Employment by Nativity · BLS Table A-7 via CPS · Jan 2007–Present

Who's actually getting the new jobs?

Counting how many months foreign-born workers "won" versus native-born workers hides the real signal: the two groups aren't the same size. This page compares each group's actual share of monthly job gains against the share they'd get if hiring simply matched their existing share of the workforce — the gap between those two numbers is what's worth paying attention to.

Native-born Foreign-born
Prefer video

Watch it explained

Two walkthroughs of this data and why the "9 out of 10 jobs go to foreign-born workers" claim doesn't hold up against BLS Table A-7.

Between Foreign Workers and American Workers, who gets the jobs?.

There is only one way to get a accurate count between foreign born employed and native born employed.

Step 1 — the baseline

Share of the total workforce

On a shared 0–100% axis, both lines look almost flat, because year-to-year shifts are small relative to the full scale. Zoomed to each group's own range, the trend across the full Jan 2007–Jun 2026 history is visible.

Native-born share
Foreign-born share

Each panel is scaled to its own data range (not 0–100%), so a small shift reads as a visible slope rather than a flat line. Watch the axis labels — the two panels are not on the same scale.

Step 2 — the actual question

Is each group over- or under-represented in new hiring?

For every month, this asks: if foreign-born workers are some share of the workforce, are they getting that same share of that month's net new jobs — or more, or less? Bars above zero mean a group is capturing more than its proportional share; below zero means less.

Zero line = hiring exactly proportional to workforce share. A bar at +15pp for foreign-born means that in that month, foreign-born workers received 15 percentage points more of the net job gains than their workforce share would predict.

Step 3 — does it add up over time

Cumulative excess jobs since Jan 2007

Single months are noisy. This sums the monthly excess/deficit (in actual jobs, not percentage points) since Jan 2007, so you can see whether over-representation in one month gets offset later — or compounds.

Methodology & data

Data: BLS Table A-7 (CPS household survey), monthly, Jan 2007 – Jun 2026, pulled live from your tablea7monthly database on every page load — this is not a static snapshot. "Proportional expectation" for a month = that month's total net job change × the group's share of the prior month's workforce. "Excess" = actual group gain − proportional expectation. Any month missing from the source table is skipped (no interpolation) rather than guessed, which also means any gap silently excludes the adjoining month-over-month comparison on both sides of it.

October 2025, if present as a gap in your data, is due to the federal government shutdown, not a tracking error: BLS was unable to conduct the Current Population Survey (the household survey behind Table A-7) for the October reference period, and stated data cannot be collected retroactively — a permanent gap in this series, not a delayed release. That's distinct from the establishment survey (nonfarm payrolls), where BLS published a combined October + November report on December 16, 2025.