Mass Immigration by the Numbers

The charts and maps below show the scale and impacts of mass immigration in the United States.

15.8%

Foreign-born share of the U.S. population

The share was 7.9% in 1990.

+33.5M

Foreign-born residents, 1990 to 2025

There were 19.8M foreign-born residents in 1990. By 2025, that figure was 53.3M.

5.0×

Median home sale price ÷ median household income

2024 national ratio, up from 4.1× in 1990 and 2.7× in 1970 (+86% since 1970).

+70%

Real median rent, 1970 to 2024

$873 became $1,487 in 2024 dollars. The increase since 1990 was 39%.

+26.4%

Real median household income, 1990 to 2023

$65,440 became $82,690 in constant dollars. Housing costs rose faster.

6.1×

Home value ÷ income in metros above 20% foreign-born

ACS 2024 average, compared with 3.2× in metros under 5% foreign-born.

1

A pattern

Periods of growth and reduction

The foreign-born share of the U.S. population has fluctuated dramatically over time. After a period of high immigration to urban areas at the start of the 20th century, the 1924 Immigration Act cut admissions by 80%. This led to a significant reduction in the foreign-born population, which started changing after the 1965 Hart-Celler Act and accelerating after the 1990 Act.

Foreign-born share of the U.S. population, 1850-2025

Periods of High and Low Immigration Levels

Before 1890, most immigrants were settlers, spreading out across the country. The 1890-1924 urban wave brought in 20 million Italian, Eastern European, and Jewish immigrants to large metro areas. The landmark 1924 Immigration Act cut admissions by ~80%, leading to a 40-year period of relatively low immigration. By 1970, the foreign-born population fell to a low of 4.7%. The 1965 Hart-Celler Act reopened the door and increased immigration, before the 1990 Act significantly accelerated it. The foreign-born population has climbed back to 15.8% as of January 2025.

Rising, and rising againThe 1924-1965 restrictionReopening, 1965-1990
0%4%8%12%16%9.7%14.8%4.7%15.8%185018901930197020102025Settlement era1850-1890First urban wave1890-1924The pause1924-1965Slow reopening1965-1990Second urban wave1990-present

Decades of the Same Policy

We've had the same immigration policy in the United States since 1991, when legal inflows were increased to over 1 milltion per year and the H1-B program was established.

Green Cards Granted per Year

Permanent-resident status granted that year; roughly half are adjustments for people already in the United States.

1.6M1.1M0.5M0
1.54M
1990
720K
1995
841K
2000
1.05M
2007
1.05M
2015
1.03M
2019
1.17M
2023
1.32M
2025

Illegal Immigrant Population (millions)

How many people are living here illegally. Nobody publishes how many entered in a given year, so this is the measure that counts people.

151050
3.5
1990
5.7
1995
8.6
2000
12.2
2007
11
2015
10.2
2019
14
2023
14
2025

Pew Research Center, residual estimates, one consistent basis.

Foreign-Born Population (millions)

Everyone born outside the United States who lives here, whatever their status. A stock, so it carries the accumulated effect of every year of arrivals.

6040200
10.3
1950
9.7
1960
9.6
1970
14.1
1980
19.8
1990
31.1
2000
40
2010
50.2
2024
53.3
2025
1990 Immigration Act
Pre-1990
Post-1990
+33.5 Million

Foreign-born residents added since the 1990 Immigration Act

Over the same period the median American home went from costing 3.0 years of median household income to 5.6. A middle-class family that could buy on one income in 1990 now needs nearly twice as many years of earnings for the same house.

2

The scale

33.5 million people, congregated in specific metro areas

The U.S. has 53.3 million foreign-born residents as of January 2025, and the immigration stock sits at about 93 million in 2024, roughly 28% of the country. About 64% of all foreign-born now live in just 20 metro areas. These areas have grown at a far faster rate - and become significantly more expensive - than the rest of the country.

Metro area population growth since 1990

Twenty metros hold about 65% of the country's foreign-born

Foreign-born population growthTotal population growth
Dallas-Fort Worth4.4M1.4M4.2x
Houston4M1.5M4.0x
Atlanta3.3M911K4.3x
New York3.1M2.7M6.5x
Phoenix2.9M587K5.2x
Miami2.4M1.6M6.3x
Washington, DC2.3M1.1M4.8x
Inland Empire (CA)2.2M696K6.4x
Los Angeles1.7M848K9.4x
Seattle1.6M712K6.6x
Denver1.4M329K5.8x
Tampa1.4M407K5.0x
Source
3

Housing

Priced out of a home

A median home ran about four years of household income through the 1990s. It costs five now, and across 109 of the 111 metros over 500,000 people the ratio rises with a metro's immigrant share.

