No — averaged across states over the last decade, immigrants have been a net fiscal positive at the federal level and roughly a wash-to-modest-net-cost at the state/local level, while consistently adding more to the economy (GDP, labor force, growth) than they subtract in service costs

Summary

The honest answer is it depends on which government's ledger you read and which immigrant generation you count - the question as posed ("have immigrants cost local services more than they've added to the economy") conflates two different accounts that the research keeps separate. The consistent finding across the last decade of economics and government literature (National Academies 2016/2017, CBO 2024, Cato 2024, Manhattan Institute 2025) is: (1) at the state and local level, immigrants - mainly first-generation, working-age arrivals with children in public schools - typically consume somewhat more in services (especially K-12 education and, for some categories, Medicaid) than they pay in state/local taxes; (2) at the federal level and for the economy overall, immigrants are consistently net-positive - they expand GDP, grow the labor force, and (via payroll and income taxes) reduce federal deficits; and (3) the state/local net-cost figure reverses within a generation - the children and grandchildren of immigrants are large net fiscal contributors. So "cost more than they added" is true of one narrow slice (first-generation immigrants' state/local balance sheet) and false of the whole picture (the economy, the federal budget, and later generations). An AskAmerica-computed check across all 50 states + DC over 2013-2023 found essentially no relationship (r = -0.14, not statistically distinguishable from zero) between how fast a state's foreign-born share grew and how fast its real GDP per capita grew - consistent with the literature's finding that immigration's effects are real but small relative to other drivers of state economic performance, and do not show up as a simple cross-state penalty or bonus.

1. What "cost" means depends entirely on which government you're asking about

The National Academies of Sciences' 2016 report The Economic and Fiscal Consequences of Immigration - the most comprehensive nonpartisan study of this question, using pooled 2011-2013 Current Population Survey data matched to the Census of Governments' state/local finance data - found a structural mismatch: immigration's fiscal impact is negative at the state and local level but positive at the federal level. First-generation immigrants and their dependent children imposed a net -$57.4 billion/year average cost on state and local governments (2011-2013 average), driven overwhelmingly by K-12 education costs for immigrants' children, who as a group have larger households and lower average incomes than third-plus-generation residents. But second-generation individuals (the U.S.-born children of immigrants, once grown) contributed +$30.5 billion/year, and third-generation-plus individuals (which include immigrants' grandchildren) contributed +$223.8 billion/year - because the report attributed each generation's costs and taxes as adults, immigrant generations "pay forward" the state/local cost of their own childhoods. (NAS Ch. 9, State and Local Fiscal Effects of Immigration)

The Congressional Budget Office's 2024 analysis of the 2021-2026 immigration surge reached the same structural conclusion for a very different period and immigrant population: "Research has generally found that increases in immigration raise state and local governments' costs more than their revenues, and CBO expects that finding to hold in the case of the current immigration surge." But at the same time, CBO projected the surge would raise nominal GDP by $8.9 trillion over 2024-2034, increase federal revenues by $1.2 trillion, and reduce federal deficits on net by $0.9 trillion over that decade - because immigrants pay federal payroll and income taxes immediately upon working (often without collecting Social Security/Medicare benefits for decades, if ever, in the case of unauthorized workers) while federal outlays for them rise much more slowly. (CBO, Effects of the Immigration Surge on the Federal Budget and the Economy, 2024)

2. State and local governments do receive real tax revenue from immigrants, including undocumented immigrants - this is separate from the net-cost question

It's important not to read "state/local net cost" as "immigrants pay nothing to state and local governments." The Institute on Taxation and Economic Policy estimates that in 2022, undocumented immigrants alone paid $37.3 billion in state and local taxes nationally (sales/excise, property, and income taxes), with six states each collecting over $1 billion: California ($8.5B), Texas ($4.9B), New York ($3.1B), Florida ($1.8B), Illinois ($1.5B), and New Jersey ($1.3B). ITEP also estimates this figure would rise by roughly 20% if these workers were granted legal work authorization, since much undocumented income currently goes unreported for payroll tax purposes. (ITEP, State and Local Tax Contributions by Undocumented Immigrants, 2024) This is a real, computed tax-paid figure, not a net-of-cost estimate - it says nothing by itself about whether that $37.3B exceeds or falls short of the services that population consumes; that comparison is exactly what the NAS/CBO studies above attempt, with more assumptions and more room for disagreement.

