Trust in government varies systematically across population groups defined by demographic background, socio-economic status, and political attitudes. Perceived political agency and partisanship most strongly correspond with levels of trust. In this chapter, 'trust gaps' are defined as the difference in the proportion of respondents reporting high or moderately high trust in different public institutions across distinct population groups. In addition to documenting variation in trust between different population groups, the chapter analyses how these trust gaps have evolved over time, paying particular attention to recent developments in partisan and education-related trust gaps.
OECD Survey on Drivers of Trust in Public Institutions 2026 Results
Navigating Rising Expectations and New Horizons
Abstract
Perceived political agency is closely associated with trust in the government. People who feel that the political system allows people like them to have a say in what government does or who are confident in their ability to participate in politics are much more likely to report high or moderately high trust in the national government in OECD countries. However, only around one in three people (31%) believe that people like them have a say in what the government does.
Partisanship influences trust. Across the OECD, trust in the national government is 26 percentage points higher among the self-reported supporters of governing parties. Partisan differences also extend to administrative institutions such as the civil service, and they have increased over time.
Socio-economic characteristics correspond with trust gaps, although to a somewhat smaller extent. Individuals reporting financial concerns are 18 percentage points less likely to trust the national government. A 14-percentage point gap is visible among individuals with different education levels, and for those who identify as part of a discriminated-against group.
Demographic characteristics show relatively smaller trust gaps. Across OECD countries, women are seven percentage points less likely to trust the national government than men. Similarly, young people (18–29) are five percentage points less trusting than older adults (50+).
Patterns in trust gaps are broadly similar across surveyed OECD and OECD accession candidate countries. In both country groups, people’s sense of political voice and partisanship are more strongly associated with trust than socio-economic characteristics or demographic factors.
Trust gaps have remained broadly stable over time. On average across OECD countries, most gaps have changed little across survey waves, although trends vary by country. The most notable changes are observed for age, where the gap between young people (18-29) and older individuals (50+) has narrowed considerably, and for political agency and education, where gaps have widened among countries surveyed in all three waves.
Governments face a dual challenge. They need to continuously improve public governance processes and outcomes to match people’s expectations, which can improve trust in public institutions. But governments must also recognise that such improvements will not necessarily build trust at the same rate across groups. For example, groups from lower socio-economic backgrounds often report lower levels of trust in public institutions and participate less in political processes, making it more difficult to design and implement policies that effectively reach them. In this context, it is critical to understand how trust differs across population groups to design effective policies.
Trust gaps, which are defined as differences in the share with high or moderately high trust in a given institution between different population groups, provide important insight into where disparities in confidence in public institutions are most pronounced. Systematically assessing these gaps help governments identify which groups feel less represented or supported by public institutions and indicates where targeted action could most effectively strengthen trust. Large gaps can indicate problematic trust deficits that could benefit from targeted policy interventions. But even in situations where group-specific interventions are not called for, examining trust gaps can help policy makers better understand whether overall trust levels are shared broadly across society or concentrated among particular groups.
This chapter analyses differences in the share of individuals in the 2025 OECD Trust Survey reporting high or moderately high trust across population groups defined by demographic characteristics (such as age), socio-economic characteristics (including perceived discrimination, educational attainment and financial concerns), and political dimensions. The latter comprise political agency – capturing perceptions of having a political voice and confidence in one’s ability to participate in politics – as well as partisanship.
Findings from the OECD Trust Survey 2025 indicate that the largest trust gaps are found along political dimensions, namely partisanship and political agency (Figure 2.1). Gaps across socio-economic characteristics, including perceived discrimination, educational attainment and financial hardship, are narrower. Finally, the smallest differences in trust are observed with regard to demographic characteristics. While Figure 2.1 shows this pattern for OECD countries, the same is true for the five participating OECD accession candidate countries.
Share of population with high or moderately high trust in the national government by level of respondents’ socioeconomic and demographic characteristics, partisanship and political agency, OECD, 2025
Note: The figure presents the unweighted OECD averages of weighted OECD country averages of 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?” by respondents’ feelings of political agency, partisanship, socio-economic background and demographic characteristics. Shown here is the proportion that have “high or moderately high trust” based on the aggregation of responses from 6-10, grouped by respondents’ feeling of political agency (feeling confident to have a say in what the government does, feeling confident to participate in politics) and partisanship (voted for government during last national elections), socio-economic background (financial concerns, education, identification as part of a discriminated group) and demographic characteristics (men and women, age groups). Financial concerns are measured by asking ‘’In general, thinking about the next year or two, how concerned are you about your household's finances and overall social and economic well-being? and aggregating responses 3 (somewhat concerned) and 4 (very concerned)’’. Low education is defined as below upper secondary educational attainment and high education as tertiary education (ISCED 2011 classification). People’s identification as part of a discriminated group is measured by responses ‘’Yes’’ to the question “Would you describe yourself as being a member of a group that is discriminated against in [Country]?” New Zealand is excluded from the OECD average for partisanship as the survey question on voting for the current government was not included there due to political neutrality requirements.
