Here is how chain migration works in practice. A software engineer from India arrives on an H-1B visa in 2005. He gets a green card through his employer in 2010. He becomes a citizen in 2018. He immediately petitions for his parents, his unmarried adult sister, and his married brother. His parents arrive within two years - they are "immediate relatives" of a U.S. citizen, so there is no cap. His sister enters the F1 queue. His brother enters the F3 queue. The sister waits 15 years. The brother waits 22.
When the sister finally arrives in 2033, she gets her own green card. She becomes a citizen. She petitions for her husband and her children. Her husband arrives and eventually becomes a citizen too. He petitions for his parents and his siblings. Each of them, upon arrival and naturalization, can petition for their own family members.
One software engineer in 2005. By 2050, the chain has brought in 15, 20, perhaps 30 people - most of whom were never evaluated for skills, English ability, or any quality other than being related to someone who was.
This is not an exaggeration. This is the documented, peer-reviewed reality of how America's immigration system works. And it is, by far, the primary driver of the numbers that have transformed the country.
The Numbers Behind the Chain
In fiscal year 2023, the United States issued 1,172,910 green cards. The breakdown reveals which factors actually determine who gets to live in America permanently:
| Category | Green Cards Issued (FY 2023) | % of Total |
|---|---|---|
| Immediate relatives of citizens (uncapped) | 551,590 | 47% |
| Family preference categories (F1-F4) | 204,240 | 17% |
| Employment-based | 196,760 | 17% |
| Diversity Visa lottery | 67,350 | 6% |
| Refugees | 59,030 | 5% |
| Asylees | 40,330 | 3% |
| Other | 53,610 | 5% |
Combined, family-based immigration - immediate relatives plus the four preference categories - accounted for 755,830 green cards. That is 64% of all permanent immigration to the United States. Employment-based immigration, the category that actually evaluates whether someone has skills America needs, accounted for just 17%.
Read that again. Roughly two-thirds of everyone who permanently settles in the United States each year is admitted not because of what they can do, but because of who they are related to.
The Immediate Relative Loophole
The most important number in the table is the one at the top: 551,590 immediate relatives. This category - spouses, minor children, and parents of U.S. citizens - has no annual cap. There is no numerical limit. However many qualify in a given year, that is how many are admitted.
This is the engine of chain migration. When Congress set a cap of 226,000 for the four family preference categories (F1 through F4), it appeared to be limiting family-based immigration. But by leaving immediate relatives uncapped, it created an open channel that grows automatically as more immigrants naturalize and petition for more relatives.
In FY 2023, uncapped immediate relatives alone exceeded the total number of employment-based immigrants by a factor of 2.8.
The Multiplier
In 2013, researchers published a study in the journal *Demographic Research*, later archived by the National Institutes of Health, that quantified what immigration analysts had long suspected: the chain migration multiplier is real, it is large, and it is growing.
The study tracked two immigrant cohorts:
- 1981-1985 cohort: 889,000 initiating immigrants generated 2.3 million additional family-sponsored migrants - a multiplier of 2.59
- 1996-2000 cohort: 979,000 initiating immigrants generated 3.4 million additional family-sponsored migrants - a multiplier of 3.45
The multiplier increased by 34% between the two cohorts. As naturalization rates improved and immigrants became more familiar with the family sponsorship system, each initial immigrant generated more subsequent arrivals.
But the national average conceals enormous variation by country of origin:
| Country of Origin | Additional Immigrants per Initial Immigrant |
|---|---|
| Mexico | 6.38 |
| Dominican Republic | 5.12 |
| Philippines | 4.97 |
| India | 4.23 |
| China | 3.87 |
| All countries (average) | 3.45 |
A single Mexican immigrant in the 1996-2000 cohort generated, on average, 6.38 additional immigrants through the family sponsorship chain. That means one initial admission eventually became more than seven people.
The mechanism is compounding. An immigrant arrives, waits five years, naturalizes, and sponsors family members. Those family members arrive, wait, naturalize, and sponsor their own family members. Each generation of the chain expands the base for the next generation. This is why the multiplier grew between cohorts - the system feeds itself.
The Backlog: A Pipeline That Never Closes
If chain migration were just a theoretical concern, the policy implications would be limited. But approximately 4 million people are currently waiting in family-based immigration backlogs at U.S. consulates abroad. They have approved petitions. They are in the queue. They are coming - the only question is when.
