Opinion
Case No. 17-cv-02664-RS Case No. 17-cv-06124-RS
2022-07-25
Andrew David Bizer, Pro Hac Vice, Garret Scott DeReus, Bizer and DeReus, LLC, New Orleans, LA, Karla A. Gilbride, Public Justice, P.C., Washington, DC, William Brock Most, Aqua Terra Aeris Law Group, Albany, CA, for Plaintiffs Stephan Namisnak, Francis Falls in 17-cv-06124-RS. William Brock Most, Aqua Terra Aeris Law Group, Albany, CA, Andrew David Bizer, Pro Hac Vice, Garret Scott DeReus, Pro Hac Vice, Bizer and DeReus, LLC, New Orleans, LA, for Plaintiffs Scott Crawford in 17-cv-02664-RS. Anne-Marie Estevez, Morgan, Lewis and Bockius, LLP, Miami, FL, Clara Noelle Kollm, Pro Hac Vice, Patrick A. Harvey, Stephanie Beth Schuster, Pro Hac Vice, Morgan, Lewis and Bockius LLP, Washington, DC, Ethel J. Johnson, Pro Hac Vice, Morgan Lewis and Bockius LLP, Houston, TX, Kathy Hua Gao, Morgan Lewis and Bockius LLP, Los Angeles, CA, for Defendants in 17-cv-06124-RS, 17-cv-02664-RS.
Andrew David Bizer, Pro Hac Vice, Garret Scott DeReus, Bizer and DeReus, LLC, New Orleans, LA, Karla A. Gilbride, Public Justice, P.C., Washington, DC, William Brock Most, Aqua Terra Aeris Law Group, Albany, CA, for Plaintiffs Stephan Namisnak, Francis Falls in 17-cv-06124-RS.
William Brock Most, Aqua Terra Aeris Law Group, Albany, CA, Andrew David Bizer, Pro Hac Vice, Garret Scott DeReus, Pro Hac Vice, Bizer and DeReus, LLC, New Orleans, LA, for Plaintiffs Scott Crawford in 17-cv-02664-RS.
Anne-Marie Estevez, Morgan, Lewis and Bockius, LLP, Miami, FL, Clara Noelle Kollm, Pro Hac Vice, Patrick A. Harvey, Stephanie Beth Schuster, Pro Hac Vice, Morgan, Lewis and Bockius LLP, Washington, DC, Ethel J. Johnson, Pro Hac Vice, Morgan Lewis and Bockius LLP, Houston, TX, Kathy Hua Gao, Morgan Lewis and Bockius LLP, Los Angeles, CA, for Defendants in 17-cv-06124-RS, 17-cv-02664-RS.
ORDER AND OPINION
RICHARD SEEBORG, Chief United States District Judge
I. Introduction
Plaintiffs are three people using electric wheelchairs who asked Defendants Uber Technologies, Inc. and its subsidiary Rasier, Inc. (collectively "Defendants" or "Uber") to provide wheelchair-accessible vehicle ("WAV") service ("UberWAV") in their home cities of New Orleans, Louisiana and Jackson, Mississippi. After Uber declined, Plaintiffs sued, accusing Uber of violating the Americans with Disabilities Act ("ADA"). The cases proceeded to a bench trial, at which the parties presented evidence of various methods for implementing WAV service, the costs of WAV service, and Uber's implementation of WAV service in other cities, among other topics. For the reasons explained below, Plaintiffs have not met their burden of establishing that their requested modification is reasonable, and thus have not prevailed on their claim for violation of 42 U.S.C. § 12184(b)(2)(A) for failure to make a reasonable modification. Further, even though Uber's vehicle requirements screened out WAVs from operating on the platform, there is no guarantee—or even a strong likelihood—that a person requiring a WAV would be matched with such a vehicle on the UberX platform, and thus Plaintiffs’ claim that Defendants screened out people with disabilities in violation of 42 U.S.C. § 12184(b)(1) fails as well. This Opinion and Order comprises the findings of fact and conclusions of law required by Federal Rule of Civil Procedure 52(a).
Uber's administrative motions to file materials under seal, Crawford Dkt. No. 235 & Namisnak Dkt. No. 195, are granted.
II. Parties
Plaintiffs are people with disabilities who rely on electric wheelchairs for mobility. Unlike users of non-electric wheelchairs, which may be folded and placed into the trunk of a vehicle, users of electric wheelchairs require a WAV to travel by car. Dr. Scott Crawford lives in Jackson, Mississippi while Stephan Namisnak and Francis Falls live in New Orleans, Louisiana. None of the Plaintiffs have downloaded the Uber app, but have all declared under oath that if Uber were to offer ride services that could accommodate electric wheelchairs, they would use them. Uber operates a ride-for-hire service that utilizes a mobile phone app to connect riders with drivers who have signed up with the app. The most popular type of ride is UberX, in which drivers provide trips in standard, four-door vehicles. Many drivers offer rides in their own personal cars resulting in a robust supply of UberX drivers, but these services vary from locality to locality. In some cities, such as San Francisco and Washington, D.C., the app includes "UberWAV," which offers app users the option to call a WAV. This option is not available to Uber app users in New Orleans or Jackson. Uber also operates other services, such as its UberEats food delivery service.
Defendants do not contest that Plaintiffs are disabled or the nature of their disabilities.