Source: OECD Survey on Drivers of Trust in Public Institutions 2025.
2.2.1. Perceived political agency and partisanship most strongly correspond with levels of trust
The most pronounced trust gaps across OECD countries are observed along dimensions of perceived political agency and partisanship. This points to the importance of people’s perceptions of political voice and their alignment with governing authorities in shaping trust in national government. The size of the trust gaps related to political agency and partisanship is broadly consistent with existing evidence showing that lower levels of perceived political empowerment are also associated with significantly lower confidence in institutions of a political nature (OECD, 2024[1]; OECD, 2022[2]).
More than two thirds of respondents (69%) across the OECD trust the national government when they believe people like them have a say in government decision-making, compared with only one fifth (22%) among those who feel they do not have a say - corresponding to a trust gap of 47 percentage points. The largest gaps are observed in Australia (60 percentage points), followed by Denmark and New Zealand (57 and 58, respectively), whereas the smallest gaps are recorded in Austria and Costa Rica (38 percentage points) and Chile (36 percentage points). In the five participating OECD accession candidate countries, the gap is largest in Brazil (53 percentage points) and smallest in Bulgaria (38 percentage points).
A similar pattern emerges when considering respondents who feel confident about participating in politics, however the gap is somewhat smaller. Around 55% of OECD respondents who feel confident about participating in politics report trusting the national government, compared with 28% of those who do not (Figure 2.2). This corresponds to a trust gap of 27 percentage points. The largest trust gap in the OECD is observed in Poland (44 percentage points), and the smallest is observed in Slovenia (13 percentage points). The gap is also comparatively small in Croatia and Peru (15 percentage points).
These trust gaps related to political agency—namely, the perceived ability to have a say in government decision-making and confidence in participating in politics—are substantial. Of particular concern is that a large share of people across OECD countries report low levels of political agency, and that this has shown little change over time.
Across the 19 countries surveyed in 2021, 2023 and 2025 for which the data point is available, only 31% (2021 and 2023) and 32% (2025) report that the political system allows them to have a say in government decision-making. Likewise, just 42% of respondents in 2021 and 40% in both 2023 and 2025 report feeling confident in their ability to participate in politics. These findings suggest that enhancing citizens' sense of political agency is an ongoing and challenging task for governments.
Share of population with high or moderately high trust in the national government by feeling they have a say in what the government does (blue) and confidence to participate in politics (yellow), 2025
Note: The figure presents the weighted country averages of 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?” by respondents’ feeling of having a say (blue) and confidence to participate in politics (yellow). Shown here is the proportion of respondents that have “high or moderately high trust’’ based on the aggregation of responses from 6-10 on the 0-10 response scale, grouped by whether respondents feel that people like them have a say (blue): ‘’How much would you say the political system in [COUNTRY] allows people like you to have a say in what the government does?’’ and feel confident to be able to participate in politics (yellow): ‘’How confident are you in your own ability to participate in politics?’’ “OECD” presents the unweighted average of weighted OECD country averages.
Source: OECD Survey on Drivers of Trust in Public Institutions 2025.
Consistent with previous OECD survey waves in 2021 and 2023, perceptions of political agency are higher among people who voted or would have voted for a governing party than among those who did not, suggesting that political agency is associated with partisan alignment (OECD, 2022[2]; OECD, 2024[1]). But notably, and even more among those who have or would have voted for a governing party, only a minority feel they have a say or are confident to participate in politics. Among government supporters, 38% report believing that people like them have a say in government decision-making, compared to 25% of non-supporters. Similarly, 45% of supporters report feeling able to participate in politics, compared with 38% of non-supporters.
Share of population with high or moderately high trust in the national government by whether they voted for a party in power or not, 2025
Note: The figure presents the weighted country averages of 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’’ by respondents’ political alignment. Shown here is the proportion of respondents that have “high or moderately high trust’’ based on the aggregation of responses from 6-10 on the 0-10 response scale, grouped by whether people voted (or would have) voted for the government in power: ‘’Is the party you voted for in the last national election on [DATE] currently part of the government?’’ or ‘Is the party you would have most likely voted for in the last national election on [DATE] currently part of the government?’ (depending on whether the respondent voted in the last national election). New Zealand is excluded from the figure as the survey question on voting for the current government was not included there due to political neutrality requirements. “OECD” presents the unweighted average of weighted OECD country averages.
