Trust and Income Inequality
There is a strong negative correlation between generalized trust and Gini indexes of income inequality, both across countries (Figure 10.6, Panel A) and across US states (Panel B). High-trusting societies are more equal (they have lower Gini coefficients), while low-trusting societies typically show higher levels of income inequality. Cross-country and cross-US states regressions controlling for income, population, education, and ethnic fractionalization confirm this correlation (Algan and Cahuc, 2014).
Alesina and La Ferrara (2000) show that this negative relationship between trust and income inequality also holds at a more local level within US counties and municipalities. Rothstein and Uslaner (2005) document a within-states correlation for the United States between the rise in income inequalities and the decline of trust over the last decades. A pending issue is that of causality. Inequality might correlate negatively with trust for several reasons. On the one hand, as suggested by Rothstein and Uslaner, high levels of trust and cooperation might go along with high preferences for redistribution and thereby contribute to lower inequality.3 On the other hand, high inequality can make individuals perceive themselves as unfairly treated by people belonging to social classes different from their own, leading them to restrict cooperative action and trust to members of their own class (Rothstein and Uslaner, 2005). Kumlin and Rothstein (2005) also show that more universalist and egalitarian welfare state regimes are associated with higher levels of trust than corporatist welfare state systems that divide social benefits by status.
Figure 10.5. Institutional Trust and Life Satisfaction, 2006–15
Note: Life satisfaction data comes from the Gallup World Poll. Data on trust in the judicial system and in government are sourced from the Eurobarometer.
Source: OECD (2017), OECD Guidelines on Measuring Trust, OECD Publishing, Paris. StatLink 2 http://dx.doi.org/10.1787/88933840114.
Research is still needed to nail down the causal effect of income inequality on generalized trust. The application of behavioral surveys looking at cooperation between individuals depending on their demographics, status, and income would improve the investigation of this causal relationship.
Figure 10.6. Income Inequality and Generalized Trust Across Countries and US States
Note: Inequality is measured by the average of the Gini index between 2005 and 2012 (World Bank). Generalized trust is measured as the country average from World Values Survey (1981–2009) and European Values Survey (1981–2008). For the United States, inequality is measured by the Gini index in 2010 (US Census Bureau). Generalized trust is taken from the General Social Survey (1973–2006).
Source: Algan, Y. and P. Cahuc (2014), “Trust, growth and well-being: New evidence and policy implications,” in Aghion, P. and S. Durlauf (eds.), Handbook of Economic Growth, Vol. 2, Elsevier, North-Holland, Amsterdam, pp. 49–120. StatLink 2 http://dx.doi.org/10.1787/888933840133.
How Can Policy Affect Trust?
Trust varies significantly within countries, depending on income, education, employment status, and household type (OECD, 2017). Both generalized trust and trust in institutions are higher among higher-income groups and among more highly educated people, and lower among unemployed people and single-person households with at least one dependent child.
While these patterns hold true across the majority of OECD countries, it is important to study the drivers of trust in the context of countries’ specific circumstances, and how policy-makers could develop such an important type of social capital. If trust plays a key role in explaining economic and social outcomes, it becomes urgent to identify the institutions and public policies needed for it to develop.
Research on this subject is still in its early stages, owing mainly to the lack of adequate behavioral measures across time and localities. Part of the literature considers trust to be a deeply rooted cultural component, whose determinants must be searched for in the long history of each country, with little room for immediate action. However, recent studies on immigrants show that their level of trust gradually converges to the average level of trust prevailing in their country of destination. This ambiguity is well illustrated by the two conflicting views of the evolution of trust given by Robert Putnam. According to “Putnam I” (Putnam, Leonardi, and Nanetti, 1993), social capital is largely determined by history. In this account, higher levels of social capital in the regions of northern Italy compared with those in the south originated in the free-city experience during the medieval period. On the other hand, according to “Putnam II” (Putnam, 2000), trust evolves from one generation to the next, and is strongly influenced by the environment. In Bowling Alone, Putnam shows that the levels of social capital, as measured by membership in associations and clubs, have starkly declined in the United States since World War II.
Depending on which perspective we take, “Putnam I” or “Putnam II,” the room for policy intervention would be small or large. In fact, both approaches have an element of truth. Trust is partly inherited from past generations and shaped by historical shocks, as the underlying beliefs regarding the benefits of trust and cooperation are transmitted in communities through families (Bisin and Verdier, 2001; Benabou and Tirole, 2006; Tabellini, 2008; Guiso, Sapienza, and Zingales, 2008). But another part of trust is shaped by personal experience from the current environment, be it social, economic, or political. In Bisin and Verdier’s terminology, both the vertical channel of transmission from parents to children and the oblique/horizontal channel from the contemporaneous environment are at play in building trust. This debate is also influenced by what generalized trust really measures. If trust consists of beliefs about the trustworthiness of others, it is likely that individuals update their beliefs depending on the environment where they live, the civic spirit of their fellow citizens, and the transparency of their institutions. If trust consists of deep preferences and moral values, transmitted in early childhood and disconnected from personal experience, as suggested by Uslaner (2002), it might take more time to adjust. Another interpretation is that there are equilibria that persist and are hard to change, unless citizens are nudged with relevant public policy (Hoff and Stiglitz, 2016).
