Can Uber Ever Deliver? Part Five: Addressing Reader Comments and Questions

Yves here. We were glad to see Hubert Horan’s series on Uber get so much reader commentary, including from some Uber drivers. Hubert provides his response to questions and observations below.

By Hubert Horan, who has 40 years of experience in the management and regulation of transportation companies (primarily airlines). Horan has no financial links with any urban car service industry competitors, investors or regulators, or any firms that work on behalf of industry participants

Today’s post is my opportunity to thank the Naked Capitalism readers who not only took the time to wade through four detailed posts about the economics of Uber, but took the extra effort to offer thoughtful questions and comments.

Best comment of them all:

<Charles Myers>—Thank You, I sent money

Yves and Lambert have created a space where these types of financial/political issues can be explored and debated. It doesn’t exist without their tireless effort, but it also doesn’t exist without money. I’ve been a financial supporter for years, and one of the reasons for contributing these articles was that it was a way to thank them for everything they’ve done over the years. I hope everyone who appreciates what Yves and Lambert have created will respond as Charles did.

Nonsensical but frequently repeated claims about the “sharing economy”

<Rusti> Most cars are parked something like 98% of the time, representing a huge lost opportunity cost.

<Matt> suggesting demand for car use is also not there for 98% of the time

It would take a book to document all of reasons why “sharing economy” claims were always ludicrous. The most important is whatever limited opportunity might exist for utilizing idle consumer goods, it is not something that could ever be scaled into a global business serving mass market demand. In order to efficient serve mass market demand, real world transportation companies require vehicles that can be intensively utilized 7 days a week. The car in your driveway cannot do that. It might be able to handle a couple taxi rides on Saturday night, but Uber (or any other taxi operator) needs a huge fleet of dedicated vehicles and drivers to serve the other 98% of demand.

Airbnb always had slightly greater opportunity to use truly spare private capacity, because there always was a portion of housing stock in big cities that sat empty for significant periods. But if Airbnb limited itself to this spare capacity, its growth would have ended years ago. To justify big corporate valuations, it had to expand into selling apartments and houses that were totally dedicated to hotel-like occupancy, while trying to maintain the fiction that its entire multi-billion dollar business was based on isolated individuals subletting their apartment when they were out of town. Total nonsense but the gap between Uber’s “it’s just folks making a few bucks driving their otherwise idle car a few hours a week when they happen to feel like it” and reality is even greater.

The amazing thing isn’t that average people don’t immediately see through the “Uber is powerfully efficient because their cars have zero marginal cost, just like when neighbors share gardening tools” BS, but that major mainstream media continually push these false claims in the face of overwhelming contrary evidence.

To cite just one example, Om Malik, the main technology writer for the New York Times is either totally economically illiterate, or simply sees his job as lending credibility to false Uber claims. In 2014, he was aggressively pushing the “Uber will soon be cheaper than private car ownership” despite totally ignorance of Uber’s actual costs, or its costs relative to private car ownership” and predicted there would soon be many other companies exploiting the power of the “sharing economy” to create wonderful new consumer benefits in food delivery and other businesses.

By 2016, there was zero evidence that Uber had, or would at any point in the future become a credible alternative to car ownership, and absolutely none of the startups that had hoped to become the “Uber of” other industries had actually created a viable business. Did he suddenly realize that maybe Uber’s actually costs were much higher than he’d realized? Did he acknowledge that maybe the “sharing economy” was just a PR concoction? No, he doubled down, attacking the failed startups for failing to meet the standard of “Uber, the hypersuccessful granddaddy of on-demand apps” without explaining how a company losing $2 billion a year qualified as “hypersuccessful”.

How many Uber drivers just drive a few hours a week?

I especially appreciate the input from folks with actual taxi driving experience including Ono, Lynn, Synoia, Tim, Watt4Bob any anyone else whose names I missed here.

Throughout my career I have been on the corporate/consulting strategic planning side of things, but I learned long ago that strategic analysis that was inconsistent with the realities faced by front line staff was pretty worthless.

<Lynn> On special event weekends in my town, when Uber is in surge pricing, it’s worth some of my spare time to drive. During non-surge time frames, not so much…

This suggests, I think accurately, that any taxi capacity owned by private individuals can’t operate profitably at traditional taxi fares; they either need higher (surged) fares, or the reduced costs that centralized fleet ownership and driver control could achieve. The question is how much of Uber’s capacity in a typical big city is actually provided by people who just drive a few hours a week.

<Tim> Uber can do something taxi/car companies can’t do which is give drivers the ability to work part-time at driver chosen hours. I drove a cab for a while and enjoyed it but you had no option to lease your cab for less than an 8-12 hour shift. Given this and that the number of cabs was fixed, the “market” had no ability to adjust supply or pricing to meet peak demand periods – usually rush hours or during conventions etc. The incremental/variable cost for a driver to use their own car is less than a fleet owner’s cost.

