464. Will Work-from-Home Work Forever?

I'll give you a little backstory. Last March, my co-authors and I were asked to write a paper about COVID. Last March. So early on. Early on. This was when we were in full-scale panic as a country. And somebody called me up and said, I'm putting together a special session at this conference. Do you have a paper on COVID that you'd like to present?

You know, nudge noise. Would you like to write one in a week, in other words? Exactly. I was like, sure, of course we have a paper on COVID. We very, very quickly realized, look, COVID's interesting, but that will go away eventually. The question is, what's permanent from COVID?

And that led us down this path of working from home might be permanent. And then it was like, well, how do we measure the essential ingredients of economic activity related to working at home? And that's what set us down this track. Morris Davis is an economist in the business school at Rutgers University. My specialty is real estate, but I also work on finance and economics. The track that Davis and his co-authors went down is the same track many of us have gone down since the pandemic began.

Today I'm issuing a Stay Home Stay Safe Executive Order for all Michiganders. We are confident that the people of the state of California will abide by it. This is the most drastic action we can take.

Now the freeze is starting to thaw. Many employees are returning to offices and other workplaces, but not all, and for some, not ever. Today on Freakadamics Radio, what is to be learned from this sudden, mandatory, global work-from-home experiment? We'll talk about productivity. Working at home is always less productive than working at the office.

always. We'll ask what this means for real estate. To be a great 21st century city, we do need high performing, high quality office space. And some surprising changes that may go even deeper. So this was like reverse brain drain happening. It's coming up right after this.

This is Freakonomics Radio, the podcast that explores the hidden side of everything. Here's your host, Steven Dubner. Let's take one profession that you're intimately familiar with, which is college professor. Have you been teaching online during the pandemic?

I have and I hate it. Because why? The students may not believe this, but when I teach, I try to figure out based on body language and the way that they look at me, what's making sense to them and what's not. I just cannot figure out how to do that online. So teaching remotely has been, for me, a terrible experience.

That, again, is more a stavis from Rutgers, so the teaching part of his job has suffered, but teaching isn't the only thing an economics professor does. We used to try to have remote seminars, and it was kind of a disaster. People would walk down the stairs or the dog would bark. Everyone now has learned how to have these remote seminars in my field. Do you think you'll ever return to those in-person seminars, at least anywhere near the same degree? I don't know.

What I think about a lot now is, if I'm invited, I could go to seminars anywhere around the world. And that has made me more productive. This is an excellent example of why working at home and working in the office are complementary for the same occupation. How does this one economics professor's work-from-home experience compare to yours? My guess is there is at least one major overlap.

Some things were better, and some things were worse. Everyone's situation is, of course, different if you happen to live in a comfortable and roomy home, maybe with people you even enjoy. Well, that's one kind of situation. If your home felt cramped even before the pandemic, and if you maybe have a couple young children who suddenly weren't going to school, that is a different story.

Perhaps the most commonly cited benefit of working from home, no commute, although some people have been told do miss commuting. What are the biggest downsides of working from home? A lot of employees really miss the camaraderie of work, and many employers suspect that working from home is a major drag on productivity, shirking from home, they call it.

Here's what the labor economist Nicholas Bloom told us in a pre-pandemic episode discussing some work-from-home research he'd done. The three great enemies are working from home as the fridge, the bed and the television. Some people can handle that and others can't. Bloom had analyzed a work-from-home experiment at a Chinese company called Sea Trip.

is very much like Expedia in the US. The sea trip jobs that moved home were call center jobs. What did Bloom find? So we found working from home raises productivity by 13%, which is massive. So that's what one economist measured in productivity within one type of work at one firm in China, again, well before the pandemic.

Bloom did a recent study, along with Jose Maria Barrero and Stephen Davis, to see what people have been doing with all the time they aren't spending on their work commutes during the pandemic. Here are the headline numbers. Before the pandemic, Americans were commuting on average 54 minutes a day.

Bloom and his co-authors found that people spent about 35% of that saved commuting time on their primary jobs and about 60% on all non-leisure activity, including housework and childcare. So that is an interesting snapshot.

