When remote work works and when it doesn't

NPR.

And the results are in for remote work.

Yes, they are. A lot of research has come in recently, all trying to answer once and for all whether remote work is better or worse than being in an office.

Yeah, I think back to the early days of the pandemic when these discussions were red hot. Our very own Stacy Vanek Smith and Greg Rosalski had this debate over it a couple of years ago.

And we actually have pretty different opinions on how things are going to shake out.

We do.

Stacy, I think the office is going to win the day.

I'm still appalled by that standpoint.

Sorry, Greg.

I am firmly with Stacy on this one.

You're much more of a work-from-home person. I'm much more of an office person. But today we put our personal prejudices aside and Just look at what the evidence says. You know, 3 years after this huge natural experiment caused by the pandemic, we want to see the results. This is The Indicator from Planet Money. I'm Darian Woods.

And I'm Adrian Ma. Today on the show, a roundup on remote work research and how the evidence is changing.

José María Barrero is one of the top experts in working from home.

I'm an assistant professor of finance at ITAM in Mexico City, which is one of the main universities here.

Some of the research that José and his colleagues conducted early in the pandemic seemed to suggest people who work fully remote could be just as productive or even more productive than people who worked in the office.

But 3 years later, with a ton more evidence coming in from around the world, José and his co-authors just released a working paper reviewing the evidence that upturns that conclusion.

José says yes, working hybrid, coming in a couple days a week, is about as productive as being fully in the office.

The big difference is kind of job satisfaction and attrition rates.

Maybe no surprise, José and his colleagues find a lot of people really like working from home, and remote workers are less likely to quit their jobs than people who only work in an office.

But working fully remotely, 5 days a week, The productivity there is a whole different story. Now, measuring productivity is tricky business, especially because the kinds of work that can be done remotely tends to be knowledge work, things like marketing or legal work. So the best evidence José looked at studies jobs where a clear output is measurable, like police dispatchers in Manchester in the UK.

Basically, the communications seem to be more efficient when these 2 people, the person who receives the call From the citizen and the person who sends the officer to attend to the incident are in the same location.

Another job that's easy to measure: call center workers. Economists Emma Harrington and Natalia Emanuel crunched the numbers there.

We spoke to Emma Harrington, who is an assistant professor at University of Virginia.

We found that pre-pandemic, the remote call center workers at this company were about 12% less productive than the on-site workers.

Okay, so we've got 12% less productive. That does not seem good.

No, 12% less productive seems pretty bad, actually. And the key here is that we're talking about pre-pandemic. Initially, Emma did not know if that 12% fewer calls answered was because of something about the nature of fully remote work, or it was a kind of selection effect happening. In other words, could it be that less productive workers were the ones choosing to work at home. The pandemic brought some clarity to this question.

Once everyone was working at home because of COVID that productivity gap shrunk, but much of it remained. And so some of the initial difference was due to the fact that people are a little less productive when they're working from home, but most of it was due to Different types of people choosing to be remote than in the office.

Emma and Natalia could now distill a closer estimate of the effects of working remotely on productivity. Their study suggested working remotely meant 4% fewer calls answered each hour.

And when you tally up this study and a bunch of other studies on fully remote work in other industries, as José has done, the evidence for worker productivity could be even worse.

It's anywhere up to a negative 10% effect on productivity.

That's significant.

I mean, that is significant, but what's important to keep in mind is that if a firm goes fully remote, and remember, this negative 10% is for fully remote, the firm can basically make that up in other dimensions. And so in particular, in saving up on floor space costs.

Right. So for some companies, savings on office space could make up more than that 10% drop in productivity. So for those companies, it would still be worthwhile to have fully remote workers.

And to understand how to make remote work work, it's helpful to know exactly why it can mean less productivity. So the first reason is communication troubles. In various studies, like the one observing police dispatchers, face-to-face meant workers could complete tasks faster.

You could write an email asking for those instructions and for that clarification, and it might delay your action a couple of minutes. That delay seems to be kind of what is dropping the productivity, kind of these frictions to communication.

Another reason, screens are just harder to focus on.

I feel this in my guts. As someone who spends all day looking at computers, my eyes just wanna fall outta my head by the end of it.

Yeah, and it's harder to be productive when your eyes are falling out.

That's true. Yes, and that brings us to a worrying third cause for productivity loss for fully remote workers, Mentoring and training. Emma Harrington and her colleagues worked on a separate study of the same company where the call center was, but this time they looked at software engineers, coders, and, and looked at how much these coders helped each other when they were working in the same space versus when they were working apart.

They write all their feedback to their coworkers about their code, and that gives us this paper trail of mentorship that you don't typically see in other contexts.

And this one was pretty interesting because it wasn't actually about working from home. It was 2 different buildings, at least initially. In one building, the software engineers were all together, and in another building, the software engineers were scattered or distributed. They're about a 10-minute walk apart from each other.

Being in the same building as all your teammates has a big impact on how much feedback you receive from your teammates. You receive about 20% less feedback if you're on a distributed team than if you're all co-located with your teammates.

So even when these coders were not working remotely from home, just when they were physically removed from their colleagues, that meant 20% less feedback. And after the pandemic started, the number of lines of feedback on the code dropped even further, about 50% less than when the team was all located together in the same room. Now, other things could be driving this 50% decline. The start of the pandemic was obviously hugely disruptive. But this does indicate massive drop-offs in the amount of training and mentoring that people get when everyone's working remotely.

We see that the effects are much more pronounced among those young workers, among those workers who are new to the company, who are the ones who have the most to learn from their more senior coworkers.

And that got us thinking, maybe some of that early productivity boost that some people found at the start of the pandemic was because when you don't mentor or train or help your colleagues very much, that means that experienced workers can do more of their own work.

It definitely accords with some of our conversations with people at the firm, some of whom said, I found remote work really productive for me. I could just go home, focus on my own tasks, just get a lot of work done. But seeing your findings makes me wonder, did that come at the expense of mentoring my junior colleagues?

Yeah, and just extrapolate these findings out a few years across the world, and you can see that there's a real long-term cost here. José agrees.

I think a lot of why companies, especially in knowledge jobs that require interaction, are going for hybrid is because kind of they need the in-person interaction to develop these qualities in workers and basically this human capital and, and kind of sense of belonging to the firm.

The way this evidence is stacking up, it's not a good look for remote work.

I mean, I want to say this is not a takedown of fully remote work. Like, there are a ton of reasons why a company might want to do it. We've talked about saving on office costs, but also you can hire from around the world. You can get talent that otherwise wouldn't be able to work for you. So I don't know, I think you're safe, Adrian, and remote work is here to stay.

Whew!

Okay. This episode was produced by Brittany Cronin, with engineering by Josh Newell and James Willetts. It was fact-checked by Ciara Juárez. Caitlin Cannon edits the show, and The Indicator is a production of NPR.