Migration, But Better: February 2026
1,000+ subscribers, the Alysa Liu argument for immigration, and the first readers' census
We hit 1,000 subscribers this month! Thank you for reading, sharing, and arguing with me in the comments. I started Popular by Design because I think the immigration debate deserves more honesty and less tribalism—and it turns out at least 1,000 of you agree (or at least enjoy disagreeing).
A few updates: I’m hiring a postdoctoral researcher to join me at Notre Dame’s Keough School to work on politically sustainable immigration. If you know someone working on immigration, public opinion, or policy design who might be a good fit, please send them my way. Immigration folks with a PhD (an unfortunate formal requirement by the admin) are more than welcome to apply. I’m also starting as an associate editor at the Journal of Ethnic and Migration Studies (JEMS)—one of the top journals in the field. Excited to help shape what kind of research gets published and how we evaluate it (at least before AI will change it all).
Since joining The Roots of Progress’s ABI fellowship, this month I published more pieces than ever before. Hopefully, I will be able to continue the streak at least until the end of my sabbatical. First, “Reflections on the Uncomfortable Truths about Immigration”—my attempt to address the most frequently asked questions about my earlier “Uncomfortable Truths” post. Second, “Student Migration Is Popular... Until It Isn’t”, on what went wrong in Canada, Australia, and elsewhere when universities started selling immigration status instead of education. Third, “Western Countries Do Not Need Immigration”—a deliberately provocative title for an argument that cuts in a surprising direction.
Before we get to the links, I have a favor to ask. I’m curious who’s actually reading this newsletter. We’re now roughly at the population of a small medieval village, which seems appropriate for some basic ideological census:
OK, here are the February links (linking does not imply endorsement):
I normally don’t follow the Olympics, but the restrictionists’ reaction to Alysa Liu has been fascinating to watch. Liberals thought MAGA celebrating her Olympic gold was a gotcha, but it’s actually a concession I’m happy to accept: no country needs immigration, but the countries that choose it wisely end up stronger. An American daughter of a Chinese dissident draped in the flag on the Olympic podium while the authoritarian government that persecuted her father watches—that’s a recruitment ad for every talented person in the world.
Noah Smith has a thoughtful piece on what a liberal immigration enforcement regime might look like. Most voters aren’t categorical restrictionists—they can support freer immigration if they believe the system is orderly and serves the national interest. The uncomfortable truth for many liberals: public opinion does support deporting most undocumented immigrants, not just criminals.
Austin Kocher also compiled a great selection of the latest immigration enforcement research—my reading list is now complete.
I often disagree with G. Elliott Morris on how to interpret public opinion on immigration, but his work this month on asking people concrete questions about enforcement is valuable. Abstract attitudes are one thing; specific policy preferences are another.
But here’s the uncomfortable truth for conservatives. More in Common US published new polling on immigration attitudes among Trump voters. Some of their findings surprised even me: 90% of Trump voters agree that “properly controlled immigration can be good for America,” and 70% want it to be easier to immigrate legally (but harder to come illegally). More room for agreement than usually assumed.
Yascha Mounk has a fascinating conversation with Ruud Koopmans on the differences between immigrant integration in Europe and the US. I’ve been beating this drum for a while: immigration is not one thing that has uniform effects—policies and contexts matter enormously.
Scott Alexander makes a related argument I’ve been pushing for years: you can’t just import conclusions from European immigration debates to the US, or vice versa. Showing a Danish crime chart says nothing about how well America integrates migrants.
The Liberal Patriot explains how Trump botched immigration. What we’re seeing is textbook thermostatic reaction to government overreach—voters still support deportation in the abstract but strongly disapprove of how it’s actually being carried out.
Case in point: the administration tried suspending TSA PreCheck and Global Entry, then caved within 24 hours after backlash. It’s honestly a bit funny, if not sad, that the trusted traveler programs are the thing that finally breaks it for normal folks who happened to vote Republican.
The Argument has several related pieces worth reading. First, Lakshya Jain shares new data on the trans rights backlash—I haven’t seen much from social scientists on what’s driving it. Is it thermostatic? Elite cue-taking? Something else? One of the most dramatic opinion shifts in recent memory, and the field has sat it out.
