Five years after the pandemic pushed office workers to their kitchen tables, remote work has settled into wildly different patterns around the globe. A college-educated employee in the United States or the United Kingdom now spends nearly two days a week working from home on average. Their counterpart in South Korea or Japan often works fewer than one. Fast internet, plentiful desk jobs, and modern digital tools are common to both groups. So what actually explains the gap?
A new analysis published in PNAS points to a less obvious answer: culture. Specifically, the degree to which a society emphasizes individual autonomy over group cohesion appears to be the single strongest predictor of how much its workers do their jobs from home.
The question behind the data
Researchers have documented big international differences in remote work for a few years now, but comparing across countries has been difficult because most national surveys ask about work arrangements in slightly different ways. To get around that, Pablo Zarate of Princeton University and colleagues at Stanford, the ifo Institute in Munich, King’s College London, and Instituto Tecnológico Autónomo de México have been running the Global Survey of Working Arrangements, a harmonized survey that asks the same questions of workers in dozens of countries.
The latest wave, conducted between November 2024 and February 2025, covers 14,427 full-time, college-educated workers across 37 countries, with samples balanced by age, gender, and education. That gave the team enough consistent data to test competing explanations for why WFH rates vary so widely.
What might explain the gap
The intuitive explanations focus on economic structure. Countries with lots of office-based knowledge work and strong digital infrastructure should have more remote work, while economies weighted toward manufacturing, agriculture, or retail should have less. Population density might matter too, since long commutes make working from home more attractive. And the severity of pandemic-era lockdowns could have produced lasting changes by forcing companies to experiment with remote setups.
But the authors note that these factors don’t fit the pattern well. Several East Asian economies have world-class digital networks, dense cities, and large white-collar sectors, yet remain among the lowest in remote work adoption. Something else seems to be going on.
Their hypothesis draws on a long tradition in cross-cultural research. Remote and hybrid arrangements mean less direct oversight from managers, fewer in-person interactions with coworkers, and more personal discretion over how tasks get done. Societies that place a high value on individual freedom and self-reliance might find that setup more comfortable. Societies that emphasize hierarchy, group cohesion, and face-to-face relationships might resist it.
To measure culture, the researchers used Geert Hofstede’s individualism index, a widely cited score developed from decades of cross-national survey work in social psychology.
How the analysis worked
The team first plotted each country’s average WFH days per week against its individualism score. The relationship was strong and positive, with a correlation of 0.539. Moving from a country near the bottom of the individualism distribution (China) to one near the top (the Netherlands) was associated with an increase of nearly half a day of remote work per week, or about 38 percent of the average WFH rate across countries.
To check whether individualism was really doing the work or just standing in for other differences between countries, the researchers ran a regression that included six factors together: the individualism index, the share of local jobs suitable for remote work (estimated from each country’s industry mix), real GDP per capita, population density, average commute time, and the strictness of pandemic lockdowns.
Individualism remained the largest and most statistically significant predictor. The industry mix showed a weaker association, roughly half as large, and was only marginally significant. GDP per capita, density, commute times, and lockdown stringency did not show statistically significant relationships once the other variables were included. The team also tested housing patterns (such as the share of single-family homes) and found no meaningful association.
The six variables together explained about 38 percent of the variation in WFH rates across countries. Individualism on its own explained roughly 29 percent, more than the combined contribution of the other five factors. The pattern held when the researchers analyzed men and women separately.
What the findings mean
The authors interpret the results as evidence that cultural values shape how societies adopt new ways of working, even when the technology and job composition would allow more flexibility. In their words, “the future of work will be shaped not only by technology and policy but also by cultural forces that influence how individuals and organizations balance autonomy, trust, and coordination.”
They also connect their findings to a recent working paper by other researchers showing that immigrants living in the same host country tend to work from home at rates that reflect the individualism of their country of origin. That study looks within a single labor market; the current analysis extends the pattern across national labor markets using direct survey measures rather than proxies.
A few caveats are worth keeping in mind. The analysis is observational, so it identifies associations rather than causal effects. Hofstede’s individualism index is one of several cultural measures, and the results reflect that particular scale. The sample also focuses on college-educated, full-time employees, the group most likely to have jobs that can be done remotely in the first place. Patterns among lower-wage or less-educated workers could look different.
Implications for firms and policymakers
For multinational employers, the results suggest that a remote-work policy imported wholesale from a headquarters in California or London may run into friction in offices where workers and managers expect more in-person coordination. Cultural fit, rather than just infrastructure or job type, appears to influence how remote setups are received.
For policymakers weighing the trade-offs of hybrid work, including its effects on urban real estate, commuting, and labor force participation among caregivers and people with disabilities, the study suggests that national outcomes will not converge simply because the technology is available. Countries starting from different cultural baselines may continue to settle at different levels of remote work for years to come.
