Magnifying glass over the work "universalism"
VoxEU Column Frontiers of economic research Political Economy

Moral universalism: Global evidence

‘Universalism’ has attracted interest across the social sciences, with both theoretical and empirical work linking it to questions of cooperation, trust, immigration, and climate change. But the existing data on universalism apply largely to a few wealthy Western countries. This column introduces a dataset of 64,000 people from 60 countries making distributive decisions in monetary trade-offs between in-groups and strangers. The authors find large variations within and across countries that suggest some systematic ways that moral universalism and politico-economic outcomes interact and universalism and policy views mutually influence each other.

‘Universalism’ captures the extent to which people account for the interests of strangers relative to those of in-group members in their decision-making. Universalism has attracted interest across the social sciences because both theoretical and empirical work has linked it to variables such as social cooperation and trust (Tabellini 2007, Tabellini 2008), voting (Haidt 2012, Enke 2020), attitudes towards redistribution, immigration, or climate change (Enke et al. 2022a, Andre et al. 2021), the internal organisation of firms (Lowes and Le Rossignol 2022), and donations (Enke et al. 2022b). Despite this growing body of work, existing efforts to collect controlled data on universalism involve only a few predominantly rich Western countries, or small convenience samples.

To further our understanding of the role of universalism in society, in a recent paper (Cappelen et al. 2022) we introduce the Global Universalism Survey (GUS), the first large-scale globally representative dataset on the extent to which people make universalist distributive decisions in monetary trade-offs between in-group members and strangers.

The Global Universalism Survey

Implemented through the infrastructure of the 2020 Gallup World Poll, the GUS covers nationally representative samples in each of 60 countries, with a total sample size of about 64,000 respondents. These countries correspond to 85% of world population and 90% of world GDP, and were selected in an attempt to move beyond the overrepresentation of Western populations that is endemic to most multinational studies.

The dataset consists of a series of disinterested distributive decisions in which the respondent is tasked with distributing the local currency equivalent of a hypothetical $1,000 between two equally well-off individuals. We employ these questions to measure two variables: domestic universalism, where the distribution is made between an in-group member (such as a friend or co-religionist) and a stranger in their own country; and foreign universalism, where the distribution is made between an unknown compatriot and an unknown foreigner. Composite universalism is the unweighted average of domestic and foreign universalism.

Variation in moral universalism across individuals

The survey data show substantial variation in distributive behaviour in the global sample. Around 26% of respondents always act in line with universalism and divide the money equally in all decisions, while 17% of respondents strongly deviate from universalism by sharing no more than a fifth of the money with the stranger across the different situations.

Figure 1 Universalism and demographics

Figure 1 Universalism and demographics

Note: The acronym WEIRD follows the cultural psychology literature and refers to countries in Western Europe, Australia, Canada, and the US

As shown in Figure 1, younger people and women in almost all countries are more universalist, and the magnitude of these relationships is very similar in high- and low-income countries. For more endogenous individual characteristics, we often find large cross-cultural variation. For example, while well-educated city dwellers are more universalist in Western high-income countries, they are significantly less universalist outside this narrow set of countries. Similarly, the negative correlation between religiosity and universalism is considerably more pronounced in the West than in other parts of the world.

Moral universalism and policy views

Enke et al. (2022a) argued for a link between universalism and economic and social policy views. The idea is that many canonical left-wing policies have a universalist focus, so that universalism should be predictive of support for these policies. For instance, redistribution by the federal government is much more universalist than the small-scale group-based redistributive mechanisms that have prevailed for most of human history. A fortiori, policies that aim at supporting immigrants, needy people abroad, or preserving the global environment are highly universalist in nature. In contrast, a strong military is in some ways antithetical to universalism because it serves to defend boundaries between ‘us’ and ‘them’.

Figure 2 indicates that universalists indeed tend to support policies associated with a more universalist focus. While these relationships are almost always quantitatively meaningful and statistically significant in our global sample as a whole, we identify large heterogeneity across cultures. In low- and middle-income countries, universalism explains very little of the variation in political views. Moreover, the correlations between universalism and political views are twice as large in rich Western societies than in rich countries outside the West, such as South Korea, Israel, or Japan. This heterogeneity points to differences in the way social identity is structured in different countries (Gennaioli and Tabellini 2019).

Figure 2 Universalism and individual-level policy views

Figure 2 Universalism and individual-level policy views

Variation in moral universalism across countries

In Figure 3, we observe a large cross-country heterogeneity in universalism: the fraction of money shared with strangers ranges from around 25% in China, India, and Israel to 45% in Ethiopia. Overall, universalism is relatively high in sub-Saharan Africa, Latin America, and to some extent Western Europe and its offshoots. In contrast, universalism is lower in East Asia, South Asia, Eastern Europe, and to some extent in the Middle East.

Figure 3 The global variation in universalism

Figure 3 The global variation in universalism

Perhaps surprisingly, as shown in Figure 4, we find that per capita income is slightly negatively correlated with universalism. This result is partly, but not entirely, driven by many sub-Saharan populations making relatively universalist decisions.

Figure 4 Universalism and GDP per capita

Figure 4 Universalism and GDP per capita

The low explanatory potential of GDP raises the question whether there are ‘deeper’ historical variables that might explain cultural heterogeneity. According to a prominent hypothesis in the literature, moral views are economically functional, and have partly evolved to support and incentivise cooperation in economic production (Tabellini 2008). Based on the literature exploring the implications of this hypothesis, we find that country-level universalism is strongly negatively correlated with tight historical kinship ties and with historical irrigation practices, both of which are associated with higher extensive collaboration with in-group members (Greif and Tabellini 2017, Enke 2017, Henrich 2020, Talhelm 2014, Buggle 2020), and positively with Christianity, which historically dissolved tight kinship systems (Schulz et al. 2019). While correlational in nature, these results are consistent with the view that historical economic incentives shaped the distribution of universalism that exists across the globe today.