III. Plaintiffs’ Claims
Under Title III of the ADA, "[n]o individual shall be discriminated against on the basis of disability in the full and equal enjoyment of specified public transportation services provided by a private entity that is primarily engaged in the business of transporting people and whose operations affect commerce." 42 U.S.C. § 12184(a). Plaintiffs bring claims under two subsections of § 12184. First, they allege Uber failed to "make reasonable modifications consistent with those required under section 12182(b)(2)(A)(ii)." Id. § 12184(b)(2)(A). Section 12182(b)(2)(A)(ii) defines as discrimination the "failure to make reasonable modifications in policies, practices, or procedures, when such modifications are necessary ... unless the entity can demonstrate that making such modifications would fundamentally alter the nature" of the goods or services provided. Second, they allege that Uber imposed "eligibility criteria that screen[ed] out or tend[ed] to screen out an individual with a disability or [a] class of individuals with disabilities from fully enjoying the specified public transportation services[.]" Id. § 12184(b)(1). Plaintiffs also pursued a third violation of § 12184, averring that Uber purchased "a new van ... that is not readily accessible to or usable by individuals with disabilities" without providing "a level of service to such individuals equivalent to the level of service provided to the general public[.]" Id. § 12184(b)(5). Summary judgment was granted to Defendants as to this third claim. Additionally, other claims were dismissed as a matter of law earlier in the litigation. See Crawford Dkt No. 80; Namisnak Dkt. Nos. 84, 102.
IV. Summary of Evidence
In this bench trial, the parties presented evidence over three days. In addition to introducing a variety of exhibits, including many internal Uber documents, each side provided testimony. Each Plaintiff testified about how his disability required him to use an electric wheelchair, and that use of an electric wheelchair necessitates travelling by WAV, rather than a normal passenger vehicle. Each Plaintiff explained his desire to use Uber, his prior request that Uber implement WAV service in his city, and the challenges of navigating his city while using a wheelchair. Plaintiffs also presented the expert testimony of Dr. James Cooper, a transportation systems professional with expertise in ground transport, taxis and for hire vehicles, accessible transportation, and transportation policy.
Additionally, Plaintiffs called three current or former Uber employees, Robert Rupp, David Reich, and Niraj Patel. Rupp is a member of Uber's Vehicle Solutions Team, and previously managed a group of analysts who assessed data concerning driver base growth. Plaintiffs questioned him about vehicle requirements for drivers in New Orleans and Jackson, Uber's vehicle rental partnerships, and driver incentives. Reich, no longer an Uber employee, previously served as Uber's director of transit, and before that as head of product strategy. Reich was never directly involved with the UberWAV program. Patel is the director of rider operations for Uber in the United States and Canada, and previously served as strategy planning lead. Since 2018, Patel has served as the person at Uber with primary responsibility for supervising and managing development of UberWAV. Among other topics, Patel testified about existing UberWAV programs and Uber's consideration of expanding its WAV program to New Orleans.
Finally, Plaintiffs introduced deposition transcripts. Plaintiffs introduced deposition testimony of Meera Joshi, formerly the Chair and CEO of the New York City Taxi and Limousine Commission, an agency which handled regulation of for-hire WAVs in New York City. Plaintiffs also introduced deposition testimony of three Uber employees who also testified at trial: Brad Rosenthal, Robert Rupp, and Niraj Patel. Uber had designated Rupp and Patel as Rule 30(b)(6) witnesses.
Uber offered the testimony of Uber employees Brad Rosenthal and Connor Fagent, and also called Niraj Patel, who testified during the Plaintiffs’ case-in-chief. Rosenthal is the director of mergers and acquisitions integration, and previously served as director of strategic initiatives and in a variety of other roles at Uber. He described various aspects of Uber's business model. Fagent currently works as the team manager for the UberWAV program, and has worked on UberWAV since 2017. Fagent described advantages and disadvantages to different methods of implementing UberWAV and his team's considerations when launching UberWAV service in a city. He also provided cost estimates for operating a WAV program in New Orleans and Jackson. Finally, Patel discussed advantages and disadvantages of different methods of implementing UberWAV in new cities.
V. Dr. Cooper's Testimony and Uber's Motion to Strike
Following the presentation of evidence, Defendant brought a motion to strike certain portions of Dr. Cooper's testimony, arguing those went beyond the scope of his expert report. When presenting testimony from an expert witness providing a written report, Federal Rule of Civil Procedure 26(a)(2)(B) requires a party to disclose "a complete statement of all opinions the witness will express and the basis and reasons for them" along with "the facts or data considered by the witness in forming them[.]" A party must supplement its disclosure "in a timely manner if the party learns that in some material respect the disclosure or response is incomplete or incorrect, and if the additional or corrective information has not otherwise been made known to the other parties during the discovery process or in writing[.]" Fed. R. Civ. P. 26(e)(1)(A). Failure to supplement a disclosure may result in exclusion of the evidence "unless the failure was substantially justified or is harmless." Fed. R. Civ. P. 37(c)(1).
The challenged testimony here falls into five categories: (1) why Uber provides a WAV marketplace in certain cities but not others; (2) whether a city's size affects its ability to have a successful WAV rideshare market; (3) what constitutes the correct demand and level of supply for WAVs in New Orleans and Jackson; (4) whether Uber's vehicle requirements prohibit WAVs and suppress a possible supply of WAVs in New Orleans and Jackson; and (5) whether an accessibility fee or price increase could cover the cost of providing WAV service and what the marginal cost of providing WAV in New Orleans and Jackson would be. Aspects of that testimony, such as whether a city's size impacts the success of a WAV marketplace, are addressed his expert report. Others, such as how high an accessibility fee would need to be to cover the cost of operating UberWAV in New Orleans and Jackson are responsive to Uber's January 2022 document production (which came just weeks before trial), making a failure to supplement Dr. Cooper's disclosures substantially justified.
