StreetLight Data, a transportation analysis firm, examined congestion in Miami with the goal of gaining a better understanding of how “gig-driving,” — the casual name for newer ride-hailing apps like Uber and Lyft, but also delivery services like Instacart or AmazonFresh — could be contributing to traffic congestion.
“That type of driving has started to cause a lot of political anxiety, and a lot of conversation,” said Laura Schewel, StreetLight Data's CEO. “And there’s a lot of chatter about, well, is this making traffic better, or is it making traffic worse? There [are] strong feelings on either side.”
It's hard to blame gig-driving as the culprit behind today's congestion woes, according to Schewel. “We think that the implications of gig-driving are far too complex to sum up in, ‘oh, it’s bad for traffic.’ Or, ‘it’s bad for business,’ and that sort of thing,” she added. “The answer is, with something this big, and something this diverse, the answer is, ‘it depends.’”
StreetLight Data hypothesizes that there is an interaction between gig-driving and congestion. However, how that interaction plays out depends on the context and makeup of the urban environment. The company examined a number of contexts and variables in Miami, looking at existing land use, location, time of day, urban density, transit options and other factors that could influence how people move through cities.
“Our goal was to show, when you analyze gig-driving at a very granular level, you find out that it has a very variable relationship to congestion, and that by looking at data in such a granular way, you can arm policymakers with data so that they can make effective decisions,” said Schewel, “so that they don’t come down and try to undo a massive mega trend and say, 'OK, this has positive and negative impacts we know.' [It's important to know] what kinds of intervention could mitigate some of the negative impacts on congestion, in an intelligent way, and not just try to blow up this whole important new thing that brings a lot of benefits to a lot of people.”
StreetLight collects location information from mobile devices and uses that data to gain insights into how drivers move through cities and regions, as well as testing whether certain traffic and transportation policies are performing the way officials want them to. In the Miami study, company officials worked with industry partners Cuebiq and Kochava Collective to follow location and travel data for anonymous users who had ride-hailing and other similar apps on their phones. By looking at driving patterns and vehicle miles traveled, the analysis was able to parse out gig drivers from ordinary motorists. The Miami metro was selected as a study region, in part, because of the variety in urban density, robust nightlife, as well as being home to major airports.
“I should say that we found so much variation in gig impacts that it’s important to say these results only hold true for Miami,” said Schewel. “A different city, like Oakland or Sacramento, might have a totally different set of relationships.”
In some of the densest parts of Miami, gig-driving did not seem to add to congestion, the study found. One theory is that these areas — by their design with dense urban settings and limited parking — seemed to discourage personal auto use, prompting visitors to use a ride-hailing service.
“But, if you look at the medium-density areas, where there’s just some grocery stores and bars and restaurants, then, the more gig-driving, the greater the congestion,” said Schewel.
This pattern seems to fly in the face of a similar recent study in San Francisco that explored the role of ride-hailing and congestion. Those findings pointed to significant congestion and traffic slow-down in the city’s dense business and financial districts.
“There’s a very complex relationship between economic activity, congestion and gig, and our core hypothesis is, you can’t make blanket statements about gig,” said Schewel. “You need to be really granular to measure it effectively.”
However, if cities truly want to address congestion — both in downtown districts and on the vast networks of highways funneling into and out of these regions — they need to find approaches that reduce the load of personal cars, said Jonathan Hopkins, executive director of Commute Seattle, which advocates for the use of transit and other sustainable forms of mobility like biking and walking.
A person traveling in a car occupies 100 square feet, said Hopkins, while someone on a bus occupies only five square feet.
“And so all of a sudden when you expand the envelope that someone is in, it’s going to cause congestion,” said Hopkins.
The role for cities and transit agencies, said Hopkins, is taking steps to incentivize congestion-mitigating transportation options, while discouraging the forms of mobility that contribute the most to traffic, offering options like congestion pricing, which places a premium on traveling through certain districts during certain times.
“The challenge for the TNCs [transportation network companies], is that places where they are liable to make the most money the most efficiently, is downtown in cities,” said Hopkins. “But they are the least efficient mode in that location … And they’re also not good for the environment. And so, we should discourage that. But if it’s to use Uber and Lyft to get to a park-and-ride so they can ride the bus, then that’s a net gain. And that allows more people to increase the efficiency of the system,” he added.
Traditionally, cities have sought to reduce congestion by investing in public transit. However, the Miami-Dade Transit Agency has experienced steep declines in ridership in recent years. From 2014 to 2018 ridership has dropped 26 percent, according to a Miami-Dade County statistic.
Officials at StreetLight Data steered clear of laying out absolute approaches to solving congestion, and continued to stress the need for data as city and transportation officials move forward with congestion mitigation plans.
“We’re trying to get the planning community interested and knowledgeable,” remarked Schewel.
“The relationship between this, congestion, social equity, climate change, VMT [vehicle miles traveled], is extremely complex,” Schewel added. “And the policies to manage it are going to have to be guided by granular data.”
