The Obama administration has offered a temporary reprieve from deportation for up to 1 and 3/4 million immigrants who came to the United States as children.
Whatever the immediate merits of the Deferred Action for Childhood Arrivals, it signals a much broader principle all immigrant receiving countries should recognize: children experience migration differently than adults, and public policy can create both great opportunity and great risks for their long-run capacity to become independent and successful adults.
The DACA initiative was announced earlier this year, but it is only now, with the release of the detailed eligibility criteria, that millions of young people, some of whom have been raised almost all their lives in the United States, are finding out if they qualify to shed their status as illegal immigrants, if only for a two year period.
The policy offers a reprieve from deportation and an authorization to work for individuals who came to the US before the age of 16, and had not turned 31 by June 15th of this year. But there are other conditions that also have to be met: in addition to not having brushes with the law, potential beneficiaries must either currently be in school or be a high school graduate.
Jeanne Batalova and Michelle Mittelstadt of the Migration Policy Institute summarize their demographic profile noting that of the 1.76 million almost three-quarters are of Mexican and Central American origin.
About 800,000 children and youth are potential beneficiaries, and the DACA education requirement gives them a very strong incentive to stay in school and graduate from high school.
The high school drop out rate is notoriously high in the United States to begin with, but particularly so among migrant children from Mexico.
Children of Hispanic origin face all sorts of challenges in successfully completing high school, some of which have to do with the way they are treated by US society, job markets, and the schooling system. This is clear from the experience of Puerto Ricans, who are not strictly speaking immigrants at all, and are likely to have some exposure to English before coming to the mainland.
But the drop-out rates of Mexicans is striking even if Puerto Ricans are used as a base case. More than one-half of children arriving after the age of 10 from Mexico will be high-school drop-outs, the proportion increasing with each subsequent year and reaching 70% for those arriving as 15 year olds.
Arriving younger is better. Those coming to the US from other Latin American countries illustrates a common pattern: age at arrival does not matter until roughly the age of 9 or 10, thereafter the chances of dropping out of high school rise steadily.
This age threshold matters because it is associated with the capacity to fluently learn a new language, and with making a series of social transitions, particularly during the teen years, that lead to high school graduation, and ultimately success as an adult.
In this sense it is little wonder that particularly high risk groups—those most likely to drop-out of high school, to then have a tenuous link to a job, and to be more likely to join a gang and commit a felony or misdemeanor—are boys from a non-English background who came to the country as young teenagers.
There is much to be applauded in the Obama administration’s recognition that child migrants need special treatment, but these patterns suggest a temporary reprieve will offer an incentive to stay in school only for a fraction of potential beneficiaries.
At the same time the eligibility criteria will leave important groups disenfranchised. Those who have already dropped out are most likely to have had the deck stacked against them by virtue of the simple fact that they arrived as young teens and faced bigger challenges learning English.
[This post uses information from a just published research paper by Audrey Beck of San Diego State University, Marta Tienda of Princeton University, and myself called “Age at Immigration and the Adult Attainments of Child Migrants to the United States .” We show that the variation in education due to age at arrival influences adult outcomes that are important markers of integration into the American mainstream: earnings, employment, and marital behaviour.]
Economics for public policy 
Excellent research ,Professor Corak et al. I was just wondering what percentage of the children aged 10 and above are really engaged in paid employment or assist their parents in the workplace and so do not have the time to attend school as other children.If this percentage is high then public policy needs to address the issue of child labour and children rights among the population in question.
Thank you for raising this. In fact many analyses of Mexican migration to the US on child migrants often restrict their discussions to those who are 15 or even 14 and younger. The reason for this is the feeling that some children come to the US looking for employment, which would implicitly suggest that you are raising a valid point worthy of a closer look.
I don’t have the data at hand to comment on the issue … indeed to the extent that this involves “illegal” migrants it might be difficult to get good figures. The analysis in our research paper is based on all individuals 17 and younger. This follows the definition of a child in the UN Convention on the Rights of the Child.
Someone who has worked on issues associated with illegal migration is Doug Massey of Princeton University, but I don’t know if he specifically addressed this issue. I would be interested to learn from other readers if they have a sense of the percentage you are asking for, and if the issue of child labour is even part of the US discourse on this issue.
Professor , your response is much appreciated. It occured to me that if the undocumented parents are largely low income earners …then they may rely on their children to supplement the income of the family . This may be the case in activities such as the harvesting of fruit or other agricultural commodities where the marginal productivity of a child may actually be quite high. This is quite a common problem in some underdeveloped countries where the short term rewards of employment can benefit the family group more than the long term gains of schooling for the child. If the issue of child labour is cultural, then another policy response may be to educate the parents of such children on the laws governing schooling and the benefits of keeping the child away from the workplace. This should keep more children in school and should make them less “at risk ” as they enter the adult world of work.
Prof Corak, Greetings from Washington DC!
I was just reading your post, and the following occurred to me: how do you draw the age-line for ability to learn a foreign language based on age of arrival. I know you indicated 9-10 years of age, but where did you dig up this age threshold and have you consulted differing views before deciding on this specific age? 🙂
Many thanks! Always enjoy reading your posts.
Hana K
Thanks for this Hana. I hope you are enjoying Washington and your work there.
The age threshold is something that is determined from the data. In particular I use information on both language usage and also on how patterns in not having a high school diploma vary with age at arrival. I am continuing to explore this in more detailed ongoing research with refined measures of language use, but you can see pictures and the statistical analysis on dropping out of high school in this paper http://ftp.iza.org/dp6072.pdf, the major objective of which was to uncover the most accurate threshold.
This said there is also some theory in the linguistic and early childhood development literatures to suggest that language acquisition ability changes discretely some point before the onset of puberty, though this is often interpreted as being just acquisition of an accent.