<Reslez>No. As TFA stated several times, the cost to an individual is higher than for a traditional taxi company which has the advantage of a standardized vehicle fleet.

This is key to understanding the “sharing economy” issues. If you drive your personal car for Uber just a couple hours a week, Tim is correct and the true cost per hour is less than the cost of a fleet owner’s vehicle for those same couple hours. But if you look at the total vehicle hours Uber needs to operate to serve total market demand, only a tiny fraction could be operated by truly spare private vehicles; the overwhelming majority require intensively utilized vehicles and drivers. If you operate your private car in Uber service a large number of hours a week, then Reslez is correct, your cost per hour is much higher than what a minimally competent fleet owner could achieve. Isolated individuals can’t possibly achieve vehicle ownership, financing and maintenance costs as low as Yellow Cab.

Do Uber drivers understand the economics of working for Uber?

<Oho>Anecdotally, it definitely does sound like for many Uber drivers: gross revenue minus Uber’s commissions = net income. ….i’ve long given up shaking my head and trying to show the math.

<Watt4Bob> One of the things Uber relies on is drivers not understanding their own economics.

Thanks to Oho and Watt4Bob for taking the trouble to lay out some of the cost details drivers face. Oho’s explanation was based on a Honda Accord which is a good starting point for the economics a Uber driver is facing. My table of the traditional taxi cost structure in part two of the series was based on a Toyota Prius, which is typical of what a large fleet operator would ideally acquire.

<Kevin>I remember what it was like in 2011 or 2012, when everyone was still saying “we don’t even know what it is yet. Is it a taxi? It’s in a grey area. Are they 1099 or W2?…. I think a lot of entrepreneurship revolves around exploiting early days, unknown zones, and this is part of what they mean by “freedom”. The freedom to tell stories that aren’t bothered by facts, because the facts take time.

<oho> ‘What is the driver buying?’ –the equivalent of a payday loan—-as if one is an Uber-Lyft driver w/the wrong cost structure, she’ll quickly go ‘upside-down’ on her car loan, especially given the post-2010 trend of 60 to 84-month car loans

<Oho>As the NakedCap uber driver emeritus, my hunch is that Uber’s cash hemorrhaging is driven by the ginormous driver subsidies that Uber spends to attract/retrain drivers. —as the base rate that passengers pay is 100% unsustainable given the cost structure of the majority of drivers, so Uber throws out supplemental cash as a bribe, er incentive. Strategy being that Uber hangs in long enough to drive out its main competitors in US/globally.

As everyone understands the Uber business model extends the “independent contractor” model that traditional cabs have used for decades so that Uber drivers have to purchase, finance and maintain the vehicles they carry Uber passengers in.

Anyone with classic economics/MBA training would assume that while there might be some confusion and uncertainty at first, Uber drivers would quickly figure out that it would make no sense to work for Uber unless their gross take home pay (fares minus direct daily costs like gas minus Uber’s 30% cut) would need to much higher than take home pay with a traditional taxi company in order to cover all the vehicle ownership, financing and maintenance costs. It would actually need to be higher than that, in order to cover the risk that Uber could terminate them at will, leaving them stuck with all the vehicle costs. Again, people trained to think that markets work perfectly might understand that drivers might underestimate these costs at first, but word would quickly get out that you need a significant premium at Uber just to reach a breakeven versus driving elsewhere.

My articles argue that the assumption of an efficient driver labor market based on good quality information about pay is wrong because there was a fundamental “information asymmetry” between a multi-billion dollar company and hundreds of thousands of isolated drivers, and Uber worked aggressively to exploit this imbalance by providing blatantly false information (drivers making $90,000) and deliberately misleading information (implying higher gross earnings directly translated into higher real take home pay).

What hasn’t been clear to me is whether good information would finally cut through the initial barrage of deliberate misinformation, and the inherent difficulty of dealing with issues like depreciation costs, and how depreciation and maintenance costs increase if you drive your personal car 65 hours a week instead of 10. Oho’s comments strongly suggest the problem is even worse than I’d imagined.

To simplify just a bit, the Uber strategy was to (1) jumpstart rapid growth with driver pay premiums that would get lots of drivers to switch from traditional operators; these premiums were real but not as large as they seemed because drivers hadn’t figured out how to properly deduct vehicle costs to determine true take home pay, and by willful falsehoods (our drivers make $90,000) (2) gradually cut back driver pay once Uber was clearly a large established play by eliminating incentive programs and increasing the percentage of fares Uber retained; but drivers can’t do anything about pay cuts because they’ve locked themselves into car payments (3) At some point—and according to the study quoted in the second article in the series, it may have already happened—true Uber take home pay (after vehicle costs) is no better or slightly worse than what Yellow Cab paid before (4) Uber achieves industry dominance, drivers have no alternatives, and take home pay falls to (or even below) minimum wage level.

Do taxi drivers focus on peak demand in order to maximize earnings?

