How Work-from-Home Became Normal: Insights from a Massive U.S. Business Survey

Researchers from the U.S. Census Bureau, Stanford University, the Instituto Tecnológico Autónomo de México (ITAM), and the National Bureau of Economic Research (NBER) have teamed up to sharpen our understanding of how work-from-home (WFH) is reshaping the American labor market. Drawing on over 150,000 responses from the Census Bureau’s Business Trends and Outlook Survey (BTOS), their study offers the most detailed and timely business-level insights yet into how firms are adapting to the post-pandemic reality of hybrid and remote work. Their findings reveal not just how widespread WFH is, but also the reasons it varies so dramatically across industries, business sizes, and operational philosophies.

One Day a Week, and Holding Steady

As of early 2025, about 31% of U.S. businesses reported having at least one employee working from home for a full workday during the previous two weeks. On average, employees at these firms worked remotely 1.04 days per week, and when asked to forecast conditions five years ahead, employers projected virtually the same level of remote work, suggesting the current hybrid model has become a new equilibrium. Interestingly, this stability in remote work patterns hides striking differences across sectors. In the Information industry, almost 70% of firms have WFH employees, compared to less than 7% in Accommodation and Food Services. A similar pattern emerges with business size: large firms with over 250 employees are nearly three times more likely to have remote workers than those with fewer than 20. These disparities underscore how remote work is far from universally adopted; it thrives in industries and organizations where digital tools, management structure, and job functions allow flexibility.

Why Many Jobs Still Resist Remote Work

Despite the wide adoption in certain industries, the biggest limiting factor remains the nature of the job itself. More than 60% of businesses cited infeasibility, meaning the tasks simply cannot be performed off-site, as the main reason for not allowing remote work. This concern was dominant even in sectors like Information, where one might expect universal flexibility. At the same time, nearly 27% of firms reported no limiting factors at all, revealing a potential for wider WFH adoption if operational or cultural norms were to shift. Other reasons cited included concerns about productivity, collaboration, training, and IT security, though each of these ranked well below feasibility in importance. Notably, productivity concerns were cited by 11.7% of respondents while mentoring and teamwork challenges were raised by about 9%.

Sparse Tracking and Loose Oversight

Contrary to popular fears of an increasingly surveilled digital workplace, the study reveals that most businesses do not rigorously monitor remote employees. Around 75% reported they do not track employee activity when working from home, and about 70% do not even monitor in-office attendance. Among those who do, most rely on basic tools such as badge swipes, manager check-ins, or meeting attendance. In the Information sector, some companies reported using output-based tracking, computer activity metrics, or online meeting participation as indicators of engagement, but even here, monitoring was minimal. Return-to-office mandates are also rare, just 4.1% of businesses nationally reported enforcing a minimum number of in-person days. Even among firms that do have such requirements, fewer than 20% actively enforce them. This relaxed approach challenges narratives of aggressive RTO enforcement and suggests that, in many workplaces, the onus remains on employees to self-manage.

Mixed Reviews on Productivity, But No Collapse

One of the most debated issues around WFH is productivity. According to the study, among firms that had remote employees and answered the productivity question, 15.6% reported no discernible difference between remote and onsite work. Among the rest, 6.6% said onsite work was more productive, compared to just 2.1% who favored remote productivity, a roughly three-to-one margin nationally. The ratios varied across industries: in the tech-heavy Information sector, the difference narrowed to nearly 1:1, while in Accommodation and Food Services, 91% of businesses either did not know or found WFH inapplicable, reflecting the on-site nature of those jobs. These findings suggest that while skepticism remains among employers, few see WFH as an outright drag on performance. Instead, the prevailing view appears to be one of neutrality or cautious acceptance, especially in sectors where WFH has become entrenched.

A Blueprint for Smarter Surveys and Future Research

The research team designed the BTOS questions with precision, drawing lessons from earlier efforts like the ACS, CPS, and SWAA. They defined a WFH day as six or more hours of remote work, excluded incidental tasks like after-hours emailing, and asked firms to reflect on past practices (2019), current behavior, and future projections (2029). This longitudinal framing offers a valuable time-series perspective that other business surveys have lacked. The study also laid the groundwork for future improvements in survey design, urging the addition of more granular WFH questions in federal household surveys. The authors plan to build on this work by analyzing microdata to explore how firm age, size, and sector influence WFH outcomes, and by comparing employer-reported productivity to actual performance data from business databases.

Overall, the paper confirms that work-from-home is now a structural component of the American labor market, though its spread remains uneven and its implications complex. It urges researchers, policymakers, and businesses to embrace this complexity and refine their tools accordingly, because WFH is not just a trend, but a transformation that demands new ways of measuring, managing, and imagining the modern workplace.