Skip to content
FourthSpace

The second place went quiet: remote work and daily company

How much of American work moved home, and how much did it stay there? What does a remote workday do to the time people spend around others? And where does Sacramento stand, given its large state workforce and the July 2026 return-to-office order?

Analysis by Andrew Ly (opens in a new tab) · Updated

Key finding

57 vs 423

Minutes with anyone outside the household on home vs workplace workdays

Remote workdays are nearly empty of other people.

Hours on a typical workday, 2021 to 2025, on days with 4 or more hours of paid work

  • Home workday
  • Workplace workday

atus_workday_company_by_type.csv

Findings

  1. Remote workdays are nearly empty of other people.

    On home workdays in 2021-2025, people spent about 1.0 hour with anyone outside their household. On workplace days it was 7.1 hours. Time alone was 10.7 hours on home days and 5.8 on workplace days. Source: ATUS 2021-2025, 2,505 home workdays and 6,912 workplace days.

  2. Remote workers do not make it up with friends.

    Time with friends was 12 minutes on home workdays and 11 on workplace days.

  3. The gap holds with controls.

    Holding year, age and work minutes constant, home workdays carry 348 fewer minutes with people outside the household (95% CI 341 to 356) and 273 more minutes alone (263 to 283).

  4. Home work settled at about four times its old level.

    In ATUS, 5% to 7% of workdays were home workdays every year from 2010 to 2019. The share was 24% in 2021 and 20% in 2025. SWAA puts the share of paid full workdays at home at 27% in October 2026, against 7% before the pandemic and 62% in May 2020.

  5. Sacramento County leans more remote than the state or the nation.

    14.7% of its workers usually worked from home in 2025, against 13.4% in California and 12.9% nationally. That is 112,103 people, up from 52,374 (7.2%) in 2019. The Sacramento metro area sits at 16.3%.

  6. The state's return-to-office order came after this data.

    Governor Newsom's order requiring most state employees to work in person four days a week, up from two, took effect July 1, 2026 (CapRadio). Sacramento County holds just under 90,000 state workers (KPBS, February 2026). The Downtown Sacramento Partnership reported downtown foot traffic up 12% from July 1 to 11 compared with the same days in 2025 (The Abridged). The 2026 ACS, due in September 2027, will be the first to show the effect.

Options

  1. Option A. Make remote and hybrid workers the first user segment

    My pick

    They lost the most daily contact, and Sacramento County has about 112,000 of them (2025).

  2. Option B. Run weekday daytime meetups

    Midday coffee and lunch meetups at suburban cafes and libraries, where remote workers live.

  3. Option C. Ride the return-to-office wave downtown

    After-work events for the state workers now downtown four days a week.

  4. Option D. Sell cafes on work-from-here hours

    Laptop-friendly tables and a weekly cowork morning, listed in the app, give venues weekday traffic.

  5. Option E. Wait for the 2026 ACS

    See how much the order cut remote work before committing to a segment.

Default pick: A, launched through B. The ATUS gap is large and shows up in every year since 2021. The people it describes are easy to find. Weekday daytime meetups solve the specific problem the data shows: long, empty home workdays. C is worth a test once the order's effect is clearer.

Method

  1. 01_atus_workday_company.py finds every diary day with 4 or more hours of paid work (activity codes 0501xx). A home workday has 90% or more of work minutes at home. A workplace workday has 90% or more at the workplace. From the who file it sums the minutes spent with anyone outside the household (person codes 40 and up), with coworkers, bosses, supervisees or customers (codes 55, 59-62), alone (codes 18-19) and with friends (54). Since 2010 ATUS records who is present during work too. A weighted regression then compares home and workplace days while holding year, age and minutes worked constant.
  2. 02_acs_wfh.py reads the 2021-2025 table-based files and the 2019 sequence file (B08301 sits in sequence 0036 at position 124), then pulls the SWAA monthly series.
  3. 03_charts.py draws the charts.

Tools

Python 3.13, pandas, NumPy, statsmodels (weighted least squares with HC1 standard errors), openpyxl, matplotlib, Plotly. ACS data comes from the Census summary-file server, which needs no API key.

Caveats

  1. ATUS counts people physically present. Video calls with coworkers do not count as time with others, which overstates the social gap for some remote jobs.
  2. Home and workplace workers differ in job, income and family situation. The regression holds age, year and work time constant. It does not control for occupation, so the gap is descriptive.
  3. ATUS samples are small for single years since 2020 (1,547 workdays in 2025). Pooling 2021-2025 gives stable comparisons.
  4. ACS asks about the usual way to work last week, so hybrid workers who commuted most days count as commuters. It undercounts partial remote work.
  5. SWAA and ATUS define a home workday differently, so their levels differ.
  6. The return-to-office figures come from news reports and the Downtown Sacramento Partnership. They are context. This project did not analyze them.

Downloads

These files come from US federal agencies and are in the public domain. Every other source has its own terms, so it is linked at the publisher under Sources.

Sources

  1. BLS American Time Use Survey: activity, who and summary files (work location TEWHERE, who was present)

    Years
    2010-2025
    Geography
    U.S. employed people

    https://www.bls.gov/tus/data.htm (opens in a new tab)

    Public domain

  2. American Community Survey 1-year, table B08301 (worked from home), Census summary files

    Years
    2019, 2021-2025
    Geography
    U.S., California, Sacramento County, Sacramento city, Sacramento metro

    https://www2.census.gov/programs-surveys/acs/summary_file/ (opens in a new tab)

    Public domain

  3. Survey of Working Arrangements and Attitudes (Barrero, Bloom and Davis), WFH time series, CC-BY 4.0

    Years
    May 2020-Oct 2026
    Geography
    U.S.

    https://wfhresearch.com/data/ (opens in a new tab)

    Publisher's terms

  4. News reports on California's return-to-office order (cited context)

    Years
    2026
    Geography
    Sacramento

    CapRadio https://www.capradio.org/articles/2026/07/01/californias-state-workers-protest-as-return-to-office-mandate-takes-effect/ (opens in a new tab) · KPBS https://www.kpbs.org/news/quality-of-life/2026/02/24/union-backed-bill-would-require-telework-options-for-california-state-workers (opens in a new tab) · The Abridged https://www.abridged.org/?p=19424 (opens in a new tab)

    Publisher's terms

Kastle Systems badge data and the Downtown Recovery project were considered and left out. Kastle covers 10 large metros, and Sacramento is not one of them. Downtown Recovery uses paid cell phone data, which the brief marks optional.

Interactive charts

  • Sacramento County still has twice as many home workers as in 2019

  • On a remote workday, people spend about an hour with anyone outside their household

  • Between a fifth and a quarter of U.S. workdays now happen at home, up from about 6% in the 2010s

Keep reading

The friendship recession: 23 years of time use

How has the time Americans spend with friends, alone and socializing changed since 2003? Did it fall for everyone or only for some ages? And what lines up with it in time?

Next study
Time with friends has fallen in every age group since 2003

Be first through the door.

We're opening in Sacramento first. Join the list and we'll email you when we open.

Fields marked * are required.

Places you'd go (optional)
How we use your details

Study07/07Join the waitlist