Understanding (Anti-) Immigration Attitudes

Survey responses and support for anti-immigration parties suggest that most voters in Western countries desire reduced immigration to their countries. Such anti-immigration attitudes are quite stable within societies and individuals. European societies have been opposed to immigration for decades and individual citizens also rarely change their immigration attitudes. Consistent with this evidence, a large literature that explicitly tried to change immigration attitudes (toward more positive views regarding immigrants) largely fails to do so. Thus, anti-immigration attitudes can be though as a persistent feature of (modern Western) politics.

But why are anti-immigration attitudes so pronounced? In the following, I will provide the best short explanation I can give. This is, of course, a huge field and I refer the interested reader to, e.g., this book by Alexander Kustov. Still, here is my take that, hopefully, includes some original ideas:

What do anti-immigration attitudes actually mean?

If you ask a representative sample of citizens from any Western country whether immigration should be reduced or increased, the majority will most certainly lean toward a decrease. This response is what scientists often mean when they say people are anti-immigration.

But this question is very vague and undifferentiated. In particular, it lumps together very different immigrant groups. I think distinguishing different types of immigrants is key to understand immigration attitudes. The most important distinction is probably between asylum seekers and other immigrants.

Asylum seekers vs other immigrants

Legal differences

Asylum seekers are a subgroup of immigrants who usually make up a small share of all immigration. Asylum seekers differ from other immigrants regarding their legal status, demographic characteristics, and their effect on the host society. Regarding their legal status, asylum seekers distinguish themselves by claiming asylum. Claiming asylum is a right that most countries acknowledge through their signing of the 1951 United Nations Geneva Convention relating to the Status of Refugees and the 1967 New York Protocol. Signature countries grant anyone the right to apply for asylum and to protect the claimant as long as his claim is being processed. Loosely speaking, the conventions guarantee asylum to anyone who is either fleeing persecution or is threatened by indiscriminate violence, like a civil war, in his homeland.

In the public discourse, asylum seekers are usually called “refugees” which can be misleading. Everyone can apply for asylum, but not everyone is legally eligible for it since not everyone is really fleeing war or persecution. To determine whether an asylum seeker is eligible, applications are processed by the national migration agency. According to these agencies, a large share (depending on the country and region but very roughly half on average) does not flee war or persecution but applies nonetheless. In practice, most asylum seekers, including those whose application is rejected, usually stay in the host country in the long run.

Hence, one can think of the immigration system of most Western countries as having two channels. First, there is the standard channel immigrants use after acquiring an entry permit, usually because they found a place to work or study. Second, there is the asylum channel that can be used by everyone, independent on whether they found a place to work, whether they have relevant qualifications or any connection to the host country at all. It is plausible to assume that most of the immigrants who benefit the host country will choose the first channel while those who would have no chance to acquire entry permit can only use the asylum channel.

Different effects on host countries

There are also major demographic differences between asylum seekers and other immigrants and the two groups have different effects on host countries. Regarding demographic characteristics, asylum seekers in Europe are much more likely than other immigrants to come from developing countries, in particular the Middle East and Africa, to be Muslim, young, male, and have much lower levels of education.

Moreover, immigration by asylum seekers has different effects on natives than other immigration. For instance, there is no evidence that immigrants overall increase crime rates, but recent studies find that immigration by asylum seekers does increase crime rates strongly.

At the same time, their fiscal impact is much more negative than the effect of other immigrants. The figure below is taken from the magazine “The Economist” but based on data from the Danish Finance Ministry. It shows the net contribution of several immigrant groups to the Danish public finances.

Figure 1: Net contribution of selected immigrant groups to Danish public finances (taken from here)

As can be seen, there is no huge difference between native Danes and other Western immigrants. However, immigrants from the Middle East, North Africa, Pakistan and Turkey have, in total, a large negative effect on the public finances. Strikingly, even working-age immigrants from these regions are, in sum, a strain on the public finances. While the overlap with asylum seekers is not perfect, many immigrants from these regions come through the asylum system. Similar results have also been found in other countries (e.g., in the Netherlands).

