The Four-Day Work Week and Productivity: What the Research from Iceland, Microsoft Japan, and 4 Day Week Global Actually Shows

Here’s a number that should reframe the entire four-day work week debate: knowledge workers average fewer than three hours of deep, productive work per eight-hour day. A 2018 global survey of 3,000 full-time workers found that 45% said they could finish their daily work in five focused, uninterrupted hours.

If that’s true — and the four-day work week research from Iceland, Microsoft Japan, and 4 Day Week Global all point in the same direction — then the question isn’t whether we can afford to work fewer days. It’s what we’re actually doing with the other five hours.

The four-day work week has moved from fringe experiment to serious policy discussion. In 2024, 22% of US employers offered a four-day option, up from 14% in 2022, according to the APA’s Work in America Survey. Congressional bills for a 32-hour standard week are on the table. But the research behind the movement deserves closer scrutiny than most headlines provide.

Let’s look at what the studies actually measured — and what they didn’t.

The Three Studies Everyone Cites

Almost every article about the four-day work week draws from three bodies of evidence. Each is real, each produced noteworthy results, and each has methodological limitations worth understanding.

Iceland’s Public Sector Trials (2015–2019)

Reykjavík City Council and the Icelandic national government ran two trials between 2015 and 2019, involving approximately 2,500 workers — over 1% of Iceland’s working population. Workers shifted from 40-hour weeks to 35–36 hours with no reduction in pay. The trials spanned diverse public-sector workplaces, including offices, hospitals, and social services.

The results, analyzed by the Association for Sustainable Democracy (Alda) and UK think tank Autonomy, were striking: productivity was maintained or improved in the majority of workplaces. Stress and burnout dropped significantly. Work-life balance improved across the board. The trials remained revenue-neutral for the government. Following the results, Icelandic unions negotiated shorter hours for 86% of the workforce.

The caveats: These were exclusively public-sector workers. There was no randomized control group — workplaces volunteered for the program. “Productivity” was measured primarily through service delivery metrics and manager assessments, not standardized output measures. And Iceland’s small, homogeneous labor market limits generalizability.

Microsoft Japan’s Work-Life Choice Challenge (August 2019)

Microsoft Japan closed its offices every Friday during August 2019, giving roughly 2,300 employees three consecutive days off per week with full pay. The company reported a 39.9% increase in labor productivity compared to August 2018, along with a 23.1% decrease in electricity consumption and a 58.7% reduction in pages printed.

The “40% productivity boost” became the most-cited data point in the four-day work week conversation. But look closer at the methodology.

Productivity was measured as output per employee per hour — not total output. When you reduce working hours by 20%, even a modest output improvement creates a dramatic per-hour figure. The trial lasted one month. It was bundled with other interventions: meetings were capped at 30 minutes, remote communication was promoted over in-person meetings, and employees received support programs for self-development. Microsoft never released a full methodological whitepaper, raw data, or independent audit.

The caveats: One month in August (already a low-activity period in Japan). Multiple simultaneous interventions make it impossible to isolate the four-day effect. No control group. No peer review. The 39.9% figure is internally reported, not independently verified.

4 Day Week Global’s UK Pilot (June–December 2022)

The largest and most structured trial: 61 UK companies and approximately 2,900 workers participated in a six-month pilot organized by 4 Day Week Global, with academic evaluation led by Professor Juliet Schor at Boston College and researchers from the University of Cambridge. Companies followed the “100:80:100” model — 100% pay, 80% time, targeting 100% productivity.

Results: 92% of companies continued the four-day week after the trial. Revenue stayed broadly flat, rising 1.4% on average. Staff turnover dropped 57%. Burnout decreased 71%. Thirty-nine percent of employees reported feeling less stressed.

Schor noted that the primary mechanism wasn’t working faster — it was working on fewer low-value tasks: “The main reason employees have maintained productivity is that companies have decreased or cut activities with questionable or low value. Meetings — a traditional source of complaints — have been a major target.”

