How are immigrants and refugees in Sweden doing?

We use administrative Swedish data to show that, conditional on parent income, immigrant children have similar incomes and higher educational attainment in adulthood than native-born Swedes. This result, however, masks the fact that immigrant children born into poor families are more likely than similar natives to both reach the top of the income distribution and to stay at the bottom. Immigrant children from high-income families are also more likely than natives to regress to the economic bottom. Notably, however, children from predominantly-refugee sending countries like Bosnia, Syria, and Iran have higher intergenerational mobility than the average immigrant child in Sweden.

That paper, co-authored with Cristina Bratu, is from Valentin Bolotnyy, who is on the job market from Harvard.  Here is my post on his job market paper.


disclosure: the bottom in Sweden is not same as the bottom you imagine

Taking Fig 4 at face value. Something odd must be going on with their income distribution / immigration system in Sweden if migrants from USA to Sweden fall at 25th percentile of Swedish income distribution and their children at 35th. Not they are alone, as it seems Norwegian parents and children in Sweden roughly at the same position.

A whole host of low income migration from Western and Northern Europe which is bizarrely the opposite to what you see in terms of migration from there to Britain and the USA. Very strange.

Immigrants form Spain also appears to be an outlier - parents and children both particularly unsuccessful. What's with that?

And similarly anomalous results for Germans. On its face, it's somewhat implausible.

I mean, imaginable could happen. Welfare shopping, hey? No ambitious people migrating because of limited prospects for ambition under social democracy? So weird exception to Western European migrants generally to be more vastly more qualified and capable than natives to find it worthwhile to migrate. But demands explanation.

America isn't sending its best to Sweden.

That makes sense. Winner take all economies means those moving to Sweden might be less economically oriented.

"conditional on parent income, immigrant children have similar incomes [...] in adulthood than native-born Swede" So, this means that immigrant children do not really better than their parents, at least no more than kids of poor Swedes in general do (regression to the mean?). Yet these immigrants kids have immense advantages over their parents: they are born in Sweden or came early; they learned the language as a kid, while their parents probably never learnt it completely; they benefited from the Swedish school system, etc. Kids of Swedes had none of these advantages over their parents since by definition they parents were already Swedes and went to Swedish school, etc. The results of this paper, if correct, point to huge failure of integration of this new generation of immigrants kid in Sweden.

do much

Good catch. And they pretend to not understand why Taleb calls economists charlatans (though to be fair to the author the poor fellow is on job market and has to preference-falsify to survive).

Here is an interview with Tino Sanandaji, a Swedish economist of Middle Eastern ancestry.

Money quotes:

" 1990 non-European immigrants accounted for only 3% of the population and any problems could be isolated and managed within the bigger framework of society. That figure has increased to some 13-14% now, and is growing at perhaps 1-2 percentage points from last year, with persistent gaps in income, unemployment and education."

"...immigrants on average earn 40% less than the natives..."

"Research shows that the tipping point for [white] flight to occur is very low: after 4% of non-European immigrants the native Swedes start to move out."

"If you look at GDP and population growth figures projected by the government, we are seeing something that I had never seen before: projected negative GDP per capita growth rates in a period of economic cycle recovery . The only reason for that is immigration; Sweden is bringing in a lot of people who consume but do not produce much."

"Just the initial cost for those asylum seekers is 1.5% of our GDP, significantly higher than our defense budget at around 1%. And that does not count the net costs associated with housing, health, welfare spending and so forth that arise later."

"’s probably 50 or 100 times more expensive in Sweden...the 3,000 people housed in these tents are going to cost more than the biggest refugee camp they built in Jordan, for perhaps 100,000 Syrian refugees."

" Since the OECD started with Pisa tests no other country has crashed in the scores as much as Sweden. In Western Europe we are already second last after Greece. Swedish policies played the major role here, but immigration may explain about 30% of that decline."

"...we have serious scientific polls that clearly state that the majority or plurality of Swedes support reducing refugee immigration, even going back as far as the 1980s ... It’s the elite opinion that forms the consensus that Sweden should take many more immigrants – it’s almost like a religion, but it is not the popular view."