Addressing each piece of challenged testimony is unnecessary, however, because the challenged opinions are largely unpersuasive, and not relied upon in the following analysis. For example, Dr. Cooper provided little support for his estimates of how many WAV vehicles would be needed successfully to implement UberWAV service in New Orleans and Jackson, or his opinion that the marginal cost to Uber of providing WAV service in New Orleans and Jackson could be zero. With a lack of support in the data, serious comparison to other cities, or other methods of analysis, many of Dr. Cooper's opinions verged on speculation. The motion to strike need not be reached, because as explained below, Plaintiffs’ evidence does not establish a violation of the ADA.
VI. The Existing UberWAV Program
Before delving into the merits of the case, it is helpful to describe the contours of the existing UberWAV program, as detailed at trial. Similar to Uber's standard UberX product, in which riders can request an on-demand ride via the Uber app, UberWAV allows riders to request an on-demand ride, but in a WAV rather than a typical passenger vehicle. UberWAV is one of a number of specialized products Uber currently or has previously offered; for example, Uber has operated an UberXL product in which riders could request a passenger vehicle with a greater number of seats, and UberBlack, through which a rider could request a luxury black car. Uber prices WAV rides at the same cost as a standard UberX ride.
UberWAV, however, is not offered in all markets in which Uber operates. As of the time of trial, UberWAV operated in eleven cities across the United States: New York, Boston, Philadelphia, San Francisco, Los Angeles, Phoenix, Houston, Austin, Portland, Chicago, and Washington, D.C. Niraj Patel testified that in the past three years, Uber spent between $40 and $60 million on its UberWAV program, including approximately $55 million in 2021. In some cities, like New York, operating a WAV option is required by local law or regulation. In other cities, operating a WAV option is not required, but regulations encourage Uber and its competitors to operate a WAV program, such as by imposing fines on Uber if it does not provide a WAV option.
In the cities where UberWAV operates, Uber has utilized four main methods to ensure WAV vehicles and drivers are available on its platform: (1) driver incentive programs; (2) partnerships with a commercial operator; (3) WAV rental programs; and (4) a dispatch program to pool WAV resources with other transportation providers. First, an incentive program provides bonuses to people who personally own WAVs to drive on the platform, via either a sign-up bonus or a per trip incentive. Uber has previously offered WAV incentives in Philadelphia, Portland, and Chicago.
Second, in a commercial operator partnership, Uber contracts with a commercial WAV operator on a per hour basis to gain access to both a WAV and a driver. Uber already contracts with a commercial operator called MV Transportation to provide WAV service in other cities, such as New York, Los Angeles, San Francisco, and Boston. The commercial operator partnerships have resulted in the provision of hundreds of thousands of rides. As Uber negotiates a specific number of hours of service in its contracts, Uber pays for the WAV and driver's time regardless of whether that driver completes any rides during the contracted time.
Third, in a rental or leasing model, prospective drivers may lease a WAV from either an Uber subsidiary or a third party rental company. Uber previously operated a WAV leasing model via its former subsidiary, Xchange Leasing, in Philadelphia. Uber also creates partnerships with third party rental companies, through which drivers may lease a car to use while driving on the Uber platform. In its current partnership with Avis, drivers can lease a WAV in Boston or Washington. Uber can incentivize the rental company to purchase WAVs, and can also subsidize the cost of WAV rental to encourage drivers to choose to rent a WAV.
Uber subsequently sold Xchange Leasing, following concerns that it was losing money.
Fourth, a dispatch program refers to the pooling of WAV resources between different transportation operators. In a dispatch model, a request for a WAV vehicle from any participating operator may be fulfilled via a vehicle from another operator. Such a program exists in New York City. In that program, Uber and its competitors Lyft and Via, along with taxi, Black Car, and limousine companies, pool their WAV resources. Uber does not participate in any dispatch programs other than in New York City.
Uber previously explored implementing UberWAV in New Orleans. At the time, city officials in New Orleans were considering an ordinance which would have required Uber to provide WAV service in the city. Uber lobbied against that ordinance, and dropped New Orleans from its WAV launch priority list immediately after the city declined to adopt the ordinance. Uber has not previously explored implementing WAV service in Jackson.
VII. Failure to Make Reasonable Modifications in Violation of 42 U.S.C. § 12184(b)(2)(A)
A. Legal Standard
Section 12182(b)(2)(A)(ii) requires covered entities "to make reasonable modifications in policies, practices, or procedures, when such modifications are necessary to afford such goods, services, facilities, privileges, advantages, or accommodations to individuals with disabilities[.]" 42 U.S.C. § 12182(b)(2)(A)(ii). An essential element of the reasonable modification claim is that a plaintiff has requested and been denied a modification from the defendant before filing suit. Namisnak v. Uber Techs, Inc. , No. 17-cv-06124-RS, 2018 WL 7200717, at *2-3 (N.D. Cal. 2018) ; accord Karczewski v. DCH Mission Valley LLC , 862 F.3d 1006, 1010 (9th Cir. 2017) ; Fortyune v. Am. Multi-Cinema, Inc. , 364 F.3d 1075, 1082 (9th Cir. 2004). A covered entity has an affirmative defense to that requirement if "making such modifications would fundamentally alter the nature of such goods, services, facilities, privileges, advantages, or accommodations." 42 U.S.C. § 12182(b)(2)(A)(ii).