It is true that most economists think that immigration can boost economic growth. One key argument is that people can go where they are most productive. But this logic does not apply generally and the asylum system in combination with generous welfare states undermines it. It is thus no wonder that immigration outside of the asylum system is probably beneficial for host countries while immigration through the asylum system is economically costly.

These negative effects on the host countries are not very surprising because the asylum system was never intended to benefit the host country but to benefit asylum seekers. It was created as a reaction to the widespread persecution and dislocation during and after World War 2 and intended to ensure that those fleeing war or persecution would find a save haven. Hence, the main argument in favor of the asylum system is a humanitarian one.

As a result of these various differences, natives might hold different attitudes toward asylum seekers and other immigrants. On the one hand, natives might be more opposed to asylum seekers than other immigrants since their effects on the host country are more negative. On the other hand, natives might also be particularly open to asylum-seeker immigration due to their special humanitarian situation.

Anti-immigrant means anti-asylum seeker

Figure 2 visualizes the attitudes of Germans regarding asylum seekers (top) and other immigrants (bottom). I elicited these responses from a representative sample of German adults in 2021.

Figure 2: Responses of a representative sample of German adults. Taken from this paper.

As can be seen, a majority of the subjects wants much fewer asylum seekers to enter their country. However, there is no clear tendency to want fewer other asylum seekers. Hence, anti-immigration are largely confined to asylum seekers.

Importantly, this paper find that most respondents think about asylum seekers when they respond to standard “do you want more or fewer immigrants” items. The author argues that this might be because asylum seekers are much easier to notice and distinguish from the native population (via visual cues) than other immigrants. Hence, when you ask people about their immigration views, they effectively tell you their views regarding asylum seekers.

This is quite important because it can be challenging to rationalize why, say, a German would be strongly opposed to immigration by Danish people who, on average, contribute as much to society as native Germans. In contrast, it is easy to come up with good reasons for why Germans would be opposed to asylum seekers from the Middle East: because it is costly for them and leads to more crime. Of course, asylum seeker immigration still has the positive effect of improving the lives of the asylum seekers.

The ingroup-outgroup trade-off

This leads to a trade-off that, I think, is key to understanding modern cultural politics: how many foreigner lives are worth the same as one native life. Or, maybe less pointed: if I gave you 1,000$ and you have to distribute them between a stranger from your country and a stranger from another, randomly selected, country, how would you split the money?

Researchers have actually done that to study how people trade-off well-being of their ingroup and outgroups. Figure 3 shows a histogram that visualises how their participants split the money. In particular, it shows what percentage subjects gave to the foreigner. Note that this histogram pools responses across broadly representative samples from 60 countries.

Figure 3: Percentage of 1,000$ people give to a stranger from another county instead of to a stranger from their own country. Taken from this paper.

As can be seen, about half of all participants do split the 1,000$ equally (in this experiment they used hypothetical money but validated this procedure in another paper). However, most of those do not split equally, give more to the stranger from their own country, and nearly 20% give everything to their compatriot. On average, there is thus a tendency to give more to compatriots than to foreigners. Note that people could not keep anything for themselves. Had the trade-off been between themselves and a foreigner, the distribution would probably shift strongly to the left.

This result applies in all countries analyzed and tends to be just as strong in European countries (even though there are interesting between-country differences). This suggests that, for many Europeans, the negative consequences of asylum seeker immigration on their country weigh particularly heavy. Potentially so heavy that they outweigh the benefits for the asylum seekers.

The ingroup-outgroup trade-off applied to asylum seekers

This is also what I usually read in anonymous responses to free text questions of the form “what are your thoughts on the immigration policy of your country”. In contrast to what the media would make you believe, people do not seem very polarized or extremist to me. When given the opportunity to express themselves anonymously, most describe complex and partly conflicting attitudes. On the one hand, they want to help asylum seekers make a better future. On the other hand, they want their countries to stay safe and wealthy. When forced to decide they prioritize their own well-being and that of their compatriots.

