
Source: OECD Trust Survey (http://oe.cd/trust).
Box 2.4. Trust in the police
According to the OECD Trust Survey, on average across OECD countries, as well as in Brazil, the police enjoy the highest levels of public trust among public institutions. These findings may seem somewhat disconnected from the incidents of police brutality and the movements protesting police misconduct reported in recent years across many surveyed countries. According to research, these events should have a strong negative impact on trust in the police, as media coverage is the primary source of information people have on the police (Mawby, 2013[24]). As with the other public institutions included in surveys, responses on trust levels towards the police may reflect a variety of diverse aspects such as personal experiences, performance and evaluations about specific parts of the institution, or individual values and expectations about the role fulfilled by the public institution.
In general, institutional trust is rooted in a broad understanding of the role and nature of the institution rather than personal experiences. Indeed a minority of people tend to have direct experience with the police; for instance, in England and Wales just under 26% of people reported to have some kind of contact with the police from April 2019 to March 2020 (Office of National Statistics, 2022[25]). Moreover, the personal experiences people do have tend to convey little information about police intentions and characteristics.
Data on trust in the police also may not fully capture the complexity of law enforcement institutions in a given country, as some countries have multiple police forces with diverse mandates and cultures, and the aggregated concept of “police” does not allow respondents to provide different evaluations. For example, in Brazil, there are multiple types of police with different mandates, responsibilities and coverage -i.e. the state "military police" forces are tasked with public order, while "civil police" are responsible solely for crime investigation-. However, data from the OECD Trust Survey, as with most existing cross-country surveys, do not differentiate between types of police.
The comparatively high levels of trust in the police may also result from the inherent feature of constitutional democracies where the political elements of government are separated from those providing public goods, with the latter serving public purposes like security and welfare. Unlike more political institutions, police forces benefit from broad agreement within the public about their commitment and operate over a sustained period of time (Warren, 2017[26])
Levels of trust in the institution are therefore most likely to change due to long-term processes or major events, and will be relatively immune to short-term factors, although once a change of direction is underway, it might be difficult to halt or reverse (Bradford and Jackson, 2010[27]). Factors that may undermine institutional trust include single incidents of malpractice, perceived changes in levels of police visibility, perceived declines in availability and readiness to intervene, and increasingly widespread ideas that the police do not treat everyone equally (Bradford and Jackson, 2010[27]).
Finally, it is worth highlighting that although trust in the police is generally high, it tends to be unevenly distributed, and may thus obscure important variations across socioeconomic status, region of origin, or racial and ethnic background (Van Craen and Skogan, 2014[28]; Tyler, 2005[29]). These variations may be context-specific and tied to the country or region’s unique history and political background. For instance, Caicedo’s (2021[30]) research found that Latin Americans' trust in the police is affected by their class, opinion on governance, and perception of corruption, with lower trust among indigenous people. In Brazil, people who do not identify themselves as part of the white population report lower levels of satisfaction with the police (de Oliveira Junior, 2011[31]).
Results from the OECD Trust Survey echo findings from other similar surveys previously implemented in the country, and find that Brazil is in line with general trends in the region (Figure 2.4). The most recent data available from Latinobarometer (2020) show that, on average across Latin American countries, the police are the most trusted institution while institutions of a political nature, such as political parties or the national congress are the least trusted. The exceptions to this regional trend are Bolivia, the Dominican Republic, Guatemala and Venezuela. Results in these countries could be explained by factors including changes at the political regime level, that is, those that involve changes in electoral systems, party systems, fundamental rules or the political community beyond governments (Mainwaring and Pérez-Liñán, 2023[32]; Corrales, 2023[33]). Uruguay is the only LAC country in which more than three in ten respondents reported trusting political parties or the national congress. The country has the highest voter turnout levels in the region and one of the most stable party systems (Piñeiro Rodríguez and Scrollini Mendez, 2019[34]).
Figure 2.4. Trust varies across institutions in LAC
Note: Figure presents responses to the survey question ‘’Please look at this card and tell me how much trust you have in each of the following groups/institutions. Would you say you have a lot, some, a little or no trust in?’’. Percentages show the aggregated responses of ‘’a lot’’ and ‘’some’’ trust in selected institutions. Data for trust in the National Congress only available for 2017.
Source: Latinobarometer, 2020.
According to the OECD Trust Survey, the third most trusted institution in Brazil was the news media. About one third of Brazilian respondents (28.2%) reported trusting the media, a higher share than for the courts and the legal system, or the Congress (Figure 2.3). These proportions are similar to those in Colombia, while in contrast, on average across OECD countries, respondents are highly sceptical about the media (the second least trusted of the nine institutions measured). This result for the news media supports academic research highlighting how the media have played a key role in the cultural and political landscape of Brazil over the last two decades (Porto, 2009[35]). This dynamic nevertheless carries some risks, most notably the possible “spectacularisation of politics” (Edelman, 1991[36]).
