What if the deportation-labor shortage thesis is wrong — what labor market dynamics could absorb the shock
When Immigrants Leave, Who Does Their Jobs? Probably Nobody — Because the Jobs Disappear Too
Based on analysis of a 91-node, 287-edge knowledge graph exploring what labor market dynamics could absorb a mass deportation shock.
The Question
There is a popular idea: if immigrants are removed from the country, the jobs they held will open up for American-born workers. Wages will rise. American workers will fill the gap. The labor market will rebalance.
This analysis maps out a knowledge graph — a web of 91 concepts connected by 287 cause-and-effect relationships — that examines whether that idea holds up. The short answer the graph encodes is: mostly no, but for reasons that are not what most people expect.
The Surprise: The Jobs Don’t Wait Around
Imagine a small town where 100 workers at a local factory also happen to be 100 customers at the local diner, the barbershop, and the grocery store. If those 100 workers leave town, yes — 100 factory jobs open up. But the diner, barbershop, and grocery store now have 100 fewer customers. So the diner cuts shifts. The barbershop lets someone go. The grocery store orders less. Suddenly there are also 30 fewer jobs there.
The net gain in available jobs is not 100. It might be 70, or 50, or even negative.
This is the central mechanism the graph keeps returning to, described in the analysis as “demand-supply co-destruction.” The same people who fill jobs also spend money that creates jobs. When they leave, both sides of that equation shrink at once.
The graph encodes a specific ratio: for every six immigrants who leave or are deported, approximately one native-born worker loses their job due to this chain of reduced spending. That does not mean deportation destroys more jobs than it opens — but it means the gap that opens is significantly smaller than the raw departure count would suggest.
Why Don’t American Workers Just Take Those Jobs?
This is where the graph’s second major finding comes in, built around a concept called “segmented labor markets.” Think of the labor market not as one big pool of interchangeable workers, but as a collection of separate smaller pools that don’t easily connect.
Picture a building with different floors. Workers on the ground floor do physical outdoor work — farm harvesting, roofing, landscaping. Workers on the second floor do service work in cities. Workers on the third floor do office work. The staircases between floors are narrow and hard to climb.
When jobs open up on the ground floor, the people on the third floor don’t automatically come down. There are real obstacles: the work is physically demanding and located in rural areas far from where most job-seekers live; the pay, while possibly higher than before, still requires moving your family; and the social infrastructure (housing, transportation, childcare) for doing that work doesn’t exist where the workers currently are.
The graph documents this through specific examples:
- Farm work requires being in rural areas for short seasonal windows. Most unemployed workers in cities cannot simply relocate for three months and then relocate back.
- Construction work relies on established subcontracting networks — crews of workers who know each other and work together efficiently. You cannot hire individual strangers off the street and expect them to perform as well as a seasoned crew.
- Long-term care and eldercare work is low-wage, emotionally demanding, and located wherever the elderly person lives — not necessarily near a job-seeking worker.
The graph also notes a specific policy trap: the “benefits cliff.” Some workers who are currently receiving government assistance (food stamps, healthcare subsidies) would actually lose money by taking a low-wage job, because taking the job disqualifies them from benefits worth more than the wage increase. So there is a pool of potentially available workers who are structurally locked out of re-entering the labor market even when wages rise.
The Fear Effect Is Bigger Than the Deportations
Here is a non-obvious finding the graph encodes: the number of people who actually get deported is less important than the number of people who believe they might be deported.
When enforcement intensifies, many undocumented workers who were not deported still stop going to work. They stop going to the grocery store. They keep their children home from school. They move in with relatives. They stay off the radar.
The graph estimates this fear-of-enforcement effect is roughly five to ten times more impactful on the labor market than the actual deportation numbers. The term used is “enforcement chilling effect” — like how a cold wind makes a room feel much colder than the thermometer reads.
One particularly sharp illustration: undocumented workers who paid taxes using an official tax ID number were actually more exposed to enforcement because the IRS began sharing data with immigration enforcement. Workers who complied with tax law faced higher risk than workers who stayed entirely off the books. That is a perverse outcome — compliance became a liability.
The Feedback Loops: When Problems Feed Themselves
Some of the most important structural findings are about what the graph calls “feedback loops” — situations where a problem causes conditions that make the same problem worse.
