Check Out Nick Bloom's LinkedIn Stats (Last 30 Days)

What do US businesses forecast on working from home? We surveyed executives at over 1,000 US businesses from Feb 10-21, 2025. Summary: 1) If a recession hits - defined as unemployment doubling - days worked from home would decline from 21.2% to 20.3% of days. A minor fall. 2) Currently, only 12% of firms plan to change their Return to Office policy. On net it would reduce WFH from 21.2% to 20.8% of days. A tiny fall. 3) Over the last six months almost as many firms increased WFH as reduced it. You only hear about the ones reducing it because they make better headlines. Finally, important *methodology* points. A) This is a panel of firms stratified across US regions, industries and sizes. B) The survey was run by the Atlanta Fed and Stanford, with no profit motive to under/over sample any types of firms/industries/clients. C) We did not use recall - the 6-month change result was based on comparing surveys 6-months apart using the same question and stratified sampling.

New paper showing a 12% productivity increase from hybrid WFH in the public sector from quieter home conditions. Alessandra Fenizia and Tom Kirchmaier study agents processing crime cases, who are randomized to home or office, finding: 1) A 12% rise in cases processed at home with no change in quality. Alongside the reductions in office space this reduces costs by about 20%. 2) A range in how individual employees respond. Some improve productivity by 25% while others are less efficient WFH. So employee choice can further increase WFH benefits. 3) Fully-remote offers no additional change vs hybrid. These employees were WFH on average 3 days a week and in the office 2 days a week, which seemed to be a sweet spot for their productivity. This highlights how Return to Office mandates for Government workers is is likely to increase costs by lowering productivity and raising real-estate costs. As a taxpayer I urge hybrid WFH for public sector workers.

"There are lies, damned lies and statistics" comes to mind for this McKinsey & Company piece. It claims WFH halved since 2023. Big data from U.S. Census Bureau, SWAA, Kastle Systems and Placer.ai all show it's almost flat. So why the difference 1) The McKinsey survey has recall bias. It was run once, using recall to generate historic data, rather than collecting data each month. When folks are reporting their "WFH last year" many are likely recalling 2020 to 2022 when WFH was a lot higher. 2) The survey also has sample bias. They sampled about 75% in office employees. It is like running a political poll but sampling 75% Republican voters - you can guess who would win. 3) Why did McKinsey not spot this, as monthly data on WFH is publicly available? Likely they did spot this and knew it was wrong, but wanted to publish this anyway to please clients who want this narrative in the media. So if anyone is asked "I thought WFH was falling. Didn't McKinsey put out a survey showing this" you might say. "No, their survey data is wrong, with recall and sample bias. Census, Kastle and Placer data show WFH has been stable since 2023" McKinsey & Company is an amazing company. I worked there in the early 2000s. I am proud of it and have worked with many great McKinsey folk since. But I am not proud of this Quarterly piece, which should have been fact checked. To see correct WFH data see links below.

Could 5-day RTOs lead to industrial action? We asked 12,000 employees in our SWAA survey this month and about one third reported they would strike over WFH. That is less than for pay, benefits or hours, but it is still large. Recruiters now say WFH is now one of the "big three perks". These are pension, healthcare and WFH, which in addition to pay, are key drivers of job attractiveness.

Great piece by Pilita Clark in the FT about Work From Home "dark matter". From the media you think WFH is ending. But in big data on surveys, swipe cards or cell-phone mobility WFH has been stable from early 2023. Three possible reasons: 1) Coverage bias - the old adage "If it bleeds it reads". Bad news sells, so perhaps only stories about the end of WFH get reported. 2) Cohort effects - young firms are more WFH friendly and grow faster, offsetting older firms in the headlines cutting back. 3) Lack of enforcement - the headlines are about RTO mandates, but in practice these get enforced for a month or so and then quietly ignored. Whatever the headlines I trust the data. WFH in the US and globally stabilized in spring 2023. Article and data in comment below.

Today there is a lot of discussion if Trump's RTO announcement for Federal workers will end WFH for everyone. The short answer is *no*. Here is why 1) This announcement was known since last Fall. But despite the advanced warning no-one has followed - WFH has been flat since early 2023 (data below - source in Comment). 2) Hybrid WFH is very profitable for firms - it has no net effect on productivity but cuts costly turnover (source Nature paper). This is why 80% of Fortune 500 firms are hybrid for managers and professionals (source Flex Index). 3) It is unclear if this will actually be enforced and persist. The RTO will lead to high quit rates, particularly amongst hot skill areas like IT. This could hit Federal websites, systems and payments. Federal IT was never that great to start with, so I can see lot of risks. Say veterans payroll or the IRS taxes processing fails - some disaster like that could lead to a quiet reversal. So, yes, Federal Workers are being forced back to the office. But I do not think many other companies will follow, and I'm not sure if the Federal policy itself will even be fully enforced or persist.