VoxEU Column Europe's nations and regions Migration

Exposure to transit migration: Public attitudes and entrepreneurship

The impact of the 2015 refugee crisis on sending and receiving societies has received significant scholarly attention. But there is little research on how the crisis affected ‘transit countries’ through which migrants travelled. This column studies 800 localities in 18 European countries to discover how local populations responded to the temporary presence of forced migrants. Data show that entrepreneurial activity of residents living closer to refugee routes fell considerably, and while anti-migrant sentiment increased in these areas, attitudes towards other minorities remained unchanged.

Does exposure to mass migration affect the economic behaviour, attitudes, and beliefs of the native population in transit countries? While there have been many studies of the implications of migration for sending or receiving societies (see e.g. Becker and Ferrara 2019), there is little empirical research on the effects of migration on transit countries, through which migratory flows move in order to reach their destination country.

Transit migration is fundamentally different from permanent migration in ways that are relevant to its socio-economic impact. Following seminal work by Allport et al. (1954), many papers have described the conditions that increase empathy and integration between in-group and out-group individuals (e.g. Barlow et al. 2009). These conditions are hard to meet in the context of transit migration. In particular, when migrant flows pass through transit locations temporarily, there is little opportunity for repeated social interactions, which may result in reinforcing rather than dispelling pre-existing stereotypes.1

In our recent paper (Ajzenman et al. 2020), we provide the first large-scale evidence on the impact of transit migration by studying localities along the Eastern Mediterranean route and its extensions in Europe that were particularly exposed to large transit migration flows in 2015. We show that exposure to transit migration increases prejudice and anti-foreigner sentiment. It reduces trust in institutions, increases the aversion to risk, and decreases entrepreneurship among the native population.

The Eastern Mediterranean route

During the 2015 refugee crisis, the Eastern Mediterranean route served as the primary gateway to Europe. Migrants who entered the EU (Bulgaria or Greece) via Turkey by land or sea then travelled through Western Balkan countries – Albania, Bosnia and Herzegovina, Croatia, North Macedonia, Kosovo, Montenegro, Serbia, and Slovenia – with the aim of reaching Western Europe (Aksoy and Poutvaara 2019).

As Figure 1 shows, prior to the refugee crisis, the flow of asylum seekers arriving in Europe was relatively small. These numbers imply a massive unexpected shock experienced by the transit country communities, most of them located along the Eastern Mediterranean route.

Figure 1 Asylum seekers in Europe

Source: Eurostat. Authors' calculations. 

In order to identify the exposure to transit migration, we use a unique locality-level panel from the 2010 and 2016 rounds of the Life in Transition Survey (EBRD 2016) and geo-localised data on the main land routes taken by migrants in 18 European countries during the refugee crisis in 2015, which let us calculate the distance of each locality to the closest migration route. Overall, we analyse a panel of 822 different localities in 18 different European countries, as shown in Figure 2.

Figure 2 Migration routes

Source: Life in Transition Survey and IOM. Notes: The map illustrates all localities used in the sample. Names and boundaries do not imply official endorsement or acceptance by the EBRD or IOM.

To identify the causal effect, we implement a distance-based instrumental variable approach (similar to Ghani et al. 2016 and Faber 2014). More specifically, we construct an instrument based on the distance from each locality in LiTS data to the ‘optimal’ migration routes, which are routes that minimise the walking time between the main origin countries (Syria, Iraq, and Afghanistan) and the main destination countries (Germany and Italy). Therefore, our instrument captures the variation in distance between each locality and the real migration routes chosen by refugees, which was induced by ex-ante and plausibly exogenous geographical determinants.

Rising anti-migrant sentiments

By comparing the same localities before and after the mass migration episode, we find important changes in attitudes and beliefs. First, we document a major rise in negative attitudes towards migrants. Halving the distance to migrant routes results in an increase of 1.8 percentage points in the proportion of people that consider migrants a burden, an increase of 2.4 percentage points in the proportion of people who would prefer not to have migrants as neighbours, while attitudes towards other minorities (people of a different race or religion, for instance) did not change.

