Below, I discuss what we’ve learned in the past year about topics previously covered in this living lit review.

US Immigration Updates

Updating Undocumented People In The United States

2025 and 2026 have seen increased immigration enforcement, with many more ICE arrests and interior deportations increasing 4.6x. This has resulted in a significant slowdown of migration to the United States, including of legal migration. Brookings estimates that for the first time in at least half a century, more people left the United States in 2025 than entered it.

Perhaps unsurprisingly, this does not look great for the American economy. The Dallas Fed estimates that declining immigration will knock a point off GDP growth, and the Penn Wharton Budget Model has produced a model of the effects of deporting 10% of undocumented people per year.

Over four years, Penn estimates this will increase the deficit by $350 billion and decrease overall GDP. The effects on GDP/capita are more ambiguous; there is a short-term increase (particularly for the remaining undocumented workers who manage not to be deported), followed by a longer-term decline.

They find that most of the high-skill population will net lose out under mass deportations. This includes both authorized immigrants and the native-born. Older high-skill workers are the exception, as their working lifetimes will not include the longer-term decline. The picture for low-skill authorized immigrants is more mixed, with most having net gains (particularly older workers or those with high wages), but younger workers ending up worse off.

Updating H-1B Visas and the American Economy

In September 2025, the US introduced a new $100,000 fee on H-1B visas. I wrote at the time that I did not think this was a good idea.

This fee was the brainchild of George Borjas, who served in the Trump administration. In a 2026 paper, he argued that H-1B visa holders earn 16% less than similarly credentialed Americans, suggesting firms might be willing to pay an additional fee to continue hiring these low-wage workers. He calculated the revenue-maximizing fee would be between $118,000 and $264,000, and would have no impact on the number of H-1B visas issued.

The Economic Innovation Group argues that Borjas’ paper has data errors, and the actual wage gap is much smaller, at around 5%. In this case, a $100,000 fee would be far more than an employer would be willing to pay for a H-1B visa.

Michael Clemens also argues that Borjas has made data errors, but he finds that H-1B visa recipients actually command a small wage premium relative to the native-born. In this case, it is extremely unlikely that an employer would wish to pay an additional fee.

I have not gone through the data in these papers in enough detail to litigate who is correct, though Ozimek and Clemens’ point that Borjas uses 2020-2023 salary data for immigrants and only 2023 for the native-born seems hard to argue with. Revealed preference also seems to suggest that the revenue-maximizing fee is considerably less than $100,000, because overall fees collected on H-1B visa applications have decreased substantially since the fee was introduced.

Brain Drain and Brain Gain Updates

Updating What happens when you send Ugandan students to Germany?

Malengo is a program that provides a loan for Ugandan students to attend university in Germany. The expectation is that many students will then get jobs in Germany, and earn perhaps 20-50x what they would in Uganda. Students that stay will also pay back some or all of that loan.

Last year, they updated their model. They now estimate that the program produces 28x the returns of a cash transfer program. The positive update largely comes from moving consumption gains forward in time - even while attending university, consumption is some 7x what it would be in Uganda.

That being said, I think the cost-effectiveness of Malengo remains fairly uncertain until their first couple of cohorts of students graduate. Since the goal is to increase income largely through students choosing to stay in Germany after graduation, we still need to know what their labor market placements will look like.

Updating Migration of Doctors and Nurses

I have previously posited that there are significantly diminishing returns to additional medical personnel in places that already have a reasonable supply of medical personnel. I still think this is true, but I was surprised to see that the effect of adding or subtracting physicians is still detectable even in high income countries.

For instance, Dodini, Lundborg, Løken and Willén 2025 estimates that losing a doctor in Sweden results in 0.21 deaths per year.1 Grabowski, Gruber and McGarry 2026 estimates that each additional immigrant health worker in the US prevents 0.07 deaths per year,2 though this may be higher than in other contexts because immigrant health workers appear to be of higher quality than US-trained physicians.3

There does not appear to be a saturation level at which point additional physicians are not helpful. This would seem to suggest that expanding the supply of health workers is important in almost all contexts.

This would seem to make the general equilibrium result found in Abarcar and Theoharides 2024 extremely important. Abarcar and Theoharides shows that you can end up with a net increase in the supply of health workers even in a country with significant emigration if the following two conditions are met:

1. High wages outside the country draw people into health professions.

2. Supply of health professionals is allowed to expand to meet demand.

Some of the newly-qualified health professionals will migrate, but not all will.

If supply is not allowed to expand, you will probably have a net brain drain effect.4 So: expanding supply is very important; without it, migration can be net negative.

UK Immigration

Updating Are recent immigrants a “ticking time bomb” for British public finances? / UK Immigration and Public Services

We have a couple new reports from the Migration Advisory Committee, focusing primarily on the 2022-2023 cohort of immigrant entries. (This is the peak of the “Boriswave”.)

This has an interesting analysis on the fiscal impacts of each visa type, with a deeper dive into care worker fiscal impacts here.

The average skilled worker main applicant will pay £689,000 more in taxes than they receive in benefits (in net present value terms). The average health and care worker main applicant will pay £54,000 more in taxes than they receive in benefits.5 Care workers are net negative (-£36,000) while higher-skilled professionals are net positive.

Dependents are somewhat less fiscally positive; the adult dependents of skilled workers (e.g. spouses) are just about break-even, while the adult dependents of health and care workers cost £67,000 more than they contribute in taxes.

People who enter on partner visas are the most fiscally negative, at NPV -£109,000. This is consistent with the low labor force participation rate for this group (<50%). It’s not overly surprising that a group that is only 50% in work is fiscally negative.6

However, when considering the types of visas that the MAC analyzes, the cohort as a whole (skilled workers, their dependents, health and care workers, partners) is highly fiscally positive. They estimate that the 2022-2023 cohort of these visas will contribute £44.4bn more in taxes than they receive in benefits.

This is not the overall fiscal effect of the 2022-2023 immigrants, though, as we don’t know the overall fiscal cost of those entering on humanitarian visas.7 2022 was the peak of the influx from Ukraine; about 40% of immigrants that year entered on a humanitarian visa.

Given how positive those who entered on work visas are, though, these entrants can be significantly fiscally negative and the whole cohort will be positive. All other visa entrants can have a fiscal impact of -£122,781 (that is, more negative than any other visa type) and the whole cohort would still be fiscally neutral.8

This led to a rather bizarre national1 conversation about if Ramaswamy himself is truly an American. Is he - a second-generation immigrant - even qualified to opine on such things?2 The conversation about what it means to be American has continued as ICE seeks to detain anyone who they believe looks like an undocumented immigrant.3

So, it seems worth asking: when do immigrants - and their children - become American?

Many asking these questions are not looking for the dates that one gains permanent residency or takes the oath of citizenship. They are asking something fuzzier: when do immigrants socially integrate into American culture such that they are “one of us”?

These are not questions that economists usually seek to answer. In general, economists tend to think about assimilation in terms of economic integration, rather than cultural integration. But no one thinks Ramaswamy has done poorly economically; rather, they question his cultural identity.

My own field, political science, tends to spend a bit more time on fuzzy concepts like culture and belonging. This post will thus pull as much from the political science literature as it does from the economics literature. And economics is not a field that believes any question is beyond its all-consuming reach; economists Leah Boustan, Ran Abramitzky, and Katherine Eriksson, in particular, have done some great work on these questions.

But even with the best data we have, we can’t truly measure when other people think someone has become an American. However, we can measure some of the traits Americans think make an American. These includes speaking English, civic engagement, and participating in American traditions and culture.

Language

Let’s begin with language. The US is an English-speaking nation, but it is not an exclusively English-speaking nation. About 22% of people in the US primarily use another language at home (though this may be in addition to at least some English). 13% speak Spanish at home, with around 9% speaking a non-English non-Spanish language at home.

A bit less than half of US immigrants speak English “very well” upon arrival. (They are, after all, generally coming from non-English-speaking countries.) One of the first signs of integration into America is generally improving one’s English skills.

There are significant economic benefits to learning English. If you only speak Spanish, you are restricted to employers who also speak Spanish at work. This is quite a significant decrease in possible employment opportunities. It also rules out many educational opportunities; while there are a handful of universities in the US that teach their B.A. degrees in Spanish language/literature entirely in Spanish, you’re probably out of luck should you want a degree in anything else.4 Being randomly assigned to an English language class increases income by $2,400;5 arriving in the US when you are young enough to become a near-native speaker increases your income significantly more.6

English language skills are also useful in social integration.7 When people are more comfortable in English, they tend to become more civically engaged - being randomly assigned to an ESL class doubled the likelihood of registering to vote. Fluent English speakers are less likely to live in ethnic enclaves and are more likely to marry people of other ethnicities.

Of course, learning English takes time. In general, the longer an immigrant has been in the United States, the more likely they are to speak English very well. These things are most useful if you plan to stay in the US long-term; if you plan to move back to your home country in a few years, you’re less likely to invest in the multi-year project that is becoming fluent in English. While we cannot always determine which immigrants want to stay in the US, there is one group of immigrants who do not want to (or cannot) return to their home countries. Refugees are generally expecting to spend the rest of their lives in the US, as their home country is no longer safe for them. They therefore learn English more comprehensively than other types of immigrants.

There does seem to be a ceiling, though; even after 30 years in the US, only about 70% of immigrants can speak English very well. The likelihood of any particular immigrant becoming fluent in English appears to be heavily dependent on age; almost all immigrants who arrived as children become fluent in English, while it is likely less than half of non-English-speaking immigrants who arrive as adults do.8

Even if they don’t become fluent, though, there is some benefit to trying to learn English.9 In an experiment, people felt more positively towards immigrants who speak heavily accented English - that is, signaling that they were in the process of making a costly investment to assimilate. Americans know English is hard to learn, and they respect the effort it takes.

Interestingly, coercing people to learn English does not seem to be very effective in turning them into Americans. Fouka 2020 examines the case of German-Americans. Before 1917, some US schools taught in German as well as English. Some states banned this practice when the US entered WWI against Germany. Forcing German-American children to learn in their second language only strengthened their identity as Germans. They were more likely to marry other people of German descent, less likely to give their children “American” names, and less likely to volunteer to fight against Germany in WWII.

Another paper that studies English-only schooling around the same time period (1910-1930) finds similar results. While English-only schooling did improve English fluency, it didn’t improve labor market outcomes or social integration.

Ethnic Enclaves

Ethnic enclaves are neighborhoods where a particular ethnic group clusters. They are common in cities where there is a large number of immigrants. The most well-known variant is probably the various Chinatowns around the United States, but there are also ethnic enclaves for Korean-Americans, Mexican-Americans, Somali-Americans, Vietnamese-Americans, as well as historic enclaves for Jews, Italian-Americans, Irish-Americans…

For decades, if not centuries, ethnic enclaves have been a common first destination after arriving in the United States. This makes some sense. If you speak limited English, you are probably going to move to an area where 1) you may know people, and 2) where other people speak the same language that you do. Many new immigrants prefer to live in communities that share at least some similarities to where they’re coming from.

Many native-born Americans also dislike ethnic enclaves. There is a perception that the residents of ethnic enclaves are not integrating into American society; indeed, that living in an ethnic enclave is a sign that you are resisting integration into American society.

This is, to some extent, true. It seems that moving to an ethnic enclave does reduce integration relative to moving to a place where there are fewer people of your ethnicity. In the twentieth century, residents of ethnic enclaves were less likely to change their names to fit in with Americans. When Italian-Americans attended Italian churches (often located in enclaves), they took longer to integrate into wider American society. They were less likely to intermarry and less likely to naturalize.

Living in an ethnic enclave might also harm economic outcomes. During the Age of Mass Migration, Norwegians who lived in ethnic enclaves earned less and were less likely to work in white-collar jobs. When Jews were funded to leave their ethnic enclaves, they earned more.10 And yet: people tended to live in enclaves for decades, perhaps significantly delaying assimilation.

Given all these negatives, are (some) Americans right when they criticize ethnic enclaves? My considered opinion is “ehhh, maybe, but also maybe not”.

There is also some evidence that enclaves can be beneficial because they provide a soft landing point for people new to the United States. Living in one can make it much easier to find a job with limited English skills, and appears to increase employment rates for newly-arrived immigrants. In some cases, it seems that people make more in enclaves than they do outside them, so it is not universal that ethnic enclaves harm economic integration.

Why are there contradictory results on the economic effects of enclaves? Some of it appears to be selection - that is, who else lives in an ethnic enclave. Is your enclave full of low-earning people or high-earning people? Do people tend to be highly educated? Did people choose to live there to avoid integration or was it just convenient? The character of the enclave seems to matter a lot.

It also seems plausible to me that enclaves are helpful in providing a soft landing when immigrants first arrive, but are net negative for immigrants who have been in the US longer and could be on the broader job market. Thus, there might be an optimal amount of time to live in an enclave.

More importantly than any of that, though, enclaves just don’t seem like a problem. Yes, they are common, but enclaves are neither more nor less prevalent now than they were in the 1920s.11

And they also seem to be temporary. The groups that populate today’s enclaves are not the same as those who populated enclaves in the 1920s. Even if Norwegians did stay in enclaves longer than would have been optimal, they eventually integrated into wider American society. In 2025, one would be very hard-pressed to find someone who complains about all those Norwegians and their distinct cultural practices.

Measures of Assimilation

The modern US is not the same as the US of the Age of Mass Migration, though. Should we expect similar integration outcomes for modern immigrants?

One way to measure cultural integration is through the names immigrants choose for their children. Naming your child “Inaya” or “Aarya” is quite popular in India, but these names are less common in the United States. If parents instead choose to name their child “Emma” or “Amelia”, this is suggestive that parents care more about their child being perceived as “normal” in the United States than they do about their child being perceived as “normal” in India. Indeed, it goes a bit further; choosing an “American” name is a sign that one wants to be perceived as “normal” by white Americans, rather than by co-ethnics.

Abramitzky, Boustan, Eriksson and Hao argue that naming is a particularly good way to measure assimilation for several reasons. Naming a child is a relatively unconstrained choice available to all types of immigrants. It does not cost more to name a child “Emma” as opposed to “Inaya”, meaning that such a choice is equally available to rich and poor immigrants. Unlike measures like intermarriage, it does not require the cooperation of anyone outside the immigrant family unit - after all, to marry someone from another ethnic group, that person also has to want to marry you. The only constraint on choosing a Generic American Name is one’s desire to have a child with a Generic American Name.12

Perhaps unsurprisingly, people are more likely to choose “American” names for their children the longer they have been in the US.13 More surprisingly, the rate at which people converge to “American” names is relatively unchanged over the last 100 years. People transitioned from the Greek “Giorgios” to the English “George” in the 1910s at roughly the same rate as people transition from “Juan” to “John” today. By this metric, then, immigrants integrate just as well today as they did in the Age of Mass Migration. Furthermore, the groups of immigrants most often accused of not being interested in assimilation are often the fastest to shift towards American names.14

There is a notable exception to the narrative that immigrants want to assimilate and be perceived as American. As they adopt mainstream cultural markers, African immigrants can be perceived as African-Americans. African-Americans are, of course, a marginalized group that also faces discrimination. If an immigrant thinks the discrimination they’d face as a (perceived) African-American would be worse than the discrimination they’d face as an immigrant, they might logically choose to maintain attributes that mark them as not-American. And this is just what Adida and Robinson find. Somali immigrants who look similar to African-Americans15 are less likely to adopt American-sounding names.

This is suggestive that name assimilation is something immigrants do only when they perceive it to be advantageous. If assimilation would lower their social standing, they will choose to assimilate less.

One last caveat to choosing an American name: if this was an investment in pursuit of better economic outcomes for one’s children, it wasn’t a very good one. When Abramitzky, Boustan, Eriksson and Hao looked at brothers whose names varied in their level of “ethnicness”, there was little difference in their level of education achieved or likelihood of employment. Learning English might matter for your integration into American society, but at least in 1920, no one cared if your name was Hyman or John.16

The US Also Assimilates To Immigrants

It would be remiss if I did not point out that assimilation is not a one-way street. Even as immigrants (mostly) move towards the dominant American culture, the dominant American culture moves towards them. If we conceptualize the US as a melting pot, the composition of the melt is affected by additional ingredients.

We can see this in names - some of the names that were once distinctively “ethnic” are now just… names. In the Age of Mass Migration, “Eric” and “Kurt” were names that one could easily identify as belonging to an immigrant from Germany or Scandinavia. In 2025, those are names one might choose in order to assimilate into American society.

It is likely we will see some of the same things in future - names, foods, and practices once thought to be foreign that become Just How Americans Do Things. Being American is a dynamic thing, but US immigrants seem more than happy to try to meet that moving target.

So: what does this all mean for Vivek Ramaswamy and the other second-generation immigrants like him? Is he - are they - truly American?

The empirical evidence suggests yes. Current immigrants are following (largely) the same path of past immigrants. Over decades, today’s immigrants will become the long-ago ancestors of tomorrow’s “just Americans.”

Indeed, Ramaswamy himself clearly views himself as American; he is so invested in the American project that he ran for office. It is true that he still follows some distinct cultural practices,17 but in ways, that only makes him more American.

Over the last two and a half centuries, millions of Americans have done the same, slowly flavoring the melting pot with a few new spices.

Many thanks to Claire Adida for sharing her politics of migrant exclusion and inclusion syllabus. All mistakes are my own.

The United States has a large population of undocumented immigrants, making up perhaps 3% of the US population. Close to one in four immigrants in the US are not documented.

This is relatively high for a rich country, and much US political discussion is devoted to the issue of undocumented migrants. Therefore, it seems worth understanding who undocumented migrants are, and how they fit into both US society and the US economy.

Two notes to start:

A note on what this is: This is designed as an overview of what the academic literature says about the US’ undocumented population at the time the constituent papers were written. It is descriptive, rather than prescriptive. It does not attempt to recommend policies, because that is not what a literature review is for.

In other parts of this Substack, I make arguments based on personal opinion; for the migration living literature review, I try to stick closely to only what the evidence says. I show that undocumented migrants in the US are largely law-abiding people who participate in the labor market, because that is what the evidence shows; it is not an attempt to make a political point.

A note on data: somewhat by definition, the empirical evidence on the undocumented population is weaker than on documented residents of a country. There is limited information available on a population that has limited interaction with the legal institutions of a country.

For instance, the method for determining how many undocumented people are in the US is fairly rudimentary. It simply involves taking the number of temporary legal immigrants2 and applying mortality table data to the number of green cards issued, to determine how many documented permanent residents there “should” be in the country. One then sums the number of permanent residents and temporary immigrants, and subtracts the sum off the total population to find the number of undocumented people.

There are several ways this could be wrong, though:

1. Our population estimates might be wrong. Undocumented people typically try to avoid interacting with the state; this is likely to include census takers. The Census makes some assumptions on what percentage of the undocumented population they are able to reach, but they could be wrong.

2. Mortality life tables might not accurately reflect the survival rates of green card holders. If immigrants live longer than expected, one might see “too many” people in the population data and assume that they’re undocumented - but actually, they are green card holders who are unexpectedly healthy.

The difficulty in accurately counting this population means that even descriptive statistics have significant error bars. For instance, the Pew Center is one of the major organizations that attempts to measure the undocumented population, and several economics papers (e.g. Borjas 2016, Borjas 2017) build on their work. But Bhandari, Feigenberg, Lubotsky and Medina-Cortina 2021 believes the Pew Center underestimates the population of undocumented migrants from Mexico by 35%; another paper argues that Pew underestimates the undocumented population by 50%.

Given the uncertainty of even basic descriptive statistics, then, this post relies less on causal inference than most of my migration literature review. This post is less about establishing the effect of changes in policy on undocumented people and more about trying to establish who undocumented people in the United States even are.

Descriptive Statistics

With that out of the way, who are undocumented immigrants in the US? Or at least, who are they (to the best of our knowledge)?

Demographic Characteristics

Borjas 2016 works with a (de-identified) version of the Current Population Survey to find that (as of the mid-2010s), undocumented migrants are:

• Disproportionately male (54%)

• Younger (on average) than either the native-born or legal immigrants.

• Have much lower education levels than either the native-born or legal immigrants. Notably, nearly half of undocumented migrants (42%) do not have a high school diploma, while only 7% of the native-born do not have a high school diploma.

• … and are concentrated in states near the border3

Country Of Origin

About 70% of undocumented migrants in the US come from Latin America, with about a third to half coming from Mexico and another ~20% from the Northern Triangle of El Salvador, Honduras and Guatemala.

However, the Discourse often neglects that perhaps one in three undocumented migrants in the US is not from Latin America. About 6% of undocumented migrants come from India; another 6% of undocumented migrants come from Canada and Europe.

Tenure

Most have been in the US a while. Only about 20% have been in the US less than five years; 40% have been in the US more than 15 years.

Some of this may (ironically) be due to increased enforcement over the last several decades. Many undocumented migrants would like to be circular migrants - coming to the US, earning some money, going back to their home country, and repeating the cycle. But as the border becomes riskier and crossing becomes more difficult, it becomes more dangerous to go back and forth - migrants are more likely to pick one side and stay there.4

Furthermore, if a migrant now does not think trying the border again is a good idea, they are more likely to bring their family with them. For a six-month stint away, it is likely one would leave a wife and children at home; for a ten-year stint abroad, one is more likely to try to bring them along.

Given the long tenure of migrants, it is also very common for households to be mixed-status. Around 70% of undocumented migrants live in a household where at least one person is either a citizen or documented migrant.

The most common situation appears to be a US citizen child living with undocumented parent(s); there are about 4.4 million such children. This means about 6% of US children live with an undocumented parent.

Population Size

As noted above, there is some disagreement on how many undocumented people live in the United States. The most commonly cited number is around 11 million, though other sources find up to 16 million.

Most sources do agree that the undocumented population likely peaked around 2007, and then decreased by several hundred thousand per year through 2021 or so. This is likely for two reasons:

• Employment in construction - the second-largest sector for undocumented immigrants in the US - dipped from 2006 to 2022.

• Demographic shifts - and lower fertility - in Mexico and Central America have meant there are simply fewer young people who want to move (Hanson, Liu and McIntosh 2017).

In recent years, it looks like undocumented migration is on the upswing again, though the total is still significantly below its 2007 high.

Employment

Undocumented men are much more likely to work than documented men. Borjas 2016 finds that 87% of undocumented men are employed, compared to 78% of legal immigrants and 73% of native-born men.

Undocumented women, however, are much less likely to work than documented women. Only 55% of undocumented women are employed, compared to 59% of legal immigrants and 67% of native-born women.

Doing the weighted average,5 this means undocumented migrants are, overall, more likely to be employed than either legal immigrants or the native-born. Nor does this seem to have changed since 2014 (the last year in the Borjas sample); in 2022, Pew estimated that 75% of undocumented migrants were working, while only about 60% of the overall US population is currently working.

The high labor force participation of undocumented migrants makes sense for several reasons:

• Legal immigrants and the native-born have access to social safety net programs that undocumented people do not. For instance, undocumented people cannot access unemployment or disability benefits. Without such benefits, work is likely to be not very optional.

• Furthermore, undocumented immigrants are in a considerably worse financial state than documented immigrants or the native-born and are thus unlikely to be able to survive on savings. In 2008, the average undocumented Latino immigrant had a net worth of $38,000; the average documented Latino immigrant had a net worth of $66,000. The gaps for other racial groups are similarly large (see table 2).