More than half of Brazilian respondents (56.5%) reported regularly using social media as a source of information on politics, 11 percentage points higher than the average across OECD countries (45%). While 48.8% of Brazilians considered “having the source of news cited” as a key criterion for judging the trustworthiness of information and 41.3% "knowing the publishing organisation of the story”, respondents using social media are less likely to consider either of these as key to trustworthiness than those reading newspapers (Figure 2.5). In contrast, users of social media are more likely than newspaper readers to value whether the “news has been widely commented on” as a criterion for trustworthiness. These results may indicate that social media users may be less willing to verify information or be able to detect disinformation (Allcott et al., 2020[37]; Smith, 2019[38]). Recent guidelines to regulate content of social media platforms could facilitate the spread of mis- and disinformation, as well as polarising content (OECD, 2023[39]).
Figure 2.5. Newspapers readers and social media users rely on slightly different criteria for gauging trustworthiness
Note: Figure presents responses to the survey question ‘When you read, see or hear news, what are the three most important aspects for you to decide if the news is credible?’’. Presents all answer options and share of respondents who mentioned each of the six possible responses. Figure is disaggregated by respondents’ use of newspapers and social media, based on the survey question ‘’ Which of the following media do you use to obtain information about politics and current affairs at least once a week?’’.
Source: OECD Trust Survey (http://oe.cd/trust).
Around one-third of Brazilian respondents (34.6%) said they trusted other people in 2022, the item with the second highest level of reported trust in the country (Figure 2.3). This result may be interpreted with a degree of optimism as a positive outcome of initiatives to promote solidarity and a sense of community that were launched during the COVID-19 pandemic. Historically trust in others in Brazil has been lower than in most Latin American countries (see Figure 1.2 in Chapter 1).
2.2.3. More vulnerable groups –those with lower income, less educated and younger- report lower levels of trust
Trust levels vary between population groups, and trust divides appear to replicate social and economic inequalities. According to the results of the OECD Trust Survey in Brazil, young people, respondents with low levels of education, and those living on low incomes report lower levels of trust than other groups in the country. As these vulnerabilities tend to be combined together (Camargo and de Hollanda Guimarães Ferreira, 2000[40]), when they shape the structural political attitudes and behaviour of those who are (or feel) excluded and marginalised, this risks eroding social cohesion and citizenship (OECD, 2021[6]).
Brazilian respondents with greater purchasing power (those included in “Group A” according to Critério Brasil) trust the federal and local governments and civil service more than those in lower groups. More than one-third of people (36.1%) with the highest purchasing power (upper class) reported high levels of trust in the federal government, compared to 22.6% of those with the lowest purchasing power (Figure 2.6). Similarly, there is a trust gap of over ten percentage points by purchasing power for local government and of 20 percentage points for the civil service.
Figure 2.6. Trust is higher among people belonging to a higher social class
Note: Figure presents the responses to the question “On a scale of 0 to 10, where 0 is not at all and 10 is completely, how much do you trust the institutions?”. “High or moderately high trust” corresponds to responses of 6-10. “BRA mean” presents the weighted average across all respondents in Brazil. Socioeconomic status (SES) is based on household's purchasing power following the Brazil Economic Classification Criteria (Critério Brasil-ABEP), which assesses respondents households’ access to public utility services, education and possession of several amenities. * means that differences in proportions are statistically significant at the 90% significance level; ** means that differences are statistically significant at the 95% level; *** means that differences are statistically significant at the 99% level. Reference group in light blue.
Source: OECD Trust Survey (http://oe.cd/trust).
Similarly, trust in public institutions also varies according to educational background (Figure 2.7). In Brazil, trust in government among those with a university degree is 9 percentage points higher than among those with below upper secondary education. However, gaps related to differences in education levels are relatively small, compared to those related to purchasing power, with the striking exception of a difference of over 20 percentage points in trust in the civil service.
Figure 2.7. People with low levels of education tend to trust public institutions less
Note: Figure presents the responses to the question “On a scale of 0 to 10, where 0 is not at all and 10 is completely, how much do you trust the institutions?” “High or moderately high trust” corresponds to responses of 6-10. BRA mean presents the weighted average across all respondents in Brazil. High education is defined as ISCED 2011 levels 5-8, i.e. university-level degrees such as bachelor’s, master’s or doctorates, and low education refers to below upper secondary attainment. * means that differences in proportions are statistically significant at the 90% significance level; ** means that differences are statistically significant at the 95% level; *** means that differences are statistically significant at the 99% level. Reference group in light blue.
Source: OECD Trust Survey (http://oe.cd/trust).
The fact that differences in level of trust in the civil service are particularly pronounced – both by purchasing power and education – could be linked to a widespread perception of a lack of fairness (see Chapter 4), and the pervasive impact of unequal treatment of the most vulnerable Brazilians.
