A quick look at immigration and crime in Germany

The Ministry of the Interior in Germany released data about crime a few months ago. Ines Laufer had a look at them on her blog and noted they showed that immigrants in general and asylum seekers in particular were massively overrepresented among criminals. I wrote a thread on Twitter where I discuss some of the patterns she found in the data. I don’t have time to write a detailed post, so I’m just going to copy the thread here.

Now, this is Twitter, so obviously I couldn’t go into details and had to ignore a lot of complications. But as I promised recently, as soon as I have more time, I plan to write extensively about the effect of immigration in Europe. In the meantime, I wanted to share these statistics, because I hadn’t seen the figures for 2016 yet and they are pretty striking.

Germany after 2015 is basically a natural experiment on what happens when a country does the kind of stupid shit libertarians and leftists advocate on immigration. It never ceases to amaze me how otherwise intelligent people can delude themselves into believing even the most absurd nonsense. For instance, after Merkel decided to open the floodgates in 2015, the cognoscenti were confidently asserting that, by letting in more than a million people who for the most part didn’t speak a word of German, were largely uneducated by Western standards and had a totally different culture, Germany would soon become a land of milk and honey. But then reality happened and, as the Financial Times recently pointed out, it wasn’t pretty:

Initially, the influx of so many working-age, highly-motivated immigrants spurred optimism that they would mitigate Germany’s acute skills shortage and solve the demographic crisis posed by its dangerously low birth rate. Dieter Zetsche, chief executive of carmaker Daimler, said the refugees could lay the foundation for the “next German economic miracle”.


But those hopes have faded as a new realism about the migrants’ lack of qualifications and language skills sinks in. “There has been a shift in perceptions,” Ms Özoğuz [the German commissioner for immigration, refugees and integration] told the FT. Many of the first Syrian refugees to arrive in Germany were doctors and engineers, but they were succeeded by “many, many more who lacked skills”.

A recent report by the Institute for Employment Research (IAB) found that only 45 per cent of Syrian refugees in Germany have a school-leaving certificate and 23 per cent a college degree.


Statistics from the Federal Labour Agency show the employment rate among refugees stands at just 17 per cent. It said 484,000 of the refugees are looking for work, up from 322,000 last July — an increase of 50 per cent.


Of those, 178,500 are officially unemployed, meaning they not only have no work but are not enrolled in any training programmes or language courses — up 27 per cent on last July.

Of course, this is not surprising in the least, what is surprising is that so many people were stupid enough as to think it wasn’t going to happen. But the most amazing thing is that you can be certain that, despite this fiasco, the sophisticates will continue to treat anyone who voice skepticism about the benefits of mass immigration in Europe as a bigoted cretin. To be convinced of your intellectual and moral superiority when you are making claims that are manifestly absurd is perhaps the worst kind of stupidity.

20 thoughts

  1. Yes, it’s very easy to be hospitable and open-minded when the people in question
    will not settle in your neighborhood nor in your apartment building.

  2. Thanks for the interesting data Philippe. I think nevertheless that it’s hard to generalize them as you seem to do, because sometimes data seems to show the reverse : http://econlog.econlib.org/archives/2011/08/immigration_and_3.html (I know that Caplan is not your cup of tea, but the data are still relevant!). More important in my view is also the fact that these data, even if generalized, would not definitely convince me that anti-immigrationists are right. Crime rate doesn’t justify banning immigrants by itself (maybe because they have a right to immigrate, or maybe because immigration is still beneficial economically for the community at large), or if it does, then it seems to prove too much: http://bleedingheartlibertarians.com/2016/01/the-crime-objection-to-open-borders/

    1. Actually, I quite like Caplan, as long as he doesn’t talk about immigration 🙂 Then, as most libertarians, he says a lot of stupid shit. In the post you mention, he makes a comparison which suggests that immigrants in the US are mostly in prison for violating immigration law and commit other crimes at lower rates than natives. But there are so many problems with his comparison that I don’t even know where to begin.