• Undocumented migrants come to the US largely for economic opportunity, balancing the higher wage-earning potential with the legal risks. If one is not currently pursuing that economic opportunity, being in legal jeopardy is less worthwhile.

However, this high labor force participation is still worth noting because of the relatively low levels of education in the group. Among legal immigrants and the native-born, employment rates are lowest for those with the least education. An undocumented man is 17 percentage points more likely to be employed than a native-born person, controlling for education, age, and socioeconomic status.6

I’ll discuss aggregate effects in more detail in the fiscal effects section, but this high labor force participation rate is a significant reason that undocumented migrants are a net fiscal positive for the US.

Wages

Borjas 2017 uses Current Population Survey data to look at the wages of undocumented migrants. Perhaps unsurprisingly, they are low; at the age of 25, perhaps 18% below documented immigrants and 26% below the native-born.

More surprisingly, they are completely flat over time. The average documented immigrant makes considerably more at 45 than at 25; the average undocumented migrant does not.

This also means the hourly wage penalty for being undocumented rises over time; by age 45, the average documented immigrant makes about double what the average undocumented migrant does. Most of this wage penalty does not appear to be due to status; documented immigrants are a different population with different (and higher) expected earnings, largely due to higher education levels among the documented population.

But there is a significant wage penalty for status alone - around 10%. Put another way, if an undocumented worker became documented tomorrow, they could probably get a job paying about 10% more.

Interestingly, this wage penalty is much lower than it used to be; Borjas estimates that in 2007, the wage penalty for being undocumented was about double what it was in 2014. He does not really have an answer for why this is, though.

Fiscal Effects

Undocumented immigrants are sometimes considered to be a drain on the US economy. This appears to be exactly the opposite of what actually happens - since undocumented immigrants often pay taxes, but are entitled to few benefits.

How do undocumented migrants pay taxes?

In order to work in many US jobs,7 you have to fill out a W-9 where you list a Social Security Number (SSN) or an Individual Taxpayer Identification Number (ITIN).

Acquiring a SSN requires legal status, so some undocumented people list a fake SSN or list a SSN belonging to someone else on their W-9. Furthermore, you don’t need legal status to acquire an ITIN. Historically, the IRS does not care about your documentation status;8 even as an undocumented person, you very much still have to pay federal income tax.

Not all undocumented people work a job that does federal tax withholding; some work solely cash-in-hand. However, even they cannot escape the long arm of taxation, as consumption taxes are levied at the point of consumption. No one checks your immigration status when you buy clothes, food or other goods; sales tax is always levied.

What do undocumented people contribute to the economy (besides taxes)?

Per Edwards and Ortega 2017, undocumented people contribute about 3% to the GDP of the US.9 This is somewhat less than their percentage of the workforce, as they are paid less than the average worker.

Still, this is a substantial amount; undocumented people contribute (roughly) similar amounts to the economy as the defense industry.

The Institute for Taxation and Economic Policy estimates that undocumented people pay about $60B in federal taxes each year, and about $30B in state and local taxes each year.

What is the overall impact of undocumented migrants on the US’ fiscal policy?

Undocumented migrants clearly pay some taxes in the US. But do they pay more in taxes than they receive in benefits? There have been a number of attempts to figure this out.

I’ll start with the most negative findings:

1. The Heritage Foundation finds that the average undocumented migrant household is a net tax recipient, because they receive more benefits than they pay in taxes. This costs the state around $15,000 per year per household.10

2. In 2006, the Texas Office of the Comptroller calculated that undocumented migrants generated $424M for the state but cost local governments $1.44B.11

   However, most other estimates are positive.

3. Hanson 2009 estimates that undocumented migrants raise native welfare by 0.03 percent of GDP. This is basically indistinguishable from zero, as it would be about $26/person/year.12

4. Alex Nowrasteh, Sarah Eckhardt, and Michael Howard have modeled the economic contributions of immigrants by education level.

If we use Borjas 2016’s decomposition of the educational makeup of undocumented migrants, 42% of undocumented migrants have less than a high school diploma (Nowrasteh et al NPV -$85,000), 29% have a high school diploma only (NPV $35,000), 13% have some college (NPV $141,000) and 16% have a college degree (NPV $375,000).

If an undocumented migrant had average earnings for someone of their education level, their NPV would be +$52,780, and thus, they would be a net contributor to the US.

However, we know that undocumented migrants make less than the average immigrant, so this is probably somewhat an overestimate.13 Given the legal-undocumented wage gap is about 10%, I’d adjust all contributions downwards by 10%. In this case, you’d end up with a still positive but smaller net contribution of about $40,000 per undocumented immigrant (in NPV).14

5. Chassamboulli and Liu 2024 uses a calibrated model to find undocumented migrants increase native welfare, but there’s not an obvious way to convert this into a dollar value.

Notably, this result does not come directly through their fiscal contributions in the form of taxes, but through increased firm profits and an overall increase in jobs.

It’s worth taking a second to dwell on this point. Undocumented migrants do pay taxes and consume benefits, as do all members of a society. But looking at only this may underestimate the impact undocumented migration has on an economy, because immigrants also consume goods and services. They participate in the economy, and it adjusts around them. Other people - who may not be undocumented - are employed in order to sell them things.

A strict Heritage-style “here are taxes paid and benefits received” accounting may underestimate what undocumented people bring into the US economy. This is discussed in more detail in Clemens 2022, but suffice to say that he finds that adding an immigrant with less than a high school diploma - the lowest-earning education group and thus, the group that pays the smallest amount of taxes - has significant and positive NPV.15

Deportations

Another way one can examine the economic impacts of undocumented migrants is to look at what happens when they aren’t there - or rather, are forcibly removed.

In 2008, a new data sharing program between local law enforcement and Immigration and Customs Enforcement (ICE) was introduced. This program, called Secure Communities, increased the likelihood that undocumented migrants would be detained and deported.16

East, Hines, Luck, Mansour and Velasquez 2022 and Velasquez and East 2024 examine the effects of this program (and its quasi-random rollout). They find that undocumented immigrants were less likely to work - but also that wages and employment for US citizens declined in response to deportations. For every 1% that employment of likely-undocumented people declined, wages for the US-born declined 0.1%.17

There are several reasons this could be true:

• As firms hire more documented workers, this increases their labor costs (since as previously noted, undocumented workers are cheaper). This in turn reduces their ability to create new jobs.

• Undocumented people consume goods and services, and if there are fewer of them, there is less employment among US citizens providing said goods. And indeed, East et al find that employment drops particularly in local consumption.

Velasquez and East 2024 adds another explanation: employing undocumented women - in childcare - allows US citizen women to work. The implementation of Secure Communities reduced the amount of childcare available, and thus, particularly affected the wages of women with young children.

Thus, undocumented people seem to fill an important role in the labor market. Combining these papers with the macro evidence above, I would say it is more likely than not that the presence of undocumented people increases the income of the US-born.

Crime

Money is, of course, not everything; as noted in my crime and immigration post, many people worry that undocumented immigrants are more likely than the native-born to commit crimes.

The evidence suggests they are not, though. There is surprisingly strong evidence for this: this is a consistent result across a variety of contexts. Undocumented migrants are less likely to be arrested and incarcerated in Texas and Georgia; another study in Texas finds the same results. There is no association between the size of the undocumented population and overall crime rates.

When cities adopt sanctuary policies (such that they may attract more undocumented immigrants), they do not experience an increase in crime rates (and may even experience a decrease). Nor does increasing deportations decrease crime - not in North Carolina, not in Arizona. There is simply no evidence that “[undocumented migrants are] bringing drugs. They’re bringing crime. They’re rapists” - and reasonably strong evidence that really quite a lot are “good people, I assume”.18

Alex Nowrasteh has written a longer post going through possible reasons why undocumented people are less likely to commit crime than the native-born, but I think the most likely explanation is simply that the punishments are worse for an undocumented person. If a native-born person commits a crime, they face possible fines and imprisonment. If an immigrant commits a crime, they face possible fines, imprisonment, and deportation. Undocumented migrants, in particular, have often faced considerable risk to move to the country they are now living in; it seems very rational to avoid anything that might put that in jeopardy.

It is undoubtedly true that there are individual undocumented people that commit crimes, just as there are individual citizens that commit crimes. As a group, though, undocumented people in the US seem to be pretty law-abiding once they arrive in the US.

Legalization

There is one last topic that is of relevance to the economics of undocumented migration. What if undocumented migrants are given legal status?

The US has done this via several programs. In 1986, the Immigration Reform and Control Act (IRCA) granted legal status to 1.7 million undocumented migrants; more recently, Deferred Action for Childhood Arrivals allowed some undocumented migrants who arrived in the US as children to gain work authorization.

Not surprisingly, having the ability to work legally increases one’s likelihood of having a job at all. It also increases wages among those who are working - consistent with the wage evidence previously discussed.

Legalization has somewhat more contradictory results in terms of educational attainment. Legalization does make youth more likely to finish high school, but it’s unclear what effects it has on college attendance. It’s possible that DACA recipients are more likely to attend college - but it also might make university less attractive to newly legalized immigrants.19

Even if it does, though, I don’t think this should be an argument against legalization. One framing of this might be “undocumented kids drop out of college when they get status”; another, equally-valid framing would be that “when undocumented people are given the opportunity to earn more money, become productive members of society20 and pay more in taxes, they take it”.

Indeed, this appears to be the through line for what we know about undocumented people in the US. Undocumented people want to work, and indeed they do. They work, they pay taxes, and (largely) abide by the laws of the country in which they live. The US benefits from their presence.

Many thanks to Akshay Narayanan and Ruth Grace Wong for their edits. As always, all mistakes and typos are mine.

These arguments tend to focus around two things:

1. Do firms employ fewer of the native-born because they employ migrants instead?

2. Do immigrants contribute more in taxes than they consume in services?

I have written elsewhere about the answers to these two questions. However, I think these debates may ignore the most important part of immigration: its effect on innovation.1

Before we get into the evidence: it’s worth specifying why we care about innovation.

Economists model innovation as the only way to improve living standards (in the long run). Innovation in this sense is not necessarily limited to patents and scientific innovation; it can also include management improvements, process improvements, or any kind of change that improves efficiency. When defined this way, this makes sense; if you do exactly the same things in exactly the same manner, you will get… the same output. For things to improve, something has to change - and those changes are labeled as “innovation”.2

Given this, it is obvious why a society would want to produce more innovation. It is hard to look at life and think that this is the best it could possibly be;3 we want to live longer, to work less, to spend more time doing things we want to do and less time doing tedious things we hate. We have to innovate for that to happen. In some sense, producing innovation might be the whole ballgame; perhaps it should be the primary goal for a society.4

Why would innovation be linked with immigration?

Theoretical Reasons

Immigrants Are Weird5

It is not obvious ex ante where people get innovative ideas. Ideas are notoriously hard to pin down; inspiration is a fleeting and fickle beast.6 It seems likely, though, that the types of ideas you produce are shaped by your environment. Perhaps an unusual combination of environments - e.g., growing up in one country and migrating to another - will produce unusual ideas. With the low-hanging fruit largely plucked,7 we might need more and more unusual thinkers to keep making breakthroughs.

One might argue that immigration isn’t the only way to produce weird people and weird ideas. And that’s true, but it is one way.

Idea Recombination

Furthermore, it is also a way to bring ideas that are old news in one area into another. Matt Clancy has written about how one of the most reliable sources of innovation is combining disparate “old” ideas into a new idea. That is, to produce new ideas, you want to maximize your idea recombination rate.

Since we know that people’s interests and knowledge vary based on their country of origin, combining source-country ideas with the ideas “in the water” in a receiving country could be fertile ground for innovation. You actively want different perspectives, backgrounds, and thought processes bumping up against each other.8

Note that this is different than the above because the first explanation involves the immigrant being more innovative; the second suggests that the combination of immigrants and natives creates more innovation.

The Practical Reasons

OK, that’s a very 30,000 ft view of why innovation and immigration might be linked. But there are more prosaic reasons to think this.

(For the below, I focus on data from the United States because we simply have more data from the US; later on in the post, I’ll discuss how much this applies in other countries.)

Selection

There are many reasons that immigrants might select into R&D careers. In the US, it is undeniably at least partially driven by the world-class university system; some percentage of US immigrants are immigrants because they want to work in the world’s R&D lab.9

The US effectively vacuums up talented people who want to innovate from all over the world and reaps the benefits of their productivity.

Unobservables

Secondly, the unobservable characteristics of immigrants are likely correlated with the characteristics of good innovators. Immigrants are people who have picked up and left their home country in search of better opportunities, which means they tend to be selected for high levels of resilience, tolerance for risk, and general desire to fuck around and find out. Since innovation also involves trying new things, it seems plausible that immigrants are particularly well-suited to this.

Immigrants Are Already The Innovation Workforce

Immigrants are more likely than the native-born to work in STEM, innovation and R&D. Immigrants are considerably more likely to pursue STEM degrees (54%) than the native-born (35%). About a quarter of the US R&D workforce consists of immigrants (compared with 15% of the population).

This means changing the nature and structure of immigration will change the nature and kind of innovation that happens even if you believe absolutely nothing of the story above.

Are immigrants actually more innovative than the native-born?

Yes. Look, I try to convey a reasonable level of uncertainty about most of my conclusions in the academic literature, but this one seems pretty clear.

Hunt 2011 finds that college-educated immigrants are a bit more than twice as likely to produce patents than the native-born. Bernstein et al. 2022 finds that immigrants who entered the US in their 20s or later10 produced 40% more patents than would be expected based on their percentage of the population.

Neither of these papers studies the full universe of immigrants, but these are large advantages in patenting. Furthermore, almost all patents are filed by the college-educated;11 it is not possible for non-college-educated natives to make up this gap in patenting.12 Therefore, I feel safe in saying that the average US immigrant is more likely to produce a patent than the average native-born American.

Why are immigrants more innovative?

There is at least some evidence for all of the practical mechanisms outlined above.

1. Immigrants are likely predisposed to want to pursue innovation.

As noted above, immigrants are more likely to pursue STEM degrees and be employed in R&D. They are also more likely to become entrepreneurs. Entrepreneurship and innovation are likely linked; generally, if you create a new firm, it is because you think you can do better than the current state of the art. Azoulay, Jones, Kim, and Miranda 2022 finds that immigrants are about 80% more likely to found a firm than the native-born. About half of venture-backed companies have at least one immigrant founder.

This pattern isn’t unique to the US; 54% of Britain’s fastest-growing companies have an immigrant founder. This is some 3.3x the percentage of immigrants in the British population.

Immigrant-founded firms are also particularly innovative. Brown et al. 2019 looks at 16 different measures of innovation. Immigrant-owned firms were more likely to produce 15 of those kinds of innovation than native-owned firms.

None of this proves that immigrants were definitely more interested in innovation than the native-born even before immigrating, but it is certainly suggestive. It seems very likely to me that immigrants are a population that are particularly likely to want to try new things, and that they choose careers that will let them pursue these innovative ambitions.13

2. Immigrants are selected to be particularly good at innovation.

The universe of people who would like to immigrate to the US is far larger than the number who are actually permitted to move to the US. US immigration policy - and particularly, US universities - can be very selective in who they choose, and they generally try to select the highest potential individuals (who will do particularly good work).

International admission rates to US universities are typically much lower than domestic admission rates. This allows universities to select for only the highest potential individuals.

For instance, Gaulé and Piacentini 2013 finds that Chinese graduate students at US universities are more productive than the average graduate student. They believe this is likely selection - that Chinese graduate students in the US come from “a restricted number of very selective elite institutions in China”.

Combining these two arguments, though, one might worry about brain drain. If the US is simply vacuuming up talent from the rest of the world, this could be a zero-sum game. The US benefits; other countries do not.

I don’t think this is the case, though, because:

Immigration also makes more people more innovative.

Most of this effect is probably “moving to opportunity”. This is often, but not exclusively, in the form of a better job, but it can also include moving to education. A new job market paper by Manfredi Aliberti shows that getting a PhD increases one’s likelihood of patenting. We also know that migrating to a higher-income country makes you more likely to get a PhD. Therefore, moving to a place where you are more likely to get a PhD will also increase your expected number of patents.

Let’s use a stylized example here. Consider a person who is clearly very talented but lives in a not great neighborhood of Kigali, Rwanda. She’s the best science student in her school by a long way, but her parents don’t make a lot of money, and she certainly doesn’t know anyone who is an academic researcher. She manages to get into a US university on a scholarship - not a great one, but a pretty decent one.

Suddenly, her world is changed; being a pharmaceutical researcher is a career that, like, actually exists. After university, she goes on to chemistry graduate school, and then becomes a biochemistry professor. Her research focuses on designing new pharmaceuticals. In this role, she almost certainly produces more innovation than if she had never left Kigali.

It is not that she would have definitely not produced anything in Rwanda. She was a great student; she likely could have attended university at home, perhaps eventually ending up in research after all. But she has more opportunities in the US than she would in Rwanda, and the likelihood of having a high-impact career is higher.

Does this particular person exist? Not that I’m aware of - this particular example is an anonymized amalgam of a number of people - but she serves as an example. It is easier to have a productive research career in a high-income country than it is in a low-income country. Moving to a high-income country both increases your likelihood of pursuing a career focused on research and development and increases your likely productivity in your career.

Indeed, Agarwal, Ganguli, Gaulé and Smith 2023 finds the migration-induced productivity increase matters more to the overall level of production than the increase in likelihood of having a research career at all. This would suggest it’s not a zero-sum game: that moving our hypothetical person from Rwanda to the US doesn’t just move her productivity from Rwanda to the US, but increases the total amount of productivity that happens.14

That’s all very nice, but I care about the UK/Europe/Australia/China.

Is all of this unique to the US? One could imagine it would be - the US has the best universities in the world15 and many of the most innovative firms. Perhaps this beneficial effect is specifically about moving to the US, not immigration in general.

In general, I don’t think so. Kahn and MacGarvie 2016 looks at US-trained scientists that return to their home country after their PhD. If a person moves back to a comparatively rich country, their output tends not to decline. If a person moves back to a much poorer country, it does decline. This would suggest that it is not that the US is unique, but rather, the US is one of a set of countries with high innovative output.16

If a country provides more opportunity than a migrant’s source country, it seems plausible to expect migration to increase innovation; if it provides fewer opportunities than the source country, well - why did they move there anyway?

What happens to natives?

Earlier, I posited the idea that the migration of innovative immigrants might increase native innovation as well, through the combination of different ideas and perspectives.

On balance, I think this is likely true - that migration of innovative people likely results in somewhat more innovation among the native-born. But there is at least some evidence for positive effects, no effects, and negative effects.

The Evidence That Immigrants Increase Native Productivity

There are four papers I really like here:

1. Hunt and Marjolaine Gauthier-Loiselle 2008 estimates that the increased likelihood of immigrants to produce patents would lead to a 6% increase in patenting for every 1 percentage point increase in the population of college-graduate immigrants. They actually find that patenting increased 15%. This suggests that patenting rates also increased among the native-born (and indeed, most of the increase in patenting is due to spillovers, not direct production by immigrant inventors).

2. Bernstein et al. 2022 finds more evidence of spillovers from immigrant inventors. When an inventor dies, their collaborators also produce fewer inventions. This effect is stronger for immigrant inventors than for native-born inventors, suggesting that immigrant inventors are more likely to cause others to produce more inventions.

3. Moser, Voena and Waldinger 2014 finds that patenting increased substantially after German-Jewish émigrés came to the US after the Nazis came to power in 1933. The émigrés seem to have drawn in new researchers into their field.

4. There’s even some direct evidence that the mix-and-match theory of innovation is correct. Posch, Schulz and Henrich 2024 finds that places with more surname diversity also produce more patents.

We can also examine total factor productivity.17 Peri, Shih and Sparber 2015 finds positive productivity impacts in areas with many H-1B recipients. Ottaviano, Peri and Wright 2018 finds similar results in the UK.

The Evidence That Immigrants Hurt Native Productivity

That’s reasonably strong evidence that immigration increases native productivity. What’s the case against?

Kerr and Lincoln 2010 finds basically no spillovers - Chinese and Indian immigrants produce more patents than the native-born, but there is no change in the number of patents produced by people with Anglo names. But that’s at worst neutral, not harmful.

There is one notable case where it seems like innovative immigrants really did hurt American incumbents. When the Soviet Union fell, about 1,000 mathematicians migrated to other countries, with the plurality going to the US. After their arrival, US mathematicians who studied similar topics experienced large declines in output and often ended up moving to worse academic institutions. There was no expansion in overall mathematics output - it was a zero-sum game where Soviets replaced Americans, and Americans suffered as a result.18

I am not sure I have a good explanation of why this case seems different from the Moser et al 2014 paper cited above. Both concern academic researchers fleeing negative conditions at home to the United States; neither was in response to local demand for researchers. Perhaps there was more slack in the academic market in the 1930s than there was in the 1990s.

Still, most of the papers in this literature seem positive, putting the careers of mathematicians who study integral equations aside. Having a diverse set of collaborators seems to be good for the native-born.

Summary

So: the evidence suggests:

1. Immigrants are more innovative than the native-born.

2. Moving to opportunity makes immigrants more innovative than they otherwise would be.

3. Immigration probably improves the productivity and innovation of the native-born.

All of this seems very good - and likely quite important. Indeed, if we only consider immigration’s fiscal effects, we may significantly underestimate the benefits of immigration.

As I did last year, I’ve gathered all the migration-related job market papers from the top 150 economics departments1 (as ranked by RePEc) and top 10 ag econ departments (as ranked by RePEc). I have also cross-checked via econ.now for papers involving “migrants”, “migration”, and “immigration”.

Earlier this week, I posted a roundup of international migration job market papers; this post focuses on domestic (within-country) migration. There is, of course, some overlap between migration and urban economics; I’ve used my own judgment for when papers about moving counted as being about domestic migration.

I’ve organized the papers by topic. The topics are listed below:

• Climate Migration

• Economic History

• Household Relocation

• Internal Migration in China (and Hukou)

• Urbanization

Climate Migration

Adaptation in Motion: Temporary migration under heat stress

Moumita Das (UCSC) and Anirban Sanya (Reserve Bank of India)

The impact of climate-induced temporary migration remains largely unexplored. Yet, this flow is widespread in developing countries and also responds to warming. The distinction from permanent migration is critical: because temporary migrants are often under-counted and unaccounted for in local administrative planning, they generate a distinct externality through the systematic underprovisioning of public services. Using a large-scale panel survey in India, we find that a one-degree rise in mean daily temperature increases temporary out-migration rates by 2%-6%. To investigate spatial spillovers under widespread climate change, we develop a model with both migration channels where temperature affects productivity and the under-provisioning of public services degrades local amenities for everyone. We use this framework to quantify the welfare costs of restricting each migration channel and compare different policy responses under climate change. Under the IPCC SSP 5-8.5 climate change scenario, restricting temporary migration generates welfare costs larger than restricting permanent migration, demonstrating that temporary flows are a critical but overlooked adaptation mechanism. Remedying the under-provisioning of services for temporary migrants delivers more than thrice the welfare gains from cost-equivalent place-based adaptation measures. These results have implications for the allocation of scarce climate adaptation funds in developing countries.