Democracy is another relevant factor, hypothesised by philosophers and cultural psychologists alike as potentially increasing support for universalist moral views (Rawls 1993, Henrich et al. 2010). Figure 5 suggests the existence of a positive link between universalism and democracy. We investigate a possible causal effect of democracy by linking country-cohort-specific variation in democracy over an individual's lifetime to universalism, holding the respondent's country and age fixed, and leveraging variation in exposure to democracy across age groups and countries. In line with the conjectures above, we find that experience with democracy is significantly predictive of universalism. This finding tentatively suggests that universalism can change based on the environment experienced by individuals, in a similar way to social trust or altruism (Helliwell et al. 2014, Giavazzi et al. 2014).

Figure 5 Universalism and democracy, controlling for log per capita income

Figure 5 Universalism and democracy, controlling for log per capita income

Does heterogeneity in universalism reflect differing moral views?

A main takeaway from the aforementioned analyses is that universalism varies widely both within and across societies. But should we think of this variation as reflecting heterogeneity in morality? One possibility is that people indeed have heterogeneous moral views: as in philosophical discussions of morality, some people may consider it morally right to give more weight to the interests of their in-group, while others consider it morally right to act in more impartial ways. Another possibility is that some people may deviate from universalism despite viewing it as morally right, simply because their in-group members are more important to them.

To shed light on this issue, we decompose heterogeneity in universalism into moral views and in-group preferences. For this purpose, we directly ask a subset of respondents how they would split the $1,000 if they were to do what they consider morally right. Strikingly, we find almost no difference in allocation decisions when we ask participants to do what they consider morally right compared to simply asking them to allocate money. This suggests that most of the heterogeneity in universalist behaviour across the world reflects disagreement about what is the morally right thing to do, perhaps because some people believe in the existence of relationship-specific moral obligations while others do not.

Takeaways and broader research agenda

Our work presents several stylised facts about moral universalism and its correlates at the individual and the country level. The findings provide evidence that support some hypotheses in the literature, but also show that some results that hold for high-income, Western countries do not generalise to a more global context. A main takeaway from our paper is that, around the world, moral universalism and politico-economic outcomes interact in systematic ways, and universalism and policy views mutually influence each other.

We believe that this interaction can be fruitfully explored in future research. Indeed, we constructed the GUS with a focus on making available to the research community a rich dataset that can be used for a broad set of analyses in behavioural, cultural, political, and development economics.


Andre, P, T Boneva, F Chopra and A Falk (2021), “Fighting Climate Change: The Role of Norms, Preferences, and Moral Values”, IZA DP No. 14518, June.

Buggle, J C (2020), “Growing Collectivism: Irrigation, Group Conformity and Technological Divergence”, Journal of Economic Growth 25 (2): 147–93.

Cappelen, A W, B Enke and B Tungodden (2022), “Moral Universalism: Global Evidence”.

Enke, B (2017), “The coevolution of kinship systems, cooperation, and culture”,, 21 September.

Enke, B (2020), “Moral Values and Voting”, Journal of Political Economy 128 (10): 3679–3729.

Enke, B, R Rodríguez-Padilla and F Zimmermann (2022a), “Moral Universalism and the Structure of Ideology”, Review of Economic Studies.

Enke, B, R Rodríguez-Padilla and F Zimmermann (2022b), “Moral Universalism: Measurement and Economic Relevance”, Management Science, 10 August.

Gennaioli, N and G Tabellini (2019), “Identity, Beliefs, and Political Conflict”,, 6 June.

Giavazzi, F, I Petkov and F Schiantarelli (2014), “Culture: Persistence and Evolution”,, 16 June.

Greif, A and G Tabellini (2017), “The Clan and the Corporation: Sustaining Cooperation in China and Europe”, Journal of Comparative Economics 45 (1): 1–35.

Haidt, J (2012), The Righteous Mind: Why Good People Are Divided by Politics and Religion, Vintage.

Helliwell, J, S Wang and J Xu (2014), “New Evidence on the Durability of Social Norms”,, 12 March.

Henrich, J, S J Heine and A Norenzayan (2010), “The Weirdest People in the World?”, Behavioral and Brain Sciences 33 (2–3): 61–83.

Lowes, S and E Le Rossignol (2022), “Ancestral Livelihoods and Moral Universalism: Evidence from Transhumant Pastoralist Societies”, CEPR Discussion Paper DP17476.

Rawls, J (1993), “The Law of Peoples”, Critical Inquiry 20 (1): 36–68.

Schulz, J F, D Bahrami-Rad, J P Beauchamp and J Henrich (2019), “The Church, Intensive Kinship, and Global Psychological Variation”, Science 366 (6466).

Tabellini, G (2007), “Morality Matters for Economic Performance”,, 22 December.

Tabellini, G (2008), “The Scope of Cooperation: Values and Incentives”, Quarterly Journal of Economics 123 (3): 905–50.

Talhelm, T, X Zhang, S Oishi, C Shimin, D Duan, X Lan and S Kitayama (2014), “Large-Scale Psychological Differences within China Explained by Rice versus Wheat Agriculture”, Science 344 (6184): 603–8.