Finally, there are also significant trust differences by age consistent with findings from OECD countries. In Brazil, there is a 12 percentage point gap in trust in government between the younger and older cohorts (Figure 2.8). This could relate to the fact that globally, young adults have higher unemployment rates and that, according to recent research, 20-24 year-olds were the population group in Brazil who experienced the highest income loss during 2014-2019 (-17.76% compared to a national average of -3.71%) (Neri, 2019[41]). Young people were also disproportionally hit by the economic consequences of the pandemic: in 2021, job losses were concentrated globally among young and temporary workers (OECD, 2021[42]), which could have increased their concerns about future economic prospects. However, an important and maybe optimistic difference to highlight is that young Brazilians report significant higher levels of trust in the Congress and political parties than older cohorts. Younger generations who have grown up in a democratic system may have more positive perceptions about politics and political institutions. In any case, this result can be an opportunity to further invest in policies that aim to strengthen democratic values and civic education in schools.
Figure 2.8. Trust in public institutions differs markedly by age
Note: Figure presents the responses to the question “On a scale of 0 to 10, where 0 is not at all and 10 is completely, how much do you trust the institutions?” “High or moderately high trust” corresponds to responses of 6-10. “BRA mean” presents the weighted average across all respondents in Brazil. * means that differences in proportions are statistically significant at the 90% significance level; ** means that differences are statistically significant at the 95% level; *** means that differences are statistically significant at the 99% level. Reference group in light blue.
Source: OECD Trust Survey (http://oe.cd/trust).
2.2.4. Economically, politically or physically insecure respondents are less trusting of public institutions
In addition to objective socio-economic indicators, people’s own perceptions of their vulnerability also matter for trust in government. Brazilian people who feel economically vulnerable report lower levels of trust across public institutions. Among respondents who reported being concerned about their household’s finances, around two in ten trust their government, compared to almost four in ten of those who are less concerned about their financial situation, a wider gap than for differences according to income or social class. This is of key relevance given recent academic research that relates economic expectations to political behaviour. According to a study using German household panel data, adults who have not reached the socio-economic status that might have been expected given their childhood background tend to vote less frequently or, if they do vote, do so more frequently for radical parties than for mainstream ones (Kurer and van Staalduinen, 2022[43]).
In a similar vein, those who feel powerless and without a say in government decision making are also understandably less trusting of public institutions. More than half of the respondents (53%) who believe they have a say in what the government does trust the federal government, compared to only 18.5% of those who do not believe they have a say (Figure 2.9). This 34-percentage point gap is one of the largest trust divides in Brazil. Perceived political inequalities seem to have as pervasive an impact on trust as economic inequalities. Respondents who were not politically aligned with the government in power also reported lower levels of trust in government, a finding which is consistent with results in other OECD countries (OECD, 2022[1]). Yet, people who reported they were politically aligned with the government consistently reported lower levels of trust in other public institutions (Figure 2.10), in contrast to results in other Latin American and OECD countries. These findings could suggest political alignment with the government in power was related to negative attitudes towards other public institutions which might exercise checks and balances on any potential abuses of power by the government.
Figure 2.9. People who feel they have a say in what the government does trust all institutions more
Note: Figure presents the responses to the question “On a scale of 0 to 10, where 0 is not at all and 10 is completely, how much do you trust the institutions?” “High or moderately high trust” corresponds to responses of 6-10. Trust responses are disaggregated by perceptions of having a say (6-10) and not having a say (0-4), based on the survey question “How much would you say the political system in Brazil allows people like you to have a say in what the government does?” Differences in proportions are statistically significant at the 95% significance level for trust in the police; and are statistically significant at the 99% level for trust in the rest of the groups and institutions included in the figure.
Source: OECD Trust Survey (http://oe.cd/trust).
Figure 2.10. Levels of trust vary between those who see themselves as politically aligned with the government and those who do not
Note: Figure presents the responses to the question “On a scale of 0 to 10, where 0 is not at all and 10 is completely, how much do you trust the institutions?” “High or moderately high trust” corresponds to responses of 6-10. Trust responses are disaggregated by whether respondents are politically aligned with the government, based on giving the same response options to the survey questions “In politics, people usually talk about "left" and "right". On a scale where 0 is left and 10 is right, where would you place the national government?” and “In politics, people usually talk about "left" and "right". On a scale where 0 is left and 10 is right, where would you place yourself?”. * means that differences in proportions are statistically significant at the 90% significance level; ** means that differences are statistically significant at the 95% level; *** means that differences are statistically significant at the 99% level.
Source: OECD Trust Survey (http://oe.cd/trust).
Finally, perceptions of insecurity matter for trust in public institutions, hindering long-term planning and increasing social barriers for the most vulnerable. People who are less concerned with “being attacked or assaulted in the neighbourhood they live in” have more trust in national and local government and the civil service (Figure 2.11). The gap between the two groups is widest for trust in the local government: only 15.6% of respondents with security concerns say they trust the local government, compared to 25.5% of those who feel safe in their neighbourhood.
