Even in the historical era of near-open borders, migration was not free of cost. But the main costs of international migration were similar to the costs of intranational migration: the cost of transporting one’s person and belongings, and of finding work, lodging, and sustenance at one’s new home. Legal restrictions on migration imposed no financial cost. At some immigration checkpoints, such as Ellis Island, officers turned back migrants who they thought would be unable to sustain themselves, but they did not charge an admission fee (beyond a small fee to cover administrative costs). So migrants simply had to demonstrate that they had enough money to take care of themselves for a few days while they looked for a job, rather than pay a huge fee to the officers.

Today, legal travel costs are quite low. One can get a one-way flight between diametrically opposite corners of the world for well under $2000. This is far from a negligible amount of money for the world’s poorest people, but it’s something that people above the very lowest rungs of poverty could probably access in loans. Particularly given the huge place premium, it seems like people would be quite willing to lend money to finance journeys, and travel costs wouldn’t be a bottleneck. For migrants with more assets back home, moving costs could be greater insofar as they may need to dispose of their assets before moving, but again, this cost is a small proportion of the migrant’s net worth. (Two somewhat related posts of mine: factors that would constrain migration if borders were opened quickly over a short span of time, and selection effects for migrants: some a priori possibilities. The latter discusses the way that moving costs could lead to selection effects for migrants).

At present, however, legal migration is a very difficult option for many people. Those who seek to migrate unlawfully have two options: either enter on a tourist visa or temporary work/study visa and then overstay it, or smuggle oneself in illegally. But temporary work/study visas are pretty hard to get too. Consular officers are quite cautious about granting tourist visas to people who seem like they are trying to use it as a pretext to migrate, and the tourist visa option is also thereby rendered unavailable to many potential migrants (in fact, consular officers have an extraordinary level of discretion here and little accountability to the visa applicants they deny, see my co-blogger John’s posts on the subject here, here, and here). Moreover, coming in on a tourist visa leaves a legal trail for the migrant, making it potentially easier for authorities to track the migrant down. Thus, the incentive for people to cross borders surreptitiously, despite the considerable physical discomfort and danger. In the United States, somewhere between a third and a half of migrants who are not currently in legal status are people who actually crossed the border without authorization.

African migrants stranded on a boat coming from Libya wait for rescue services near Sfax, on the Tunisian coast, on June 4, via CNN

As discussed on our human smuggling fees page, the costs of smuggling oneself in illegally are nontrivial (if you’re just interested in the prices, see this Havocscope page). Costs vary heavily by source country but are generally at least ten times the legal travel costs. Illegal immigration from China or India to the United States costs in the range of $50,000. And these are the pure financial costs, not the other costs of the journey, including the physical risks. This raises a natural question: who but the very wealthy could afford such sums? After all, most of the people who migrate illegally aren’t well-off. Where are they getting this money?

The most obvious answer (which is a non-answer of sorts) is that most people don’t. The legal barriers to migration are successful in keeping a significant fraction, probably a vast majority, of potential migrants, out. This is exactly why we are far from open borders and it’s such a radical proposal. As my co-blogger David Bennion wrote:

The immigration system isn’t broken, it is working as intended. But it needs to be broken; we need to break it.

Nonetheless, enough people do migrate in this manner for the phenomenon to be worth understanding. As I discussed in my blog post on snakeheads such as Sister Ping, a fair number of people migrate all the way from China to the United States in cramped ships, paying the huge human smuggling fees.

I outline three interconnected answers below. Here’s the list:

1. Personal and family savings back home
2. Financing by family members already in the destination country
3. Loans from outside the family

#1: Personal and family savings back home

Many migrants save up money for over a decade in order to be able to finance their journey. In some cases, they liquidate their assets for the purpose.

In many cases, parents save up money over the course of their lives so that their children can access better opportunities. Just as some parents save up for their children’s college education while others save up for their children’s marriage, some parents save up so their children can be smuggled out of the country. For instance:

Liu’s father, a fisherman, had scraped together every penny he could possibly find to fund his son’s journey to the U.S. – using his savings, borrowing from families and friends. He paid $30,000 upfront to the smuggler – an astronomical figure for an average Chinese villager, whose annual income was a few hundred dollars at the time.

Also, the human smuggling and money transfer business have close connections, as relatives in the home country may send money to smugglers to pay off debts for successfully smuggled family members:

When the families of new arrivals wanted to send money to America to satisfy a relative’s snakehead debt, Sister Ping [a human smuggler] could pay that sum out of her reserves in New York.