But what about the bigger picture? What have been the larger effects of this year-long work-from-home experiment? Or, as Morris Davis put it earlier, how do we measure the essential ingredients of economic activity related to working at home? So that's what Davis set out to do, along with his colleagues, Andrew Ghent and Jesse Gregory. How do you even begin to answer such a broad question? Well, okay. Taking a step back.

One reason many respectable social scientists hate economics is because... I love any sentence that begins with that, just so you know. We rely a lot on assumptions about how people behave. They care about and how they make decisions. The thing that bothers people about models of economics is that we don't write about everything. We write about a few things.

When I talk to my mom or my sisters, it's always, well, what about this? Or what about that? I said, well, I just ignore those things. And they go, well, why does this paper have any value if you ignore these things? Yeah, I'm with mom. So what's your answer for that?

I mean, I understand your aversion to what aboutism, which strikes me as been a growing scourge because no one believes anything at all, even if there's ample evidence. But if you're presenting a finding based on primarily a model that omits some real behaviors, why should we pay so much attention to it? I'll give you two reasons. The first is

I think of economics as a science, and I think science builds on itself. So you as the researcher write down the things that you think are important, and then you let the world know what the implications of that are. And that's how we build a body of evidence. The other thing is, while everything might matter, everything might not matter to the same degree. And so our job as economists is to try to pick out the two or three or 10 things that really, really matter. So that's what we try to do in our paper.

So Davis and his colleagues built a complex model using all sorts of data and assumptions, hoping to gauge the productivity trends of this work-from-home experience. They focused on what they call high-skill workers, essentially college graduates, who do the kind of work that can be done at home or in an office. The office, they assumed, was in some sort of central business district, and there was one more big assumption in the model.

So just for full transparency, we don't observe productivity. We infer it. We infer that productivity of working at home was, on average, 50% of what productivity of working in the offices. So that's quite different from the higher productivity for those Chinese call center jobs. When you measure across the entire work spectrum, Davis says, working at home is always less productive than working at the office.

always. So if that would make you think, oh, then you always want to work 100% at the office. But the nature of these goods is that they're complimentary. So on average, that's true. But at the margin, that's not true. That's a little confusing. Let's back up. The central argument here is that the office creates certain benefits that your home cannot beat.

The real benefit to being at the office is face-to-face interaction, which might be painful if it's your boss reprimanding you, but this concept of a knowledge spillover, all of that causes we think productivity to be higher at the office than at home. But we also think working at home is not as unproductive as it used to be because we have all of these tools at our disposal.

So Davis and his colleagues argue that productivity is substantially lower when people work from home, but and this is a big, but work from home productivity has increased since the start of the pandemic. And not by just a little bit. They argue it has increased 46% relative to the productivity of working in the office.

Here's the way to think about it. Whatever the productivity was in 2019, we're better at working from home than we used to be. This was only possible because of technologies that already existed. My co-authors and I, and I think many people are now viewing the technology that has enabled us to work from home, almost as important a revolution as, say, electricity. We think it's going to permanently change the landscape.

of where we work, how we value housing, how we think about our commutes. What would you consider the most important components of that? There's a whole sequence of inventions that have led to us being able to work remotely, the PCs that stopped crashing in the early 90s.

And then you had the Netscape IPO, which led to a whole set of new firms exploring the World Wide Web. Email was introduced right around the same time. Then we had smartphones, as compared to just regular old cell phones, enabling us to have instant communication anywhere. And then finally, really, the things that have mattered right now are high-speed internet, which became cheap enough to become widespread cloud computing, which enabled massive instantaneous sharing of data.

and then video conferencing technology. In other words, we picked a pretty good time to have a pandemic is what you're saying. In the paper, we asked, what would have happened if the pandemic had occurred in 1990? And it was just much harder to work from home in 1990. We're not, in fact, sure how it could have been done. The ability to work from home buffered the impact of the pandemic

on many people's productivity and incomes, as well as sickness too. For some people, the fact that you have access to good coffee or you can take a brisk walk, that makes working at home have other benefits too. Theoretically, if you're more satisfied or happier, that may lead long-term to more productivity that we haven't even seen yet, right? That's right.