Climate Change, Income Inequality, and Migration in a Spatial Economy

Michael Tueting (University of St. Gallen)

Negative local labor market shocks create strong incentives to migrate, yet low-income households often remain in place. This paper studies how income shapes migration responses to climate change and the resulting welfare effects. Using Brazilian Census data, I show that high-income individuals are systematically more mobile than low-income individuals from the same origin. To interpret this pattern and quantify the impacts of climate change, I develop a dynamic spatial general equilibrium model with monetary migration costs and liquidity constraints, embedded in a two-sector trade framework in which rising temperatures depress agricultural productivity. The quantitative results imply sharply regressive welfare losses: in already hot regions, low-income households experience permanent consumption declines of about 0.6-0.9% per period, with worst-case losses near 3.3%, while richer households are largely insulated. Two mechanisms drive these disparities: climate shocks reduce agricultural wages in hot areas, and monetary migration costs disproportionately burden low-income individuals, limiting their ability to relocate in response. A counterfactual policy that makes low-income individuals as mobile as high income individuals–modeled as a targeted subsidy–raises low-income individuals’ welfare by 24%, offsets 4% of climate losses, and increases aggregate output by relocating labor to more productive regions. Adaptation to climate change depends not only on where productivity shocks occur but also on who can afford to relocate. Consequently, adverse shocks tend to fall disproportionately on those unable to afford relocation.

Economic History

Lifetime and Intergenerational Effects of Place: Evidence from the Orphan Train Movement

Maxwell Bullard (Texas A&M) and Jacob Van Leeuwen (BYU)

Where children grow up shapes their economic outcomes, but identifying causal place effects is difficult because families choose where to live. To overcome this challenge, we study the Orphan Train Movement, a large-scale child welfare program from 1853-1929 that relocated orphaned children from northeastern cities to families across the United States. Institutional procedures resulted in quasi-random variation in placement locations based on arrival timing to an orphanage. We digitize archival records and link riders to Census data to measure long-run outcomes. We define place opportunity using county-level characteristics capturing education, urbanization, wealth, and labor market size. Riders placed in high-opportunity counties earn more lifetime income, have fewer children, and are less likely to work in agriculture. These effects persist into the second generation. Examining dimensions separately reveals that urbanization, wealth, and labor market size drive effects. We find important age heterogeneity, where older children show larger marginal gains from high-opportunity places despite younger children having higher adult baseline earnings. Decomposing place effects by geographic scale shows household factors are approximately five times larger than county-level measures, though both are independently significant. Intergenerational transmission operates through a persistent change to individuals rather than geographic persistence, as effects continue despite high migration from riders’ original placement counties. Our findings provide the first causal evidence that place effects transmit across generations.

The Great Migration and Those Left Behind

Gabrielle Grafton (Brown) and Michael Neubauer (Brown)

Despite the extensive literature on the Great Migration, very little is known about

the economic and social consequences on Southern non-migrants. We study the impact

out-migration imposes on those who remain behind, focusing on Black Americans who remained in the South during the first decade of the Great Migration. We build a shift-share instrument for the net Black out-migration rate using shocks to manufacturing employment in Northern cities. We find that both Black and White non-migrants experienced increases in wages due to out-migration. Black out-migration also led to farm closures and less racial hostility in Southern communities. We identify labor supply shocks, occupation switching, and changes in the production process as the mechanisms driving our results. At least in the short-term, out-migration can be beneficial for non-migrants.

Forced Removal or Moved to Opportunity?

Dante Yasui (Oregon)

During World War II a large scale forced migration of over 100,000 Japanese American immigrants and their citizen children lead to relocation away from the West Coast. I analyse the effect that this forced relocation had on the post-war earnings of a linked sample of likely internees. My empirical strategy uses an individual-level panel of Japanese and Chinese men and women to compare earnings before and after internment. I find that individuals who were likely interned experienced faster growth in earnings compared to their counterfactual earnings if they had not been interned. The possible mechanisms which might explain this seeming adaptation include occupational change and later migration decisions.

Household Relocation

Parental Job Loss and Geographic Mobility

Andrew Joung (Michigan) and James Reeves (University of Colorado Denver)

Using restricted Census microdata, we examine the role of location and mobility in the intergenerational transmission of parental job loss on children’s long-run human capital. Following job loss, children experience a permanent rise in commuting zone outmigration. Without accounting for the location of the job loss, we find that this mobility leads children to worse neighborhoods with insignificant changes to college attendance. However, this masks sharply divergent impacts between families across different labor markets. Exploiting labor market scarring from the Great Recession, we show that children whose parents lost jobs in scarred markets experience greater outmigration, improved neighborhood quality, and increased college attendance.

The Gendered Effects of Children on Household Relocation

Jiwon Lee (NYU)

This paper studies how fertility and children reshape couples’ relocation decisions and their gendered consequences for mobility, wages, and welfare. Using Australian panel data, I document that before children, either spouse’s preferences similarly predict long-distance moves, whereas after the first birth, husbands’ preferences dominate, and the gender wage gap widens following relocation. To interpret these patterns, I develop a dynamic collective household model with endogenous fertility, joint job search, and bargaining over relocation under limited commitment, in which decision weights respond endogenously to spouses’ outside options. The model highlights two mechanisms: children raise the payoff to specialization and weaken the primary caregiver’s outside option, tilting decisions toward the male partner’s preferences and gains. I provide causal evidence for these channels using a 2015 reform to Australia’s Family Tax Benefit that tightened eligibility for single-earner couples for a benefit paid to the non-working spouse. The reform, which reduced the payoff to specialization, decreased husbands’ job-related moves among couples with young children but increased them for couples with low-earning wives, where the bargaining channel dominated. Estimates from the model show that joint-search frictions align relocations with the fertility window and that bargaining frictions generate substantial inefficiencies, including moves that disproportionately lower women’s wages and welfare. Policy counterfactuals reveal that moving subsidies raise mobility but amplify within-household inequality, whereas insuring caregivers against career interruptions, such as through childcare support, promotes moves with more equal wage growth and welfare without materially reducing overall mobility.

Internal Migration In China (and Hukou)

Take to the Road for the Newborn: The Effect of the Two-Child Policy on Internal Migration in China

Fangxu Duan (Boston College)

A large literature finds that migrants tend to have fewer children, typically by comparing migrants with non-migrants or by exploiting variation in moving costs to study how migration barriers influence fertility. In this paper, I reverse the channel and ask: how do fertility constraints affect migration? Using China’s shift from the One-Child to the Two-Child Policy as a natural experiment in a difference-in-differences design, I find that relaxing the fertility bound increases the probability that one spouse migrates by 3.10 percentage points, relative to a 16.45% baseline migration rate in the 2010 census. This result remains robust even after accounting for the contemporaneous policy of 2014 Hukou Reform. I argue that looser fertility constraints increase the expected number of children by couples, thereby strengthening their incentives for temporary migration to earn higher income and finance child-rearing costs. To interpret and quantify these dynamics, I develop and estimate a household-level dynamic joint discrete choice model of migration and fertility. The model predicts that further relaxing fertility limits (e.g., from the Three-Child Policy to full removal of restrictions) would have limited effects on both fertility and internal migration. In contrast, it shows that reducing non-financial migration barriers (e.g., the Hukou policy) would be a more effective channel for promoting both internal migration and fertility.

Migrant Integration and Social Stability: Evidence from China

Weizheng Lai (Bowdoin) and Yu Qiu (Pitt)

Contemporary public discourse often raises concerns that migration may threaten social stability, fueling support for exclusionary integration policies. We study this issue by estimating the causal effect of China’s recent reform of its internal migration institutions on labor unrest (e.g., strikes). Exploiting variation from the reform’s population-based discontinuity rule, we find that the reform significantly reduced labor unrest. A key mechanism is migrants’ enhanced settlement intentions: to secure the opportunity of settlement offered by the reform, migrants have weaker incentives to engage in unrest. We show that the reform increased the likelihood of migrants remaining in migration destinations, and through a novel causal mediation analysis, we find that enhanced settlement intentions can explain 63 percent of the reform’s effect on labor unrest. Moreover, the reform’s effect on labor unrest is more pronounced in places where migrants live closer to their origins or are culturally similar to natives, making them more inclined to stay once the reform lowers institutional integration barriers. We find no evidence that the reform changed migrant composition, significantly improved migrants’ wellbeing, or prompted governments to tighten social control.

Migration Restrictions, College Choices, and Spatial Skill Sorting

Junni Pan (Purdue)

College education is widely regarded as a pathway to local labor markets, particularly where migration frictions limit labor mobility. This paper examines how such frictions shape college choices in China, where mobility is constrained by both formal migration restrictions and informal barriers. Using a national administrative dataset on four-year college admissions from 2005 to 2011, I show that relaxing migration restrictions through hukou reforms enabled colleges in reformed cities to attract higher-quality students. The largest gains occurred in colleges located in economically more developed cities relative to students’ origins, consistent with the mechanism of improved local labor market prospects. Counterfactual analysis based on a college choice model indicates that easing migration restrictions in major cities strengthens the sorting of stronger students into treated colleges and raises aggregate welfare, though the gains are unevenly distributed. Welfare increases further when students can freely access the highest-paying labor markets. These results highlight the role of both formal and informal migration frictions in shaping spatial skill sorting and welfare.

Parental Rural-Urban Migration and Child Education

Conghan Zheng (UAB/BSE)

Parental migration and family separation are key factors affecting the outcomes of the next generation. This paper examines the joint household decision of parental rural-urban migration and children’s education in China, where the Hukou system restricts migrants’ access to urban public services. I develop a nested discrete choice model that incorporates expected returns to children’s education as part of the parental migration decision. Estimation results using household panel data show that rural parents migrate for better educational opportunities for their children and a wage premium, avoiding high costs but still concentrating in the most restrictive and congested destinations. Counterfactual analyses suggest that education subsidies at the rural origin of migrants are more effective than subsidies at the destination, or even a universal subsidy, in reducing family separation and improving children’s school performance. And all education subsidies are more effective than mobility restrictions in controlling migration flows without harming the usually hidden but highly vulnerable group in labor migration - children, suggesting that policies targeting the motivation for migration are more effective than mobility frictions in controlling migration.

Urbanization

Density and Diversity in African Cities

Andre Gray (UCSD)

The impact of migration is not just a function of how many people migrate, but where they come from. Migrants carry region-specific identities, traits and skills that shape outcomes in receiving areas. In rapidly urbanizing African cities, the composition of migrants may play a negative role, as ethnic and linguistic divisions drive conflict and counteract classic agglomeration forces. This paper disentangles the effects of migrant flows and migrant composition on productivity in destinations. I build a subnational panel of internal migrant flows across Africa and develop a nonlinear shift-share instrument that identifies shocks to both levels and the birthplace composition of migrants. Using exogenous variation from climate, commodity and conflict shocks, I identify changes to the size and composition of migrants. I find that cities that receive migrants from more diverse birthplaces have lower short-run growth, but experience long-run urbanization benefits. The effects of migrant composition are heterogeneous, with more diverse cities experiencing higher ethnic conflict, but also higher rates of structural transformation. The methods proposed have broad applications to identifying nonlinear effects of migration, when relative group sizes matter for outcomes.

As I did last year, I’ve gathered all the migration-related job market papers from the top 150 economics departments1 (as ranked by RePEc) and top 10 ag econ departments (as ranked by RePEc).2 I have also cross-checked via econ.now for papers involving “migrants”, “migration”, and “immigration”.

Economists are, of course, not the only people who study migration; political scientists and sociologists do valuable work here. Since job market papers are relatively rare in those disciplines (and frankly, looking for every working paper written by a job market candidate in any discipline would take approx. the next year), this post only includes economists. If you are a political scientist with a JMP on migration, please email me at lagilbert@gmail.com!

This post contains all the international migration papers; my next post will focus on domestic migration.

Last year, I organized papers by the country studied. This year, I’ve organized by topic instead.

The topics are listed below:

• Assimilation & Integration

• Economic History

• Entrepreneurship

• External Shocks

• Health Systems

• International Students

• Refugee Integration

• Tax Policy

• Temporary Migration

• Undocumented Migration

Assimilation & Integration

Between Arab and White: Syrians and the U.S. Naturalization Law

Donia Kamel (PSE)

This paper examines how legal inclusion shapes immigrant assimilation. I study the 1915 Dow v. United States ruling, which classified Arabs as white and thus eligible for naturalization. Using U.S. Census data, I show that Arab children born after 1915 received significantly less foreign-sounding names, with within-family estimates indicating effects comparable to the naming assimilation typically associated with decades of additional parental residence in the United States. Difference-in-differences analyses relative to Poles and other immigrant groups corroborate these findings. Among first-generation immigrants, intermarriage with natives increased following the ruling, while residential integration lagged behind. These cultural shifts translated into economic payoffs: post-Dow cohorts with less foreign names earned about 37 percent higher hourly wages and were significantly less concentrated in immigrant-intensive manufacturing jobs. Finally, I assemble a new corpus of Arab-American newspapers (1890–1940) to study identity debates, providing the first systematic text-based evidence of how migrants internalized legal reclassification. Together, the findings show that access to naturalization both encouraged and rewarded assimilation—an insight of enduring relevance as contemporary pathways to citizenship narrow worldwide.

Choosing Distance: How Natives Reduce Migrant Exposure in Daily Life

Artur Obminski (PSE)

Interactions with natives are critical for migrants’ economic and social integration, yet segregation often limits such interactions. This paper asks whether natives adjust their choices of restaurants, stores, and other consumption locations in ways that change their exposure to migrants. I address this question using granular data on 43,000 individuals’ consumption across Poland, exploiting the sudden arrival of over one million Ukrainian refugees in 2022. I find that natives shifted away from locations with higher migrant presence, especially those that facilitate social mixing. New establishment openings reinforced this pattern, underscoring how individual choices can constrain the potential benefits of intergroup contact.

Networks, Sorting, and the Productivity Implications of Immigrant Assimilation

Luke Rawling (Queen’s University)

I study how barriers to labor market integration shape immigrant outcomes and the aggregate economy. Using matched employer–employee data from Canada, I document new and detailed differences between immigrants and natives in terms of earnings, labor market dynamics, and sorting patterns across firms. Guided by these facts, I develop and estimate a search model with two-sided heterogeneity, sorting, referral networks, and immigrant assimilation. A quarter of the gap in earnings and half of the gap in firm sorting is driven by allocational barriers--a combination of lower effective job search, higher job instability, and segmented referral networks--rather than differences in human capital. Eliminating these barriers would raise immigrant output by 8.4% and total output by 1.1% by increasing immigrant employment and allowing them to reallocate into higher productivity firms. Feasible integration programs that combine human capital investments with search assistance deliver similar output gains. These gains are achieved without harming native workers due to firms’ endogenous job-creation response. The counterfactuals highlight that integration programs should target unemployed low-skill immigrants.

Economic History

Immigration, job sorting, and health: Evidence from 1920s US immigration policy

Patrick Szurkowski (Pitt)

This project examines how an exogenous decline in immigration flows induces sorting across the distribution of health conditions in the local labor market adjustment process and documents significant changes in the community health indicators. In the 1920s, a set of immigration laws in the United States imposed quotas based on national origin and restricted immigrant flows. These policies cut immigration flows from constrained countries to around 150,000 individuals a year. The effects of this policy change are identified in a difference-in-difference framework utilizing the LIFE-M, historical labor survey, and full count census data. Results indicate that individuals in areas facing larger declines in immigration are observed transitioning into occupations and industries with worse associated health measures. Increasing native-born employment in high health cost (HHC) positions offset lost immigrant labor leading to no change in HHC employment share. Further, policy exposure was associated with declining average lifespan and increasing mortality rates among counties’ working age US-born population.

Immigration Raids and Immigrant Assimilation: Evidence from the 1920 Palmer Raids

João Tampellini (Vanderbilt)

How do immigration raids affect immigrant assimilation? This paper studies the Palmer Raids, a nationwide campaign in 1920 that arrested thousands of immigrants suspected of radicalism but resulted in few deportations. I digitize government records for over 4,000 arrested individuals and link them to the 1920 Census, which allows me to determine their neighborhood of residence. I compare immigrants in neighborhoods where at least one resident was arrested to immigrants in other neighborhoods in the same city. I show that exposure discouraged assimilation in both the short- and long-run. In the short-run, treated fathers gave their children significantly more foreign-sounding names. In the long-run, treated immigrants were less likely to naturalize, more likely to marry a spouse from the same birthplace rather than U.S.-born spouses, and more likely to live in ethnic enclaves. These choices had lasting economic consequences: immigrants in treated neighborhoods had slower occupational upgrading and lower rates of entrepreneurship. These findings show that immigration raids can undermine assimilation.

Entrepreneurship

Curbing Tax Flight? Aggregate Effects of Taxing Entrepreneur Migration

Christine Blandhol (Princeton)

This paper examines the trade-offs policymakers face when imposing out-migration taxes with the goal of preventing tax flight. I use Norway as a laboratory—a small open economy where tax flight is a key policy concern. Exploiting a recent increase in the wealth tax rate at the top of the wealth distribution, I document a significant migration response. The out-migration rate of affected households increased from 0.2% in the pre-period to more than 2% in the year of the reform. In addition to out-migration reducing the size of the tax base, 40% of out-migrating households are firm owners. Firms of out-migrating owners experience on average a 12.6% decrease in firm revenue compared with firms where the owners stay. To analyze the aggregate effects of the reform and the effectiveness of out-migration taxes, I develop a dynamic equilibrium model where heterogeneous entrepreneurs make forward-looking savings and migration choices. Entrepreneurs who choose to operate their firm in a different location than they are currently residing may suffer a hair-cut to their productivity. I estimate the key model parameters using the quasi-experimental evidence from the tax reform. In the aggregate, the wealth tax reform decreases aggregate output in the long-run by 1.3%. Introducing a tax on the market value of the firm when the entrepreneur out-migrates reduces tax flight, especially for more productive entrepreneurs, and increases aggregate output.

Creating Opportunity: The Impact of Immigration on Native Entrepreneurship

Gabriel Chaves Bosch (Queen Mary University of London)

This paper examines the impact of immigration on native entrepreneurship using rich social security data and a unique immigration episode in Spain. Using variation across local labour markets and employing a shift-share instrumental variable for identification, I find that immigration has a positive effect on native entrepreneurship. The effect is primarily driven by wage workers who transition into entrepreneurship following the immigration episode. To rationalise these findings, I propose and calibrate a model of occupational choice and immigration. The model shows the empirical results are consistent with an opportunity channel: an immigration-induced labour supply expansion lowers immigrant wages but has a limited impact on native wages. As a result, immigration lowers labour costs, enabling the creation of businesses that would not otherwise be profitable.

External Shocks

Are Recent Immigrants More Resilient to Job Loss? Evidence from Mass Layoffs in Canada

Olivier Gagnon (McGill)

This paper measures the effect of job loss on the subsequent labor market outcomes of immigrants as a function of the time spent in the host country at the time of displacement. The evidence comes from yearly employer-employee administrative data from Canadian taxes (2001-2019), linked to immigration records. I look at immigrants displaced during mass layoffs, which provide plausibly exogenous job separations. I estimate the impact of displacement in two distinct ways. First, through an event study approach. Second, through a regression-based approach that allows me to quantify how differences in the composition of pre-displacement characteristics contribute to the heterogeneous treatment effects and how the heterogeneity in earnings loss is linked to specific differences in post-displacement outcomes. I find that recent immigrants experience smaller and less persistent earnings losses from displacement, with a 21% decrease in earnings one year after displacement, compared to 26% for those who have been in the host country longer. Recent immigrants also display better post-displacement outcomes in other dimensions, such as lower time spent nonemployed and higher geographic mobility. I show that differences in pre-displacement characteristics account for 50% of the heterogeneous treatment effects on earnings: each additional year in Canada results in 0.8 percentage points larger earnings losses, but only 0.4 percentage points when controlling for pre-displacement characteristics. Age at displacement alone explains half of this difference. Differences in post-displacement outcomes account for an additional 40% of the heterogeneity in earnings losses, with time spent nonemployed being the most important mechanism.

The impact of the 2015 earthquake on internal and international migration in Nepal

Shailee Manandhar (Rutgers)

This study examines the impact of the 2015 earthquake on internal and international migration in Nepal. Using ground shake intensity and district-level death rates as measures of earthquake impact, I analyze two datasets with two-way fixed effects, difference-in-differences, and synthetic difference-in-differences models. The results show a significant decline in international migration from affected districts, particularly among men, with the probability of male migration declining by 3–6 percentage points (a 28-57% decrease relative to pre-earthquake mean) and male labor permits issued for employment in countries other than India declining by 8%. In contrast, internal migration and the low female migration were largely unaffected. These findings provide insight into the labor market dynamics and sensitivity of migration to large shocks, emphasizing the need for stronger social protection systems and creation of domestic employment opportunities to support recovery and resilience.

Health Systems

Quality and Location Choice of Immigrant Doctors

Jason Chen (Columbia)

Doctor shortages are a widespread and growing concern in the healthcare systems of many developed countries, including in the United States. Allowing for immigration of working doctors is a common policy in such by which to expand doctor supply. In the US however, such immigration is bottle-necked by licensing requirements that require domestic retraining, ostensibly due to quality concerns. I study the quality of domestic vs. immigrant emergency medicine doctors in the US. I find quality premiums associated with care provided by immigrant doctors, both within a given hospital and across the entire distribution of ER doctors. Notably, I do not find such quality premiums for US citizen medical students educated abroad. I also find immigrant doctors are significantly more likely to work in locations with worse patient outcomes and in designated health provider shortage areas. Estimates of doctor location preferences suggest that this affinity cannot be explained by initial location or vertical matching. My results imply that current licensing restrictions on doctor immigrants are too strict, and policies allowing for alternative licensing pathways for doctor immigrants could alleviate doctor shortages in the areas of greatest need at no cost to healthcare quality.

International Students

International Undergraduate Student Inflows and College Pricing Strategies

Sheng Qu (CU Boulder)

This paper examines how growth in international undergraduate enrollment affects both sticker-price and net-price tuition at U.S. PhD-granting institutions. Leveraging the relaxation of U.S. visa policy and the appreciation of the Chinese yuan as natural experiments that drove a rise in Chinese undergraduate enrollment beginning in 2005, I use institution-level panel data from 2000 to 2019 and employ difference-in-differences and instrumental variable approaches to identify the causal effects of rising international undergraduate enrollment on tuition outcomes. I find that increases in international undergraduate enrollment raise out-of-state sticker-price tuition at public PhD-granting universities but reduce it at private PhD-granting institutions. Private PhD-granting institutions with greater exposure to international undergraduate enrollment growth also experience reductions in average net-price tuition, while public PhD-granting institutions show no significant change. These divergent responses highlight differing institutional priorities: private universities appear to prioritize school quality and student subsidization, while public institutions emphasize in-state access and budget stability. The findings suggest that domestic students at private universities benefit more from international undergraduate student growth than their counterparts at public institutions.

Refugee Integration

Refugees, Amenities, and the Skill Premium

Elif Basaran (Penn State)

This paper examines how intra-national native migration patterns and region-specific welfare respond to large inflows of immigrants. Leveraging the case of Turkiye, which experienced a substantial influx of Syrian refugees following the 2011 Syrian Civil War, I first provide reduced-form evidence on the effects of the influx on local labor markets and housing rents across skill groups. I then document an increase in native outmigration from refugee-concentrated areas, particularly among the high-skilled, alongside a significant deterioration in local amenities. These changes disproportionately burden the low-skilled natives, deepening pre-existing disparities between skill groups. Finally, to quantify the role of amenity changes in shaping native outmigration, I develop a dynamic spatial general equilibrium model in which amenities evolve endogenously and affect natives’ migration decisions through estimated, skill-specific amenity taste parameters. The model highlights amenity deterioration as a key mechanism behind native flight, and shows how differential mobility and amenity preferences reinforce rising skill premiums. It also provides a basis for counterfactual experiments that explore the effects of refugee reallocation policies and targeted subsidies. These demonstrate the potential for policy interventions to reduce regional distributional gaps and welfare losses.

