The Economic Consequences of Increasing Sleep Among the Urban Poor

Gautam Rao, associate professor at the Haas School of Business and the Department of Economics at UC Berkeley, has a research focus on integrating psychological insights into economics, particularly on topics relevant to developing countries. In this short interview with Ulrike Malmendier, Rao describes his work exploring chronic sleep deprivation in urban India. His findings show that the urban poor in India experience severely fragmented sleep, comparable to people with clinical sleep apnea or to a 90-year-old from the U.S. His team found that increasing sleep duration by 30 minutes per night had no significant cognitive, productivity, or health benefits, which suggests that poor sleep quality, rather than insufficient sleep, is the primary concern. Short afternoon naps (20–30 minutes) led to notable improvements in productivity, psychological well-being, cognitive function, and financial decision-making. The study highlights the restorative potential of naps compared with extended nighttime sleep. Gautam’s work challenges assumptions in sleep science and he advocates for field-based research that prioritizes real-world interventions.

Ulrike Malmendier:

How did you first become interested in research on sleep deprivation? What’s the motivation behind you working on it?

Gautam Rao:

The motivation comes from my collaborators, Frank Schillbach and Heather Schofield, who were grad students at the time. They were walking around Chennai, India, looking for research ideas. They noticed that in the middle of the day­–in 40-plus degrees Celsius heat and high humidity­­–people were napping in the back of a cycle rickshaw or on the side of the street. They were struck by how tired people must be in these conditions, and how these are the conditions in which they need some rest. I thought their observation was great and it’s one of the reasons why I’m so happy to be working in the U.S. on topics often based in India, because it means I end up collaborating with people from around the world and different people notice different things. I’ve seen people passed out, sleeping in the afternoon or in difficult conditions in India millions of times. I saw it so much growing up in India that it became completely invisible to me. It took an outsider’s perspective for me to see it. Once they pointed it out to me, it set my mind on fire – we looked into research in sleep medicine, which told us that being up all night is basically as bad for your cognitive function as being legally drunk.  We asked, “What is the world really like? What if people in the world are actually severely sleep deprived all the time?” We realized that nobody had documented or objectively measured sleep systematically in the developing country context. We didn’t know the first thing about it.

We found sleep science research very exciting and got into it and thought this could be really important for people’s lives. So it became an obvious thing to work on, and that’s how it started.

Ulrike Malmendier:

And it’s interesting, given that in my mind, I always connect sleep research to mental health research.  I always thought there was some grand initial vision where we need to take the whole body and mind more into account, but it was actually specifically the observation of the naps. That’s very cool.

My next questions: What are two or three of the key takeaways from your research? And what surprised you the most?

Gautam Rao:

The most important takeaway is that people in cities and developing countries – particularly in a major city like Chennai, India, they have absolutely terrible sleep quality. If you look at their sleep objectively measured over the course of a night using actigraphs, they sleep like a 90-plus year old in the U.S., or they sleep like someone with severe clinical sleep apnea, which means highly interrupted sleep. These are not people selected because they are concerned about their sleep or because they have sleep issues or individuals selected because they are getting treatment for their sleep. So in one night of sleep, even though people are in bed for about eight hours a night, which is what you’re supposed to be doing, they’re only getting about five and a half hours of actual physical sleep out of it. And they are waking up for at least a minute or two or longer for 30 plus times a night. Their sleep is very fragmented, which means it’s very hard to get the full benefits of sleep. I would say that’s the most important finding from our paper.

Then what we try to do is ask, what can we do to improve people’s sleep? Can we get them to get more sleep at night? Can we instead give them a more comfortable place to get some rest at their workplace in the afternoon? Can we offer them naps?

And we were quite successful at increasing the total amount of sleep people get in both these ways. But the finding that quite surprised us is getting more sleep at night––this is about 30 minutes more of sleep per night–– which is more than the effect of sleeping pills––didn’t translate into any discernible benefit to people’s lives, at least not over the course of a month. There were no improvements in their cognitive function, in their productivity at work, in their decision making, or in measures of their physical or mental health. Now, in the longer term, some further benefits may emerge, but this was a surprising finding. In fact, we documented that it’s surprising by having experts from both sleep science as well as from economics to help make predictions about what they expected we would find if we got people to sleep half an hour more per night.

The experts thought there would be big positive effects. If anything, there’s an important negative effect, which is if you’re spending more time in bed, you just have less time available to do other things you value in life, including work. So, we found that sleeping more at night really doesn’t do much. But in contrast to that, getting people to take a short nap at around 1.30 p.m. for about 20 to 30 minutes every day led to improvements in productivity, which made up for the time lost sleeping. You got as much work done and made as much money, even though you were taking time out of your day to take a nap.

Napping improved psychological well-being, mental health, and cognitive function. It made people more patient, and less likely to delay work for later, but also more likely to invest money in attractive savings and instruments, which we provided people with, so they saved more money.

There were a range of benefits from naps, which we didn’t see from sleeping at night, even though the naps were shorter. We can’t say exactly why that is, but we think part of the reason is that naps provided a higher quality of sleep.

We were able to get people to sleep more at night in their home environments, but we couldn’t improve their quality of sleep. One interpretation is sleeping more, but with bad quality sleep is just not helpful, but higher quality sleep, and maybe particularly in the afternoon, is different than night sleep. This had a lot of benefits.

