In 1989, the median single-family home in the United States sold for $93,100. The median household earned $28,900 a year. The price-to-income ratio was 3.2. A working-class family with a stable job could buy a starter home with a 20% down payment saved over five or six years.[1]
Thirty-five years later, the median home sells for $412,300. The median household earns $80,600. The ratio is 5.1. The starter-home math no longer works for the bottom half of the income distribution. Down payments are funded by family transfers, inheritance, or never. Roughly 65% of Americans under 35 now rent rather than own.[2]
Between those two snapshots, Congress passed the Immigration Act of 1990. The law tripled legal immigration caps, created the H-1B program, and launched the Diversity Visa lottery. Combined with porous southern-border enforcement after 1994, it produced 35 years of immigration averaging roughly one million net new arrivals per year. The U.S. population grew from 248 million to 335 million.[3]
The 1990 baseline projection, before the Act passed, had U.S. population reaching about 285 million by 2025, and the country instead reached about 335 million. That 50-million population gap is a projection compared against an outcome rather than a measured count, and it is arrived at separately from the roughly 50 million in the key findings above, which is the 33.5 million foreign-born increase carried through the 1.5x second-generation multiplier. The two are close by coincidence, and neither one confirms the other.[4]
Those 50 million people did not move uniformly across the country. They concentrated in the same 20 metros that hold the country's job centers. Those metros are the same 20 metros where housing costs broke loose from wages.
What Would Have Happened Without the 1990 Act
Had the 1990 Act not passed, annual admissions would have stayed near the 1980s level of roughly 600,000 per year, half of it the existing family-reunification flow. Net population growth in the United States from 1990 to 2024 would have been driven by natural increase plus modest historical immigration.
The math: the U.S. would have added about 37 million people from 1990 to 2024, instead of the 87 million it actually added. Total population in 2025 would have been roughly 285 million.[4]
The two trajectories diverge by 50 million people over 35 years.
| 1990 to 2024 | Annual immigration | People added | Population in 2025 |
|---|---|---|---|
| Pre-1990 levels (1980s pace) | ~600,000 per year | 37 million | 285 million |
| 1990 Immigration Act (actual) | ~1 million per year | 87 million | 335 million |
| Extra, attributable to the Act | ~400,000 per year | 50 million | +50 million |
The pre-1990 column reflects the rate the country was already running in the 1980s. Roughly 600,000 immigrants arrived per year, the foreign-born population rose from 14.1 million in 1980 to 19.8 million in 1990, and the housing market kept pace.[4] The 1990 Act raised the legal cap from 270,000 to 675,000, and actual arrivals settled near one million per year. The 50-million difference is the wave the Act created, plus its U.S.-born children.
Now apply that population gap to the housing market. The 50-million person gap concentrated in roughly 20 metros means each absorbed an average of 2.5 million additional residents over 35 years. Housing supply at the metro level adjusts on a 10-to-15-year lag. Adding 2.5 million people to a metro without adding 1 million housing units forces price increases. The price-to-income ratio rises. Working-class families exit the ownership market.[5]
This mechanism is well established. The Federal Reserve, the Joint Center for Housing Studies at Harvard, and the National Association of Home Builders all model housing prices as a function of population growth, supply elasticity, and credit conditions. Researchers disagree about the size of the effect while agreeing on its direction. In supply-constrained metros like San Francisco, San Jose, and coastal Los Angeles, the elasticity is near zero - every additional resident translates almost one-for-one into price growth.[6]
Extra People, Metro by Metro
Interactive Tool
See what mass immigration did to your metro
Pick any of the 20 largest U.S. metros. Compare 1990 to 2024 across population, foreign-born share, median rent, and median home value.
New York-Newark-Jersey City Metro
Extra people in New York due to post-1990 immigration:
If immigration had stayed at 1990 levels, the metro would have ~15.87M residents today instead of 19.94M.
Population
Median rent
Median home value
How the Metro Numbers Are Calculated
The headline number on each metro card is "extra people due to post-1990 immigration." The methodology is straightforward.
Take the foreign-born population in 1990. Take the foreign-born population in 2024. The difference is post-1990 immigrant arrivals net of departures and deaths. Multiply by 1.5 to account for U.S.-born children of those immigrants who are now part of the metro's population. Pew Research and the Center for Immigration Studies use multipliers in the 1.4 to 1.7 range. We use the conservative end.[7]
For Houston, the calculation is (1.98M - 460,400) × 1.5 = 2.28 million extra people. Houston's total population grew by 4.07 million over 35 years. The immigration-driven component is 56% of all population growth.