Sheltering Refugees in Cities: Crime, Public Services, and Voting

Chris Brito (UC Davis)

An extensive literature documents that immigration flows frequently trigger native backlash, often reflected in heightened support for far-right parties. Yet, the mechanisms driving this response remain unclear, whether rooted in economic competition or cultural threat. Because these channels operate locally and through daily contact, identifying them requires granular data that are typically scarce. This paper examines how the reception of refugees affects local public education, crime, and ultimately voting behavior, exploiting detailed within-city data and plausibly exogenous variation from the quasi-random location of refugee shelters. Using a differences-in-differences strategy, I find that refugee shelters had no impact on crime or congestion in local public schools. However, they boosted natives’ support for far-right candidates and reduced votes for the incumbent. Effects are largely driven by shelters hosting culturally diverse refugees comprising different indigenous ethnicities. Using UN reports data, locals’ exposure to vulnerability - children outside of school, child labor, and homelessness - can be behind the results. Together, the results reveal that cultural perceptions can dominate economic channels in shaping local political responses to migration, and that aggregated data can miss essential nuances in locals’ attitudes towards migrants.

Employers and Refugee Economic Integration: The Effect of Early Employer Quality

Alessandro Caiumi (UC Davis) and Emil A.L. Simonsen (University of Copenhagen)

Drawing on matched employer–employee data from Denmark, we study the role of early employers in the labor market integration of refugees. First, using the two-way fixed effects model of Abowd, Kramarz, and Margolis (1999), we estimate firm-specific wage premia that we use as a proxy for workplace quality. Second, we leverage the role of social connections and a dispersal policy implemented between 1986 and 1998, which quasi-randomly allocated refugees across municipalities, to obtain exogenous variation in their exposure to the quality of first potential employers. We find that placement in a municipality where, at arrival, co-nationals are employed by high-quality employers has positive and statistically significant effects on refugees’ employment and earnings for up to ten years. We also present a set of novel stylized facts on refugees and the firm ladder, highlighting the lasting influence of first employers for this group of workers and discussing potential implications for two-way fixed effects models. Incorporating our insights into a data-driven algorithm to optimally match refugees with Danish municipalities leads to a 46% increase in short-run employment probability relative to the status quo dispersal policy. Our results imply that the type of employers available upon refugees’ first entry is an important determinant of their success in host countries.

Migrant Networks, Liquidity Constraints, and Integration Frictions in the Venezuelan Exodus

Agustín Deambrosi (Penn State)

Forced migrants concentrate persistently in overcrowded border regions despite superior opportunities elsewhere. This paper shows that the interaction of liquidity constraints, network externalities, and location-specific integration capital creates powerful path dependence in refugee settlement, and that both the design and timing of assistance are first-order for policy effectiveness. I develop and estimate a dynamic structural model where credit-constrained migrants make forward-looking location decisions, networks endogenously reduce migration costs, and integration capital accumulates but resets upon relocation. Using data from specialized migrant surveys across four South American countries during the Venezuelan exodus, I show how these forces interact: binding liquidity constraints trap early migrants in proximate but unproductive locations; their presence generates network externalities drawing subsequent cohorts to the same destinations; accumulated integration capital then locks both groups in place even as savings permit moves to better locations. The model reveals that intervention timing dramatically affects program effectiveness. Early transport assistance---subsidies for reaching productive destinations deployed in the crisis’s first two years---increases welfare by 4.7% of lifetime consumption. The same budget deployed two years later generates only 2.7% gains, implying a timing multiplier of 1.7. This difference emerges because early assistance triggers compounding network effects that amplify program impact through successive cohorts. Critically, conditional transport assistance dominates unconditional cash transfers of equal value by leveraging positive network externalities that forward-looking migrants do not internalize. These findings speak directly to the refugee recognition debate: Venezuelans’ exclusion from UNHCR programs denies them access to coordinated mobility support, generating substantial welfare costs through spatial misallocation.

Refugees’ Right to Work: Efficiency and Equity in Host Country Labor Markets

Sarah Winton (LSE)

One in every 200 people in the world is a refugee. In most host countries, refugees face legal barriers to work, confining them to informal work or unemployment. This paper studies how granting refugees the right to work reshapes the allocation of refugee and host labor across occupations. I leverage a unique natural experiment – a large-scale work permit scheme for Syrian refugees in Jordan – and assemble a novel dataset to study how the policy impacted the labor market outcomes of both refugees and hosts. Using a shift–share measure of exposure to refugee competition, I document three main effects on Jordanian workers. First, Jordanians exit occupations that are highly exposed to refugees, re-sorting elsewhere. Second, consistent with a standard sorting model, this exit coincides with an increase in the average wage of Jordanians in exposed occupations. Third, re-sorting leads to occupational upgrading, as college-educated Jordanians move into less exposed, higher-paying jobs. To separate the effects of refugee entry from locals’ re-sorting, I build a model of occupational choice nested in general equilibrium. The estimated model implies Jordanians experience modest wage gains and a small rise in unemployment from the policy. Distributionally, the poorest Jordanian workers benefit the most from the work permit scheme, despite being those who lose in a benchmark without re-sorting. Aggregate output increases by nearly 11%, driven by improved utilization of refugee labor and translating into large wage gains for refugees. Work permits unlock aggregate efficiency gains and, through re-sorting, reduce host country income inequality.

Tax Policy

Elasticity of Taxable Income and Social and Cultural Norms: Evidence from immigrants in Canada

Kuot D. Manyang (University of Calgary)

Understanding how cultural and social norms shape taxpayer responses to changes in tax policy is vital to promoting tax compliance and morale. Exploiting exogenous variation in the tax rate from Canadian reforms and detailed administrative data, I estimate taxable income elasticity (ETI) and find that immigrants have a larger ETI (0.094) than non-immigrants (0.078). Then, I estimate the ETI for immigrants’ countries of origin separately and examine how it varies with its cultural norms. Immigrants from countries with cultural norms such as individualism, uncertainty avoidance, trust in others and government, and religiosity have lower elasticities. In contrast, those from countries with cultural norms such as high-power distance, masculinity, long-term orientation, and political corruption are highly sensitive to tax changes. These findings emphasize differentiating between pure and culturally induced behavioural responses for effective tax policy. In addition, promoting economic integration and trust in government initiatives within immigrant communities is essential to encourage compliance.

Temporary Migration

Development Effects of Social Eligibility Criteria for Temporary Migration

Daliah Al-Shakhshir (Stanford GSB)

Temporary migration offers a pathway to higher income, but can be disruptive to communities of origin. I investigate distortions in origin communities that arise in response to social eligibility criteria that gatekeep access to those opportunities. Specifically, I study the effects of age and marital status eligibility requirements in the case of commuting Palestinian migrant labor to Israel. Over time, the minimum age threshold dropped while holding marriage fixed. Using individual-level Palestinian census data and a record of work permit regulations, I employ a triple-differences design that leverages differential cohort and regional exposure to the policy change. First, I show that men bring forward their timing of marriage in line with changes in the minimum age threshold in more highly exposed areas. Second, I find that women, nonparticipants in the work permit program, are more likely to get married earlier, have children earlier, and register declines in educational attainment in those same areas. I do not find effects on their labor market status. These findings demonstrate how policies that condition labor market access on demographics can have potentially adverse consequences for non-target populations through general equilibrium dynamics.

Political Impact of Curtailing Low-skilled Immigrants: Evidence from the Bracero Program

Rui Zhong (GWU)

A central question in U.S. immigration policy is how a reduction in low-skilled immigrant labor shapes the economy and politics. This paper examines the political impact of restricting the supply of immigrant workers under the Bracero Program, a U.S.–Mexico agreement that provided Mexican farm labor to address postwar shortages. This restriction occurred in two phases: first, a 1962 minimum wage increase for Mexican workers and second, the program’s official termination. Using cross-county variation in exposure to Mexican farm workers, I employ a difference-in-differences model to compare electoral outcomes in high- and low-exposure counties after 1962. In the short run, these policies led to a 2.8 percentage point increase in the vote share for the Republican party, which had opposed the program’s termination. This effect is likely driven by voter reaction to higher agricultural prices and by the mobilization of voters through Republican-affiliated media.

Undocumented Migration In The US

Uncertain futures: How did the threat of rescinding DACA affect eligible immigrants’ outcomes?

Mate Szurop (CU Boulder)

Since its introduction in 2012, Deferred Action for Childhood Arrivals (DACA) has been one of the most contested U.S. immigration policies. This article examines how the uncertainty arising from the 2017 attempted rescission of the policy impacted the labor market outcomes of eligible immigrants. Using American Community Survey data, I implement a difference-in-differences research design that exploits the sharp eligibility cutoffs of the policy to define treatment and comparison groups. I find that the threat to end the program had statistically significant negative effects on eligible immigrants’ employment, labor force participation, and total income. I further investigate how state-level support for DACA recipients mitigated these effects and explore heterogeneity by sex, age, and education. My results are robust across a range of different specifications, samples, and undocumented proxies, and pass placebo tests. My paper demonstrates that the outcomes of DACA-eligible workers respond to legislative uncertainty, strengthening the argument for a more permanent legal structure.

Immigration Policies and Human Capital: The Impact on Undocumented College Attendance

David Titus (Cornell)

I estimate the impact of Universal E-Verify laws on the college attendance of undocumented Hispanics in the United States. To do so, I implement a series of event studies that account for staggered adoption over time, and I use a random forest algorithm as my primary approach for predicting undocumented status. My results indicate that Universal E-Verify laws lower the college attendance of undocumented Hispanics ages 18–24 by about 3.7 percentage points. This is a substantial effect: only 15.7 percent of undocumented Hispanics ages 18–24 in treated states were enrolled in college following the passage of the laws. This effect is robust to using logical imputation on non-citizen Hispanics to proxy for undocumented immigrants, using a logit model instead of random forest, testing for migration spillover effects on bordering states, and considering potentially confounding impacts of other state-level policies. I develop a theoretical model that explains the mechanisms through which Universal E-Verify affects college education, and I test this model’s implications. I find suggestive evidence that the effect is driven by a negative labor market shock on undocumented adults ages 25–54, which likely leads to worse schooling for their children and renders college less attainable. These findings indicate that employment restrictions targeting working-age undocumented adults hinder the human capital development of undocumented youth.

Here, I explore what we know about this reverse flow (return migration) - how many migrants will eventually return to their home country, how long they stay abroad, how these migrants differ from those that stay and what happens to both them and their home countries when they return.

How common is return migration?

Somewhere between 30% and 60% of migrants will likely return to their home country within ten years. This appears to have been true for decades if not centuries.

The largest multi-country dataset of return migration seems to be Amanzadeh, Kermani and McQuade 2024. This uses LinkedIn data to track 350 million skilled international migrants across 180 countries.1 After a decade, 50% of migrants are still in the country they moved to; 38% have returned to their home countries; 12% have moved to a third country.

This paper does only consider skilled immigrants, but the 40% return number is remarkably durable across contexts. Ward 2016 found that 40% of migrants who entered the US in 1917-1924 eventually returned home. Akee and Jones 2019 found that 40% of all US migrants in 2005-2007 return migrated within 10 years. There is 80 years between these two measurements!

Obviously, it is not universal that exactly 40% of migrants return to their home countries. In some cases, the return rate is lower; Dustmann and Görlach 2016 finds only a 20% return rate from Australia, Canada, New Zealand and the United States (combining a variety of studies in figure 1). In some cases, it is higher - 85% (!) of Greeks who migrated to Germany from 1960 to 1984 eventually returned to Greece. Across contexts and skill levels, average return rates seem to be around one-third.

This bears some emphasis; for many, many immigrants, immigration is not a permanent decision. Indeed, most workers who go abroad do not intend to stay abroad forever. It is a relatively common pattern for workers to migrate to higher wages, work there long enough to achieve their financial goals and then go home.

We should not treat migration as a one-off, permanent choice; rather, it is sometimes a permanent relocation and sometimes a temporary economic choice.

Who stays and who returns to home countries?

So: who tends to return and who stays?

At the country level

At the country level, the predictors of likelihood of return are not very surprising. Many emigrants from wealthy countries with strong economies eventually return; poor, war-torn countries see much lower rates of return.2

However, that simplification elides the role of the receiving country. It is the difference in your fortunes at home and abroad that appears to make you more (or less) likely to stay. The richer the destination country, the likelier you are to stay long-term; the richer your home country, the likelier you are to go home.

GDP isn’t destiny, though; Indonesia, Zambia and Mongolia stand out as being poor countries with high rates of return, while wealthy3 England has about the same rate of return as Albania.4

Interestingly, cultural similarity does not seem to make you more likely to stay; indeed, the reverse is true. Migrants from another country that speaks the same language - for instance, Nigerian migrants to the UK - are more likely to go home, not less.5

At the individual level

More interesting than looking at the country level, though, is the predictors of an individual staying long-term. What kinds of people are likely to stay?

One could imagine two different worlds. In one, the highest skilled migrants are the most likely to go home, because they have the best options at home and don’t need to stay abroad for good opportunities.

In another, the lowest skilled migrants are likeliest to go home because they are struggling in the new country. For instance, if you’re down on your luck, it’s relatively common to move in with family for a bit to help you get back on your feet. If your family is in another country, back you go to that country.

Which one of these is a better reflection of what actually happens?

On average, return migrants appear to be negatively selected relative to the migrant population. While not universally true, this pattern seems to be relatively robust across countries and time periods. It has been true in the US in the age of the mass migration and it is still true in the modern US, in the Netherlands, in Egypt…

To me, this makes sense.

Return rates are relatively high across the board; a reasonable percentage of returns will be driven by external shocks - for instance, one might have an ailing parent at home and you end up returning to take care of them.6 This is likely to happen with roughly the same frequency among both above-average and below-average migrants; your parents’ health is unlikely to be correlated with your relative success in your new country.

But of the returns driven by skill level, I think the below-average are more likely to return than the above-average. Most people migrate to places where they will make more money than in their home country; staying is probably the earnings-optimizing path if you’re doing OK.

It is the people who are struggling - who might have struck out in the new job market - that are more likely to look around and go “fuck this, I’m going home”. Indeed, this can be driven by the visa process itself - some people have to have employment to stay in the new country,7 and if they’re fired, they must return home.

To the recipient country, this means that people who stay long-term tend to be positively selected relative to the overall immigrant population. If you are a fiscally motivated country, this pattern is clearly great for you. You’ve ended up with doubly selected migrants - you first selected the migrants you thought were valuable to admit and then those that didn’t succeed go home. Amazing for you!8

However, it does raise the spectre of brain drain from sending countries. If only the most successful migrants stay, does that mean that advanced economies are siphoning off the talent of the developing world?

Migration, Innovation and Skill Acquisition

In some sense, yes, this is true by construction. In most countries, migrants are more educated and richer than average; if some of this population never comes home, the country has lost people who are educated and wealthy.

I think this explanation ignores two important things, though:

1. The general equilibrium effects of migration

I’ve written about this elsewhere, but the case for general equilibrium effects boils down to: you are more likely to spend time and effort on training if there are higher returns to that training.

Let’s say you are a person with a high school education in Uganda. You make perhaps $1000 a year. Getting a degree will cost you $500/year, so a total of around $2000 (excluding foregone wages and opportunity cost), but it will increase your earnings to $1500 a year. This is clearly worthwhile, but paying back a loan for tuition would likely take you some time.

Let us contrast the case where you have a 10% chance of migrating to somewhere where you make $10,000 a year. Your expected value of a degree is now considerably higher and you should be more willing to spend money and take out loans for training.

Thus, the number of skilled people in an economy should scale with the returns to skill. Since migration raises the returns to skill, it pulls more people into pursuing additional training. And indeed, this is what we observe in the Philippines; the possibility of becoming a nurse in the US led more Filipinos to train as nurses.

Thus, it is the prospect of being able to migrate that creates some of those educated and wealthy individuals.

2. The skill effects of migration

This is the key value of return migration to a country: migrants gain important skills while abroad. Some of these skills might be impossible to obtain in one’s home country; some skills might simply be easier to obtain in other economies.

The most obvious version of this is for specialized training. For instance, some low-income countries do not have some kinds of medical training (e.g. anesthesia). If one is to become an anesthesiologist, one must migrate to another country. Some of those migrants will stay in the new country; some will eventually return to their home country. But even if only a few come home, you’ll still end up with more anesthesiologists than you started with in the home country, because return migration is literally the only way you acquire anesthesiologists at all.

There are less drastic examples of this as well. There have been a number of industries that were kickstarted by bringing back skills from another country. People go abroad, gain skills and then bring those skills back to improve the efficiency of firms in their home country.

Bahar, Özgüzel, Hauptmann and Rapoport 2019 highlights “the textile sector in Prussia, of the IT sector in India or Israel, of the garment industry in Bangladesh, or of the car-parts industry in Bosnia” as examples of such. They also show that when migrants from Yugoslavia returned from Germany, they brought knowledge that increased the productivity of Yugoslav firms.

That is, return migration is how technological and knowledge diffusion happens. We can see this in knowledge production specifically. Choudhury 2016 finds that migrants in India spurred innovation; Fry 2022 also finds that returning scientists build bridges and collaboration between the production frontier and home countries.

This is in effect a theory of neither “brain drain” nor “brain gain”, but rather “brain circulation” - “human capital enhancement via (temporary) mobility which, implicitly, is used more effectively upon return”.

What’s the counterfactual?

Would firms be better off if those people had stayed in the country the whole time? This is a harder question to answer, but I think the answer is probably not.

In general, having advanced-economy job experience is highly valued in emerging economies. Indeed, it’s considerably more highly valued than having worked the same number of years at home (with returns to international experience up to 200% of that from home experience).

This is very widely reported: in addition to Amanzadeh, Kermani and McQuade 2024’s multi-country data, there are papers showing an earnings premium for return migrants in Hungary, Albania (1, 2), Romania, Ireland, Mexico, “several West African countries” and Egypt. If we think employers are acting rationally, this suggests that skills gained by migrants really are very valuable and home countries benefit from having return migrants.9

Return migrants also appear to be more likely to become entrepreneurs (and employers of others) than those who have never left home. A review found that the likelihood of self-employment is higher among return migrants than non-migrants, even when the migration itself is quasi-randomly assigned.10

Notably, it is not just that return migrants are more likely to employ themselves; they become more likely to employ others as well. Return migrants appear to be particularly likely to be “job creators”, thus raising employment levels in the areas they return to.11

Note that all of this is true even though return migrants are negatively selected.12 Even migrants who didn’t do all that well as migrants - who earned less than average - appear to benefit from the experience.

For instance, during the age of mass migration in the US, Norwegians were more likely to move back when they weren’t very successful in the US. But “upon returning to Norway, return migrants held higher-paid occupations than Norwegians who never moved, despite hailing from poorer backgrounds. They were also more likely to get married after return.”

Beyond economics: what are the cultural impacts of return migration?

This is largely a literature review about the economics of migration, but I am a political scientist. Therefore, it’s worth noting one other impact from return migration: cultural change.

In general, migrants tend to move to more democratic countries (relative to their home country). This is not particularly because people have a strong preference for democracies, but rather because most rich countries are democracies. If you are maximizing earnings, you want to go to Europe or the US and so you want to go to a democracy.

When people migrate, they begin to integrate into their new society, adopting at least some of that society’s norms. When people migrate to democracies, they seem to develop a taste for political accountability.

Batista and Vicente 2011, Chauvet and Mercier 2014 and Tuccio, Wahba and Hamdouch 2019 all find that return migrants keep this taste for accountability and demand more from their home governments.

Conclusions

I worry sometimes that this literature review - of which this post is a part - comes across as too positive about migration. Surely migration cannot be good on all fronts; someone must be worse off because they have lost (or gained) population.

And it is true that there are possible negatives to migration. I think it’s plausible that migration would be net negative for sending countries if all migrants stayed away for their entire lives. It’s also plausible that the fiscal impacts of migration would be worse if every receiving country kept every migrant forever.

Return migration blunts both these criticisms. Sending countries gain innovation and skills; the receiving country keeps doubly-selected migrants. Migrants that stay abroad generally have much higher incomes than if they had stayed home, but migrants that return home also have higher income than if they’d never left.

Return migration is a key part of what makes migration the little intervention that could.

Like last year, I will be doing an annual roundup of economics job market papers on immigration and migration. I would like to do this again, and I would particularly like to include non-economists.1

Please help me out by sending me any and all migration PhD job market papers. Send me your paper. To quote Matt: send me your student’s paper. Send me your friend’s paper. Send me a cool paper you saw online or at a seminar.

Email me at lagilbert@gmail.com.

The administration claims that this will make firms hire Americans instead, and therefore this will increase the economic outcomes for the native-born.

I do not think this is true. I have previously written that the entry of H-1B holders into the US is good for the American economy. It is possible that Trump is right that firms that hire H-1Bs do hire fewer Americans, but it’s equally possible they don’t. What is clear is that the H-1B program increases American innovation. Considering economists believe innovation is the way that living standards increase, that’s a big deal.4 Decreasing American innovation is not worth maybe a few more jobs going to Americans.

There is not some fixed pool of jobs in the United States, where hiring an immigrant means definitely not hiring an American. Immigration exists in general equilibrium, where the labor market changes over time and new industries, technologies, and innovations can be created. Immigrants can (and often do) create jobs when they create new industries and firms. And of course, immigrants also consume goods and services from the native-born. An Indian software engineer does not simply collect his paycheck and go sit in a cardboard box on the street, spending none of it; he goes to restaurants, he hires accountants, he also creates economic activity.

Most immigration policy experts I respect have had similar reactions. David Bier has a post on how this new fee is both a bad policy idea and illegal. Jeremy Neufeld is also tweeting on this.

Part of the reason I’m writing this post is a question from my friend Oscar Sykes,5 which I hadn’t seen discussed yet. He texted me tonight to ask if this would really be that bad. He agrees that more immigration would be better, but since the US has a statutory cap on the number of H-1B visas issued per year, and it’s already vastly over-subscribed, this might serve as a sort of salary sorter. Under this policy, firms would only offer jobs to those they’d be willing to spend an extra $100,000 to employ - clearly, these would be the highest value immigrants.

Indeed, Jeremy Neufeld has previously argued that the US should instead sort H-1B applicants by salary.6 Perhaps this fee could ensure that only the highest-paid applicants get through rather than the current random selection process. If only the highest-salary applicants stay in the pool, that could eliminate the participation of low-wage off-shoring firms in the H-1B program but keep the participation of high-wage product firms in the program. If there are still more than 85,000 H-1B applicants that would get through this new Trumpian salary sorter, overall immigration wouldn’t even decrease.

Unfortunately, I don’t think that the fee would do this, for two reasons:

1. The above logic relies on the H-1B being oversubscribed. It is only true if you can swap out one H-1B applicant for a new one willing to pay the fee. At the moment, this is true for private sector organizations, which must go through the H-1B lottery. Last year, there were about 339,000 people who wanted one of the 85,000 available visas.

