Konomu Suido is a rarity among Tokyo executives: most evenings, he gets home in time to cook dinner with his girlfriend. “There’s an expectation in Tokyo that you shouldn’t leave work before your boss,” he explains. For many of the city’s suits, sleep is often fleeting.

It’s the same in many cities around the world, with working hours and lengthy commutes encroaching on bedtimes. Sleeplessness seems a particular problem – even a badge of honour – in urban centres of power and innovation, with rest long derided as an opportunity cost by society’s leaders. Margaret Thatcher claimed to sleep just four hours a night – now it’s Elon Musk defending his 120-hour work weeks, telling Arianna Huffington that sleep is not an option.

A growing industry has emerged to offer short-term respite to weary workers. In Tokyo, Madrid, New York and many other cities around the world, nap time is being monetised through sleep pods like those belonging to Pop & Rest, a new startup in London. At its Shoreditch base customers pay £15 to sleep for an hour while listening to ocean waves and inhaling scented oils.

When scoping out cities for further expansion, founder Mauricio Villamizar tells me he maps urban commuting times; the longer people commute to work, the less time they spend in bed. “As long as a city has a dense population of commuters and a business district, there’s a market for us,” he says.

But commuting times alone don’t show which city gets the least shuteye. Noise, light pollution, sunset times and individual factors like income all play a part. Where researchers once depended on labs for data, the arrival of fitness-tracking wristbands has made it possible to study entire cities – and find out which one sleeps the least.

According to 2014 data from Jawbone wearable technology, it is Japan’s capital city, where people clock an average five hours and 46 minutes each night. The data, collected from 1,600 people across 21 cities, found that Melbourne gets the most sleep, averaging 7 hours and 5 minutes each night. Close second is London, where people sleep for 7 hours and 2 minutes. Tokyo, Seoul and Dubai sit at the bottom end of the sleep spectrum. But the Jawbone data set, while sizeable, only represents those who were able to afford one of the gadgets.

Matthew Gibson, an economist who studies sleep and productivity, reckons that rest is linked to a city’s sunset time – and its wages. Gibson began thinking about sleep late one night while driving to a conference with his colleague, Jeffrey Shrader. The two professors were discussing the papers that their advisors wouldn’t want them to write. “We thought: ‘sleep takes up a third of our lives, but nobody in economics is looking at it’,” he says by phone from his Massachusetts office.

Together, Gibson and Shrader studied sunset times across different American cities and found that people tend to get less sleep in places where the sun sets later. By comparing sunset times with wage data, they show that in the long-run, sleeping just one extra hour per night can increase wages by up to 4.9%, with the amount of sleep a city gets also linked to the productivity – and pay grade – of its residents.

Using Salt Lake City and Denver as examples, Gibson and Shrader’s theory is striking. The two cities occupy the same US time zone, but in Denver the sun sets almost 30 minutes earlier. According to 2016 data from the US Census Bureau, Denver’s average household income was $7,446 (£5,695) higher than that of Salt Lake City.

But there’s a chicken-and-egg relationship between sleep and wages, Gibson cautions – sleep can be both a cause and consequence of income. Getting little sleep may be one of the reasons why someone is paid less, because it runs down their productivity and physical health. Conversely, concerns about money may make it more difficult to sleep.

This two-way relationship complicates researchers’ efforts to determine whether sleep deprivation causes poverty, or vice versa. It’s also tricky to figure out how much sleep deprivation among the urban poor is due to environmental factors like noise, crime and light pollution, which all affect one’s ability to sleep easy at night.

But if poor sleep is generally correlated with lower wages, do poorer cities sleep less? It’s hard to test the hypothesis, there being far less information about sleep patterns in cities near the equator than about western cities.

In Chennai, on India’s east coast, people fall asleep in improbable locations – by roadsides, on the back of cycle rickshaws, and in noisy and crowded streets. When development economist Heather Schofield first visited the city, she was struck by its numerous daytime sleepers. “They got me thinking – how tired must people be to fall asleep in these crazy places?” she says.

Night-time is frenetic in Chennai, where overcrowding and noise stem from rapid urbanisation and widespread poverty. Schofield describes how four or five people might be trying to sleep in a single room, with trucks rumbling past their door, in 90F (32C) degree heat, circled by thirsty mosquitos. On top of this are the compounded stresses of living in poverty.

Among development economists, sleep isn’t a mainstream subject. Schofield wants to change that, and as a result, the lab run by Schofield and colleagues Gautam Rao and Frank Schilbach in Chennai is pioneering a study into the impact of improved sleep on poverty reduction. So far, the lab has found that participants sleep an average of five and a half hours a night. By distributing sleep mats and earplugs, it has managed to increase night-time sleep by an extra 30 minutes.

Yet if less sleep is associated with lower wages, shouldn’t people in Tokyo, where average incomes are higher, be getting more sleep than those in Chennai?

The results don’t necessarily tally with what we expect. In fact, sleep may have more to do with whether people feel in control of their circumstances than the amount of money they earn. Michael Grandner, a psychologist at the University of Arizona, explains that exceeding a comfortable income bracket tends to be associated with sleeping less – not more.

This pattern can be explained by how much control individuals feel they have over their sleep patterns. As Grandner puts it, “Some people need to work three jobs to put food on the table. For others, their time is too valuable to take a break. Neither feels like they can afford to sleep”.

Ultimately, that means for people living in crowded Chennai or overworked Tokyo, sleep is more likely to feel beyond their control – and therefore be fleeting.

Follow Guardian Cities on Twitter, Facebook and Instagram to join the discussion, and explore our archive here