      First, I may not have been clear, but I wasn’t trying to generalize. I was talking about the effect of immigration on crime in Europe, whereas Caplan’s data are about the US. All the data on immigration and crime I have ever looked at, including the data about Germany I discuss in my tweetstorm, show that crime rates vary a lot depending on national origin. Since immigrants in the US tend to come from different countries than in Europe, there is no reason to expect that they’d also be overrepresented among criminals. So even if Caplan was right about the US, which he isn’t (more on that shortly), it wouldn’t give us any reason to doubt that immigrants are overrepresented among criminals in Europe.

      Also, just to be clear, I agree that the fact that immigrants are overrepresented among criminals by itself doesn’t show that immigration is bad. But 1) I don’t think I ever made that claim, 2) it’s still a relevant consideration and 3) the elites constantly deny it. (As you know, in France, anyone who makes the connection between immigration and crime is vilified.) If only because of 3, I think it’s worth pointing out what the data actually show, which is quite different from the nonsense you can find in the media on that issue. (To be clear, I’m not trying to suggest that you disagree with that, I’m sure you don’t.)

      Okay, now that I have clarified this point, let me turn to Caplan’s comparison and explain why it doesn’t show what he thinks it shows or wants people to believe it shows. First, in this blog post, the data he is looking at are about prisoners in federal prisons. However, federal prisons in the US hold only a small minority of prisoners, the vast majority of whom are held in state prisons and county jails. Moreover, if we’re looking at the kinds of crime for which people are incarcerated (as Caplan is doing), then federal prisons are totally unrepresentative. In particular, people who are incarcerated for both immigration law violations and drug-related crimes are vastly overrepresented, which is why Caplan’s comparison misleadingly gives the impression that immigrants in the US are mostly incarcerated for those kinds of crimes. It’s hard to believe that Caplan doesn’t know that, so I strongly suspect he is just willingly trying to mislead people.

      Finally, although people constantly say that data about the carceral population in the US show immigrants commit less crimes than natives, the truth is that the data we have show no such thing because they are total garbage. Indeed, as a Center for Immigration Studies report I mentioned a few months ago explains, most of the data on immigration status is imputed, because this information is often not available. The reason why it’s often not available is largely because people like Caplan, who are pushing a pro-immigration agenda, are doing their best so that it’s not collected… As I explained when I shared that report, since immigrants tend to have socio-economic characteristics associated with higher crime rates, the safest bet is that they are overrepresented among criminals. But the truth is that, in the case of the US, we simply don’t know for sure. In the case of Europe, on the other hand, we do know.

      1. Interesting post.

        I don’t think we can claim a correlation between higher crime and immigration from these meager statistics. These German statistics capture only one year. Also, as someone else on your Twitter feed pointed out, are these crimes simply someone being charged or the outcome of a court case. I also think that the word foreigner and asylum seeker can mean any kind of “immigrant.”

        I would also caution against using Center for Immigration Studied reports. They are notoriously bad and have close to zero scholarly value.

        It shouldn’t surprise anyone that later waves of migrants to Germany would present a problem, especially in learning the language and acquiring a job. German is notoriously difficult to learn, and I imagine that acquiring the necessary language skills would be a time consuming process. The real problem is that German policymakers have not put plans in place to deal with these under skilled migrants. I’d be shocked if anyone in Germany thought that immigrants would be a magic bullet cure all for the demographic problem plaguing many European countries. The German minister you quote above said just as much in further interviews.

        1. These German statistics capture only one year.

          Do you think that disparities such as these are going to disappear from one year to the next? (To be clear, that’s a rhetorical question.)

          Also, as someone else on your Twitter feed pointed out, are these crimes simply someone being charged or the outcome of a court case.

          They’re people who were investigated by the police and prosecuted. It’s unclear to me whether they had to be convicted in order to be counted. In any case, bias is a possibility, but it’s unlikely to make much of a difference for reasons I explained on Twitter. Another reason to doubt that bias plays a large role is that crime rates vary a lot depending on national origin, even though cops often can’t distinguish between immigrants who are ethnically similar but come from different countries.

          I also think that the word foreigner and asylum seeker can mean any kind of “immigrant.”

          I’m not sure what your point is here.

          I would also caution against using Center for Immigration Studied reports. They are notoriously bad and have close to zero scholarly value.