However, it is not true for many users of H-1Bs. Non-profits and universities are exempt from that 85,000 visa cap, and can currently hire as many people on H-1Bs as they please. In 2024, I’d estimate that there were about 40,000 new entrants from cap-exempt employers.7

There isn't a pool of 50,000 additional qualified applicants for these visas that one could seamlessly select instead. If you charge large fees to this population, an employer with only a fixed amount of money to spend on salary will hire fewer people.

This seems especially likely to hit postdoc and tenure track faculty employment. The text of the proclamation suggests that it is a one-time payment of $100,000 for a three-year duration of status.8 This means a de facto fee of $33,333 a year.

Academic salaries are notoriously low. The average US postdoc makes around $60,000 a year;9 the average new professor probably around $100,000.

Adding a $33,333 per year immigration fee will obviously decrease the number of foreign academics hired. This will decrease the competitiveness of US science and academia, as postdocs are a key part of the scientific ecosystem.

2. The private-sector H-1B lottery is about 4x over-subscribed (339,000 entrants for 85,000 slots in 2024). For overall immigration to stay at the same level, I don’t think that >25% of private sector H-1B applicants would still be economically viable with this large of a fee.

The H-1B visa generally requires that one pay $60,000 or the local prevailing wage, whichever is highest. With the fee, the employer would end up paying (at least) the local prevailing wage… plus $33,333 a year.

Certainly, some H-1B applicants probably are so good that they make $30k more than would be expected in their area. I doubt it’s 25%, though.

On an emotional rather than economic level, it also makes the US seem very hostile to immigrants. I recognize that this is why Trump did it, but also: it will make talented, high-earning immigrants with other options take those options more seriously.

If the UK did this, I would not want to stay in the UK long-term, even if my employer would pay the fee for me. I would simply not want to live in a country that would treat me like that.

I fear that many highly talented US immigrants may feel the same way. The US will be worse for losing them.

I joined Malengo for their annual retreat1 this year, and had a chance to speak to a number of young international migrants about their experience.

What is Malengo?

Malengo is an NGO facilitating international migration. It supports Ugandan and refugee students2 in obtaining a German university degree and then working in Germany.

When a student moves to Germany, their life prospects improve markedly. A Ugandan university graduate will be lucky to earn $5,000 a year; a German university graduate averages >$50,000 a year.

The mechanics of the Malengo program are as follows:

• Germany has many open-access universities, where if you have a high school diploma, you can enroll.

• German universities have no tuition fees (even for international students).

• However, most Ugandan students don’t have the money upfront to move to Germany. The average Ugandan makes about $1,000 a year; the flights to Germany alone would be cost-prohibitive.

• Malengo provides support in enrolling at a German university, flights to Germany, some language training,3 and a stipend for the first year’s living expenses.

• After the first year, students are expected to get a job and support themselves.

• After graduation, students repay Malengo a portion of their income. Specifically, once they earn >€27,000, they repay 14% of income (up to a maximum amount paid or maximum amount of time).4

This structure is not dissimilar to how student loans work in England, where high-earning students cross-subsidize the lower-earning students. However, it fills an important market niche for international student migration.

Scholarships and loans are rather thin on the ground for the merely good, rather than the exceptional. If you are Malala Yousafzai, or indeed any young person who is admitted to Harvard for university, there are probably financial options available for you to attend.5 Malengo does not target these students.

Rather, it focuses on students that come from humble backgrounds that otherwise would not have access to university abroad. They come from families making, on average, $42 per person per month. For a family of six - not uncommon in Uganda6 - this would mean a household income of about $3,200 per year. Self-funding a degree abroad is simply not an option when your household makes $3,200 a year. Indeed, even attending university within Uganda is often financially difficult.

Most Malengo students will probably not become politicians, global leaders, or CEOs after they graduate;7 rather, they will go to solid white-collar jobs that make a good - but not exceptional - living. These are also the kind of jobs Germany is most likely to need to fill in the coming decades.

Like many European countries, Germany is aging and is likely to need many skilled migrants in the coming decades. There are only so many Malalas in the world, but BMW is going to continue to need quite a lot of engineers. Malengo students may fill some of these gaps - working good professional jobs, paying taxes, contributing to the German economy.

The program is intended to support international migration in a scalable - and hopefully not unpopular!8 - way.

How It’s Going

So far, the descriptive statistics look like the model is working. Of the ~250 Malengo scholars currently abroad, almost all are making progress towards their degree. 85% are actively employed, with average earnings of €880/month. This is already many times more than they would earn in Uganda.

So far, their grades are good but not amazing. Germany uses a 1-5 grading scale, where 1 is best. The average GPA is currently 2.37. A reasonable number will take more than the nominal time to graduate, because they switched majors or failed exams,9 but they are nonetheless on track to graduate eventually. Only two - of the 250 currently in Germany - have fully dropped out.10 95% want to stay in Germany after graduation.

Of course, the ultimate goal is not how well they do in school, but how they do afterwards. Malengo has embedded an RCT into the program to compare the life outcomes of those who attend university in Germany to similar students who did not go to Germany.

The first cohort of Malengo scholars started in 2021. This means the first cohort is approaching graduation.11 We don’t yet know what kind of jobs they will get, and how successful Malengo will be - but the students I talked to made me feel hopeful.

Malengo Student Interviews

Here are just a few of the students I spoke to over the weekend:

Jasmine, 21, Uganda

Studying international business at Hochschule Anhalt in Bernburg

Starting her second year in Germany

You know when you meet someone and you can just tell they’re a leader? That’s how I felt meeting Jasmine. In her first year in Germany, she’s already started working in her university’s international students office, run for student council, and is the president of the Erasmus club.

She’s the daughter of a pastor and grew up in Entebbe, Uganda, on the shores of Lake Victoria. She has four siblings, so money was always tight growing up. She thinks she probably could have attended university in Uganda, but she wouldn’t have had nearly the opportunities she’s had in Germany.

To her, living in Germany is “a fairytale”. She especially loves the nature in Germany - how there’s a park in every city, and how you can just spend your evenings in the forest.12 She loves her university, where she says she now works with people from countries “she didn’t even know existed a year ago”.

It hasn’t always been easy - she misses her family and Ugandan food - but since moving to Germany, she’s found a community in both her university and her church. She now feels like she can aim for the skies. She says that the Malengo program has taught her that “I’m actually stronger than I thought I am”, and that you “can get through anything as long as you take it one step at a time”.

After graduation, she hopes to work for the UN or an NGO, supporting young people. For now, though, she’s trying to take advantage of every opportunity she gets.

Peter Ngoth Anyang, 23, South Sudan

Studying applied biology at Hochschule Bonn-Rhein-Sieg in Rheinbach, Germany

Starting his second year in Germany

Ngoth is originally from South Sudan, but South Sudan has had significant conflict much of its (short) existence. In the 14 years South Sudan has been a country, some one million South Sudanese have fled to Uganda. Ngoth’s family were among them.

He spent much of his adolescence in Uganda’s Kiryandongo Refugee Settlement. At the time he moved to Germany, he had been in Uganda for 10 years. He finished high school in 2022, but he didn’t have money to pursue a degree in Uganda. Most people pay for their tuition upfront, but Ngoth, being a refugee, didn’t have funds for this.

Without Malengo, he likely wouldn’t have been able to attend university at all, let alone study in Europe. He told me that “you can’t picture someone like me being [able to study] in Germany.” He is incredibly cognizant of how lucky he is - he told me that “there might be people better than me but they don’t have this chance. I’m here representing the whole community of people who have dreams.”

He hopes to become a genetic engineer. When he graduates, he hopes to spend some time in both South Sudan and Germany. South Sudan’s science community is quite small, and he thinks he could do a lot for his country with the quality of training he has received in Germany.

Joseph, 28, Uganda

Studying international taxation and law at Rhine-Waal University of Applied Sciences in Kleve, living in Düsseldorf

Starting his fourth year in Germany

Joseph has been supporting himself since he was 15, after his mother passed away. With some support from his grandmother, a friend and a part-time job, he was able to scrape together enough to attend university in Uganda.

He was in his third year of law school13 when the COVID-19 pandemic hit. He didn’t have enough money to continue, and so he dropped out, left the city where he was studying, and took up farming. At the time, he was living with an uncle and his cousins. It wasn’t much of a living; when I asked him what he would be doing if he hadn’t found Malengo, he said “surviving”. He thinks it is unlikely he would have ever finished his degree. If he had miraculously managed to make it, it would have taken longer, and he doubts he would have managed to sit for the state bar exam due to the high costs involved.

In Germany, he is thriving, though. Like others, he mentioned that the transition wasn’t easy; he found the first year lonely, and his classes were harder and moved faster than they would in Uganda. The first winter was rough - he says, “I didn’t know a country could be that cold”. After three years, though, he’s adjusted - remarking philosophically “would we want more sun [in Düsseldorf]? Yes. Will we get it? Probably no.”

Indeed, he loves Germany; he says that he’s one of those immigrants who becomes a little too enthusiastic about their adopted home. He loves German food - he went on a small digression on “leberkäse, schnitzel, and the bread - Jesus” - but more than the material conditions, he loves the freedom. He says that “where I come from, everyone has opinions on how you should live your life; you can’t be anything you choose.” But in Germany, “you can do anything”.

He’s currently working at a global engineering firm in the tax, customs and foreign trade department, and is set to graduate in the next year. He’s still deciding what he wants to do next - either continue through a master’s degree (taught in German) before moving to full-time employment, or gain more work experience and continue with the master’s later.

He is definitely sure he wants to stay in Germany, though; indeed, he plans on taking the German equivalent of the bar exam once he’s eligible.14 This credential virtually guarantees he will stay in Germany, as it does not transfer elsewhere. Once he passes, he wants to start his own tax practice.

For more information on Malengo (or to donate to Malengo), see their website.

If these new immigrants are a significant fiscal drain, the UK probably does not want them to stay long-term. Thus, there have been moves to reform the indefinite leave to remain system, so that the UK is not stuck with a large non-working population.

I decided to check the data on how worried the UK should be about this. Is the UK actually the world’s welfare state, having admitted vast numbers of people who will take out more in benefits than they pay in taxes? Will the British-born be on the hook for the astronomical cost of mass migration?

I decided to look at the data. Handily for me, the Home Office has collected data about this specific wave of immigrants, including information on their wages and labor force participation.

tldr: I think these concerns are overblown. Even considering only new Boriswave immigrants, I calculate that the average new immigrant salary was still higher than the average British national’s. This is true even when one considers dependents and corrects for their likelihood to take paid work.

Considering that immigrants are also cheaper to the state than the native-born, and they outearn the native-born, migrants who arrived 2019-2023 will make a net contribution to British finances.

I do not think that the UK should restrict immigration for fiscal reasons. If the UK does choose to restrict immigration - as it seems inclined to do - it should be clear about why. It is not because immigrants have placed “public services and housing access… under too much pressure” and immigration threatens the welfare state. Rather, it is simply because voters don’t like it very much.

First, let’s start with basic statistics.

The UK Labor Force

Before discussing migrants, it is useful to calibrate expectations.1 If one is going to talk about the wages and labor force participation of a new group, one must also know what the labor market looked like before.

The median employed worker made about £613 a week, across full- and part-time workers.2 This is equivalent to an annual salary of £31,891.3

This data excludes the self-employed. I don’t have great data about what self-employed residents make, so I've assumed that they make similar amounts to those who are not-self-employed. This is almost certainly too generous, because 2023 data suggests that the self-employed generally make less than the non-self-employed, but it’s what I have, so I’m going with it.

Of course, not everyone is currently in work. According to ONS, about 75% of UK citizens aged 16-64 were employed in 2024. The remaining 25% might be unemployed, in school, providing care, or indeed anything else that takes you out of the labor force.

Considering this, I estimate the median working-age UK citizen made about £23,918 in 2024. This is probably somewhat of an overestimate, though.

What happened with migration from 2019 to 2024?

During this period net migration to the UK was very high by historical standards. In the mid 2000s, net migration hovered around 250,000; in 2023, it hit a peak of 906,000. That is a significant change; perhaps it is not surprising that there has been equally-significant backlash.

Between 2019 and 2023, new entrants to the UK4 were:

• ~11% people on skilled worker visas

• ~8.5% dependents of people on skilled worker visas

• ~15% people on health and care worker visas

• ~18% dependents of people on health and care worker visas

• ~12% family visas (e.g. the spouse of a UK citizen)

• ~7% other visas (business mobility, investor, global talent, high potential individual, etc)

• ~4% dependents on other visa types

• ~24% people on humanitarian visas (Ukraine, Hong Kong)

Below, I split this into three categories: main applicants, dependents, and people on humanitarian visas.

Main Applicants

I’ve included four types of visa in this category: skilled worker, health and care worker, family (usually spouses of British citizens), and other main applicants.

There is considerable variation in salary by visa type. The highest median incomes are for skilled workers and business mobility visas, at £56,600 and £47,900 respectively.5 The lowest is for those who are joining family, at £20,200.6

Some of this variation is structural; a skilled worker or health worker must have a job offer in hand in order to acquire a visa, but someone on a family visa does not.

Overall Wages

I calculate the average employed main applicant’s annualized salary at £37,168.7 Since not all main applicants are required to be in work, the average income is slightly lower, at £34,376.8

Still, this is higher than the median UK worker (£31,891), let alone the median person in the UK (when one corrects for labor force non-participation). Given that the British state doesn’t have to pay to educate migrants, this is clearly an excellent fiscal deal.

My calculation isn’t perfect; the Home Office provides data on only one kind of “other main applicant”, so I’ve applied that income mean to the whole category.

Still, even if we set the median wage for “other main applicants” to £0, the overall main applicant average would be above the average wage for the British-born, solely on the strength of skilled worker incomes. The median skilled worker makes more than twice what the median British citizen does.

Wait a second - a ton of these people don’t have any income.

According to the Home Office, ~20-30% of main applicants have no PAYE earnings.9 Are there actually vast numbers of people moving to the UK on skilled worker visas that then proceed not to work?

Probably not, no.

This is for two reasons:

First, and most simply, the Home Office’s matching process with PAYE data isn’t perfect. They even say that themselves; there are definitely people working (legally!) in the UK that the data merge just didn’t catch.10

However, I think the more significant reason is that many of the migrants who don’t have UK earnings have already left the UK.

Most types of main applicant require a job offer in order to move, so one cannot enter without a job. Quitting your job and living on welfare isn’t much of an option either, because 1) once you quit your job, you have 60 days to leave the country, 2) you and your family aren’t permitted to access most types of assistance.11

As anyone who lives in the UK will tell you, it is an expensive place to live. It is unlikely that many people have moved to the UK, with a job in hand, then quit that job, ignored the requirement to leave and decided to become an illegal immigrant instead.

Rather, many of the main applicants - on routes that require jobs - do not have PAYE earnings because they either have not yet entered the UK or have already left.

The latter is particularly likely because return migration - that is, leaving the country one immigrated to and returning to one’s native country - is quite common. Based on older UK data, 10% of immigrants leave within a year; 40% within five years.

Given this, it seems very likely that many of ~20-30% of main applicants that are required to have a PAYE job while in the UK that do not appear to have a PAYE job are simply… not in the UK anymore. Thus, our conclusion that migrants are very good for public finances is unchanged.

Let’s Talk About Dependents

OK, so main applicants earn reasonable wages. However, much of the public concern about immigration has not been about the main applicants, but their dependents.

Who are dependents?

To be a dependent, one must be the spouse, unmarried partner,12 or a minor child of the main applicant. Note that parents, cousins, siblings, adult children13 or other relatives cannot qualify for a dependent visa.

Over 2019-2023, there were 537,900 dependent visas issued to 776,900 main applicants. That is, the average migrant to the UK entering in 2023-2024 had 0.69 dependents, though, in some years, there were more dependents than main applicants.

Are dependents dependent?

Does this mean that - as Katie Lam MP would have - that millions came here without intent to work? Not really. The term dependent is a bit of a misnomer here. They are a dependent in terms of immigration status only.

In some countries (cough cough, the US), dependents are barred from work. This is not true in the UK. Indeed, in a lot of ways, your options for employment are better as a dependent than as the main applicant.

For instance, I am on a skilled worker visa. Any job I take must:

• Be for a UK employer that’s been approved by the Home Office

• Be for an employer that has a ‘certificate of sponsorship’ (CoS)14

• Be on the list of eligible occupations

• Be paid at least a minimum salary at the start of the job (in 2023, £26,200; in 2024, £38,700; in 2025, £41,700, or the “prevailing wage” in your field, whichever is higher)

My employer must also pay the certificate of sponsorship fee (£525) and the Immigration Skills Charge (£364 - £1,000 for each year they intend to employ me).15

By contrast, the requirements for my (nonexistent) spouse's job would be:

• It is a job in the UK.

Their employer also does not have to pay any fees to employ them. This leads to an interesting conclusion: if you are a couple where both people could plausibly get UK work visas, it is probably a better idea for one person to get a work visa and one to enter as a dependent. Being a dependent for the purposes of immigration does not mean that you are economically dependent.

Labor Force Participation Rate

The fact that dependents can work does not necessarily mean that they do, though. Every UK citizen is eligible to work; this does not mean that every UK citizen does. Indeed, a significant fraction of people are not in work at any given time.

To figure this out, I again use Home Office data. I calculate the labor force participation rate for each category from:

1. The percentage of dependents with PAYE earnings in that tax year (by visa category).

2. The percentage of main applicants with PAYE earnings in that tax year (in a visa category)

The latter adjustment is because it seems unlikely to me that the dependents of a worker will receive PAYE earnings if the main applicant (required to have a PAYE job) is not receiving PAYE earnings.16

As above, the majority of those on visas with no PAYE earnings have probably left the country. Their spouse and children are quite unlikely to have stayed in the UK if they have departed. Therefore, I adjust the labor force participation rate of dependents by the percentage of main applicants with PAYE earnings.

I calculate the labor force participation rate for dependents in this sheet. All of these are likely to be underestimates, as the Home Office data excludes self-employment,17 but it does appear to vary substantially by visa type. Only 37% of dependents of those on business mobility visas work, while 83% of dependents on health and care worker visas have some earnings.

That last point bears some emphasis, as there has been considerable discourse about the dependents of low-wage care workers being a drain on the British economy. And it is true that there were many such dependents; a full third of non-student visas issued in 2023-2024 went to dependents of someone on a health and care worker visa. But very few of these migrants seem to sit at home - indeed, the dependents of those on health and care worker visas are considerably more likely to be in work than British citizens.

Labor force participation seems to be quite sensitive to the earnings of the main applicant. Adult dependents are very likely to work in migrant households where the main applicant is not a particularly high earner (e.g. health and care worker visa holders), but are less likely to work the more money the main applicant makes. 83% of health and care worker dependents work, while only 60% of skilled worker dependents do. This seems reasonable; a health worker on £30,000 can’t afford a stay-at-home spouse, but a skilled worker on £56,600 might be able to.

Weighting by the number of dependents of each type, 70% of dependents that are likely to be in the country do have PAYE earnings. This is similar to the percentage of the overall population with PAYE earnings.18

Individual Wages

There is less variation in wages by type of visa among dependents than main applicants. This makes sense - as noted above, main applicants can have significant restrictions on where they can work and what types of jobs they can take, while dependents experience the general UK labor market.

Dependents of those on skilled worker visas are the highest earners, taking jobs paying £30,200 on average; dependents of those on health and care worker visas make the least, taking jobs that pay £22,100. As above, these are underestimates; this excludes any self-employment incomes, such as Deliveroo earnings.19

The average employed dependent makes £25,534; the average dependent (accounting for labor force non-participation) makes £17,898. This is lower than the average UK wages, by £6,020.

However, that’s not as concerning for public finances as it sounds for two reasons.

1. Our data excludes self-employment earnings. I’ve included an imputed self-employment earning for UK residents (indeed, I’ve probably overestimated self-employment earnings for UK residents), but not for migrants. Dependents are free to start businesses, and this could make up some of the gap.

2. Dependents - and other migrants - are quite a bit younger than the British-born. I couldn’t find data for 2023-2024, but in 2020, the median new migrant was some 14 years younger than the median British person. One would expect a younger group to have lower current earnings even if they have similar lifetime earnings.

Considering the average wage for a person in their 30s is some £7,600 more than someone in their 20s, recent immigrants actually seem like they’re on a perfectly reasonable earnings trajectory. In 14 years, it seems reasonable that they will close the £6,020 gap with the native-born.20

Household Wages

Of course, one does not have just main applicants or just dependents. People are part of families; you cannot simply say “I would like your high-earning spouse but the low earning one can stay in Pakistan”. (Or you can, but then you take the risk neither spouse will move to the United Kingdom.)

From our wage calculations, we can estimate the average household wages for skilled workers, health and care workers, and those on other visas.

Skilled worker households are relatively small, at 1.8 individuals. For every four households with a skilled worker main applicant, three will contain a dependent; one will not. The average household income is about £70,203. This puts the average skilled worker household at about the 70th percentile in the UK.

Despite the stereotype, the average health and care worker household isn’t that much larger, at 2.24 individuals. Assuming that households have two adults and 0.24 children, they have an average household income of around £49,180. This is very similar to the UK average - Britain-wide, the average household contains 2.36 individuals and has an income of £53,140.

Other work visa holder households are smallest, at 1.6 individuals. As noted above, I’m making a bit of a heroic assumption, applying business mobility visa wages to all other work visas - but if this holds, these households have an income of £54,213. This is nearly identical to the average British household.

Humanitarian Visas

All of the above ignores the largest single visa category for 2019-2023: humanitarian visas. I always feel a bit weird writing about the fiscal dynamics for humanitarian visas, because that is not why countries take humanitarian migrants.

It is true that the median Ukrainian migrant doesn’t make very much money. The government itself estimates the median Ukrainian only makes £16,000 per year, less than even the much-complained-about dependents of care workers.

This is one of the lowest wage categories; only the dependents on other visas have lower expected wages, and that number is a little bit fictional. (I only had data for the dependents of those on business mobility visas; I’ve assumed the dependents on other types of more obscure work visas had similar labor force participation, but that’s… not terribly well justified.)

The good news for British public finances is: if one wants to limit the number of new refugees entering the country, that problem has already been solved. Refugee numbers spiked in 2022 due to the war in Ukraine, but have declined quickly. The number of humanitarian visas issued in 2023 was just a quarter of the number issued in 2022. It continued to decline in 2024.

As long as Vladimir Putin manages to avoid invading any other countries, the 2022 spike in humanitarian visas is unlikely to be repeated.

Summary

• The average main applicant who received a visa 2019-2023 made about 45% more than the average person in the UK.

• The average dependent who received a visa 2019-2023 made about 25% less than the average person in the UK.

• The average person who received a humanitarian visa made about 33% less than the average person in the UK.

When we take a weighted average of all categories of 2019-2023, the average new migrant who arrived 2019-2023 made £24,881 in FY 2023-2024. This is about 4% more than average earnings - even when I have made optimistic assumptions about native-born earnings and excluded any self-employment income for migrants.

It is difficult to see how admitting people who outearn the native-born would be catastrophic for public finances, particularly given that the British state need not pay to educate them. The data certainly does not seem to suggest this cohort is a “ticking time bomb” for British public finances.

This is relatively high for a rich country, and much US political discussion is devoted to the issue of undocumented migrants. Therefore, it seems worth understanding who undocumented migrants are, and how they fit into both US society and the US economy.

Two notes to start,1 though:

A few words on what this piece is not: it is not written in response to any current events, nor can it respond to them. The piece itself is being written in mid-April 2025; it cannot and will not respond to any events that may occur in the future or are happening now.

