Equitable Electrification: Early Findings from the Uber-Hertz Partnership
As one of the largest mobility platforms in the world, Uber has a responsibility to join global leaders to aggressively tackle the challenge of climate change. Uber has committed to become a fully zero-emission platform by 2030 in the US, Canada, and Europe. In October 2021, we launched a partnership with Hertz that will make up to 50,000 fully electric Teslas available to rent by 2023, exclusively for drivers using the Uber platform in the US and Canada.
This article highlights early findings, showcasing the sustainability and equity implications of the partnership.
Program growth has led to emissions reductions
Since we announced the partnership in October 2021, Teslas rented through Hertz have now completed over 5 million trips and traveled more than 40 million fully electric miles in over 30 different U.S. markets. Figure 1 showcases trip growth in the Bay Area, where trips today now cover nearly all neighborhoods throughout San Francisco.
With the growth of this program, the total amount of avoided emissions have correspondingly grown. As of August 1, 2022, over 19.9k metric tonnes of on-trip CO2 tailpipe emissions have been avoided across the United States.¹ This is equivalent to avoiding roughly 2.1 million gallons of gasoline.
The partnership has also coincided with a large growth in the percentage of zero emission vehicle (ZEV²) use across the Uber platform. In Q2 2022 ZEVs accounted for more than 2.5% of on-trip miles in the US, up from 0.34% from the previous year (Figure 2). While the Teslas rented from Hertz make up a significant portion of this growth, it’s worth noting that use of other, non-rental ZEVs on the platform has also increased. In California, for example, total non-Hertz Tesla ZEV utilization by drivers using Uber’s app nearly tripled since the Hertz Tesla program inception, from 0.97% of all statewide on-trip miles during November 2021 to 2.61% in July 2022. To provide context, in 2021 ZEVs made up 0.5% of all vehicles across the US and 1.6% of all vehicles in California.
Balanced adoption across neighborhood demographics
The impacts of climate change and localized air pollution due to vehicle emissions hits low-income communities and communities of color hardest. These communities disproportionately live near roads and highways, leading to higher rates of asthma, cancer, and other pulmonary illnesses (Holland et al, 2019).
Inequitable electric vehicle (EV) adoption only serves to further widen these divides. In the state of California, EVs are currently distributed inequitably. Using publicly available vehicle registration data and U.S. Census Bureau household income data, we find that EV adoption is highly skewed to wealthier neighborhoods. Zip Code Tabulation Areas (ZCTAs) within the two lowest-income quintiles (<$70k) contain 44% of all registered vehicles but only 21% of EVs, while ZCTAs within the highest-income quintile ($109k<) contain 17% of all vehicles but a staggering 38% of all EVs (Figure 3, left).
These disparities are partly attributable to the high cost of EVs, which are more aligned with luxury vehicle prices than mainstream vehicle prices. However, some programs and policies intended to increase the proliferation of EVs are not constructed with equity in mind, and may also contribute to EV inequality. For instance, many government incentive programs for electric vehicles are designed to be received post-purchase, often excluding lower-income people from accessing the subsidy.
We conducted an analysis to examine driver participation in our partnership with Hertz to better understand equity related outcomes. We do not regularly collect demographics for drivers who use the Uber platform, so as a proxy, income levels and racial distribution are imputed by their home addresses (obtained from their driver’s licenses). Addresses are aggregated by ZCTA. For each ZCTA, the percentage of drivers that rented a Tesla vehicle through Hertz during May 2022 is calculated. This information is then compared with ZCTA demographic data from the U.S. Census Bureau. For this deeper dive analysis, we only looked at California data.
We find that the distribution of Hertz Tesla drivers relative to all drivers who use the Uber platform is generally balanced, skewing only slightly to higher-income quintiles compared to all drivers using the app. 58% of all CA drivers live in the two lowest-income quintiles versus 55% of Hertz Tesla drivers (Figure 3, right).
Visual inspection of heatmaps across Los Angeles reveals similar patterns. Figure 4 demonstrates overall EV adoption rates (left) and Tesla rentals with Hertz through the Uber partnership (right) at the ZCTA level. ZCTAs in the bottom two quintiles of median household income are outlined in red. Lower income ZCTAs generally correspond with EV ownership deserts. This clear correspondence does not exist for Tesla rental adoption, which varies throughout all ZCTAs.
Regression analysis was performed to statistically validate the relationships described by Figure 3 and Figure 4. Neighborhood racial demographics were also factored in, with ZCTA-level Census data used to determine the percentage of population that is non-white. Two multiple linear regressions were performed, one modeling EV adoption percentage and the other modeling Hertz Tesla adoption percentage. Table 1 shows the results of these analyses.
Racial and socioeconomic inputs have strong explanatory power over EV adoption percentage, as demonstrated by the model’s high R² value of 0.562. Median household income and racial demographics both have low P-Values under 0.001, implying they correlate significantly with ZCTA-level EV adoption. Higher median household income and a lower non-white population percentage both correlate with higher levels of EV adoption.
Using these same two factors to model ZCTA-level Hertz Tesla adoption produces a model with far less explanatory power. The model has a low R² value of 0.020, and both factors have high P-Values, implying that neither median household income nor racial demographics correlate significantly with Hertz partnership adoption.
A springboard towards long-term EV adoption
Not only does the rental partnership with Hertz have lower equity and access barriers for drivers hoping to utilize EVs, early evidence suggests it may be a springboard towards long-term EV adoption. For 95% of drivers in the program, this was the first-ever EV they had driven on Uber’s platform. 77% of these drivers plan to either continue using their Hertz Tesla vehicle or switch to another EV.³ In fact, since program inception, we have already seen over 200 drivers graduate from a Tesla rental with Hertz and have begun to drive an externally acquired EV (either bought or leased) on the Uber platform.
In an Uber Newsroom post, we interviewed several drivers who expressed satisfaction with their experience renting a Tesla through Hertz. Results from an internal phone survey of over 1,800 drivers suggest that these sentiments are broadly shared. In fact, 94% of all drivers report a positive experience driving a Hertz Tesla vehicle on the Uber platform.
One difficulty some EV drivers on the Uber platform face is equitable access to charging. Only 19% of Hertz Tesla drivers report charging at home³, meaning the vast majority of drivers must utilize public charging options. Research conducted by RMI and Bloomberg found that in Los Angeles and New York City, respectively, the distribution of public fast chargers is highly skewed towards wealthier areas. In order to better serve the growing population of EV drivers on the Uber platform, resources must be spent towards developing a more equitable charging network.
While the opportunity of our partnership with Hertz is actively being realized, we recognize that this model of electrification will not work for every driver on the Uber platform. This program is designed for rental drivers, who typically spend more hours on the platform than drivers who own their vehicles. For other drivers hoping to electrify, we have a suite of other programs including charging discounts with EVgo and Wallbox, electric vehicle discounts with GM and Nissan, and an extra $1/trip up to $4k a year for drivers on Uber in zero emissions vehicles. We’re excited to continue to expand EV options for drivers on Uber, as we work together to reach zero emissions.
[1]: The amount of emissions “avoided” by Hertz Tesla drivers is calculated by assuming a counterfactual whereby they would have driven the same number of miles in a vehicle with average fuel efficiency for vehicles on Uber per metropolitan market (weighted average carbon intensity of 299 gCO2/mi).
[2]: ZEVs include battery EVs and hydrogen fuel cell vehicles. Well over 99% of ZEV drivers using Uber’s app in the US and Canada use battery EVs.
[3]: Results from an internal phone survey of over 1,800 drivers.