The wait times reveal the scale of demand the system has created:
| Category | Worldwide Wait | Mexico | Philippines |
|---|---|---|---|
| F1 (unmarried adult children of citizens) | ~8 years | ~22 years | ~15 years |
| F2B (unmarried adult children of residents) | ~6 years | ~22 years | ~10 years |
| F3 (married children of citizens) | ~15 years | ~24 years | ~22 years |
| F4 (siblings of citizens) | ~15 years | ~24 years | ~23 years |
For Mexicans applying under the F4 sibling category, the current cutoff date is approximately November 2000. People who applied 24 years ago are only now being processed. But here is the critical point: every year, new petitions are filed faster than old ones are processed. The backlog is not shrinking. It is growing.
This means the chain migration pipeline is essentially permanent. Even if Congress eliminated all family preference categories tomorrow, it would take decades to process the existing backlog. The 4 million people in the queue represent future immigration that is already committed - a demographic pipeline that is extremely difficult to shut off.
Who Actually Gets In
The skills profile of family-sponsored immigrants compared to employment-based immigrants tells the clearest story about what kind of system America is running.
Education levels at admission:
- Employment-sponsored principal applicants: 80% have a bachelor's degree or higher
- Diversity visa holders: 55% have a bachelor's or higher
- All new green card holders combined: 46%
English proficiency at admission (percentage with limited English):
- Employer-sponsored immigrants: 45% have limited English
- Siblings of U.S. citizens (F4 category): 87% have limited English
- Parents of U.S. citizens: 89% have limited English
Nearly nine out of ten parents admitted through chain migration cannot speak English proficiently. These are not people who were selected because America needed their skills. They were admitted because their adult child became a citizen and filed a petition.
And the quality gap extends beyond raw education credentials. Research from the Center for Immigration Studies found that roughly 40% of immigrants with college degrees received those degrees from foreign institutions - and on standardized literacy and computer skills assessments, foreign-educated immigrants with college degrees scored at the level of American-born high school graduates.
What Other Countries Do
The contrast with peer nations is stark. Every other major Western democracy has moved toward skills-based immigration. The United States is alone in maintaining a system dominated by family connections.
| Country | Skills-Based | Family-Based |
|---|---|---|
| Australia | 72% | 28% |
| Canada | 58% | 23% |
| United Kingdom | ~65% | ~25% |
| United States | 17% | 64% |
Australia admits 72% of its immigrants through skills-based selection - a points system that evaluates age, education, English proficiency, work experience, and whether the applicant's occupation is on a skills shortage list. Family-based immigration exists but is capped at roughly 28% of the total program.
Canada runs a similar system. Its Express Entry program scores applicants on a comprehensive ranking scale. Family reunification accounts for 23% of admissions. The remaining 58% are economic-class immigrants selected for their ability to contribute.
The United States does it backwards. 64% family, 17% skills. America admits nearly four family-sponsored immigrants for every one skills-based immigrant. Australia and Canada do the opposite.
This is not an ideological argument. It is an arithmetic one. When the majority of your immigration flow is determined by family connections rather than skills, you get a population that is, on average, less educated, less English-proficient, and less economically productive than what a skills-based system would produce. And because the family system is self-reinforcing - each admission creates new petitions for more admissions - the dynamic accelerates over time.
How the 1965 Act Built This Machine
The family preference system was not an accident. It was a deliberate policy choice embedded in the Immigration and Nationality Act of 1965.
Before 1965, the national-origins quota system established by the Immigration Act of 1924 controlled both the total number of immigrants and their countries of origin. Each country received a quota based on its share of the existing U.S. population. The system was designed to maintain demographic continuity.
The 1965 Act abolished the national-origins quotas and replaced them with a preference system that prioritized family reunification. The sponsors of the bill argued that family reunification was a humane, apolitical principle - who could be against keeping families together?
But family reunification as an immigration principle has a structural consequence that its sponsors either didn't foresee or didn't care about: it makes immigration self-selecting for source country rather than skills. When you give priority to relatives of existing immigrants, future immigration is determined by the demographics of past immigration. The system reproduces itself.
In 1965, the foreign-born population was 9.6 million, predominantly European. The family preference system initially processed mostly European applicants. But as immigrants from Latin America, Asia, and Africa arrived under the new preference categories and naturalized, they petitioned for their own families. Within a generation, the family system had completely transformed the source-country composition of American immigration - exactly as critics of the 1965 Act had warned.
The 1990 Immigration Act compounded the problem by raising the overall cap from 270,000 to 675,000 and adding the Diversity Visa lottery. But the fundamental engine - family reunification as the dominant path to permanent residency - was already built in 1965.
