Working from Home: Before and After the Pandemic

Abstract

Hilary Silver on pandemic trends.

People have always worked at home. The spatial separation of workplace and residence, fueled by the Industrial Revolution and reinforced by zoning codes, has never been complete. Employers build worker housing, small businesspeople live above the store, and generations of women take in piecework while tending to their families. The family farm persists.

In the United States, as agriculture contracted and factories and offices expanded, paid work in the home did dwindle over time. This trend began to reverse in the 1980s with the increasing application of information technology to many types of service work. But working from home really took off with the coronavirus pandemic.

Trends in homework

Decennial censuses show home-working was declining, both absolutely and relatively, from 4.7 million in 1960 to 2.2 million in 1980 (at right, top). Between 1980 and 1990, however, the Census Bureau reports a 56% rise, to 3.4 million people. In the 2000 census, nearly 4.2 million, or 3.2% of American workers, labored where they lived. Since then, the American Community Survey (ACS) 5-year estimates document a continuous rise. By 2020, over 11 million people, or 7.3% of the U.S. labor force, reported their primary job was mostly performed at home. The 2021 ACS one-year estimate was 27.6 million people primarily working from home nationwide, or 17.9% of employees.

The Census Bureau’s Household Pulse Surveys (HPS) were designed to collect data quickly and efficiently on the impact of the coronavirus pandemic on American households. From August 2020 to July 2021, they asked, “Did any adults in this household substitute some or all of their typical in-person work for telework because of the coronavirus?“ Responses documented the rise in remote work, followed by a decline and levelling off above pre-pandemic levels (second Figure below).

The U.S. Bureau of Labor Statistics (BLS) asked a slightly different question in its Current Population Survey (CPS): “At any time in the last four weeks, did you telework or work at home for pay because of the coronavirus pandemic?” Approximately 20% of respondents answered affirmatively as pandemic lock-downs entered their sixth month, though that portion fell to 17% by November 2021. The BLS American Time Use Survey reported that in 2021, 38% of employed persons did some or all of their work at home on days they worked, for an average of 5.6 hours, and 68% did some or all of their work at their workplace, for 7.8 hours on average. This compared to 24% at home and 82% onsite, respectively, in 2019.

Number of people working at home (principal place of work in primary job)

Note: Prior to the 2010 Census, decennial censuses asked about commuting. Commuting data is no longer included in censuses, but rather in the ACS. Five year ACS estimates are averaged over the years included in the range, so the latest ACS mostly represents pre-pandemic practices. Source: U.S. Census Bureau. Decennial Censuses and the five-year American Community Survey 2010-15 and 2016-2020.

U.S. teleworkers due to the coronavirus pandemic, by week (August 2020-July 2021)

Note: Weeks 13-27 refers to “Some adult in household substituted some or all of their typical in-person work for telework because of the coronavirus pandemic.” Weeks 28-32 refer to “some adult in household teleworked in the last 7 days” and “because of the coronavirus pandemic,” excluding those teleworking not because of the pandemic.

Source: Household Pulse Surveys Week 13 August 19-31, 2020, through Week 27 March 17-29, 2021, and Week 28 April 14-26, 2021, through Week 31 July 23-August 8, 2021.

Government figures on homeworking diverge from those in private surveys. For example, Gallup’s State of the American Workforce Report, found that 43% of employees spent “at least some of their time” working remotely in 2016, up from 39% in 2012. Gallup’s June 2022 poll found 70 million or 56% of fulltime employees said they can do their job remotely. Of those, three in ten (21 million) currently worked exclusively at home, down from four in ten in February 2022, while another five in ten (35 million) worked at home part of the time.

The discrepancies in official statistics have a number of sources. First, the Census Bureau has concentrated on people’s usual means of transportation to their main job. Transportation data show a precipitous drop in commuting in the early pandemic, with taxis the only form of transit that increased. But many people have secondary, home-based jobs or businesses, or perform part of their primary job at home and part elsewhere. Some surveys confine attention to full-time, continuously employed, wage-and-salary employees, excluding part-time, moonlighting, and self-employed workers. Second, sample sizes, methodologies, and terminology differ. Online surveys oversample the more educated and can capture responses from business owners or employers right alongside employees. Vague language is a further complication. Using synonyms like remote work or telecommuting is consequential, as not everyone working remotely is doing so at home or via information technology.

