For Lyft users, time is money—now, we know how much
This story was originally published by UChicago News. Read it on their site here.
Study suggests the U.S. may undervalue timesaving transportation infrastructure
How much money is time worth? It’s an age-old question, but University of Chicago economists and their collaborators think they have found the answer—$19.38 per hour. It’s a finding with policy implications, according to the economists, because it suggests the U.S. government may currently be underestimating the value of travel time.
To quantify the value of time, the economists created a field experiment involving 3.7 million users of the Lyft app in 13 cities. By manipulating the price quotes and wait times displayed on the app, the researchers were able to study how much customers were willing to pay for a faster rideshare pickup.
Released by the Becker Friedman Institute for Economics, the new working paper calculated the $19 figure by modeling the relationship between wait times and prices and extrapolating to find the value of an hour. The average wait time for a ride in the study was 3.08 minutes and the average price was $13.83.
The broad nature of the experiment, said Prof. John List, allowed the economists to estimate the value of time across different people, choice circumstances and market conditions. The authors estimated time to be 10% more valuable on weekdays than on weekends, 20% more valuable in downtowns as compared to suburbs and 50% more valuable during rush hour as compared to off-peak hours.
A pioneer in the use of field experiments in economics, List co-authored the paper with UChicago alumni Ariel Goldschmidt, Ian Muir and Jenny Wang—who now work as data scientists at Lyft—and former UChicago postdoctoral researcher Robert Metcalfe, now at Boston University. V. Kerry Smith, an emeritus professor of economics at Arizona State University, also collaborated on the paper.
In total, the scholars factored more than 14.8 million instances of user engagement with the Lyft app into their analysis. Their findings relied on data collected before the onset of the COVID-19 pandemic.
As a general rule, time was also more valuable for Lyft riders when other modes of transportation were available—meaning there was greater competition—and when it was raining or snowing. Rain increased the value of time about 15%, while snow had a bigger effect, increasing time’s value by about 20%, because people prefer not to stand outside in the cold.
The U.S. Department of Transportation uses its own value of time estimates to evaluate what new infrastructure projects it should take on. The government values time at about $14.20 per hour at most, according to the authors, because the government’s estimates vary between one-third and one-half of the median wage rate, depending on the policy. The $14 figure is still about 35% lower than the $19 amount calculated in the study, which is close to the median hourly wage rate in the U.S. after taxes.
“Our value of time estimate is larger than that which is currently used by the U.S. government, suggesting that society is undervaluing time improvements and subsequently underinvesting public resources in timesaving infrastructure projects and technologies,” said List, the Kenneth C. Griffin Distinguished Service Professor in Economics.
“Our value of time estimate is larger than that which is currently used by the U.S. government.”
On the basis of these findings, the study’s authors recommend that policymakers account for differences in the value of time based on location and time of day when estimating the costs and benefits of new public projects, and increase their rule-of-thumb value of time estimate to 75% of the mean wage rate.
“Time is the ultimate scarce resource, and its value has deep implications for a range of economic phenomena and investment decisions,” the authors wrote. Among these, commute time is especially important because it impacts how people decide where to live.
Although the study did not evaluate the importance of transportation reliability, the authors acknowledged that such factors—including the accuracy of estimated pickup and arrival times—could be important in determining rideshare usage.
They argue that future studies should consider the value of reliability, possibly in the context of decision making about both rideshares and other travel options like buses, trains and bikes.