Cost of a U.S. home, relative to income

A home costs five years of income now, and more where immigration concentrated

Nationally a median home ran about 4x median household income through the 1990s and about 5x today. The comparison below uses the median metro in each foreign-born-share band, across 109 metro areas with more than 500,000 residents.

3.11xMedian home value / income under 5% foreign-born
6.42xMedian home value / income at 25%+ foreign-born
MedianEvery band statistic

By year, nationally

Median sale price divided by median household income. It held near 4x through the 1990s, then stepped up after 2000 and has not come back down.

1990 Immigration Act
3.3x
3.7x
3.6x
4.1x
3.9x
4.0x
5.1x
4.5x
5.2x
4.8x
5.0x
1975
1980
1985
1990
1995
2000
2005
2010
2015
2020
2024

Sorted into seven foreign-born-share bands

Each color marks a band from lower to higher foreign-born share. The bars show median home value relative to median household income; the table adds median annual rent relative to that income.

Median home value / median household income

Under 5% foreign-born3.11x
5 to 7% foreign-born3.81x
7 to 10% foreign-born3.82x
10 to 15% foreign-born4.30x
15 to 20% foreign-born4.82x
20 to 25% foreign-born5.64x
25%+ foreign-born6.42x
BandMetrosHome / incomeRent / income
Under 5%43.11x18.8%
5 to 7%173.81x19.7%
7 to 10%273.82x19.1%
10 to 15%254.30x21.3%
15 to 20%114.82x21.4%
20 to 25%175.64x24.2%
25%+86.42x22.0%

Mass Immigration = Higher Home Prices

109 of the 111 U.S. metro areas over 500,000 people.

3.2xMedian home price to income in metros under 5% foreign-born
6.1xMedian home price to income in metros over 20% foreign-born
2x4x6x8x10x0%10%20%30%40%Foreign-born share of metro populationMedian home value ÷ median household income
r = 0.69
4

Jobs and wages

The raise is real. It does not survive the rent.

0%. 9%: local prices take two thirds of the difference. Across the same 109 metros, moving that way lifts median household income about 23% and lifts the median home price about 97%. Pay is broadly similar across the country. Housing is not, and that gap is where a wage goes. The occupational split below shows where the added workers landed, which is the jobs that never required a degree.

Foreign-born share of each occupation group, 2025

The competition is heaviest in the jobs that never needed a degree

Foreign-born workers are 19.2% of everyone employed. In construction, maintenance and farm work they are 28.5%. In service jobs, 24.4%. On factory floors, in warehouses and behind the wheel, 23.9%. In management and professional work they are 16.2%, below their share of the workforce as a whole. These are the only five groups the Labor Department publishes by nativity, and each one is listed with the jobs it actually contains.

Above the 19.2% national averageBelow it
0%8%16%24%32%Construction, trades, farm workBuilders, electricians, plumbers, mechanics, farm work28.5%4.2M of 14.6MService jobsCooks, cleaners, home health aides, childcare, security24.4%6.6M of 27.3MFactory, warehouse, truckingMachine operators, truck and delivery drivers23.9%4.7M of 19.8MManagement and professionalSoftware, engineers, teachers, nurses, doctors16.2%11.5M of 71.3MSales and officeSales, cashiers, clerical and administrative support13.9%4.3M of 30.6MAll occupations 19.2%
5

The H-1B pipeline

No longer representing "the best of the best"

When the H-1B Visa was introduced in 1990, it was positioned as a narrow way to fill specialty occupations no American could fill. It replaced the existing H-1 Visa, which was a 'best of the best' test requiring distinguished merit and ability, read by the immigration courts as preeminence in the field, and which carried no numerical cap at all. H-1B in practice only requires a bachelor's degree and is a way to bring in millions of Indian immigrants to compete with Americans for IT and engineering jobs.