Workers who teleworked or worked from home in the last 7 days, by days of home work

Source: U.S. Census Bureau. Household Pulse Survey. Weeks 46-48

A third reason for figures to vary is the unit of analysis. The new HPS surveys households, not individuals, so it undercounts homeworkers living with other homeworkers. The Census Bureau’s 2022 Annual Business Survey is based on employers, finding the percentage of businesses with employees who worked from home rose from 28.1 in 2019 to 41.9 in 2020, a huge surge even before the pandemic.

Finally, changes in survey wording over time complicate the data. HPS surveys, for instance, initially focused on substituting telework due to COVID-19, but in the summer of 2022, began to ask simply how many days in the last week people had teleworked or worked from home (see above). In these data, we see that since the early 2022 Omicron surge, the number of homeworkers stabilized. As of July 2022, 67,977,728 people, or more than a quarter of the U.S. labor force, were working at home at least one day a week. Over half of those (38 million individuals) reported working at home five or more days in the week (far exceeding the 11 million Americans who responded in the 2016-2020 wave of the Census Bureau’s American Community Survey that their primary place of work was their home). Still, full-time home-working is slowly declining compared to hybrid and workplace-based work.

Surveys conducted by non-government agencies do seem to confirm the Yes, 5 or more days trends seen in government data. For example, the monthly online Survey of Working Arrangements and Attitudes reported that less than 5% of days were worked at home before the pandemic, rising to a high watermark on May 1, 2020, with fully 61.5% of full-paid working days worked at home, then a decline to 37% by the end of the year and 29.5% by August 2022, holding steady since. (European data, by the way, also found increases in working from home, though the rises were relatively more modest: Eurofound’s fifth round of the Living, Working, and COVID-19 e-survey of EU workers reported just over 33% working exclusively from home in mid-2020, falling to 12% by spring 2022, and a fairly constant rate of hybrid home-working of 18%.)

A sociological profile of homeworkers

What do contemporary homework-ers look like? In the past, homeworkers were thought of as low-paid contingent workers, marginal small business owners, or independent contractors. Nowadays, they tend to have a slightly higher social status and be more engaged in privileged, well-paid occupations and knowledge industries than on-site workers (see next page). Homeworkers earn more and have more assets than the average worker. In general, they are older, non-Hispanic White, highly-educated, and in better health. Indeed, the Centers for Disease Control and Prevention’s Morbidity and Mortality Weekly Report for November 6, 2020 stated that teleworking significantly reduced the likelihood of testing positive for COVID-19.

Social profile of U.S. labor force and homeworkers

Social Characteristics	Total Labor Force	Worked from Home
Workers 16+ years old	153,665,654	11,153,095
Female	47.14%	51.03%
Male	52.86%	48.97%
Median Age	41.7	45.9
Workers 65+ years old	5.88%	9.60%
% Foreign Born	17%	15%
Speak English less than very well	8.4%	4.94%
Hispanic or Latino workers 16 years and over	17.3%	11.3%
White alone workers 16 years and over	72.1%	76.4%
White alone, not Hispanic or Latino workers 16 years and over	62.2%	70.7%
Black or African American alone workers 16 years and over	11.7%	7.9%
American Indian and Alaska Native alone workers 16 years and over	0.7%	0.4%
Asian alone workers 16 years and over	6.0%	6.9%
Two or more races, workers 16 years and over	4.3%	5.5%
Some other race alone workers 16 years and over	0.05	0.027
% Below poverty line	24%	17%
Median Earnings in past 12 months	$40,122	$51,673
% Householders living in owner-occupied housing units	66%	73%
% No vehicles available	4%	4%

Source: U.S. Census Bureau, American Community Survey 2016-2020, five-year estimates.