3. Contested cost estimates exist, and one of the highest-profile ones does not hold up

The Federation for American Immigration Reform (FAIR), an immigration-restriction advocacy group, has published estimates putting the annual taxpayer cost of illegal immigration as high as $151 billion (2023 study). These figures are widely disputed by economists across the political spectrum. The Cato Institute's review found FAIR's methodology "rejected by all economists who work on this subject," citing inflated population counts (including millions of U.S. citizens misclassified as unauthorized), undercounted tax revenue, and an assumption that immigrants use every public good (national defense, interest on the debt) at the same absolute cost as natives rather than at the marginal cost of one more person. Cato calculates that correcting just the population-count and tax-rate errors lowers FAIR's net-cost figure by 87-97%, to $15.6 billion or $3.3 billion depending on assumption. (Cato, "FAIR's 'Fiscal Burden' Study Is Fatally Flawed") (FAIR, 2023 Cost Study, for the original claim)

On the other side, Cato's own longer-run historical estimate finds that over 1994-2023, immigrants (across all legal statuses, all levels of government combined) generated a cumulative fiscal surplus of $14.5 trillion in 2024 dollars, including $3.9 trillion in reduced federal interest costs from a smaller national debt than would otherwise exist. (Cato, Immigrants' Recent Effects on Government Budgets: 1994-2023) The gap between these estimates - roughly $150B/year net cost vs. a multi-trillion-dollar 30-year net surplus - illustrates how sensitive this question is to methodology (marginal vs. average public-goods cost, which government level, whose descendants count, what time horizon), not a genuine empirical disagreement about the same measured quantity.

4. Education level and age at arrival, not immigration status per se, drive whether an individual immigrant is a net fiscal contributor

The Manhattan Institute's October 2025 update - built on a CBO-comparable 10-/30-year fiscal-scoring methodology - finds that the direction of an individual immigrant's lifetime fiscal effect depends far more on education and age at arrival than on immigration category. Immigrants with a bachelor's degree or higher, especially those who arrive before age 40, produce large fiscal surpluses over both 10- and 30-year windows; immigrants with a high school diploma or less, and their descendants, tend to receive more in benefits than they pay in taxes over the same windows. Legal-status categories matter too: employment-based immigrants are the most fiscally positive category, while the average unauthorized immigrant still grows GDP but adds less to the deficit reduction than a legal immigrant with the same education level. (Manhattan Institute, The Fiscal Impact of Immigration: 2025 Update) This finding — that skill composition, not immigrant status, is the main driver — is broadly consistent with the NAS report's identical conclusion nearly a decade earlier.

5. An AskAmerica-computed check: does a state's changing immigrant share track its economic growth?

None of the studies above are broken out at fine state grain over a full ten-year panel, so as an additive check we computed, for all 50 states + DC, the change in foreign-born population share (Census ACS 5-year, 2013 vintage vs. 2023 vintage) against the growth in real GDP per capita (BEA SAGDP1, chained 2017 dollars) over the same window. The correlation is r = -0.14 (r² = 0.02, n = 51) - a weak negative slope, not distinguishable from zero at this sample size, and not evidence that states absorbing more immigrants grew their per-capita economies more slowly (or faster). This null result is itself informative: it means the state/local fiscal-cost finding from NAS/CBO is not visible as a simple cross-state growth penalty - immigration's net effects operate through channels (specific service lines like K-12 education and Medicaid, specific labor-market segments) that a state-level GDP-per-capita comparison is too coarse to detect, and are swamped by every other driver of state economic growth (industry mix, housing supply, tech-sector concentration, etc.). This is a descriptive correlation, not a causal test, and a 51-unit cross-section cannot separate immigration's effect from these confounds - it only shows that no crude aggregate penalty is visible in the data.

6. Bottom line

Over the last decade, immigrants have consistently added more to the U.S. economy overall - GDP, labor force size, federal tax revenue - than they have cost federal, state, and local governments combined, when following any of the mainstream, peer-reviewed or government-agency methodologies (NAS, CBO, Cato's historical estimate, Manhattan Institute). The one place where "cost more than added" is a defensible reading of the consensus literature is specifically the state and local government budget, and specifically for first-generation immigrants, largely because of K-12 education costs for their U.S.-born children - a cost that is recovered, and then some, once those children and grandchildren become taxpaying adults. Higher-profile claims of a much larger net cost (e.g., FAIR's $151B figure) rely on methodology that is rejected by economists across the ideological spectrum and inflates the result by roughly an order of magnitude once corrected.