The cap against the reality

The 85,000 cap never governed how many H-1B workers are here

The cap counts only new cap-subject petitions. Renewals and cap-exempt employers are not counted against it, so the number of people in the country was never governed by the number Congress set.

The cap Congress set on new visas
85,000
Petitions approved in 2025
406,348
H-1B workers living here now
730,000
Including spouses and children
1,300,000

Petitions approved each year

Every year since 1999 has run above the cap. Approvals count a worker again each time the petition is renewed.

Approved2026, projected85,000 cap
0100K200K300K400K85,000 cap on new visas199220002008201620222026

The cap was never the constraint. What the number of visas cannot show is who receives them and what they are paid, and those are the two facts that decide whether the program is filling a gap or lowering a wage.

H-1B approvals, origin and pay, 2025

70% from one country, 83% below the local median wage, 72% renewals

The 1990 Act replaced the H-1 visa, which required "distinguished merit and ability", with the H-1B, which requires a bachelor's degree. USCIS approved 406,348 petitions under it in 2025. Three questions decide whether it is bringing in talent the country cannot find: who gets the visas, what they are paid, and whether they are new workers at all.

1. Who gets them. Country of birth, 2025

IndiaEvery other country
0%20%40%60%India283,77269.9%China49,16112.1%Philippines5,5461.4%Canada4,3821.1%South Korea4,1951%Mexico3,3900.8%Pakistan3,3800.8%Taiwan3,1610.8%Brazil2,8760.7%Nigeria2,4200.6%All other countries43,62810.8%

2. What they are paid. Wage level of new H-1B jobs, five-year average 2020-2024

83% pay below the local median wage for the job28%55%12%5%local median wageShare of cap-subject petitions
Level I17th percentile · Below local median
Level II34th percentile · Below local median
Level III50th percentile · At local median
Level IV67th percentile · Above local median

3. Whether they are new workers. 406,348 approvals against an 85,000 cap, 2025

28% new72% extensions, amendments and job changes for workers already here85,000 annual cap65,000 plus 20,000 for U.S. master's degreesTotal petitions approved: 406,348Median salary offered: $133,000
6

Assimilation

The other half of the question

Assimilation is directly related to how many immigrants are coming in. When fewer entries are allowed, assimilation is more likely, as immigrants have to adapt to the broader society around them. If you move to an area with hundreds of thousands of people who speak your language and share your customs, you are unlikely to adapt to the American way of life. Language is the most direct measure of that, and it is the one the country stopped collecting for forty years. The modern question dates only to 1980, so there is no comparable reading from the restriction era to measure today against. What can be measured is the present: about 47% of immigrant adults speak English less than very well, and the naturalization test asks for a vocabulary far smaller than the ones other countries require.

English ability among immigrant adults, 2023

Almost half cannot hold a conversation in English

47% of immigrant adults, about 22.3 million people, speak English less than very well. They account for 81% of everyone in the country with limited English. The barrier follows them into the doctor's office, the job market, and the checkout line.

How well immigrant adults speak English

53%
15%
19%
13%
Speaks only English, or speaks it very well53%
Speaks English well15%
Speaks English not well19%
Speaks no English at all13%

47% have limited English, about 22.3 million adults

81% of every adult in the United States who speaks English less than very well is an immigrant.

Citizenship Language Requirements: Vocabulary Comparison

Approximate vocabulary required for citizenship, mapped to CEFR levels

United States~A1 or below
78 words
Canada~A2 (CLB 4)
1,500 words
Speaking/listening only
Australia~A2-B1
2,000 words
No formal test
NetherlandsA2-B1
2,000 words
B1 target under 2021 Act
GermanyB1
2,500 words
C1 for fast-track
NorwayB1 (oral)
2,500 words
Test-based since 2025
SwitzerlandB1 / A2
2,500 words
Cantons can require more
AustriaB1 → B2
3,500 words
Moving to B2 by 2029
FranceB2
5,000 words
Raised from B1 in 2025
DenmarkB2
5,000 words
Strictest in Europe

The U.S. requires 78 words for its citizenship test, while Denmark requires ~5,000.

A typical B1 test, which most of Europe requires as a minimum, demands 30-40x more vocabulary than the U.S. naturalization test.

