Can social cohesion—the social ties that weave citizens together—be maintained in the face of an increasingly ethnically diverse populace? With continued immigration, this question is more relevant than ever. A key derivative of the general question relates to whether everyday exposure to immigrants in residential areas reduces natives’ trust in other people in general (“social trust”)—a key component of social cohesion. This has been examined extensively in previous studies, but only at high levels of geographical aggregation—for example, municipalities or regions—that do not index actual exposure to people of other ethnic background. Thus, no conclusive evidence yet exists. This project seeks to answer whether local exposure to people of different ethnic background erodes trust in other people by zooming in on the immediate residential environment (down to within a radius of 80 metres of the place of residence), where interethnic exposure is unavoidable in ethnically diverse contexts. It does so using data on contextual diversity from the national Danish population registers linked to survey data on trust. The results show that interethnic exposure in immediate residential surroundings—down to a few hundred metres within the place of residence—reduces social trust among native Danes. When zooming out to more aggregate residential contexts, the negative effect is gradually reduced and ultimately disappears in the most aggregate contexts (of a radius of 2500 metres), which further substantiates the claim that interethnic exposure in the micro-context lowers social trust.
Immigration, Ethnic Diversity, and Social Cohesion
“Ethnic Diversity and Social Cohesion: The Influence of Ethnic Diversity in the Neighbourhood on Social Trust” is led by Professor (MSO) Peter Thisted Dinesen (Department of Political Science, University of Copenhagen) and Professor Kim Mannemar Sønderskov (Department of Political Science, Aarhus University).
As illustrated by the recent heated discussions over the increased influx of asylum seekers in Denmark and other EU countries, concerns over the consequences of continued immigration loom large in the destination countries. A key concern relates to the financial strain on the welfare state imposed by immigration. However, the discussion also relates to the consequences for social cohesion in the receiving societies in guise of the social ties weaving citizens together, thereby allowing them to collaborate for the common good. This project, which is supported by the Carlsberg Foundation and led by Professor (MSO) Peter Thisted Dinesen (University of Copenhagen) and Professor Kim Mannemar Sønderskov (Aarhus University), has examined whether one of the most manifest consequences of continued immigration—increased ethnic diversity in residential areas—influence a key component of social cohesion, namely trust in other people in general, or what is commonly referred to as “social trust”, among native Danes (Dinesen & Sønderskov, 2015).
The Previous Literature: Inconsistent Findings
Research on the consequences of residential exposure to ethnic diversity for social trust has grown at an exponential rate over the last decade and a half, gaining particular momentum with Putnam’s (2007) controversial findings from the United States. Putnam showed that living in more ethnically diverse residential settings reduces trust in others as well as a number of other civic attitudes and behaviours. This spawned an intense discussion—in academic circles and in the public at large—and a wealth of studies (re)examining Putnam’s claim with more refined research designs and/or in other geographical settings. The results from these studies are, however, inconsistent: some find a similar negative relationship between residential ethnic diversity and social trust, while others find no relationship.
A large number of studies have examined the relationship between ethnic diversity and social trust, but none has done so in the immediate residential context. It is therefore unknown if actual exposure to people of different ethnic background erodes trust.
There are multiple potential explanations for the inconsistent relationship between residential ethnic diversity and trust, but one stands out as particularly straightforward: the level of aggregation of the residential areas in which diversity is measured. More specifically, due to data limitations, previous studies have been confined to examining the diversity-trust relationship in very aggregate contexts such as municipalities or regions. Consequently, these studies fail to examine the key mechanism typically invoked to account for the relationship between residential ethnic diversity and social trust: actual residential exposure to people of different ethnic background, which people then use to form perceptions about the trustworthiness of others more generally. If ethnic diversity is measured in overly expansive contexts, we cannot know if it in fact indexes actual experiences with interethnic others. Or put differently: even in ethnically diverse aggregate contexts, e.g. municipalities, one can live in an ethnically homogenous neighbourhood with little interaction with people of different ethnic background. This in turn implies that previous studies—whether finding a relationship or not—do not speak directly to the question of whether residential interethnic exposure erodes social trust.
A New Approach: Measuring Ethnic Diversity in the Micro-Context
Against the backdrop of the shortcomings of crude measurement of context in previous studies, this project has zoomed in on the immediate residential environment—where interethnic exposure is unavoidable in ethnically diverse contexts—in order to test the posited influence of residential ethnic diversity on social trust, and to substantiate interethnic exposure as the linking mechanism. It has done so by utilising the extremely elaborate data in the national Danish population registers, which contain (anonymized) information about the exact location of residences of everyone living in Denmark, as well as their ethnic background (country of origin of themselves and their parents), and a host of other relevant information. By linking survey data on social trust with the population registers, we have been able to calculate measures of contextual ethnic diversity for individualised contexts of different size; from more aggregate (a radius of 2500 metres) to previously unseen levels of disaggregation (down to within a radius of 80 metres of the place of residence).
Danish register data linked with survey data on trust allow us to zoom in on the relationship between ethnic diversity in the residential micro-context (down to within a radius of 80 metres of the place of residence) and social trust.