Figure 2.11. Those who feel safer tend to have higher levels of trust in public institutions
Note: Figure presents the responses to the question “On a scale of 0 to 10, where 0 is not at all and 10 is completely, how much do you trust the institutions?” “High or moderately high trust” corresponds to responses of 6-10. Trust responses are disaggregated by whether respondents feel insecure or safe, based on the survey question “Speaking of the neighbourhood where you live and thinking about the possibility of being assaulted or robbed, do you feel very safe, a little safe, a little insecure or very insecure?” “BRA mean” presents the weighted average across all respondents in Brazil. * means that differences in proportions are statistically significant at the 90% significance level; ** means that differences are statistically significant at the 95% level; *** means that differences are statistically significant at the 99% level. Reference group in light blue.
Source: OECD Trust Survey (http://oe.cd/trust).
2.3. The drivers of trust in public institutions in Brazil
This section focuses on how public governance affects public trust in the federal government, local government and the civil service in Brazil. Using observational data to establish a causal relationship between public governance drivers and trust is complicated. Causal relationships between drivers and trust are likely to move in both directions, and many variables may not be observable. Notwithstanding this caveat, this section presents the results of an aggregate analysis on the drivers of trust to measure which aspects of people’s perception of government reliability, responsiveness, integrity, openness and fairness have a significant effect on their trust levels (Box 2.5). Therefore, the results in this section provide a general indication of which policy areas would produce the greatest dividend in terms of trust in the federal government, local government or the civil service. This section also provides an assessment of the efforts required to achieve these gains, based on how Brazilian respondents judge the institutional performance tied to the various public governance drivers.
Based on the results of the trust survey and interviews with national stakeholders carried out for this study, Chapters 3 and 4 go on to provide detailed evidence on the relationship between trust and people’s perceptions of government reliability and responsiveness (Chapter 3) and of its integrity, openness and fairness (Chapter 4) and discuss actions that the country can take to enhance trust.
Box 2.5. Notes on the model for the drivers of trust in public institutions
The results presented in this section are based on three logistic regression analyses for establishing the main drivers of trust in the federal government, local government, and civil service in Brazil.
Based on the OECD Framework on the Drivers of Trust, respondents’ perceptions of the responsiveness, reliability, openness, integrity and fairness of government and public institutions are expected to be the main drivers of trust in the three institutions (federal government, local government and the civil service). Trust in each of the institutions is recoded as a binary variable (low or no trust: 0-4 and high or moderately high trust: 6-10). Neutral responses (5) and “don’t know” are excluded. The analysis operationalises government competencies and values through 15 variables, measured on a 0-10 response scale and standardised for the analysis. The model also includes six further variables: internal political efficacy, external political efficacy, satisfaction with administrative services, the perceived relevance of knowing a broker to access public services, willingness to formalise any new business and confidence in the country’s ability to tackle environmental challenges.
All the drivers of trust are included in the three baseline regression models and those which are not statistically significant are deleted (stepwise deletion process). The results in this section show all the significant drivers of trust in the three institutions.
All models include survey weights and control variables for individuals’ socio-demographic characteristics (age, gender, education, social class and region of residence), interpersonal trust, perceptions of economic and physical insecurity, and whether the respondent voted for the parties in government. Missing data are excluded using listwise deletion.
Technical interpretation
The statistically significant drivers are shown as average marginal means. The technical interpretation of the effect of government effectiveness to reduce greenhouse emissions on trust, for example, is that one standard deviation increase in the perceived likelihood that the government is effective is associated with a 10 percentage point increase in trust in the federal government. Or – taking into consideration all other variables in the model – all else being constant, moving from the typical citizen to one who is one-standard-deviation more satisfied, results in a 10 percentage point increase in trust in the federal government in Brazil.
The analysis of the drivers of trust in Brazil suggests four main results:
First, the determinants of trust differ for different institutions, suggesting different policies might be needed to build or strengthen trust in the federal government, the local government or the civil service.
Second, government fairness – either expressed as people’s perceptions of fair treatment of their applications to social benefits or about civil servants treating them equally regardless of their income – is a key driver of trust in both the federal government and the civil service. This result underscores the relevance of policies and initiatives aiming at tackling inequalities in the country, not only in economic terms, but also ensuring equal access to services, as a first and basic function of the state.
Third, and linked to the previous finding, satisfaction with administrative services is a significant driver of trust in the local government and the civil service. This supports the argument that this is an area where improvements are needed to enhance trust in Brazil.
Finally, perceptions about government openness and participation, especially the feeling that one’s own inputs are considered, or confidence in one’s own ability to participate in politics, are crucial drivers of trust in all public institutions. This suggests that strengthening initiatives to make people feel heard and ensuring they can voice their views could significantly improve levels of trust in Brazil.
2.3.1. Trust in the federal government
The main drivers of trust in the federal government in Brazil are perceptions related to its reliability, fairness and openness. The matrix in Figure 2.12 shows the main drivers of trust in government and how people perceive them. Points closest to the green top right corner highlights drivers with possible high trust gains, that would require less investment, as the starting points of how people currently perceive these drivers are already higher than the others.
