What can we learn from the great working-from-home experiment?

28th January, 2021

In February 2014, London’s Underground was partially shut down by a strike that forced many commuters to find new ways to get to work. The disruption lasted just 48 hours, but when three economists (Shaun Larcom, Ferdinand Rauch and Tim Willems) studied data from the city’s transport network, they discovered something interesting.

Tens of thousands of commuters did not return to their original routes, presumably having found faster or more pleasant ways to reach their destination. A few hours of disruption were enough to make them realise that they had been doing commuting wrong their entire adult lives.

I mention this because we are at a turning point in the pandemic. Many people, myself included, have largely been working from home. For months it has been hard to shake the feeling that this will last for ever. Now we are contemplating a vaccine-fuelled return to normality — maybe not today, maybe not tomorrow, but soon.

What the 2014 Tube strike teaches us is that temporary disruptions can have permanent effects. Sometimes there are scars that do not heal; sometimes a crisis teaches us lessons we can use when it has passed. So what have we learnt from the remote-working experiment? And to what extent will it continue after the virus has retreated?

One obvious point is that, like a strike-delayed commuter who invests in a new bike, both workers and employers have sunk considerable time and effort into acquiring the equipment and skills necessary to support the change. Such investments will make working from home cheaper and more tempting in future.

I suspect, however, that the crucial step is not investment but information. We have learnt that working from home is more productive than we had guessed.

Emma Harrington and Natalia Emanuel, two young economists at Harvard University, found that before the pandemic remote workers at one large company had been less productive than office-based workers. Yet when everyone switched to remote work, overall productivity increased. The explanation of the apparent contradiction is that working from home is intrinsically more productive, but this truth was obscured by the fact that fewer productive workers were attracted to working from home. Now that employers have discovered the apparent productivity penalty is illusory, perhaps remote work will be far more popular in future.

Similarly, a famous study of remote work by Nicholas Bloom and colleagues analysed a randomised experiment at Ctrip, a large Chinese travel agency, in which some staff were assigned to work from home. The expectation was that productivity would fall but the costs of providing office space would fall too. Instead, Bloom and colleagues found that workers became sharply more productive at home.

All this suggests that the pandemic, like the Tube strike, will be the jolt that pushes us into doing the remote working we should have been doing all along. But I am not so sure.

One point that is easily overlooked is that, in both these studies, the workers in question were moving from taking calls in a call centre to taking calls at home. In Bloom’s research, the home workers and office workers used the same equipment and order-flow software, did the same tasks and were rewarded with the same bonuses.

This should also be a warning not to draw conclusions that are too broad. In a well-run call centre, the protocols for assigning, monitoring and closing out tasks are well established. They do not require a chain of group emails to figure out what is happening or to schedule a Zoom call. The same is not true of much knowledge work.

As Cal Newport, author of the forthcoming A World Without Email, pointed out in the New Yorker last May: “The knowledge work pursued in many modern offices — thinking, investigating, synthesizing, writing, planning, organizing, and so on — tends to be fuzzy and disorganized compared to the structured processes of, say, industrial manufacturing.”

Newport’s view is that this is a solvable problem, but most offices simply haven’t got their act together. Sitting in an open-plan office allows a team to muddle through without realising quite how much time they waste on busywork and co-ordination.

A few knowledge jobs, such as IT support, are properly systematised to allow focused work without endless ad hoc emails. Newport believes that others will follow once we all wise up. Or we may find that certain kinds of knowledge work are too unruly to systematise. Improvisation will remain the only mode of working — and, for that, face-to-face contact seems essential.

A recent survey by Bloom, with Jose Maria Barrero and Steven Davis, estimates that remote work in the US will become more than four times as common after the pandemic, increasing from 5 per cent to 22 per cent of work days. That would be a big swing back towards normality — the researchers estimate that in May 2020 more than 60 per cent of paid workers in the US were operating from home. But it would still be a seismic fall in demand for commuting and city-centre office space.

I hope that the crisis teaches us how to do productive and fulfilling work from home. But it seems that most of us, most of the time, are destined to return to the office in due course. If so, I hope the crisis teaches us how to do productive and fulfilling work wherever we may be.


Written for and first published in the Financial Times on 31 December 2020.

My new book, “The Data Detective” is published in the US/Canada on 2nd February. “Nobody makes the statistics of everyday life more fascinating and enjoyable than Tim Harford.” – Bill Bryson

“This entertaining, engrossing book about the power of numbers, logic and genuine curiosity”- Maria Konnikova

I’ve set up a storefront on Bookshop in the United States and the United Kingdom – have a look and see all my recommendations; Bookshop is set up to support local independent retailers.

Pin It on Pinterest

Share This