<Temporarily Sane> I was taking a cab last week on a rainy and windy weeknight and the cabbie remarked that he was going to end his shift early because the fares were far and in between. So it appears that cab drivers are not universally forced to drive for a full shift regardless of demand.

Speaking about the questions of whether academics understand real world taxi driving, there is a curious subliterature in the economic journals as to whether taxi drivers actually fit the neoclassic model whereby market participants rigorously focus on revenue/profit maximization. If they did, there would be clear evidence of drivers busting their butt on busy/rainy days and avoiding slow periods. Studies clearly demonstrated that this wasn’t true, and it was very common to see “daily income targeting” where drivers had a good idea of the gross earnings they’d need to cover the lease costs of the cab, gas, and justify the work they’d put in. On peak demand days, many drivers wouldn’t work the maximum 12 hours, but would quit early because they’d hit their daily target.

These articles got very angry rebuttals from doctrinaire free-market types who didn’t want their beloved “markets work perfectly in every and all situations” assumptions challenged but the rebuttals I’ve read were pretty weak. In simple terms, all the neoclassical models of the labor market assumed every hour of “labor” was exactly the same as every other hour of labor, thus higher peak demand should induce more labor. In reality, an hour of low paying, exhausting labor is different, and workers make the rational choice to blow off a few more dollars when it involves a lot more exhaustion, but the neoclassical models couldn’t deal with variables like exhaustion or nasty working conditions.

Here are some references in case anyone is interested. Camerer is the key author, Richard Thaler later became known as a leader in trying to incorporate behavioral issues into mainstream economics (and had a cameo in the movie version of The Big Short)

Camerer, C. (1997). Taxi Drivers And Beauty Contests. Engineering and science, 60(1), 10-19, Camerer, C., Babcock, L., Loewenstein, G., & Thaler, R. (1997). Labor supply of New York City cabdrivers: One day at a time. The Quarterly Journal of Economics, 407-441, Crawford, V. P., & Meng, J. (2011). New York City Cab Drivers’ Labor Supply Revisited: Reference-Dependent Preferences With Rational Expectations Targets For Hours And Income. The American Economic Review, 101(5), 1912-1932

Does surge pricing increase taxi supply?

<John Thacker>“Given the short notice this does nothing to increase total taxi supply” This seems contradictory. You repeatedly argue that the times that demand increases and that Uber is likely to engage in surge pricing are extremely well-known and cannot be changed by people needing transit. Fair enough, but if it’s that predictable, then surely drivers can change their behavior over the long run and thus increase supply. Do you really think that it’s impossible that drivers could say, “Hey, this time is a time that is likely to have surge pricing, so I’m going to be ready to drive?” The inelastic nature of the demand that you postulate helps nullify the short notice problem. And surely one effect of the “using your own car” aspect of Uber (and Uber taking a percentage for trips instead of a fixed daily cost common in regular taxi services) is that drivers are at least *somewhat* able to alter their behavior on shorter notice than in the taxi industry. In the non-owner operator taxi model (which is common in many cities) it is impossible for drivers to work anything less than a full day, due to the fixed costs paid to the cab company to use the vehicle for the day. This may be not enough, but you seem to be overselling your point claiming that it doesn’t increase supply at all.

As the question suggests, drivers know there will be more revenue out there on Saturday night than there will be on Monday afternoon, but they didn’t need surge pricing software to tell them that. The studies I’ve seen suggest Uber actually isn’t all that predictable about surge pricing as the duration of the surge and the small geographic zones that get surged aren’t plannable by drivers. Surge pricing responds to sudden demand spikes in the last few minutes. If demand suddenly spikes west of Central Park, but not east of Central Park, drivers on the east side can react to the surge and head west. But it isn’t going to get drivers who live on Long Island or New Jersey who weren’t already on the road to suddenly head into the city. So surges can shift existing capacity around a bit, but total capacity doesn’t increase.

In the third article I tried to make a major distinction between Uber surge pricing and airline revenue management. The latter plans both capacity and fares months in advance based on lots of historical data about demand on a given day at a given hour, and it is based on data for the entire market, not just the airline doing the planning. Customers then use the schedule and pricing information that all the airlines publish well in advance and sort themselves out. Both capacity and demand gets allocated efficiently but only because everyone made plans well in advance and had perfect information about all the options. Uber drivers have no idea what prices they can charge on Wednesday or Thursday, aside from their gut feel as to whether prices might surge or not. Uber passengers who head to dinner at 6pm have no idea what Uber will charge them to get home.

You’ve explained that Uber is fundamentally unprofitable. Maybe they’ve driven a lot of traditional operators out of business. But won’t surviving competitors and new entrants eventually kill them?

<Vlade> There’s little cost of switching for users and potentially drivers. The moat Uber is trying to create is that it drives prices down so much that it kills all competition – but that is not a viable long term strategy – because it present it with two choices only – keep losing money (not viable long term – yes, I know Amazon is technically losing money, but it has IIRC positive cashflow, and the loss is mostly due to investments – it could turn itself profitable anytime it wanted w/o impact on end customers), or raise prices (which would invite competition back).

