Note: This article is deliberately structured so that each section marks a point where a different party would have stopped analyzing. The golden boxes show where. If you already know which party you're voting for, skip to the matching box and stop there. (Please don't actually do that.)

If you want the whole truth, you'll have to read everything, I'm afraid. Yes, I know. Reading is hard work. We all have to make sacrifices.

Dr. Dubium asks his minions

Dr. Dubium has spent the last few years in a cave on the moon, doing nothing but AI alignment research. Now he wants to go into politics. Step one: get briefed by the minions.

"Minions! What's this immigration issue everyone's so worked up about? Give me facts, not opinions."
— Dr. Dubium

What follows is the result. Not an opinion, but a method — applied to one of the most charged topics in German politics. Summarized and verified in minutes, with the help of AI.

What the statistics say at first glance

The facts that conservatives cite:

That sounds pretty clear-cut. 15% of the population, but 34-41% of crime suspects — and up to 18% for sexual offenses. Not much to debate, is there?

↳ This is where the AfD (far-right) would stop. The numbers speak for themselves. We keep going.

What happens when you look more closely

The facts that progressives cite:

  • Several peer-reviewed studies (including an ifo Institute study from 2025) find no correlation between the share of foreigners in a district and the local crime rate
  • When you control for age, gender, and place of residence, the gap shrinks dramatically or disappears entirely
  • Migrants are also disproportionately victims of crime — attacks on refugees rose 75% in 2023

Why the raw numbers are misleading:

  • Demographics: Immigrants are disproportionately young and male — the group that commits the most crime regardless of origin. A 25-year-old German man commits offenses at a similar rate to a 25-year-old Syrian man.
  • Geography: Migrants predominantly live in cities, where the crime rate is higher for everyone.
  • Reporting bias: Studies show that Germans are twice as likely to file a police report when the perpetrator looks foreign.
  • Immigration-specific offenses: The statistics include violations of residency laws — crimes that Germans cannot commit by definition.

So: no real problem? All just statistical artifacts?

↳ This is where the Grüne (the Greens) would stop. The studies are clear. We keep going.

We ask our minions: Dig deeper

At this point, we asked an AI to formulate the strongest counterarguments against the progressive position — not because we agree with them, but because an honest analysis takes both sides seriously.

The AI comes back with a crucial distinction: Not all immigrants arrive by the same route.

Annual immigration to Germany (OECD data 2024):

  • 47% EU free movement (no screening beyond EU citizenship)
  • 11% Labor migrants (vetted, job offer required)
  • 18% Family reunification (relationship verified, not qualifications)
  • 24% Humanitarian migrants (asylum seekers — minimal upfront screening)

This fundamentally changes the analysis:

Vetted labor migrants do well — often better than average. This makes sense: anyone who goes through an immigration process with qualification requirements brings qualifications. Research confirms this directly.

Among asylum seekers, the picture varies by country of origin: Immigrants from Syria, Afghanistan, and Iraq — the three largest source countries — commit significantly fewer offenses than the average across all immigrants. Roughly one in three immigrants in Germany is Syrian, but only one in five crime suspects among immigrants is.

Overrepresented, by contrast, are immigrants from countries with low asylum recognition rates — Nigeria, Algeria, Morocco, Georgia, Tunisia. These are people who know their asylum application will likely be rejected but stay in the country anyway.

This is also confirmed over time: A study in the Journal of Economic Behavior & Organization (2023) shows: Before the refugee crisis (2008-2014), a rising share of immigrants correlated with higher crime. During and after the 2015 wave, the effect was negative or insignificant — presumably because the wave was dominated by Syrians with legitimate asylum claims. Since 2022, the numbers have again been rising faster than the population is growing. This supports the adverse selection thesis: the more widely the system becomes known as exploitable, the more it gets exploited.

The insight: It's not "asylum seekers" as a category. It's specific subgroups — particularly those who use the asylum system as an immigration pathway even though they don't qualify.

↳ This is the point most debates never even reach. Too nuanced for a campaign poster. We keep going.

Wait: Suspects are not convicts

Another follow-up question for the minions: All the numbers so far are based on suspects. Being a suspect is not the same as being guilty. If the police suspect foreigners more often, that would be a measurement artifact, not a real problem.

What do the conviction statistics say?

This is revealing: If policing bias were the main explanation, the conviction rate among foreigners would have to be significantly lower than the suspect rate — they would be suspected more often but convicted less often. That's not the case. The suspect-to-conviction ratio is similar for foreigners and Germans.

At the same time, there are documented biases:

  • Against foreigners: Germans are twice as likely to file a police report when the perpetrator looks foreign. Foreigners receive somewhat harsher sentences than comparable Germans.
  • In favor of foreigners: Migrants are less likely to file reports when they themselves are victims. Crime within migrant communities is reported less often.

The biases partially cancel each other out, but not completely. The 2.6x overrepresentation (15% of the population, 39% of convictions) cannot be fully explained by the 2x reporting effect — not least because the reporting effect only applies to victim-reported crimes, not to crimes discovered by police.

↳ Most politicians would not even have asked the question about convictions.

Where our analysis stands

If you say "All immigrants are criminals" — the data does not support that. The majority of immigrants, including most refugees, integrate successfully over time. Crime correlates with demographics and living conditions, not with origin.

If you say "There is no immigration problem" — that's not true either. Germany has real integration challenges, certain subgroups are overrepresented in crime statistics, and public services are overburdened.

The nuanced answer that fits on no campaign poster:

A thoughtful policy position would be: "We want labor migration, should take in genuine war refugees, and need a way to actually exclude migrants who exploit the asylum system as a back door."

But that position fits on no campaign poster. Conservatives don't want to admit that Syrian refugees aren't the problem. Progressives don't want to admit that "refugees" is not a homogeneous category and that some subgroups are genuinely problematic.

Sometimes complex issues have complex solutions

This analysis cost you 10 minutes of your time. It contains data that both sides find uncomfortable. It arrives at no simple conclusion.

That's not a flaw. That's a good sign. Reality is complex. Anyone who compresses a topic this complex into a catchy slogan does not have your best interests at heart: they're trying to sell you something.

Most political debates end at one of the golden markers above. Not because the analysis would be too difficult, but because continuing would mean contradicting your own side.

"We don't have a position. We have a methodology."
— Dr. Dubium

It turns out that immigration is a complicated topic that doesn't lend itself to being summarized in a few lines. If you don't want to do this work yourself — vote for us, and we'll do it for you. And if this is exactly the kind of thing you've always wanted to see in politics, then join the League of Evil.