Source: OECD Survey on Drivers of Trust in Public Institutions 2025.
These partisan differences in political agency are also mirrored in trust outcomes. As shown in Figure 2.3, on average 54% of respondents who voted for the governing party report trusting the national government, compared with only 28% among those who did not, implying a partisan trust gap of 26 percentage points. This gap is particularly large in Italy (39 percentage points) and Ireland (38 percentage points), while it is considerably smaller in Lithuania (7 percentage points) and the Netherlands (12 percentage points). In the participating OECD accession candidate countries, the gap is large in Brazil (38 percentage points) and substantially smaller in Bulgaria (11 percentage points).
Trust gaps associated with partisanship are perhaps unsurprising for political institutions, such as the national government or legislature. However, such gaps are less anticipated for administrative institutions such as the police or civil service, which are designed to operate on a non-partisan basis. Figure 2.4 shows that partisan trust gaps are present across all public institutions, although their magnitude varies substantially. For political institutions, partisan trust gaps are substantial. The gap amounts to 26 percentage points for the national government and 19 percentage points for the national legislature, and 18 percentage points for the regional government. It is somewhat smaller for political parties (15 percentage points) and local or municipal government (14 percentage points). Partisan trust gaps, however, also exist for institutions that are perceived as operating on a more neutral or administrative basis. These include the national civil service (a difference of 15 percentage points), the police (12 percentage points), and the regional civil service (12 percentage points).
Share of population with high or moderately high trust into different institutions by whether they voted or would have voted for a party in power or not, 2025
Note: The figure presents the unweighted OECD average of the weighted OECD country averages of 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 [institution]’’ by respondents’ political alignment. Shown here is the proportion of respondents that have “high or moderately high trust’’ based on the aggregation of responses from 6-10 on the 0-10 response scale, grouped by whether people voted (or would have) voted for the government in power: ‘’Is the party you voted for in the last national election on [DATE] currently part of the government?’’ or ‘Is the party you would have most likely voted for in the last national election on [DATE] currently part of the government?’ (depending on whether the respondent voted in the last national election). New Zealand is excluded from the OECD average as the survey question on voting for the current government was not included there due to political neutrality requirements.
Source: OECD Survey on Drivers of Trust in Public Institutions 2025.
Beyond identifying support for the current government in the most recent election, the OECD Trust Survey does not include measures of broader political preferences, such as respondents’ ideological orientation (presented for example in (Esteban and Ray, 2011[3])). Differences in trust between supporters and non-supporters of a governing party therefore offer a proxy for examining how trust varies across political groups. While such differences are commonly observed with respect to political institutions, for example the national government, they are particularly relevant when observed for administrative institutions, such as the civil service. Trust in the national civil service is generally expected to reflect perceptions of professionalism, impartiality and performance rather than partisan alignment (OECD, 2025[4]). Large differences in trust between supporters and non-supporters of a governing party may suggest that political attitudes extend beyond political to administrative institutions that are expected to operate in a politically neutral manner (OECD, 2024[1]).
Share of population with high or moderately high trust into the national civil service by whether they voted for a party in power or not, 2021, 2023 and 2025
Note: The figure presents the weighted country averages of 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 civil service” by respondents’ political alignment. Shown here is the proportion of respondents that have “high or moderately high trust’’ based on the aggregation of responses from 6-10, grouped by whether people voted (or would have voted) for the government in power: Shown are 17 OECD countries that participated in all three survey waves and where information on partisan alignment is available for all three years. “OECD” presents the unweighted average of weighted OECD country averages.
Source: OECD Survey on Drivers of Trust in Public Institutions 2021, 2023 and 2025.
Partisan trust gaps with respect to the national civil service exist in nearly all OECD countries covered by the survey (Figure 2.5). On average, these gaps have also widened, from 10 percentage points in 2021, to 13 percentage points in 2023 and further to 15 percentage points in 2025. This widening is not driven by rising trust among government supporters, among whom trust levels remained broadly stable. Rather, it reflects declining trust among respondents who did not vote for the government, with trust levels falling from 47% in 2021 to 43% in 2023 and 42% in 2025. Such differences in trust across political groups may have broader implications for democratic governance, including most importantly for policy making and implementation.
As discussed in Box 2.1, partisan differences in trust in the national civil service warrant particular attention, as they may signal broader divergence in trust across political groups, meaning that supporters of different political parties express systematically different levels of trust in public institutions and not only into the national government or national legislature.
