Only 25% of Global Workers Feel Their Job Is Safe — ADP's 39,000-Worker Survey

Only 25% of workers worldwide feel their job is safe. Read that again. After two years of "the labor market is strong" headlines and surging payroll numbers, ADP Research went out and asked 39,000 workers across the globe how secure they actually feel — and three out of four said: not very. That gap, between the macro statistics and the worker on the ground, is the most important labor-market signal of 2026 so far.

The April 2026 study from ADP chief economist Nela Richardson, Ph.D., reframes how we should read this cycle. Hiring numbers can stay solid while worker confidence cracks underneath. That's not just a sentiment problem — it's a leading indicator. Workers who don't feel safe make different decisions: they don't change jobs, they don't negotiate, they don't take risks, and over time, those defensive postures show up in productivity, engagement, and eventually labor supply.

The headline number, in context

[Fact] In ADP Research's Global Workforce Survey, only 25% of respondents globally reported feeling their jobs were safe. In the United States, the figure was marginally better at 28% — still under one in three workers.

Compare that to the unemployment rate most economies are reporting. The U.S. headline unemployment rate is sitting in the low 4% range. Job openings remain elevated by historical standards. By the traditional metrics, this is a tight labor market. Workers should feel like they have options. They don't.

The disconnect is real, and it's not new. Worker sentiment surveys have been signaling stress under the surface of solid headline data for most of the post-pandemic period. What's different in 2026 is the concentration of three simultaneous pressure sources: AI deployment in white-collar workflows, geopolitical risk affecting trade and supply chains, and lingering inflation memory that hasn't fully reset household budgets. Workers are reading their environment, not the headline statistics.

The age breakdown tells a sharper story

The aggregate 25% obscures generational variation that matters for career planning:

• Ages 18–26: only 26% feel secure. The cohort just entering the labor market is sitting right at the global average — not the privileged position early-career workers usually occupy in tight markets.
• Ages 27–39 and 40–54: both at 30% feeling secure. Mid-career workers, with the most domain expertise and the strongest network effects, are the most confident group — but still less than one in three.
• Ages 55–64: drops back to 23%, the lowest cohort.

The U-shape is consistent with what we see in AI exposure data. [Claim] Workers just starting out worry about the entry-level pipeline being thinned by AI assistance. Workers near retirement worry about losing positions they can't easily replace. Mid-career workers, with the most leverage and the most current skills, sit at the local maximum of confidence — but the absolute level is still low.

If you're early-career, the read isn't "the market is bad." It's that the traditional advantages of starting in a tight market are muted. You'll need to be more deliberate about where you land, and faster to build the kind of role-specific competence that compounds. See our occupation-level analysis at /en/occupation/computer-programmers for one of the categories where this signal is sharpest.

The factors workers actually cite

Richardson's analysis identifies three contributing factors to the insecurity reading, though the survey doesn't quantify their relative weight:

1. Geopolitical concerns — trade frictions, sector-specific tariff exposure, supply chain volatility affecting hiring confidence in manufacturing, logistics, and energy
2. AI impact — visible deployment of AI tools across knowledge-work occupations, with workers reading their employer's adoption pace as a signal about their own roles
3. Ongoing inflation concerns — real-wage erosion from 2022-2024 is still being remembered in household budgets, even as headline inflation has normalized

What the survey doesn't tell us is whether AI ranks first, second, or third in any given respondent's mind. [Estimate] Based on the rapid acceleration of AI adoption in office occupations through 2025, it's reasonable to infer that AI is rising fastest as a stated worry — but the survey design doesn't isolate it from the others.

Why "feeling safe" matters for performance

Here's the part most coverage of this study has missed. ADP also measured the engagement-and-productivity correlate of feeling secure, and the numbers are striking. [Fact] Workers who feel secure are 6 times more likely to be fully engaged at work, and 3.3 times more likely to report being highly productive.

That multiplier isn't a soft-skill aphorism. It's a hard business signal. If three-quarters of your workforce sits below the security threshold, the engagement and productivity drag at the company level is real and measurable. The implication for employers is the opposite of the cost-cutting reflex many are running: short-term workforce reductions to absorb AI deployment costs may be eroding the engagement of the workers being asked to absorb additional scope.

For workers, the signal is different. Feeling secure isn't just a psychological win — it's a precondition for the kind of work that builds long-term career equity. If you're stuck in a role where you can't shake the insecurity, that's a leading indicator that the role is wrong for you, not that you should grind harder.

What this means for your career strategy

If you're a worker, three practical reads:

First, treat sentiment data as a leading indicator. Worker confidence usually moves before employment statistics do. If the people inside your sector are quietly hedging, that's information about where the sector is heading.

Second, the engagement multiplier cuts both ways. Find a role where you feel secure enough to do your best work — even if it pays modestly less than a high-stress alternative. The compounding effect on output, promotion velocity, and skill formation is much larger than the salary differential over five years.

Third, AI fluency is no longer optional in white-collar work. The workers who feel most secure are the ones who can describe how AI changes their workflow rather than threatens it. The transition window for picking up that fluency is still open, but it's narrowing.

If you're an employer, the read is simpler: the 6× engagement multiplier suggests that psychological safety during AI transitions is a higher-leverage investment than additional tooling. Workers who feel like AI is being deployed alongside them, not on top of them, are dramatically more productive than the ones who don't.

How fresh is this signal?

Published April 7, 2026, the survey reflects worker sentiment captured in Q1 2026 — squarely in the window after the major AI deployment announcements of late 2025 but before the policy responses now being debated. The 35-day-old age of the report means it's still capturing the contemporary mood, not a stale signal. ADP Research is one of the few sources with direct payroll exposure to validate sentiment data against actual behavior, which makes this study unusually credible as a labor-market read.

Sources

• ADP Research (April 7, 2026), "Is your job safe?", by Nela Richardson, Ph.D., based on Global Workforce Survey of 39,000+ workers: https://www.adpresearch.com/is-your-job-safe/
• Methodology: stratified random survey, global sample

Update History

• 2026-05-13: Initial publication based on April 7, 2026 ADP Research Global Workforce Survey findings.

---

_This analysis is AI-assisted and synthesizes findings from the cited primary source. All numerical claims are sourced directly to ADP Research. Interpretation and career framing are editorial._