78
U.S. vocabulary list
93%
U.S. pass rate
5,000
Denmark vocabulary
0
Countries lowering bars
7

Current Changes

Where we're at now

The Trump Administration has been making progress through executive measures, namely:

  • Net immigration has fallen from a 2.7 million peak in 2024 to a projected 570,000 in 2026, a drop of roughly 80% and the sharpest two-year fall in the Census Bureau's series.
  • It was done through executive authority: border enforcement, the country restrictions, and a rebuilt vetting and visa-processing regime.
  • The housing market has already turned. Asking rents have fallen year over year for 28 straight months, the national multifamily vacancy rate hit a record 7.4%, and Zillow reports rent affordability at a four-year high.
  • The sharpest drops are in the metros that took the most arrivals. Every other section on this page describes a cost that rising demand imposed; this is the same mechanism running backwards.
  • What remains is to hold the number there and write it into law, so the next Congress inherits a number rather than a policy that ends with the administration that issued it.

U.S. net international migration

Net migration is down 83% from its 2023 peak

Net migration averaged about 920,000 a year in the 2010s, peaked at 3.3 million in 2023, and fell to 1.3 million in 2025. For 2026 the CBO projects about 570,000 and the Census Bureau about 321,000. The first two bars are decade averages; the rest are single years. Hover any bar for detail.

Decade averageSurge yearsAfter the fallProjected
~1.0M
~920K
2.8M
3.3M
2.7M
1.3M
~570K
2000s
avg/yr
2010-19
avg/yr
2022
2023
Peak
2024
2025
2026
projected

The full analysis is in the articles

Every figure on this page is drawn from Census, BLS, and other government data, and each article works through one part of it in detail.

