Invested in the WFH argument? Home in on the evidence

The writer is a science commentator

It was striking to read this week that many Gen Z workers regard video calls as “face-to-face” meetings. The older segment of this cohort — born in the late 1990s, raised on social media and coming of age in the Covid-19 pandemic — often have a distinctive take on many aspects of office life, such as corporate loyalty (meh) and work-life balance (100 per cent).

They generally expect at least some remote working, an aspiration at odds with the many companies and executives who prefer employees in the office. Now, a new study has found that hybrid working, with three days in the office and two at home, does not lower performance or hinder promotion prospects — and may save money by improving retention.

The research brings some evidence to bear on the contentious issue of working from home, which has become something of an ideological battleground. Significantly, the study shows how a scientific approach can help to evaluate the impact of social and economic interventions. That should happen more widely: robust evidence gathered from well-designed trials can test assumptions, expose poor policies and promote effective ones, whether in corporate management, health, crime or education.

Nicholas Bloom and Ruobing Han, from Stanford University and the Chinese University of Hong Kong in Shenzhen respectively, led a six-month trial in 2021 with the Chinese technology company Trip.com to test how hybrid working affected productivity, employee satisfaction and retention. Trip.com, today worth about $33bn, is an online travel agency that owns several booking websites, including Skyscanner. 

About 1,600 employees — typically graduates in their mid-thirties — were randomised into either hybrid working or coming full-time to the office. Participants with odd-numbered birthdays were assigned to the hybrid “treatment” group, with three days in the office and two at home; those with even-numbered birthdays to the “control” group based five days a week in the Shanghai office. 

Bloom and Han reported last week in the journal Nature that, over the subsequent two years, hybrid workers showed no difference in performance grades or promotion prospects compared to office-bound colleagues; the company’s computer engineers, for example, did not differ in their coding output across the two groups. 

Hybrid workers, however, reported higher job satisfaction and were less likely to quit, especially if they were non-managers, female, or had long commutes. The authors conclude that “a hybrid schedule with two days a week working from home does not damage performance”, adding the result would probably extend to other organisations. 

The caveat is that this trial looked at only one company — one that wanted to road-test hybrid working after seeing American rivals adopt it, possibly willing its success. It is hard to extrapolate across other company cultures. Nonetheless, the study appeals because it was a randomised control trial, often described as a “gold standard” method of judging whether an intervention works. This approach aims to tweak just one variable — such as a drug — while keeping everything else, as far as possible, the same.

While past studies have tended to focus on fully remote working in repetitive jobs like data entry or call handling, this study examined the more usual hybrid pattern, and among highly skilled workers in creative team roles like software engineering, marketing and finance. Older studies of full-time remote working have been linked to worse productivity.

Importantly, the experiment shifted mindset. The perception of hybrid working changed among managers, from negative before the trial to mildly positive afterwards. The travel company has since made the offering permanent on the basis of improved retention; each lost employee, it calculates, costs $20,000 in recruitment and training.

Collecting good-quality evidence on whether interventions are effective is also critical in the public sector. That thinking underpinned the UK government’s What Works Network, introduced in 2013. Thirteen What Works Centres now collect data and run pilot schemes to guide decision-making, such as the move to equip police in London with bodycams.

However, as a WWN report lamented last year, “high-quality evidence on the impact of government policies remains the exception rather than the rule”. Given the imperative to improve public services in straitened times, the next UK government must be serious and systematic about putting evidence at the heart of decision-making.