Can TNC Drivers affect Vehicle Ownership in the U.S.?

January 31, 2019

Transportation network companies (TNCs) such as Uber and Lyft are disrupting transportation at a large scale – both, in how passengers use this service, and how it provides labor opportunities to millions of drivers. Despite TNCs’ promises to end car ownership (Dudley, 2016; Hutt, 2016), the U.S. is experiencing increases in vehicle ownership rates (Descant, 2018; Schaller, 2019). Based on my experience as a TNC driver for my PhD dissertation and continuing doing research in this topic (I also receive many emails trying to re-recruit me as a driver), I would like to offer a hypothesis that has not been further explored. Besides looking at the people that might not own a car due to choice (the “car-free” or “car-light”), we should be looking at those that might not own a car due to constraints (the “car-less”) since the socioeconomic and mobility differences among those low-car households are different (Brown, 2017). These trends in the opposite direction is most clear in places such as Southern California, where: 1) Los Angeles is the largest TNC market of any metro area in the U.S., and 2) the region added 2.3 million people and 2.1 million vehicles from 2000 to 2015, or an average of 0.95 vehicles per new resident. That’s a substantial change from 1990 to 2000, when the region added 1.8 million people but only 456,000 vehicles, or 0.25 vehicles per new resident.

           

The conventional hypothesis aiming at understanding the effect of TNCs on car ownership states that vehicle ownership is declining (or will decline) since passengers are able (or will be able) to rely less on privately-owned vehicles shifting from a car owner/driver pattern to a user/rider pattern (potentially shared and multimodal). This hypothesis is supported by survey-based studies showing that thanks to TNCs, some passengers are owning less vehicles. For example, a report surveying over 4,000 adults in major U.S. metropolitan areas (data from August 2015 – January 2016) found that 9% of TNC users had made reductions in vehicle ownership rates (Clewlow & Mishra, 2017); my own ethnographic study in Colorado (Fall 2016 data) shows that approximately 13% of 311 TNC users report owning fewer cars due to the availability of TNCs (Henao & Marshall, 2018); and a 2017 Reuters/Ipsos poll shows that of 584 Americans who said they disposed their personal vehicle within the last 12 months, 9% turned to TNCs as their main way to get around (Henderson, 2017). Furthermore, an analysis of U.S. state-level data found that per-capita vehicle registrations decreased 3% after TNC entry (Ward, Michalek, Azevedo, Samaras, & Ferreira, 2018).

 

However, causation between TNCs and changes in car ownership is difficult to discern due to many factors other than just TNC presence influencing the decision to own a car. For several years, factors have been noted in how an individual (or household) evaluates vehicle ownership including abilities, values, preferences, attitudes, and perceptions; as well as several other more specific determinants such as life stage, location of residence and workplace, access to transit, children. More recently studies have explored association with safety, economics and parking, travel time, quality, and convenience of available modes. And while vehicle ownership/registration, and vehicle use are highly correlated, they are not the same. For example, the rates of vehicle ownership can stay constant, but the frequency of use can vary; or the rate of vehicle ownership can vary but the frequency of vehicle use stays the same.

The focus of this post is to formulate a new and opposing hypothesis: “an increase in vehicle ownership is associated with TNCs (specifically, from people signing up to drive for TNCs)”. This hypothesis posits that certain individuals see TNCs as an opportunity to earn an income and buy/lease vehicles that they would otherwise not be able to afford. This could be especially relevant for people with low/moderate-income, unemployed and/or non-car owners. Nationwide, car ownership has been rising the most among people with low-incomes and foreign-born households. For example, the share of households without vehicles fell by two-thirds between 2000 and 2015 among certain subpopulations (e.g. among foreign born from Mexico), especially those with a previously high propensity to ride transit (Manville, Taylor, & Blumenberg, 2018). Finally, greater prevalence of car ownership changes has been associated with households experiencing changes in employment (Clark et al, 2016). To build on the ‘why’ for key explanations to this hypothesis, the following four points are further explored:

 

  1. TNC passenger growth means TNC driver growth

  2. Low driver retention means TNCs need to constantly recruit drivers

  3. Income as a main motivation to sign-up and low hourly wages

  4. People might need/want to own a vehicle (but they can’t!)

 

1. TNC passenger growth means TNC driver growth

Uber and Lyft in the U.S. market have been growing exponentially. In order to meet the passenger demand, Uber and Lyft need to provide adequate driver supply. TNC drivers in average provide approximately 1.75 passenger trips per h

