Daniel's grandfather ran a synagogue in Cork. Born in Dublin, raised in Cork, moved to Israel at twenty-five, and he says the move was never about opportunity.
Right, and he's making a sharper claim than that. He's saying he'd rather have avoided it.
He wrote in this week, and the short version is this. His migration to Israel was rooted in ideology. Not in seeking a better life, not because Israel was a good option that happened to be available. He thought Israel is where Jewish people should live, and he acted on that. And then he makes the argument that as a rational actor, migration is something he would rather have avoided. Immigrating to a country where you don't necessarily have a more prosperous future is, in his words, stacking the odds against you.
Which is a strange thing to hear from a man who did exactly that.
It is. And he knows it. He reckons ideological immigrants are the exception among immigrants, and he wants the conversation taken well beyond the Jewish case. So, four things. Why do people move countries generally? Has anyone actually quantified the most common reasons, and what percentage each cohort makes up? Besides Jewish immigration to Israel, what are the other examples of groups that move for ideological reasons? And then the big one, why might a rational actor prefer not to move at all?
That's a lot of question for one email.
He's thorough. So let's start with the standard model, the one that treats migration as an investment.
And that's the right place to start, because the rational-actor framing Daniel is using is the mainstream one in migration economics. Migration is a costly, risky investment. You only do it when the expected gains outweigh the costs, and the costs are substantial. Financial, psychological, social, legal. So the people who actually move are self-selected. They're the ones for whom the net expected return comes out positive.
Which means the population of migrants is not a random sample of the population of origin countries.
Not remotely. That's the whole insight. If migration were costless, you'd see everyone from a poor country move to a rich one. You don't, because the cost is enormous and the return is uncertain. So the people who go are the ones with the highest expected payoff, or the lowest attachment to where they were, or both.
And the dominant academic model for sorting those drivers is push-pull.
Push-pull, yes. Push factors from the origin country. War, persecution, political instability, inadequate pay, limited professional growth, poor working conditions, lack of services. Pull factors from the destination. Stability, economic opportunity, better remuneration, education, quality of life, family ties.
The European Psychiatric Association put out a policy paper last year that lays the taxonomy out quite cleanly. Push factors as war, persecution, political instability. Pull factors as stability, economic opportunities, education, favorable living conditions. And then it splits migration into three broad categories. Voluntary, forced, and workforce migration.
That three-way split is useful, and it's also where the trouble starts, because those categories overlap badly in practice. A Syrian doctor who leaves because the hospital was bombed and then takes a job in Germany, is that forced or workforce? The paper would say forced. The German labor ministry would say workforce. Both are true.
And that's a preview of the data problem, which we'll get to. But first, the drivers themselves. What does the evidence actually show about why people move?
The clearest recent picture comes from health workers, because they're the most studied migrant population on earth. There's a scoping review from this year on Africa-born health workers, and the push factors were exactly what you'd predict. Inadequate pay, limited professional growth, political instability. The pull factors were better remuneration and career opportunities abroad.
Nothing surprising there.
No, but the interesting part is the return migration. What drove people back? High costs of living, discrimination, and deskilling.
Deskilling meaning their qualifications didn't transfer.
A surgeon who arrives and can't practice as a surgeon. The calculus reverses. The thing that pulled them out becomes the thing that pushes them home, and often the pull of home is weaker than the push of the new country.
So the model isn't static. It's a running calculation.
It's a running calculation, and that shows up even more clearly in the UK doctors literature. There's a scoping review from a few years back, forty studies, and it found migration is determined at three levels. Macro, so global and national. Meso, so profession-led. And micro, so personal.
And the striking finding?
The same factors drive both immigration to the UK and emigration from the UK. Poor working conditions, employment opportunities, training, quality of life, and this one phrase that keeps appearing, desire for a life change.
So the same country is simultaneously pulling people in and pushing people out, for the same reasons.
A doctor in Lagos looks at the NHS and sees opportunity. A doctor in Manchester looks at the NHS and sees burnout, and moves to Australia. Same system, opposite verdicts.
Which tells you push-pull isn't a property of countries. It's a property of the match between a person and a country at a moment.
That's a better way to put it than most of the papers do.
What about the studies that aren't about professionals? Because doctors are a peculiar sample. Highly credentialed, internationally portable, in demand everywhere.
There's a rural-urban study from Uttarakhand, survey of a hundred migrant families. Major reasons were education, employment opportunities and income, and facilities. Basic services.
And the environmental factor?
Ranked very low. Which is fascinating, because the region has environmental problems. Landslides, water stress, out-migration from hill villages. But when you ask people why they left, they don't say the mountain is falling down. They say the school is bad and there's no work.
So the real driver is present but not articulated.