The Role of Education
The bulk of the existing policy-relevant evidence on the drivers of trust is on education programs. There is some evidence that more education is associated with higher social capital (Helliwell and Putnam, 2007; Glaeser, Ponzetto, and Shleifer, 2007). However, variation in the average years of education of the population across developed countries is too small to explain the observed cross-country differences in trust. Algan, Cahuc, and Shleifer (2013) propose a complementary explanation by looking at the relationship between how students are taught and students’ beliefs in cooperation. They show that methods of teaching differ widely across countries, both between schools and within schools in a country. Some schools and teachers emphasize vertical teaching practices, whereby teachers primarily lecture, students take notes or read textbooks, and teachers ask students questions. In this model the central relationship in the classroom is between the teacher and the student. Other schools and teachers emphasize horizontal teaching practices, whereby students work in groups, do projects together, and ask teachers questions. In this model, the central relationship in the classroom is among students.
Consistent with the idea that beliefs underlying social capital are acquired through the practice of cooperation, and that social skills are acquired in early childhood, Algan, Cahuc, and Shleifer (2013) show that horizontal teaching practices can develop social capital. This evidence calls for adding questions on social capital and teaching methods in traditional cross-country educational surveys such as PISA.
Several studies provide justification for policy intervention in the form of early childhood programs aimed at developing children’s social skills. Recent longitudinal studies suggest that much of the impact of programs that improve adult achievement (such as the Perry Preschool program or Project STAR in the Unite
d States) flows through some sort of noncognitive channel (Heckman and Kautz, 2012; and Heckman et al., 2013). Algan et al. (2012) use data from a large and detailed longitudinal study following the social, cognitive, and emotional development of men who were kindergarteners in neighborhoods of low socio-economic status in Montreal in 1984. The study incorporates a randomized evaluation of an intensive two-year social skills training program at the beginning of elementary school for the most disruptive children. Those who participated in the training program had significantly more favorable social and economic outcomes upon reaching adult age. By distinguishing between the different cognitive and noncognitive channels through which this intervention operates, the authors conclude that noncognitive skills are the main channel shaping economic outcomes in adult life.
The Role of Institutions
This chapter has so far treated inter-personal trust and institutional trust separately. But as we have mentioned, a key ingredient of inter-personal trust is the belief that others will behave in a fair and cooperative way. The role of institutions is crucial to strengthen cooperation. This is a real policy lever to build trust in the short run by improving the integrity and transparency of institutions.
Figure 10.7 first shows a strong positive correlation between generalized interpersonal trust and the quality of the legal system for a sample of 100 countries. This robustly correlates to using different measures of institutional quality commonly used in the economic literature (such as the rule of law, the strength of property rights, and the enforcement of contracts, as well as government effectiveness, accountability, and corruption) and to controlling for other influences of institutional quality.
Several papers try to go beyond this correlation by showing a causal impact of legal enforcement on generalized inter-personal trust. Tabellini (2008) provides evidence that suggests that generalized morality is more widespread in European regions that used to be ruled by nondespotic political institutions in the distant past. Weak legal enforcement also forces citizens to rely on informal and local rules, and to develop limited trust as opposed to generalized trust. This pattern is well illustrated by the experience of the Italian Mafia. According to Gambetta (1993), feudalism was formally abolished in Sicily much later than in the rest of Europe, and the state was too weak to enforce private property rights concerning land. The Mafia benefited from this institutional vacuum by offering local protection through informal patronage.
Other evidence shows that the transparency and integrity of institutions are important drivers of generalized trust not only in a cross-section of countries (Rothstein and Stolle, 2008), but also in an experimental context that isolates causality (Rothstein and Eek, 2009). The main theory behind this channel is that citizens who think that civil servants are corrupt extrapolate the same belief to others and to the population in general (Sønderskov and Dinesen, 2016).
Democratic institutions also have an impact on cooperative behavior. Bardhan (2000) finds that farmers are less likely to violate irrigation rules when they themselves have set up those rules. Frey (1998) shows that tax evasion in Swiss cantons is lower when democratic participation is greater. All these different works suggest an impact of democracy on cooperation.