          The Center for Immigration Studies is frequently vilified, especially by groups such as the SPLC, but they usually rest on defamation. I think the report I mentioned is excellent and conclusively shows that we can’t draw any conclusions about the rates at which immigrants offend in the US based on data about prisoners/inmates. If you think this report does not show that, you need to explain why, because otherwise I don’t see why I should take your warning seriously.

          I’d be shocked if anyone in Germany thought that immigrants would be a magic bullet cure all for the demographic problem plaguing many European countries.

          I could literally quoted hundreds of people who say just that. More generally, the belief that immigration is good for Europe because it’s the only way to prevent the aging of the population is widespread among the sophisticates, despite the fact that every single article on that issue showed it wasn’t the case. I will probably write about this question eventually, since it’s such a widespread myth.

    2. The linked argument is the one that proves too much. Essentially no restrictions on the movement of citizens within the United States are permitted by our system, so once you equate controls on immigration with controls on internal movement by citizens, no restriction on immigration for any reason can survive. Suspected Russian spies, known Nazi war criminals, whatever: all are welcome to enter the country at any time and remain indefinitely, just as citizens are.

      1. Yes, I didn’t say anything about Brennan’s post in my reply because I plan to publish a detailed criticism of this line of argument, which libertarians use all the time to defend open borders. (Brennan also makes empirical assertions which are false, although he doesn’t know it, but that’s another story, which I will also address at length in the future.) Beyond the interest of debunking the conclusion of this argument, I think it’s important to discuss it because the assumption on which it implicitly rests shows that libertarians simply don’t understand what a political society is, which is why they are never taken seriously by more than a small minority of people and never will be. Anyway, I don’t want to say too much here, since again I plan to write a post about this.

        1. I like both what you write and what Brennan, Huemer, (David) Friedman among other empirically oriented libertarians write. So far I’ve been convinced that there’s a prima facie moral case in favor of open borders, but I happily grant that there can be overriding considerations that could in the end weight in favor of some kind of restricted immigration (indeed, I’m not very much convinced that there are no major shortcomings of a global open borders policy given the current geopolitical state of affairs). But wheter or not one buys the moral argument seems to me kind of secondary to usual discussions on immigration policy; instead I think the central issue is empirical: liberals and libertarians usually say there’s not only a moral presumption in favor of free immigration, but also that there are relevant consequentialist considerations as well, namely that the global economic welfare will substantially increase under an open borders policy. Of course, one might reply that even if the economic argument is sound, there are non-economic empirical considerations that should give us pause, and I think violence is probably one of the main worries. Since I know little about the relevant empirical literature on the effects of immigration on violence (I have seen conflicting data from the little I know, though), I try to remain as agnostic as possible. But anyway, I’m just writing this comment to ask you if you think that libertarians tend to overstate the economic consequences of open borders. Also, even if open borders are not a good idea, do you think current immigration policies should in general be revised in order to be less restrictive in some respects? Or do you think there’s no way to make a reasonable general policy recommendation on this issue?

          1. The claim that open borders would double the world GDP, which Brennan makes, is precisely the one I had in mind when I said he made empirical claims that are false. I also plan to write a post specifically against this argument, because we hear that claim all the time, yet the argument used to reach that conclusion is terrible. It’s also a great example of how economists sometimes use models that are totally worthless to bamboozle people. There are empirical questions about immigration, such as the effect it has on wages, which are difficult to settle, but this isn’t one of them. (To be clear, I’m not trying to imply that it’s easy to determine the effect that open borders would have on the world GDP, which it isn’t, only that it’s easy to show that the argument used to reach the conclusion that open borders would double the world GDP is terrible.) Anyway, as I said, I will publish a detailed post about this argument when I start my series on immigration. It will probably take a while before I write it, although I already have a draft, but I will do it eventually.

