Smartphones vs. Taxi Drivers

Regulators and entrenched interests scramble to cope with the e-hailing revolution.


Raido Kaldma / photo on flickr

Barry Korengold, president of the San Francisco Cab Drivers Association, has been driving for hire around the Bay Area for 35 years. These days, he's none too happy.

Korengold owns a coveted city-issued medallion; it's the only way to legally drive a metered taxi around Baghdad by the Bay. If you don't have your own, you need to work for someone who does. This license to cab is currently available at the jaw-dropping price of $250,000. For years, that barrier to entry produced enough scarcity to make Korengold and his grizzled brethren a solid middle-class living.

But over the past year or so, revenues have plunged a Lombard Street-like 50 percent, Korengold and other local drivers tell me. Press reports suggest the cabbies are exaggerating, but only a bit. What happened?

New, unmedallioned drivers dominate the streets of San Francisco, Korengold complains, "competing with us for the same customers, providing the same service but allowed to play under different rules-or no rules. They undercut us because they have hundreds of millions in venture capital behind them."

In other words: It's all Danetta's fault.

Danetta (who wished only to use her first name, in order to speak more freely without offending potential customers), is a 32-year-old bookkeeper who for the last year and a half has been driving passengers around the Bay Area and later San Diego for both UberX and Sidecar, two entrants in the hot new business model of e-hailing, or providing rides via smartphone. (See page 23 for how these companies work.) Danetta also drove a while last year for a third market leader, Lyft, a self-consciously wacky brand whose drivers frequently strap whimsical giant pink moustaches to the front of their cars. "We had candy and we had water and every ride we treated like our friends," she says.

It's pretty easy to meet the basic requirements for being an e-hail driver: You must have a post-2004 model car for UberX (post-2000 for Lyft) and be able to pass a background check on both your driving and criminal records. And the pay is good-Danetta takes home 80 percent of her Sidecar fares, and when I interviewed her in March she was getting 95 percent from UberX, though the latter goes through periods of being extra generous to lure more drivers. And the work is pleasant: After more than 4,000 rides, she remembers only five even slightly negative experiences.

Medallioned taxi drivers in the Bay Area don't have it so easy. As Korengold explains, "The cost of getting a driver's permit in San Francisco is around $350, which includes a fingerprinted background check, taxi school, taxi permit, and DMV printout. To get a taxi medallion you must be next in line on the waiting list and have put in the driving requirement of 800 hours a year during at least four out of the last five years." And that's before you hand over your quarter-million dollars to the city.

Cabs in San Francisco these days are required to either be hybrids or to use an alternative fuel, and they have to be equipped with cameras. If you're renting a vehicle from a fleet, you often pay up to $100 a shift in "gate fees," in addition to having to tip dispatchers in order to get good calls-a practice that is pervasive, despite being technically illegal.

Korengold and Danetta now compete on a playing field that the state of California in September 2013 declared level, after months of bitter political wrangling. Taxi drivers who have long enjoyed the perks of protectionism have not taken well to the new reality. (Students of American transportation history might feel some deja vu. When the practice of individual cars carrying passengers for hire at all began in the 1910s, the electric street car industry condemned them as, among other things, "a menace," "a malignant growth," and "this Frankenstein of transportation.")

Cabbies routinely photograph e-hailers' license plates, publish the results on the Internet, and in some cases report drivers to their insurance companies hoping to get policies cancelled on grounds of operating personal cars for commercial purposes. A glimpse of Lyft's fuzzy pink moustache causes cabbies to go apoplectic, sometimes swerving to cut off what many of them view as scabs. The streets of San Francisco are filled with bad juju.

But so far, the e-hailers seem to be winning the livery war. "The California framework is a great example for a regulatory framework that holds safety as a top priority, yet allows for consumer choice," says Erin Simpson, a spokeswoman for Lyft.

In June, Colorado became the second state to impose statewide regulations on this 21st century business model. But for every normalization, multiple fronts of an ongoing war continue to rage. The e-hailers are besieged by city councils trying to ban them, by "anarchist" activists blocking their cars, by state transportation boards who find them confusing, and sometimes even by their own sticker-shocked customers.