An alternative approach for identifying the effect of institutions on cooperation is to mimic formal and legal rules in experimental games. Formal and legal rules implemented in experimental games obviously differ from real institutions. But this setting has the advantage of providing a controlled experiment to estimate how people change their cooperation and trust depending on exogenous variations in the rules of the games. Fehr and Gachter (2000) analyze cooperation in a public goods game, showing that free riders are heavily punished even if punishment is costly and does not provide any material benefits to the punisher. The opportunity for costly punishment causes a large increase in cooperation levels because potential free riders face a credible threat. In the presence of a costly punishment opportunity, almost complete cooperation can be achieved and maintained during the games. Herrmann, Thöni, and Gächter (2008) have used this setup to measure conditional cooperation in 16 different cities across the world. They find that cooperation for the funding of the public good is the highest in Boston and Melbourne and lowest in Athens and Muscat. This ordering is highly correlated with the rule of law and the transparency of institutions in the corresponding country. Similarly, Rothstein (2011) uses various experiments with students in Sweden and Romania that show their generalized trust and trust in civil servants decline substantially when students witness a police officer accepting a bribe. His interpretation is that the absence of transparency in institutions and of civic spirit by public officials can have very large damaging effects on generalized trust: if public officials, who are expected to represent the law, are corrupt, people infer that most other people cannot be trusted either.
Figure 10.7. Generalized Trust and Quality of Institutions
Note: Measures of the quality of the legal system are taken from the Economic Freedom of the World Index (2007). Generalized trust is measured as the country average from World Values Survey (1981–2009).
Source: Algan, Y. and P. Cahuc (2014), “Trust, growth and well-being: New evidence and policy implications,” in Aghion, P. and S. Durlauf (eds.), Handbook of Economic Growth, Vol. 2, Elsevier, North-Holland, Amsterdam, pp. 49–120. StatLink 2 http://dx.doi.org/10.1787/888933840152.
Conclusions
We propose three main recommendations to make progress on the measurement and analysis of inter-personal trust and institutional trust, and on how to reinforce the complementarities between survey measures and experimental measures.
Survey Measures of Trust
Survey measures will always out-perform experimental platforms in terms of sample size and coverage, but there is room for improvement, harmonization, and expansion. As discussed above, current survey measures have various shortcomings: their small sample size, which makes it impossible to get a comprehensive description of the level of trust at the local level and to analyze the economic, social, and policy determinants of trust; their relatively poor coverage, particularly over time, which makes it difficult to analyze how policy reforms affect the evolution of inter-personal trust and trust in institutions; and the heterogeneity across surveys in question wording and response scales. We recommend two steps:
• Invest in research about methodological issues such as question wording, scale use, and priming effects (i.e., how memory of a preceding stimulus influences response to the question being asked) in trust questions; and move toward a common and integrated approach to measurement for data producers. Institutional trust in particular would benefit from additional methodological research. The OECD Guidelines on Measuring Trust are an important step in this direction and include a set of prototype question modules that cover both inter-personal and institutional trust and can be readily inserted in household surveys.
• Include trust measures, especially the generalized trust question, in official (and unofficial) surveys. As shown above, the validity of the questions about inter-personal trust has been firmly established, and this dimension is critical for social progress and well-being. To maximize the use of these data, we need larger sample sizes, more detailed geographic locations, and time variation to provide policy-makers with more useful conclusions about the impact of trust and how to best support it. More research should be done on institutional trust. In addition to the shortcomings in coverage shared by inter-personal trust, we lack a sufficiently deep theoretical understanding of what the construct of institutional trust is. While institutional trust measures based on self-reported surveys are still worth collecting, further developing experimental measures of trust in institutions (discussed below) will be an important part of this line of research.
Experimental Measures of Trust
The important value added of experimental measures is that they describe observed behaviors and provide a true sense of bilateral cooperation between different
individuals and groups within a society. They are not better than survey measures in all respects, but they have a different and independent set of biases, so that we can learn something important from looking at the two approaches combined. We thus recommend expanding these measures, in particular TrustLab, which combines classical laboratory experiments with a more traditional internet-based questionnaire based on a representative sample. Progress in this field requires actions to improve the validity of these two approaches:
• Develop behavioral measures of trust in institutions. Implicit Association Tests are a promising step in this direction, because they allow for the measurement of attitudes difficult to capture through explicit self-reporting (Greenwald et al., 2002). Implicit Association Tests have been successfully applied to measure perceptions, stereotypes, and attitudes toward commonly stigmatized groups such as black people, women, and old people. The combination of these behavioral measures of institutional trust with survey questions and their application in different geographic contexts and over time will be of great help.
• Reinforce the complementarities between survey and experimental measures by harmonizing questionnaires. Falk et al. (2016) have conducted representative surveys with a few questions first validated from small experimental samples of students showing a consistency between behaviors in the lab and self-reported surveys. It would be important to enlarge the focus beyond trust by developing and refining a short survey module on norms and values and include such modules in official surveys. The Guidelines on Measuring Trust refer to some questions of Falk’s Global Preference module in their prototype question modules.
For Good Measure Page 49