Or the environmental pressure is mediated through the economic one. The land fails, so the income fails, so you leave for income. The stated reason is the proximate one.
And the outcome? Did moving work for them?
Mostly yes. Most migrants reported significantly improved socio-economic situations after moving. That's an important counterweight to Daniel's thesis, and we'll come back to it.
What about the cases where the driver is violence rather than economics?
There's a multilevel study from Deep South Thailand showing violence accelerates migration on top of long-standing economic motives. Which is the honest picture in most conflict zones. It's not that people were content and then war broke out. It's that they were already inclined to leave for work, and the violence compressed the timeline.
The war doesn't create the migration. It accelerates it.
That's a good way to say it. And then there's the softer end of the spectrum. There's a study of Italian medical students from last year, and more than half had considered emigrating. Main reason wasn't money. It was higher-quality training. The barrier they cited was bureaucratic procedure.
So opportunity-seeking migration is often about professional development, not just salary.
And that blurs the line Daniel is trying to draw. If a medical student leaves Italy for a better residency in Germany, is that economic migration? Yes. Is it about identity, about becoming the kind of doctor they want to be? Also yes. The categories don't separate cleanly.
Which brings us to the percentages. Daniel asked whether research has quantified the most common reasons and the percentage each cohort comprised. And the honest answer is no.
Not in the way he's asking. There is no single authoritative global dataset that cleanly assigns percentages to economic versus family versus ideological versus forced. And the reasons are structural, not just a gap in the research.
Break that down.
Three reasons. First, definitions vary. The UN and the International Organization for Migration distinguish voluntary from forced, but economic and family categories overlap heavily. A person who moves to join a spouse and then works, are they family or economic? Depends who's counting.
Second?
Self-reported reasons are unreliable. People give the acceptable reason. If you're asked at a border why you're entering, you say family reunification or study, because those are the categories that get you in. The real driver might be that you couldn't stand your hometown anymore.
So the data is shaped by the admission system it's collected through.
And that's the third point. Most large datasets, UN DESA, OECD, Eurostat, categorize by legal admission channel, not by motive. They're counting visas, not reasons.
So what can we actually say?
We can say things about channels. Family reunification is typically the largest legal admission category in OECD countries, somewhere around forty to fifty percent of permanent flows in many years. Labour and economic migration is next. Humanitarian and asylum is a much smaller share of total legal flows, typically under fifteen percent in most OECD countries in a given year, even though it dominates public attention.
And free movement?
Enormous, but often counted separately. Intra-EU movement doesn't show up in the same tables because there's no visa to count.
So the honest answer to Daniel's second question is that the categories are policy constructs and the motive data is soft.
That's the honest answer. And it's not a dodge, it's a finding. The reason nobody has produced the percentage breakdown is that the thing being measured isn't a stable quantity. Motive and channel are different variables, and the datasets mostly track the second one.
Which means when someone tells you what percentage of migrants are economic versus ideological, they're either using a proxy or making it up.
Usually the second one.
So that's the standard model. Push and pull, costs and benefits, and a data infrastructure that tells us about visas rather than reasons. But Daniel's own case doesn't fit the framework at all.
No. He moved to a country where he expected a harder material life, on purpose, because he thought it was where Jewish people should be. That's not utility maximization. That's something else.
So let's define it. Ideological migration. People who move for identity congruence, political or religious fulfillment, ideological conviction. They're not maximizing expected utility in the standard sense. They're maximizing something else.
And the cleanest empirical illustration of that is a study from this year on older French immigrants to Israel. Aliyah after sixty-five.
People moving in their retirement years.
Which is the opposite of what the economic model predicts. If you're sixty-eight and moving countries, the expected lifetime return on that investment is tiny. You're not going for a career. You're not going for income. You're going because of what the move means.
What did the study find?
That the migration was the fulfillment of a lifelong identity project, centered on family, Jewish identity, and belonging. And here's the crucial part. The participants experienced persistent linguistic, social, and institutional challenges. Persistent. Not transient. They struggled with Hebrew, with social integration, with bureaucracy, and it did not diminish their strong sense of attachment to Israel.
So the costs were real and ongoing, and the payoff was somewhere the cost-benefit model doesn't look.
The payoff was belonging. And belonging doesn't depreciate the way income does. That's the key. A salary increase fades into the baseline within a year. A sense of being in the right place doesn't.
So Daniel's thesis holds up empirically. Ideological migrants accept worse material outcomes because the payoff is identity.
For this cohort, yes. And it's worth being precise about what that means. It doesn't mean they're irrational. It means they're running a different objective function. If your utility includes identity congruence, then moving to Israel at sixty-eight with a worse pension and no Hebrew is a rational move. It's just not rational under the model that only counts money.
Daniel said he'd rather have avoided migration as a rational actor. But if identity is in the utility function, then he wasn't violating rationality. He was weighting it differently.