The caveats: The study was organized by a nonprofit whose mission is to promote the four-day week. Companies self-selected into the trial — creating significant selection bias toward organizations already motivated to make it work. The results were not peer-reviewed at the time of release. Productivity was largely self-reported via manager surveys, not measured through standardized output metrics. And six months may not capture long-term effects like the fading of novelty or workload creep.

The Measurement Problem

The most important question in four-day work week research is also the one most often glossed over: what does “productivity” actually mean in these studies?

In Iceland, it meant that public services continued to be delivered at the same level — a reasonable but low bar. At Microsoft Japan, it meant output-per-hour compared to the same month a year earlier — a metric that conflates efficiency gains with reduced denominator hours. In the 4 Day Week Global trial, it was largely self-reported by managers who had opted into a program designed to succeed.

A 2023 working paper from the European Commission’s Joint Research Centre put it bluntly: the major four-day week pilots suffer from “too short a time of research implementation,” “subjective assessments of employers and employees,” and “the lack of examination of potentially negative effects.” A systematic review published in Management Review Quarterly (2023) reached a similar conclusion: “The scholarly research paints a more complicated and ambiguous picture compared to that presented by 4DWW advocates and the media.”

None of this means the four-day week doesn’t work. It means we don’t yet have the kind of evidence — longitudinal, randomized, with objective output metrics — that would let us say so with confidence. As Dennis Stolle of the American Psychological Association noted in the APA Monitor (January 2025): “Longitudinal, randomized studies are needed to better assess outcomes.”

If you’re making decisions about managing attention rather than just hours, understanding what was actually measured matters.

The Real Mechanism: Parkinson’s Law and the Elimination of Low-Value Work

If the studies are methodologically limited, why do the results keep pointing in the same direction? The most compelling explanation isn’t that humans can magically produce five days of output in four. It’s that the five-day week contains substantial padding — and compressing the timeline forces its removal.

This is Parkinson’s Law in action: work expands to fill the time available for its completion. C. Northcote Parkinson coined the phrase in 1955, but behavioral economics has since provided the empirical backbone.

Sendhil Mullainathan and Eldar Shafir, in their 2013 book Scarcity, demonstrated through lab experiments that time constraints create a “scarcity mindset” that redirects cognitive bandwidth toward essentials. Participants allocated eight hours for a task completed it in four under scarcity conditions, with 20–30% efficiency gains. Kahneman and Lovallo found in Psychological Science (2003) that consultants given flexible 10-day deadlines averaged 12 days, while those given strict 5-day deadlines averaged 4.8 days — with equivalent quality.

The mechanism shows up clearly in timeboxing research: a 2018 study in Organizational Behavior and Human Decision Processes found that professionals given 50% less time for tasks completed 68% of them on schedule with no quality reduction. A McKinsey analysis of 1,500+ firms found that loose deadlines correlated with 20–30% “slack time” filled by low-value work.

This is exactly what Juliet Schor observed in the four-day week pilots. The productivity didn’t come from working faster. It came from cutting meetings, eliminating redundant reporting, replacing synchronous check-ins with async updates, and forcing sharper prioritization of what actually matters.

What Actually Changes in Four-Day Adopters

Across the studies, the operational changes are remarkably consistent. They’re worth cataloging because they reveal where five-day weeks leak time — and because most of them don’t require a company-wide policy change to implement.

Meeting reduction is the primary lever. Microsoft Japan capped meetings at 30 minutes. Companies in the 4 Day Week Global trial reported cutting meeting volume and duration as their first move. Average knowledge workers spend 21.5 hours per week in meetings, and 78% say they attend so many that it’s hard to get their actual work done (Atlassian). Eliminating even a fraction of this is a massive time recovery.

Asynchronous communication replaces real-time interruptions. Status updates move from meetings to written channels. Decisions that previously required a 30-minute call get resolved in a thread. This isn’t just faster — it reduces the context-switching cost that fragments deep work. Research from Gloria Mark at UC Irvine shows it takes an average of 23 minutes to refocus after an interruption.

“Work about work” gets audited. Redundant reporting, unnecessary approvals, status meetings that could be emails — these activities fill time without producing output. The four-day constraint forces teams to confront them.