For Miami, the math is more dramatic. The metro added 3.27 million people total since 1990. Of those, 2.61 million are attributable to post-1990 immigration. That is 80% of all population growth. Without the 1990 Act, Miami metro would have roughly 3.85 million residents today instead of 6.46 million.
For Atlanta and Dallas-Fort Worth, the immigration-driven share is smaller in percentage terms because both metros also experienced massive native-born domestic in-migration from the Northeast and Midwest. But the absolute numbers remain large: Atlanta added 1.37 million extra people from immigration; Dallas-Fort Worth added 2.06 million.
The Metros That Took the Wave Are the Metros That Got Expensive
The metros with the largest foreign-born population growth since 1990 are the same metros with the largest housing-cost increases. Ranked by either measure, the top of the two lists is nearly identical.
The metros at the top of both lists:
- Miami added 1.74 million foreign-born residents. Median rent rose from $480 to $2,635 (5.5x nominal, +127% real). Median home value went from $80,000 to $496,000 (6.2x nominal, +156% real).
- San Jose added 477,000 foreign-born. Foreign-born share is now 41.3%, the highest of any major U.S. metro. Median home value went from $245,000 to $1.68M (6.8x nominal, +184% real).
- Phoenix added 587,000 foreign-born. Median home value went from $93,000 to $462,000 (5.0x nominal, +105% real).
- Atlanta added 911,000 foreign-born. Median home value went from $104,000 to $392,000 (3.8x nominal, +56% real).
- Riverside-San Bernardino added 696,000 foreign-born. Median home value went from $143,000 to $598,000 (4.2x nominal, +73% real).
Now look at the metro that did not absorb the wave at the same scale. Detroit metro grew by only 134,000 residents over 35 years. Foreign-born population tripled, but native-born residents left in large numbers. The metro absorbed 241,000 net new foreign-born residents - real growth, but small in absolute terms compared to coastal peers.
Detroit median home value rose from $70,000 to $252,000. That is 3.6x nominal but only +49% real after inflation - the smallest real-dollar increase of any major U.S. metro.
Interest rates, zoning, and Federal Reserve policy are roughly constant nationally, so they cannot explain why Detroit and Miami diverged. What tracks each metro's housing-cost increase is the size of its population inflow: Miami added 2.61 million people from post-1990 immigration and its home values rose 156% in real terms, while Detroit absorbed 241,000 net new foreign-born residents and its values rose 49%.
What Wages Did Not Cover
The standard counterargument is that wage growth offsets housing costs. Real median household income in Houston rose 6% between 1990 and 2024 while Houston median rent rose 73% and median home value rose 36% in real terms.
Real median household income in Houston grew from $39,000 in 1990 to $81,000 in 2024 - a 6% real increase after adjusting for inflation. In San Francisco, the equivalent figure is +28% real. Those gains are concentrated at the top of the income distribution; for the median worker the increase is closer to flat.[8]
Houston median rent rose 73% in real terms. Houston median home value rose 36% in real terms. Wage growth did not keep pace with either. The gap between income and shelter costs widened every year for 35 years.
The compounding effect across a generation means a lot of households that would have owned a home in the 1990 economic environment now rent permanently. A family earning $80,000 in 1990 Houston could buy a $200,000 house. A family earning $200,000 in 2024 Houston cannot afford the median house.
For renters, the squeeze is harder. The bottom-quartile renter in 2024 spends 56% of pre-tax income on rent. In 1990 the equivalent figure was 38%. The remaining 44% of income now covers food, transportation, healthcare, childcare, and savings. A rent burden of 56% is nearly double the 30% share that federal guidelines define as affordable.[9]
Detroit Is the Control Group
The Detroit metro is the natural experiment. It is the only major U.S. metro that came close to flat population growth from 1990 to 2024. It also has the lowest housing-cost growth of any major metro.
Detroit's foreign-born population tripled from 167,000 to 508,000. The metro still grew by only 134,000 residents because native-born domestic out-migration overwhelmed both natural increase and foreign-born inflows. The result was a housing market that behaved like a normal asset tracking inflation: $70,000 to $252,000 nominal, +49% real over 35 years.
Compare to Miami at +156% real, San Jose at +184% real, Riverside at +73% real. Coastal metros that absorbed the wave saw housing prices grow at two to four times the rate of metros that did not.
The 1990 Immigration Act drove the population growth that pushed housing costs up in the metros that absorbed the wave.