 

our, so for every 100 passenger requests in an hour TNCs would need to have approximately 57 drivers in operation during that hour (Cook, Diamond, Hall, List, & Oyer, 2018; Henao, 2017). This picture is not simple as TNC demand is very dynamic (high and low peaks), and working hours per driver varies from casual (only a few hours) to frequent drivers with an average of 17.06 hours per week and 29.83 trips per week based on Uber data from millions of drivers (Cook et al., 2018), so TNC would need to have a number of drivers equal to 3.35% total passenger trip demand per week. The same Uber dataset shows that drivers only work in average for 13.25 weeks in a year, so the percentage of drivers needed during the study is equal to 0.25% the passenger trips during the study period. To have an idea of the number of passenger trips, reports filed by the companies with the city of Seattle shows that TNCs provided more than 91,000 rides on an average day in King County, or approximately more than 33 million in 2018 (Dickey, 2018; Gutman, 2018). In Massachusetts where TNC data is publicly available, there were approximately 64.8 million ride-hailing trips in 2017 (State of Massachusetts, 2017). Based on Uber and Lyft billionth rides milestones (Dickey, 2018; Lyft Blog, 2019), Figure 1 presents the growth in passenger rides. With these rates, and using U.S. market share of Uber and Lyft (Gessner, 2018), TNCs are giving rides at a pace of 180 million rides per month or over 2 billion rides per year in the U.S. To be able to provide this number of rides, TNCs need at least 5 million drivers per year in the U.S. (but could be much higher).

 

Figure 1. TNCs growth

 

 

2. Low driver retention means Uber and Lyft need to constantly recruit drivers

Uber and Lyft need to constantly recruit drivers to keep up with growing passenger demand and an inability to retain drivers. A study co-authored by an Uber employee and an Uber consultant using Uber data found that only half of its Uber-driver partners stay active after a year (Hall & Krueger, 2015) and a report from SherpaShare – an app platform for ride-hailing drivers – surveyed 963 drivers and found that the turnover rates for TNCs was notably high, with about only 35% of drivers being active more than 6 months, and only 20% of drivers being active more than 12 months (SherpaShare, 2015). A more recent study, co-authored by Uber economists with access to Uber data for 1.87 million drivers across 196 cities, shows that the 6-month retention rate for Uber drivers is 31.9% (Cook et al., 2018), which is similar to the SherpaShare report. Retention rates this low means that a company needs to hire several times the employer based needed over a year span.

 

3. Income as a main motivation to sign-up for TNCs and low hourly wages

While there are different motivations for drivers to sign-up for ride-hailing, the main motivation is income (Hall & Krueger, 2015; SherpaShare, 2015). And when it comes to Uber and Lyft, driver income has become a hot topic. For instance, a 2013 Wall Street Journal article stated that a typical Uber driver makes $100,000 a year in gross sales (MacMillan, 2013). After this estimate was questioned, TNCs reduced this income characterization to around $25 – $35 per hour. The previously cited Uber data articles shows that drivers grossed approximately $17.40 an hour across 20 cities (Hall & Krueger, 2015) and more recently,  estimated gross hourly earnings of $15.80 for the 1.87 million drivers across 196 cities (Cook et al., 2018) (although the study misleads readers by stating $21.07 per hour on the paper while neglecting the ~25% service fee that Uber charges its drivers). One issue with the studies using Uber data is that the ride-hailing wage rates only includes the times when the Uber app is turned on or when the drivers are “active” for specific times (Chen et al., 2017); this ignore times when drivers are traveling at the beginning/end of shift, driving to reposition to areas with high ridership (including with the app-off), and/or times when passenger requests are very low for an extended period. As a comparison, SherpaShare survey respondents working 21 to 25 hours a week collected in average $1,376 monthly before expenses for an hourly rate of about $15 per hour (SherpaShare, 2015). Another survey of around 1,000 drivers from a blog called the “Rideshare Guy 2017 report” estimates hourly earnings of $15.68 (Campbell, 2017). Early in 2018, there was a controversy over an MIT working paper entitled “The Economics of Ride-Hailing” where researchers used ride-hailing driver survey responses acquired via that blog to estimate median net earnings of $3.37 per hour (Zoepf et al., 2018). Uber’s chief economist criticized the report (Hall, 2018), and the report even garnered a response from Uber CEO, Dara Khosrowshashi, who tweeted: “MIT = Mathematically Incompetent Theories”. MIT admitted fault and revised their estimations, finding the median profit to be on the order of $8.55/hour and $10/hour (Zoepf, 2018). This example shows the delicacy and importance of this topic. My own ethnographic study – examining a unique and detailed dataset – estimates gross wages averaging $15.57 per hour, and net hourly wages ranging between $5.72 and $10.46 per hour given common expense scenarios. The numbers from all these reports suggest that many drivers earn less than the federal minimum wage.

 

4. People might need/want to own a vehicle (but they can’t!)

While buying a vehicle in the U.S. seems straight-forward, a few requirements need to be in place. Financial status is the main requirement, comprising of factors such as income, employment, and a healthy credit rating. So, what happens if a person needs or wants to own a vehicle but does not have a permanent full-time job, a decent income, a decent credit history, or someone to co-sign? People under these circumstances can’t be car owners. But what if a company – such as TNCs – helps with all these requirements? Wouldn’t that break those barriers to vehicle ownership? Financial schemes provided by TNCs might be helping many people obtain access to a car.