Through all the hubbub, Uber in particular remains a tech darling, with a fresh $1.2 billion investment in the second quarter of 2014 and a current estimated market value of $18 billion. (Uber has different levels of service, including limos and actual taxis, but for the purposes of this article we are discussing UberX, whose drivers are private citizens using their personal vehicles.) The company's controversial founder, Travis Kalanick, spouts big plans for fleets of driverless cars. His vision: to eventually become the one-stop shop for anyone who needs to get any physical object to any physical location, period. Lyft, lagging far behind in total investment value, still was able to raise a quarter billion in the second quarter of 2014.

If venture capital firms are hot on the future of these companies, they have an unlikely bureaucracy to thank: the California Public Utilities Commission (CPUC). For once, the Golden State has offered an instructive example to regulators across the country of how to encourage tech-driven innovation in an overregulated industry.

Fist Bumps vs. Medallions

California's first e-hailing services appeared in summer 2012. By December, the CPUC had heard enough complaints about them to initiate a rulemaking proceeding on the nascent industry. (While taxis in California are regulated at the local level, CPUC sets the rules governing limousines and other non-metered, non-street-hail "charter party carriers.")

A multi-month process of filings and meetings commenced, with the three major ride services (and some smaller ones) on one side and pretty much everyone with a lobbyist on the other, including the Center for Accessible Technology, the International Association of Transportation Regulators, the San Francisco Cab Drivers Association, the San Francisco Municipal Transportation Agency, the San Francisco Airport Commission, the Taxicab Paratransit Association of California, and the United Taxicab Workers. Thousands of pages of filings, counterfilings, replies, and responses were generated, but the line of argument was ultimately not that complicated.

The new ride services insisted as an opening gambit that since they did not technically own any cars or even employ any drivers-every operator is an independent contractor-they were not liable to CPUC control at all. They were merely tech businesses that facilitated communication between drivers and passengers, not transportation companies. You wouldn't conflate Priceline with an actual hotel or subject Open Table to restaurant health codes, would you?

The taxi interests scoffed. The companies that made the hailing apps set their brand's prices, collected the money, and remitted the driver's cut. Whether they own a fleet of vehicles or not, Team Cabbie argued, the likes of Uber and Lyft run transportation companies.

The Taxicab Paratransit Association of California charged that e-hail services "have merely added a simple technological component to 'bandit cab' operations that have plagued municipalities for many decades." If the CPUC didn't regulate these new entrants with the same level of picayune detail as municipalities did taxis, the organization said, then all cab services would switch to this new model and abandon existing local restrictions on the number of hired cars on California roads. While the taxi interests suggested that this would be a terrible outcome, potential customers might not agree.

Cab companies argued that since their work is dangerous, all e-hailers needed to be outfitted with the same bulletproof safety partitions and cameras as taxis in municipalities where such specifications are required. But this ignores a fundamental attribute of e-hailing apps: Driver and rider know each other by both name and reputation.

Operators and customers of Uber and Lyft are pre-identified and crowd-rated. The driver sees who his passenger is and how other drivers have judged him; the passenger sees who the driver is, what car she is driving, and how fellow passengers have judged her. This is a vital reason why people, especially the techy/hip/bourgeois crowd that quickly made these services a national sensation, prefer e-hails to taxis. They feel instantly part of something more than a hired service. They are members of a community, communicating and exchanging value. Different services' communities have their own qualities. Uber's vibe is more high-end and perhaps a bit pretentious; Lyft more down-to-earth and friendly. (These distinctions, important at the start, seem to be fading in California as more people drive for both companies.)

Uber, Lyft, and their rivals offer convenience, certainly. But they also offer an appealing brand identity. Cabs are rapidly becoming obsolete among a certain class of educated urbanites, even in neighborhoods where standing on a street corner and sticking out a hand is likely to yield a traditional yellow cab in short order.

The cab companies' inability to grasp this dimension has bordered on the comical: "Taxi companies would never condone physical contact such as the 'obligatory fist bump' of Lyft," the San Francisco-based Luxor harrumphed in a filing to the CPUC, "because it encourages a degree of intimacy that is inappropriate in commercial for-hire transportation."

But that intimacy also puts drivers on their best behavior to a degree unimaginable in yellow cabs. Danetta says she was cut off from Lyft's app because her aggregate driver rating fell below 4.7 (out of 5) during what she insists were "a couple of bad days." The rating system, as many Lyft drivers told me, is unforgiving. Another Bay Area Lyft driver says "riders should understand that if the experience was anything better than awful, they should give five stars." Just a few less-than-perfect ratings greatly increase the chance of being booted by the company or rejected by riders.