Sources

  1. Sources: U.S. Bureau of Labor Statistics, Foreign-Born Workers: Labor Force Characteristics, 2025, for how foreign-born and native-born workers are distributed across the five broad occupation groups; Current Population Survey annual averages for employment by nativity, 31.3 million and 132.2 million. The share in each group is the foreign-born count divided by the total, computed from those two published figures. The Labor Department built the 2025 annual averages from eleven months because of the federal government shutdown that October, and notes they are not strictly comparable with other years. The five groups above are the only occupational split the Labor Department publishes by nativity; the jobs listed beside each bar are the detailed occupations it rolls into that group.
  2. Sources: Migration Policy Institute; Census Bureau Historical Data; DHS Yearbook
  3. Sources: USCIS, Characteristics of H-1B Specialty Occupation Workers, 2025 annual report to Congress (April 2026), for country of birth, the 406,348 approvals, the 28%-72% split between new and continuing employment, and the $133,000 median. Wage levels are the Department of Homeland Security's own figures in the weighted-selection final rule, 90 Fed. Reg. 60864 (Dec. 29, 2025), Table 12, a five-year average of cap-subject petition receipts for 2020 through 2024. The Labor Department sets Level I at about the 17th percentile of the local wage for the occupation, Level II at the 34th, Level III at the 50th, and Level IV at the 67th. Universities and nonprofit research employers are exempt from the 85,000 cap, which is why approvals run far above it.
  4. Total approvals include renewals and cap-exempt employers (universities, non-profits), which are not capped, so the total runs well above the 85,000 cap on new visas (65,000 plus 20,000 reserved for U.S. advanced-degree holders). The program began in 1992. Approvals fell to 328,185 in 2025; the 2026 figure annualizes the 47,821 approvals recorded in October-December 2025 and is an estimate, not reported data. Sources: USCIS H-1B Employer Data Hub and cap history; Pew Research Center (March 2025); FWD.us analysis of American Community Survey data for the resident H-1B population, which USCIS does not publish directly.
  5. By year: median sales price of houses sold (U.S. Census Bureau and HUD, FRED series MSPUS, annual average of the four quarters) divided by median household income (Census Current Population Survey). One consistent basis across every point from 1975 to 2024; nothing is spliced. FRED's income series begins in 1984, so the 1975 and 1980 points are taken from Census Table H-6, which carries the same household series back to 1975. That is as far back as it goes: earlier years exist only for family income, a different and lower basis. By metro: American Community Survey 2024 one-year estimates, median home VALUE divided by median household income across all 109 metros over 500,000 people. Value-based ratios run about a point below sale-price ratios, so a metro reading 5.2x is not directly comparable with the national 5.0x above it.
  6. Source: U.S. Census Bureau, Historical Census of Housing, homeownership rates, 1900-2020, with the 2024 rate from the Housing Vacancies and Homeownership survey.
  7. Bubbles show foreign-born population for the 30 U.S. metro areas with the largest foreign-born populations, U.S. Census Bureau ACS 2024 1-year: each metro's foreign-born share applied to its population, one basis for all thirty. Those 30 metros hold about 81% of the foreign-born counted across the 109 metro areas over 500,000 people. The national total and its growth from 1990 to January 2025 are from the U.S. Census decennial counts, ACS, and CIS analysis of the Current Population Survey. Immigrant stock including U.S.-born children is anchored to Migration Policy Institute's 2024 estimate of about 93M (51.9M foreign-born plus 41.1M U.S.-born children). Because no annual stock series exists, the animation applies measured first-plus-second-generation ratios from Pew's historical series to this map's foreign-born series and interpolates between benchmark years; the animated stock is therefore an estimate, not a Census count. The three largest countries of birth in each popup are ACS 2024 table B05006, via the Census Reporter API: one-year for 26 of the 30 metro areas, the same vintage as the bubble counts, and the 2024 five-year table for Austin, San Antonio, Providence and Fresno, which do not publish country-level place of birth at one-year. The table's regional roll-ups ("Caribbean", "Eastern Africa") and its residual "Other" rows are excluded before ranking, so every entry is a country.
  8. Green Cards granted: DHS Yearbook of Immigration Statistics 2024, Table 1, plus the Legal Immigration and Adjustment of Status Report for 2025. Apprehensions: U.S. Border Patrol Total Apprehensions FY1925-FY2020 and CBP nationwide encounters from 2021. The 2020-2023 figures include Title 42 expulsions; 2026 runs through July. Foreign-born population: U.S. Census Bureau decennial census and American Community Survey.
  9. U.S. Census Bureau net international migration estimates (2000s and 2010s annual averages, and 2022 through 2025); Congressional Budget Office, The Demographic Outlook 2026-2056, for the 2026 projection of about 570,000, against the Census Bureau’s about 321,000. Net migration counts everyone entering minus everyone leaving, in every category including temporary visas and illegal entry. It is not Green Cards granted and not border encounters.
  10. Sources: U.S. Census Bureau, American Community Survey 2023, for the distribution of English ability among immigrant adults and the 22.3 million total; KFF/LA Times Survey of Immigrants (2023) for the reported difficulties, insurance coverage, and self-reported health. Both series count adults aged 18 and over born outside the United States.
  11. Sources: Council of Europe CEFR descriptors; USCIS Reading & Writing Vocabulary Lists; national immigration authorities
  12. Population, 1990: U.S. Census Bureau, County Population Estimates 1990-2000 (CO-99-8), the April 1 1990 census base, summed over the counties of each metro's 2023 OMB delineation so both years sit on the same footprint. Population and foreign-born, 2024: American Community Survey 2024 1-year estimates, at metropolitan statistical area level. Foreign-born, 1990: U.S. Census Bureau Working Paper POP-twps0029 (Gibson and Lennon, 1999), Table 16, "Nativity of the Population for Metropolitan Areas With 500,000 or More Population: 1990", summed over the 1990 metro areas that make up the modern footprint. The native-born column is the difference between the two measured series, which makes it exactly the change in the native-born population: total 2024 less foreign-born 2024, against the same subtraction for 1990. Census counts as native anyone born in the United States, Puerto Rico or a U.S. Island Area, or abroad to a U.S.-citizen parent. Counties that were rural in 1990 sit outside the 1990 nativity table, which affects Atlanta (8.1% of its modern population), Phoenix (5.2%), Minneapolis (4.5%) and San Jose (2.4%, San Benito County) and slightly overstates their foreign-born growth. Median home price to income is ACS 2024 1-year median home value divided by median household income, the same basis as the metro scatter; it runs about 1.2 points below the national median SALE-price ratio and is not comparable to it.
  13. U.S. Census Bureau, American Community Survey 2024 1-year (home value, income, rent, foreign-born); Census population estimates (1990, 2025). Home price-to-income here is the ACS home-VALUE basis, which runs about 1.2 points below the median sale-price basis used for the national figure and is not comparable to it. Two of the 111 metros over 500,000 people are absent: Killeen, TX and Springfield, MO, which lack matched home-value and income data in this release, leaving 109.