As explained above, 42 U.S.C. § 12184(b)(2)(A), which addresses public transportation services operated by private entities, cross-references section 12182(b)(2)(A)(ii) for the definition of reasonable modifications.
B. Discussion
This Court previously ruled that Uber is a covered entity within the meaning of the ADA. See Summary Judgment Order, pp. 7-11. The parties do not dispute that Plaintiffs each submitted a request to Uber for it to provide UberWAV service in their cities. Plaintiffs’ reasonable modification claim thus turns on three questions: (1) whether Plaintiffs’ proposal that Uber provide WAV service is a modification within the meaning of the ADA; (2) if it is a modification, then whether that modification is reasonable; and (3) whether providing WAV service in New Orleans and Jackson would fundamentally alter Uber's business. The questions of whether Plaintiffs’ proposal is a cognizable modification and whether the modification would fundamentally alter Uber's business are addressed prior to discussing reasonableness.
1. Whether Plaintiffs Have Proposed a Cognizable Modification
Plaintiffs’ requested modification is for Uber to "turn on" the UberWAV option in New Orleans and Jackson. As a threshold matter, Uber argues that Plaintiffs have not requested a modification, and instead propose an "iterative process" to arrive at a desired metric. In making this argument, Defendant relies on a recent case from this district, Independent Living Resource Center v. Lyft , No. C 19-01438 WHA, 2021 WL 3910719 (N.D. Cal. Sept. 1, 2021) (" Lyft "), in which electric wheelchair users requested Lyft, a rideshare provider like Uber, to implement WAV service in two California counties. In Lyft , the wheelchair users requested that Lyft not only provide WAV service but also meet certain wait time benchmarks, which would then entitle Lyft to funds from the state government. The court explained that this request was "not a concrete proposal or modification" but rather "a performance standard" which "beg[ged] the question of how it could be done." Lyft , 2021 WL 3910719, at *11. The court concluded that plaintiff in that case "fail[ed] to show" how Lyft could meet the performance standard, "leaving it to Lyft and trial and error to figure it out." Id.
The Court previously rejected Uber's "outcome" vs. "modification" argument at summary judgment. See Summary Judgment Order, pp. 17-18.
Unlike in Lyft , there is not a specific performance standard that Uber must meet. Plaintiffs point to different mechanisms Uber could use to implement WAV service—the same mechanisms that Uber has used in other cities, such as leasing programs or commercial fleet partnerships. The existence of multiple paths Uber could take to implement WAV service does not make Plaintiffs’ request an "iterative, experimental, or trial-and-error proposal[.]" Id. at *9-10. After all, if a plaintiff's proposed modification specified a single mechanism to implement WAV service, but Uber could achieve the same WAV service using a cheaper mechanism, an injunction requiring the plaintiff's modification would prohibit Uber from using the more cost-effective mechanism, despite its ability to achieve the same on-the-ground result for the plaintiff. In other ADA contexts, an injunction does not need to specify the precise plan for implementing a modification. Indeed, a building owner required to install a ramp so wheelchair users may enter the building could choose between a variety of construction materials, designs, and contractors to build the ramp. Plaintiffs’ requested modification is indeed a large modification. But it is nevertheless a modification within the meaning of section 12182(b)(2)(A)(ii). The size and breadth of this modification bears on its reasonableness, not whether it is a modification.
2. Uber's Fundamental Alteration Defense
Uber's fundamental alteration defense also fails. Uber argues that Plaintiffs’ requested modification would fundamentally alter its business because (1) it does not offer UberWAV in Jackson or New Orleans, and (2) to implement the modification Uber would need to use a commercial fleet operator, and "[s]uch partnerships are a substantial deviation from Uber's business model[.]" Defendant's Post-Trial Brief, pp. 16-17. Concerning the first argument—that Uber does not currently offer UberWAV in Jackson and New Orleans—this argument simply states that because there is no UberWAV service, implementing it would fundamentally alter Uber's business. Providing no authority in support of this argument, Uber apparently contends that any requested modification of its services is a fundamental alteration, which would allow it to escape ADA liability for any requested modification. This argument is therefore rejected. As for Uber's argument that using commercial fleet partnerships is a fundamental alteration, Uber has used such partnerships in other cities. Even if Uber had not previously used that model, Plaintiffs’ proposed modification is still within the ambit of its business model: providing on-demand rides to people who request them via Uber's app. See Lyft , 2021 WL 3910719, at *8 ("WAV services would not fundamentally alter Lyft's business because WAVs are not a ‘new transportation service.’ Lyft already provides WAV services where compelled to do so by local governments and voluntarily in Los Angeles and San Francisco[.]"). In short, the proposed modification would not fundamentally alter Uber's business.
3. Whether the Modification is Reasonable
The ADA requires only the provision of "accommodations that are reasonable." Baughman v. Walt Disney World Co. , 685 F.3d 1131, 1135 (9th Cir. 2012). "[D]etermination of whether a particular modification is ‘reasonable’ involves a fact-specific, case-by-case inquiry that considers, among other factors, the effectiveness of the modification in light of the nature of the disability in question and the cost to the organization that would implement it." Fortyune , 364 F.3d at 1083 (citation omitted).
