ANALYSISApril 13, 2026
Has outsourcing hurt the American middle class?
Outsourcing of American jobs — both offshore and domestic — has grown substantially over decades, with approximately 300,000 jobs outsourced annually as of 2024 and the industry valued at $300 billion. Research consistently finds that outsourced workers earn 10–20% less than directly employed counterparts doing identical work. Meanwhile, a January 2026 AEI study found the core middle class shrank from 36% to 31% of households between 1979 and 2024, and a December 2025 Brookings report found one-third of middle-income households struggle to afford basic expenses.
Decades of outsourcing promised cheaper goods and global prosperity — but if the American middle class is hollower than it was in 1980, who actually won? And if we wanted to reverse it, could we — or would the cure be worse than the disease?
Not familiar with this story? Get context →
Domestic fissuring suppresses wages structurally
C
The janitor employed directly by Ford in 1965 shared in Ford's productivity gains through union contracts and internal pay norms. The janitor employed by a subcontractor in 2024 earns 15–20% less for identical work, with no claim on the firm whose floor he cleans. That wage gap is not a market signal reflecting different skill levels — it is a contractual architecture that lets firms disclaim responsibility for workers they functionally employ.
L
We agree on the mechanism, which makes the conservative framing puzzling: if the fissured workplace is 'a structural artifact,' as you put it, that was 'policy-shaped' rather than market-determined, then why has the default conservative position been to resist the policy corrections that would undo it? You've just described a rigged labor market — the question is whether you're willing to actually fix it.
C
Identifying a problem and endorsing any particular remedy are different things. Restoring the direct employer-employee relationship doesn't require a federal mandate — tax treatment of subcontracting arrangements, joint employer standards, and procurement rules are all levers that work within market logic rather than against it.
L
Those levers are real, but they've been available for decades and largely unused under conservative governance — so the question isn't whether the tools exist, it's whether there's political will to deploy them.
Mexico's outsourcing ban as proof of concept
C
Mexico's April 2021 ban on outsourcing core business functions was followed by approximately a 35% rise in average manufacturing wages. That outcome is not what a pure efficiency model predicts — it demonstrates that the wage suppression documented in American labor markets is reversible. If the penalty was created by policy, it can be corrected by policy.
L
That 35% figure is exactly the point conservatives should be most uncomfortable with, because it means the wages workers were receiving before the ban were not market wages — they were suppressed wages. The market was not clearing efficiently; it was clearing in favor of capital because the contractual structure gave workers no leverage.
C
Agreed, but Mexico's mandate is a blunt instrument — it requires direct employment of all core functions, which forecloses legitimate specialization arrangements that actually do generate efficiency. The lesson isn't 'ban outsourcing'; it's that restoring bargaining power produces wage gains, which can be achieved through multiple policy levers.
L
If the lesson is that bargaining power matters, then the follow-up question is why American workers have systematically lost it over four decades while conservatives opposed almost every policy that would have preserved it.
Aggregate efficiency gains versus distributional outcomes
C
The strongest pro-outsourcing case rests on McKinsey's finding that $1.14 returns to the domestic economy per dollar outsourced. But efficiency gains and distributional outcomes are not the same thing — the core middle class shrank from 36% to 31% of households between 1979 and 2024, and a third of middle-income households now can't afford basic expenses. A booming aggregate economy and a hollowing middle class are not contradictory; they are precisely what the data show.
L
The McKinsey multiplier doesn't tell you who captures the reinvestment — and you've answered that yourself: shareholders and high-skill workers, not the displaced production employee. But I'd push further: conflating GDP growth with worker prosperity isn't an honest mistake, it's a rhetorical move that has justified four decades of policy inaction on wages.
C
Fair, but the distributional critique cuts both ways. The AEI finding you cite shows the middle class shrank partly because households moved upward into higher income brackets — which means the aggregate growth wasn't entirely captured by capital. Some workers did win, and a policy framework that ignores that risks protecting job structures at the cost of the dynamism that moved them up.
L
Upward mobility for some doesn't explain away a 15–20% wage penalty for subcontracted workers doing identical jobs — those two facts coexist, and using the mobility story to dilute the wage suppression story is precisely the kind of averaging that makes inequality invisible in aggregate data.