That's the wrinkle. The rational-actor model isn't wrong, it's underspecified. It assumes we know what people are maximizing.
So what are the other examples? Daniel asked specifically for cases beyond Jewish immigration to Israel. And there are a lot of them.
There are, and they cluster into a few types. The first is political exiles and dissidents. People who leave because they reject the regime. Anti-apartheid South Africans. The Iranian diaspora after 1979. Cuban exiles. Russian dissidents after 2022. Hong Kongers after 2020, via the BN(O) route to the UK.
Those are people moving toward something, but the primary driver is moving away from a political order they can't live under.
Right, and the distinction matters. A Cuban exile in Miami might have better economic prospects than in Havana, but that's not why they left. They left because they couldn't stay. The economics is downstream of the politics.
Second type?
Religious migration. Mormons to Utah historically. Puritans to New England. Muslim migration to be in a Muslim-majority society. The driver is the desire to live among co-religionists, in a place where the institutions match your beliefs.
Which is interesting, because that's not exile and it's not persecution. It's a preference for congruence.
It's the same thing Daniel is describing. It's not that Cork was hostile. It's that Israel was where he thought Jews should be. The pull is toward the place that matches the identity, not away from a place that doesn't.
Third type?
Repatriation and homeland return movements. Liberian and Sierra Leonean returnees. Armenian repatriation. The Greek and Turkish population exchanges of 1923. German Aussiedler returning from the former USSR.
The 1923 exchanges are interesting because they were state-organized. That's ideology imposed from above rather than chosen from below.
Which is a different phenomenon, and worth flagging. Some ideological migration is voluntary, and some is engineered. The population exchanges were the second kind. And then there's ideological settler migration. Nineteenth-century European colonists motivated by a civilizing-mission ideology. Some Israeli settlement movements. Back-to-the-land ideological communities.
And there's a reverse case worth naming. People who stay for ideological reasons.
Yes. People who refuse to leave a homeland under occupation or oppression. That's ideological non-migration, and it's the same variable. If you'd move for identity, you'd also stay for identity. The model has to cut both ways.
So the distinction is clean. Ideological migrants are not maximizing expected utility in the standard sense. They're maximizing identity congruence or political or religious fulfillment. Which is why they're exceptions to the rational-actor model.
They're exceptions to a specific version of it. And Daniel's argument, that he'd rather have avoided migration as a rational actor, is really an argument that the standard version is the correct one. Which is defensible. It's just not the whole picture.
So let's take the rational-actor case against migration seriously. What's the evidence that migration is something to avoid?
The supporting evidence is substantial. Migration is disruptive. You lose your social networks, which are a real asset. You hit language barriers. Your credentials get devalued, which is the deskilling problem from the health worker review. You face discrimination. You deal with bureaucracy, which the Italian medical students named as the main barrier.
And there's the health literature.
There's the healthy migrant paradox. Migrants often start healthier than natives, because the people healthy enough to move are a selected group. But their health advantage erodes over time in the destination. So the initial selection advantage gets eaten by the stress of the move.
That's a real cost, and it's invisible in the salary comparison.
It's invisible in almost every comparison. Nobody puts the erosion of your health advantage in the spreadsheet when they're deciding whether to take the job in Frankfurt.
So Daniel's case holds. But the evidence is mixed, and you flagged that earlier.
The evidence is mixed. The Uttarakhand study found most migrants' socio-economic situations significantly improved. That's not a small thing. For most economic migrants, migration works. That's why they keep doing it.
And there's the caregiver study.
From Taiwan, foreign caregivers were slightly happier and had higher life satisfaction than native caregivers. Small study, suggestive rather than conclusive, but it points the same direction. Migration can improve subjective well-being even when the objective conditions are hard.
And the UK doctors review found desire for a life change as a driver.
Which is the blur. If a doctor moves to Australia because they want a life change, is that economic or ideological? It's neither, or it's both. The line Daniel is drawing is real, but it's not as clean as the categories suggest.
So where does that leave the synthesis?
Daniel's rational-actor case against migration holds for many people. The costs are real, they're large, and they're often underestimated. But for most economic migrants, migration works. The socio-economic situation improves. The disruption is worth it.
And for ideological migrants?
For ideological migrants, it works differently. The payoff is identity, not utility. The French Aliyah study shows they accept persistent challenges because belonging is the return. And that's not a lesser return. It's just a different one.
So the rational actor who migrates against economic logic isn't breaking the model. He's running a different one.
He's running a different one, and the model he's running is the one the standard framework can't see. Which is why Daniel's own case is so useful. He's the data point that shows the limits of the framework he's using to describe himself.
Which is a strange kind of self-awareness.
It's the good kind.
