Amazon workers on food stamps have tripled despite its record revenue—and it’s just the latest evidence of the new economy of shrinking labor shares

U.S. labor compensation fell to 52.8% of economic output, its lowest share since BLS records began in 1947, while the S&P 500 has gained 600% since 2000 versus only 12.5% real wage growth. Amazon's profits rose from $11.6B to $77.7B while 12,346 employees used SNAP and 11,338 used Medicaid in the GAO's 11-state sample; Amazon reported 2025 revenue of $717B, up 12% year over year. Analysts cite weakened union representation, subcontracting and gig work, and potential AI-related wage compression, with high-AI-exposure occupations seeing a 6.7% decline in real wage growth after 2023.

Analysis

The investable issue is not the labor-share datapoint itself but whether political and consumer constraints begin to tax the margin model of labor-light, high-volume platforms. AMZN, WMT and FDX face asymmetric risk: wage/benefit mandates, joint-employer enforcement, or state-level Medicaid/SNAP cost recapture would raise the effective cost of fulfillment labor while their value propositions limit pass-through. A 100bp increase in labor/fulfillment expense as a percentage of sales is materially more damaging to low-margin retail and parcel operations than to AMZN's consolidated earnings, but it would weaken the narrative supporting continued retail-margin expansion.

Over the next 1-3 months, this is primarily a Washington/state-policy and reputational-risk watch rather than an earnings trade; company assertions about benefit enrollment do not resolve exposure among part-time, contractor, and subcontracted workforces. The more important 6-18 month transmission is weaker lower-income real consumption: spending shifts toward essentials and value, favoring WMT traffic but potentially pressuring discretionary mix and supplier pricing. This also raises delinquency and deposit-balance risk at consumer-facing banks, though BAC's diversified balance sheet makes it an indirect rather than clean expression.

The consensus may over-attribute labor-share pressure to imminent AI job losses. The near-term earnings-positive channel is wage restraint and automation-driven operating leverage, favoring AMZN and AI infrastructure beneficiaries; the underpriced counterforce is regulatory repricing once productivity gains become visible without broad wage gains. A reversal in real wage growth, easing inflation, or a looser labor market without legislative action would preserve current margins and invalidate the policy-risk leg of the thesis.

Watch quarterly fulfillment and transportation cost per unit at AMZN/FDX, WMT operating-margin guidance, real hourly earnings versus CPI, state joint-employer actions, and any federal budget proposals targeting employer reliance on safety-net programs. Absent measurable cost inflation or policy action, this article alone is insufficient basis for a directional short.

Key Decisions for Investors

• Maintain AMZN as a core long only with a policy-risk hedge: buy 6-9 month AMZN put spreads around 10-15% out of the money after strength. The hedge becomes attractive if retail/fulfillment margin is being priced as structurally expanding; exit if next two quarters show stable cost per unit and no material regulatory escalation.
• Use a 3-6 month pair trade only on confirmation: long WMT / short FDX, targeting a 8-12% relative move if real wage growth weakens and low-income consumption trades down. WMT's essentials mix should defend traffic better, while FDX has less pricing flexibility against labor and contractor costs; stop out on FDX margin guidance improving or WMT discretionary-mix deterioration.
• Set an event-driven short alert on FDX rather than initiate now: act only if a state or federal joint-employer/contractor rule is proposed alongside transportation-cost inflation. A 50-100bp operating-margin risk would be meaningful against a business with cyclical volume exposure; falsified by sustained yield improvement that offsets cost pressure.
• Monitor consumer-credit proxies at BAC and broader card lenders after real-wage and inflation releases; do not short BAC on this thesis alone. A sustained rise in lower-income delinquencies or deposit outflows would create a cleaner 6-18 month credit-cycle signal than the labor-share statistic.