Eleven percent of homeworkers are Hispanic or Latino American, in contrast to 17% of the American labor force. Eighteen percent of homeworkers are Black, compared to 12% of U.S. workers, and 7% are Asian. According to a November 2020 analysis of data from the second phase of the HPS by the UCLA Center for Neighborhood Knowledge, racial disparities in access to remote work were explained mainly, but not completely, by differences in income and education, age, and location. Similarly, with CPS data, a National Institute for Occupational Safety and Health (NIOSH) study found occupation explained more than 60% of the total effect of race on tele-working for Black and Hispanic workers.

Historically, women were more likely to work at home than men. During the early Industrial Revolution, women performed piece work at home and were paid a fixed rate for each unit produced or action performed instead of an hourly wage. This way, they could combine child care and earning in one place. In 1980, 53.6% of homeworkers were female; in 2020, just before the pandemic, 51% were. Coronavirus kept more men home.

If gender is defined more subtly, 49.1% of teleworkers were female “at birth.” According to the June 2022 HPS, 32% of cisgender males, 29% of cisgender females, and 30% of the 2 million transgender respondents teleworked in the last 7 days, compared to 30% of the entire labor force. When asked about one’s “sexual orientation” rather than gender, 39% of LGBT respondents teleworked at least one day in the last week, considerably more than the 30% of straight respondents. It is important to recall, however, that the HPS is still experimenting with measures of gender and sexual orientation and so, these figures should be regarded with caution.

Home work today, unlike in the past, is associated with higher than average pay. The ACS data at left show these workers’ median earnings in the past twelve months in 2020 inflation-adjusted dollars was $51,673 compared to the national median of $40,122. Looked at another way, 24% of the U.S. labor force lives below the poverty line, but just 17% of homeworkers do.

Homeworkers have more assets, too. Perhaps unsurprising is the fact that nearly three-quarters of homeworkers own their homes compared to two-thirds of the overall labor force. More unexpected is the finding that, even though they do not commute to work, only 4% of homeworkers do not own a car, the same percentage as among the general labor force. In fact, like workers overall, three-quarters of their households own two or more cars.

Although pre-pandemic time use data show average full-time workers spent more time alone when they teleworked, which led to an increase in loneliness for some, recent ACS demographics imply that homeworkers are not especially isolated. Their households are somewhat larger than households of those who work on-site. Only 5% of homeworkers live alone, compared to 10% of non-homeworkers. In fact, 27% of teleworkers lived in two-person households, compared to 33% of non-teleworkers. Teleworkers are more likely to be married (60%) compared to non-teleworkers (51%), but are about as likely to have children (42% vs. 40%). Forty percent of the married households and 38% of the households with children teleworked. This implies that child care is not the only reason people are working at home.

An analysis of pre-pandemic time use diaries by Sabrina Pabilonia and Victoria Vernon examined hours of child care by homeworking mothers and fathers. While they saved time on commuting and grooming, fathers increased time on primary child care, while women increased household chores. Since the pandemic, these authors found with the 2021 American Time Use Survey, women were more likely than men to do some or all of their work at home on work days—42% of women, 35% of men. Parents in dual-career couples working from home alone increased childcare hours compared to on-site workers, but mothers also reduced paid work hours. When both parents worked from home, mothers and fathers maintained their paid hours and spent more time on childcare. Parents working from home did equally more household chores, although fathers working from home alone did more compared to on-site working fathers.

Class of worker of U.S. labor force and homeworkers

Industrial distribution of labor force and homeworkers

Occupational distribution of labor force and homeworkers

Workers 16 years and older

Source: U.S. Census Bureau, American Community Survey five-year estimates 2016-2020

When examining homeworkers’ employment, one distinction stands out: they are more likely to be self-employed. This has long been the case. In 1990, 54% of those working at home were self-employed, 10 times the rate of self-employment among those who worked away from home, while 17% were government workers, compared to 6% of non-homeworkers. In 2019, one-fourth of those working mainly from home were self-employed compared to 6% of the overall U.S. labor force. Only 7% of home-workers are government employees.