"...[the far right] party has its roots in racism and it was taboo to vote for them. Despite that they went from 3% around ten years ago to perhaps 20% now."

There's also a significant cost in mounting insecurity and crime. " About 58% of men convicted in Sweden of rape and attempted rape over the past five years were born abroad, according to data from Swedish national TV."

Similarly in Germany ( from the WSJ Oct 15 2018): " Refugees and asy­lum seek­ers make up only about 2% of Ger­many’s 82 mil­lion strong pop­u­la­tion but 8.5% of all crime sus­pects in 2017, ac­cord­ing to po­lice sta­tistics. They made up 14% of sus­pects for un­law­ful killing and as­sault, and a sim­i­lar pro­por­tion of rob­bery sus­pects.

At some point, the damage the “elites” are inflicting on their own society begins to look intentional. One wonders what they expect the end state to look like.

All depends on what they consider "their society", doesn't it? If you could reinvent manorialism on a global scale, wouldn't you?

I couldn't help but notice that the number of convictions on the Sweden article is incredibly small compared to the number of accusations. I'd want to see more information before I'll fully accept cause and effect there, however. It could also be that foreigners are easier to identify and convict, so they could still form an outsize portion of convictions but are still a small minority of perpetrators. It's too inconclusive without knowing more about the demographics of the reported.

This paper - and most of the immigration discussions I've seen - appears to be silent on dosage effect. They do note the very high fraction of immigrants now in the population.

I would expect to see a difference in integration success between the initial relatively rare immigrants, and the later not rare (perhaps even overwhelming to integration mechanisms) immigrants.

It would also be interesting to assess the cultural gap between the immigrants and the Sweden, for the various immigrant groups, by country and time frame, and whether the Swedish expectations for immigrant integration (becoming Swedish) vs. maintenance of cultural identity has changed.

It would be very interesting to study dosage. The conventional wisdom is that more dosage means assimilation is harder, but sometimes the reverse is true. Smaller groups can sometimes have more social cohesion than large groups, and are also more vulnerable to discrimination from the broader society. In some cases, this could mean that smaller groups of immigrants would actually assimilate less well than larger groups.

For example, Asian immigrants are much better assimilated in California where there are a lot of them. In your typical Midwestern town with only a few Asians, they will often have a relatively narrow set of similar professions, stick together, and rarely venture into the broader society. But in California you see Asians with all different kinds of professions and social circles, and mixed-race friendships are all over the place. Asians in California are even about as Republican as whites, in some ways the ultimate sign of assimilation.

"For example, Asian immigrants are much better assimilated in California where there are a lot of them. In your typical Midwestern town with only a few Asians"

California has had a large Asian population for generations. That's not true of a typical midwestern town.

Conditioning on parents' income is interesting for academic purposes but not what critics care about. What matters is the unconditional convergence of immigrants. What we learn from this study is that children of good families do well. Children from low income households -- undoubtedly biased towards Africa and the Mideast or their descendants who came through other countries -- don't converge to the average. Which means they do not become ordinary Swedes. This is very bad news because this means that these immigrants are like low end Latino immigrants in the US who don't converge to the US norm. Related work in the US on Latino convergence also controls for income and education, which means that the uncontrolled convergence, which you might be more interested in as an average person, is not great. Note that historically most other immigrant groups to the US have converged to the US average or, in the case of Asians, done even better.

It would appear that if you are motivated to be financially successful you don't emigrate to Sweden - unless you are a refugee and are desperate to change your address. Then some highly motivated refugees do well.

Yes, agreed. If you're successful in your home country, you don't have reason to move. However, if you're a good, productive person, but happen to live in a warzone, then migrating to Sweden is a great idea. But migrating to Sweden is a great idea no matter who you are, if you live in a warzone. But if you don't? Then migrating to Sweden is likely a step backward, unless their safety net is better than your home country's, and you operate on the safety-net level of society.

Comments for this post are closed