  3. You didn’t even check for the accuracy of these numbers. That’s mind-boggling to me, sorry.
    The crime rate for Germans and foreigners seems to be accurate, because there are indeed currently around 9 mio foreigners living in Germany.
    Also the number of crimes excerpted from the Police statistics is probably accurate, I haven’t looked for it myself.
    BUT Refugees and asylum seekers almost certainly don’t have such a crime rate.
    How did Ines Laufer (whoever that is) made these numbers up?
    Well it seems that she just estimated the number of asylum seekers living in Germany at around 1,1 Million People.
    Seriously? First of all there were ca. 500.000 – 700.000 asylum seekers living in German prior to 2015 (source available).
    Then an additional 900.000 asylum seekers came to Germany in 2015 (refugee crisis) and then 300.000 – 400.000 asylum seekers in 2016.
    1,1 Million asylum seekers at the end of 2016 seems like an extremely low estimate.

    How do you except people to take you seriously on this rather controversal Topic (although it shouldn’t be deemed controversial) if you can’t even look for the numbers yourself, and then draw conclusive statements from that?
    Maybe you’re saying that even with more accurate numbers it is still clear that immigrants on average commit more crime than native Germans.
    And you’re right. I’ve nothing to add to that. But I feel just way more confident about the discussion if we’re operating with somewhat accurate numbers.

    1. You should have read Laufer’s article more carefully. As she explains in footnote 1, asylum seekers who have been granted asylum, or have obtained protection under some other status, are removed from the category “asylum seekers” for the purposes of the statistics about crime. This is presumably why the number of people who fall under that category used by Laufer to calculate the crime rates is lower than you would expect.

      The important point is that, since these people are no longer considered asylum seekers for the purposes of these statistics, the crimes for which they are suspects are not counted toward calculating the crime rates of asylum seekers and it would therefore have been inappropriate to add them to the denominator. Doing what you suggest would have resulted in underestimating the crime rates for asylum seekers.

      Given that people who have been processed are not considered asylum seekers for the purpose of these statistics, the figure Laufer used seems in the right ballpark. (Presumably, most of the asylum seekers already in Germany in 2015 had already been processed, while only 1.2-1.3 millions of them arrived after that.) I will try to email her to see if she can explain how she calculated the number of asylum seekers in the sense of the BKA exactly, but I have no reason to think the figure she used for the calculation is wrong.

      I also see no reason to think that asylum seekers who have already been processed have a significantly lower crime rate than asylum seekers who haven’t been processed yet, so I doubt this quirk about the way in which the BKA classify asylum seekers makes a meaningful difference. Perhaps the actual rates are somewhat lower, perhaps they are somewhat higher, but it’s unlikely to be very different and in any case it’s much higher than natives.

    2. Hi Konstantin,

      thank you for your concerns about the numbers of Asylum seekers of 1,1 Mio. which I used to calculate the crime factors for this group.

      Well, Philippe already explained correctly why this number only includes those people who are still in the asylum process – waiting for the decision of the authorities. People who been granted asylum or who have been accepted as refugees, do not appear in this number – while those who have been refused but given temorarly acceptance do.

      At the end of 2015, 980.000 people had to be considered “asylum seekers” according to the Federal Statistical Office. Then, during the whole year of 2016, another 280.000 people arrived in Germany for seeking asylum. At the same time, more than 400.000 have been granted asylum or accepted as refugees – and they would not any longer appear as “asylum seekers” in the criminal statistics. In fact, I could have calculated with 860.000 aslymum seekers and would not have been wrong! But as the numbers continiously changed during the year, I decided to uses an average rate of 1,1 Mio. – and I think it is quite generously in favour of the asylum seekers.

      If I would have wanted to be more “accurate”, I would have calculated the monthly rates of arriving asylum seekers (addition) and those who have been accepted (deduction) and then made a proper annual average – which would be below 1,1 Mio.

      Anyway, even my slightly biased number in favour of the refugees is giving a clear picture of the immense discrepancy between the criminal behaviour of Germans and refugees

  4. As I see what happens in US – (In Europe seems to be different), all the Dems, the left, the ‘good compassionate souls’ that set up and maintain sanctuaries that do not allow deportation under almost ANY reason and refuse to collaborate with the feds – are fighting to BREAK immigration laws, and – weirdly, not trying to CHANGE them… what do you think about this?
    That also reminds how Obama opened an embassy in Havana and visited Cuba, which I think is actually illegal, there is a law voted by the Congress and signed in the ’90s by the president Clinton, that forbids connections to Cuba as long as the Castros are still ruling. He did not even bother to cancel that law, that as far as I know is still on the books. And…I didn’t see Republicans make much use of this specific fact… did you?
    I suppose if you research this, there might be other examples.