We were surprised all the way through with what we learned. Many of my priors turned out to be wrong. And in fact, the priors of the sleep scientists and a lot of economists also turned out to be wrong.

So, that for me is the joy of research- I keep discovering all the ways I’m wrong, and I find that very exciting.

Ulrike Malmendier:

That’s the true researcher’s attitude. And what you said about the quality restoration reminds me of this whole debate in health economics about how it’s not the lifespan we want to extend, but the health span. And you’re saying it’s not the sleep span but the sleep quality that we want to improve.

The other thing I take away from it: on the one hand, I’m desperately trying to increase my Fitbit measure of sleep duration at night and not managing it, and maybe it’s not so important after all.

Gautam Rao:

Yes, that’s right

Ulrike Malmendier:

My next questions is, where do you want to go next from here? What’s your view on where research should be going–other than further debunking myths established by sleep researchers.

Gautam Rao:

We wrote a short perspective piece in Science on this. And I’ve been trying to present our results to sleep scientists and so on. I think the one thing that sleep science really would benefit a lot from doing is moving to research in the field.

A vast majority of sleep science is either correlational studies, that is correlating people’s sleep with their health outcomes and economic outcomes, which has the limitation that it’s very difficult to infer causality, or there are experimental studies done in the lab, where you bring someone in and the modal is to keep people up all night to study a night of total sleep deprivation, which ends up being so different from a policy relevant sense. Moving to the field, studying the effects on outcomes that people naturally care about in life, like your actual work performance and so on, rather than just lab measures of performance, I think that that’s something sleep science needs to do.

I also think that expanding the scope of whom you’re studying to include these settings that have been previously not considered in developing countries is important. There’s a huge welfare gain, and there’s a lot of interesting science to be done there. Economists have started working on sleep a bit more. Sally Sadoff, Osea Giuntella, and co-authors have been doing work on college students in the U.S. These students are not sleeping for very different reasons from the people in India that I’m studying. College students in America may not be sleeping enough because of self-control problems, so can we give them incentives to sleep more? Does that translate into changes in educational outcomes? There’s a lot of exciting work happening there, which is studying sleep in the field and trying to see how it matters for real-world outcomes we care about.

Ulrike Malmendier:

I couldn’t agree more. You allude to the fact that more and more people are interested in the area and working on it. And I personally find this very encouraging to hear that from you. As somebody who doesn’t work on sleep but also thinks about her own work as going beyond traditional behavioral economics, to take into account what’s happening in both our mind and in our body, we’re not all just about preferences and updating priors, but a lot of stuff happens to us as we walk through life.

It’s still a little bit of work to convince people that this is part of economics. In fact, this is a super crucial part of economics and it’s a big oversight that we haven’t done more work on it already. So, the last question: Would you have any advice for junior researchers who share your interest and maybe our joint view that this is an important contribution to economics, widening this perspective if they want to get started in this area? Are there any pitfalls to avoid? Any piece of advice you might want to share?

Gautam Rao:

I’m always positively surprised by how open economists end up being to whatever thing I’m studying. I fear that once I start telling economists about what I’m doing, they’ll say, Why is this economics? But in practice, people are interested in things about the world. There are huge benefits to working on novel questions. \ One topic I’ve been working on a lot nowadays has to do with mental health, and then it turns out that tons of economists are starting to work or are working more on mental health. Whenever I feel like I’m working on something truly innovative and novel, it turns out it’s just that the idea is in the air and plenty of other people are also working on it, maybe six months ahead of me or one year behind me.

The whole point of being an academic and a researcher is to pursue what you find interesting, and I feel like economics ends up being very open, especially if you take a step beyond economics and look at the broader scientific role. You might say, well, what role do economists have here? What is our comparative advantage? With sleep, it’s very evident.

Here’s another comparative advantage we have; we are experts at using natural experiments. In public health and medical research, you either have clinical trials or you have correlational studies, and there’s really nothing in between, whereas we are very good at natural experiments.

So that’s something where we or economists are also preaching to the folks in public health and so on. We can study the field and real-world outcomes. We can think about things like opportunity costs. For example, in my work on sleep, what comes out of it is that sleep is actually pretty costly. When economist Dan Hamermesh first wrote down these Chicago style models, price theory models of sleep, it was obvious that sleep comes with some cost. That notion that sleep has a cost is completely missing from sleep science and sleep medicine.

I would say, just chase the things that seem important about what could matter for people’s lives that seem understudied, especially those that excite you. If you’re a young economist, this is your time to go do those things. I think people can be convinced. Economists end up being interested in everything. We always do things with our own sensibility and end up adding some value.

One other piece of advice is don’t do anything quite as complicated as my sleep experiment – a lot of blood, sweat, and tears went into it. Looking back, we could have done one fifth of what we did and still had a very nice paper.  In fact, a paper that would be easier for readers and referees to comprehend. Don’t think that one paper that you write necessarily needs to answer every possible question. That would be one way in which you should not emulate me.

Ulrike Malmendier:

I love the advice. Just do good economics. Use the great tools economics has to offer and otherwise follow your passion and follow what you think is interesting. I think that’s beautiful! Thanks so much, Gautam.

Read Paper Here