Deportations Were Supposed to Create Jobs. The First Read Is 2.8 Million Fewer Jobs.

The promise was simpler than the first scoreboard

The pitch was straightforward: crack down on illegal immigration, take the open jobs, and push wages up for native-born workers. The first scoreboard does not support that. In the first year of Trump's second term, unemployment rose, hiring slowed and wage growth stagnated. Rather than create a cleaner handoff into higher-paying native-born employment, the enforcement-driven labor supply shock looks contractionary.

What the early data show

In regions with bigger ICE surges, employment among remaining immigrants fell by 4% on average, likely because fear kept people from showing up. More important for the average voter, U.S.-born men with no more than a high school degree saw employment fall. The evidence also showed no increase in either employment or wages among U.S. citizens in areas with heavier enforcement.

That is the opposite of the simple "jobs left behind" story. When workers leave or stop showing up, and demand for labor does not shift the other way, the result is not a boom for remaining workers. It is slower hiring, weaker project flow, and softer local demand.

A shrinking workforce can make bad labor-market conditions look better than they are

When the labor force shrinks, unemployment can behave in deceptively calm ways even as actual jobs disappear. That is the main risk for investors: a headline can look acceptable while the economy underneath gets smaller, not tighter.

Why labor data can mislead after a shock

Earlier evidence already pointed in that direction. In areas with heavier ICE pressure, employment among remaining immigrants fell, and the number of male U.S.-born workers at work also declined as local economic activity cooled. So the key question is not whether the unemployment rate looks clean. It is whether businesses still have enough workers to keep operations running.

The fiscal hit is part of the demand problem

There is also a revenue side to this shock. Undocumented workers paid about US$66 billion in federal taxes in fiscal 2023, mostly through payroll levies. The new IRS-sharing setup with immigration authorities could weaken tax compliance in that group, and the IRS may lose US$12 billion this fiscal year and more than US$313 billion over the next decade. Less revenue means less funding for services and infrastructure that Main Street already depends on.

Enforcement can turn into supply disruption quickly

The same logic shows up at the border. The U.S. halted inspections needed for Michoacan avocado exports, a reminder that policy pressure can disrupt trade flows fast. Avocados are small in portfolio terms, but the example shows how enforcement can move from politics to shelves, prices, and restaurant menus.

That is the mix investors fear: stronger job growth on the surface, alongside a labor market that may be masking a smaller workforce, while earlier data already showed unemployment rose and hiring slowed in the first year. If growth slows while prices stay firm, the market is left with stagnation risks and no easy Fed rescue.

The real-economy damage shows up in complementarities and local demand

Job losses are only the first layer. The next layer is how labor shocks move through the real economy.

Why teams matter more than headcounts

A job site does not just need bodies. It needs a working team: supervisors, carpenters, equipment operators, laborers, suppliers, inspectors, truckers, and the local businesses that serve them. Remove one group too quickly, and the whole chain gets less efficient.

In Colorado, model estimates show that taking 1,000 immigrant men from their jobs would also shrink the labor force by an additional 2,700 immigrant men through chilling effects, while cutting U.S.-born male employment by 6,800. The report also found that wages for those remaining employed are not predicted to change. If displaced workers were simply being replaced at the same pay, this would look like a clean reset. The data do not show that.

Healthcare is an early warning sector

The next place this shows up is not an abstract macro variable. It is the sectors that need workers on site every day. Healthcare added just 22,000 jobs in July, below its 36,000 average monthly gain in the prior year. That matters because the sector has been one of the few supports for the labor market, even as demand for care rises with an aging population.

The investor transmission path is fairly direct:

• fewer workers available
• slower staffing growth in care and other hands-on sectors
• weaker subcontractor hours, supplier demand, and local spending
• earnings pressure that arrives before Washington calls the cycle healthy

What would change this read?

This call weakens if hiring steadies after October, construction and care staffing hold up, and labor-market conditions stop deteriorating in enforced regions. It also weakens if the broader demand picture improves instead of fragmenting further.

For now, the clearest risk is not a fake boom. It is a labor market that looks manageable on the surface while the underlying workforce, project flow, and local demand keep weakening.