Brazilians’ confidence in their government’s ability to reduce greenhouse gas emissions (reliability) has the largest effect on trust in the federal government (10 percentage points). Despite many ongoing initiatives in this area, few people (39.6%) are confident in the government’s effectiveness at addressing such long-term challenges as climate change. There is a great deal of room for improvement in this area that would yield large increases in trust. Improving public perceptions of the country’s capacity to tackle climate change could also include a renewed public communication strategy (see Chapters 3 and 4).
An increase in people’s confidence that they would be fairly treated when applying for social services and benefits would similarly have a large effect on trust in federal government (6 percentage points). Currently, around one-third of Brazilian respondents reported being satisfied with the fairness of treatment in this area; thus this is another area where investment would bring a large pay-off in terms of trust.
An increase in respondents’ confidence that their views would be considered in public consultations would have a similar effect on trust in government (4.7 percentage points). However, very few people currently feel heard in such areas (21.5%) and investment in making them feel listened to would require considerable effort to yield the potential trust gains.
People’s confidence in business conditions remaining stable (reliability), and their willingness to formalise any new business (reliability), as well as confidence in their own ability to participate in politics (internal efficacy) are the other three relevant drivers of trust in the federal government in Brazil. Around one-quarter of Brazilian respondents perceive business conditions as predictable – with a potential trust gain of 4.4 percentage points – and more than half are willing to formalise any new business (potential gain of 4.1 percentage points). More than one-third of Brazilian respondents are confident in their own ability to participate in politics, and moving from the typical respondent to one slightly more confident could increase trust in government by 2.8 percentage points. Levels of internal efficacy are closer to OECD average than all other “starting points”. This result may suggest that providing people with information about political process, allowing them to better understand how politics works, could be crucial to increasing trust.
Figure 2.12. Perceptions of effectiveness in reducing greenhouse emissions and fairness are the most important determinants of trust in the federal government
Note: The figure shows the most robust determinants of self-reported trust in federal government in a logistic estimation that controls for individual characteristics and self-reported levels of interpersonal trust. All variables depicted are statistically significant at 99%. Only questions derived from the OECD Trust Framework are depicted, while individual characteristics such as age, gender, education, which also may be statistically significant, are not shown. The X-axis shows the share of the population that answered the question positively (6-10 on a 0-10 response scale).
Source: OECD Trust Survey (http://oe.cd/trust).
2.3.2. Trust in the civil service
Perceptions that public institutions treat people fairly, and are open and willing to receive inputs from the public are the main drivers of trust in the civil service in Brazil. Improving people’s perception that civil servants would treat people equally regardless their income (fairness) would yield a 7.4 percentage point gain in trust in the civil service, and there is broad room for improvement in this area, as very few Brazilians (19.1%) perceive institutions as fair (Figure 2.13). People’s confidence that their inputs would be taken into account in public consultations also shows a significant impact (4.9 percentage points). This is another area where Brazilian institutions are lagging, with only 2 in 10 respondents feeling heard. These results point to the importance of greater investment in making people feel they are not left behind (whether economically or politically) if aiming to enhance trust in the civil service.
Two other drivers of trust in the civil service are also related to openness and participation: ease of access to information on administrative procedures and people’s confidence in their ability to participate in politics. Improving perceptions about how easy it is to access administrative information is associated with an increase of 3.5 percentage points in levels of trust in the civil service. Although the starting point on this measure is higher than for the other drivers discussed above (34.9%), it is low compared to other Latin American countries, despite Brazil’s multiple initiatives to promote transparency (see Chapter 4). People’s confidence in their own ability to participate in politics has a similar starting point and would yield a similar trust gain in the civil service (3.7 percentage points).
A final driver of trust in the civil service in Brazil, not surprisingly, is people’s satisfaction with administrative services. About one-third of Brazilians are satisfied with administrative services in the country (they expressed levels of satisfaction with administrative services between 6-10 along the 0-10 scale), and further investment in increasing citizens’ satisfaction would yield a potential trust gain of 3.7 percentage points. Although it seems intuitive that satisfaction with administrative services would be an important driver of trust in the civil service, the size of this impact is relatively small compared to other drivers, as well as compared to average results for OECD countries (OECD, 2022[1]). The results of the econometric analysis suggest that investing in activities related to the underlying values guiding actions and procedures would be of greater relevance to improving trust in the civil service than those related to actual results or outcomes by themselves.
Figure 2.13. Perceptions of equal treatment and openness are key to explaining trust in the civil service
Note: The figure shows the most robust determinants of self-reported trust in civil service in a logistic estimation that controls for individual characteristics and self-reported levels of interpersonal trust. All variables depicted are statistically significant at 99%. Only questions derived from the OECD Trust Framework are depicted, while individual characteristics such as age, gender, education, which also may be statistically significant, are not shown. The x-axis shows the share of the population that answered the question positively (6-10 on a 0-10 response scale).
