The Case for Nuclear-Family Immigration Over Extended Family Chains

There is a line at the State Department that some people have been standing in longer than some of their eventual neighbors have been alive. Nobody in the line is breaking a rule. Nobody is cutting. They filed the paperwork, they paid the fees, they waited, and the wait is measured, for some, in decades.

A system that makes honest people wait that long is not generous. It is broken, and the break is not in the processing. It is in the design.

The claim

Let me state it plainly. American immigration law should protect the nuclear family, meaning a married couple and their minor children, and should stop extending sponsorship to adult siblings and married adult children of citizens. The conservative case rests on three points: that family unity is best defined by the household, that open-ended categories produce queues that make nonsense of the promise of reunification, and that a nation has the right and the duty to choose how many people it admits and on what basis.

I am not making this case because family does not matter. I am making it because it does.

The law as written

You cannot argue about the structure without knowing the structure. Here it is.

Immediate relatives of United States citizens are the spouses, unmarried children under twenty-one, and parents of citizens who are at least twenty-one. They are not subject to the annual numerical limits. A citizen can petition for them, and as long as the relationship is genuine and they are otherwise admissible, no quota stands in the way.

The four family preference categories cover everyone else with a qualifying relationship. They are numbered in section 203(a) of the Immigration and Nationality Act:

• F1: unmarried sons and daughters, age twenty-one or older, of citizens.
• F2A: spouses and minor children of lawful permanent residents.
• F2B: unmarried sons and daughters, age twenty-one or older, of lawful permanent residents.
• F3: married sons and daughters of citizens.
• F4: brothers and sisters of citizens, where the citizen is at least twenty-one.

The statute sets annual numbers for these. The figures are 23,400 for F1, 114,200 for F2 as a whole, 23,400 for F3 and 65,000 for F4, which together equal the 226,000 floor for family preference visas. And each category is also subject to the per-country cap in section 202, under which no single country may receive more than seven percent of the total in the family and employment preferences combined.

Now do the arithmetic that the statute forces. Suppose a country has far more people with qualifying relatives in the United States than the cap will admit. The supply of visas is fixed. The demand is not. The line lengthens every year, because each year's new petitions join the old ones.

The visa bulletin and what it shows

Each month the State Department publishes the Visa Bulletin, a table that lists, for each category and each major country, a "cutoff date." If your petition was filed before the cutoff date, your priority date is current and you may proceed. If it was filed after, you wait. The bulletin is public and is worth reading once, because nothing explains the system better.

I will not quote the current cutoff dates, because they change every month and I would be handing you a stale number. I will tell you what you will find when you look. In the sibling category and the married-adult-child category, for applicants from the countries with the most petitions, the cutoff dates sit many years behind the calendar. In some country columns, the gap is measured in decades. A petition filed today by a citizen for a brother in one of those countries will not, on present trends, be reached in the petitioner's working life. Check the bulletin yourself and see.

Think about what that means as a policy. A law that tells a woman her brother may come, and then tells her he will be called in a generation, has not reunited a family. It has taught a family to wait, and in practice it has taught some families that there is another path, which is to enter without waiting. The queue itself is an incentive to break the queue.

The steelman: why people defend the chains

Let me give the other side its best argument, as its defenders would put it.

One, family is the strongest institution there is. A newcomer who arrives with relatives nearby has support in finding work, housing and church, and is less likely to depend on public benefits. Family networks are how America absorbed its great waves of immigrants.

Two, a citizen's love for a brother or a sister is not a lesser love because the sibling is an adult. A mother does not stop being a mother when her son marries. Why should the law?

Three, the family categories are the one part of the system that ordinary Americans, not employers or officials, control. A citizen who sponsors a relative is vouching for him, usually in a binding legal document, and is financially responsible for him.

Four, there is no clean line. Once you say that an adult sibling is out, someone will ask about an adult child who cares for an aging parent, or an unmarried brother with a disability. The moral claims blur at the edges.

I take those arguments seriously. Now let me answer them.

The answer

First, the nuclear family is a principled line, not an arbitrary one. The household formed by a married couple and their dependent children is the basic unit of society. It is where children are raised and where the duty of support runs most directly. A husband and wife cannot be expected to live on separate continents, and a minor child cannot be expected to be raised without a parent. Those claims are different in kind from the claim of an adult sibling, who has a life, a household and a career of his own and has been living them without the citizen for years. The law should recognize that difference. Most immigration systems in the developed world do. Many systems put spouses and dependent children first and treat adult relatives more like ordinary applicants who must qualify on their own.

Second, the chain is the problem. The phrase "chain migration" is political, and I do not need it. What I mean is the plain mechanism: a person immigrates, becomes a citizen, petitions for a sibling, who immigrates and, in time, petitions for his own spouse and children and later his own siblings. Each admission creates the right to petition for more. The flow is not chosen by Congress as a number. It is chosen by the family tree. And since each family's tree branches differently, the system produces an outcome that no one designed: a pattern of admissions determined by who happened to be admitted before. That is not a policy. It is a feedback loop.

Third, the queue makes the promise a lie. If the point of the family categories is to reunite families, then a twenty-year wait defeats the point. A system that admitted fewer categories and processed them promptly would reunite more families, sooner, than a system that promises everything and delivers in a generation. A shorter list with a short line is more generous than a longer list with a long one.