Although Plaintiffs have proposed a variety of ways Uber may implement WAV service in New Orleans and Jackson, these mechanisms come with varying advantages and drawbacks, and the parties have introduced evidence in favor of or against their likelihood of success. Thus, to understand whether the modification is reasonable, it is first necessary to evaluate the evidence on the different implementation mechanisms to determine which present viable possibilities.
Uber argues that the only reliable method for implementing WAV service in New Orleans and Jackson is via a commercial fleet partnership, in which Uber contracts with another company to have WAVs and drivers at the ready to respond to riders’ requests. According to estimates Uber obtained in the weeks leading up to trial from TowerWAV, a commercial WAV fleet operator, providing such service would cost $800,000 per year (approximately $400 per ride) in New Orleans and $550,000 per year (approximately $1,000 per ride) in Jackson.
These estimates are based on providing 300 supply hours of WAVs and drivers per week. A supply hour is a metric that represents an hour a vehicle is available to receive requests and complete rides. According to the TowerWAV quote, 300 supply hours per week would cost $52.21 per hour in both Jackson and New Orleans. The 300 supply hour quote is premised on having three WAVs and drivers on the road during operating hours. Fagent called 300 supply hours "a bare minimum[.]" Day Three Transcript, 395:10. He explained that "with just a few cars on the road, even if multiple requests came in at the same time, a trip could easily go unfulfilled or not completed[.]" Day Three Transcript, 395:14-16.
Plaintiffs argue that they presented evidence showing Uber could implement a WAV program in New Orleans and Jackson via any of the four implementation mechanisms Uber has used in other cities, or through a combination of any of the methods. More specifically, Plaintiffs argue that the non-commercial fleet partnership options—incentive programs, rental partnership, and a dispatch program—would be cheaper than a commercial fleet partnership. Plaintiffs also argue that the cost of a WAV program could be mitigated through the implementation of an accessibility fee on all rides, or through implementing cross-dispatching.
Plaintiffs insist that incentive programs—which could include sign-up bonuses, per trip bonuses, or waiving Uber's collection of a service fee—would encourage people with personally owned WAVs to drive on the platform. As Uber employee Niraj Patel testified, however, Uber's incentives work "when there is a sufficient existing base of both supply and customers to bring into [the Uber] marketplace," Day Two Transcript, 234:13-17, and further testified that "incentive programs haven't been very effective." Day Three Transcript, 343:1. Plaintiffs have provided scant evidence of a sufficient existing base of WAVs in New Orleans and Jackson. Dr. Crawford testified that in his home of Jackson, he is aware of seven or eight personal friends who own WAVs, along with a handful of independent operators who provided rides currently or in the past. In New Orleans, Plaintiffs point to Uber documents which show five WAVs were signed up on the Uber platform. There is also no indication that incentives alone would be effective at encouraging people to purchase WAVs. When asked whether "incentives [have] been effective at causing individuals to go out and buy WAVs and use them on Uber's platform?" Fagent answered "[n]o, they certainly have not been able to do that." Day Three Transcript, 402:7-10. In sum, there is not sufficient evidence that incentive programs for drivers with personally-owned WAVs would be effective.
Existing personally-owned WAVs may not even qualify for use on the platform, for example if they are an older vehicle model than is allowed on the platform. Plaintiffs have also not presented any evidence to establish that the ability to drive WAVs on Uber's platform, with incentives that would provide more compensation than for a typical UberX ride, would encourage people to purchase WAVs for use on the platform. For example, Dr. Crawford testified that in his experience, a new WAV would cost $45,000 to $60,000, and retrofitting an existing minivan would cost around $20,000. Day One Transcript, 51:3-52:3.
Similarly, Plaintiffs have not provided evidence that a rental or leasing model would lead to a significant number of WAVs on the platform. To start, Uber no longer owns Xchange Leasing, so a leasing model would require a partnership with an outside company, as is the case with Uber's current partnership with Avis in Washington and Boston. Uber employees testified about challenges with the Avis partnership. Rupp testified that Avis is "not particularly interested in expanding because they don't own these vehicles for their retail rental business, and the vehicles that they have offered in these two cities have not been very popular among drivers. In fact, they've had difficulty maintaining a profitable or satisfactory level of utilization on the vehicles." Day One Transcript, 175:17-22. Patel testified that Uber has not expanded the Avis partnership beyond Washington and Boston because "we have not been able to conclude that it's an effective method for getting WAVs onto the platform." Day Two Transcript, 275:11-13. He further elaborated that "[w]e haven't been able to conclude that there is sufficient driver demand. We haven't been able to conclude that if a driver does choose a WAV, that they will continue, because these are short -- these rentals can be returned." Day Two Transcript, 269:14-18. Without evidence that a rental partnership is succeeding in other markets or evidence supporting the idea that a partnership in New Orleans or Jackson would be more successful, there is insufficient evidence to support Plaintiffs’ theory that Uber could pursue a rental program to implement WAV service in New Orleans and Jackson.
As for a dispatch model, Plaintiffs have provided no evidence that such a model could work outside of New York City, where the dispatch model is required by local ordinance. In New York, companies such as Uber and Lyft, along with hundreds of other transportation companies, are required either to be an accessible vehicle dispatcher or affiliate with such a dispatcher. See N.Y.C. Rules, tit. 35, § 59B-17(c), (f). Plaintiffs have not provided evidence that a central dispatch scheme could succeed without the kind of regulatory requirement present in New York City. After all, Uber could not force any other company to participate in such a scheme with it.