In September 2013, after 10 months of testimony and lobbying, the CPUC carved out a new regulatory classification for e-hail services: "Transportation Network Carriers." That is, neither taxis nor limos. All TNCs (the companies, not the individuals who drive for them) are required to register with the state and to impose a new set of demands on drivers, including full criminal and automotive background checks and a 19-point mechanical inspection of their cars. Taxi companies wanted to force the companies to establish minimum fares and to wait a mandatory 30 to 60 minutes from the time of a summons to pick up passengers. The CPUC thankfully declined to do anything that stupid.

There is still some regulatory ambiguity when it comes to two of the biggest sticking points in the proceedings, disabled access and insurance. While the TNCs cannot now guarantee that users in, say, wheelchairs will be able to comfortably ride in all of their driver's cars, they are required to keep records and to show an ongoing plan of improvement for disabled use of both the apps and the vehicles in service.

Prior to the CPUC regulations, insurance for e-hail drivers existed in a troublesome gray area. Nearly all personal auto insurance policies contain exclusions for people driving for hire; distinct commercial policies are usually required. Among Lyft and Uber drivers it was well understood that you should keep your for-profit driving discreet.

While the CPUC reviewed and approved supplemental insurance of up to a million dollars per incident that the e-hail companies keep for their drivers, cabbies were suspicious when their competitors refused to let them examine the policies. Representatives of both Lyft and Uber assured me they have indeed successfully used their new insurance, though they declined to identify specific cases or stories.

As of mid-March, both Uber and Lyft had rolled out new blanket insurance policies to cover drivers actively on call for the apps even if they do not have a rider at the time. When not on call, the assumption is that the drivers' personal policies, which California law insists all drivers have, should cover them. (Cabbies have suggested using the car professionally at all will give insurance companies an excuse to never pay or to deny coverage.) Nicole Mahrt Ganley, a spokeswoman for the Association of California Insurance Companies, told me in July that a standard personal policy ought to cover even e-hail drivers, as long as they aren't driving paying passengers and don't have the app open soliciting rides for hire.

The California legislature in August passed regulations tougher than the ones the CPUC adopted. A bill from Assemblywoman Susan Bonilla (D-Contra Costa) passed with no opposing votes in the state House in May; the bill passed the Senate in late August, as this story was going to press. Her legislation requires the TNCs to cover their drivers during the entire time the drivers have their apps open, and at higher levels than UberX and Lyft began offering in March. At a June press conference, Bonilla hosted the parents of a young girl, Sophia Liu, who became the symbol of e-hailing chaos after an Uber driver-not carrying a passenger but with the app on-struck and killed her in an intersection in San Francisco on New Year's Eve. This was prior to the new insurance policy, and Uber refused responsibility. A suit against both Uber and the driver is in process.

Meanwhile, the Taxicab Paratransit Association of California has a suit in process in the Supreme Court of California, claiming that the CPUC regulations should be overturned since the body didn't conduct a report on the environmental impact of these new cars-for-hire.

Uber spokeswoman Eva Behrend told the Los Angeles Times in June that Bonilla's and similar bills are "thinly veiled attempts to end ride-sharing in the state. This legislation is not about safety or consumers; it's about protecting entrenched Sacramento special interests from competition." Uber sent mailers to Bonilla's district accusing her of caring more about insurers than consumers. However, at the last minute, Uber and Lyft both decided they could live with Bonilla's bill, after the amount of excess insurance coverage the companies would have to provide when the app was on but no ride has been picked up was lowered to $200,000.

The Death of Rideshare

Prior to their formal classification as TNCs, Lyft and Sidecar liked to use the term rideshare. It has a touchy-feely, progressive aura about it. Similar businesses are often lumped together as the sharing economy, the peer-to-peer transaction model that has similarly upended paid lodging through Airbnb, grocery shopping with Instacart, and clothing consignment via Poshmark. That kind of socially conscious overlay can lead a certain sort of economically progressive young urbanite to come down on the side of a bunch of for-profit companies.

