New Immigration Estimates Are Not the Last Word
Prepare for another year of uncertainty and confusion over the rate of net US immigration.
There’s a fierce debate over the current rate of US immigration and its implications for the labor market and the interpretation of labor market data. Yesterday and today, the two government agencies that produce the most important estimates of immigration published updates. While these updates are sorely needed and get us a couple steps closer to truth, together they portend another year of uncertainty and confusion about current immigration trends.
This short overview blogpost links to earlier, more detailed posts on Slow Boring, the Peterson Institute for International Economics, and my own Substack; follow those links for technical details.
One estimate gets revised down, while another hits a new high
Yesterday the Congressional Budget Office revised its 2025 estimate for net immigration down significantly, from 2.0 million in its January 2025 report to 408 thousand in its latest report. Projections for immigration in 2026 through 2030 were also revised down. The estimate of lawful permanent residents was unchanged, while estimates of other groups including people who entered illegally, or under parole authority, or via other pathways were revised sharply downward.
Meanwhile, the Census Bureau’s newly released American Community Survey (ACS) for 2024 shows a new high in the number of foreign-born US residents who lived abroad one year ago: 1.8 million.1 This estimate is based on surveys conducted throughout the entire year of 2024 and will be the key input into the Census immigration estimate for 2025, which (confusingly) covers the period July 1, 2024, to June 30, 2025, and will be released in December 2025. Alongside this key input, other inputs like (1) an estimate of foreign-born residents emigrating from the US and (2) several administrative data sources on immigrants, particularly “humanitarian migrants”, are included in the final net immigration estimate. These additional inputs account for two issues: first, the estimate for the year ending in mid-2025 is based on lagged survey data collected as early as January 2024; second, many types of migrants are less likely to be included in the ACS.
When Census publishes its 2025 immigration estimate in December, it could be well below the 1.8m number of new immigrants reported in the 2024 ACS — especially with the decline in net immigration likely to be added from administrative sources — but we’ll have to wait and see.
The 2025 immigration rate remains uncertain. The stakes are high
The new CBO number helpfully narrows the range of plausible net immigration estimates. The CBO estimate of 408k is above the range of -525k to +115k estimated by a team of leading immigration researchers, but still in the same ballpark. In contrast, the reported January-to-July drop of 2.2m in the foreign-born population in the Current Population Survey (CPS) of the jobs report looks implausibly negative relative to the new CBO estimate.
After Census releases its 2025 estimates in December 2025, and CBO issues its next annual update in January 2026 (if they follow their historical cadence), the Census and CBO estimates won’t necessarily agree. They use different methodologies and have disagreed before: notably, Census’s estimates for 2022 and 2023 immigration were far below CBO’s until Census revised its historical estimates upward in December 2024. We might face a similar situation in reverse, with Census estimates for 2025 immigration possibly above CBO’s estimates until Census issues its first revision for 2025 in December 2026.
The stakes are high. Immigration is the unpredictable and volatile component of population estimates and projections. CBO’s demographic estimates underpin the CBO’s economic outlook, which will be updated tomorrow followed by a fuller budget and economic outlook in winter 2026. Census population estimates have numerous official uses: they are the population controls for the Current Population Survey; they are the denominator for per-capita economic measures reported by the Bureau of Economic Analysis; and the local-area population estimates derived from the national estimates affect funding allocations for numerous federal programs. And the rate of immigration and population growth changes the interpretation of headline economic indicators like monthly job growth and GDP. Because the true immigration rate is unknown and is likely to stay uncertain, the fog surrounding key economic data won’t lift soon.
This estimate is in Table B07007 of the 2024 American Community Survey.