Plaintiffs also presented evidence that the cost of a commercial fleet partnership could be mitigated in two ways: via implementation of an accessibility fee, or by cross dispatching. Starting with an accessibility fee, Plaintiffs rely on a 2017 internal Uber presentation that states a three-to-four cent accessibility fee could "fully fund" the UberWAV program. Defendants explain that this calculation was based on earlier WAV service projections, when Uber projected to spend $15.2 million per year on WAV, rather than the more than $50 million per year it actually spent in 2019 and 2020. In addition, this calculation is not specific to WAV programs in New Orleans or Jackson. Dr. Cooper, Plaintiffs’ expert, opined that a ten-cent fee would be sufficient to fund WAV efforts in New Orleans. Although his testimony was purportedly based on Uber's disclosure of quotes from TowerWAV, it is unclear how Dr. Cooper arrived at his opinion. Uber employee Fagent later testified that a ten-cent fee would not fully fund the WAV program, explaining that "looking at how many trips New Orleans completed last year, that was about 5 million trips; adding a 10-cent fee to the $5 million trips is only $500,000, when the cost to Uber was $800,000." Day Three Transcript, 398:24 – 399:2.
The broader issue with relying on an accessibility fee to counter any additional costs Uber may incur is the unforeseen ramifications on demand for Uber's services. Unlike the "fees" collected in jurisdictions required by regulation, Plaintiffs’ proposed accessibility fee would apply only to Uber, rather than equally across all competitors. Uber employee Niraj Patel testified that an analysis of how much money would be raised via an accessibility fee would need to consider "how much that would generate versus how much business we would lose." Day Two Transcript, 295:5-6. Plaintiffs have presented no evidence to demonstrate that an increase in price would not result in a decrease in business. That is not to say that an accessibility fee could not succeed and either cover, or greatly diminish, Uber's expenses in implementing a WAV program in Jackson or New Orleans. Yet, Plaintiffs have not presented reliable evidence that such a fee could diminish the cost of a WAV program without impacting revenue.
Finally, Plaintiffs argue that the cost of a WAV program could be diminished through cross-dispatching. Cross-dispatching refers to using UberWAV vehicles for other purposes, such as UberX rides or UberEats deliveries, during downtime between requests for UberWAV rides. When cross-dispatching is permitted, a WAV on the road would bring in more revenue for Uber, because its downtime would be partially filled with other trips. As Uber employee Connor Fagent testified, however, this increased revenue would come at the price of reliability, as a WAV would be more likely to be completing another ride elsewhere at the time of an UberWAV request. See Day Three Transcript, 386:15-19 (explaining that when cross-dispatching is enabled, a WAV drive "might be preoccupied quite a bit with those UberX requests and so during the times that they're on UberX requests, they're not on or eligible for WAV requests. So you're trading off maybe costs for not as much availability"). Thus, while cross-dispatching could reduce costs, it would also decrease the effectiveness of the service.
Based on the evidence before the Court, implementation of a successful WAV program in New Orleans and Jackson would likely require a commercial fleet partnership. Uber may be able to reduce the cost of such a partnership via combining this model with other options, like driver incentives, a leasing partner, or a dispatch model, but based on the evidence presented at trial, it does not appear likely that one of those models could succeed on its own. Uber may also be able to reduce its own costs via an accessibility fee or implementing cross-dispatching to use the time of WAV vehicles and drivers more efficiently. It is unclear how much money would be saved while maintaining desired levels of service—especially given the already limited nature of service from a commercial operator in terms of vehicles on the road and hours served. Thus, the reasonableness of the proposed modification is considered in light of Uber's likely need to implement a commercial vehicle partnership.
Uber argues that for a modification to be reasonable, it must only impose a de minimis cost, but this argument is unsupported by caselaw. Uber points to a variety of cases in which the cost of a modification was de minimis. See PGA Tour, Inc. v. Martin , 532 U.S. 661, 682, 121 S.Ct. 1879, 149 L.Ed.2d 904 (2001) (allowing professional golfer to use a golf cart rather than walking in certain tournaments); Fortyune , 364 F.3d at 1084 ; Doud v. Yellow Cab Co. of Reno , No. 3:13-CV-00664-MMD, 2014 WL 4302552, at *5 (D. Nev. Aug. 28, 2014) (explaining that taxi company's "practice of refusing to transport disabled individuals who use motorized scooters in standard taxis when their disassembled mobility equipment fits in the trunk of a standard taxi" likely violates the ADA); Anderson v. Franklin Institute , 185 F.Supp.3d 628, 652-53 (E.D. Pa. 2016) (requiring museum to waive general admission fee for personal care attendant, when such a waiver would not cost the museum more than $1 per waived admission). In all of these cases, however, the courts did not find that the modification was reasonable only because there was little or no cost. Cost is certainly a key consideration when evaluating the reasonableness of a modification, see Fortyune , 364 F.3d at 1083 ("An accommodation is not reasonable if it imposes undue financial and administrative burdens." (internal quotation marks and citation omitted)), but it is not the only one.
Plaintiffs point to a number of cases which arise in the context of transportation where a defendant was ordered to incur a significant financial burden to comply with the ADA. In Cupolo v. Bay Area Rapid Transit , 5 F.Supp.2d 1078 (N.D. Cal. 1997), Bay Area Rapid Transit, the light rail system serving San Francisco and the surrounding area, was required to spend significant resources on elevator maintenance to ensure compliance with the ADA. In American Council of the Blind of New York v. City of New York , 495 F.Supp.3d 211 (S.D.N.Y. 2020), the City of New York was required to install accessible signals at street intersections, at an estimated cost of $60,930 per intersection or $870 million total, per the City's estimates. Id. at 224, 241. The district court in Culvahouse v. City of LaPorte , 679 F.Supp.2d 931 (N.D. Ind. 2009) ordered an Indiana city to render its existing sidewalks accessible to wheelchairs, at an estimated cost of $5.8 million. Id. at 935.