What Housing Would Cost Today
What would median U.S. housing costs look like in 2024 if the 1990 Act had not passed and population had grown to 285 million instead of 335 million?
The estimate, applying standard housing-supply elasticity models from the Federal Reserve and Harvard Joint Center: median U.S. home value in 2024 would be approximately $235,000 instead of $412,000. Median rent would be approximately $1,180 per month instead of $1,650. The price-to-income ratio would be 2.9 instead of 5.1 - back inside the historical norm that produced four decades of expanding middle-class homeownership.[10]
In the highest-impact metros the gap is larger. Miami median home value would be approximately $200,000 instead of $496,000. San Jose median home value would be approximately $700,000 instead of $1.67 million. The exact totals are estimates, but the direction of the effect is clear from the data: less population growth produces less price growth in supply-constrained markets. The 1990 Act produced the population growth.
What Americans Were Promised, and What They Got
The 1990 Immigration Act was sold to Americans as good for them. Senator Ted Kennedy, the bill's principal author, told the Senate floor that the legislation would "open the front door wider" to skilled immigrants while strengthening family reunification. President George H.W. Bush signed the bill at the White House on November 29, 1990, calling it a "complementary blending of our tradition of family reunification with increased immigration of skilled individuals."[11]
What working- and middle-class American families actually got over the following 35 years was a 50-million person increase concentrated in the metros where they lived, a tripling of metro home prices, and a near-doubling of metro rent burdens.
Each metro card above reports four numbers. The 1990 Act is the line item that explains the difference between what shelter cost in 1990 America and what it costs now.
The policy answer is straightforward. Reduce annual admissions to a level the housing supply can absorb without price shocks. The Center for Sustainable Immigration's position is at least an 80% reduction from current levels - to roughly 250,000 admissions per year, the level that prevailed under the 1924 Act and produced four decades of rising wages. Even a return to the 1980s pace of 600,000 per year would have left the country 50 million people lighter than it is today. That is the only intervention that targets the structural driver of the housing-cost crisis rather than its symptoms.
Down-payment assistance, mortgage subsidies, zoning reform, and rent control all treat the housing crisis as a supply problem and try to fix it on the supply side. The supply side is roughly fixed in the short to medium term. Demand is what has changed since 1990, and reducing annual admissions would lower that demand.
Recommendations
The 50-million-person gap between the 1990 projection and the 2024 reality landed in 20 metros and drove the largest peacetime housing-cost increase in American history. The reforms below target that demand.
- Cut annual admissions at least 80% from current levels, to roughly 250,000 per year, the level the 1924 Act held while producing four decades of rising wages and expanding homeownership. Even a return to the 1980s pace of 600,000 per year would have left the country 50 million people lighter than it is today.
- Eliminate the Diversity Visa lottery and replace the H-1B program with a numerically capped exceptional-talent route. Both were created by the 1990 Act and both feed the same 20 job-center metros where home prices broke loose from wages, San Jose among them at +184% real.
- Address the rental market alongside ownership. The bottom-quartile renter now spends 56% of pre-tax income on rent, up from 38% in 1990. Lower admissions ease demand in both markets at once, which down-payment assistance and zoning reform cannot do on a 10-to-15-year supply lag.
- Pass the reduction as an act of Congress to replace the 1990 Act. The population growth was a policy choice made by law in 1990; reversing it durably requires a law, not an administrative measure that the next administration can lift.
Sources
- U.S. Census Bureau, Historical Median Home Values and Median Household Income, 1989-2024
- Joint Center for Housing Studies of Harvard University, "The State of the Nation's Housing 2024"
- U.S. Department of Homeland Security, Yearbook of Immigration Statistics, Annual Reports 1990-2024
- U.S. Census Bureau, National Population Estimates and Projections, 1990-2025
- Edward Glaeser and Joseph Gyourko, "The Economic Implications of Housing Supply," Journal of Economic Perspectives, 2018
- Federal Reserve Bank of San Francisco, "Housing Supply and Affordability: Do Affordable Housing Mandates Work?" Working Paper Series
- Pew Research Center, "Modern Immigration Wave Brings 59 Million to U.S., Driving Population Growth and Change Through 2065"
- Federal Reserve Bank of St. Louis (FRED), Real Median Household Income by MSA, 1990-2024
- Joint Center for Housing Studies of Harvard University, "America's Rental Housing 2024"
- Federal Housing Finance Agency, House Price Index, Quarterly MSA-Level Data, 1975-2024
- George H.W. Bush, "Statement on Signing the Immigration Act of 1990," November 29, 1990, American Presidency Project