I think what's rapidly changed is our ability to use the existing technology. It's never been adopted in mass like this. Expectations can be reset about what can be achieved. Let's say that you're the CEO or the owner of a firm where working from home is a possibility and you have a thousand employees.

And you realize that their work from home productivity is decent, although not as good as when they come in. How do you think about the balance of productivity versus your real estate? You used to have to house during the day of thousand people. What do you do now?

Boy, that is an awesome question. And that's a question that every employer is asking right now. How much space do I need and where should it be located? So here's something we assume in the paper that non-economists will bristle at. But maybe if we just talk it out, you'll get comfortable with it. We assume that workers are paid their marginal product for labor. Now, if workers invest more in a home office, according to our assumptions in the model, that makes them more productive. And their wages go up.

So the chain is firms will pay less for office space, but they'll have to pay workers more because workers are more productive. So firms will still be paying for office space. They'll be paying it to workers, for workers to rent their own. So what happens to the price of commercial office space? What are you predicting? Let me just give a caveat that if the space can be converted, then the impact on the prices differs. But if the space can't be converted,

The price of office space will fall by about 20% in the central business district. Would we see a corresponding rise in exurban or suburban office space? Well, we don't take a stand in that.

I've talked with employers around the state of New Jersey just to ask them what's going on. How are you handling this? We don't know if employers are going to set up small satellite offices so that people have the option of leaving the house but going somewhere close by, which I think matters a lot if you have small kids.

But we do talk about the price of housing. We predict that the price of housing will increase between 11 and 20%, depending on where the housing is located. People expect to work from home more frequently, and they want more space in their house when they work from home.

I would almost think that if central business areas are going to decline as you predict, and therefore people have more flexibility about where they live, that theoretically housing prices wouldn't have to rise so much because there's now more flexibilities, more options, and I can choose to live in other places. This is the great question of our time, which is how far can people move away?

And if they do move away, what productivity levels will they have? There's plenty of relatively inexpensive metropolitan areas in the United States right now. Detroit, Milwaukee, Columbus, Memphis, Oklahoma City. There's lots of inexpensive, nice metro areas. With a few exceptions, people aren't moving there.

At least they weren't. So there were strong economic forces driving people to places that were already expensive. I'm curious how you think the changes may affect the tax bases of those different areas. Do city centers inevitably suffer? Do suburbs boom? If so, what are they going to spend all that money on? Do the schools get better? I'm curious about any knock-on effects you might see. The first knock-on effect is that

to the extent that cities' finances depend on property taxes from commercial properties. They're going to suffer, probably greatly, because the value of office buildings will fall. The rents will fall, and the simple reason is the space just isn't as useful as it used to be. People are not going to commute as much as they were. I think it'll be a real strain on New York City's finances, for example.

No, I'm probably wrong, but I could imagine it going the other way. So hear me out for one minute if you don't mind. A lot of offices pre-pandemic were set up with open floor plans. Executives and designers say that was meant to optimize flow and collaboration, but as we've discussed on this show, that actually doesn't always happen.

It's a lot cheaper to build open office plans, but many people who work there don't like that openness. It's noisy. It's intrusive. And they actually become less collaborative and often less productive. Now, after the pandemic, you could imagine that a lot of people

who will be willing and even eager to come back to an office may not be as comfortable with those open floor plans for pandemic reasons and also just they've gotten used to having a little bit more control over the environment. And if that were the case, maybe firms won't be downsizing. Maybe they'll need more space. That's my crazy theory. It's not crazy at all. It's the nicest thing anybody's ever said to me.

But the one place where I might disagree with you is, will a firm be willing to have a large office for one person that's only occupied two days a week? If you're going to have all these offices that are only half occupied, are you really going to waste all that money? As of April, more than 15% of office space in Manhattan was unrented, the highest level in nearly two decades.

Will that number fall as workers return to offices, and as some companies, including Google and Facebook, continue to expand their New York office footprint? Or will vacancies increase as more firms decide that at least some of their employees will keep working at least some of their hours at home?