Sidecar and Lyft used to crow that they weren't technically engaged in market transactions; they were friends giving other friends a ride, even if these friends met two minutes ago via a smartphone app. But by the end of last November, both Sidecar and Lyft had announced their shift in California from a donation model to one where-like market leader UberX-you were fully committed to paying for your ride. Sidecar now uses a pay model across the country, and Lyft has a pay model in most of the cities where it operates. Some cynical industry watchers think it's no coincidence that as soon as California unequivocally said the business model is legal, the services who used to glory in their groovy communitarianism abandoned that pose and acknowledged that, yes, they sell rides to people.

Some cities, such as Portland and New Orleans, have blocked e-hailers entirely. But in June Colorado became the first state to regulate-and normalize-the ride services statewide, with legislation that passed both the state's House and Senate with mere single-digit opposition.

The regulations in Colorado are similar to those in California. The driving force behind the new legislation, Rep. Dan Pabon (D-Denver), noted that taxi drivers in Denver needed to make $35,000 a year just to break even from their typical medallion-leasing costs. If the new rules mean there will be fewer taxi drivers on the road in Colorado, he says, then so be it. "I don't think the Colorado legislature is in charge of fashioning categories of where people should work and how much they should make," Pabon explains. "Let the free market operate and make those decisions."

Pabon says the politics of the bill were pretty straightforward. Insurance companies, unions, and bankers demanded an insurance solution, which the legislation contains. But "to the extent that consumers were savvy enough to understand the issue, I had lots of thank yous on social networks. Equally fervent in support were the technology companies, even folks like Google and Facebook, not in this particular space but recognizing that technology and entrepreneurship won the day and that's a good thing for them."

I joined Lyft in Los Angeles to see what the service is like. My first driver, who didn't want to be identified by name, told me he values the gig for how it fits with the rest of his life. He's an entertainer, and if he suddenly has a show booked it's great to have a job he can freely walk away from, then freely return to. Like other drivers I spoke to, he likes the lack of micromanagement and the ability to choose his own hours and level of effort.

My subsequent Lyft drivers included screenwriters, actors, newcomers to town still feeling their way around, and a couple of former cabbies. Most of my more recent drivers also take gigs from both UberX and Lyft. Both companies indulge in regular promotions trying to win drivers over from the other by offering signing bonuses and special deals for turncoats. In San Francisco, they offer free lunches. The Wall Street Journal reported in August that Uber was offering a $500 bounty to drivers who poached colleagues from Lyft, and Uber will sometimes indulge in dirty tricks such as calling and then cancelling rides to make Lyft driving aggravating.

My first driver found Lyft's communication with its drivers a bit inadequate and erratic, but over the course of a couple of months he managed to apply, go through Lyft's "mentoring," and get on the road. (By his recollection, Lyft was not quite as thorough in the mechanical inspection as the new regulations demand.) He's doing four or five shifts a week and generally makes $20 to $25 an hour.

When e-hails run smoothly for the passenger-which it usually seems to do, and always did for me-it can be an example of mutually beneficial tech-facilitated interaction at its finest. But a bit of the law of the jungle still applies to drivers. Despite their frequent attempts to socialize together in the real and virtual world, on the road every Lyft driver is a rival, every pink moustache a potential fare lost.

My driver showed me how the app works from his end, with the ominous shadows of other Lyft drivers floating around the roughly 10-block square area of Hollywood he was examining. As we sat and talked after our six-mile round-trip drive-it cost $19 pre-tip-some of those drivers lurched jerkily closer to him on his app while others slid further away. The more drivers are crammed into an area, the less likely enough rides will be requested in the vicinity to feed all the hungry drivers. "Sometimes I'll see a moustache behind me in the rear-view mirror and I'm, like, why are you staying so close to me? It's better for all of us if we split up a little." (Over time, I've noticed that fewer Lyft drivers in Los Angeles seem to be sporting those moustaches.)

Cabs and limos, which used to have the field to themselves, now compete with e-hail services. The e-hail services in turn compete with one another, and within those services the drivers face off against each other too. Free markets create a nerve-wracking world for everyone but the customer.

Taxis in the Real World

If you believe the legal filings of California cab companies, taxis are consistently clean and sweet-smelling, their cars and meters function reliably, and all the drivers are above average.