This is designed as an overview of what the academic literature says about the US’ undocumented population at the time the constituent papers were written. It is descriptive, rather than prescriptive. It does not attempt to recommend policies, because that is not what a literature review is for.

In other parts of this Substack, I make arguments based on personal opinion; for the migration living literature review, I try to stick closely to only what the evidence says. I show that undocumented migrants in the US are largely law-abiding people who participate in the labor market, because that is what the evidence shows; it is not an attempt to make a political point.

A note on data: somewhat by definition, the empirical evidence on the undocumented population is weaker than on documented residents of a country. There is limited information available on a population that has limited interaction with the legal institutions of a country.

For instance, the method for determining how many undocumented people are in the US is fairly rudimentary. It simply involves taking the number of temporary legal immigrants2 and applying mortality table data to the number of green cards issued, to determine how many documented permanent residents there “should” be in the country. One then sums the number of permanent residents and temporary immigrants, and subtracts the sum off the total population to find the number of undocumented people.

There are several ways this could be wrong, though:

1. Our population estimates might be wrong. Undocumented people typically try to avoid interacting with the state; this is likely to include census takers. The Census makes some assumptions on what percentage of the undocumented population they are able to reach, but they could be wrong.

2. Mortality life tables might not accurately reflect the survival rates of green card holders. If immigrants live longer than expected, one might see “too many” people in the population data and assume that they’re undocumented - but actually, they are green card holders who are unexpectedly healthy.

The difficulty in accurately counting this population means that even descriptive statistics have significant error bars. For instance, the Pew Center is one of the major organizations that attempts to measure the undocumented population, and several economics papers (e.g. Borjas 2016, Borjas 2017) build on their work. But Bhandari, Feigenberg, Lubotsky and Medina-Cortina 2021 believes the Pew Center underestimates the population of undocumented migrants from Mexico by 35%; another paper argues that Pew underestimates the undocumented population by 50%.

Given the uncertainty of even basic descriptive statistics, then, this post relies less on causal inference than most of my migration literature review. This post is less about establishing the effect of changes in policy on undocumented people and more about trying to establish who undocumented people in the United States even are.

Descriptive Statistics

With that out of the way, who are undocumented immigrants in the US? Or at least, who are they (to the best of our knowledge)?

Demographic Characteristics

Borjas 2016 works with a (de-identified) version of the Current Population Survey to find that (as of the mid-2010s), undocumented migrants are:

• Disproportionately male (54%)

• Younger (on average) than either the native-born or legal immigrants.

• Have much lower education levels than either the native-born or legal immigrants. Notably, nearly half of undocumented migrants (42%) do not have a high school diploma, while only 7% of the native-born do not have a high school diploma.

• … and are concentrated in states near the border3

Country Of Origin

About 70% of undocumented migrants in the US come from Latin America, with about a third to half coming from Mexico and another ~20% from the Northern Triangle of El Salvador, Honduras and Guatemala.

However, the Discourse often neglects that perhaps one in three undocumented migrants in the US is not from Latin America. About 6% of undocumented migrants come from India; another 6% of undocumented migrants come from Canada and Europe.

Tenure

Most have been in the US a while. Only about 20% have been in the US less than five years; 40% have been in the US more than 15 years.

Some of this may (ironically) be due to increased enforcement over the last several decades. Many undocumented migrants would like to be circular migrants - coming to the US, earning some money, going back to their home country, and repeating the cycle. But as the border becomes riskier and crossing becomes more difficult, it becomes more dangerous to go back and forth - migrants are more likely to pick one side and stay there.4

Furthermore, if a migrant now does not think trying the border again is a good idea, they are more likely to bring their family with them. For a six month stint away, it is likely one would leave a wife and children at home; for a ten year stint abroad, one is more likely to try to bring them along.

Given the long tenure of migrants, it is also very common for households to be mixed status. Around 70% of undocumented migrants live in a household where at least one person is either a citizen or documented migrant.

The most common situation appears to be a US citizen child living with undocumented parent(s); there are about 4.4 million such children. This means about 6% of US children live with an undocumented parent.

Population Size

As noted above, there is some disagreement on how many undocumented people live in the United States. The most commonly cited number is around 11 million, though other sources find up to 16 million.

Most sources do agree that the undocumented population likely peaked around 2007, and then decreased by several hundred thousand per year through 2021 or so. This is likely for two reasons:

• Employment in construction - the second-largest sector for undocumented immigrants in the US - dipped from 2006 to 2022.

• Demographic shifts - and lower fertility - in Mexico and central America have meant there are simply fewer young people who want to move (Hanson, Liu and McIntosh 2017).

In recent years, it looks like undocumented migration is on the upswing again, though the total is still significantly below its 2007 high.

Employment

Undocumented men are much more likely to work than documented men. Borjas 2016 finds that 87% of undocumented men are employed, compared to 78% of legal immigrants and 73% of native-born men.

Undocumented women, however, are much less likely to work than documented women. Only 55% of undocumented women are employed, compared to 59% of legal immigrants and 67% of native-born women.

Doing the weighted average,5 this means undocumented migrants are, overall, more likely to be employed than either legal immigrants or the native-born. Nor does this seem to have changed since 2014 (the last year in the Borjas sample); in 2022, Pew estimated that 75% of undocumented migrants were working, while only about 60% of the overall US population is currently working.

The high labor force participation of undocumented migrants makes sense for several reasons:

• Legal immigrants and the native-born have access to social safety net programs that undocumented people do not. For instance, undocumented people cannot access unemployment or disability benefits. Without such benefits, work is likely to be not very optional.

• Furthermore, undocumented immigrants are in a considerably worse financial state than documented immigrants or the native-born and are thus unlikely to be able to survive on savings. In 2008, the average undocumented Latino immigrant had a net worth of $38,000; the average documented Latino immigrant had a net worth of $66,000. The gaps for other racial groups are similarly large (see table 2).

• Undocumented migrants come to the US largely for economic opportunity, balancing the higher wage-earning potential with the legal risks. If one is not currently pursuing that economic opportunity, being in legal jeopardy is less worthwhile.

However, this high labor force participation is still worth noting because of the relatively low levels of education in the group. Among legal immigrants and the native-born, employment rates are lowest for those with the least education. An undocumented man is 17 percentage points more likely to be employed than a native-born person, controlling for education, age, and socioeconomic status.6

I’ll discuss aggregate effects in more detail in the fiscal effects section, but this high labor force participation rate is a significant reason that undocumented migrants are a net fiscal positive for the US.

Wages

Borjas 2017 uses Current Population Survey data to look at the wages of undocumented migrants. Perhaps unsurprisingly, they are low; at the age of 25, perhaps 18% below documented immigrants and 26% below the native-born.

More surprisingly, they are completely flat over time. The average documented immigrant makes considerably more at 45 than at 25; the average undocumented migrant does not.

This also means the hourly wage penalty for being undocumented rises over time; by age 45, the average documented immigrant makes about double what the average undocumented migrant does. Most of this wage penalty does not appear to be due to status; documented immigrants are a different population with different (and higher) expected earnings, largely due to higher education levels among the documented population.

But there is a significant wage penalty for status alone - around 10%. Put another way, if an undocumented worker became documented tomorrow, they could probably get a job paying about 10% more.

Interestingly, this wage penalty is much lower than it used to be; Borjas estimates that in 2007, the wage penalty for being undocumented was about double what it was in 2014. He does not really have an answer for why this is, though.

Fiscal Effects

Undocumented immigrants are sometimes considered to be a drain on the US economy. This appears to be exactly the opposite of what actually happens - since undocumented immigrants often pay taxes, but are entitled to few benefits.

How do undocumented migrants pay taxes?

In order to work in most US jobs,7 you have to fill out a W-9 where you list a Social Security Number (SSN) or an Individual Taxpayer Identification Number (ITIN).

Acquiring a SSN requires legal status, so some undocumented people list a fake SSN or list a SSN belonging to someone else on their W-9. Furthermore, you don’t need legal status to acquire an ITIN. Historically, the IRS does not care about your documentation status;8 even as an undocumented person, you very much still have to pay federal income tax.

Not all undocumented people work a job that does federal tax withholding; some work solely cash-in-hand. However, even they cannot escape the long arm of taxation, as consumption taxes are levied at the point of consumption. No one checks your immigration status when you buy clothes, food or other goods; sales tax is always levied.

What do undocumented people contribute to the economy (besides taxes)?

Per Edwards and Ortega 2017, undocumented people contribute about 3% to the GDP of the US.9 This is somewhat less than their percentage of the workforce, as they are paid less than the average worker.

Still, this is a substantial amount; undocumented people contribute (roughly) similar amounts to the economy as the defense industry.

The Institute for Taxation and Economic Policy estimates that undocumented people pay about $60B in federal taxes each year, and about $30B in state and local taxes each year.

What is the overall impact of undocumented migrants on the US’ fiscal policy?

Undocumented migrants clearly pay some taxes in the US. But do they pay more in taxes than they receive in benefits? There have been a number of attempts to figure this out.

I’ll start with the most negative findings:

1. The Heritage Foundation finds that the average undocumented migrant household is a net tax recipient, because they receive more benefits than they pay in taxes. This costs the state around $15,000 per year per household.10

2. In 2006, the Texas Office of the Comptroller calculated that undocumented migrants generated $424M for the state but cost local governments $1.44B.11

   However, most other estimates are positive.

3. Hanson 2009 estimates that undocumented migrants raise native welfare by 0.03 percent of GDP. This is basically indistinguishable from zero, as it would be about $26/person/year.12

4. Alex Nowrasteh, Sarah Eckhardt, and Michael Howard have modeled the economic contributions of immigrants by education level.

If we use Borjas 2016’s decomposition of the educational makeup of undocumented migrants, 42% of undocumented migrants have less than a high school diploma (Nowrasteh et al NPV -$85,000), 29% have a high school diploma only (NPV $35,000), 13% have some college (NPV $141,000) and 16% have a college degree (NPV $375,000).

If an undocumented migrant had average earnings for someone of their education level, their NPV would be +$52,780, and thus, they would be a net contributor to the US.

However, we know that undocumented migrants make less than the average immigrant, so this is probably somewhat an overestimate.13 Given the legal-undocumented wage gap is about 10%, I’d adjust all contributions downwards by 10%. In this case, you’d end up with a still positive but smaller net contribution of about $40,000 per undocumented immigrant (in NPV).14

5. Chassamboulli and Liu 2024 uses a calibrated model to find undocumented migrants increase native welfare, but there’s not an obvious way to convert this into a dollar value.

Notably, this result does not come directly through their fiscal contributions in the form of taxes, but through increased firm profits and an overall increase in jobs.

It’s worth taking a second to dwell on this point. Undocumented migrants do pay taxes and consume benefits, as do all members of a society. But looking at only this may underestimate the impact undocumented migration has on an economy, because immigrants also consume goods and services. They participate in the economy, and it adjusts around them. Other people - who may not be undocumented - are employed in order to sell them things.

A strict Heritage-style “here are taxes paid and benefits received” accounting may underestimate what undocumented people bring into the US economy. This is discussed in more detail in Clemens 2022, but suffice to say that he finds that adding an immigrant with less than a high school diploma - the lowest-earning education group and thus, the group that pays the smallest amount of taxes - has significant and positive NPV.15

Deportations

Another way one can examine the economic impacts of undocumented migrants is to look at what happens when they aren’t there - or rather, are forcibly removed.

In 2008, a new data sharing program between local law enforcement and Immigration and Customs Enforcement (ICE) was introduced. This program, called Secure Communities, increased the likelihood that undocumented migrants would be detained and deported.16

East, Hines, Luck, Mansour and Velasquez 2022 and Velasquez and East 2024 examine the effects of this program (and its quasi-random rollout). They find that undocumented immigrants were less likely to work - but also that wages and employment for US citizens declined in response to deportations. For every 1% that employment of likely-undocumented people declined, wages for the US-born declined 0.1%.17

There are several reasons this could be true:

• As firms hire more documented workers, this increases their labor costs (since as previously noted, undocumented workers are cheaper). This in turn reduces their ability to create new jobs.

• Undocumented people consume goods and services, and if there are fewer of them, there is less employment among US citizens providing said goods. And indeed, East et al find that employment drops particularly in local consumption.

Velasquez and East 2024 adds another explanation: employing undocumented women - in childcare - allows US citizen women to work. The implementation of Secure Communities reduced the amount of childcare available, and thus, particularly affected the wages of women with young children.

Thus, undocumented people seem to fill an important role in the labor market. Combining these papers with the macro evidence above, I would say it is more likely than not that the presence of undocumented people increases the income of the US-born.

Crime

Money is, of course, not everything; as noted in my crime and immigration post, many people worry that undocumented immigrants are more likely than the native-born to commit crimes.

The evidence suggests they are not, though. There is surprisingly strong evidence for this: this is a consistent result across a variety of contexts. Undocumented migrants are less likely to be arrested and incarcerated in Texas and Georgia; another study in Texas finds the same results. There is no association between the size of the undocumented population and overall crime rates.

When cities adopt sanctuary policies (such that they may attract more undocumented immigrants), they do not experience an increase in crime rates (and may even experience a decrease). Nor does increasing deportations decrease crime - not in North Carolina, not in Arizona. There is simply no evidence that “[undocumented migrants are] bringing drugs. They’re bringing crime. They’re rapists” - and reasonably strong evidence that really quite a lot are “good people, I assume”.18

Alex Nowrasteh has written a longer post going through possible reasons why undocumented people are less likely to commit crime than the native-born, but I think the most likely explanation is simply that the punishments are worse for an undocumented person. If a native-born person commits a crime, they face possible fines and imprisonment. If an immigrant commits a crime, they face possible fines, imprisonment, and deportation. Undocumented migrants, in particular, have often faced considerable risk to move to the country they are now living in; it seems very rational to avoid anything that might put that in jeopardy.

It is undoubtedly true that there are individual undocumented people that commit crimes, just as there are individual citizens that commit crimes. As a group, though, undocumented people in the US seem to be pretty law-abiding once they arrive in the US.

Legalization

There is one last topic that is of relevance to the economics of undocumented migration. What if undocumented migrants are given legal status?

The US has done this via several programs. In 1986, the Immigration Reform and Control Act (IRCA) granted legal status to 1.7 million undocumented migrants; more recently, Deferred Action for Childhood Arrivals allowed some undocumented migrants who arrived in the US as children to gain work authorization.

Not surprisingly, having the ability to work legally increases one’s likelihood of having a job at all. It also increases wages among those who are working - consistent with the wage evidence previously discussed.

Legalization has somewhat more contradictory results in terms of educational attainment. Legalization does make youth more likely to finish high school, but it’s unclear what effects it has on college attendance. It’s possible that DACA recipients are more likely to attend college - but it also might make university less attractive to newly legalized immigrants.19

Even if it does, though, I don’t think this should be an argument against legalization. One framing of this might be “undocumented kids drop out of college when they get status”; another, equally-valid framing would be that “when undocumented people are given the opportunity to earn more money, become productive members of society20 and pay more in taxes, they take it”.

Indeed, this appears to be the through line for what we know about undocumented people in the US. Undocumented people want to work, and indeed they do. They work, they pay taxes, and (largely) abide by the laws of the country in which they live. The US benefits from their presence.

Many thanks to Akshay Narayanan and Ruth Grace Wong for their edits. As always, all mistakes and typos are mine.

This was not entirely the post I was planning on writing this week,1 but: H-1B visas have recently been In The News.

The incoming Trump administration seems to be signaling that it is at least somewhat sympathetic to the “tech right” with the appointment of Sriram Krishnan to an advisory position on AI. Far-right activists were not thrilled with this choice, as Krishnan has advocated to make it easier for foreign workers to come to the US, particularly if they work in tech.

Thus sparked a “civil war” among Trump supporters. I think it would be more accurate to describe this as a large number of people shouting at each other on the platform formerly known as Twitter.com, but nonetheless: there has been extensive discussion of skilled migration generally and H-1B visas in particular.

Elon Musk is vocally in favor of expanding the program. Steve Bannon is opposed. Trump says he’s a big believer in H-1B visas, though he moved to restrict them during his first term. Bernie Sanders has also joined the fray, saying Musk just wants cheap immigrant labor.

So it seemed worthwhile spending some time summarizing what we do - and do not - know about H-1B visas and their effects on the American labor market.

What is an H-1B visa?

The short version

1. A H-1B visa is the US’ primary “professional worker” visa.2 It is issued to people with at least a college degree3 in a specialty occupation, for a specific job; the visa requires a job offer in hand. H-1B visas are relatively common, compared to other types of work-based visas; if someone is working on a visa in the US, it is likely an H-1B.

2. However, H-1B visa holders are relatively rare in the US population as a whole. There are about 600,000 workers currently on a H-1B (or at least, there were in 2019). Since there are 167 million members of the US labor force, this means about 0.4% of the US workforce is on an H-1B visa. Most people in the US probably don’t know anyone on an H-1B,4 and most firms don’t employ any workers on one.5

3. Jobs that offer an H-1B visa must pay at least $60,000, or the prevailing wage for that job in their area, whichever is higher.6 This means that by construction, a person on a H-1B visa makes more than the average American.7

4. H-1B visas are not permanent. They expire, and no one goes directly from an H-1B visa to US citizenship. If you don’t have sponsorship for green card status (not part of H-1B status), you cannot stay in the US indefinitely on a H-1B visa.8

Wait, I thought we were talking about people who move to the US to stay here.

Well, sort of. Like most skilled worker programs, a H-1B visa is issued for a particular job opening. It is considered to be a non-immigrant visa, because you can’t stay or renew indefinitely.9

If your maximum six year period of H-1B status is going to expire, you can either 1) depart the US for one year, and then reapply for a new H-1B, or 2) one’s employer can petition for an individual to be classified as an immigrant that would allow the individual, when a green card number becomes available, to file for an adjustment of status that grants permanent residency.10 (Note that they would, at that point, no longer have or need H-1B status, as they would be a permanent resident.)

The likelihood of option 1 or option 2 varies significantly by your employer. Some firms eventually apply for employment-based green cards for nearly all their H-1B visa holders (for instance, Qualcomm and Microsoft), while others send nearly all of their workers home after the visa is up (for instance, Tata Consultancy Services and Wipro).11

Indeed, this is one of the complications of looking at the H-1B visa program as a whole; different actors and firms use the H-1B programs in different ways. Some firms treat it as a way to bring in highly skilled immigrants with skills they could not find elsewhere; others treat it as a guest worker visa. This is one of the reasons evaluating the whole H-1B program is difficult; it is true that in some cases, it is (as per Elon Musk) to bring in top engineering talent; others (as per Bernie Sanders) really do use it as an offshoring method. There is no one, single way in which H-1B recipients fit into the US economy.

OK, but I’ve heard about a H-1B lottery. Why do people talk about a visa lottery if this is for a specific job opening?

If you are one of the Americans who knows anyone on an H-1B, you will have heard people discuss the H-1B lottery. And, yes, H-1B visas are determined partially by lottery.

Like most US visa types, there is a limited number of H-1B visas that can be issued in a single year. Since 2004, there have been 65,000 H-1B visas issued to any skilled worker, with an additional 20,000 reserved for workers with U.S. advanced degrees, an additional 5,400 reserved for Singaporean nationals, and an additional 1,400 reserved for Chilean nationals.12 (If, by this point, you are thinking that US immigration law is needlessly complicated: congratulations! You are correct!)

However, there are far more than 91,800 foreign nationals who receive job offers that are conditional on receipt of a H-1B visa per year. Indeed, there are about 450,000 such people every year. Thus, there is a lottery to assign these 91,800 visas to 450,000 people. Or rather, several lotteries, as there is both a “regular” lottery and an “U.S. advanced degree” lottery. Because who ends up being selected is random, the lottery serves to assign H-1Bs to some firms and not others by pure chance. Several of the papers I discuss below will use this to determine what the effect of having an additional H-1B worker is.

There is one last caveat to how many H-1B visas are issued: some categories of employers – nonprofit research organizations, universities - are not subject to the cap. This means despite the H-1B “cap” being 91,800 visas/year, 386,000 H-1B visas were issued in FY 2023.

OK, enough background. What are the impacts of H-1B labor on the American economy?

So: who’s more right? Is Elon Musk right, and H-1B workers provide useful skills that aren’t available in the American labor market? Are they advancing the American economy? Or is Bernie Sanders right, and firms use H-1B workers as cheap foreign labor?13

The Labor Market

Many Americans want to know if H-1B visa holders are “taking American jobs”. I want to be a bit more precise than this. I want to know if H-1B visa holders complement American workers (perhaps by providing useful skills they do not have, allowing for more output or increasing overall output) or substitutes (thus, decreasing American employment or wages). Most likely, they are some combination of the two - but how much of each?

Native-Born Employment And Wages At Firms That Hire H-1Bs

Positive Supply Shocks

We begin by trying to answer if firms that successfully employed H-1B labor employed fewer natives as a result. The below papers use getting an additional H-1B visa (or many H-1B visas) to examine this question.

Doran, Gelber and Isen 2022 and Bound, Braga, Golden and Khanna 2015 both say that yes, native employment decreases as a result of H-1B labor.

Doran, Gelber and Isen 2022 finds that an additional H-1B visa crowds out 1.5 other workers at the firm and Bound, Braga, Golden and Khanna 2015 finds that the H-1B program reduced native-born employment in computer science in 2004 by 7-13.6% and wages for native-born computer scientists by 2.6%-5.1% relative to a counterfactual where immigration stayed at 1994 rates.

However, Mahajan, Morales, Shih and Chen 2024 finds the opposite - that firms that hire an H-1B worker end up hiring another 0.83 additional employees. These employees tend to be non-college-educated workers, perhaps providing complementary skills to those of H-1B visa holders (who are, by definition, college educated).

Of these three papers, I like the design of Doran, Gelber and Isen 2022 the best, because it uses the universe of firms with H-1B applications (albeit only in two years). However, it has possible external validity problems, as the identification strategy uses variation between outcomes of firms that applied the day the cap was reached. These firms differed somewhat from the universe of firms, which might mean that their results don’t apply broadly.

Still, I have more issues with the other two papers; Mahajan, Morales, Shih and Chen 2024 looks at only a single year and a subset of firms, and does not have additional data to determine how much the included firms might differ from firms not included in their data. Bound, Braga, Golden and Khanna 2015 uses a model rather than using the natural experiment.

Negative Supply Shocks (or “Great! Let’s fire all the H-1Bs and get back those jobs!”)

OK, so if a marginal H-1B visa recipient leads to 1.5 fewer Americans being hired by a firm, perhaps we should have fewer H-1B visas

The US government has, in fact, already sort of tried this by capping the number of H-1B visas. For the first few years of the current iteration of the H-1B visa program, starting in FY1992, the cap was not reached but since FY1997 H-1B numbers have been used before the end of the allotted fiscal year except for three years when Congress temporarily raised the cap (FY2001-2003). However, in 2007 when filings were made for FY2008, the cap was met before the start of the fiscal year and the government introduced the visa lottery.

Since a lottery is, by definition, random, some areas received fewer denied H-1B visas and some regions received more. Did the regions where more H-1B visas were denied hire more computer workers?

According to Peri, Shih and Sparber 2015, no, they did not. In fact, native-born computer based employment may have even fallen when firms had more H-1B visas denied, suggesting complementarities between native-born and immigrant computer workers.

