June 2026

From typewriters to telecommuting: insights from the National Longitudinal Survey of Youth 1979 on four decades of remote work

Drawing on decades of longitudinal data, this article examines how telework trends have evolved over the working lives of the National Longitudinal Survey of Youth 1979 (NLSY79) cohort. The analysis highlights changes in the frequency of working from home, with particular attention to the impact of the COVID-19 pandemic. Telework patterns within the cohort are also examined by sex, educational attainment, race and ethnicity, and region.

In the 1980s, when typewriters were still common in the workplace and personal computers were only beginning to replace them, it was difficult to find jobs that allowed for telework.¹ In those days, the working lives of the National Longitudinal Survey of Youth 1979 (NLSY79) cohort were just beginning. The survey began in 1979, when respondents were 14 to 21 years old, and has followed these individuals for more than four decades.

This article tracks the NLSY79 cohort over time to examine how the prevalence of telework has evolved—from early career development in the 1980s, through midcareer stages in the 1990s and 2000s, and into later career phases in the 2010s and 2020s. (See table 1.)

The NLSY79 survey provides a rich dataset for analyzing long-term labor market experiences across the life course of a specific birth cohort.² The survey follows a nationally representative sample of Americans born between 1957 and 1964.

Table 1. Ages of the National Longitudinal Survey of Youth 1979 cohort over time

Year	Survey round	Ages of sample cohort
1979	1	14–21
1990	12	26–33
2000	19	36–43
2010	24	46–53
2020	29	56–63
2022	30	57–64
Source: U.S. Bureau of Labor Statistics.

Trends in teleworking incidence for private and public sector workers

Over the years, telework has consistently been present—though, not dominant—in the labor market experiences of the NLSY79 cohort. For decades, a modest share of workers engaged in some form of remote work. Prior to 2020, telework adoption increased incrementally and followed similar trajectories in both the private and public sectors. However, the data reveal a marked surge in the prevalence of telework beginning in 2020, coinciding with the onset of the COVID-19 pandemic. During this late stage of cohort members’ careers, when they were 56 to 63 years old, private sector workers and government workers reported that they teleworked for 36.8 percent and 30.9 percent of their work time, respectively. (See chart 1.)

Trends in the amount of time spent teleworking

The average proportion of work time spent teleworking remained relatively flat for many years and increased significantly in recent years—mainly between 2018 and 2020. Among all employed workers in the NLSY79 cohort, the share of total work time spent at home rose from 2.0 percent in 1988 to 8.7 percent in 2020—a 6.7-percentage-point increase.

Among those who reported at least some telework, the growth was even more pronounced, with the average share of work time spent teleworking rising from 11.2 percent to 25.3 percent from 1988 to 2020.

Telework remained relatively uncommon between 1988 and 2010, with the share of work time performed from home declining slightly from 11.2 percent in 1988 to 9.7 percent in 2010 among those with at least some time working from home. By 2018, this share had increased to 13.5 percent, indicating modest growth in telework prior to the COVID-19 pandemic. The most dramatic increase occurred in 2020, when the share of work time performed from home rose to 25.3 percent, coinciding with widespread pandemic-related workplace disruptions. Between 2020 and 2022, this telework share declined to 21.7 percent, suggesting some reversion as workplaces reopened, although levels remained well above those observed prior to 2020. (See chart 2.)

Full-time telework begins to take off in 2020

Among employed workers, the vast majority of NLSY79 workers have historically reported working 2 hours or less per week from home. For example, in 1988, 3.7 percent reported a small amount of telework that didn’t exceed two hours per week, and 82.4 percent reported no telework hour at all. This pattern remained stable from 1988 through 2008, after which a slow and noticeable shift began to emerge. In 2020, the share of workers who reported no telework and having worked 2 hours or less dropped to 68.6 percent. This shift was driven by full-time teleworkers, who saw their share rise from 3.9 percent in 2018 to 15.3 percent in 2020—a surge of 11.4 percentage points. That surge represented the most dramatic change observed over the four-decade span. The occurrence of full-time and half-time telework increased while the prevalence of smaller amounts of telework hours declined (those working fewer than 8 hours).

The share of full-time teleworkers declined 3.9 percentage points from 2020 to 2022, but several categories of part-time telework continued to rise, indicating the emergence of hybrid arrangements. Interestingly, the share of workers with minimal telework time did not return to prepandemic levels, suggesting a structural shift toward more widespread telework adoption, even if not for full-time workers. (See chart 3.)