Source: OECD Trust Survey (http://oe.cd/trust).
2.3.3. Trust in the local government
In contrast to the results for the federal government and the civil service, competence plays more of a role in explaining trust in local government in Brazil. Apart from perceptions about opportunities to voice views in community decisions, which is the most significant determinant of trust in local government, all other significant drivers are related to the reliability and responsiveness of government (Figure 2.14). Improving people’s perceptions about being able to have a say in community decisions would improve trust in local government by 5.7 percentage points, but this improvement would require considerable effort given Brazilian respondents’ low perceptions on this aspect (see Chapter 4).
Satisfaction with services is the next most important driver of trust in the local government in Brazil, which could be related to the federal structure of the country. Local governments are responsible for education, healthcare, policing and a myriad of other services. An increase in levels of satisfaction with administrative services could yield a trust gain of 5.4 percentage points. Similarly, investment in people’s confidence that innovative ideas would be adopted to improve service delivery would also have returns on trust in local government (4.5 percentage points). However, fewer Brazilian respondents are currently confident about capacity for innovation in public services (27.9%) than are satisfied with the delivery of administrative services (31.8%). Therefore, greater investment in improving innovative capacities may be required to yield potential trust gains.
The final relevant driver of trust in the local government is the government’s preparedness to fight future infectious diseases (reliability) with a potential trust gain of 4 percentage points. This is a very topical point in the context of the pandemic, and the potential for future crises. Considering the comparatively low levels of satisfaction on this aspect – only one in five respondents perceive the government as prepared – Brazil would need to invest considerably in preparedness for to affect trust levels (see Chapter 3).
Figure 2.14. Openness and satisfaction with administrative services are the main drivers of trust in the local government
Note: The figure shows the most robust determinants of self-reported trust in local government in a logistic estimation that controls for individual characteristics and self-reported levels of interpersonal trust. All variables depicted are statistically significant at 99%. Only questions derived from the OECD Trust Framework are depicted, while individual characteristics such as age, gender, education, which also may be statistically significant, are not shown. The x-axis shows the share of the population that answered the question positively (6-10 on a 0-10 response scale).
Source: OECD Trust Survey (http://oe.cd/trust).
2.3.4. Comparative analysis of the drivers of trust in public institutions in Brazil
Figure 2.15 shows average responses of people expressing confidence in government’s responsiveness, reliability, openness, integrity and fairness.
Figure 2.15. Perceptions of public responsiveness are the highest among Brazilians, and integrity the lowest
Note: Figure presents Brazil’s average of "likely" responses across questions related to reliability, responsiveness, integrity, openness and fairness (see OECD Trust Framework in Chapter 1). The questions used to assess these five dimensions are included in Chapter 3 (responsiveness and reliability) and Chapter 4 (openness, integrity and fairness); the whole questionnaire can be found in the Annex.
Source: OECD Trust Survey (http://oe.cd/trust).
Table 2.3 summarises the results of the analysis on the drivers of trust in the national government, the local government and the civil service in Brazil. Yellow cells indicate that the variable is a statistically significant driver of trust in the corresponding institution, while a positive sign (+) in the cell flags which institution has the largest coefficient of the three for that variable.
Table 2.3. Drivers of institutional trust in Brazil
Aspects related to openness and participation, especially meaningful engagement, and internal efficacy have a slightly higher relative effect on trust in the civil service than on trust in government. In contrast, satisfaction with administrative services has almost twice the coefficient for trust in the local government than for the civil service, which may be linked to institutional settings and the federal structure.
When considering specific personal characteristics that were used as control variables, only interpersonal trust is consistently significant across all institutions, showing a high correlation between public and interpersonal trust. These findings are in line with previous research (Keefer and Scartascini, 2022[44]) and further highlight the need to set trust as an explicit policy objective. Political alignment with the government only shows significance for trust in federal government. In addition, age – in particular being older – also has greater explanatory power for trust in federal government.
Finally, holding a university degree is significant only for trust in the civil service, while perceptions of insecurity are only significant for trust in the local government. This last finding could be again linked to the different functions and responsibilities allocated to different levels of government.
2.3.5. Opportunities to upgrade the measurement of trust to build a robust evidence base
This section summarises key results and presents guidance on how Brazil could upgrade how it measures trust and build a robust evidence base.
The OECD Trust Survey finds that perceptions of reliability, fairness and openness are significant drivers of people trust in the federal government, civil service and local governments. As part of its mandate to perform internal control, the Office of the Comptroller General (Controladoria-Geral da União, CGU) could:
Regularly collecting data on the drivers of trust to monitor the evolution of trust, and to identify where to invest to preserve and strengthen trust in public institutions.
People with lower incomes and lower levels of education, young people, and those with security concerns have consistently lower levels of trust in institutions.
Ensuring survey samples are representative of the diversity of the population, by combining socio-economic characteristics. This would allow systematic analysis of the differences within and across populations groups.