This triumphalism ignores lived reality. Uber, Lyft, and Sidecar drivers told me at length how much their customers prefer them to cabs-the quality of the vehicle, the convenience, the speed of arrival, the friendliness. As one Lyft driver put it, "the stories I hear of bad experiences with SF cabbies-they are not nice, demand cash, don't show up-honestly I feel they created this situation themselves, they created a need for a new system." And it's not just the e-hailers who say things like that.

Consider a January 2013 report from the Center for Investigative Reporting's The Bay Citizen that studied more than 1,700 citizen complaints about San Francisco taxis over a one-year period. It found that "San Francisco taxi drivers routinely flout the law by refusing rides, declining to take credit cards, charging unauthorized fees, speeding, smoking, and talking and texting on cellphones while driving.…Taxis infested with bed bugs, drivers falling asleep at the wheel, rude behavior and difficulty getting a cab also were among the complaints. One patron reported that a cab driver allegedly stole his credit card number and used it to make purchases in Brazil. And two friends were upset when a driver offered them a 10 percent discount if they made out in front of him."

Those officially filed complaints merely scratch the surface. No one I interviewed had ever bothered to file a complaint about a bad cab experience. They just switched to e-hailing and never looked back.

The problems with taxis can be systematic, which is why the San Francisco specialty rideshare service Homobile arose-years before-to get around the discrimination and even violence faced by openly gay, transgendered, and transvestite taxi passengers, even in San Francisco. And contrary to some of the lawsuits filed in California, many disabled people report that e-hailers have been a life-saver, since no longer do they need worry about a taxi passing them by to avoid the hassle of slow passengers.

Airports, always a big moneymaker for hired drivers, are still off-limits to e-hail services in California. After a year of steady scofflawing by e-hail drivers, and hundreds of citations to UberX drivers in the prior month, UberX in January officially announced it was no longer picking up at Los Angeles International. Prior to the new TNC classification, an UberX driver told me the company promised to pay any fines they accrued for illegally delivering passengers to San Francisco International Airport. Still, he said, "we take precautions, take our phone out of the cradle and hide it. If you can get the passenger to sit up front, do so. Ask them to give me a hug and pretend we're friends. I joke with my passengers I feel like a coyote taking illegal aliens across the border."

A Bay Area Lyft driver once told me of being caught, but getting off with a warning, when an airport snoop noticed the pink moustache stuffed in her trunk as her passenger took out his luggage. "They told me what they look for-for people in back seats, for fist bumps, for phone holders on the dash," she says. A spokesman for SFO told me in March that the airport is considering integrating TNCs into its transportation services, but hasn't yet. Neither has any other California airport. In June, the CPUC sent a harsh letter to the TNCs reminding them that they were actively breaking the law by serving airports and that they had better stop.

A Nation of Ubers

When it comes to transportation regulation, there is no capital to conquer. In the city of Washington, D.C., the e-hail services won a brutal political showdown in December 2012 and now operate freely. But each locality remains its own battlefield.

So it's a bruising fight across the nation, with some victories and some defeats for customers. As of late summer, Uber is operating in 88 American cities, Lyft in 64, and Sidecar in 10. One independent researcher, Nate Good, who studied Philadelphia data, is sure he's already seeing signs that the ready availability of easy and convenient and quick hired transportation is reducing drunk driving. Good found an 11 percent drop over the course of 2013 after e-hailing services began operating there-mostly among the under-30 crowd, the likeliest users of smartphone apps. Uber followed up with its own study in July indicating a disproportionate number of Uber rides being summoned to bars during what are likely prime drunk-driving hours, near the close of bar times.

It doesn't matter how many people love it, how many people are making a living from it, or even how many lives might be saved. These technologies disrupt existing business and regulatory interests, so the fight rages on. Arizona and Washington are both contemplating following a Colorado model of manageable statewide regulations. Virginia imposed and then removed a statewide cease-and-desist order. Pittsburgh told Lyft and Uber they could no longer legally operate and began citing drivers through the summer. Both Lyft and Uber continued operating and vowed to help their drivers with legal expenses. Memphis announced a task force to actively seek out and punish e-hail drivers.