<Oho> network effect. If I launch Humanity Rides LLC tomorrow, any rider using my puny network of drivers is going to likely wait 10-20 minutes for a ride. …Versus the 1 to 8 minutes for Uber. People are a lot more impatient than you think.

<Tim>I am wondering if Uber’s anti-labor knife will ultimately cut both ways. Both customers and drivers can and will easily switch to and among alternatives as they arise.

<Lynn>Uber will have issues retaining drivers, hence the interest in going to driverless passenger pods.

<Damian> – this business whether Lyft or Uber et al – is merely a dispatcher gone digital –the attempt to take rentier profits by Uber will fail. This model will revert to essentially a Cooperative (“COOP”) of Medallion Owners with the economy of scale associated with: insurance / maintenance / acquisition of vehicles / use of vehicles 24 / 7 by multiple drivers and the Dispatcher APP used for $1.50 per ride.

<voteforno6>What I’ve been trying to figure out is, what is the endgame? If Uber does somehow manage to gain market dominance, then presumably they’ll raise fares in order to achieve profitability. Wouldn’t that just encourage smaller operators to jump into the market? I suppose that Uber could then push for regulatory barriers to entry for these smaller operators, but that would certainly fly in the face of how they’ve been running their business up to this point.

<Samuel Connor>I speculate that after the traditional fleet operators are driven out, it may be hard to resurrect successors when fares “normalize” since the up-front investment to re-constitute the local taxi fleets may be prohibitive
Taxi markets fall into three main segments that have totally different competitive dynamics.

“Streethail” dominates in Manhattan, similarly dense European/Asian cities, and in the central business/entertainment core of a few other large US cities.

“Taxiranks” are what you see at airports, train stations, convention centers, and other places where you have very large, very concentrated pockets of demand.

In the US, the overwhelmingly dominant model is “Dispatch” where demand and trip patterns are of very low density, so cabs are ordered by phone (or now, by smartphone app) and cabs assigned by a dispatcher (or dispatching software). Individual owner/operators can easily serve airports or the taxiranks at big hotels, but will have little ability to get any business from widely dispersed houses and businesses that occasionally need cabs.

Thus in the vast majority of cities, the taxi industry was organized around a bunch of dispatch companies. The predatory competition funded by Uber’s investors has devastated these dispatch companies.

The commenters above were implicitly asking that if Uber abused its dominant market position, couldn’t new taxi companies sprout back up overnight and reclaim a big chunk of the market.

There’s no problem with new entry at airports and other dense locations, and it is unlikely that Uber would ever completely eliminate competition in these segments. But it is next to impossible for drivers who are limited to these markets to actually make a living—without a broader base of revenue you can’t cover the cost of a full day’s driving and fares will be held down by drivers just trying to earn a few extra bucks. The question is whether new dispatch competition could ever emerge. In a competitive market where Uber isn’t a dominant player, the Uber app has little or no impact on Uber’s ability to operate profitably. But as many people understand, the value of the app seriously kicks in after Uber becomes dominant. Once everyone who uses (or drives) cabs regularly has to have the Uber app on their phone it becomes much much harder for newer, smaller companies to get to the scale where they can provide a strong competitive challenge. As Izabella Kaminska of the Financial Times put it, with dominance the app would provide the basis for controlling “a rent-extraction business of the highest middle-man order.” And of course, Uber’s billionaire owners could just cut prices until the new competitors failed.

Is Uber really a software company? Is it actually any good at software?

<voteforno6>I think that Uber has a better chance at becoming profitable if it focused on developing and licensing its software for real taxi companies.

<jon s>As a software engineer by trade, Uber’s software is not that difficult to replicate.

<dave> What’s to stop clever coders from creating their own Ubers,

<Temporarily Sane>Uber is about as “digital” as Domino’s Pizza, after all they have an app too, right?

<Kevskos> Seems that Uber has a few things going for it. An app that many people have installed, a name they can remember and they can use it everywhere they go. Seems like all they have is a dispatch service. Seems like the profitably way to run the business would be to charge taxi companies a very small fee (somewhat lower than dispatch and cc fees) and operate as a dispatch company instead recreating existing infrastructure.

A few years ago “we are a software company not a transportation company” was a major Uber PR theme, supporting the larger theme that regulations that applied to every other company that took money in exchange for rides in cars. How could local regulators who dealt with Yellow Cab ever figure out how to regulate a software company. Kalanick insisted they were just a passive intermediary ““Are we American Airlines or are we Expedia? It became clear, we are Expedia.” This was palpable nonsense when uttered back in 2013; with hindsight it is clear Uber hasn’t developed any software other than what was necessary to run its transportation operations.