Prioritization sharpens. With less time available, the difference between high-value and low-value tasks becomes impossible to ignore. This is the scarcity effect Mullainathan describes: constraints don’t just remove slack, they improve decision-making about what deserves attention.

The Individual Application: Your Personal Four-Day Week Experiment

Here’s the practical takeaway: you don’t need your company to adopt a four-day week to benefit from the same mechanisms. The research points clearly at what creates the productivity effect — time constraints, meeting elimination, and forced prioritization. You can engineer all three individually.

Run a time audit first. Before changing anything, spend one week tracking how you actually spend your time in 30-minute blocks. Most knowledge workers who do this discover that 30–40% of their week goes to activities they’d classify as low-value if forced to rank them. This is the same discovery companies make when they pilot a four-day week — they just do it collectively.

Block Fridays as a “no-meeting, admin-only” day. You may not be able to take Fridays off, but you can restructure them. Decline meeting invitations for Fridays. Use the day exclusively for email processing, administrative tasks, weekly review, and next-week planning. This effectively compresses your production work into Monday through Thursday — creating the same Parkinson’s Law pressure that drives the four-day week results.

Timebox your production work into four days. This is where time blocking becomes essential. Block 2–3 hour deep work sessions on Monday through Thursday for your highest-value work. Protect those blocks the way four-day week companies protect their shortened schedule — by treating them as non-negotiable.

Daybook’s time blocking features are built for exactly this kind of constraint-setting. You can plan your week in plain text, block your deep work sessions, and see immediately whether your schedule reflects your actual priorities — or whether meetings and admin have quietly consumed the space you need for real work.

Audit your meetings ruthlessly. For every recurring meeting on your calendar, ask: does this produce a decision, transfer critical information, or solve a problem that async communication can’t? If the answer is no, decline it, convert it to a written update, or cut it to 15 minutes.

Honest Caveats: What the Research Doesn’t (Yet) Tell Us

Intellectual honesty requires acknowledging the gaps.

Selection bias is the biggest problem. Every company in these trials volunteered. As a 2023 JRC working paper notes, this means the results tell us what happens when motivated organizations try a four-day week — not what happens when the average company does. The 92% continuation rate in the 4 Day Week Global trial is impressive, but a program where resistant or unprepared companies were required to participate would likely produce very different numbers.

Industry and role variation is underexplored. Most trial participants were knowledge workers in office-based roles — the population most likely to benefit from meeting cuts and async communication. The evidence for client-facing roles, manufacturing, healthcare, or shift work is sparse. As researchers at the University of Queensland noted: “The four-day work week doesn’t suit every business model, and the extent of gains may depend on the tasks being performed.”

Long-term effects are unknown. The longest trial (Iceland) ran four years, but most are six months or less. We don’t know whether the productivity gains persist once the novelty fades, whether workload creep gradually extends the four remaining days, or whether certain types of work (creative, relational, research-intensive) suffer in ways that take years to surface.

The benefits may come from restructuring, not from the fifth day off. This is the most important caveat. Multiple analyses have concluded that the gains in these trials come from eliminating low-value work, reducing meetings, and improving focus — not from the extra day of rest per se. If that’s true, the implication is powerful but different from the headline: you can capture most of the benefits by restructuring how you work, regardless of how many days you do it.

What This Means for How You Plan Your Week

The four-day work week debate will continue. Governments will run more trials. Academics will (hopefully) conduct the randomized studies we need. But you don’t have to wait for the evidence to become airtight before applying its core insight.

The research consistently shows that when people have less time, they don’t produce less — they eliminate waste. They cut meetings. They stop doing “work about work.” They prioritize ruthlessly. These are behaviors you can adopt tomorrow.

The real lesson of the four-day work week research isn’t about the number of days. It’s about the power of constraints. Whether you use time blocking to protect your deep work hours, timebox your tasks to create artificial deadlines, or simply audit where your time actually goes — the mechanism is the same.

Work expands to fill the time available. Give it less time, and watch what gets cut. It’s almost never the work that matters.