Distribution of hours teleworked is bimodal

The distribution of telework hours among NLSY79 participants has historically been bimodal, with a large share of teleworkers reporting either very limited amounts of telework or working remotely nearly full time. In particular, much of the lower peak in the distribution is driven by respondents who report teleworking two hours or less per week, suggesting that telework for many workers historically consisted of occasional or supplemental work from home rather than a regular work arrangement. By contrast, the upper peak reflects workers whose jobs allowed them to perform most of their work remotely, indicating a smaller group for whom telework functioned as a primary mode of work. This pattern highlights longstanding heterogeneity in how telework has been incorporated into jobs, even prior to the pandemic.

By 2018, this pattern began to shift as intermediate levels of telework became increasingly common, indicating a move away from a strict dichotomy of occasional teleworker versus full-time teleworker.

In 2020, workers reporting near–full-time telework constituted the largest share of teleworkers, reflecting the widespread shift to remote work during the COVID-19 pandemic; specifically, individuals working from home between half time and full time accounted for 54 percent of all teleworkers. (See chart 4.)

By 2022, the distribution shifted again, with growth concentrated in the intermediate categories—specifically, workers reporting telework of more than 8 hours per week but less than full time, including both those working between 8 hours and half time and those working between half time and nearly full time from home. This change indicates an increase in part-time telework arrangements relative to both minimal and full-time telework.

For many workers capable of performing their jobs remotely, telework has evolved from a marginal option to a routine component of employment.

How the composition of telework hours shifted

A transition matrix like table 2 provides the probability that an event in row category A moved to column category B. For example, among those who reported no telework in 2018 (row category A), 81 percent remained in the no telework category in 2020 (column category B). This allows us to see the transition of workers between 2018 and 2020 across categories of hours teleworked. While there was considerable persistence in telework patterns between 2018 and 2020, the table also shows notable transitions into full-time telework from most starting categories.

Workers who teleworked for just 2 hours or less in 2018 exhibited substantially greater volatility in their work arrangements as 37.4 percent reported no telework by 2020. This pattern suggests that limited telework prior to the pandemic often reflected informal or episodic arrangements rather than jobs well suited for sustained remote work. In contrast, telework arrangements that were more intensive proved far more durable. Among workers who reported full-time telework in 2018, only 4.8 percent transitioned to no remote work by 2020, underscoring the persistence of telework in occupations and jobs where remote work was feasible on a large scale. (See table 2.)

For full-time teleworkers in 2018, 81.5 percent remained full-time teleworkers in 2020, and 56.5 percent of those in the half- to full-time range moved into full-time remote work by 2020. The data show a consistent relationship: the greater the degree of telework in 2018, the more likely individuals were to either remain at that level or move into higher categories by 2020.

Table 2. Work from home category transition matrix, 2018 and 2020

Row category (2018)	Column category (2020)
Teleworked hours	No telework	2 hours or less	Between 2 and 8 hours	Between 8 hours and half time	Between half and full time	Full time
No telework	81.0	2.0	2.6	2.5	2.8	8.9
2 hours or less	37.4	16.4	15.0	10.1	5.7	15.5
Between 2 and 8 hours	19.0	6.1	23.5	15.0	7.1	29.3
Between 8 hours and half time	11.0	3.1	9.5	32.1	9.2	35.1
Between half and full time	11.0	0.0	5.6	7.3	19.6	56.5
Full time	4.8	0.0	5.4	3.7	4.7	81.5
Source: U.S. Bureau of Labor Statistics.

While full-time telework arrangements declined in 2022, there remains a strong relationship between hours of telework in 2020 and 2022. Many workers who were either full-time teleworking or teleworking 2 hours or less in 2020 transitioned into more intermediate levels of telework by 2022. Persistence is strongest among workers in the lower and middle ranges of telework frequency: 45.2 percent of those teleworking between 2 and 8 hours and 46.6 percent of those teleworking between 8 hours and half time in 2020 remained in the same categories in 2022. By contrast, transitions were more common among those teleworking between half time and full time, suggesting greater fluidity at higher—but not full-time—levels of remote work. (See table 3.)

Table 3. Work from home category transition matrix, 2020 and 2022

Row category (2020)	Column category (2022)
Teleworked hours	No telework	2 hours or less	Between 2 and 8 hours	Between 8 hours and half time	Between half and full time	Full time
No telework	90.4	2.5	3.5	1.5	0.7	1.4
2 hours or less	44.3	23.8	22.6	3.1	2.9	3.3
Between 2 and 8 hours	28.1	8.4	45.2	7.7	4.1	6.5
Between 8 hours and half time	17.7	4.6	17.8	46.6	10.2	3.1
Between half and full time	20.7	3.3	12.5	28.1	24.0	11.3
Full time	11.1	0.6	5.1	10.5	16.1	56.6
Source: U.S. Bureau of Labor Statistics.