The software behind Uber’s app appears fine, and people like the app, but as correctly noted, Domino’s Pizza has a nice app too, and it didn’t allow Dominos to drive all its competitors out of business and create $69 billion in corporate value. Yes, Uber could have just decided to be an “app” company, and could have sold software to taxi operators all over the world. But that wouldn’t have created a $1 billion business, much less a $69 billion one. In my airline career I’ve run pricing departments and helped design revenue management systems. The systems behind Uber’s ordering app are incredibly simple compared to what airlines and hotels have been using for decades.

<Silas Barta>They take people from point A to B in cars, but not “like traditional car operators always have”. Name me a service operating before 2010 that had smartphone integration, reliable, tip-free service, actual responsiveness to complaints, and which was willing to drive you to far-out places, and which had mutual rating to weed out bad drivers/passengers.

<Art Eclectic> I disagree with the premise that Uber has no meaningful competitive advantage. What Uber has is a ubiquitous app that works in whatever city the service operates in. Until I can say, “Okay, Google. Call me a cab. This location to destination SFO” Uber has an advantage.

As mentioned the app is nice and people seem to like it. But does it make Uber profitable? Does it create a huge efficiency or pricing advantage that competitors couldn’t readily match? The smartphone integration is certainly useful, but did this create a 2% improvement over traditional telephone dispatching, or 25% better efficiency? I’m not sure but I’m guessing that it is closer to 2%.

How much has Uber spent on software, and more importantly how much did it cost to scale all the behind the scenes scheduling and pricing software so they could function in every city in the world? The jury is still out, but I’m guessing the costs to date have been much higher than the cost savings and competitive benefits they’ve generated.

And remember the service improvements many attribute to the app were really due to other hugely subsidized service advantages. The Uber app produced cars sooner, and at lower cost than Yellow Cab could offer because of those subsidies. If smartphone requests for Uber cars led to the same long, unreliable wait times as Yellow Cab, no one would be talking about how wonderful the app was. Nobody would be talking how wonderful it was that you could use the app in different cities.

What about carpooling as a future growth opportunity?

<Alex> What do you think about uber pool service (when you share a car with fellow passengers going approximately in the same direction)? It seems to be beneficial, as less energy is super and pollution generated per passenger. Do you think it could be replicated by traditional companies and could it be a source of competitive advantage for Uber?

Urban transport people have been touting multi-rider options (carpools, jitneys, paratransit, dial-a-ride) for approximately forever. The occasional places you see actual operations require much larger subsidies than regular transit services. No one has the slightest clue how to create a commercial service here, outside of very isolated cases like SuperShuttle.

Problems include (1) people don’t like sharing vehicles (2) people really don’t like it when their trip takes twice a long because of the other people; because of (1) and (2) nobody with access to a car ever uses the service (3) the normal time wasted in single-trip taxi service (taxi takes 20 minutes to arrive instead of the promised 10, passenger isn’t ready when they said they would be) gets hugely magnified here (4) software can’t solve very much of the trip inefficiency problems because it doesn’t have accurate info in advance, and the demand density isn’t big enough to create a lot of opportunities for more efficient trips. (5) in a big dense city, where there’d be enough people going in the same general direction at any given time the efficient solution has already been found. It is called “public transportation”

When traditional taxi regulations were established in big cities in the 1920s and 30s, multi-destination options were banned because the (then privately owned) transit operators didn’t competitors weakening low density routes. When libertarian/anti-public transit advocacy groups fought for taxi deregulation in the 90s, they prominently pointed to these vestigial protections and claimed that regulation had stifled industry “innovation” for example these jitney/paratransit services. These rules were eliminated everywhere; absolutely no innovative services emerged.

Is Uber like Theranos?

<Ibrahim> Do we have another Theranos on our hand?

The similarity is that the much of business media completely swallowed a company narrative about heroic, cutting-edge innovation that would bring huge benefits to consumers, and didn’t bother to look at basic economics to figure out where the huge benefits were coming from. But aside from that, hugely different. Theranos was promising its customers specific product features that it couldn’t deliver; they could blow smoke for a while, but eventually people would figure out the product didn’t work.

Uber hasn’t promised customers anything but a ride from point A to point B, and it has delivered on that, and given the subsidies that have increased cab availability and held down fares, customers are happy. People may discover that the low prices and cleaner cabs and shorter waits aren’t sustainable, and people may discover that a dominant Uber makes their cable TV and phone providers seem like efficient lovable companies, but that’s a totally different set of issues than Theranos raised.

Did Uber make an ill-fated wrong turn a few years ago?

<Ben Mandell> instead of pursuing a business model that actually would make money, they made an unfortunate decision to put everyone in the taxi profession out of business. In the beginning, Uber was a superior product (in many locations) to a traditional taxi service. Instead of capitalizing on that superior product offering and charging what is actually needed to provide that product and make a profit,

<Art Eclectic>If Uber stayed with their original model and hadn’t gone after global domination, the picture would different. If they stayed with major metropolitan markets and focused on solving a real problem where there was unmet market demand for better transportation options, we might be having a different conversation.