Hilbert: Can I ask you something.
Go ahead.
Hilbert: What happens to the ones who move for the money. Not the ideology cases. The ordinary ones. The engineer with the offer in another country and a wife and two kids and a mortgage.
That's the bulk of it, isn't it.
Hilbert: I spent a few years doing relocation work for an engineering firm. Mid-sized. Maybe four hundred employees. My job was to help people move to new countries for work. Visas, housing, schools, shipping. The whole thing.
And what did you see?
Hilbert: The ones who were most excited about moving were often the ones who had the least concrete reason to go. And the ones with the best offers, the biggest salary bumps, the best career moves, were often the most miserable after they arrived.
That's the opposite of what the model predicts.
Hilbert: It is. I had a man, a design engineer, took a forty percent raise to move to a plant in another country. Good job. Nice house provided. He was back inside two years. Couldn't tell you exactly why. His wife missed her sister. His kid wasn't settling. He said the work was fine. He just wanted to go home.
The ones who stayed?
Hilbert: The ones who stayed were the ones who moved for something they couldn't put on a spreadsheet. A partner. A faith. A cause. A feeling that they belonged somewhere else. I had a woman who moved for a church. She found work when she got there. She's still there.
The pattern is the same as the research, just at a different scale.
Hilbert: It's the same pattern. The people who moved for reasons they couldn't quantify were the ones who stayed. The ones who moved for the number came back.
Did you track it?
Hilbert: I kept a list. Handwritten. Every client who moved, and whether they stayed or came back. Still have it somewhere.
What does it say?
Hilbert: It's more complicated than a percentage.
Of course it is.
Hilbert: It is, though. Because the ones who came back weren't failures. Some of them did very well and chose to come home. And some of the ones who stayed were miserable and just didn't have the money to leave. The list doesn't sort into stayed equals happy and returned equals sad. It sorts into people who had a reason that outlasted the move and people who didn't.
The reason that outlasted the move. That's the variable.
Hilbert: It doesn't show up on the offer letter. Nobody puts it in the contract. The firm I worked for never asked about it. They asked about salary expectations and notice periods. They never asked what the person was moving toward.
Which is the same blind spot as the datasets. They count visas, not reasons.
Hilbert: They count visas. And then they're surprised when the visa holder goes home.
The practical version of the research finding is that the thing that predicts staying isn't the size of the offer. It's whether the move is attached to something durable.
Hilbert: It's whether there's something on the other side that the person is actually moving toward. The ones who were running from something, a bad job, a bad town, a bad marriage, they didn't stay either. Running from isn't enough. You have to be running to.
The ideological migrants aren't the exception after all. They're the clearest case of something true for everyone.
Hilbert: That's how I'd put it. The ideology is just the reason made visible. Most people have a reason. They just don't call it ideology.
The ones who don't have one are the ones who come back.
Hilbert: In my experience. Two years, usually. Sometimes eighteen months.
The list is more complicated than a percentage, you said.
Hilbert: It is. And I'm not going to read it to you.
Fair enough.
Hilbert: But I'll say this. If you ever want to know whether a move is going to work, don't ask about the salary. Ask what the person is moving toward. If they can't answer in one sentence, the odds aren't good.
That's a better screening question than anything in the literature.
Hilbert: It's cheaper, anyway.
If the data on motive is soft, and the categories are policy constructs, how should we think about the categories themselves?
With humility, mostly. They're useful for counting visas. They're not useful for predicting who stays.
The distinction between ideological and economic migrants. Is it real, or is it just the visible edge of something universal?
I think it's the visible edge. The French Aliyah study shows the mechanism clearly because the material sacrifice is so obvious. But the same mechanism is running in the engineer who moves for a church, and probably in a lot of people who say they moved for work.
Daniel's framing, that ideological migrants are the exception, might be backwards. They might be the control group that shows what's actually driving everyone.
That's a strong way to put it, and I think it's close to right. The economic model describes the vehicle. It doesn't describe the destination.
As migration patterns shift. Climate change, political instability, remote work. Does the distinction hold up?
Remote work is the interesting one. If you can keep your job and change your country, the economic cost of moving collapses. Which means the decision gets made on the other variables. The ones that used to be buried under the salary calculation.
We'd expect more migration that looks ideological, because the economic barrier stops filtering it out.
We'd expect more migration that looks like a preference, because the thing that used to force the decision is gone. And that's going to make the datasets even less useful, because the channel categories assume the move is for work.
The categories get worse just as the phenomenon gets more interesting.
That's the pattern in most of these fields.
My Weird Prompts is available on all podcast platforms. If you enjoyed this episode, leave a review and tell a friend.
Thanks to our producer, Hilbert Flumingtop.
This has been My Weird Prompts.
We'll be back soon.
