Lightnews — Scholar-powered news

Daniel Zhao @danielzhao.bsky.social · 1h

No way the "More than 30" group is only 3%

YouGov @yougov.co.uk · 1h

How many browsing tabs do you typically have open?*

1: 6%
2-5: 54%
6-10: 14%
11-20: 8%
21-30: 2%
More than 30: 3%

*across all windows, on desktop/laptop

yougov.co.uk/topics/techn...

1 1 7

Daniel Zhao @danielzhao.bsky.social · 2h

Unclear, vast majority of establishment survey responses come from online or via data file which don't require an employee to manually collect, but idk what setup needs to happen for each new month:
www.bls.gov/web/empsit/c...

Data sources

www.bls.gov

2 1 2

Daniel Zhao @danielzhao.bsky.social · 2h

By contrast, furloughed federal workers were classified as unemployed on temporary layoff in the household survey, which boosted the unemployment rate during the 2019 shutdown.

Screenshot:
"3. Household survey: How did the partial federal government shutdown impact January estimates?
Unlike the establishment survey, some of the estimates from the household survey reflect an impact of the partial federal
government shutdown. Both the unemployment rate, at 4.0 percent, and the number of unemployed people, at 6.5 million,
edged up in January (on a seasonally adjusted basis). The impact of the shutdown contributed to the uptick in the measures.
In the household survey, workers who indicate that they were not working during the entire reference week due to a
shutdown-related furlough and expect to be recalled to their jobs should be classified as unemployed on temporary layoff,
whether or not they are paid for the time they were off work. (The household survey reference week was January 6–12.)
In January, many furloughed federal employees were classified as unemployed on temporary layoff, contributing to a rise in
the overall number of unemployed people on temporary layoff. (See details in item 4 below.)
However, as was the case with the 2013 partial federal government shutdown, some federal workers who were not at work
during the entire reference week were not classified as unemployed on temporary layoff in January. Rather, they were
classified as employed but absent from work. BLS review of the underlying data indicates that most of these workers were
misclassified; they should have been classified as unemployed on temporary layoff. (See details in item 5 below.) "

2 3

Daniel Zhao @danielzhao.bsky.social · 2h

Reading through the BLS's 2019 FAQ on govt shutdown and in the establishment survey, furloughed federal employees were still counted as employed because they were expected to receive back pay for the period.

www.bls.gov/bls/shutdown...

Screenshot of text
"2. Establishment survey: How did the partial federal government shutdown impact January estimates?
In the establishment survey, there were no discernible impacts of the partial federal government shutdown on the January
estimates of employment, hours, and earnings.
There was no impact on the establishment survey estimate of federal employment, which was essentially unchanged in
January (+1,000). Federal employees on furlough during the shutdown were considered employed because they worked or
received pay (or will receive pay) for the pay period that included the 12th of the month."

3 2 3

Daniel Zhao @danielzhao.bsky.social · 3h

Some good details on the mechanics of when pay arrives for government workers from @gregorykorte.com

www.bloomberg.com/news/article...

3 3

Daniel Zhao @danielzhao.bsky.social · 20h

I think making it enticing for people to get emotionally invested in the Jets is NOT a social good

source: @matt-levine.bsky.social, www.bloomberg.com/opinion/news...

2

Daniel Zhao @danielzhao.bsky.social · 1d

Bank of America Institute released internal data on the job market which showed further slowing in Sep.

Their continuing UI claims estimate is up +10% YoY vs. +5% YoY in recent govt data. And jobs growth is at about +0.5% YoY vs. +1% in Aug BLS data.

institute.bankofamerica.com/content/dam/...

3 8

Reposted by Daniel Zhao

Amanda Mull @amandamull.bsky.social · 1d

Maybe the weirdest touchdown I’ve ever seen

7 5 140

Reposted by Daniel Zhao

Conor Sen @conorsen.bsky.social · 4d

“Overall, the administration expects the federal workforce to end the year with hundreds of thousands fewer employees, attributed to a hiring freeze, layoffs and voluntary departures.” www.wsj.com/economy/jobs...

Federal-Worker Buyouts Are Kicking In, Darkening the U.S. Employment Picture

About 100,000 federal workers came off the payroll last week, adding strain to an already weakening labor market.

www.wsj.com

2 10 38

Daniel Zhao @danielzhao.bsky.social · 5d

Again, alternative data is not a replacement for BLS data. Even in the best of times, it's a useful complement, but still requires BLS data to benchmark and set the context.

Write-up on the above Glassdoor data: www.glassdoor.com/blog/glassdo...

No jobs report: What does Glassdoor data say? - Glassdoor US

www.glassdoor.com

7

Daniel Zhao @danielzhao.bsky.social · 5d

[4] In interview reviews on Glassdoor submitted in Sep, the % of candidates that received a job offer but declined it fell to 17%, down from 20% in Aug and down from 26% in Sep 2024.

As hiring declines, workers have less leverage when interviewing for a new job and may be more inclined to settle.

1

Daniel Zhao @danielzhao.bsky.social · 5d

[3] The % of Glassdoor reviews mentioning layoffs fell 2.3% month-over-month from Aug to Sep. Feels like layoffs have come out of the headlines a little vs earlier in the year + employee sentiment has followed.

However, the % of reviews that mention layoffs is still up 26% year-over-year.

1

Daniel Zhao @danielzhao.bsky.social · 5d

[2] Salaries* on Glassdoor declined slightly in Sep: Salaries averaged $71,831/yr, down 0.4% from $72,128 in Aug.
Salaries grew 4.9% year-over-year in Sep, slower than +5.4% in Aug and the slowest annual pace of pay growth since Apr 2025

*Self-reported, reweighted to CPS industry-hourly mix

1

Daniel Zhao @danielzhao.bsky.social · 5d

We're not getting a jobs report today, so what does Glassdoor data say about the Sep job market? It's a mixed bag (which highlights how important it is even for alternative data to have BLS data set the context):

[1] The Glassdoor Employee Confidence Index improved in Sep rising to 47.7% from 45.6%

1 3 6

Reposted by Daniel Zhao

John Gramlich @johngramlich.bsky.social · 6d

% of US men under 30 who say legal sports betting is a *bad thing* for society

22% in 2022
47% in 2025

No other demographic group has seen a bigger increase.

Americans increasingly see legal sports betting as a bad thing for society and sports

Today, 43% of U.S. adults say the fact that sports betting is now legal in much of the country is a bad thing for society, up from 34% in 2022.

www.pewresearch.org

180 820 3.7K

Reposted by Daniel Zhao

Erika McEntarfer @erikamcentarfer.bsky.social · 6d

Big picture, we do not know the exact answer to 'how good/bad is the ADP data if BLS data disappears?' but the private data providers themselves are pretty clear in every conference/panel that the quality of their own data will erode if they cannot benchmark to BLS estimates.

4 22