Here's one clue. In December, the real estate board of New York, the lobbying group for the industry, called on New York City to allow older, less desirable office towers to be more easily converted to residential use. I think it's a winning idea because, you know, the average age of our office stock is just over 70 years. That's Marianne Gilmartin. To be a great 21st century city, we do need high performing, high quality office space.

Gilmartin is a big real estate developer in New York who recently started her own firm, MAG Partners. Before that, she was a lead developer on projects like the Barclays Center in Brooklyn and the New York Times office tower in Times Square. She mentioned one of the biggest and flashiest new developments in New York, Hudson Yards, a complex of office and residential and commercial properties on the far west side of Manhattan. It opened about a year before the pandemic,

Lately, much of it has set empty, but Gilmartin is bullish on its future.

I do think that Hudson Yards is really an example of high performing space and a flight to quality, because let's be clear, Hudson Yards was never considered convenient. And so this flight to quality means that some of the older office stock in certain locations should be considered for conversions, because if we convert some of the older office buildings, we will take down the supply of office space. And by doing that, we can also improve the fabric

the diversity and the social equity of our central business districts by bringing people into those areas to live. But what if the whole idea of the central business district has begun to permanently erode?

Yes, this idea has a long and profitable history and economists have shown there are huge benefits to density. Indeed, the world has become more and more urban over the past few centuries as a place to work and live and play. How much will the pandemic change that?

for all the advantages that density offers, the pandemic has shown the downsides too. Also, cities are a habit, and people can sometimes take up new habits. So, coming up after the break, we hear from another researcher who has been studying remote work for years. The first thing I would say is that we need to dissect what remote means.

That's coming up in a minute. In the meantime, if you've been enjoying the special episodes of the Freakonomics Radio book club we've been making, you should know that we have just created a separate Freakonomics Radio book club podcast feed. So if you have friends who love books but hate economics, now you have got somewhere to send them. The Freakonomics Radio book club is available on Apple podcasts, Stitcher, Spotify, wherever you get your podcasts, we'll be right back.

As of today, roughly 50% of US workers are still virtual. A year ago, that number was around 70%. What will it be a year from now? No one really knows. Some predictions say that office work will eventually return to pre-pandemic levels,

Other predictions say that office work has been broken forever that too many workers have seen the light. One new research paper based on survey data estimates that 20% of full workdays will now be from home versus just 5% before the pandemic. Here again is the economist Morris Davis. I live in a commuting suburb to Manhattan. And what I hear is there were a set of my neighbors that were commuting five days a week. And they all say, I'm never doing that again.

Never doing five. Never doing five. They might do one or two. Yeah, somewhere between one and three is the number that I hear. Everybody makes predictions about everything all the time. Do you think that the prediction, the stated desire will bear any resemblance to the reality? Yes. Because why? Because people do not like to commute.

Many predictions about the future of work do lean toward this hybrid model, two or three days in the office, two or three at home. Many companies have already moved in this direction. This creates all sorts of coordination and communication and strategic issues to work out. As one software executive recently told Wall Street Journal, there's going to be a bunch of unintended consequences. It's going to be a mess.

But does it have to be a mess? Why can't the technology that made working from home possible during the pandemic make it more probable in the future? To answer that question, we need help. What we need is a researcher who was thinking about these issues long before anyone ever heard of COVID-19.

Hello, I'm Rad Chaudhuri. I'm the Lumri Family Associate Professor at the Harvard Business School. And you teach what exactly? So my research and teaching is focused on the future of work, and especially on the changing geography of work. I hate to say it, but given your pre-existing research interests, it sounds like the pandemic has probably been good for your work, yes?

Yes, so it's been the silver lining, I would say, and it's really been an interesting time to see how the phenomenon of remote work is playing across industries and tasks. Choudary studies what he calls work from anywhere. It's not quite the same as work from home.

and work from anywhere you could be working from an office, you could be working from a co-working space, you could be working from your home. But most importantly, you're working in a city or town or village that you want to live, not where the company has an office.

The pandemic has led many people in many countries to leave the area where their office is located, at least temporarily. Some firms have seen their employees literally scattered around the world, creating what's called a distributed workforce. Chowderay grew up in part in India, which for decades has housed a distributed workforce for many US companies. I asked him if there's much to be learned from that history.