Decrease in entrepreneurship

We use two different measures of entrepreneurship: (i) having tried to set up a business and (ii) self-employment status. Figure 3 shows that both outcomes, which were very similar pre-crisis across localities – whether close to or far from migration routes – behaved very differently after the crisis. In the localities that were relatively close to migration routes (within a radius of 25 km), the proportion of entrepreneurs dropped significantly in 2016 (after the crisis). In the localities that were farther away, both measures of entrepreneurship remained unchanged.

Figure 3 Effect of exposure to transit migration on entrepreneurship

Notes: ‘Tried to start a business’ takes a value of one if the individual answered positively to the question “Have you ever tried to set up a business?” and zero otherwise. ‘Self-employment’ takes a value of one if the individual declared themselves self-employed at the moment the survey was conducted and zero otherwise. The charts show the percentages of positive responses. In both charts, ‘Before’ refers to 2010 and ‘After’ to 2016. ‘Treated’ are localities within a radius of 25 km from the closest migration route. The chart shows 95% confidence intervals.

This pattern is confirmed when estimating our instrumental variables model: halving the distance to migrant routes decreases the propensity to set up a business by 2.8 percentage points and the likelihood of being self-employed by 2.4 percentage points (the respective averages of these two variables before the refugee crisis were 14% and 9%).

Why does transit migration affect entrepreneurship?

Migration could affect entrepreneurship through several channels, such as a demand shock, a drop in interpersonal or institutional trust, or a change in risk attitudes. We explore each of these potential mechanisms and find that the fall in entrepreneurship is likely explained by changes in attitudes and beliefs related to trust in institutions and confidence in government, and a decline in the willingness to take risks. In particular, we show that individuals exposed to the migration shock increased their perception of political instability, and reduced their belief that law and order existed in their countries. They also had less trust in all levels of government.

On the other hand, we find no evidence that exposure to transit migration affects interpersonal trust or local labour market outcomes, such as unemployment or labour force participation. Using luminosity data, we also document a null effect on local economic activity.

Policy implications

Our results highlight the importance of understanding the impact of different types of migration. Policymakers should be aware of and try to mitigate the negative implications of exposure to large transit migration flows, when there is little opportunity for repeated social interactions between the native population and migrants/refugees.


Ajzenman, N, C G Aksoy and S Guriev (2020), “Exposure to Transit Migration, Public Attitudes and Entrepreneurship”, CEPR Working Paper No. 14605.

Aksoy, C G and P Poutvaara (2019), “Refugees’ and Irregular Migrants’ Self-selection into Europe: Who Migrates Where?”, CESifo Working Paper No. 7781.

Allport, G W, K Clark and T Pettigrew (1954), The Nature of Prejudice, Addison-Wesley Reading, MA.

Barlow, F K, W R Louis and M Hewstone (2009), “Rejected! Cognitions of rejection and intergroup anxiety as mediators of the impact of cross-group friendships on prejudice”, British Journal of Social Psychology 48(3): 389-405.

Becker, S O and A Ferrara (2019), “Consequences of forced migration: A survey of recent findings”, Labour Economics 59(C): 1-16.

European Bank for Reconstruction and Development (2016), Life in Transition Survey

Pew Research Center (2016), “Number of Refugees to Europe Surges to Record 1.3 Million in 2015”, Washington, DC, 2 August.

Steinmayr, A (2020), “Contact versus Exposure: Refugee Presence and Voting for the Far-Right”, Review of Economics and Statistics, forthcoming.


1 For instance, Steinmayr (2020) shows that vote shares of the far-right political party increased in Austrian municipalities that were exposed to mass transit migration, while the opposite was true in communities where refugees actually settled.

1,260 Reads