Fourth, the argument about moral claims blurs at the edges, but so does every line. The age of majority is arbitrary. So is the voting age. We draw lines because law requires them. A nuclear-family rule can have sensible exceptions, such as for dependents with a disability, without abandoning the rule.

Fifth, family sponsorship does not guarantee economic self-sufficiency. The affidavit of support that a sponsor signs is a legal promise, but enforcing it has been difficult, and the question of when newcomers rely on public programs is contested. The very first rule that the republic applied to immigrants was that they must not become a public burden, a story told in Self-Sufficiency Was the First Immigration Rule, the Public Charge Story. A category that has no test of any kind except relationship is at odds with that tradition.

What the Jordan Commission recommended

This is not a partisan invention. In the 1990s, the U.S. Commission on Immigration Reform, chaired by Barbara Jordan, a Texas Democrat and a civil-rights figure, studied the system for years. Its 1995 recommendations on legal immigration, as I recall them, urged that family-based admission be focused on the nuclear family: spouses and minor children of citizens and permanent residents. It recommended ending the preferences for adult sons and daughters and for siblings, and it recommended lowering the overall level of legal immigration. The reasoning was that the national interest, not the extent of a relative's family tree, should determine who comes. I tell her story, and why Congress set the recommendations aside, in Barbara Jordan's Commission Wanted Less Immigration and a Workplace Check.

I mention this for a reason. A commission chaired by a Democratic icon of civil rights, appointed to study the question without a political agenda, reached the same conclusion that restrictionist conservatives reach today. The idea that limiting extended-family chains is an extremist proposal does not survive a look at who proposed it first.

A worked example, labelled as a hypothetical

Picture an engineer who immigrates in 2000, becomes a citizen in 2006, and in 2007 petitions for his sister, who is married with two children. The sister's petition is under F4. Her priority date is fixed as the filing date in 2007. If her country's cutoff is moving slowly, she waits. When her turn comes, her husband and her two children, who are derivative beneficiaries, may come with her. They now have the right, once they become citizens, to petition for their own siblings. Each of those petitions begins a new line.

Now take the same engineer and ask what the nuclear-family rule would have done. His wife and any minor children would have come immediately, as immediate relatives. His sister would not have a claim by relationship. She could apply on her own, under the employment categories or a points system, if she qualified. The engineer's family would be united. The chain would not exist.

Which of these two is more respectful of the engineer? I think the second, because it takes the family he is actually raising seriously, and it does not leave his sister standing in a line for seventeen years.

What a good reform would include

A reform that honored the nuclear family would do four things.

1. Keep immediate relatives unlimited. Spouses and minor children of citizens would continue to come without a cap, and the spouses and minor children of permanent residents would be moved forward, by raising the F2A numbers or treating them as immediate relatives, so that the husband of a green-card holder does not wait for years.
2. Close the adult-sibling and married-adult-child categories to new petitions, with a fair transition for those whose petitions are already pending, so that no one who has waited honestly is told the line has simply ended.
3. Permit a narrow dependency exception, for an adult child with a disability who depends on the parent.
4. Open a merit channel so that siblings and adult children can apply as individuals, on their own qualifications. The design of such a channel is the subject of What an American Points System Would Look Like Written Into Law.

The transition is the hard part. The honest way to handle it is to say what the law will do with those already in line, rather than to leave them in doubt. Other countries have faced the same choice, and a sensible one allows pending cases to be processed over a fixed period.

For readers who want the practical side of how a citizen sponsors a spouse or child under the present law, How to Bring a Family Member to America the Legal Way, Step by Step walks through the steps.

The Founding and the family

The Constitution says nothing of families in its naturalization clause. It gives Congress the power "To establish an uniform Rule of Naturalization," and leaves the rest to legislation. That means the shape of family immigration is a decision for Congress, and Congress has changed it before. The 1965 Act, which established the family preferences, was passed in the belief that it would not greatly change the numbers of admissions, a prediction that turned out to be wrong. If a prediction was wrong, the law should be reviewed.

The Declaration of Independence says that governments derive "their just powers from the consent of the governed." Consent of the governed includes consent about who joins the governed. Under the family chains, the consent is given not by the public through its representatives but by the accident of a family tree. I do not think that is what the Founders had in mind.

The case in brief

1. Immediate relatives of citizens, including spouses and minor children, are admitted without numerical limit, and the four family preference categories, F1 to F4, cover adult and married children and siblings under a 226,000 floor.
2. The monthly Visa Bulletin shows cutoff dates that, in the sibling and married-child categories, sit many years behind the calendar, so the promise of reunification becomes a wait of decades.
3. The nuclear family, a couple and their dependent children, is a principled line, and extended chains admit people by family tree rather than by a number Congress chose.
4. The Jordan Commission, chaired by a Democratic civil-rights leader, recommended focusing family admission on the nuclear family in the 1990s.
5. A good reform would keep immediate relatives unlimited, close the sibling and married-adult-child categories to new petitions with a fair transition, allow a disability exception and offer a merit channel.

What you can do: Open the current Visa Bulletin on the State Department's website, find the F4 and F3 rows for the country with the longest wait, and send that table to your representative with a question about which categories Congress should keep.