Famously liberal Austin has never allowed e-hailing to legally operate. In March, Sidecar sued the Texas town, echoing the claims made in the CPUC proceedings: that it is a tech company, not a transportation company, and thus doesn't fall under the transportation department's jurisdiction. Meanwhile, other e-hailers found ways to keep their brands in the minds of Austin's technorati even if the actual service can't legally function: During the annual South by Southwest festival in March, UberX ran branded pedicab services, and Lyft had mustachioed Lyfters literally offering piggyback rides.

In May, Lyft went ahead and entered the Austin market anyway. Within days, the city began citing and impounding the cars of its drivers. Lyft spokeswoman Katie Dally told the TV station KXAN in June that the company would handle drivers' legal fees and that "Lyft firmly believes our platform is operating legally, as Austin's transportation code does not contemplate our peer-to-peer model." In June, Uber followed Lyft into the market. It's a bracing example of civil disobedience in action: At least 15,000 Austinites have downloaded the Lyft app since the service entered the city, even though every transaction is technically illegal. August saw the highest number of citations of e-hail drivers in Austin-29-of any month since they entered the market.

Whether any of these companies are profit powerhouses yet is unclear. Both Lyft and Uber have been systematically cutting their cut of fares, for long periods to zero, in order to win driver loyalty. Both appear to be playing the tech-innovator game of fighting for market share and awareness first, profits later. And they do command politically engaged consumer bases of well-to-do tech savvy people, as well as the support of an impassioned body of national activists who want to promote the sharing economy.

E-hail companies are becoming a political force to be reckoned with. In Colorado, the leading sponsors of the new law told me they were brought into the process by e-hailing companies eager to be regulated in a sensible way. Both Uber and Lyft claim to be satisfied with the regulations they have helped impose on themselves. "Colorado has sent a clear message that it embraces innovation, supports consumer choice and empowers small business owners," Uber crowed on its website.

But the companies don't always play politics politely. In a March conference call, discussing Seattle's then-ongoing attempt to restrict any given e-hailing company to just 150 drivers on the road at once, Uber's combative CEO Travis Kalanick said that Seattle needed to realize that "Uber makes a better quality of living, it makes a city a better place." The presence of e-hailers, he continued, "is good for cities. I'm going to talk to every regulator, city council, mayor out there, also with governors and legislators. We will not compromise our principles against those who are trying to limit supply, limit choice, make cities harder to get around, just so a few individuals can run a government-sanctioned monopoly." (Seattle ended up creating a California-like TNC classification for the services, with no caps on how many drivers could be on the road.)

Still, even Kalanick can yield to public pressure. UberX used to impose "surge pricing" during rush hours, storms, or other situations in which drivers might be less inclined to bother dragging themselves out to find fares. Without such pricing, shortages of available drivers would become inevitable. But people freaked out when their surge-priced rides cost more than they thought was reasonable. Kalanick-who for a while used a Twitter avatar image from the cover of Ayn Rand's The Fountainhead-mocked the outrage with a sarcastic tweet reacting to an airplane flight that cost different amounts on different times and days: "OMG so f'in EVIL OMG OMG I can't believe this shit, so horrible holy shit." But in the face of possibly running afoul of New York state's price-gouging laws, Kalanick agreed in July to eliminate surge pricing during emergencies, as the state put it, "resulting from stress of weather, convulsion of nature, failure or shortage of electric power or other source of energy, strike, civil disorder, war, military action, national or local emergency, or other cause of an abnormal disruption of the market which results in the declaration of a state of emergency by the governor." That's good PR, but it may be bad news for any citizen of New York who can't find a hired ride during the next emergency.

Each day, e-hails face the rigor of peer-to-peer judgment and review, open for anyone to see. It's the difference between trust in government and trust in a decentralized, information-rich network of consumer choice and power. Customers have more variety and choice, drivers like Danetta can create schedules flexible enough to meet her needs, while incumbent stalwarts like Barry Korengold see their longtime business models evaporate overnight.

The taxi companies' counterarguments against this seemingly unstoppable new way of hiring rides basically boil down to this: Don't pay any attention to the facts about your experience with taxis. We are regulated and cartelized, so we must be safe, pleasant, fair, and affordable. As the United Taxi Workers rhetorically asked in one of its California filings: "Does the CPUC wish to be the agent responsible for the degradation and possible collapse of the taxi industry in San Francisco and elsewhere in the State of California?" Some questions you shouldn't ask, if you are afraid of what the answer might be.