2.2.2. Lower socio-economic status is associated with lower trust in public institutions
Socio-economic characteristics are associated with smaller trust gaps in national government than those linked to political agency and partisanship. The most pronounced disparities refer to financial concerns, while differences related to self-identified membership of a discriminated-against group and educational attainment are evident but comparatively smaller. At the same time, socio-economic characteristics are often closely interrelated. For instance, individuals who identify as belonging to a group that experiences discrimination are also more likely to report financial concerns; the share reporting financial concerns is 13 percentage points higher compared to those who do not identify this way. Despite these overlaps, examining trust gaps separately across each dimension remains informative, as it helps identify which socio-economics factors are most strongly associated with differences in trust in the national government.
Figure 2.6 shows that perceived financial insecurity is one factor that significantly affects trust in the national government. Of the socio-economic characteristics examined, financial concerns are associated with the largest trust gaps in national government. One half of respondents (53%) without financial concerns trust the national government, but only one third (35%) of those experiencing financial concerns trust the national government, yielding a substantial trust gap of 18 percentage points (Figure 2.6). Respondents with financial concerns are also more likely to identify economic pressures as a major issue. For example, 57% of people cite inflation as a major concern, compared to 44% of those without financial concerns. Those with financial concerns are also less optimistic about receiving financial support in the event of an economic crisis (36% find this likely, compared to 53% of those without concerns) (see Chapter 4).
There is, however, substantial cross-country variation in the trust gap associated with financial concerns. The largest gaps are observed in Ireland (33 percentage points), Portugal (28 percentage points) and Greece (27 percentage points). By contrast, the smallest gaps are found in France (eight percentage points), the United Kingdom (eight percentage points) as well as Croatia (10 percentage points).
Share of population with high or moderately high trust in the national government by financial concerns, 2025
Note: The figure presents the weighted country averages of 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?”. Shown here is the proportion of respondents that have “high or moderately high trust” based on the aggregation of responses from 6-10 on the 0-10 response scale, grouped by whether individuals responded 3 or 4 (somewhat or very concerned concerned) or 1 and 2 (not at all or little concerned) to the question ‘’In general, thinking about the next year or two, how concerned are you about your household’s finances and overall social and economic well-being?’’. “OECD” presents the unweighted average of weighted OECD country averages.
Source: OECD Survey on Drivers of Trust in Public Institutions 2025.
Educational attainment also corresponds with trust.1 For the purpose of this analysis, the comparison focuses on respondents with low levels of education (no upper secondary degree) and those with high levels of education (post-secondary degree). Respondents with upper secondary / high school education (medium level of education) are excluded for this comparison, as they fall between the other two levels and trust disparities are most pronounced between the lowest and highest education group. The full set of results, including the middle education group, is reported in Table 2.1 at the end of the chapter.
Figure 2.7 shows that 48% of respondents across the OECD who completed a post-secondary degree (i.e. a further degree after upper secondary/high school) trust the national government compared to 34% of respondents who did not graduate from upper secondary school. This amounts to an education trust gap of 14 percentage points. The education trust gap varies substantially across countries and is particularly large in Switzerland (35 percentage points), Portugal and the United Kingdom (both 31 percentage points). By contrast, trust in the national government is higher among respondents with lower levels of education for example in Mexico (eighteen percentage points) and Bulgaria (seven percentage points).
These cross-sectional differences occur against a backdrop of consistently increasing educational attainment. By 2024, nearly half (48%) of young adults had completed tertiary education – up from 27% in 2000 – and this figure is expected to rise further on average across the OECD (OECD, 2025[5]). Given the positive correlation between education and trust, the ongoing increase in educational attainment could lead to higher overall levels of trust over time. However, this does not necessarily imply a reduction in trust disparities between educational groups in the long run.
Share of population with high or moderately high trust in the national government by education group, 2025
Note: The figure presents the weighted country averages of 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?'. Shown here is the proportion of respondents that have 'high or moderately high trust' based on the aggregation of responses from 6-10 on the 0-10 response scale, grouped by respondents' highest education level attained: higher education (post-secondary degree) or lower education (no upper-secondary degree). “OECD” presents the unweighted average of weighted OECD country averages.
Source: OECD Survey on Drivers of Trust in Public Institutions 2025.
The trust gap associated with perceived discrimination is similar to the education trust gap. In the OECD, only 31% of people who identify as a member of a discriminated-against group trust the national government, compared to 45% of people who do not identify this way (Figure 2.8).2 This corresponds to a gap of 14 percentage points. These results are consistent with previous findings that suggest people self-identifying as belonging to a marginalised group may feel excluded from political processes and perceive public institutions as less responsive or accessible (OECD, 2025[4]; Tyrberg, 2024[6]).