Keefe, Patrick Radden (2009-07-15). The Snakehead: An Epic Tale of the Chinatown Underworld and the American Dream (p. 46). Knopf Doubleday Publishing Group. Kindle Edition.

One phenomenon observed in education is that, in poor families, the family often makes a bet on which child’s education it’ll finance (beyond the very basic level it can afford for all children). For instance, they may bet on the child who is smarter and more studious and therefore use combined family savings to finance her college education. The hope is that this child, once she gets a good job (made feasible by the human capital/signaling/networking benefits of a college education) can in turn provide support to the rest of the family, perhaps by financing the education of her siblings. This kind of triage makes some rational sense in places where there’s a huge income premium for those who get a college education, so that having one college-educated kid and one middle-school-educated kid is a better strategy than having two high-school-educated kids. In some cases, the belief may be factually irrational and based on inaccurate beliefs about the return to education. Here’s a quote from Poor Economics on the phenomenon:

Parents also tend to believe that the first few years of education pay much less than the next ones. For example, in Madagascar, parents believed that each year of primary education would increase a child’s income by 6 percent, each year of junior high education by 12 percent, and each year of senior secondary education by 20 percent. We found a very similar pattern in Morocco. There, parents believed that each year of primary education would increase a boy’s earning by 5 percent, and each year of secondary education by 15 percent. The pattern was even more extreme for girls. In the view of parents, each year of primary education was worth almost nothing for them: 0.4 percent. But each year of secondary education was perceived to increase earnings 17 percent.

[…]

This belief in the S—shape means that unless parents are unwilling to treat their children differently from one another, it makes sense for them to put all their educational eggs in the basket of the child they perceive to be the most promising, making sure that she gets enough education, rather than spreading the investment evenly across all their children. A few doors down from Shantarama (the widow whose two children were not in school), in the village of Naganadgi, we met a farming household with seven children. None of them had studied past second grade, except the youngest, a twelve-year-old boy. They were not satisfied with the quality of the government high school, where had spent a year. So the boy was attending seventh grade in a private boarding school located in the village. A year at school cost the family more than 10 percent of its total income from farming, a considerable commitment for just one child and clearly an impossible expense for seven. The lucky boy’s mother explained to us that he was the only intelligent child in the family. The willingness to use words like “stupid” and “intelligent” to refer to one’s own children, often in their presence, is entirely consistent with a worldview that puts a large premium on picking a winner (and in getting everyone else in the family to back the winner). This belief creates a strange form of sibling rivalry. In Burkina Faso, a study found that adolescents were more likely to be enrolled in school when they scored high on a test of intelligence, but they were less likely to be enrolled in school when their siblings had scored high.28

Banerjee, Abhijit; Duflo, Esther (2011-04-26). Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty (pp. 88-89). Perseus Books Group. Kindle Edition.

Anecdotal evidence suggests something similar for migration. Rather than making an egalitarian effort to save enough money for everybody to migrate simultaneously, families try to save enough so that one person (typically a young male) can make the move. The hope and expectation is that once the person is settled, he can help finance the migration of others, and/or provide financial assistance to them so they can have a better standard of living back home.

Both aspects of this lead to the conclusion that even this apparently unselected migration is likely to be heavily selected. Individuals who can save a large chunk of their income for over a decade to pursue a dream of smuggling themselves into a distant land are unlikely to be very ordinary, even if their qualifications on paper look no different from those of others around them. Insofar as families pool their resources to finance the migration of one member, there is also intra-family selection. This does mean that the selectivity of such migration may be stronger than what we might believe a priori. It also means that opening the borders might lead to lower selectivity even for “unskilled” migration.

Political refugees are likely to fall disproportionately in this category. Many of them are relatively well-off by the standards of their home country and the main reason they are fleeing is that they belong to a persecuted minority (sometimes directly on account of their relative wealth).

Another implication worth noting is the potential for differential selectivity. Given the low human smuggling fees from Mexico to the United States, those who are able to pay the fee may not be as heavily selected as those from China who pay the much higher fees. This suggests that, even for illegal immigration arising through unauthorized entry, migration from China may be more heavily selected than migration from Mexico. Therefore, we should be cautious about using the comparative performance of migrants from these countries to draw strong conclusions about the culture of the source countries and we should be careful before generalizing group differences for observed immigrants, even illegal ones, to the open borders scenario.

UPDATE: I raised some of these issues in an Open Borders Action Group post, and Carl Shulman commented on the post, linking to Cynthia Feliciano’s research, and noting that treating illegal immigration as unselected is a reasonable assumption overall in the US context because Mexican migration dominates numerically. A later comment by him suggested that migration to the UAE might provide a better idea of the level of selectivity we would see under open borders. See also Carl’s blog post on migration to the UAE.