References
[37] Allcott, H. et al. (2020), “The Welfare Effects of Social Media”, American Economic Review, Vol. 110/3, pp. 629-676, https://doi.org/10.1257/aer.20190658.
[15] Almond, G. and S. Verba (1963), The Civic Culture: Political Attitudes and Democracy in Five Nations., Princeton University Pres, Princeton, New Jersey.
[14] Avritzer, L. (2017), “Participation in democratic Brazil: from popular hegemony and innovation to middle-class protest”, Opinião Pública, Vol. 23/1, pp. 43-59, https://doi.org/10.1590/1807-0191201723143.
[27] Bradford, B. and J. Jackson (2010), “Trust and Confidence in the Police: A Conceptual Review”, SSRN Electronic Journal, https://doi.org/10.2139/SSRN.1684508.
[4] Brezzi, M. et al. (2021), “An updated OECD framework on drivers of trust in public institutions to meet current and future challenges”, OECD Working Papers on Public Governance, No. 48, OECD, Paris, https://doi.org/10.1787/b6c5478c-en.
[30] Caicedo, J. (2021), “Autoridad y privilegio: confianza en la policía en Latinoamérica”, https://doi.org/10.7440/colombiaint110.2022.06.
[40] Camargo, J. and F. de Hollanda Guimarães Ferreira (2000), “The poverty reduction strategy of the government of Brazil: A rapid appraisal”, No. 417, Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio), Departamento de Economia.
[33] Corrales, J. (2023), Autocracy Rising. How Venezuela Transitioned to Authoritarianism, Brookings.
[31] de Oliveira Junior, A. (2011), “Dá para confiar nas polícias? Confiança e percepção social da polícia no Brasil”, Revista Brasileira de Segurança Pública, Vol. 5/9.
[20] Delhey, J. and K. Newton (2003), “Who trusts?: The origins of social trust in seven societies”, European Societies, Vol. 5/2, pp. 93-137, https://doi.org/10.1080/1461669032000072256.
[17] Duque, D. and A. Smith (2019), “The Establishment Upside Down: A Year of Change in Brazil”, Revista de ciencia política (Santiago), Vol. 39/2, pp. 165-189, https://doi.org/10.4067/s0718-090x2019000200165.
[36] Edelman, M. (1991), La construcción del espectáculo político, Manantial, Buenos Aires.
[22] Güemes, C. (2016), “Confianza”, Eunomía. Revista en Cultura de la Legalidad, Vol. 10, pp. 132-143.
[23] IPEA (2019), Implementando desigualdades : reprodução Desigualdades na implementação de políticas públicas, https://repositorio.ipea.gov.br/bitstream/11058/9323/1/Implementando%20desigualdades_reprodu%C3%A7%C3%A3o%20de%20desigualdades%20na%20implementa%C3%A7%C3%A3o%20de%20pol%C3%ADticas%20p%C3%BAblicas.pdf (accessed on 20 February 2023).
[44] Keefer, P. and C. Scartascini (eds.) (2022), Trust: The Key to Social Cohesion and Growth in Latin America and the Caribbean (Executive Summary), Inter-American Development Bank, https://doi.org/10.18235/0003911.
[11] Koster, M. and F. Eiró (2021), “Clientelism in Northeast Brazil: brokerage within and outside electoral times”, Contemporary Social Science, Vol. 17/3, pp. 222-234, https://doi.org/10.1080/21582041.2021.1876244.
[43] Kurer, T. and B. van Staalduinen (2022), “Disappointed Expectations: Downward Mobility and Electoral Change”, American Political Science Review, Vol. 116/4, pp. 1340-1356, https://doi.org/10.1017/s0003055422000077.
[32] Mainwaring, S. and A. Pérez-Liñán (2023), “Why Latin America’s Democracies Are Stuck”, Journal of Democracy, Vol. 34/1, pp. 156-170, https://doi.org/10.1353/jod.2023.0010.
[24] Mawby, R. (2013), Policing Images, Willan, https://doi.org/10.4324/9781843924661.
[12] Medard, J. (1976), “Le rapport de clientèle : du phénomène social à l’analyse politique”, Revue française de science politique, Vol. 26/1, pp. 103-131.
[19] Muggah, R. and K. Aguirre Tobón (2018), Citizen Security in Latin America: Facts and Figures. Strategic Paper 33, IGARAPÉ INSTITUTE, Rio de Janeiro.
[3] Murtin, F. et al. (2018), ““Trust and its determinants: Evidence from the Trustlab experiment””, OECD Statistics Working Papers, No. 2018/02, OECD, Paris.
[41] Neri, M. (2019), A escalada da Desigualdade: Qual foi o impacto da crise sobre a distribuicao de renda e a pobreza?, FGV Social, Rio de Janeiro.