However, there is considerable cross-country variation in the magnitude of this gap. The gap is particularly large in Estonia (24 percentage points) and Luxembourg (26 percentage points), while it is negligible in Brazil (two percentage points) and Colombia (one percentage point). Japan is the only participating country where the share with high or moderately high trust in the national government is higher among respondents who identify as part of a discriminated-against group, with a trust gap of seven percentage points.
Share of respondents with high or moderately high trust in the national government by those self-identifying as belonging to a discriminated group, 2025
Note: The figure presents the weighted country averages of responses to the question “On a scale from 0 to 10, where 0 is not at all and 10 is completely, in general how much do you trust the government?” by respondents’ perceptions of belonging to a discriminated group. Shown here is the proportion of respondents that have “high or moderately high trust” based on the aggregation of responses from 6-10 on the 0-10 response scale, grouped by whether respondents stated whether they feel they belong to a discriminated group: ‘’Would you describe yourself as being a member of a group that is discriminated against in [COUNTRY]?’’. “OECD” presents the unweighted average of weighted OECD country averages.
Source: OECD Survey on Drivers of Trust in Public Institutions 2025.
Taken together, these findings emphasise the significance of socio-economic factors in shaping trust in government. Although these characteristics are correlated, regression analysis shows that each factor remains statistically significant when demographic, socioeconomic and political variables are controlled for simultaneously.3 In other words, perceived financial concerns, perceived discrimination and education levels each contribute to variation in trust.
Across the OECD, respondents with low levels of interpersonal trust – i.e., trust in other people – report substantially lower confidence in all public institutions. However, the size of the gap differs markedly between institutions perceived as neutral and implementation-oriented and those that are explicitly political and representational. For example, among respondents with high interpersonal trust, 75% report trusting the police compared with 38% among those with low interpersonal trust. Trust in the national government shows an even larger gap, falling from 54% to 14% between these two groups (Figure 2.9). In relative terms, individuals with high interpersonal trust are roughly twice as likely as those with low interpersonal trust to trust the police and four times as likely to trust the national government. This gradient suggests that interpersonal trust serves as a baseline through which institutions are evaluated. Institutions perceived as impartial rule enforcers tend to sustain comparatively higher confidence, even among more sceptical individuals (Rothstein and Teorell, 2008[7]). Political institutions, by contrast, are more closely associated with contestation and discretionary authority, so confidence in them depends more strongly on prior beliefs about the trustworthiness of others and the fairness of the system, amplifying interpersonal trust divides (Hwang, 2017[8]).
Share of population with high or moderately high trust into different institutions dependent on respondents’ interpersonal trust levels, OECD 2025
Note: The figure presents the unweighted OECD average of the weighted OECD country averages of 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 following [Institution]?'. Shown here is the proportion of respondents that have 'high or moderately high interpersonal trust' based on the aggregation of responses from 6-10 on the 0-10 response scale or ‘low or no interpersonal trust’ based on the aggregation of responses from 0-4 on the 0-10 scale to the question ‘On a scale from 0 to 10, where 0 is not at all and 10 is completely, in general how much do you trust most people?’.
Source: OECD Survey on Drivers of Trust in Public Institutions 2025.
2.2.3. The smallest trust gaps are associated with demographic characteristics
Looking more closely at demographic differences, trust in national government varies to a comparatively small extent between men and women. 44% of men in OECD countries place high or moderately high trust the national government, compared to 37% of women – resulting in a trust gap of seven percentage points (Figure 2.10). The largest of these trust gaps is found in New Zealand with 24 percentage points, while there is no trust gap in Estonia and Latvia. In three countries, the pattern is reversed, with women reporting higher levels of trust than men. This applies to Norway (six percentage points), Iceland (five percentage points) as well as Brazil (four percentage points).
A closer examination shows that the trust gap between men and women is more pronounced among younger and middle-aged respondents.4 Trust levels among men in the OECD are relatively stable across age groups, ranging from 44% among younger respondents (aged 18-29) to 42% among those aged 30-49 and 45% among older respondents (50 and above). For women in the OECD, by contrast, trust is highest among those aged 50 and above (41%), compared with 32% among middle-aged women (30-49) and 34% among younger women (18-29). As a result, the difference in trust in the national government between men and women is wider among younger respondents (ten percentage points) and middle-aged respondents (ten percentage points) than among older ones (four percentage points).
Share of men and of women with high or moderately high trust in the national government, 2025
Note: The figure presents the weighted country averages of responses to the “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?” for people of different age groups. Shown here is the proportion of respondents that have “high or moderately high trust” based on the aggregation of responses from 6-10 on the 0-10 response scale, grouped by their age. “OECD” presents the unweighted average of weighted OECD country averages.
Source: OECD Survey on Drivers of Trust in Public Institutions 2025.