This paper assessed such trade-offs in a more structural way. To that end, they conducted a “vignette-study with representative samples from 15 European countries. They showed subjects vignettes of hypothetical asylum seekers and asked participants whether they want the asylum seeker to come to their country or not. Each vignette contained attributes such as the asylum seeker’s gender, country of origin, and former occupation. Importantly, the authors randomly assigned one of several potential attributes to each of these categories. For instance, one subject would see an asylum seeker vignette of a young Muslim man and then an old Christian woman. Another subject might first see a middle-aged Agnostic man and so on. In practice, vignettes contained 8 different categories (shown in Figure 4) and each subject saw many different vignettes.

From this data, it is possible to estimate how each attribute affects the probability for an asylum seekers to be accepted by Europeans. Figure 4 visualizes these effects as points, always in comparison to a reference within each category. For instance, the point next to “Major inconsistencies” near the top means that, if the testimony of the asylum seeker has major inconsistencies (suggesting he is not fleeing war or persecution) this lowers the probability that Europeans accept him by about 10percentage points.

Figure 4: Effects of asylum seeker characteristics on the probability that Europeans accept them. Taken from this paper.

The authors included humanitarian and economic considerations in their analysis (but not crime). As the estimates for the “Vulnerability” and “Reason for migrating” categories show, Europeans do take humanitarian considerations serious. They are much more likely to accept those fleeing persecution and those who would be tortured at home. However, economic concerns also matter a great deal. Europeans are much less likely to accept those who do not speak their country (and most asylum seekers don’t) and worked in simple occupations (and most asylum seekers did).

If you map these preferences to the actual attributes asylum seekers have, you find that Europeans would not accept asylum seekers if they knew them. This is, to a large extent, driven by the fact that they are an economic burden and that humanitarian concerns do not weigh heavily enough to compensate for this. Europeans are willing to make some limited sacrifices to benefit foreigners, many donate, for instance. However, all preferences have a limit and the huge costs that asylum immigration has placed on the host countries may have simply overstretched the limits of altruism.

Why are people anti-immigrant

Let me now answer the question I mentioned at the beginning: why are anti-immigration attitudes so pronounced? I think people are strongly opposed to asylum seeker immigration because such immigration makes their lives and those of their compatriots much worse, in the form of increased crime and taxes, which are needed finance the asylum seekers who are responsible for the increased crime. One can easily imagine how such a situation can make Europeans feel like they are being invaded and oppressed by an foreign horde, that comes to their country against their will, takes more economic goods than it gives back and brings crime. This is what many people have in their heads when you talk to them about immigration.

In contrast, researchers and probably also most mainstream politicians think about the average immigrants, i.e. they are less likely to overweigh asylum seekers. I think this can partly account for the vastly different immigration attitudes that the societal leadership and “ordinary” people have.

A Misunderstanding

However, the more important implication might be that the anti-immigration attitudes of voters and the pro-immigration attitudes of leaders might be reconcilable. When voters say they want less immigration, what they really mean is that they want fewer asylum seekers, e.g., through tougher asylum laws, more deportation of former asylum seekers, less social benefits for them etc. When researchers oppose anti-immigration sentiments what they mostly want is to preserve the free movement of workers across countries, e.g., they want it to be easy for a Koran scientist to get a Visa in the USA. These attitudes are reconcilable since they refer to different types of immigration. Hence, anti-immigration movements and the rise of anti-immigration parties do not necessarily conflict with the goals of most experts.

I would even go one step further and claim that both groups, experts and “ordinary” people are right. Most voters basically just say that their societies would be better off if we would abolish or at least severely cut down the asylum system. I think they are right because asylum immigration has negative economic and societal effects on the host countries. Most experts basically say that societies would be much worse off if we would hinder regular migration, thereby making it harder for people to work where they are most efficient. They are also right. Migration outside of the asylum system largely has positive effects on the host country.