The Assimilation Connection
Chain migration doesn't just affect numbers. It affects assimilation.
When immigrants arrive individually or in small numbers, they have strong incentives to learn English, adopt American customs, and integrate into the broader society. There is no alternative. The surrounding community is American, and survival requires adaptation.
When immigrants arrive as part of a family chain - parents, siblings, cousins, in-laws - the incentive structure reverses. The family unit is self-sufficient. They speak the home language among themselves. They shop at ethnic grocery stores, attend ethnic houses of worship, and consume media from the home country. The surrounding American community becomes optional.
This is the connection between chain migration and the ethnic enclaves documented throughout this publication's research. The Harvard diversity study found that diverse communities have lower trust and civic participation. The Minnesota fraud investigation documented how concentrated ethnic communities develop parallel institutional structures. The elections analysis showed how unassimilated communities vote as ethnic blocs rather than as individual Americans.
Chain migration is the mechanism that creates these outcomes. It is not random individuals arriving and dispersing. It is extended family networks transplanting themselves from one country to another, maintaining their original social structure in a new location.
The 89% limited-English rate among parents admitted through chain migration is not a failure of the immigrants themselves. It is a failure of the system. A 65-year-old parent arriving from rural Mexico or Bangladesh was never going to learn English. The question is why the system admits them at all when the predictable result is a person who cannot participate in American civic life, who will depend on their children for translation and navigation, and who will live in an ethnic enclave rather than integrating into the broader society.
What Reform Looks Like
The Center for Assimilation advocates for a fundamental restructuring of the immigration system away from family reunification and toward skills-based selection. The specific reforms:
- Cap immediate relatives. The uncapped "immediate relative" category is the single largest driver of chain migration. Placing an annual cap - even a generous one - would impose the numerical discipline that currently does not exist.
- Eliminate the sibling and adult child categories (F3 and F4). No other major immigration system in the world allows citizens to sponsor adult married siblings and their families. These categories are pure chain migration with no skills component whatsoever. They should be eliminated entirely.
- Adopt a points-based system for the majority of admissions. Follow the Australian and Canadian model: evaluate applicants based on age, education, English proficiency, work experience, and skills alignment with national needs. Family connection can be one factor among many, not the dominant one.
- Process the existing backlog, then close the pipeline. The 4 million people in the family-based queue represent a commitment that will be difficult to break. But processing the backlog while simultaneously closing the categories that feed it would allow the pipeline to drain over time rather than refill indefinitely.
- Require English proficiency for all adult immigrants. If 89% of parents admitted through family sponsorship have limited English, the system is admitting people who cannot assimilate by definition. Advanced English proficiency should be a non-negotiable requirement for permanent residency.
America does not have an immigration system designed to serve American interests. It has a family reunification system designed to serve the interests of immigrants who are already here. The difference between 17% skills-based and Australia's 72% skills-based is the difference between a country that selects immigrants and a country that lets immigrants select themselves.
The chain needs to be broken. Not out of hostility to families, but out of recognition that a system which admits two-thirds of its immigrants based solely on family connections - regardless of skills, language ability, or capacity to assimilate - is a system that has abandoned any pretense of serving the national interest.
Sources
- Department of Homeland Security, Office of Homeland Security Statistics, Yearbook of Immigration Statistics, Table 6: Persons Obtaining Lawful Permanent Resident Status by Type and Major Class of Admission, FY 2023
- DHS OHSS, Lawful Permanent Residents Annual Flow Report, FY 2023
- Demographic Research / National Institutes of Health, "Family Sponsorship and Late-Age Immigration in Aging America: Revised and Expanded Estimates of Chained Migration," 2013
- Center for Immigration Studies, "Immigration Multipliers: Trends in Chain Migration," Jessica Vaughan
- USCIS, "Green Card for Family Preference Immigrants," family preference category definitions and annual caps
- Fwd.us, "Family-Based Immigration Backlogs: 5 Things to Know," backlog estimates and wait time data
- Department of State, Visa Bulletin and National Visa Center backlog reports, 2024
- Australian Department of Home Affairs, 2023-24 Migration Program Report
- Institute for Canadian Citizenship, Immigration Dashboard 2023
- Russell Sage Foundation Journal, "Do Employer-Sponsored Immigrants Fare Better Than Family-Sponsored Immigrants?"
- Center for Immigration Studies, "Foreign-Educated Immigrants Are Less Skilled Than U.S. Degree Holders"
- Immigration and Nationality Act of 1965, Public Law 89-236
- Immigration Act of 1990, Public Law 101-649