Colorado non-citizen households used welfare at a far higher rate than the U.S.-born.

By Shaina Cole | Contributing Writer, Rocky Mountain Voice

Half of Colorado households headed by a non-citizen, regardless of legal status, either used a traditional welfare program or qualified for federal tax credits aimed at low-wage workers. That is the finding of a new national report the Center for Immigration Studies released June 11. It is built on five years of Census Bureau survey data, pooled to get state samples large enough to be meaningful.

The numbers come at a moment when Colorado is still absorbing the fiscal fallout from its own immigrant health coverage program, one that was projected to cost $27 million and enrolled more than eight times as many people as the legislature anticipated. And they arrive just weeks after Colorado began implementing federal food-stamp restrictions that ended SNAP eligibility for many non-citizens.

The data offers the clearest picture yet of what non-citizen households in Colorado were accessing — and what is now being cut.

What the data shows

CIS drew on five years of the Census Bureau’s Current Population Survey, 2021 through 2025, asking respondents about program use over the prior calendar year. The combined sample was large enough to analyze Colorado separately. The result was that 42 percent of non-citizen-headed households in the state used at least one traditional welfare program, against 25 percent of U.S.-born households. 

When eligibility for the Earned Income Tax Credit or Additional Child Tax Credit is included, refundable credits for low-wage workers who may owe little or no federal income tax, the non-citizen rate hits 50 percent versus 30 percent for U.S.-born households.

Within traditional welfare, food assistance showed the sharpest split. 

Non-citizen households used SNAP, WIC or free school meals at a rate of 30 percent, more than double the 14 percent rate for U.S.-born households. Health coverage followed the same pattern. Non-citizen households enrolled in Medicaid were at 30 percent compared to U.S.-born households at 17 percent.

Cash assistance is a different story. 

Non-citizen households used programs like TANF and Supplemental Security Income at a lower rate than U.S.-born households, 2.8 percent versus 3.6 percent. That is largely a function of eligibility. The 1996 federal welfare reform law sharply restricted non-citizen access to both programs, though eligibility varies by immigration status. The welfare gap in Colorado runs through food and health coverage, not cash.

One number that might seem counterintuitive: 89 percent of Colorado non-citizen households have at least one person working. That is higher than the 76 percent rate for U.S.-born households. Yet 42 percent of those working non-citizen households still used traditional welfare programs — nearly double the 23 percent rate for working U.S.-born households.

CIS researchers point to education as the explanation. 

Non-citizen household heads in Colorado averaged 12.5 years of schooling, roughly a high school diploma, compared to 14.8 years for U.S.-born household heads, closer to an associate’s degree. Lower education tends to mean lower wages. Lower wages tend to mean program eligibility, even for households where someone is working full time.

The survey counts a household as a welfare user if anyone in it receives a benefit, including U.S.-born children of non-citizen parents. 

Critics of household-level comparisons, including the libertarian Cato Institute, argue that immigrant welfare use looks different when measured per person and by benefit value rather than by whether anyone in a household received a benefit. 

CIS responds that the cost to taxpayers is real regardless of which household member technically receives the benefit.

Colorado’s numbers also sit in the middle of the national distribution. 

Nationally, 47 percent of non-citizen households use traditional welfare, compared to Colorado’s 42 percent. The state ranks 31st out of 50 states and Washington, D.C. Colorado is not among the highest, but well above the U.S.-born rate in food assistance and Medicaid, and above it in housing.

What Colorado built — and what it cost

Colorado has been running its own immigration-related health program since January 2025, and its cost trajectory explains a lot about where the state’s Medicaid budget went.

Cover All Coloradans was created by the legislature in 2022 to extend Medicaid-equivalent health coverage to children, and pregnant or postpartum women, who would otherwise qualify for Medicaid or CHP+, Colorado’s children’s health insurance program, except for their immigration status. 

Undocumented immigrants are generally barred from Medicaid under federal law. Cover All Coloradans was Colorado’s workaround.

Legislative staff initially assumed roughly 3,600 enrollees and projected about $27 million in FY 2025-26 HCPF costs tied to the bill, the first full fiscal year after the January 2025 coverage expansion took effect.

Neither figure held.

By early 2026, nearly 30,000 women and children had enrolled, against the 3,600 the fiscal note assumed. Even against the health department‘s own later projections, enrollment came in “nearly three times higher” than the department had forecast. 

Without policy changes, the program was headed toward a cost of $127 million in the coming fiscal year, according to Colorado Politics. That is nearly five times the $27 million projection. 

House Speaker Julie McCluskie did not dispute the numbers. “Clearly, enrollment in Cover All Coloradans has been significantly higher than nonpartisan staff predicted,” she told Colorado Politics. “It’s important to note that this covered population is entirely pregnant women, new mothers and young children. I support the steps the JBC has taken to cut spending in this program, but eliminating it entirely would simply shift the cost of providing this care onto consumers, providers and private insurance plans,” she said.

In a statement to Colorado Politics, an HCPF spokesperson explained the miscalculation. Officials had assumed this population would be “relatively healthier than the traditional Medicaid children population, based on early data from a similar program in Oregon and empirical evidence from academic research that immigrants are a generally healthier population than non-immigrants.” That assumption did not hold in practice.

The legislature responded this spring. House Bill 26-1411, passed April 28, capped children’s enrollment at 25,000, imposed a $1,100 annual dental limit beginning July 1, 2026 — raised from $750 by a conference committee — and, beginning January 1, 2027, shifted behavioral health services to fee-for-service only while removing accountable care and managed care coverage.

What changes now

The CIS surveys cover 2021 through 2025. Two significant policy changes took effect after that window closed.

Colorado began implementing new federal SNAP restrictions in the spring, with refugees, asylees, parolees and most other non-U.S. citizens losing eligibility beginning May 1, 2026 under the One Big Beautiful Bill, the federal budget legislation signed into law last year. 

The Cover All Coloradans benefit reductions will work their way into future health coverage surveys as well.

That makes the CIS figures a baseline, not a current snapshot. They document what Colorado’s non-citizen households were accessing before the two most significant cuts to non-citizen benefits in recent years. 

Future survey cycles will show lower numbers. How much lower depends on how many non-citizens lose eligibility, whether states backfill federal cuts with their own dollars and whether programs like Cover All Coloradans continue to shrink.

Colorado’s Medicaid spending reached an estimated $16 billion in 2025, a 101 percent increase from 2015, even as enrollment returned to near-2015 levels. 

The Common Sense Institute, a Denver-based fiscal research organization, attributed $858 million in annual cost growth to 182 health care bills the legislature passed since 2019, a figure that includes federal matching funds covering about 36 percent of the total.

Cover All Coloradans was one of them.