Age likewise is associated with a modest trust gap. Across OECD countries, 43% of respondents aged 50 and above report trusting the national government, compared with 37% among those aged 30–49 and 38% among those aged 18–29 (Figure 2.11). This corresponds to a five-percentage-point difference between the oldest and youngest generation. The pattern of lower trust in the national government among younger respondents compared with older ones holds in 23 OECD countries. However, several countries display the opposite pattern, with higher trust among younger people: In the OECD, these are Australia, Chile, France, Japan, the Netherlands and the United Kingdom. In Australia, for example, 60% of those aged 18-29 and 45% among those aged 50 and above have high or moderately high trust in the national government. In the five participating OECD accession candidate countries, the level of trust among younger people is also higher than among older people in Bulgaria and Peru.
Share of respondents with high or moderately high trust in the national government by respondent’s age group, 2025
Note: The figure presents weighted country averages of responses to the “On a scale of 0 to 10, how confident are you that the national government adequately balances the interests of current and future generations?” by respondents’ age. Shown here is the proportion of respondents that are “confident’’ based on the aggregation of responses from 6-10 on the 0-10 response scale, grouped by three age groups: 1) 18-29; 2) 30-49; 3) 50 and above. “OECD” presents the unweighted average of weighted OECD country averages.
Source: OECD Survey on Drivers of Trust in Public Institutions 2025.
2.2.4. Many trust gaps have remained broadly stable since 2021, but the educational trust gap is widening and the age-related trust gap is narrowing
Differences between population groups often reflect enduring predispositions shaped through early socialization processes (van der Meer, la Roi and van Alebeek, 2025[9]). While individuals’ trust may increase or decrease later in life in response to political or institutional developments, these shifts often leave relative differences between groups intact. Consequently, trust gaps may be resistant to short-term change, even when aggregate trust fluctuates.
Share of population with high or moderately high trust in the national government by level of respondents’ socioeconomic and demographic characteristics, partisanship and political agency, 2021, 2023 and 2025
Note: The figure presents the unweighted average of weighted OECD country averages of 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?” by respondents’ feelings of political agency, partisanship, socio-economic background and demographic characteristics. Shown here is the proportion that have “high or moderately high trust” based on the aggregation of responses from 6-10 on the 0-10 response scale, grouped by respondents’ feeling of political agency (feeling confident to have a say in what the government does, feeling confident to participate in politics) and partisanship (voted for government during last national elections), socio-economic background (financial concerns, education, identification as part of a discriminated group) and demographic characteristics (men and women, age group). Financial concerns are measured by asking ‘’In general, thinking about the next year or two, how concerned are you about your household's finances and overall social and economic well-being?’’ and aggregating responses 3 (somewhat concerned) and 4 (very concerned). Low education is defined as below upper secondary educational attainment and high education as tertiary education, following the ISCED 2011 classification. People’s identification of a discriminated group is measured by responses ‘’Yes’’ to the question “Would you describe yourself as being a member of a group that is discriminated against in [Country]?” The figure shows changes over time for all countries where relevant information is available in 2021, 2023 and 2025.
Source: OECD Survey on Drivers of Trust in Public Institutions 2021, 2023 and 2025.
Consistent with this expectation, trust gaps across population groups tend to persist. To ensure comparability over time, the longitudinal analysis presented in this section includes only those 18 OECD countries for which consistent data on trust and population group characteristics are available across survey waves. For completeness, Table 2.1 reports trust gaps not only for these 18 countries but also for the 29 countries that participated in both the 2023 and 2025 survey waves.
Figure 2.12 shows that most trust gaps have remained broadly stable between 2021 and 2025. The relative ranking of gaps has also remained largely unchanged, with the largest differences consistently observed for political agency and partisanship. At the same time, the magnitude of specific gaps within these categories has evolved to a limited extent for some population groups, in particular with regard to age, education, and political agency.
The only trust gap that has constantly decreased over time is defined by age. The age-related trust gap, which reflects higher levels of trust among respondents aged 50 and over compared to those aged 18–29, has declined markedly, falling from ten percentage points in 2021 to four in 2023 and then to just two in 2025. This is driven by rising trust among younger respondents coupled with a slight decrease among older people. In 2021, 37% of 18–29-year-old respondents reported trusting the national government. This figure increased to 39% in 2023 and 42% in 2025. Trust among older respondents decreased slightly, falling from 47% in 2021 to 43% in 2023 and then to 44% in 2025.
Unlike the convergence observed in the age-related trust gap, trust gaps with regard to political agency have widened over time. These trends point to a gradual widening of the trust divide between individuals who perceive themselves as politically empowered and those who do not. The trust gap relating to confidence in participating in politics has steadily widened over time, growing from 22 percentage points in 2021 to 25 in 2023 and 27 in 2025. By contrast, the trust gap relating to having a say in government decision-making primarily increased between 2021 (44%) and 2023 (48%), remaining stable since then at 48% in 2025.