[9] Nguyen, D. et al. (2022), “Survey design and technical documentation supporting the 2021 OECD Survey on Drivers of Trust in Government Institutions”, OECD Working Papers on Public Governance, No. 53, OECD Publishing, Paris, https://doi.org/10.1787/6f6093c5-en.
[8] OECD (2023), Drivers of Trust in Public Institutions in New Zealand, Building Trust in Public Institutions, OECD Publishing, Paris, https://doi.org/10.1787/948accf8-en.
[39] OECD (2023), “Good practice principles for public communication responses to mis- and disinformation”, OECD Public Governance Policy Papers, No. 30, OECD Publishing, Paris, https://doi.org/10.1787/6d141b44-en.
[1] OECD (2022), Building Trust to Reinforce Democracy: Main Findings from the 2021 OECD Survey on Drivers of Trust in Public Institutions, Building Trust in Public Institutions, OECD Publishing, Paris, https://doi.org/10.1787/b407f99c-en.
[7] OECD (2022), Drivers of Trust in Public Institutions in Norway, Building Trust in Public Institutions, OECD Publishing, Paris, https://doi.org/10.1787/81b01318-en.
[6] OECD (2021), Drivers of Trust in Public Institutions in Finland, Building Trust in Public Institutions, OECD Publishing, Paris, https://doi.org/10.1787/52600c9e-en.
[42] OECD (2021), OECD Economic Surveys: Portugal 2021, OECD Publishing, Paris, https://doi.org/10.1787/13b842d6-en.
[21] OECD (2021), Public Employment and Management 2021: The Future of the Public Service, OECD Publishing, Paris, https://doi.org/10.1787/938f0d65-en.
[2] OECD (2017), OECD Guidelines on Measuring Trust, OECD Publishing, Paris, https://doi.org/10.1787/9789264278219-en.
[13] OECD et al. (2022), Latin American Economic Outlook 2022: Towards a Green and Just Transition, OECD Publishing, Paris, https://doi.org/10.1787/3d5554fc-en.
[5] OECD/KDI (2018), Understanding the Drivers of Trust in Government Institutions in Korea, Building Trust in Public Institutions, OECD Publishing, Paris, https://doi.org/10.1787/9789264308992-en.
[25] Office of National Statistics (2022), Contact and trust with the police, year ending March 2016 to March 2020 - Office for National Statistics, https://www.ons.gov.uk/peoplepopulationandcommunity/crimeandjustice/adhocs/15340contactandtrustwiththepoliceyearendingmarch2016tomarch2020 (accessed on 24 August 2023).
[34] Piñeiro Rodríguez, R. and F. Scrollini Mendez (2019), Uruguay’s Stability and Change in an Institutionalized Party System, Oxford University Press, https://doi.org/10.1093/acrefore/9780190228637.013.1695.
[35] Porto, M. (2009), “La crisis de confianza en la política y sus instituciones: los medios y la legitimidad de la democracia en Brasil”, América Latina Hoy, Vol. 25, https://doi.org/10.14201/alh.2647.
[16] Prats, M. and A. Meunier (2021), “Political efficacy and participation: An empirical analysis in European countries”, OECD Working Papers on Public Governance, No. 46, OECD Publishing, Paris, https://doi.org/10.1787/4548cad8-en.
[10] Rocha, M., A. Souza and P. Araújo (2019), “Clientelism and Local Politics: Interactions Between Municipal Councilors and Voters in the State of Minas Gerais”, Brazilian Political Science Review, Vol. 13/3, https://doi.org/10.1590/1981-3821201900030002.
[38] Smith, R. (2019), Rage Inside the Machine: The Prejudice of Algorithms, and How to Stop the Internet Making Bigots of Us All, Bloomsbury Business, Bloomsbury, London.
[29] Tyler, T. (2005), “Policing in Black and White: Ethnic Group Differences in Trust and Confidence in the Police”, Vol. 8/3, pp. 322-342, https://doi.org/10.1177/1098611104271105.
[28] Van Craen, M. and W. Skogan (2014), “Differences and similarities in the explanation of ethnic minority groups’ trust in the police”, Vol. 12/3, pp. 300-323, https://doi.org/10.1177/1477370814535375.
[26] Warren, M. (2017), “Trust and Democracy”, in Uslaner, E. (ed.), The Oxford Handbook of Social and Political Trust, Oxford University Press, https://doi.org/10.1093/oxfordhb/9780190274801.001.0001.
[18] We Are Social (2019), “Digital in 2019: Global Internet Use Accelerates”, https://wearesocial.com/uk/blog/2019/01/digital-in-2019-global-internet-use-accelerates/ (accessed on 13 March 2023).
Note
← 1. Results for levels of trust in the federal government from the OECD Trust Survey are slightly below to those reported by Gallup in 2022 (39%). This difference could be explained by the fact that in the World Gallup Poll the wording of the question “Do you trust the government?” is slightly different, and it has a binary answer “yes or no”; while the one included in the OECD Trust Survey offers an eleven-point scale, allowing for a more nuanced response.