The housing construction loop. Fewer immigrant construction workers means fewer homes get built. Fewer homes get built means fewer people can afford to move to take construction jobs. Fewer people moving means the network of immigrant construction crews — who recruited each other, trained each other, and coordinated work — collapses further. The collapse of those networks means even if wages rise, there is no functional crew to hire. The price signal (higher wages) stops working because the organizational infrastructure to respond to it no longer exists.
The farm exit loop. When farms lose workers and face higher tariffs on imports simultaneously, some farms shut down entirely. When farms shut down, workers in those regions who sent money home to other countries send less. When those origin communities receive less money, more people in those communities become desperate and attempt to migrate — which intensifies the political pressure for enforcement — which intensifies the original problem. The loop feeds itself.
The tax and compliance loop. IRS-ICE data sharing causes undocumented workers to withdraw from formal economic participation. That withdrawal contracts the shadow economy (informal jobs, cash payments, informal markets). That contraction reduces spending, which reduces jobs, which increases economic desperation among the native-born population, which creates political conditions for more enforcement. Again, the loop sustains itself.
The Twisted Irony of the Guest Worker Program
The United States has a legal guest worker program for agricultural labor called H-2A. The idea is: if farms can’t find enough American workers, they can hire temporary legal foreign workers.
The graph notes something structurally strange about this program. Because guest workers on H-2A visas are tied to a specific employer and cannot easily switch jobs, employers have unusual power over them. This is called “monopsony” — when a buyer (here, the employer) has so much market power that they can set the price (here, the wage) rather than compete for workers.
When the government lowered the required H-2A wage rate to make the program more affordable for farmers, it inadvertently made the program less effective: workers already in the country chose not to enroll because the new wages were too low to make the arrangement worthwhile. The program designed to solve the labor shortage ended up worsening it by cutting the price that would have attracted workers. The graph describes this as a “self-defeat paradox.”
The Immigrant Entrepreneur Angle
The standard framing of immigration and jobs focuses on competition: immigrants take jobs, or immigrants free up jobs. But the graph encodes a different relationship that is easy to miss.
Many immigrants are not just workers — they are business owners. A restaurant owner employs kitchen staff. A construction firm owner employs crews. A cleaning company owner employs cleaners. When that business owner is deported, it is not just one job that disappears — it is ten or twenty jobs that depended on that person’s organizational role.
The graph calls this the “immigrant entrepreneur native employment multiplier.” The point is that deportation removes demand-creators, not just labor-suppliers. This inverts the usual framing.
Wrong Now, Right Later
One of the more nuanced findings: the graph does not say the deportation-creates-jobs thesis is entirely and permanently wrong. It says the thesis is wrong about timing.
The demand-destruction mechanism dominates in the short run — the next several years. But there is a longer-run demographic reality: the American-born population is aging, birth rates are low, and the working-age population will genuinely contract over the next decade. By roughly 2031, the analysis suggests, the labor shortage the thesis predicts for now may actually arrive — but it will arrive from the aging-population side, not from the deportation side, and years later than claimed.
There is also a wild card: automation. Robots and AI could theoretically replace some of the physical labor that immigrants currently do. But the graph notes a gap — the technology exists in principle but is not yet practically deployable in, for example, strawberry harvesting or elder home care. Whether automation catches up before the 2031 demographic crunch is an open question the graph does not resolve.
Bottom Line
The knowledge graph encodes several structural findings that are non-obvious and worth holding together:
The labor gap largely closes itself, not through native substitution but through reduced demand. Deported workers were also customers and business owners. The job openings they leave behind are smaller than the raw numbers suggest.
The workers who are available cannot easily take the jobs that open, because the jobs and the workers are separated by geography, skill mismatch, network infrastructure, and policy traps — not just by wages.
The fear effect dwarfs the deportation effect in magnitude. Enforcement chills participation far more broadly than the enforcement actions themselves.
The guest worker legal alternative is currently undermining itself through wage policy that makes the program unattractive.
The thesis is probably correct about direction but wrong about timing — genuine labor scarcity is a real long-run demographic outcome, but it is being delayed, not accelerated, by the demand-destruction mechanism.
The aggregate numbers hide significant geographic and sectoral variation. Some specific communities and industries will experience genuine labor crunches even as national averages look stable. The national picture is, in this way, misleading.
The graph does not say immigration policy is good or bad. It maps the structural mechanisms by which labor markets actually respond — and those mechanisms do not match the simple prediction that deportation creates proportional job openings for native workers.