An even more pronounced increase in trust gaps is observed with regard to educational attainment. The trust gap between respondents with high and low levels of education increased from 10 percentage points in 2021 to 13 percentage points in 2023 and 16 percentage points in 2025. With regard to the national government, this represents the largest widening of trust disparities observed across all population groups analysed in this chapter. Importantly, this trend is not limited to trust in the national government. A similar pattern is evident for trust in the national parliament, where the education-related gap grew from ten percentage points in 2021 to 14 in 2023 and 17 in 2025 (see Chapter 7).
The widening gap in trust in the national government is in parts driven by declining trust among respondents with lower levels of education. While trust among highly educated respondents remained relatively stable (49% in 2021, 47% in 2023 and 51% in 2025), the share of lower-educated respondents expressing trust fell from 39% in 2021 to 34% in 2023, with only a modest recovery to 35% in 2025. These differences may reflect variations in how individuals’ access and interpret information (OECD, 2024[10]). Differences in literacy and critical thinking, related to education, may influence how people engage with digital and social media environments. In return, this could have implications for trust, in particular among those with lower levels of education.
There is, nonetheless, considerable cross-country variation in this gap (Figure 2.13). In Portugal, for example, the education-related trust gap increased steadily, from 11% in 2021 to 14% in 2023 and 31% in 2025. By contrast, the gap decreased in Norway, falling from 21% in 2021 to 20% in 2023 and 15% in 2025. In the United Kingdom, the relationship between education and trust has shifted over time: In 2021 and 2023, respondents with lower levels of education reported higher levels of trust in the national government, with an eight percentage point gap in 2021 and a seven-percentage-point gap in 2023. By 2025, the situation had reversed, with those with higher levels of education reporting higher trust, resulting in a 31-percentage-point gap.
Share of population with high or moderately high trust in the national government by level of respondents’ education, 2021, 2023 and 2025
Note: The figure presents the weighted country averages of 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?” by respondents’ level of education, grouped by respondents' highest education level attained: higher education (post-secondary education degree) or lower education (no upper-secondary education degree). The figure shows changes over time for all countries where relevant information is available in 2021, 2023 and 2025. “OECD” presents the unweighted average of weighted OECD country averages.
Source: OECD Survey on Drivers of Trust in Public Institutions 2021, 2023 and 2025.
The following measures could serve to improve trust in public institutions across different population groups:
Reinforce perceptions of political agency and responsiveness, while expanding opportunities for participation. The share among individuals who feel that people like them have a say in government decisions who trust the national government is more than three times as high as among those who feel they lack voice. As is also demonstrated by regression results (see Chapter 1), strengthening the perceptions of political agency could be a powerful lever to improve institutional trust. At the same time, these perceptions are shaped by the extent to which individuals have real opportunities to participate in decision making and feel heard by decisionmakers. Expanding accessible channels for public participation – and meaningful responses from policymakers – can therefore help strengthen both perceived and actual political agency. Chapter 6 provides more detailed insights into the perceived barriers to political participation and impact and the potential policy actions that can be taken to address them.
Strengthen the perception of neutrality of administrative institutions. In order to prevent partisan divides from eroding broader institutional confidence, governments could consider ways to reinforce perceptions of the neutrality, transparency and accountability of public institutions. This may include safeguarding the independence of oversight bodies, strengthening integrity frameworks, and promoting structured cross-partisan dialogue.
Strengthen the involvement of socio-economically disadvantaged groups in policymaking processes. Trust levels are lower among individuals with financial concerns, those who identify as belonging to a discriminated-against group, and those with lower levels of education. To improve public trust among these population groups, policies aimed at better reaching them, enhancing their participation in the political process and making them “feel heard” may help to narrow trust disparities.
Share of respondents that trust the national government by subgroup in 2021, 2023 and 2025
Note: The table presents how trust in the national government varies across different population sub-groups. It highlights differences between individuals who believe they have a say in what the government does compared to those who do not, between those wo feel confident to participate in politics and those who do not (both defined as political agency gap); between those who voted for a party in power and those who did not (partisan gap); and between individuals who are not concerned about their financial well-being and those who are (financial concern gap). It also compares trust levels between individuals with tertiary education and those without (education gap), between men and women, and between older respondents (50 years and above) and younger respondents (18–29 years old) (age gap). Trust in the government is measured on a 0 to 10 scale, where 6 to 10 correspond to expressing moderately high to high trust in the government. Voted for a party in power is measured as people that have voted for the party currently in power or would vote for the party currently in power. Being concerned about financial well-being is measured on a 0 to 4 scale, where 3 to 4 correspond to people that are concerned about their financial wellbeing. “OECD (all countries)” refers to all countries participating in the OECD Trust Survey in a given wave; “OECD (29 countries)” refers to the countries participating in both 2023 and 2025; and “OECD (18 countries)” refers to those participating in 2021, 2023 and 2025. Please note that 20 countries participated in 2021, 2023 and 2025, but for Finland and Mexico, “Trust in national government” was not asked in 2021 implying that trust gaps are only available for 18 OECD countries. All reported averages are unweighted means across the respective country groups.