 

Assuming TNCs only need the minimum of 5 million drivers in the U.S. per year, this represents a rate of at least 2% of all light-duty vehicle registrations in the U.S. Given the four considerations – TNC industry growth, low retention rates, low earnings, and people that might need/want to own a vehicle but can’t – who would be the best candidates for Uber and Lyft to recruit? Perhaps populations with low/moderate-income, unemployed and/or people that need/want to own a car, but due to their personal economic circumstances can’t? And since the TNC industry needs MILLIONS of drivers per year in the U.S. alone, the question is: How many of those drivers are new vehicle buyers/leasers (thanks to TNCs financial opportunities) that would otherwise not be able to afford a vehicle? As I mentioned earlier on this post: “causation between TNCs and car ownership is difficult to discern since many factors other than just TNC presence influence the decision to own a vehicle”. However, researchers should at least consider this hypothesis as one of the many factors when exploring car ownership in the U.S. and potential implications for the future of mobility and energy impacts in cities worldwide.

 

 

References

Brown, A. E. (2017). Car-less or car-free? Socioeconomic and mobility differences among zero-car households. Transport Policy, 60, 152-159. doi:https://doi.org/10.1016/j.tranpol.2017.09.016

 

Clewlow, R. R., & Mishra, G. S. (2017). Disruptive Transportation: The Adoption, Utilization, and Impacts of Ride-Hailing in the United States. Retrieved from

 

Cook, C., Diamond, R., Hall, J., List, J. A., & Oyer, P. (2018). The Gender Earnings Gap in the Gig Economy: Evidence from over a Million Rideshare Drivers.

 

Descant, S. (2018, Sep 6). Are Drivers Ditching Their Cars for Uber? Not So Fast, Say Experts. GovTech.com. Retrieved from http://www.govtech.com/fs/transportation/Are-Drivers-Ditching-Their-Cars-for-Uber-Not-So-Fast-Say-Experts.html

 

Dickey, M. R. (2018, July). Uber hits 10 billion trips. Tech Crunch. Retrieved from https://techcrunch.com/2018/07/24/uber-hits-10-billion-trips/

 

Dudley, D. (2016, Sep 19). The Guy From Lyft Is Coming for Your Car. CityLab. Retrieved from https://www.citylab.com/transportation/2016/09/the-guy-from-lyft-is-coming-for-your-car/500600/

 

Gessner, K. (2018, Dec 17). Rideshare: With IPOs looming, Uber leads market share, but Lyft has gained ground. Second Measure. Retrieved from https://blog.secondmeasure.com/2018/12/17/rideshare-industry-overview/

 

Gutman, D. (2018, Nov 5). How popular are Uber and Lyft in Seattle? Ridership numbers kept secret until recently give us a clue. Seattle Times. Retrieved from https://www.seattletimes.com/seattle-news/transportation/how-popular-are-uber-and-lyft-in-seattle-ridership-numbers-kept-secret-until-recently-give-us-a-clue/

 

Hall, J. V., & Krueger, A. B. (2015). An Analysis of the Labor Market for Uber’s Driver-Partners in the United States. Princeton University Industrial Relations Section Working Paper, 587.

 

Henao, A. (2017). Impacts of Ridesourcing - Lyft and Uber - on Transportation Including VMT, Mode Replacement, Parking, and Travel Behavior. University of Colorado at Denver.  

 

Henao, A., & Marshall, W. E. (2018). The impact of ride-hailing on vehicle miles traveled. Transportation. doi:10.1007/s11116-018-9923-2

 

Henderson, P. (2017, May 25). Some Uber and Lyft riders are giving up their own cars. Reuters/Ipsos poll.

Retrieved from https://www.reuters.com/article/us-autos-rideservices-poll/some-uber-and-lyft-riders-are-giving-up-their-own-cars-reuters-ipsos-poll-idUSKBN18L1DA

 

Hutt, R. (2016, June 28). Uber CEO Travis Kalanick: soon, nobody will own a car. World Economic Forum. Retrieved from https://www.weforum.org/agenda/2016/06/uber-travis-kalanick-driverless-cars/

 

Lyft Blog. (2019). 2018 in Review: Putting Our Vision Into Action.  Retrieved from https://blog.lyft.com/posts/2018/12/19/2018-year-in-review

 

Schaller, B. (2019, Jan 7). In a Reversal, ‘Car-Rich’ Households Are Growing. CityLab. Retrieved from https://www.citylab.com/perspective/2019/01/uber-lyft-make-traffic-worse-more-people-own-cars-transit/579481/

 

SherpaShare. (2015). The top demographic trends of the on-demand workforce. Retrieved from https://www.sherpashare.com/share/the-top-demographic-trends-of-the-on-demand-workforce/

 

State of Massachusetts. (2017). Rideshare in Massachusetts. Retrieved from https://tnc.sites.digital.mass.gov/?_ga=2.262765444.905628367.1540430223-999829066.1529339334

 

Ward, J. W., Michalek, J. J., Azevedo, I. L., Samaras, C., & Ferreira, P. (2018). On-Demand Ridesourcing Has Reduced Per-Capita Vehicle Registrations and Gasoline Use in US States. Paper presented at the Transportation Research Board 97th Annual Meeting, Washington DC, United States.

Please reload

2018 by Alejandro Henao