<cnchal>. . .They take people from point A to B in cars, but not “like traditional car operators always have”.

Traditionally, they were called taxis, and the taxi operators were called “cabbies” and in the old days, when a cabbie gave a customer a rough ride, a call to the dispatcher by the customer would land the cabbie at the bottom of the call order list, waiting a loooong time for the next fare, and gasping for money to pay the daily rate. Cab companies even had their own communications infrastructure in the form of 100 foot tall aerials and radios in every car, so not even cell phones, never mind smart phones, but so what? That got the jawb done, which was pick up your fare and take them to where they want to go, the further the better.

Go ahead and believe the whole Uber turd is worth 15 Nimitz class aircraft carriers, after they only invested in three and then sunk a few to gain “market share”. The Fed works in mysterious ways, and before you know it, the finance flim flam artists will have sold it to you.

Have Uber’s investors been fully aligned with what management has been doing?

<Matthew> So the thesis of this series is that Uber’s business model is to: * Raise lots of money through “the bezzle”.* Use this funding to drive out the competition through predatory pricing. * Extract monopoly rents once this is done. I am inclined to agree to agree.

I’ve argued with Lambert about this, and it is really a semantic question. “Bezzle” implies either a Theranos-type situation where the product itself is fundamentally fraudulent, or a situation where the early round investors have been totally snookered by an investment proposal that fundamentally misrepresented where investor returns would come from. This is certainly debatable, but I believe that Uber’s investors and senior managers have had very strongly aligned views of what the long term business objectives were, and where ROI would come from. The long term objective was to create a quasi-monopoly company and to undermine all traditional legal/regulatory obstacles to exploiting anti-competitive market power. Undoubtedly lots of different perceptions among investors about which elements of the business model would be most powerful, and clearly frustration among investors about Kalanick’s slow path towards an IPO, but from any strategic perspective Uber’s managers and investors have been very much on the same page.1 But aside from this semantic quibble about the “bezzle”, I think Matthew summary was spot on.

<Tim> Exactly! It’s devil take the hindmost like most unicorn companies today. The early/smart guys have gotten out or will before the rest of the world figures it out.

<Teddy> is actually just a ruse to con wealthy investors out of their money and only a select group of insiders is expected to make a profit?

<Tongorad>I’m guessing Uber has already delivered, in spades, to a relatively small number of pirates at the top.

<Larry> I agree that for deep pocketed investors, an IPO is a must. Multiply the value of their private shares on the public market thanks to multiple hype machines (Wall Street banks, fawning media, etc) and then dump the shares for fantastic gains.

Given past unicorns, the concerns here are certainly understandable. One of the themes of this series is that Uber has been completely different from previous tech startups. The important differences involve strategy, politics and the sources of investor returns. But the difference in the timing of investor returns are also important to understand, and I think Kalanick deserves credit here, even if the differences aren’t creating any broad economic benefits.

As these comments suggest, many previous startups have been structured to create a fairly quick opportunity for monetizing the corporate value created. Put in money, start a business with a bit of buzz, quickly go public before the difficult bits of building a business hit home, then the company shrinks/collapses/sells out at a discount. Uber has been playing a much longer game. Uber hasn’t gone public, isn’t going to anytime soon, and makes it very difficult for investors to sell stock privately. The monetization target isn’t peak media coverage (2014 for Uber) but when full industry dominance has been secured. None of Uber’s early investors have cashed out with big profits, which has been very frustrating for several of them. But waiting for dominance could produce much bigger returns than any previous unicorn saw.

<LT> You have to go a step further and realize that there are investors that are not investing in the success of Uber, but they are investing in Uber as a vehicle to get rid of certain laws and regulations. The wool was never pulled over their eyes about “profitability”. It’s an investment based on ideology and they would deny that every step of the way.

This series has focused on the economics of Uber in the marketplace. If you wanted to better understand Uber’s origins and many of the tactics it has used, and why it has always been focused on monopoly, and the extreme nature of the “deregulation” it has been pursuing, the political/ideological worldview of its founders and investors becomes very relevant. Also relevant to my point earlier that investors and management have always been strongly aligned about Uber’s basic strategies and objectives. But I’ve seen no evidence that any investors put money into Uber just to pursue political ends. I think “possibility of big financial returns” was the primary motive in every case. But I’m sure the worldview compatibility made it easier to conclude that this management team might actually produce those big financial returns.

If Uber has all these problems couldn’t we short it and make a lot of money?

<Lyman>This reminds me of the story during the runup to the financial crisis where potential investors met with the execs of some company issuing CDOs and asked how they were going to make money. The execs couldn’t offer an explanation in plain English at which point the investors realized the execs had no idea what they were doing and went short instead.