Homework does not compose a majority of any significant industry, but as the 2016-20 ACS shows, 11% of those working in agriculture, forestry, fishing, hunting, and mining; 14% in the information sector; 13.6% in finance, insurance, and real estate; and 15.5% in professional services work primarily at home. Indeed, compared to the 7.3% of the American labor force working at home just before the pandemic, home-workers are especially overrepresented in professional and financial services. While remote work increased in most industries, it is much more common in those with better educated and better paid labor (for instance, technology, communications, professional services, and finance and insurance workers).

Similarly, occupations that work with goods, whether producing, maintaining, or moving them, are much less likely to be performed at home. Three-fifths of all homeworking occupations are in management, business, science, and the arts.

Finally, the industrial distribution of homework is reflected in its bicoastal geography. Based on both the May 2020 and February 2021 CPS, workers in the New England, Mid-Atlantic, and Pacific regions, were most likely to be teleworking, while those in East South Central states were significantly least likely.

Prospects for homework

As the pandemic subsides, many firms are calling employees back to the workplace. Bosses worry about productivity and creativity. Two-thirds of executives in PwC’s Remote Work Survey said a typical employee should be in the office at least three days a week to maintain a distinct company culture. Cities are worrying about vacant office space and shuttered downtown businesses. But polls like Gallup’s or McKinsey’s American Opportunity Survey and job search data from Lightcast suggest a majority of workers would prefer to labor remotely all or part of the week, with many remote workers even willing to quit if forced to return to the workplace full time. Long commutes, care responsibilities, and lifestyle contribute to the resistance. Given the leveling off of homeworking trends, remote work—at least for part of the work week—may be here to stay.

Faced with worker shortages and the cost of office space, employers are also offering telework to broaden recruitment pools, reduce turnover, and moderate employee compensation. Around 40% of U.S. firms surveyed over the last year reported they had expanded opportunities to work from home or other remote locations (see chart). Eurofound’s e-survey found employers plan to permit an average of 0.7 days per week at home after the pandemic, although workers, especially women, parents, and those with longer commutes, want 1.7 days.

Days per week of working from home that employers are planning for after the pandemic, as of each survey wave

Source: Survey of Working Arrangements and Attitudes, Jose Maria Barrero, Nicholas Bloom, and Steven J. Davis, 2021. “Why working from home will stick,” National Bureau of Economic Research Working Paper 28731. These data are made available under the CC-BY 4.0 license https://creativecommons.org/licenses/by/4.0/

There are limits. Researchers Jonathan Dingel and Brent Neiman have estimated that 37% of jobs in the United States can be performed entirely at home. The Global Survey of Working Arrangements asked full-time workers in 27 countries whether their job could theoretically be done from home. Of the 22% of workers in totally teleworkable jobs, 43% worked exclusively at home already, but of the 28% who said their jobs are partially teleworkable, only 8% were totally homeworking (another 35% worked partially from home).

The downsides of homeworking should not be underestimated, though. It fragments the labor force and impedes teamwork and unionization. It can impose long hours and multi-tasking on employees, as well as parental overload or neglect of children or elders confined to the home. Rather than work flexibility, it can extend the workday and encroach into leisure and family time and space. An April 2022 Conference Board survey found 47% of remote workers in the United States were concerned about the blurred boundaries between their jobs and personal lives. Workers may demand a “right to disconnect.” Indeed, a new law in Portugal bans out-of-hours employer contact, requires employers to contribute to the increase in utility costs linked to working from home, and restricts remote monitoring of workers. At the same time, younger workers may feel cut off from pre-pandemic office life, unable to acquire the social capital and mutual learning needed to get ahead. The surge in loneliness and mental illness resulting from the pandemic lockdowns, social distancing, and greater reliance on social media may also have some home-workers simply pining for the office water cooler.

As the pandemic eases, all these factors will influence the desirability and practice of homework. The traditional separation of public and private spheres may yet reappear.

Biography

Hilary Silver is in the sociology department at George Washington University. She has studied paid home work since the 1990s.