Source: OECD Survey on Drivers of Trust in Public Institutions 2021, 2023 and 2025.
References
[3] Esteban, J. and D. Ray (2011), “Linking Conflict to Inequality and Polarization”, American Economic Review, Vol. 101/4, pp. 1345-1374, https://doi.org/10.1257/aer.101.4.1345.
[8] Hwang, I. (2017), “Which Type of Trust Matters?:Interpersonal vs. Institutional vs. Political Trust”, Economic Research Institute, Working Papers 2017-15.
[5] OECD (2025), Education at a Glance 2025: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/1c0d9c79-en.
[4] OECD (2025), OECD Survey on Drivers of Trust in Public Institutions in Latin America and the Caribbean 2025 Results, OECD Publishing, Paris, https://doi.org/10.1787/ea3385cf-en.
[1] OECD (2024), OECD Survey on Drivers of Trust in Public Institutions – 2024 Results: Building Trust in a Complex Policy Environment, OECD Publishing, Paris, https://doi.org/10.1787/9a20554b-en.
[10] OECD (2024), What Does Child Empowerment Mean Today?: Implications for Education and Well-being, Educational Research and Innovation, OECD Publishing, Paris, https://doi.org/10.1787/8f80ce38-en.
[2] 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] Rothstein, B. and J. Teorell (2008), “What is quality of government? A theory of impartial government institutions”, Governance, Vol. 21/2, pp. 165-190, https://doi.org/10.1111/j.1468-0491.2008.00391.x.
[6] Tyrberg, M. (2024), “The impact of discrimination and support on immigrant trust and belonging”, European Political Science Review, Vol. 16/1, pp. 18-34, https://doi.org/10.1017/S1755773923000139.
[9] van der Meer, T., C. la Roi and C. van Alebeek (2025), “Socialized to (Dis)trust? A Panel Study into the Origins of Dispositional Institutional Trust”, Social Indicators Research, Vol. 178, pp. 371-391, https://doi.org/10.1007/s11205-025-03564-3.
← 1. It should be noted that, for Bulgaria, Costa Rica and Croatia, the quota for respondents with lower levels of education (i.e. those without an upper secondary education) was not met in 2025, even after applying post-stratification weights. Similar, the education quotas for lower education were not met in Chile, Costa Rica, Colombia, Denmark, Greece, Estonia, Portugal, and South Korea in 2023. Also, the education quotas in Estonia, Finland, Ireland, Latvia, Luxembourg, New Zealand and Norway were not met in 2021. Therefore, for these countries and years, the results presented need to be taken with caution.
← 2. The question about self-identified belonging to a discriminated-against group was based on a question in the European Social Survey (ESS). In the ESS, it is followed-up by a question of on what the group is discriminated against, with answer options including colour or race, nationality, religion, language etc. The composition of the group of respondents who identify as part of a discriminated-against group is likely to vary from country to country. Due to space constraints and the difficulty of achieving consistent estimates for small population groups given the 2000 respondents sample size, such a follow-up question was omitted in the OECD Trust Survey.
← 3. More specifically, results from a logit model estimating the probability of trusting the national government show that all explanatory variables are statistically significant at the 5% level or better. The model includes standardised measures of perceived political voice (feeling to have a say in the political system), perceived ability to participate in politics and interpersonal trust, as well as controls for age, age squared, educational attainment, financial concerns and support for the governing party.
← 4. The OECD Trust Survey applies representative quotas to ensure that key population groups are adequately reflected in each participating country. However, while quotas are met at the aggregate level for characteristics such as sex and age, the survey design does not guarantee full representativeness for combinations of characteristics (e.g. that the share of women within each age group among respondents corresponds to their share in the adult population). An examination of the weighted data suggests that quotas for men and women are broadly respected across age categories. Overall, the sample comprises approximately 49% men and 51% women. Within age groups, women account for 48% of respondents aged 18–29, 51% of those aged 30–49 and 53% of those aged 50 and above. Nevertheless, as with any survey-based analysis, results for subgroups, particularly when combining multiple characteristics, should be interpreted with appropriate caution.