<Just an Observer>Yeah. Trouble is that Uber (& other pointy headed startups) are trying to act like quasi-public companies – selling shares in private placements – while rigidly avoiding publishing any real financial info. To short it you’d have to find one of the private investors willing to lend you their shares *and* find and even more idiotic investor to buy the borrowed shares from you. Not gonna happen so one of the ways Mr. Market can start shouting that Emperor Kalanick is naked is not available … which is, of course, another reason Uber’s IPO will happen sometime after the heat death of the Universe

Excellent question, and there’s a very good article that answers it in detail.

LeVine, Steve, Investors have placed a one-way bet on Uber—which made us want to find a way to short it, Quartz, 5 Aug 2016.

LeVine is dubious about Uber—he doesn’t argue as I do that the entire business model is based on hopelessly uncompetitive economics, but he’s sees that the kind of positive evidence that past successful unicorns could easily produce is missing, and he recognizes that even if Uber has come up with important insights, that doesn’t mean that they will be the ones to successfully exploit them commercially.

I’ve argued that Kalanick’s refusal to pursue an IPO—over the objections of powerful investors who would like to take some of their paper profits—is fully rational because Kalanick understands that the data that would accompany an IPO would burst Uber’s carefully developed PR narratives. Some of the threat would come from the media, who would be looking for headline grabbing news out of the spreadsheets. But as these comments suggest, the bigger threat would be from shorts, who could destroy the illusion that Uber had created $68 bn in corporate value a lot more effectively than any journalist. For a company that portrays itself as a paradigm of objectivist competitive virtues, Uber is terrified of the possibility of actual scrutiny from the capital markets.

Again, the game plan here is to hold out until Uber has a strong enough market dominance, so that many of the risks shorts would be focusing on go away, and the IPO price is inflated by all the monopoly power it can readily exploit.

_____

1 I hate to differ with Hubert, but a venture capitalist specializing in auto-related technology took issue with the idea that any unity of vision among Uber’s investors was necessarily all that well informed. Via e-mail:

We have experienced this in our own research where we have the sense that something is amiss but because of lack of complete information we are a bit off with the facts. First, it is we’ll known that many Silicon Vallyers are libertarians, therefore they a service like Uber would be appealing philosophically, and of course they would advocate for less regulation. Second, VCs are like lemmings, they follow the lead of other VCs whom they perceive to be sophisticated, and often nobody performance any due diligence, I have seen this first hand. They all kinda assume that someone has looked at the deal carefully. Third, smart entrepreneurs always create a sense of urgency when raising money, so with oversubscribed deals VCs have very little time for perform due diligence, some times they only have days to decide, so again diligence is spotty. Forth, VCs know that hot deals are good PR when raising money. Limited partners, particularly the institutional kind, think that a VC must be good if they can get into hot deals, never mind the economics. Finally, VC community is very small and tight, so no one will ever say another bad about anybody, even when they are total crooks.

Print Friendly
Tweet about this on TwitterDigg thisShare on Reddit0Share on StumbleUpon0Share on Facebook0Share on LinkedIn12Share on Google+0Buffer this pageEmail this to someone

51 comments

  1. TheCatSaid

    Great summary and response to comments. Thank you Hubert Horan for connecting all the dots, and to the commenters.

  2. DolleyMadison

    I think you have a “big city” approach to this discussion. From a customer’s point of view, I love that I can get an Uber anywhere, anytime — especially when visiting suburban and even rural areas where my family lives in Kentucky where NO cab would evehn come and if they did it would be 100 dollars to get to the airport when an Uber charges 15.00. Every Uber driver I have had said they do it while not working their other job – kind of taking a “if I get a fare fine if I don’t fine” approach. And even in my hometown, Uptown, it seems to work better. I had to take treatments at a hospital 3 miles from my office in the middle of the day for a month. The first 2 times I walked across the street to where the cabs were lined up waiting for the banksters to come out for rides to the airport. No one wanted to take me on such a short ride and it took a few minutes to convince them to drive me. And they got “lost” both times when it is one turn on a main road. So I started calling Uber. They came immediately and charged LESS THAN HALF. And they always asked when I would be finished and came back to get me. (I always tip pretty well)

    1. reslez

      As the series points out, those drivers are only available because Uber massively subsidizes both drivers and riders, who pay only 41% of the fare. As soon as Uber achieves a market dominant position they take away the subsidies and the drivers disappear, because those rural routes are money losers. Which is why they aren’t served well by traditional companies today.

      Uber doesn’t offer any kind of revolutionary technology that turns a money losing rural route into a profitable one. All they have is billions in investor subsidies to drive their competition out of business. And your driver almost certainly wasn’t properly insured.

      1. NotSoSure

        As they said, you can hire one half of the poor to kill the other half. At the end of the day though you have to treat “subsidies” like love. Get it while you can :)

    2. Anon

      And those Uber drivers will disappear from those rural roads as soon as the subsidized fares and real auto operating costs are recognized.

      Enjoy the interim serendipity.