Crucially, because most immigrants do not come through the asylum system, immigration overall might often have a positive net effect on the host countries too. However, that does not mean that all types of immigration are positive and asylum immigration is a particular, salient type that has negative consequences. In that sense, we can think of voters as having identified a way to improve the immigration system, by abolishing a particular type of immigration that is largely negative.

To make this point even clearer, since I think it is really important, consider the example of a company with several workers. Of course, the company would not be profitable if it would fire all of its workers. However, there may be a few workers who really harm the company e.g., because they do a particularly bad job, steal etc. In that example, voters identified a group of such bad workers and demand that they are fired. Experts respond that the company needs workers, since it cannot operate without any workers. Both are right and there is no contradiction at all because one could just fire the few bad workers and keep the rest, thereby increasing productivity.

In that sense, the “puzzle” —evidence suggests immigration is good while people are anti-immigration— is no puzzle at all. It only arises in the heads of researchers because they do not make the effort to understand well what voters’ anti-immigration attitudes mean.

Potentially, this misunderstanding results from a tendency of researchers to take everything literally (and, taken literally, immigration does not strongly increase crime rates) and to be of low social intelligence. These traits prevent them from understanding that voters do not mean everything literally. Political correctness may also be a reason for this misunderstanding. Since asylum seekers happen to come from specific regions and to follow particular religions, criticising the asylum system is often misunderstood as racism or something similar. Hence, the average vote should not reveal his true opinion on immigration because if he did so, he must fear to be insulted and harassed. Ironically, this fear also stems from the behavior of researchers who are at the forefront of political correctness. It is precisely the atmosphere of fear they helped to create that now makes people hide their true opinions, thereby hindering researchers to understand attitudes and design a working immigration system.

A new immigration literature

I think it is high-time for researchers to change their strategy. One can approach the seeming disconnect between the scientific evidence on (i) the benefits of immigration and (ii) popular anti-immigration attitudes in two ways. First, one can ask “why are voters so stupid and how can we change them?” Second, one can ask “what part of voters’ dissatisfaction is due to real problems and how can we change the immigration system to solve these problems?”

The existing literature has, nearly without exception, tried to answer the first question. In the beginning, researchers tried to change people using light-touch interventions, like Information campaigns. But since these interventions did not really work, researchers tended to suggest ever more extreme approaches, entering a slippery road. For instance, during a conference dinner, one assistant professor from a leading European university suggested to order the army to force people at gunpoint into cinemas where they would have to watch propaganda movies in which all natives would be portrayed as evil while all immigrants would be portrayed as good.

In addition to raising concerns about the sanity of such researchers, this example illustrates that we have reached the end of this slippery road. Following it further down toward more paternalistic measures would turn the field into a left-wing propaganda machine and destroy the already low trust many voters have in researchers. Why then should voters finance such “research” with their taxes and why should politicians not severely cut funds as a result?

It is therefore high-time to turn to the other question “what part of voters’ dissatisfaction is due to real problems and how can we change the immigration system to solve these problems?” Instead of insulting voters we should use their attitudes as valuable inputs that can help to guide us toward real problems. Following my observations, I propose that voters identified a part of our immigration system, the asylum system, that does have mainly negative effects. Hence, let us do research on whether this proposition is correct. If so, let us work on how to reform or abolish the asylum system.

Thanks to Javier Granados Samayoa for his comment on this related post: Fake News Don't Decide Elections. I wrote the current post to answer his question and I hope it provides some food for thought! As I just joined this platform, my number of followers is still small. That gives me the opportunity go answer questions in detail. So please feel free to leave a question or comment. Again, there is a lot to unpack here and I can already see many potential responses that require their own post. That’s great, isn’t it?