    1. Well, to be fair with the Democrats, nobody really cares about the rule of law as long as they think the law is violated for a good reason. On Cuba, Obama knew he couldn’t change the law because Congress was controlled by the Republicans, and they would never go along with that. On immigration, most of them wouldn’t want to change the law, even if they had a majority in Congress. That’s because 1) they know that it would be political suicide and 2) they know it can be violated without any bad consequences for them. But I have to say that their defense of sanctuary cities is one of the most retarded things I have ever seen. I also think Democrats who run for a national election are going to pay the price electorally unless they distance themselves from this kind of nonsense, but the Democratic base has been so brainwashed on immigration that it’s also difficult for them not to defend sanctuary cities. The worst part is that, as I point out in my post, liberals/libertarians really think they are being smart and better informed about this, yet their arguments are embarrassing. I may write a post about it eventually, because it drives me crazy to see otherwise intelligent people defend such a patently absurd position.

  5. Dear Mr.Lemoine,

    you are absolutely right,with the exception that
    the reality is much worse.

    Look at the data at this blog of a left leaning
    guy,but with interest in the truth and educated
    in working with statistics.

    Or,a lot easier,listen to the minister of justice of the german
    member-state HESSEN,where I live.
    She is lamenting in the public,that she has got a big problem
    in her prison-system.Only 30% of the convicted inmates are
    speaking more than 2 or 3 words german.
    Plus,you have to know,as a refugee in Germany you have what we call Narrenfreiheit.
    You do not go to jail for lying about your country of origin,
    shoplifting a dozen times,stabbing your christian co-refugee with a knife or your first little rape.

  6. Hi Philippe, first I would like to say “Thank you” for sharing and discussing my findings about the different crime rates of Germans, foreigners and asylum seekers (=”refugees”) for the year 2016.

    These figures and numbers are clearly and quite brutally giving proof of what we actually experience, hear and see in our daily lives: That our country has been flooded with one of the most criminal and violent group of people you can find at this planet (= young male Muslims, coming from very violent family backgrounds and “cultures” of very high inner familiar violence, especially against women and children, with no or very low education and low social status).
    To this conclusion, I had to come not by only evaluating the crime rates themselves but furthermore by going into the so-called “correction factors” which left politicians, authorities and experts use to mention in order to put those immense crime rates into perspective. In fact, the “correction factors” are not suited at all to relativise the criminal behaviour of the asylum seekers – but they explain WHY this group of people is so criminal – and why we do not have any influence on it (not even with the most generous “integration efforts”):
    You can find the information in my first publication about this issue, dealing with the crime rates of the year 2015, chapter C.): https://reisejunkie.info/loft/wp-content/uploads/2017/02/Zahlen_Fakten_Beweise_Fluechtlingskriminalitaet.pdf It is in German only as the translation into English or French would exceed my current capacities.

    1. Hi Ines!

      Thanks for your excellent article, this is very good work. I was actually going to email you to ask for details about how you did some calculations, but you found me before I had the chance and explained what I wanted to ask you. But I actually have a few more questions, though I’m not sure if you know the answer.

      First, you mention that journalists incorrectly infer from the fact that 80% of the people who commit a crime against refugees are themselves refugees that 80% of the people who are victim of a crime committed by a refugee are refugees, which as you point out doesn’t follow. But I was wondering if you knew what the actual figure was. More precisely, I would like to know 1) what proportion of the victims of crimes committed by foreigners are German and 2) what proportion of the victims of crimes committed by asylum seekers are German. I would be particularly interested in have that information for violent crime.

      I’m also wondering if you know what the crime rate is for different categories of foreigners. In particular, I would like to know the crime rate of foreigners who are citizens of the European Union, as well as the crime rate of foreigners who come from other regions of the world. Again, I’m particularly interested in having that information for violent crime, even more so than for crime in general.

Comments are closed.