Immigration Enforcement: Native Employment Impacts and the Complementarity Trap
A central argument for immigration enforcement is that removing foreign-born workers draws native-born workers into jobs and raises their wages. The accumulating evidence keeps pointing the other way: enforcement shocks reduce measured work overall, including for natives, because immigrant and native labor are complements in the affected sectors. This cycle adds a new local natural experiment from Minneapolis.
Operation Metro Surge (Minneapolis–St. Paul, Jan–Mar 2026): work fell when agents arrived
In December 2025 the administration began sending federal immigration agents to Minneapolis; in early January it announced 2,000 agents — "the largest DHS operation ever" in the city — with over a thousand more soon after. Aaron Sojourner (W.E. Upjohn Institute) and Aaron Rosenthal (North Star Policy Action) measured what happened to work in the metro (Upjohn research highlight, July 30, 2026):
"In the weeks after the surge of DHS agents started, employment declined 2.8 percent, the number of operating businesses fell 1.7 percent, and the number of hours worked fell 1.9 percent1."
Sojourner: "If you take that impact on the number of hours and put that into dollars… workers in the Minneapolis–St. Paul area lost over $100 million in wages."
The leisure-and-hospitality sector was hit hardest: 4,600 jobs lost across Minnesota during the surge, 3.8 million work hours, and $71 million in lost wages for workers in January–March 2026.
Design: daily data from a timekeeping/payroll platform covering small businesses (hours worked, employees working, locations open), comparing Twin Cities trends against 49 other metro areas; the leisure-and-hospitality estimates use monthly BLS data. Before January 3, 2026, Twin Cities staffing tracked comparison metros; it fell after the surge began.
How much weight this can bear — adjudication
- Not peer-reviewed, think-tank-affiliated (Upjohn + North Star Policy Action). The direction is consistent with the peer-reviewed Secure Communities literature (enforcement reduced native employment shares, concentrated in immigrant-reliant sectors), but the specific magnitudes are preprint-quality.
- The sharpest single-day drop (−17% on January 23) coincided with an organized stay-home protest ("Day of Truth and Freedom"), so part of the measured decline is protest behavior, not enforcement per se.
- Hours-worked data conflates channels: fear-driven absence (supply), business closure (demand), and deportation are observationally equivalent in payroll data. The study measures that work contracted; it cannot cleanly apportion why.
- The complementarity reading stands: enforcement did not shift work toward remaining workers — total hours, employment, and operating businesses all fell. This matches the firm-level Korea evidence in Low-Skilled Immigration: Complementarity, Technology Adoption, and Output Mix Adjustments: when immigrant labor is removed, natives do not backfill the vacated tasks at scale.
Aggregate context: this occurred against a labor market with 4.1% unemployment in August 2026 (macro view, Sept 21, 2026) — a cooling but not collapsing market, which makes sector-level enforcement effects easier to see.
-
An instance of Restricting low-skilled immigration collapses the firm-level production scale that supports native jobs. — Enforcement removed immigrant labor and total work contracted outright — natives did not backfill the vacated tasks, confirming complementarity rather than substitution. ↩︎