There are some underlying similarities of course, so there was the issue of time zones because distributor teams were distributed across time zones. I think it's just a dose issue. Now, if a lot of employees spread out and start working from locations of their choice,

All these issues would be on steroids. How do you manage communication? How do you manage coordination? How do you make the whole team feel socialized and part of the same culture? And there are solutions to all of these. There are management solutions to all of these problems.

Several years ago, Choudhary came across an interesting program that allowed him to learn a lot about working from anywhere. The program was run by the US Patent and Trademark Office, which is headquartered in Alexandria, Virginia, just outside of Washington, D.C. It didn't start out as a work from anywhere program. The Patent Office allowed their patent examiners to work from home in 2007.

The deal was you work from home four days a week, but then you come to office on Friday or Monday. But then in 2012, in 2012, they said, now you can leave Alexandria, Virginia and live in any part of the continental US. And I found that fascinating as a migration scholar because many of these people had gone back to smaller towns. So this was like reverse brain drain happening.

Reversing brain drain is an issue for a lot of places all over the world, because when smart, talented, highly educated people migrate from rural areas or smaller towns to larger cities, their hometowns suffer. It's potentially great for emerging markets to get talent back. That's what I'm most excited about.

I think India could get a lot of talent back from the West, but it's not only India. I think the Indians small towns could be the winners because there are tier two, tier three cities, which have enough of an infrastructure that you could work remotely. And there's all the benefits of lower cost of living. And I think they could be the real winners instead of Bangalore or Hyderabad or Delhi.

Again, you can see the upsides for employees, but how does work from anywhere work out for the employers? With the US Patent Office program, Choudhury had a nice natural experiment to answer that question. The best news, not all patent examiners were eligible for the program at the same time.

as an economist that creates a random timing for the treatment. Choudhury and his co-authors, Siris Farugi and Barbara Larson, set about to analyze the data. What happened to the productivity levels of the patent examiners who had chosen to leave the DC area? We estimate a 4.4% increase in productivity.

4.4% increase in productivity and how is that measured so the 4.4% is measured based on the average number of actions as they call it which is essentially case files that the examiner examines every month.

So it's very objective. That's great news. Do you know the why, however? Yes. So what we essentially say that the mechanism driving this was effort. How was effort measured in this case? We looked at what kinds of case files were being examined.

And what we found was that they were increasing the number of what the patent office calls first office actions. The first office action is the first time the patent examiner is responding to the lawyer on the other side, the company. And that needs the most effort for the examiner. What we found was that the first office actions went up and the subsequent revisions didn't go up as much. And if the patent examiner wanted to slack, they could have done the reverse.

There's one really important thing to know about this program. The patent office did not adjust the income of the patent examiners based on where they chose to live. And that was clearly a benefit because real income went up. That's because the Washington DC area is expensive to live.

You may have heard that Facebook, for instance, has announced its employees can live wherever they want for the foreseeable future. But if they're no longer living in somewhere like Silicon Valley, they will no longer receive Silicon Valley salaries. Choudary prefers the patent office model. My intuition tells me that that is the right way to structure salary.

Because if you adjust salaries based on location and not task, then the risk is that your right tail of the distribution of talent will jump to your competitor if your competitor gives them the same salary for where they're living. Chowderie's analysis of the patent office program didn't include just quantitative data. He also interviewed patent examiners. And the story that came, Stephen, was one of loyalty.

that I was really helped by this policy because now I could move to Philly and my daughter needs some medical treatment which is only available in Philly. No other organization will let me work in Philly and do the kind of work I'm doing. So I have to give something back.

That is so interesting, and it also implies that this increasingly fractious relationship between firms and employees may be turning a corner, at least for some sectors. We actually frame the work from anywhere policy as a non-pigunity benefit.

It was highly valued by the patent examiners, especially women. So the other group I spoke to were the military spouses and diplomatic spouses. And their story was that we had to change our jobs every few years because our spouse was moving some base. And now we don't have to because now we can work from anywhere. So it was a story of being happy and working harder.