Second, Jerusalem Demsas on why thoughtless moderation is a mistake—voters aren’t dumb or unnecessarily cruel. Being thoughtlessly harsh on trans rights or immigration doesn’t win elections.
Ross Douthat has a great conversation at The Argument about the end of conservatism. His point on immigration: the current conundrums have nothing to do with big ideological debates about what we owe each other and foreigners. Trump’s “own, concrete choices, not just the spirit of nationalism, have led his administration to become very unpopular.”
Ryan Puzycki has a beautiful piece on Tokyo as a megacity at human scale. The “city of doorways, not vistas” framing is great. After living there, I came away with a similar sense that the intimacy is mostly about zoning and land use, not some mysterious cultural essence. Where I’d push further is toward what happens when demographics undermine that vibrancy—outside greater Tokyo, depopulation is devastating.
Rory Truex has a sobering piece on why comparativists should speak up. As someone who ended up studying US politics from a comparative perspective, I’m struck by how often the “it can’t happen here” instinct comes from people who simply haven’t looked at how things have unfolded elsewhere.
Richard Hanania came out against white culture. It’s kinda funny since this is pretty much exactly what the critical studies people have been saying all along, just without the unnecessary jargon. But well…sometimes it takes a different messenger.
Daniel Di Martino pushes back on the idea that immigration will fix fiscal problems. David J. Bier at Cato responds with a different set of assumptions that flip the results dramatically. My sense is that fiscal impacts are genuinely hard to estimate, and reasonable people can disagree on assumptions. But I’ve definitely learned from both Cato’s and Manhattan Institute’s work on this, and I wish more of our immigration debates were this technical rather than vibes-based.
Adam Ozimek and Jiaxin (Jason) He at EIG found major data errors in George Borjas’s paper making the case for a $100,000 H-1B fee. Ozimek’s broader point is one I keep returning to: you have to do the policy right. Many people who think they understand high-skilled immigration actually don’t understand the economics or the mechanism design.
Ilya Somin won his tariff case at the Supreme Court—a 6-3 decision challenging presidential tariff power. A Soviet-born constitutional scholar turning ideas into action by defending limits on executive authority. Worth reading his initial thoughts at Reason and his inspirational profile at IHS.
Taylor Trummel has a new paper using survey evidence with a conjoint experiment to test how state-level immigrant integration policy features affect perceptions of fairness and support. Important new evidence that US attitudes are much more inclusive than conventional debates suggest—but this support is conditional on policy design. People are more supportive of integration when it includes social support and clear eligibility criteria, which they view as fair. Another reminder that how you design and implement policies probably matters more than how you talk about it.
As before, if you want me to write more about one of these or other related topics, let me know!
The point where "Voters support deportations in the abstract but not the way it is right now" is where we need to take a seat back and look at how will actually functions.
I wrote an article where I dissected it into two components: needs and means. Voters have a better grasp of how a good country roughly should look like, but not so much of how to achieve that and what price we actually pay for things. In this case, deportations are a need of a second order because it's a mean for a exclusionary migration system. The question now is, is it actually possible to deport in the scale we need with appropriate means that don't violate, for example, limited damage to other human beings?
Migrants obviously don't want to be deported, so they try to hide themselves, but the more they hide, the more invasive has the government to be in order to actually achieve anything. The result is that more system-compliant migrants, which are more likely those that we want, are those who get deported first, while the deviant ones remain under the radar and refuse to interact with government systems in order to prevent leak of information about themselves. It also likely requires more violence to enforce those stricter rules.
My thesis, which can be wrong of course, is that any system which wants to minimize deviancy (aka make people interact with the system without the need to search them) needs to set an positive incentive first and does not punish for the sharing of true information in any way. This would, for example, mean, that the government needs to give people protection from deportation if they are going to the government and want to document themselves. But this goes against what the restrictive migration laws want to achieve.
Actually I notice that I write something that might belong in its own article lol
I'd be happy to take your poll, but the irony of a small number of vague categories is biting. I'm libertarian - how would you code that? I want lots of immigration but only with high assimilation - where does that fit?