Remote Work Statistics (2026): 38 Sourced Figures

This page collects figures on how many people work remotely, how often they do it and how employers respond. Every number links to the document or dataset that first published it. Most figures come from official statistics: the Bureau of Labor Statistics (BLS), the US Census Bureau, Eurostat and the UK Office for National Statistics (ONS). Others come from named research groups and one vendor, which is labeled. The closing section explains what the figures mean for employers that plan hybrid offices.

Method

The data window runs from 2019 to September 2026. The 2019 figures serve only as a pre-pandemic baseline. StatWharf used primary sources only. That means government statistical agencies, the original publisher of each survey, and raw data released by the publisher. Statistics aggregators and blogs that re-quote other sources were excluded. Each figure was checked against its source in October 2026. Figures that did not appear on a primary source were dropped. No value on this page is modeled, estimated or calculated by StatWharf. Values from data files are rounded to one decimal place.

The sources measure different things. This is the main reason remote work figures appear to conflict. The BLS counts anyone who teleworked for any hours in a reference week. The Census Bureau counts people who usually worked from home as their main way of getting to work. Gallup asks only employees in remote-capable jobs. WFH Research counts full paid days worked from home. Each table below names the measure it uses. Comparisons are valid within one source, not across sources.

How many Americans work remotely

The BLS asks about telework in the monthly Current Population Survey since October 2022. The BLS telework page defines the telework rate as the number of people who telework or work at home for pay as a share of those employed and at work. These figures are not seasonally adjusted.

In September 2026, 158,451,000 people were at work. Of those, about one in five teleworked. The split between full and partial telework is close to even. Slightly more people teleworked some hours than all hours. The telework rate for men was 20.1%, below the rate for women.

The series rose from 17.9% in October 2022 to 23.8% in October 2024. Since then it stayed between about 21% and 24%. The rate in September 2025 was 22.3%. The BLS published no value for October 2025 because of the 2025 lapse in appropriations.

Remote work before and after COVID

The Census Bureau’s American Community Survey (ACS) asks how people usually got to work in the prior week. “Worked from home” is one answer. This measure captures people who mainly work from home. It misses most hybrid workers. The ACS figures are therefore lower than the BLS telework rate.

The Census Bureau reported that the number of people who primarily worked from home tripled between 2019 and 2021. In 2021 that share equaled 27.6 million people. The share then fell to 13.8% in 2023 and 13.3% in 2024. At the check date, the Commuting at a Glance page had not yet shown a 2025 figure.

The American Time Use Survey uses a different method. Respondents report what they did on one diary day. The two percentages in the table add up to more than 100 because some people worked in both places on the same day. Education makes a large difference. Only 19% of workers with a high school diploma and no college did some work at home on days worked. Employed women (38%) were more likely than employed men to work at home.

Which jobs and industries work remotely

Remote work is concentrated in office-based occupations. The BLS publishes the telework rate by occupation and industry each month. The figures below are from September 2026.

Professional and technical services had a telework rate of 56.9%. Retail trade had a rate of 10.4%. Self-employed workers teleworked at a rate of 36.9%, well above the rate for government workers. The pattern matters for office planning. A firm with a large finance, technology or professional staff often has many empty desks on a given day.

Hybrid, fully remote and on-site

Two sources describe the mix of work arrangements. Gallup surveys US employees in remote-capable jobs. WFH Research runs the Survey of Working Arrangements and Attitudes (SWAA), a monthly survey led by economists Jose Maria Barrero, Nicholas Bloom, Shelby Buckman and Steven J. Davis. Its target population is US residents aged 20 to 64 who earned at least US$10,000 in the prior year.

Gallup defines a hybrid employee as one who works remotely 10% to less than 100% of the time. In the November 2025 data, 33% of remote-capable employees preferred to work exclusively remote. Only 6% preferred on-site work. Gallup also reports engagement by arrangement. In May 2026, 37% of exclusively remote employees were engaged, compared with 29% of remote-capable on-site employees.

The SWAA figures cover all full-time wage and salary employees, not only those in remote-capable jobs. In the same period, 62% of full-time employees worked fully on site. The SWAA also reports that employer plans for work from home “hover around 1.3-1.5 days per week”. By education, employees with a graduate degree worked from home on 33.0% of full paid workdays in 2026, against 19.2% for those with a high school degree or less. The 2026 values average the months available at the October 2026 release.

Employers, offices and job postings

Three sources describe the employer side. The Flex Index tracks published office policies at US firms. It is published by Work Forward, a think tank and advisory group, in partnership with Boston Consulting Group. Its Q3 2025 report covered over 9,000 US firms. Kastle Systems, a building security vendor, measures office entries from access card data. Indeed Hiring Lab publishes the share of job postings that mention remote or hybrid work.

The Flex Index found that 42% of US firms use a structured hybrid model. Under that model, staff are not in the office full time, but the firm sets a minimum. The report also states that required office time “climbed 12% since early 2024, but attendance increased only 1-3%”. Smaller firms are more flexible. The report found that 67% of companies under 500 employees are fully flexible.

Kastle tracks keycard, fob and app entries from over 300,000 users in ten major metro areas. Each figure compares entries with a pre-pandemic baseline from early 2020. Kastle states that the Barometer is not a national statistical sample. In the same week, Class A+ buildings averaged 77.5% occupancy. The gap between the Tuesday peak of 65.5% and the weekly average of 55.4% is the core planning problem for a hybrid office. An office sized for the peak day sits partly empty on the other days.

Indeed counts a posting as remote or hybrid when its text includes related keywords, such as “remote work” or “hybrid role”. The figures are seven-day trailing averages. A year earlier, on August 31, 2025, the US share was 8.2%. In the United Kingdom, the share on August 31, 2026 was 16.3%. The US share on the same date was 8.4%.

Remote work outside the United States

Eurostat publishes working-from-home shares from the EU Labour Force Survey. It reports two groups: people who usually work from home and people who sometimes do. The ONS measures arrangements in Great Britain with the Opinions and Lifestyle Survey. It asks working adults whether they worked from home or travelled to work in the past seven days.

The EU share of people who usually work from home was 5.4% in 2019. It reached 13.3% in 2021 and stayed near 9% from 2023 to 2025. The share who sometimes work from home rose each year from 2020 to 2025. In 2025, 77.0% of employed persons aged 15 to 64 in the EU never worked from home.

In Great Britain, 38% of working adults travelled to work only in August 2026. The hybrid share was 28% in May 2026. August falls in the summer holiday period, and the ONS publishes confidence intervals for each estimate. A monthly change of a few points is not always significant.

What the figures show

Three points stand out. First, remote work in the United States is stable. The BLS telework rate stayed between about 21% and 24% in the last two years. It stood at 21.9% in September 2026. That equals about 34.7 million people.

Second, hybrid is the most common arrangement for people in remote-capable jobs. Gallup puts it at 52% of remote-capable employees. The Flex Index found that the average US firm requires 2.87 days a week in the office. Office attendance rose more slowly than the policies that require it.

Third, office use is uneven across the week. Kastle reports a Tuesday peak of 65.5% against a weekly average of 55.4%. A firm that plans desks around attendance rather than headcount needs daily data for its own buildings. Desk booking software lets staff reserve a desk for the days they attend and records actual use. Features and published pricing are compared on the StatWharf page for desk booking software.

Where to go next

Other StatWharf data pages are listed in the research directory. The next update follows two releases: the BLS telework tables for October 2026 and the 2025 American Community Survey figures on the Census commuting page. Send corrections to any figure to the editorial address below.

Factual corrections are free: editorial@statwharf.com.