Okay, but what about any kind of objective measure of quality? Was there one? Yes, we looked at two. We looked at these requests for continued examination, which come from the firms and the patent lawyers on their side, and we didn't find any change. So quality on that measure didn't go up or down.

And the other thing we looked at is the most objective measure of quality on pattern examination is how many citations to prior art is being added by the examiner. And we found that didn't change. So quality didn't improve, but it also did not deteriorate.

Can you just tell me in terms of sectors where work from anywhere is possible? Do you think there were particularities about that kind of work, the patent examiner's work, that made it more viable than another kind of work? So my priors have changed. So prior to the pandemic, my thinking was that work from anywhere is more amenable to tasks which can be performed more independently.

such as the pattern examination task or think about call center workers. What I believe now is that there is also a very robust way of engineering collaboration and social interactions in the virtual world, which makes me think that this is probably a much more pervasive phenomenon that will be early imagine this to be.

Why do you believe that now? So a specific experiment I ran last year was around this phenomenon of virtual water coolers, which engineered these random interactions in the workplace between really senior managers and new employees. And we found fascinating effects on performance, on the probability of getting an offer for a full-time job,

Would you say that it was much more likely for let's say a low level or new employee to interact with a senior executive at this virtual water cooler than if they actually were working at the office. So these are interns all participating in this virtual internship last year.

And what they write in the surveys is that these interactions facilitated knowledge sharing, which would normally would not be written down in any manual. It's stuff that the senior manager would tell you only when they talk to you. This actually goes back to research in the 1970s by Tom Allen at MIT. He studied our social interactions in the physical office.

And the amazing insight there was, yes, we do have the serendipitous water cooler conversations or cafeteria conversations, but it's almost always with people in very close proximity of us in the office.

So it exponentially decays with distance in the office. And if there's a floor between the two people, forget about it. But in the virtual world, you can bring anyone together. So presumably you can have much richer and more virtual interactions than real physical ones.

So if I'm a senior manager and I'm thinking the pandemic is hopefully winding down soon and I plan to gather all of my employees back in the office. What's a key lesson about this virtual water cooler that I should try to replicate in the office. So first off I think that's a terrible idea.

Wait, you think it's a terrible idea to have all the workers back or to replicate the virtual water cooler? No, to go back in time to 2019. Even if workers are being brought back to the office in a hybrid remote context, I feel virtual water coolers are a great tool to facilitate discussions that wouldn't normally not happen.

Why is it a terrible idea to go back to 2019, and persuade me that your answer is not influenced by the fact that you are a work from anywhere scholar. I think I am conditioned by that, but I'll make a case for you. The case is that for the individual worker,

It would be sort of taking away the flexibility that workers are craving, especially women, because there's tons of research which has shown that in the past women have borne the brunt of dual career situations.

So if you had a promotion opportunity that might make you move from, I don't know, Columbus, Ohio to New York, and your spouse doesn't want to move there, do you forego that promotion opportunity? So work from anywhere allows companies to hire from anywhere and create a more inclusive workforce.

based on gender, based on disabilities. So my prediction, and of course this is testable, is that companies that do not offer this option are going to lose the right tail of the talent distribution.

You know, I found that with our project, with free economics radio, we are built to have remote work, although pre-pandemic, I was the only one that worked remotely. And that's just because I like being alone, and I'm a little bit anti-social. In the last, whatever, 14 months, we found that hiring is a totally different prospect because, of course, anyone can be anywhere.

If I can hire from not just the New York area, but anywhere else in the country or indeed the world, I have access to a much higher caliber pool and larger pool of labor. Yeah. That is absolutely correct. Also, it mitigates the frictions of immigration because you don't need to get workers on an H1B visa. You're not subject to that lottery system, which makes no sense, but it's also great for workers because they can make more real income.

I have a feeling that if someone were to listen to this conversation and you being really enthusiastic about the upsides of working from anywhere with data to back it up for sure. But if we were to take out every time you said the word work and plug in the word.