Each of these cases involved an application of Title II of the ADA, rather than Title III, which is applicable here. Title II applies to state and local government entities, while Title III applies to privately-run public accommodations. These Title II cases, however, are nonetheless helpful comparisons because they involve transportation or the ability to move from one place to another. Access to transportation does not just allow access to that specific service, it facilitates engagement with the world at large. As the district court in Cupolo concluded, "[t]he difficulties class members have encountered with BART's elevators have [ ] interfered with the accomplishment of the ADA's policy of assuring equal opportunity, full participation, independent living, and economic self-sufficiency to individuals with disabilities." 5 F.Supp.2d at 1084. The opportunity to move from place to place is central to the idea of full participation in society, independent living, and self-sufficiency. Thus, when considering "the effectiveness of the modification in light of the nature of the disability in question[,]" Fortyune , 364 F.3d at 1083, creating reliable access to transportation when no such access was consistently available is a highly effective modification. Cases concerning access to transportation should not be viewed as analogous to cases involving access to amenities, such as museums, amusement parks, and movie theaters. That implementing UberWAV requires more than a de minimis cost does not automatically make the modification unreasonable.
When considering both the cost and the effectiveness of the proposal, however, Plaintiffs’ proposed modification is unreasonable. Based on the evidence Uber presented, a commercial operator model would cost $800,000 per year (approximately $400 per ride) in New Orleans and $550,000 per year (approximately $1,000 per ride) in Jackson. Notably, these costs would not even provide a level of service equal to the level of service available on the UberX platform. The estimates were for 16 hours of service on weekdays, and 10 hours of service on weekends. There would thus be large gaps of time in which no UberWAV service is available. Further, as described in testimony from Uber employee Fagent, having only three vehicles on the road at a given time would be the bare minimum to provide service, and even during the "online" hours some ride requests would go unfulfilled if there were more than three simultaneous requests.
In the Lyft case, the anticipated cost-per-WAV-ride was disputed, but the lowest estimate was $622. The court held that this amount was unreasonable, regardless of whether the court looked to Lyft's size, wealth, or level of resources, stating that "[e]ven a vast bottom line does not transform exorbitant modifications into reasonable ones." Lyft , 2021 WL 3910719, at *16. The cost-per-ride estimates here are similar to those in Lyft. Further, cost-per-ride is not the only metric to consider when evaluating reasonableness. Here, the estimated costs of providing WAV service anticipate a barebones WAV program that would likely require significant wait times, and service only at certain times of day. Courts must consider "the effectiveness of the modification" when assessing reasonableness. Fortyune , 364 F.3d at 1083. Although WAV service has the potential to transform lives through creating easy access to WAV rides, the high cost here would not even provide wheelchair users with the kind of 24/7 access UberX provides. The anticipated cost here is too high for the limited service that would result, making the proposed modification unreasonable.
It is very possible that the TowerWAV quote is inflated. Uber procured a quote from one company just weeks before trial. Uber did not request quotes from multiple companies or attempt to engage in negotiations to receive a lower price. Any idea of what that lower price would be, however, is speculation, in the absence of other estimates of a what a commercial operator would charge to provide service. Further, as explained in the discussion of each operation model, there is inadequate evidence that any of the other models would be viable in New Orleans and Jackson. Plaintiffs’ argument that the use of a commercial operator in tandem with another model would reduce costs is thus similarly speculative based on the evidence introduced at trial.
Even if Uber had not procured the quote from TowerWAV, the result in this case would be the same. Plaintiffs bear the burden of proving reasonableness, and part of that analysis is cost. For the reasons explained earlier, Plaintiffs’ evidence concerning minimization of cost, such as via accessibility fees, is not persuasive. To succeed, Plaintiffs needed to provide some kind of non-speculatory evidence showing the cost to implement WAV service would not be beyond the bounds of reasonableness. Plaintiffs have failed to do so.
This decision should not, however, be read to adopt Uber's argument that any cost must be de minimis , nor that a lawsuit demanding a rideshare company implement WAV service can never succeed. Uber has placed itself in the business of transportation, and transportation implicates each person's "full participation, independent living, and economic self-sufficiency" in society, including people with disabilities. Cupolo , 5 F.Supp.2d at 1084. A request for modification so that a person may access all parts of his or her hometown must be viewed differently than a request to access a single amenity. Cost is a factor in assessing reasonableness, however, and a cost of $800,000 per year in New Orleans and $550,000 per year in Jackson for service that is markedly different than UberX service—most notably, in that there are large spans of days in which no coverage is provided—is not reasonable.