So: hiring more immigrants - possibly bad for native-born employment. Not hiring immigrants? Also possibly bad for native-born employment. Unsatisfying as this conclusion is: the evidence base here is just kind of messy; I don’t really know that I can say for certain how firms respond to getting or being denied H-1B visas.

Native-Born Employment In The Economy As A Whole

There are also complex interactions between firms within an economy; a decrease in employment at a particular firm doesn’t necessarily mean a decrease in employment across the economy, if there are new needs as a result of either migration or productivity.

As an example, let’s say that H-1B workers are much more productive than Americans at designing some new invention. The firm that makes invention X is going to probably hire more H-1B workers and fewer Americans. But if invention X needs input Y, you might also see an expansion in firms that produce input Y - and some of the displaced Americans might work there.

There is also a simpler example of how the economy might adjust to more H-1B workers even if native-born employees are displaced from a firm. H-1B workers are, in fact, people; they will need to consume goods and services. They will need accountants and barbers and childcare and lawyers. Indeed, since H-1B visa holders are relatively high-income in comparison to the average American (see the fiscal impact section below), they probably consume more goods and services than the average American.14

However, this is, unfortunately, a really hard question to study, though, and the evidence is very limited.

Peri, Shih, and Sparber 2015 (a different paper, same authors) finds that in the long-run, an increased number of H-1B visa holders in an area leads to higher wages for college graduates.

However, this is a shift-share paper, and I’ve previously discussed why I don’t love shift-share designs in migration. Still: it’s the evidence we have, and it is weak evidence that the US economy does adjust over time - and that H-1B visas are good for the US.

Welfare and Output

Furthermore, productive firms are good for the US economy. Bound, Braga, Golden and Khanna 2015 estimates that increased immigration of IT workers in the 1990s resulted in the prices of IT goods and services being 1.9% lower than they otherwise would have been.

This may sound small, but IT is a significant portion of the US economy. In 2024, the US government alone spent about $100B on IT investments, and thus, immigration in the 1990s has saved US taxpayers perhaps $2B a year.

Furthermore, it also increased output in information technology by 2.5% relative to a world where immigration did not expand in the 1990s. There are times I am not sure more information technology is good - generally, whenever I am using a Microsoft product - but overall, an increase in information technology seems probably beneficial.

These immigrants produced new things, and that made goods and services more common and also cheaper. Seems good.

Innovation

The H-1B visa is often pitched as a way to increase innovation, as it is used for skilled workers, and is used particularly extensively in tech. Elon Musk says he wants to use H-1Bs to hire the top 0.1% of engineering talent with H-1Bs - that should increase innovation. But does it really?

The empirical evidence suggests that, yes, H-1B visas are good for American innovation.15 Hunt 2011 finds that immigrants who enter on skilled work visas are much more likely to produce patents and scientific papers than the native-born. This seems like strong evidence that H-1B visas do increase innovation; if you add more of a population that overproduces innovation, you end up with a higher rate of innovation.

Furthermore, Chen, Hshieh and Zhang 2021 and Dimmock, Huang, and Weisbenner 2019 find that winning the H-1B lottery significantly increases patenting rates and financial performance of startups.16

However, this conclusion is somewhat tempered by Doran, Gelber and Isen 2022. This paper looks at the universe of firms that use H-1B labor and finds that firms that receive an additional H-1B worker do not increase patenting rates (though this is not a tight null). However, given the lack of statistical significance of their patenting results, I don’t want to lean on this result that hard - and thus, still believe that H-1B visas increase innovation in the US.

Fiscal Impact

Another common worry about migrants is their impact on public finances. Do immigrants use more in public resources than they pay in taxes?

H-1B visa recipients are perhaps a best case for those worried about the effects of migration on the public purse. H-1B holders are relatively young (median age of 33), high earning (average income $108,000 in 2021) and have relatively few dependents (there are far fewer dependents of people on H-1B visas than H-1B visa recipients). All of these traits mean they are likely to pay more in taxes than they receive in benefits.

Furthermore, because it is a non-immigrant visa, and requires employment, H-1B recipients cannot be unemployed and in the country for more than 60 days. Furthermore, the minimum salary threshold of $60,00017 means that H-1B visa holders and their families18 would be unlikely to qualify for much income-based government assistance.

I have been unable to find a calculation of net fiscal impact of H-1B visa holders in particular, but the net fiscal impact of immigrants with a bachelor’s degree or higher who arrive between ages 25 and 44 is highly positive. Such an immigrant is expected to pay about $500,000 more in taxes than they receive in benefits in net present value terms.19

High Skilled Immigration More Broadly

There is a further caveat to all discussion of H-1B visas. H-1B visas are a gateway for other types of skilled immigration. The other large category of employment-based skilled immigration, EB visas, largely do not go to people living in other countries at the time of application. They go to workers already in the US, largely on H-1B visas.

This means that innovation and economic growth in the US due to immigrants who have green cards are also, kind of, due to H-1B visas. Indeed, H-1B visas tend to be early career; other types of visa tend to be more common later in one’s career; H-1B visa holders turn into other types of skilled immigration.

I haven’t focused on the broader literature on the effects of skilled migration on the US here, since I wanted to address the literature on H-1Bs specifically, but skilled migrants (on all visa types) contribute to the US in many ways. Immigrants are more likely to be inventors than the native-born, and the patents they produce are more innovative. Immigrants are more likely to be entrepreneurs, and foreign masters students increase the number of US startups. 38% of US Nobel Prize winners are immigrants - even though only 15% of the US population are immigrants.20

I’ll discuss this literature more fully in a future post, but if we think that H-1Bs are a “gateway drug” to more skilled migrants, this also seems positive.

Takeaways

So: is Elon right, or is Bernie? Are H-1B visas good or bad for the American economy?

My read of the evidence is as follows:

• Do firms hire H-1B workers instead of Americans in the short term? I genuinely do not know; the state of the evidence is murkier than I would like.

• Does that mean that American workers suffer and have permanently lower wages? I don’t think so; one paper finds increased immigrant employment in STEM has increased American wages and employment. However, this conclusion is based on a single paper that I don’t love, and my level of certainty here is low. If another paper found otherwise, I’d be willing to believe that result.

• Did you write 3000 words telling me that you don’t know how H-1B visas have affected American job outcomes? Unfortunately yes.

• Do H-1B visas increase American innovation? Yes, this seems likely.

• Do H-1B recipients pay more in taxes than they receive in benefits? This I actually am sure about - yes.

• Is any of this evidence going to stop people arguing on the website formerly known as Twitter.com? I doubt it.

Many, many thanks to Amy Nice and Alex Nowrasteh for fact-checking this; any remaining mistakes are my own. Thank you to Nathan Barnard, Phoebe Arslanagić-Little, and Pradyumna Prasad for their comments; all remaining incoherence is due to who I am as a person.

This post continues the previous post (on international migration) with all of the job market papers on domestic migration. They are sorted by modern day/historical, country studied, and then alphabetized by author.

I’ve tried to include as many papers as I’ve found; if I’ve missed your paper; please email me at lagilbert@gmail.com!

Domestic Migration Job Market Papers

Here is a list of titles; abstracts and links are below.

Modern Day:

1. China

   1. Housing Price, Internal Migration, and Intergenerational Mobility

   2. Internal Migration Restrictions, Aggregate Productivity, and Spatial Growth

   3. Parental Rural-Urban Migration and Child Education

2. Ethiopia

   1. Internal Displacement and Conflict in Ethiopia

3. Ghana

   1. Migration, Networks, and Religious Choice

4. India

   1. Human Capital of Children from Short-Term Migrant Families in India

   2. Language Barriers, Internal Migration, and Labor Markets in General Equilibrium

   3. Farming, Non-Farm Enterprises, and Migration under Incomplete Markets

5. Kenya

   1. Overcoming Migration Barriers: The Impact of an Income Smoothing Program for Kenyan Migrants

6. Mexico

   1. Escape the Heat: The Dynamics of Migration as Adaptation to Climate Change

7. Tanzania

   1. Spatial Labor Market Power in Sub-Saharan Africa: The Roles of Self-Employment and Migration

8. US

   1. Coworker Networks and Internal Migration in the United States: Evidence from US Administrative Data

   2. Migration as Climate Adaptation: Evidence from California Wildfires

Historical:

1. China

   1. Reshaping Partition into Partnership: Forced Diversity and Development in Rural China

2. Malaysia

   1. Coercive Growth: Forced Resettlement and Ethnicity-Based Agglomeration

   2. Ethnic Proximity and Politics: Evidence from Colonial Resettlement in Malaysia

3. US

   1. Networks and Geographic Mobility: Evidence from World War II Navy Ships

   2. Time to Accumulate: The Great Migration and the Rise of the American South

Modern Day

China

Housing Price, Internal Migration, and Intergenerational Mobility

Qingyuan Chai (Boston University)

This paper examines the role of housing affordability in shaping intergenerational mobility by affecting internal migration in China. It further explores how housing policies might address the challenges that rising housing costs pose for social mobility and inequality. To address the endogeneity of housing prices, I employ an instrumental variables approach, exploiting the Housing Purchase Restriction policy as a natural experiment. This policy limited the number of properties households could purchase in selected prefectures, thereby creating quasi-exogenous variation in housing price growth. I find that higher destination housing costs deter migration, with a more pronounced effect on individuals from disadvantaged families. As a result, these individuals earn lower incomes than their counterparts from more affluent backgrounds. Therefore, higher housing costs reduce intergenerational mobility. Furthermore, to distinguish among destinations and evaluate the effects of various housing policies, I adopt a structural approach to complement the aggregate-level reduced-form results. Unlike the reduced-form analysis, the structural approach reveals that the impact varies across different destinations. Rent subsidies in megacities primarily increase migration among advantaged individuals than disadvantaged ones, thereby exacerbating income disparities. Conversely, policies targeting disadvantaged groups or offering non-targeted subsidies in non-megacities help increase migration for disadvantaged people and raise intergenerational mobility.

Internal Migration Restrictions, Aggregate Productivity, and Spatial Growth

Yanbin (Tracy) Xu (Georgetown)

This paper investigates the impact of China’s hukou system on internal migration, aggregate productivity, and spatial economic growth using a dynamic spatial equilibrium model. The hukou system, a household registration policy, imposes varying degrees of restrictions on labor mobility across regions, leading to a mismatch between labor supply and demand across regions. By incorporating institutional mobility frictions into a spatial economic framework, this research quantifies the effects of hukou constraints on the spatial distribution of labor, productivity, and output and explores the potential long-term economic growth resulting from spatial redistribution. The model is calibrated using data from Chinese prefectures spanning 1990 to 2020, allowing for a comparison of simulated outcomes with observed economic and demographic trends. The findings indicate that the hukou system dampens mobility by about 36% and results in long-term productivity loss of approximately 17.38% and a welfare reduction of about 106.9%. The abolition of the hukou system is projected to increase aggregate real output. However, the transition is marked by initial declines in welfare, highlighting the short-term challenges associated with such a reform. The analysis further reveals the heterogeneous effects of hukou abolition across regions, underscoring the importance of complementary policies to fully realize the potential benefits of reform.

Parental Rural-Urban Migration and Child Education

Conghan Zheng (Barcelona School of Economics)

Household migration involves both parental and child location choices, which raises more issues of selectivity and endogeneity. Sending children to different locations determines the quality of schooling and parent-child separation. A nested discrete choice model is developed that incorporates the expected returns to children’s education as part of the parents’ migration decision. Estimation results using panel data of Chinese rural households show that the impact of parental internal migration on children’s education differs by children’s stage of education. Policies on the destination side of migration have the greatest impact on households with primary school-age children, while parents of middle school-age children are the group most motivated to migrate for better educational opportunities for their children. And high school-age children are the group most sensitive to budget constraints, with parents having the lowest substitution between education and migration resources within the household budget. The results also suggest that migration frictions are not effective in controlling rural-urban migration flows as intended.

Ethiopia

Internal Displacement and Conflict in Ethiopia

Anne Krahn (Tufts)

Conflict is often a driving factor behind forced displacement. However, it is less understood whether hosting displaced populations cause further conflict. In this paper, I address this open question focusing on internally displaced people (IDPs) who are displaced but remain within the borders of their own country. I use Difference-in-Differences (DiD) methods to assess the impact of opening IDP settlements on conflict in Ethiopia, a country grappling with significant internal displacement. I find that districts hosting an IDP settlement experience a 4.3 percentage point increase in the probability of conflict compared to those that do not, representing a 205% increase over the control risk of conflict. Importantly, I rule out that this effect is entirely due to population shocks and also that it is simply a continuation of existing conflict, since districts where IDPs are primarily displaced due to climate disasters also experience an increase in risk of conflict after IDP settlements open. The finding that climate-driven displacement also contributes to new conflict poses a policy challenge, as global climate displacement is set to increase in the coming years.

Ghana

Migration, Networks, and Religious Choice

Timothy Hersey (Yale)

Over the last half century, affiliation to Pentecostal denominations has grown rapidly, with this growth concentrated in low income countries. This paper investigates one economic factor influencing this growth - rural-urban migration: religious institutions attract members by serving as hosts for economic and social networks, a service which is especially attractive for migrants into the city. I present new evidence from Ghana to test this hypothesis, using data from the ISSER Northwestern-Yale Panel Survey (GSPS) and a survey of Pentecostal churches in Accra. I construct a series of shift-share instruments for migration, using exposure to price shocks in the mining, construction and manufacturing industries. I find that migration significantly increases Pentecostal affiliation, especially for migrants with few network connections in the city. The process of migration also has implications for village networks. I find that out-migration induces conversion for households at the origin. I present evidence consistent with the story that out-migrant households convert in response to weakening ties with the village network.

India

Human Capital of Children from Short-Term Migrant Families in India

Leena Bhattacharya (Tilburg University)

Short-term or seasonal migration is a primary livelihood strategy for poor households in rural areas, driven by seasonal income fluctuations. There are over 10 million short-term migrants in India but the effect of migration on the education of children from these families remains under-researched. Using the large-scale nationally representative India Human Development Survey 2011-12 data, I examine the effect of short-term migration of household members on the literacy and numeracy skills of children from these families in rural India. Instrumental variable (IV) estimates show that children from short-term migrant families acquire lower levels of language and mathematics skills than children from non-migrant families. Next, I use data from a unique primary survey from Odisha, a relatively poor and high out-migration-prone state in Eastern India, in 2019. I examine children’s performance in language and mathematics, in terms of percentile ranks, and its association with their living arrangement in the months of migration. The findings show that the percentile ranks of children who stayed behind in source villages were not different from those of children from non-migrant families. The results indicate that encouraging migrants to leave their children behind can effectively protect a child’s right to education.

Language Barriers, Internal Migration, and Labor Markets in General Equilibrium

Amrutha Manjunath (Penn State)

This paper studies how language barriers impact internal migration, the skill premium, and aggregate welfare using rich microdata from India applied to a quantitative spatial general equilibrium framework. I first document four empirical facts: (1) workers migrate less often to locations where they face high language barriers; (2) migrants with high language barriers are employed less often in speaking-intensive occupations; (3) migrants with high language barriers get a wage premium; and (4) these patterns are strongest for unskilled workers. To explain these facts, I then develop and estimate a static migration model in which heterogeneous workers sort across occupations and locations by skill and language, with wages accounting for worker selection and adjusting in general equilibrium. I show through the lens of the model how language barriers, by increasing worker sorting and selection, significantly obstruct internal migration, augment skill premium, and reduce aggregate welfare. As economies shift towards services, language barriers increasingly impede aggregate gains due to the rising prevalence of speaking-intensive occupations. In the absence of language barriers—relative to observed changes—structural change would have increased aggregate welfare by 1.9 percent. Finally, I calibrate costs of both program provision and learning languages to evaluate potential benefits of language programs for unskilled migrants. Using the calibrated model, I argue that welfare benefits of implementing language programs would outweigh costs.

Farming, Non-Farm Enterprises, and Migration under Incomplete Markets

Davide Pietrobon (Lund University; with Giacomo De Giorgi, Salvatore Di Falco)

We study how rural households in developing countries use non-farm entrepreneurship to smooth consumption following shocks to farm income. Using survey panel data from rural India, we show that farm households respond to transitory shocks to agricultural productivity by reducing farm labor hours, increasing non-farm labor hours, and becoming more likely to engage in non-farm entrepreneurship and temporary migration. Unlike temporary migration, these shocks have persistent effects on non-farm entrepreneurship, as engaging in non-farm activities enables households to accumulate skills that enhance their non-farm productivity. We then structurally estimate a dynamic model of household labor supply decisions across farming, non-farm activities, and temporary migration. Counterfactual exercises reveal that over 30% of non-farm output in rural India comes from activities that households engage in to protect their consumption in response to agricultural productivity shocks. In terms of policy, improving the functioning of insurance markets can substantially promote the modernization of village economies. Specifically, weather insurance contracts and a minimum income guarantee increase non-farm output by about 70% and 40%, respectively. As a potential implication of climate change, we find that a negative shift in the rainfall distribution leads to more than a 30% increase in non-farm output, underscoring the role of climate-driven structural transformation.

Kenya

Overcoming Migration Barriers: The Impact of an Income Smoothing Program for Kenyan Migrants

Gwyneth Miner (UC Berkeley)

Migration to cities can be a gamble for the rural poor in developing countries: a trade-off between the potential upside of higher income growth and the downside of foregone rural earnings. This paper tests whether mitigating potential short-term income loss associated with migration can induce more young rural Kenyans to move to cities. In rural Kenya, I offer 700 young male workers across 111 randomized villages access to an income smoothing program, which provides workers 4 USD for each day of unemployment, conditional on migration to Nairobi. Access to short-term income support in the city more than triples the migration rate over seven months and results in more persistent moves, even after the program ends. These migration effects are also larger than those of an unconditional cash transfer of comparable magnitude. The unemployment program induces migrants who were relatively richer at baseline to migrate compared to control, suggesting that a safety net may lower the opportunity cost and perceived risk of migrating. Overall, the evidence indicates that appropriately designed migration unemployment benefits can induce persistent moves to urban labor markets, potentially leading to larger wage growth over time.

Mexico

Escape the Heat: The Dynamics of Migration as Adaptation to Climate Change

Robert Baluja (University of Arizona)

Earth’s climate is changing, which is widely expected to drive net reductions to human welfare. In this paper, I study how effectively migration will reduce experienced climate damages. To provide answers to my research questions, I develop and estimate a dynamic lifecycle model of migration within Mexico. I combine this with a non-stationary and spatially varying model of the climate, in which I allow for both fully informed and naive expectations of the future progression of climate change. Estimation of the climate model uses daily-level historical weather data and output from state-of-the-art climate simulations. Estimation of the lifecycle model uses a sample of life histories, covering the years 1950–2019, and follows a nested full solution pseudo-maximum likelihood routine. I find that climate damages from business-as-usual warming would be 28% higher if domestic migration within Mexico was no longer available as a tool of adaptation to climate change. Moreover, the fraction of the population that I estimate as forming naive expectations of the climate system would experience an average of 2% less lifetime climate damages from becoming fully informed on the climate transition. Given that most of the increased damages this population faces come from a reduced propensity to migrate, one way to reduce these losses is to subsidize migration. I find that subsidizing migration at the average level of the internality reduces their welfare losses by 8–19%. The exact value of this reduction depends on whether the policy forces people to use the subsidy in a particular period. Policies that allow individuals to choose when to use them are over twice as valuable to the affected population because they do not overly incentivize dynamically suboptimal moves. This sort of dynamically-available policy is common; examples include provisions from the recent Inflation Reduction Act and first-time homeowners tax credits.

Tanzania

Spatial Labor Market Power in Sub-Saharan Africa: The Roles of Self-Employment and Migration

Samuel Marshall (University of Warwick)

Labor markets in low-income countries are characterized by large gaps between rural and urban income, between wage and self-employment income, and high rates of self-employment. Standard explanations for these features are frictions that prevent the efficient allocation of resources. I propose an alternative mechanism: firm labor market power. I develop a spatial general equilibrium model of monopsony to disentangle the role of labor market power from migration costs and job search costs. I identify the labor supply curve using Tanzania’s 2010 sectoral minimum wage law. I find that rural labor markets are less competitive than their urban counterparts. This finding is driven by the higher share of wage workers employed in large firms. Moving to the competitive equilibrium causes total output to rise by 4.8%. Conversely, reducing migration costs by 10% reduces total output by 4.2%. This counterintuitive finding is explained by the fact that workers choose where to live and work based on the total value of wages and amenities. This creates a wedge between the productively efficient and welfare maximizing labor allocations. The standard result that reducing migration costs causes output to rise is reconciled through either a unidirectional decrease in migration costs in the direction of the city or a symmetric reduction and competitive labor markets.

US

Coworker Networks and Internal Migration in the United States: Evidence from US Administrative Data

Billy Huang (University of Michigan)

Using administrative Census data, I present new empirical evidence of the effects of coworker networks on the location choice of migrants, consistent with information provision about opportunities in other local labor markets. Workers who migrate are significantly more likely to choose the same destination as that of their coworkers, than the destinations of non-coworkers from the same industry and location. Additionally, coworkers who have worked together longer have a greater likelihood of migrating to the same locations. Moves to the same destination firms account for only a small portion of the network effect, suggesting that direct job referrals play only a small role in facilitating such migration. The estimated network effects are stronger among workers who move further distances, are older, and have higher earnings. I find that excluding moves associated with relocation within the same firm, and excluding establishment closures, do not significantly alter the results, consistent with a primary role of information provision about destination markets through coworkers.

Migration as Climate Adaptation: Evidence from California Wildfires

Victoria Wang (UCLA)

Natural disasters are expected to impact a large and increasing number of people with climate change. The movement from risky to safe areas – “adaptive migration” – is theorized to be a key strategy for minimizing the costs of natural disasters. This paper provides novel empirical estimates of the extent of adaptive migration, highlights financial constraints as a barrier, and identifies the causal impact of government disaster aid receipt on migration in the context of California, where fires have become more frequent and severe. Using detailed individual-level geographic data, I estimate the effect of wildfires on migration using a difference-in-differences (DID) event study design by comparing the migration behavior before and after a fire of individuals in census blocks that are burned for the first time with that of those in never-burned blocks within a census tract. I find that an individual experiencing a first fire has a 6.5-percentage-point (p.p.) higher probability of out-migration after four years, an 18.5% increase. Those who experience a fire are not more likely to be in safe areas. However, individuals less likely to be financially constrained after the fire (i.e. those with high credit scores) are more likely to move to safe areas after four years. Leveraging a new instrument for aid receipt – taking advantage of the fact that politically competitive counties are more likely to receive aid – I find that government aid is associated with higher migration, albeit not to safe areas. Overall, migration is occurring, but adaptive responses are small. Moreover, government aid could be redesigned to improve the level of adaptive migration out of risky areas.

Historical Papers

China

Reshaping Partition into Partnership: Forced Diversity and Development in Rural China

Bin Huang (UZH)

What is the impact of ethnic diversity on cooperation and development? This paper provides micro-level evidence from Communist China's rural forced integration campaign during the 1950s-1980s, which brought together farmers from diverse ethnic backgrounds to live and work collectively in agriculture and public goods projects. Using a regression discontinuity design, the findings reveal that institutional context shapes the effects of ethnic diversity on cooperation and development. During the campaign, forced ethnic diversity had a negative impact on interethnic marriage rates and GDP per capita. However, after the campaign ended, these effects shifted to positive. In the long run, the experience of ethnic diversity due to the forced integration campaign led to increased voluntary participation in interethnic farmer cooperatives, which grew larger and more efficient in production. The positive effects are attributed to strengthened interethnic networks and reduced inequality in human capital accumulation. Marketization, as a key institutional change, played a crucial role in transforming the impact of ethnic diversity from negative to positive. This paper highlights the importance of institutions in shaping the effects of ethnic diversity.