Learn or online education. I think people would laugh up roriously and think that you do not know what you're talking about because I think a lot of people during the pandemic have experienced that online education is really hard for young children, older college and so on. What do you think the work from anywhere revolution has to teach the more rudimentary online education revolution?

That's a very interesting question. I've not studied the education space. So to be honest, I don't have a research based answer, but one related answer is.

I think it really opens up the opportunity to consider colleges and universities which are not normally in the radar of the HR departments, because now you really have an opportunity to go and hire from Idaho or Kenya. And I think that's good news for the education industry. And maybe our biases will prevent us from doing that. I mean, says the guy at Harvard Business School, right? We should take this with a grain of salt.

But I think that is the real opportunity, right? And, you know, I've actually done some empirical work there. And what we found was that the right tale of the distribution in these lesser known colleges in India, we found based on a standardized test score of math that they were much higher than the middle of the distribution in the elite colleges. And I've not tested that in the US. That should be the same. So I think that's the real opportunity with respect to higher education.

Raj Choudhury, as much as he supports the work from anywhere revolution, doesn't deny the importance of face-to-face interaction, at least in certain contexts. He once did a study of inventors based in India. He found that when an inventor's manager visited a company's US headquarters, the Indian inventors were twice as likely to receive a patent.

Chatterie is also working on a study that looks at what happens when coworkers travel to a company retreat in the same car to and from the airport. When we find that the people who travel together, they help each other during the COVID months. We're still writing that paper up, but I think face-to-face is here to stay as is remote work. The equilibrium is to find ways to facilitate face-to-face within the remote work model.

There are these ideas, collision of ideas, some of it's deliberate, some of it's spontaneous. But there is no Eureka coming over a Zoom call, I would submit. That, again, is the New York real estate developer Marianne Gilmartin. Remember, she recently opened her own firm.

Mentoring and upskilling really requires interactions. And my decision to rent space in April and not wait is because of the age and the ambition of the people that work for me. And they want to be part of something bigger than themselves. And they don't want to be in their apartments. They want to be amongst other bright people. If I can look at my workforce and say, craft your perfect arrangement, how do you want your life?

to be lived, tell me what works. I'll get different answers, so one person on my team would love to go to Maine in August with her family and her children. I probably could do that for her now because I know that it's going to be okay. But that's different than saying over a very long period of time, people not being with people is sustainable and that talent is going to find that invigorating

And that great ideas are going to be surfaced and great things achieved. I think it's not true. That may not be true, but some places are acting as if it is. They are acting as if anyone can live anywhere, regardless of their employer.

Places like Tulsa, Oklahoma. Here again is Raj Choudhury. So I was studying work from anywhere in 2016, 2017, and then I stumbled upon Tulsa remote in late 2018. Tulsa remote is a project funded by the George Kaiser Family Foundation, which promotes equality and inclusivity in Tulsa, a city of about 400,000 people.

In the years just before the pandemic, the state of Oklahoma had its worst population outflow in years as college graduates kept moving away. Tulsa remote wanted to reverse this brain drain, although you didn't have to be from Tulsa to take advantage. You just had to prove you lived outside the state of Oklahoma.

And the incentive was a financial one. They offer $10,000 to each remote worker who would move to Tulsa and stay there at least for a year. A lot of cities use incentive schemes and tax breaks to attract employers, but Tulsa remote taking advantage of work from anywhere technology was going after employees. The reason I like the Tulsa remote experiment is it talks to this chicken and egg problem

when you're trying to attract companies, because the company might get a huge tax break, but the company also needs workers. And traditionally, the problem has been companies don't want to come to these small towns because there's not enough talent. And then workers don't want to come there because they're not enough companies. But if you can move workers,

who could remotely work for other companies. Then you're breaking this vicious cycle. In 2018, the first year of Tulsa remote, there were 25 slots available for the $10,000 bounty. After receiving 10,000 applications, they bumped it up to 100 slots. Choudhury is now studying these new Oklahomans.

When I interviewed them, disproportionately, the reason they gave for why they were moving to Tulsa was either cost of living or getting a much cheaper house to raise their family, but also this opportunity to contribute to the community.