#2: Financing by family members already in the destination country

In many cases, the families of potential migrants that are already in the destination country finance their journeys. They’re trying to get their spouses, siblings, children, or parents to join them. For close relationships, where resources are pooled anyway, the financing may be a gift. For somewhat more distant relationships, the financing may be in the form of a loan, albeit one made informally with flexible payment terms. Because we’re talking of family members in the destination country who have probably been there for a while, the amount of money is still large but not unthinkably exorbitant, thanks to the huge place premium.

This ties in very neatly with #1, because the implicit deal or payout in #1 is usually #2. It also ties in with #3, loans.

Note that the degree of selectivity here is lower. This also ties in somewhat with diaspora dynamics. In my blog post on snakeheads who facilitate smuggling from China to the United States, I mentioned this source of financing as an important one for many migrants. Here is the quote from the book:

One misconception about the snakehead business is that the smugglers will bring people over and then force them to work as indentured servants for years in order to pay off their debt. Such an arrangement would make very little sense from the smuggler’s point of view. A busy smuggler like Sister Ping didn’t want to keep track of scores of debtors at various stages of repayment, any of whom might skip town during the months, or more often years, that it took them to come up with $18,000. Instead, the smugglers would hold passengers once they arrived in the United States, giving them thirty-six or seventy-two hours to satisfy the debt. Such an arrangement might be unimaginable in any other ethnic community, but familial and communal ties among the Chinese in America were so strong that a new arrival could count on a guarantor cobbling together a five-figure fee by borrowing small amounts from many people—$1,000 here, $500 there. The immigrant was thus indentured not so much to the snakehead as to his own family.

Keefe, Patrick Radden (2009-07-15). The Snakehead: An Epic Tale of the Chinatown Underworld and the American Dream (pp. 44-45). Knopf Doubleday Publishing Group. Kindle Edition.

Elsewhere in the book, Keefe notes that many migrants who were smuggled in recently spend the first few years saving up money so as to be able to pay for their relatives to be smuggled out:

older relatives [in Chinatown] who were largely absent, working day and night to pay off snakehead debts or raise money to send for more relatives.

Keefe, Patrick Radden (2009-07-15). The Snakehead: An Epic Tale of the Chinatown Underworld and the American Dream (p. 65). Knopf Doubleday Publishing Group. Kindle Edition.

#3: Loans from outside the family

Even with the huge smuggling fees, migration pays off, thanks to the place premium. This makes it possible, at least on paper, to have a good business model of lending to such migrants. The main problems relate to enforcement: how does the lender collect on delinquent buyers, given the underground nature of the whole operation? Some ways this can be done:

• Using family members in the source or destination country as guarantors, and collecting from them either immediately or if the lender fails to repay (cf. the quote above).
• Having connections with gangs in the local community in the source or destination country, that can use violence or threats of violence if people fail to pay up.
• Having connections with the employment networks used by migrants (such as Chinatown’s kitchen network), making it difficult for people who haven’t repaid loans to be able to get a job or otherwise survive in the community.

This method of financing is most common for semi-legal or gray-area migration, and might describe migration to places such as Malaysia and the UAE (the latter even being legal). It also describes a decent fraction of illegal migration from China to the United States, with close connections between the lenders and the snakeheads who facilitate the actual human smuggling.

Loan financing has also been common historically. Back in the era where the main cost of migration was the travel cost, the credit-ticket system financed journeys through loans, that needed to be repaid through what often came close to indentured servitude. The legal immigration to places such as the UAE is often financed through loans in a similar way, with the lenders often confiscating the passports of the migrants until the time that the migrants repay the loans. There has been considerable debate surrounding the ethics and legality of the enforcement mechanisms used. On the one hand, it could be argued that in the absence of explicit legal recourse, such coercive measures are necessary to make lending profitable. On the other hand, the prima facie ethics of the actions is questionable. This is an issue worth a separate, detailed exploration. (Incidentally, Carl Shulman’s blog post on the UAE is worth reading if you’re interested in a comprehensive and balanced assessment).

See also an Open Borders Action Group post where I provide a summary of the three explanations and various commenters offer their thoughts on them.

Related reading

• Snakeheads as high-impact entrepreneurs by Vipul Naik, November 16, 2014.
• Human Smuggling and Border Crossings by Gabriella Sanchez by Bryan Caplan, October 22, 2015, on EconLog.