Malaysia

Coercive Growth: Forced Resettlement and Ethnicity-Based Agglomeration

Shanon Hsuan-Ming Hsu (University of Chicago)

How do social divisions affect the benefits of agglomeration? While the clustering of people can enhance productivity through social interactions, social divisions such as ethnic segregation and tension may limit these gains. To answer this question, I leverage an ethnic-based resettlement program that forcibly relocated 600,000 rural Chinese into compact villages in 1950s British Malaya. I find that, decades later, areas with higher resettlement had persistently higher population densities and concentrations of Chinese, driven by both the program’s direct impact and internal migration. Moreover, these areas were wealthier, more industrialized, and exhibited greater labor market specialization. However, the economic benefits primarily accrued to the Chinese population, while other ethnic groups saw only marginal gains when geographically integrated with the Chinese and working in non-agricultural sectors. To assess the overall impact of the program, I estimate a quantitative spatial model that allows agglomeration externalities to vary by sector and ethnic composition. While the resettlement increased aggregate output, the gains were insufficient to offset the welfare losses from the program’s coercive nature.

Ethnic Proximity and Politics: Evidence from Colonial Resettlement in Malaysia

Chun Chee Kok (Monash University; with Gedeon Lim)

This paper studies the long-run effects of a colonial-era resettlement program of ethnic minorities, on contemporary economic outcomes and political preferences of ethnic majority individuals in receiving areas. In ethnic Malay-majority Malaysia, the colonial British relocated 500,000 rural ethnic Chinese minorities into fenced-up, isolated, mono-ethnic camps (1948 – 1960) all across rural Malaysia. This brought some pre-existing ethnic Malay-majority areas into closer contact with ethnic Chinese minorities but not others. Site selection criteria were largely military in nature. Using a spatial randomization inference-type approach, we construct counterfactual village locations based on this criteria. Areas located immediately next to Chinese New Villages (0-2km) experienced better economic outcomes and, in turn, had lower vote shares for the ethno-nationalistic coalition, than polling districts located next to similarly suitable, counterfactual locations. We provide suggestive evidence that these lower vote shares were driven by all voters, not just the ethnic Chinese. Together, our results suggest that persistent differences in inter-ethnic proximity can have a lasting, negative impact on voter preferences for ethno-nationalistic politics, by improving economic outcomes and creating opportunities for sustained but transient inter-ethnic interactions.

US

Networks and Geographic Mobility: Evidence from World War II Navy Ships

Allison Green (Princeton)

This paper uses quasi-random assignment to World War II Navy ships during World War II to study how personal networks shape migration patterns. Using newly constructed data on 1.4 million sailors, I measure exposure to geographically diverse shipmates and estimate its impact on post-war migration. A one-standard-deviation increase in a sailor’s exposure to shipmates from different states raises the probability of out-migration from his own state by 4-5% by 1950. Effects on directed migration are larger but heterogeneous by destination, increasing moves to fast-growing Census divisions by over 15%. I then estimate a discrete choice migration model with embedded networks, revealing Navy ties encouraged long-distance moves, in part substituting short-distance moves that would have otherwise occurred. Using variation from Navy networks to construct instruments for the probability of migrating, I estimate large returns to network-facilitated migration, suggesting Navy ties enabled moves to higher opportunity areas.

Time to Accumulate: The Great Migration and the Rise of the American South

Dongkyu Yang (University of Colorado)

The idea that labor scarcity can induce economic development has been long hypothesized (Hicks, 1932; Habakkuk, 1962), but the evidence remains limited. This paper examines how the Second Great Migration (1940–1970) spurred structural change in the American South between 1970 and 2010. Empirical results using shift-share instruments show that out-migration incentivized capital investment and capital-augmenting technical change, increasing capital per worker and output in both agriculture and manufacturing at least until 2010. Labor was reallocated from agriculture to non-agriculture. I then develop a dynamic spatial equilibrium model that allows for substitution between factors of production, factor-biased technical change, and factor abundance-based trade to characterize this process. The quantitative analysis indicates that labor-capital substitution played a major role in adjustments to South-to-North migration.

Economists are, of course, not the only people who study migration; political scientists and sociologists do valuable work here.3 Since job market papers are relatively rare in those disciplines (and frankly, looking for every working paper written by a job market candidate in any discipline would take approx. the next year), this post only includes economists.

I’ve sorted these posts by type of migration (international/domestic4), time period (modern/historical5), and then alphabetized by country and author name within country.

I’ve tried to include as many papers as I’ve found; if I’ve missed your paper; please email me at lagilbert@gmail.com! So: let’s begin with international migration. (Domestic migration JMPs are in the next post.)

International Migration Job Market Papers

Here is a list of titles; abstracts and links are below.

Modern Day:

1. Canada

   1. Labor Market Power, Firm Productivity, and the Immigrant-Native Pay Gap

   2. Understanding Firm Responses to Immigration Shocks

2. Chile

   1. Immigrants’ skills and Capital Deepening: Evidence from Chile

3. Colombia

   1. Immigration Shocks and Housing Dynamics: Evidence from Bogota

4. Germany

   1. Immigrants’ Return Intentions and Labor Market Behavior When the Home Country is Unsafe

5. Netherlands

   1. The Effect of Initial Location Assignment on Healthcare Utilization of Refugees

6. New Zealand

   1. How Restricting Migrants’ Job Options Affects Both Migrants and Existing Residents

7. Spain

   1. The Cost of Waiting for Nationality: Impact on Immigrant’s Labor Market Outcomes in Spain

8. Sweden

   1. In the Shadow of Brothers: Unintended Impacts of a School Entry Policy on Migrant Girls

9. Turkey

   1. Effects of Immigrants on Non-host Regions: Evidence from the Syrian Refugees in Turkey

10. US

    1. Student and Institutional Responses to STEM Optional Practical Training

    2. Global Roommates, Local Outcomes: How Foreign Peers Influence Domestic Students in Higher Education

    3. Does the “Melting Pot” Still Melt? Internet and Immigrants’ Integration

Historical:

1. UK

   1. Organizational Practices and Technology Adoption: Evidence from Jewish Immigration and the Tailoring Industry in England

2. US

   1. Immigrant Entrepreneurship in the US 1910-1940

   2. Ethnic-Occupational Niches: Evidence from the Age of Mass Migration

Modern Day

Canada

Labor Market Power, Firm Productivity, and the Immigrant-Native Pay Gap

Stephen Tino (University of Toronto)

This paper examines the importance of labor market power and firm productivity for understanding the immigrant-native pay gap. Using matched employer-employee data from Canada, I estimate a wage-posting model that incorporates two-sided heterogeneity and strategic interactions in wage setting. In the model, firms mark down the wage below the marginal revenue product of labor (MRPL), and the equilibrium immigrant-native pay gap arises due to differences in wage markdowns and MRPL. The findings suggest that immigrants earn 77% of their MRPL on average, compared to 84% for natives. In addition, immigrants tend to work at more productive firms compared to natives, although they are less productive on average relative to natives within the same firm. To decompose the pay gap into labor supply and demand factors, I conduct counterfactual analyses that take into account general equilibrium effects. The results suggest that within-firm productivity increases the gap, while between-firm productivity decreases it. Differences in between-firm productivity are driven by immigrants sorting into cities with more productive firms, although they tend to work at less productive firms compared to natives within the same city. When all productivity heterogeneity is eliminated, the gap widens, suggesting that differences in labor supply contribute significantly to the immigrant-native pay gap.

Understanding Firm Responses to Immigration Shocks

Xiner Xu (University of Toronto)

I study the effects of immigration on the performance of local firms and their workers leveraging a sharp increase in Canada’s immigration targets in 2016. The policy led to an influx of predominantly high-skilled workers and generated unexpected variation in the growth of the foreign-born population across regions and nationalities. Motivated by the role of immigrant networks as vital channels for transmitting job information, I quantify firms’ exposure to the immigration shock using a shift-share instrument that combines these differential inflows with the ethnic and spatial composition of firms’ existing workforce. Using a novel decomposition of this measure, I draw comparisons across firms that operate within the same labor market based on differences in worker origins. Employers more exposed to the shock accelerated the hiring of recent arrivals who lacked locally accumulated human capital, increased employment and compensation for both immigrant and native workers, and experienced expansions in total output and output per worker. These results are consistent with firms benefiting from immigration through workplace ethnic networks, which may help identify workers’ productivity characteristics that are otherwise overlooked in the labor market.

Chile

Immigrants’ skills and Capital Deepening: Evidence from Chile

Alessandra Anna Palazzo (University of Maryland, College Park; with Carlos Madeira)

We study the effect of migration on firms’ capital deepening, leveraging a massive influx of immigrant workers with varying skill sets in Chile between 2015 and 2019. Using a theoretical framework that allows for differential productivity between native and immigrant workers, we estimate how firm capital responds to increases in the supply of skilled and unskilled workers. Identification relies on the shift-share instrument to estimate firms’ exposure to immigration shocks. We use unique employer-employee data to match worker characteristics, including country of origin and education level, to balance sheet data from firms’ tax returns, providing highly detailed information about the labor composition and capital in the firm that is not available in other contexts. Our findings show that a 1 p.p. increase in the proportion of skilled immigrants relative to total skilled workers in the firm raises firms’ capital per worker by 1 to 2%. The impact of skilled labor on capital deepening is nonlinear and declines as firms’ immigrant concentration increases, consistent with our model’s predictions. This nonlinear relationship is driven by the imperfect substitutability between immigrants and native workers within the same skill group.

Colombia

Immigration Shocks and Housing Dynamics: Evidence from Bogota

Daniel M. Angel (USC; with Lukas Delgado-Prieto and Juan M. Jimenez)

This paper examines the impact of immigration shocks on housing dynamics in middle-income cities. Using detailed data from Bogota at the planning unit level (UPZ), we analyze the effects of the Venezuelan diaspora on housing prices, rents, and residential construction between 2013 and 2018. Employing a shift-share instrumental variable approach, we find that the immigration shock led to a 0.07% increase in housing values and a 0.3% rise in low-cost housing construction. In contrast, rental prices remained relatively stable, likely due to regulatory constraints and the expansion of low-quality housing, which helped absorb the increased demand. Additionally, we found no evidence of native resident displacement or negative effects on native wages as a result of the immigrant influx.

Germany

Immigrants’ Return Intentions and Labor Market Behavior When the Home Country is Unsafe

Teresa Freitas-Monteiro (University of Copenhagen)

Migration is often temporary, and the intended length of stay in the host country is an important determinant of immigrants’ integration. This paper investigates whether shocks to safety conditions in the home country affect immigrants’ return intentions and labor market behavior. We combine administrative and survey data with precise information on violent events worldwide and exploit the quasi-random occurrence of violent events in the home country relative to the timing of interviews and job separations in Germany. We show that immigrants interviewed after a violent event in their home country are 12 percentage points more likely to wish to remain in Germany permanently. The effects are stronger if immigrants are less integrated in Germany and have close family members in their home country. Consistent with the prediction that revisions to the intended length of stay affect immigrants’ labor market behavior, we show that immigrants who enter unemployment when a violent event hits their home country increase their job search effort and find employment faster. However, the same immigrants trade immediate job security for lower earnings, less stable jobs and less productive firms.

Netherlands

The Effect of Initial Location Assignment on Healthcare Utilization of Refugees

Shobhit Kulshreshtha (Tilburg University)

Characteristics of a place, such as healthcare access and the local environment, influence healthcare utilization. Refugees resettled in developed countries are often assigned locations based on the host country’s assignment policies, yet the impact of initial placement on their healthcare usage remains understudied. I use Dutch administrative data to examine the effect of conditions in the initial municipality on healthcare utilization of refugees, leveraging the random assignment of refugees. I show that 10% of the total variation in hospital visits among refugees can be explained by municipality effects. Additionally, being assigned to a municipality with a higher hospital visit rate among non-refugees increases a refugee’s probability of hospital visits. There is significant heterogeneity in the results for other measures, such as depression medication use and general practitioner costs. This study highlights the role of local healthcare access in shaping healthcare usage among refugees, contributing to policy debates aiming to provide separate and more targeted healthcare services for this vulnerable population at the municipality level.

New Zealand

How Restricting Migrants’ Job Options Affects Both Migrants and Existing Residents

Wilbur Townsend (Harvard; with Corey Allan)

Governments often restrict international migrants’ job options. This paper shows that these restrictions can hurt not only migrants but also the existing residents whom they aim to protect. We study New Zealand’s ‘Essential Skills’ work visa, which was New Zealand’s main work visa between 2008 and 2022. Essential Skills migrants could only work for firms which could not find New Zealanders. Loosening restrictions for a single individual has no impact on their wages: migrants who win an unrestricted resident visa through a lottery switch jobs more frequently, but receive no gain in wages. However, when the Essential Skills job restrictions were loosened for all migrants in an occupation, both job-switching and wages typically grew. These results are consistent with a wage-posting model in which each firm pays migrants and residents equally; in such a model, the wage received by each worker will not depend directly on her own outside option but rather on the distribution of outside options among her colleagues. We estimate a wage-posting model, and compare equilibrium wages under the Essential Skills job restrictions to a counterfactual simulation in which migrants’ job options are unrestricted. The restrictions decreased migrants’ average wage by 8%. Although most residents were unaffected by the restrictions, 2.1% had their wage decreased by more than 2%. The restrictions increased profits, especially in firms which employed many migrants. The restrictions decreased annual welfare by $292m NZD — 30% of migrants’ earnings — largely because migrants could not move to firms which they preferred for non-pecuniary reasons.

Spain

The Cost of Waiting for Nationality: Impact on Immigrant’s Labor Market Outcomes in Spain

Yanina Domenella (CEMFI)

In this paper, I examine the impact of administrative delays in obtaining Spanish nationality on the long-term labor market outcomes of legal immigrants. Using Social Security data from 2006 to 2019 and an instrumental variable strategy, I find that longer delays in nationality acquisition result in significantly lower accumulated earnings over a ten-year period, driven by both lower wages and fewer days worked. Specifically, one additional year of delay reduces accumulated earnings over 10 years by 3.8 to 6.7 percent. To understand the underlying mechanisms, I study the short-term effects of nationality acquisition on job mobility and job quality. The results suggest that delays prolong the period of restricted mobility, hindering access to better employment opportunities. After obtaining the nationality, immigrants can afford a more selective and longer job search that pays off in the long run. These findings underscore the importance of timely nationality acquisition for improving economic outcomes and highlight the need for efficient administrative processes to support immigrant integration.

Sweden

In the Shadow of Brothers: Unintended Impacts of a School Entry Policy on Migrant Girls

Anna Hasselqvist (LMU Munich)

When parents prioritize investments in sons over daughters, this gender bias can render otherwise beneficial educational policies ineffective for girls or even lead to unintended negative consequences. This study examines how Sweden's school entry policy interacts with family structure to shape the educational outcomes of second-generation migrant girls. Using a regression discontinuity design on high-quality administrative data, I first assess the direct effects of late school entry, showing that it benefits migrant girls with younger sisters but not those with younger brothers. Furthermore, by investigating sibling spillover effects from an older sibling's late school entry, I demonstrate that spending more time at home with an older brother who enters school late has a strong negative effect on the educational outcomes of younger sisters. I propose a simple theory to explain these results, highlighting gender bias in parental preferences as a key factor. Supporting this interpretation, I present evidence showing that these negative impacts are specific to migrant girls, with neither migrant boys nor native children experiencing similar effects. Moreover, the effects are more pronounced in migrant families with traditional backgrounds and are also reflected in mothers' labor supply decisions when sons, rather than daughters, enter school late.

Turkey

Effects of Immigrants on Non-host Regions: Evidence from the Syrian Refugees in Turkey

Ahmet Gulek (MIT; with Tishara Garg)

This paper investigates how immigration-induced wage shocks can propagate beyond the regions receiving immigrants through the production network. Using the Syrian refugee crisis in Turkey as a quasi-experiment and the near universe of domestic firm-to-firm transaction data from VAT records, we show that the immigration shock propagates both forward and backward along the supply chain. Firms in non-host regions who directly or indirectly buy from host regions demand more labor. Firms who sell to host regions weakly increase their sales. Estimates imply an elasticity of substitution between labor and intermediate goods of 0.76 and an elasticity of substitution of nearly 1 between intermediates. Counterfactual analyses show that the spillover effects on non-host regions are economically meaningful when the host regions are central nodes of the domestic trade network. For example, a 1% increase in labor supply in Istanbul decreases real wages in Istanbul by 0.56% and increases real wages in the average non-host city by 0.38%.

US

Student and Institutional Responses to STEM Optional Practical Training

Kamal Bookwala (University of California Irvine)

The United States is a globally popular location for going to college. In fact, almost 350,000 international students were working towards their bachelors degree in the United States in the 2023-2024 academic year. In 2008, the Department of Homeland Security introduced an extension to the Optional Practical Training (OPT) program, which allows international students to work in the US after earning a higher education degree. This extension applies to any international student who receives a degree in a STEM field from a United States higher education institution. STEM students are now able to seamlessly work in the US for an additional 24 months (above the standard non-STEM 12-months). This paper examines the effects of the OPT extensions using the College and Beyond II dataset, supplemented with some University of California-Irvine administrative data. I primarily employ a cohort-based difference in difference approach to explore the relationship between the OPT policy change and undergraduate students’ course and degree choices. I also track institution-level responses to the policy changes. My main finding is an increase in enrollment in STEM majors after the OPT extension for international as well as domestic students. Additional findings show that after the extensions, institutions with higher international student proportions have greater STEM cohort sizes for domestic students. Supply-side impacts of these policies are that new majors are more often STEM and maximum seat allotment for STEM courses increases after the OPT extensions. Results from this research shed light on how the OPT extension has impacted students and institutions and identify the effectiveness of OPT extension policies.

Global Roommates, Local Outcomes: How Foreign Peers Influence Domestic Students in Higher Education

Zeyu Li (University of California Irvine)

This paper investigates the causal impact of foreign roommates on the academic outcomes of domestic students at a large U.S. public university. Leveraging quasi-random roommate assignment, I estimate the effects of living with foreign peers on major choices and academic performance, with a particular focus on STEM participation. The results show that male students assigned foreign roommates are 3.5 percentage points more likely to pursue a STEM major and to graduate in STEM fields, while no significant effect is found for female students. The positive effect for male students is primarily observed among those who initially declared STEM majors. Further analysis reveals that domestic students from areas with higher proportions of foreign-born residents are more responsive to the influence of foreign roommates, suggesting that prior exposure to diversity is associated with a greater likelihood of being influenced by peer effects in college. GPA and overall graduation rates show minimal impacts, though male students benefit from a reduced time to graduation. The effect is likely through affecting students’ perceptions of their abilities relative to their higher-achieving foreign peers, especially in STEM. These findings contribute to the literature on peer effects and gender disparities in education, highlighting the broader role of diversity in shaping academic trajectories in higher education.

Does the “Melting Pot” Still Melt? Internet and Immigrants’ Integration

Alexander Yarkin (Brown University)

The global spread of the Internet and the rising salience of immigration are two of the biggest trends of the last decades. And yet, the effects of new digital technologies on immigrants - their social integration, spatial segregation, and economic outcomes- remain unknown. This paper addresses this gap: it shows how home-country Internet expansion affects immigrants’ socio-economic integration in the US. Using DID and event-study methods, I find that home-country Internet expansion lowers immigrants’ linguistic proficiency, naturalization rates, and economic integration. The effect is driven by younger and less educated immigrants. However, home-country Internet also decreases spatial and occupational segregation, and increases subjective well-being of immigrants. The time use data suggests that the Internet changing immigrants’ networking is part of the story. I also show the role of return intentions and Facebook usage, among other factors. These findings align with a Roy model of migration, augmented with a choice between host- vs. home-country ties. Overall, this paper shows how digital technologies transform the immigration, diversity, and social cohesion nexus.

Historical Papers

UK

Organizational Practices and Technology Adoption: Evidence from Jewish Immigration and the Tailoring Industry in England

Yannis Kastis (University College London; with Hillary Vipond)

This paper provides causal evidence on the role of organizational practices in driving technology adoption. We examine a shift in practices in the English tailoring industry, prompted by the arrival of Jewish immigrant tailors who fled pogroms in the Russian Empire between 1881 and 1905. By the time Jewish tailors arrived in England, garment production was predominantly bespoke and native tailors were using sewing machines - introduced in the 1860s - to increase individual productivity. In Russia, where sewing machines were unavailable, Jewish tailors specialized in ready-to-wear production, which involved a greater division of labor into specialized tasks than bespoke work. Upon arriving in England, they combined the available sewing machines with their organizational practice to scale up ready-to-wear production. Using original data on production tasks and firm-level data, we study how this shift influenced the adoption of the sewing machine and the transition to mass production of garments in England. Our findings show that organizational practices are instrumental in integrating new technologies into production.

US

Immigrant Entrepreneurship in the US 1910-1940

Zhen Gu (University of Arizona)

I study immigrant entrepreneurship in the early twentieth century. Using the full count US Censuses, I document the widening gap between immigrants and native whites in self-employment from 1910 to 1940. Immigrants from Asia and Southern and Eastern Europe were most entrepreneurial. The tightening immigration restrictions and a series of economic and political events changed the demographics of immigrants staying in the US, which contributed to the higher business ownership rate of immigrants. I propose two factors to explain the remaining immigrant-native difference in self-employment. The pull factor are expanding ethnic enclaves that create markets for ethnic goods, increase labor supply for immigrant business owners, and build up social connection among immigrant entrepreneurs. The push factor is the deterioration of job opportunities in the wage sector of labor market which forces immigrants into self-employment for survival. I find higher wage expectation decreased the probability of immigrant being self-employed while a larger immigrant community size at city level was positively correlated with self-employment. I also show self-employed immigrants and native whites had different preferences for industries. Immigrants tended to concentrate in industries that required less education and financial investment, suggesting immigrants in the period of study used self-employment more as a survival strategy.

Ethnic-Occupational Niches: Evidence from the Age of Mass Migration

Elijah Locke (Boston University)

Why are some ethnic or immigrant groups vastly over-represented in certain occupations? This paper analyzes the drivers of such ethnic-occupational niches by turning to the United States during the Age of Mass Migration. I posit that niches begin due to immigrants' pre-migration skills, and persist via social networks. Between 1850 and 1940, I find that 46% of all immigrants were in niches. Immigrants niched in occupations ranging from tailors to managers and bakers, and the degree of their over-representation varied widely across nationality, time, and occupation. In an occupational choice framework, a skill background alone may generate a niche, but social networks have critical implications for misallocation and intergenerational mobility. Finally, I estimate the precise impact of social networks on occupational choice. A one-unit increase in the clustering of one’s incumbent co-ethnics in a niched occupation corresponds to a 7.3% increase in the likelihood a newcomer chooses that occupation. Immigrants with the least information about labor markets are especially reliant on their networks. A novel IV setup leveraging newly digitized data points to minimal OLS bias. Altogether, my findings are in line with those regarding contemporary immigrant networks, thereby reaffirming the connection between niches past and present.