So these were very pro-social people. They really wanted to give back to the Tulsa community in meaningful ways. In his preliminary analysis of the program, Choudhury has found that the state of Oklahoma gained around $2,000 to $3,000 a year from each new Tulsa worker. And so as a result of that, the state of Oklahoma got really excited about the program. And I believe they just passed this bill

which would reimburse Tulsa remote from the government if the remote worker stays for at least a year. So now the government is sort of getting involved. Remember, the experiment had been funded so far by a foundation. And I think it's great news for the financial sustainability of this program because now Tulsa remote could scale up and get more workers to relocate to Tulsa.

All this is good news, good ammunition for a work from anywhere argument. The same could be said for Raj Choudhary's patent examiner study and for Morris Davis's study too, which showed a huge increase in work from home productivity since the start of the pandemic. Still, all these studies revolved around a certain kind of worker, a worker whose talents in education and training made them eligible for working from home in the first place.

But that does not describe all workers. This was one more assumption that Morris Davis and his colleagues factored into their predictive model. We expect income inequality to widen as a result of this work-from-home revolution. You're much more likely to benefit from work-from-home technologies if you're already a highly skilled, well-compensated employee. We've boosted the productivity of people that were already productive and made a lot of income.

You know, work from home technology doesn't help you if you're a barber or if you are a delivery person. Because the work from home technology disproportionately affects people that are already high wage and high skill, we expect income inequality to widen by a significant margin. Do you have a number? Yeah, we do. Right now, the ratio of income, of high skill to low skill, where high skill is defined as a Fourier college degree. Right now, the number is about 1.8.

We expect that number to increase by 7% to a number like 1.9 to 1.93. And moreover, talking about the high-income high-skill gains, I mean, I don't mean to sound immodest. We are them, right? It's the high-income high-skill workers who are telling the story of this working-from-home revolution. That's right. By the way, this work-from-home revolution is a continuation of trends that have started since the mid-70s.

There's a concept called skill bias technical change. It means that the inventions that have come online over the past 40, 50 years, they've primarily benefited high skill workers. So this is just an unfortunate continuation of trends that have been in place.

As we've heard today, there are a variety of moving targets, a variety of unknowable variables when you think about the future of work. If you are a big believer in technology per se, you may think that lower skilled workers will also benefit, eventually. If you're a skeptic, you might see the gap widening.

We've been exploring the relationship between technology and labor for years on this show. Most recently, in episode 461, how to stop worrying and love the robot apocalypse, and we'll continue to talk to smart researchers like Morris Davis and Raj Choudhury to figure things out. Thanks to them and to Marianne Gilmartin for all their wisdom today. Meanwhile, coming up next time on Freakonomics Radio,

the debut of a brand new podcast that's going to ask and answer some fascinating questions. Questions like why children with summer birthdays are more likely to get the flu, or why patients with heart problems do better when cardiologists leave town.

The Doctor is in, but this Doctor is also an economist. A new show from the Harvard Double Threat Bapugena. It's next time on Freakonomics Radio. Until then, take care of yourself, and if you can, someone else too.

Freakonomics Radio is produced by Stitcher and Renbud Radio, and it's part of the Freakonomics Radio Network, which also includes no stupid questions. People I mostly admire and Sudhir breaks the internet. We can be reached at radio at freakonomics.com. This episode was produced by Mary DeDuke.

Our staff also includes Allison Craiglow, Greg Rippin, Joel Meyer, Trisha Bobita, Mark McCluskey, Zach Lipinski, Rebecca Lee Douglas, Brent Katz, Morgan Levy, Emma Terrell, Lyric Bowditch, Jasmine Klinger, and Jacob Clemente. Our theme song is Mr. Fortune by the Hitchhikers. All the other music you hear was composed by Luis Guerra. You can get the entire archive of Freakonomics Radio or any of our shows on any podcast app.

If you'd like to read a transcript or the show notes of Freakonomics Radio, that's at Freakonomics.com. As always, thanks for listening. I have a blank slate to just talk. You do? We'll chop it up and butcher it, don't worry. We'll misquote it in all kinds of ways. Oh, perfect, perfect. The Freakonomics Radio Network. Stitcher.