How did demographics change for teleworkers over time?

Telework patterns also vary systematically across demographic and geographic groups. To highlight these differences while accounting for year-to-year volatility, the analysis shows average telework outcomes across three periods: 1988 to 2008, 2010 to 2018, and 2020 to 2022. These intervals roughly correspond to early-career, mid-career, and late-career stages for the NLSY79 cohort. Examining averages over these timeframes allows for clearer comparisons by sex, educational attainment, race and ethnicity, and region, and provides insight into how access to and intensity of telework evolved for different groups as the cohort aged and as remote work became more technologically and institutionally feasible.

Differences in telework by sex

During the earlier stages of the NLSY79 cohort’s careers (1988–2008), men accounted for 71.6 percent of full-time teleworkers. This imbalance diminished as the cohort aged, and by 2010–18, the share of men in full-time telework dropped to 56.4 percent. From 2020 to 2022, their share of full-time telework declined slightly to 54 percent. (See chart 5.) These shifts may reflect broader changes in the workforce over time, including increasing participation of women in occupations compatible with telework and evolving patterns of labor force engagement across sexes, rather than any specific causal effect measured in the survey.

Women’s share in the half- to full-time and full-time telework categories rose steadily over the years, indicating broader inclusion in remote work practices. These shifts suggest that as telework became more normalized, sex differences in access and usage became less pronounced, particularly in near–full-time roles.

Differences in telework by educational attainment

Between 1988 and 2008, employed workers with a high school diploma in the NLSY79 cohort made up the majority of both the full-time (57.8 percent) and no telework groups (59.6 percent). Over time, the share of workers with a bachelor’s or master’s degree grew in the full-time telework category, while high school graduates remained the majority in the no telework group. (See chart 6.)

This shift portrays changes in occupational structure and differences in telework eligibility by skill level, with higher-educated workers becoming more likely to hold jobs conducive to remote work. Educational attainment thus became a more salient factor in determining access to full-time telework, especially in the 2010s and beyond.

Differences in telework by race and ethnicity

The racial and ethnic composition of telework categories has remained relatively stable throughout the decades. Between 1988 and 2008, 88.4 percent of full-time teleworkers in the NLSY79 cohort were classified as non-Hispanic/non-Black, a category that also includes individuals identified as mixed race due to small sample sizes. By 2022, this figure had declined slightly to 84.2 percent. (See chart 7.) The survey does not allow further separation within this combined group.

In contrast, the average no telework group in 2020–22 included 15.1 percent Black and 7 percent Hispanic workers—higher than their representation in the full-time telework group.

Given that around 60 percent of the NLSY79 cohort identifies as non-Hispanic/non-Black, these figures suggest persistent disparities in access to telework opportunities by race and ethnicity.

Differences in telework by region

Telework patterns also vary across geographic regions. For consistency with prior U.S. Bureau of Labor Statistics analyses, regions are defined using the U.S. Census Bureau classification:

• Northeast: Connecticut, Maine, Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, and Vermont;

• Midwest: Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Missouri, Nebraska, Ohio, North Dakota, South Dakota, and Wisconsin;

• South: Alabama, Arkansas, Delaware, District of Columbia, Florida, Georgia, Kentucky, Louisiana, Maryland, Missouri, North Carolina, Oklahoma, South Carolina, Tennessee, Texas, Virginia, and West Virginia; and

• West: Alaska, Arizona, California, Colorado, Hawaii, Idaho, Montana, Nevada, New Mexico, Oregon, Utah, Washington, and Wyoming.³

Telework prevalence and intensity are compared across these regions using data aggregated over the periods 1988–2008, 2010–2018, and 2020–2022.

Regional patterns in telework participation have remained largely consistent across decades, with some modest changes. From 1988 to 2008, the Midwest region held the largest share of full-time teleworkers (33.3 percent) and was closely followed by the South region (27.0 percent).⁴

From 2020 to 2022, the South region led all regions with the highest share of full-time teleworkers at 34.0 percent. Its share in the 2-to-8-hour telework group grew from an average of 32.8 in the 1988 to 2008 period to 38.9 percent from 2020 to 2022. (See chart 8.) These shifts may reflect changes in industry composition, broadband access, or pandemic-era policy responses across regions.

Conclusion

Using more than four decades of longitudinal data from the NLSY79, this article documents how telework evolved from a marginal work arrangement into a persistent feature of employment for a single cohort of workers. While the COVID-19 pandemic triggered a sharp increase in remote work, the data show that telework participation remained elevated after 2020, with growing prevalence of hybrid arrangements. Differences by education, sex, race and ethnicity, and region highlight that access to telework has expanded over time but remains unevenly distributed.