High-cost offshoring undermines exploitation narrative
C
The UC Berkeley finding that most international outsourcing flows to high-cost destinations like Canada and Western Europe is real and deserves honest engagement. But it actually strengthens the domestic fissuring case: the 15–20% wage penalty on outsourced janitors and security guards has nothing to do with Guadalajara. The harm mechanism was always about the severed employer-employee relationship inside the United States, not about wage arbitrage with low-cost countries.
L
That's a genuine concession, and it matters — it means the anti-outsourcing case has to stand on domestic fissuring rather than a race-to-the-bottom narrative. But that's actually a stronger foundation, because it means the harm is happening right here, in contracts between American firms and American subcontractors, with no trade agreement required to fix it.
C
Exactly — which is why the policy response should be targeted at the fissuring mechanism rather than trade restrictions, tariffs, or reshoring mandates that address a problem that isn't primarily a trade problem.
L
On that we can agree, which makes the persistent conservative opposition to joint employer standards and subcontracting regulations all the more difficult to explain as anything other than protecting the arrangement that benefits the firms.
Labor market fissuring as a freedom question
C
Free markets require genuine competition, and labor markets in which workers are systematically stripped of bargaining power through structural fissuring are not free — they are rigged toward capital. The wage gap between nonproduction and production workers in manufacturing widened from 50% to 85% between 1980 and 2011. Limited government does not mean indifferent government, and conservatives who care about ordered liberty have every reason to treat systematic wage suppression as a rule-of-law problem.
L
The framing of fissuring as a freedom problem is the most interesting move in the conservative case, and I want to take it seriously — but it requires conservatives to acknowledge that the 'free market' they've been defending was never free for the production worker. If the wage gap widened by 35 percentage points over three decades, something structural was happening that markets alone weren't correcting.
C
That's precisely the argument — markets weren't correcting it because the policy environment enabled the fissuring. NAFTA's corporate protections, trade normalization with China, regulatory frameworks permitting domestic subcontracting without joint employer liability: these are policy inputs, not market outputs. Correcting them is not anti-market, it is pro-market.
L
Then the test is simple: support joint employer liability, close the subcontracting loophole, and restore bargaining power — if conservative governance delivers those outcomes, the framing holds; if it doesn't, 'the rule of law applies to labor markets too' remains a rhetorical position rather than a governing one.
Conservative's hardest question
The most honest challenge to this argument is the AEI finding itself: the middle class shrank partly because households moved upward into higher income brackets, which would suggest outsourcing did not uniformly suppress mobility. If upward mobility explains a meaningful share of the middle-class contraction, the wage suppression narrative is incomplete — and any policy response calibrated solely to protecting existing job structures risks impeding the very dynamism that moved some workers into better positions.
Liberal's hardest question
The UC Berkeley finding that most international outsourcing flows to high-cost destinations like Canada and Western Europe directly undermines the low-wage exploitation narrative that makes the anti-outsourcing case emotionally and analytically compelling. If the dominant form of offshoring is not a race to the bottom on wages, then the harm mechanism is more limited than critics — including me — tend to suggest, and domestic fissuring must carry most of the weight of the argument.
Both sides agree: Both sides accept that the documented 15–20% wage penalty for outsourced workers performing identical jobs is real and reflects a structural contractual arrangement — not a natural market outcome determined by skill or productivity differences.
The real conflict: The sides genuinely disagree on whether middle-class shrinkage is primarily a story of upward mobility or downward wage suppression — a factual dispute about the AEI data that determines whether outsourcing harmed the middle class or merely restructured it.
What nobody has answered: If the AEI data shows the middle class shrank partly because households moved upward, and Brookings simultaneously shows a third of middle-income households cannot cover basic expenses, which trend is the real signal — and does any proposed outsourcing policy actually target the group that is genuinely falling rather than merely reclassifying?
Sources
- AEI household income study tracking 1979–2024 data, January 2026
- Brookings Institution report on middle-income household expenses, December 2025
- Economist David Weil's 'fissured workplace' framework on domestic outsourcing
- McKinsey Global Institute study on U.S. economic returns per dollar outsourced
- UC Berkeley study on international outsourcing destinations
- U.S. Bureau of Labor Statistics manufacturing employment data (1980–2017)
- Empirical wage studies on outsourced janitors and security guards (U.S., France, Germany comparisons)
- Mexico outsourcing ban legislation and wage outcome data, April 2021 onward
- Political statements by Senator Josh Hawley (December 2025) and Secretary of State Marco Rubio (February 2026)