By examining the influence of residential ethnic diversity in extremely local environments on social trust, and, moreover, comparing it to the relationship in more aggregate environments, we can assess the conjecture that interethnic exposure underlies the (negative) relationship between ethnic diversity and trust. If a relationship exists in the immediate micro-context, where interethnic exposure is essentially unavoidable due to the local nature of these contexts, but not in more aggregate contexts, where such exposure may or may not take place, this indicates that interethnic exposure triggers the negative relationship between residential ethnic diversity and trust.
The Finding: Ethnic Diversity in the Residential Micro-Context Erodes Trust
Ethnic diversity in the immediate micro-context influences social trust negatively. In more aggregate contexts, the relationship disappears. This indicates that exposure to people of different ethnic background is the mechanism linking residential ethnic diversity and trust.
The results of the empirical analysis strongly support the notion that exposure to people of different ethnic background in the immediate residential context erodes social trust among native Danes: a negative relationship is observed in highly local contexts (up to within a radius of 180 metres of the place of residence) after which the effect attenuates and completely vanishes in the most aggregate contexts examined (within a radius of 2500 metres of the place of residence). In quantitative terms, a 15-percentage point increase in concentration of immigrants in the local residential context corresponds, ceteris paribus, to approximately a one-percentage point decrease in trust. Similar effects are found for alternative measures of contextual ethnic diversity. This effect is relatively modest, but as the same time non-negligible compared to other common explanations of trust. Especially given that the analysis takes into account an extensive list of factors—at the individual level (e.g. education or income) as well as the contextual level (e.g. unemployment and crime)—which might plausibly confound the relationship between diversity and trust. As another point of concern, we have also—to the extent possible with the data at hand—assessed whether our results reflect self-selection (reverse causality) of more trustful individuals into less diverse local contexts (or other dynamics of residential selection). While we cannot provide definitive evidence of a causal relationship, we find no indication that such dynamics of residential selection confound our findings.
Apocalyptic Prophecies are Premature
The finding of a negative effect of local ethnic diversity on trust, and its potential ramifications, should be balanced against the steep increase in trust in Denmark from 1979 to 2009 (from 47% to 79% expressing trust in others).
The finding that residential ethnic diversity tends to erode social trust should obviously raise concern for diversifying societies. However, before drawing too gloomy conclusions, it is important to put the negative diversity effect in context. From 1979 to 2009, social trust (measured as share of individuals indicating to trust most other people) in Denmark increased from 47% to 79% (Sønderskov & Dinesen, 2014)—to our knowledge, the highest level ever recorded worldwide. This implies that other factors—our research particularly points to incorrupt state institutions and the public trust in these institutions that flows from it (Sønderskov & Dinesen, 2014, 2015)—overshadow the negative effects of diversity on trust. On balance, it is therefore fair to conclude that local ethnic diversity has a negative impact on social trust, which should be taken seriously, but at the same time not overestimated compared to other forces simultaneously increasing trust.
Further Perspectives
“The grant from the Carlsberg Foundation has provided a unique opportunity to develop new ways of measuring the immediate residential context. This innovation has subsequently been used not only to analyse the consequences of ethnic diversity for trust, but also to answer other important questions. In subsequent studies we have (together with co-authors) applied and extended this research design to the question of how people form perceptions of the national economy (Bisgaard, Dinesen & Sønderskov, Forthcoming) as well as to the pertinent question of how residential ethnic context influences attitudes toward immigration (Danckert, Dinesen & Sønderskov, Forthcoming). In ongoing research we hope to refine the unique measurement of residential context even further and supplement it with measures of other salient contexts in which individuals are embedded,” comment Peter Thisted Dinesen and Kim Mannemar Sønderskov.
References
- Bisgaard, M., Dinesen, P. T. & Sønderskov, K. M. (Forthcoming). “Reconsidering the Neighborhood Effect: Does Exposure to Residential Unemployment Influence Voters' Perceptions of the National Economy?” Journal of Politics.
- Danckert, B., Dinesen, P. T. & Sønderskov, K. M. (Forthcoming). “Reacting to Neighborhood Cues? Political Sophistication Moderates the Effect of Exposure to Immigrants.” Public Opinion Quarterly.
- Dinesen, P. T. & Sønderskov, K. M. (2015). “Ethnic diversity and Social Trust: Evidence from the Micro-Context.” American Sociological Review, Vol. 80 (3), pp. 550-573. http://asr.sagepub.com/content/80/3/550.abstract
- Sønderskov, K. M. & Dinesen, P. T. (Forthcoming). “Trusting the state, trusting each other? The effect of institutional trust on social trust” Political Behavior. http://link.springer.com/article/10.1007/s11109-015-9322-8
- Sønderskov, K. M. & Dinesen, P. T. (2014). “Danish Exceptionalism: Explaining the Unique Increase in Social Trust over the Past 30 Years.” European Sociological Review, Vol. 30 (6), pp. 782-795. http://esr.oxfordjournals.org/cgi/doi/10.1093/esr/jcu073