The flexibility debate has been hijacked by the wrong workers. Almost every think-piece on "the future of flexible work" implicitly addresses the 35% of the U.S. labor force whose jobs can be done from a laptop. The other 65% — the warehouse picker, the bus driver, the CNA, the line cook, the field technician — are mostly absent from these conversations, even though their lives would be transformed more by predictable schedules than by any work-from-home policy. The Bureau of Labor Statistics' American Time Use Survey makes the asymmetry plain: in 2024, 53% of workers in management and professional occupations did at least some work from home; in service occupations, the figure was 9%.
The argument of this piece is that "flexible work for all" cannot mean "remote work for all," because remote work is structurally unavailable to the majority of the workforce. It has to mean something more specific: predictability, control over schedule, and a real say in how time gets allocated. That is achievable across job types, but it requires employers and policymakers to think about flexibility as a portfolio of policies, not as a synonym for hybrid.
The scheduling crisis hiding inside the flexibility conversation
The Shift Project at Harvard, led by Daniel Schneider and Kristen Harknett, has produced the most consequential body of research on hourly-worker schedules in the past decade. Their multi-year surveys of frontline workers at large retailers, restaurants, and service-sector employers consistently find that roughly two-thirds of hourly workers receive their schedules less than two weeks in advance, more than half report shift changes within the same week, and a quarter routinely close one night and open the next morning — the "clopen" shift that gives workers fewer than 11 hours between assignments.
Those numbers correlate with effects you can measure in worker health and family life. The Shift Project's 2022 American Journal of Public Health publication linked unpredictable scheduling to higher rates of psychological distress, poor sleep, and family conflict, controlling for income and demographics. EPI's analysis of the same data estimated that unpredictable scheduling costs frontline workers roughly $1,000–$3,000 per year in lost wages from missed shifts and childcare-driven absences.
Oregon, Seattle, San Francisco, New York, Chicago, Philadelphia, and Los Angeles have passed predictable-scheduling laws over the past decade. The early evaluations are positive: a 2020 study by Schneider and Harknett of Seattle's Secure Scheduling Ordinance found a 30% reduction in last-minute shift changes and meaningful improvements in worker-reported sleep and happiness, with no measurable employment loss. That is roughly the strongest result a workforce-regulation evaluation can produce — significant worker gains, no detectable employment cost.
What "flexibility for all" actually requires
Predictable schedules with the right to refuse
Schedules posted 14 days in advance. The right to decline shifts added after posting without retaliation. Premium pay for last-minute changes. These are the core elements of every predictable-scheduling law that has been evaluated favorably, and they are the floor. Employers operating in jurisdictions that don't require them can adopt them voluntarily; the operational cost is real but bounded, and the retention benefit is sizable. Gallup's 2024 State of the Global Workplace data shows that schedule predictability is the single largest non-pay driver of frontline-worker engagement.
Self-scheduling within constraints
Software like Legion, Shyft, and even Microsoft Shifts has made it possible for workers to swap shifts, pick up open shifts, and indicate preferences inside an employer-defined envelope. Walmart, Starbucks, and several large healthcare systems have deployed self-scheduling at scale; published case studies suggest meaningful retention improvements and lower scheduling overhead for managers. This is the place where technology most clearly improves frontline-worker flexibility without removing the structural constraints of the job.
Task-bundled time-off
Hourly workers in the U.S. have, on average, far less paid leave than salaried peers. The BLS National Compensation Survey shows that the lowest-paid quartile of private-sector workers averages 7 paid vacation days a year, versus 18 for the top quartile. Closing some of that gap — and making the leave usable on short notice without manager negotiation — does more for "flexibility" than any hybrid policy.
Hybrid where it is genuinely feasible
For roles that mix on-site and remote tasks (field service, IT support, healthcare administration, several skilled trades), the question is not "remote yes/no" but which tasks belong in which mode. Microsoft's 2024 Work Trend Index and McKinsey's "American Opportunity Survey" both find that workers report the highest satisfaction with hybrid structures that designate purpose for on-site days — collaboration, training, mentorship — rather than mandate them by quota.
The competing models employers are now choosing between
The post-2022 stabilization has produced four broad models. The full-remote model survives at firms whose talent strategy depends on it. The 3-2 hybrid (3 days in, 2 remote) is now the U.S. modal pattern for knowledge work, per Nick Bloom at Stanford's WFH Research. The four-day workweek has been seriously trialed by more than 200 firms across the UK, Iceland, and the U.S.; the 2023 Cambridge University evaluation of the UK pilot found that 92% of participating firms continued the policy and saw stable or improved productivity. And the predictable-schedule model is now the legal floor in roughly 10% of the U.S. labor market.
None of these is universally applicable, which is the point. Flexibility is not a single policy. It is a portfolio that should be tailored to the job, with the goal of maximizing worker control over the highest-friction parts of their week.
For a deeper look at how the white-collar end of this debate is settling out, see our flagship piece on Remote Work, Year Six →.
The harder question: who bears the cost?
The strongest counter-argument to "flexibility for all" is that the cost of flexibility — schedule premiums, predictability penalties, training to enable cross-coverage — falls on operators who run on thin margins, particularly in food service and retail. That is real. But the empirical record from the predictable-scheduling laws does not show employment loss, and the labor-cost increases have been modest enough that they have not driven measurable price increases above the comparable-jurisdiction baseline.
The cost that the current system imposes on workers is documented and large: unpredictable schedules drive turnover, which costs employers an average of about $4,000 per replacement hire at the hourly level, per SHRM's 2024 talent-management benchmarks. The status quo is not free. It is just paid for by workers rather than by firms.
Flexibility for all does not mean remote for all. It means predictability for the cashier, real time off for the line cook, and self-scheduling for the home health aide — the workers the WFH conversation has spent five years ignoring.
What to push for next
Two policy moves would do the most to extend flexibility to workers who currently have least of it. First, a federal predictable-scheduling minimum — modeled on Oregon's Fair Work Week Act — that establishes a 14-day notice floor and premium pay for last-minute changes. The Schedules That Work Act has been introduced in Congress repeatedly; the empirical evidence for it is now stronger than it has ever been. Second, a serious expansion of paid leave: the U.S. remains the only OECD country without national paid parental leave, and the gap in paid sick days between high- and low-wage workers is among the largest in the developed world.
Neither of these is a remote-work policy. Both would do more for the average American worker's flexibility than any further iteration of hybrid debate.
Updated May 21, 2026. This piece was substantively rewritten as part of NWLB's 2026 editorial refresh.