One final note is necessary. The first argument section of Uber's Post-Trial Brief is entitled "Uber Has Done More Than Its Fair Share on WAV And Should Not Be Punished For It." Defendant's Post-Trial Brief, p. 2. Defendant argues that "Plaintiffs seek to weaponize Uber's positive steps and efforts against Uber, by forcing the company to do that work—and suffer even greater financial losses—in New Orleans and Jackson." Id. While as explained above, the modification would not be reasonable, and no injunction requiring Uber to provide WAV service in the applicable cities is warranted, such an injunction would not constitute "punishment" as characterized by Uber. Complying with the ADA and providing access to people with disabilities is not a punishment, it is the law. Further, the argument that Uber has done its "fair share" in providing WAV access in other cities mischaracterizes the purpose and design of the ADA. Although Uber is correct that just because it has taken an action in one city does not mean it is obligated to take that action in another city, Lyft , 2021 WL 3910719, at *17, the language "fair share" implies that WAV users are due a finite number of resources. The ADA does not adopt such an approach, however, and evaluates each proposed modification for its reasonableness. WAV service must be provided whenever a plaintiff meets the burden of proving an ADA violation, regardless of whether the defendant already provides more or less WAV service than others.
In short, there has been insufficient evidence presented in this particular case to demonstrate that the proposed modification is reasonable. The claim for failure to make reasonable modifications in violation of 42 U.S.C. § 12184(b)(2)(A) therefore fails.
VIII. Screening Out People With Disabilities in Violation of 42 U.S.C. § 12184(b)(1)
A. Legal Standard
Section 12184(b)(1) prohibits discrimination through the "imposition or application ... of eligibility criteria that screen out or tend to screen out an individual with a disability or any class of individuals with disabilities from fully enjoying the specified public transportation services provided by the entity[.]" 42 U.S.C. § 12184(b)(1). An entity has an affirmative defense to the requirement if "such criteria can be shown to be necessary for the provision of the services being offered[.]" Id.
Section 12184(a) also requires that the defendant be "a private entity that is primarily engaged in the business of transporting people and whose operations affect commerce." This Court previously ruled that Uber is a covered entity under the statute and that its operations affect commerce. See Summary Judgment Order, at pp. 7-11.
B. Discussion
Plaintiffs argue that Uber's vehicle criteria screens out WAVs by (1) banning vans and (2) prohibiting aftermarket seating modifications. Most WAVs are minivans or vans which receive aftermarket modifications to remove seats and thus create space for an electric wheelchair. In both New Orleans and Jackson, Uber's listed vehicle requirements prohibited vans and vehicles with aftermarket seating alterations, amongst other restrictions.
Plaintiffs argue that the ban on vans includes both vans and minivans—and thus encompasses almost all WAVs. This argument is unconvincing. Plaintiffs have not shown that the term "van" is synonymous with minivans. Further, Uber presented evidence that minivans were shown on its lists of accepted vehicles, which prospective drivers could review to determine whether their vehicle would meet Uber's requirements. Thus, the ban on vans did not alone screen out most WAVs.
The ban on aftermarket seating modifications, however, would effectively screen WAVs from the platform. Witnesses testified that WAVs are frequently made by removing seats from a minivan, and thus undergo aftermarket seating alterations. Uber argues its ban on aftermarket seating modifications only concerns the addition of new seats. Yet, nowhere in Uber's policy does it state that "aftermarket seating modifications" only refers to "aftermarket seating additions."
Uber argues that had a WAV owner inquired about driving for Uber, it would have allowed that person to drive for the platform. That argument, however, stands in contrast to testimony from Uber employee Robert Rupp, who stated that Uber did not permit any exceptions to its requirements. Rupp agreed with Plaintiffs’ statement that Uber's "requirements apply to vehicles such that if the vehicle doesn't meet the requirements, that vehicle can't be used in Uber's products[.]" Day One Transcript, 151:15-18. Rupp also agreed that Uber implements and enforces the listed vehicle requirement.
Uber contends that even if it was screening out WAVs from its platform, enforcement of its criteria is necessary to maintain its insurance contracts. While maintenance of insurance policies may be important to Uber's business, Uber has nonetheless failed to carry its burden of showing it cannot maintain those policies while allowing WAVs that have undergone aftermarket seating modifications by licensed technicians and according to safety standards to operate on its platform. Indeed, WAVs operating on UberWAV in cities where the platform exists must have certification that the aftermarket modifications adhere to safety guidelines. Further, Uber also argues that it is necessary to have standard criteria—such as the availability of four passenger seats—such that any vehicle can accommodate four passengers. Uber, however, would not have to abandon this criteria to accept WAVs onto its platform; instead, it could still prohibit WAVs that did not contain the designated number of passenger seats.
Specifically, a driver seeking to use a WAV on the UberWAV platform would need to "show proof that it was modified by a [National Mobility Equipment Dealers Association]-certified organization[.]" Day Three Transcript, 411:24-25.
The failure to allow WAVs onto the existing UberX platform, however, does not alone "screen out" people who use electric wheelchairs. As explained above, when a customer requests a ride on the UberX platform, any driver who is approved to drive on that platform could be matched to provide a ride, regardless of whether that driver operates a WAV. Explicitly allowing WAVs on the UberX platform, when those vehicles otherwise meet the UberX requirements, would provide no guarantee of WAV service for Plaintiffs. Further, the scant evidence of how many WAVs are currently operated in New Orleans and Jackson means that Plaintiffs have not even established a possible likelihood that a WAV would be matched to their request by chance. Plaintiffs have presented evidence of less than 10 WAVs in operation in each city. In effect, the lack of WAVs in operation would screen Plaintiffs out of using the platform, rather than the policy which does not allow WAVs. Thus, based on the evidence, the "eligibility criteria" does not itself screen out people with disabilities in violation of § 12184(b)(1).
IX. Conclusion
Plaintiffs have failed to present adequate evidence that Uber has violated 42 U.S.C. §§ 12184(b)(1) and 12184(b)(2). Judgment will therefore be entered in favor of Defendants.