      1. WindyGirl

        I hate to sound ignorant, but I really want to understand what you mean by Uber “subsidizing” the drivers. I am a driver, and Uber’s claims of $90,000 per year is ridiculous even for those who drive several hundred miles to a distant busy city and even this gig is slowing down. Those of us who must remain close to home in the “heartland” make an average of $6 – $8 per hour before expenses. As the author stated, less than minimum wage and I’m locked in to a car payment. Granted, what little I do make does pay for my car, but the grinding wear and tear on the car alone is unbearable, but to hear people like you rationalize that Uber is somehow subsidizing me, fries me.

        Perhaps you mean Uber’s offering up to $5,000 to new drivers? Word has gotten out and new potential drivers are no longer abundant? If by the huge bonuses you mean Uber is subsidizing drivers, it is at the expense of the little man at the bottom of the totem pole.

        1. aab

          I read this series when it was published a couple of weeks ago, but the subsidies TO drivers was the first wave of the roll-out, IIRC — basically, the subsidy allowed them to recruit existing high-quality drivers at launch by paying them slightly more than existing taxi companies in major cities. The subsidy issue now is more that if not for the subsidies from the investors to the company’s functioning, it would have already collapsed, because the business model incurs massive costs that typical taxi companies do not. So the subsidies were never primarily about better pay for drivers long-term. It’s about keeping Uber in the game long enough to develop monopoly power that will crush its competitors AND its employees and gouge its customers.

          I’m really sorry you’re stuck driving for such an exploitative company. I wish had helpful advice on how to get away from Uber. If setting up a worker coop might be feasible, I’ll go look for links.

          1. Skip Intro

            Better to be unemployed, than to lose money and subsidize Uber by working for less than the expenses (including req’d insurance, and wear&tear on the vehicle) cost.

  3. Schtub

    This has been an outstanding series, and I’m grateful for it.

    I’ve never used Uber, and won’t until forced. I’ve long understood that Uber was a bad deal for its operators, if only for the insurance reasons. I also resented the shredding of regulation in my city – there were some initial indignant protestations from the major and city council when Uber began operating counter to law, and then evidently a corrupt payoff price was reached and everything reversed.

    What I hadn’t understood clearly was how epic a money loser they currently are, or the monopoly-or-bust exit strategy. And now I like them even less.

    1. JohnnySacks

      I’ve actually been wondering how many Uber drivers have actually done the necessary homework and informed their auto insurance company that they were using their personal vehicle as a (quasi) commercial livery service. Yet another instance of a hidden operating cost, papered over by blatant cheating.

      1. oho

        ‘ and informed their auto insurance company that they were using their personal vehicle as a (quasi) commercial livery service.’

        big likelihood that your personal insurance coverage would be terminated or unilaterally not renewed. hence no driver in their right mind will tell State Farm/Geico/Allstate.

        Obviously big liability risk.

        1. Phil Martin

          Anecdotal data here, but an acquaintance of mine driving for Uber was sideswiped by a vehicle that drove off. One of the three passengers suffered some minor injury, and demanded the driver call police. The passenger needed a police report for her health insurer.

          The Uber driver sustained $4,000 damage to the vehicle, almost all of it cosmetic. The driver was under the impression that the $1.00 insurance charge on each ride contributed to the million dollar insurance policy. She was correct, however, the policy indemnifies Uber from claims by passengers and drivers. Uber told her to contact her insurance carrier first before they would even consider her claim (apparently, Uber sometimes makes accident victims whole, especially if the accident is news-worthy, and results in negative publicity. This accident wasn’t in those categories). Bottom line: the auto insurance company was not notified, because the existing police report would have shown she was operating her vehicle commercially, not as a private vehicle. She had to stop driving, and has to pay to fix her vehicle.

      2. weinerdog43

        I was wondering when someone got to this point. I can catagorically state for the record that if you have an accident while operating as an Uber driver in your own vehicle, and you have not updated your personal auto policy, YOU HAVE NO COVERAGE FOR THAT ACCIDENT. All personal auto policies contain an exclusion for ‘..transporting people or property for a fee’.

        You need to purchase either a commercial auto policy, or investigate an endorsement to your own policy to cover this type of situation. Ka-ching. Another expense Uber does not address. Also, I wonder how many passengers know that it is very likely their own Uninsured Motorist coverage is going to be primary?

        1. shinola

          Coverage for “commercial” use such as being an Uber driver or food delivery is not provided under a Personal Auto Policy and is not an option available under the PAP – at least not in any of the states I was licensed for (30+ years in insurance biz).

          There is a “business” use classification available but it is intended for “personal & incidental” use such as making sales calls or a construction worker who might have to drive to multiple sites in one day.

          I suspect that Uber provides some sort of coverage for its drivers – probably “self insuring” as most taxi co’s do. The cost for an individual to provide their own commercial insurance would be prohibitive – even for that theoretical driver making $90k per year.

  4. Schtub

    Oh yeah, one typo:

    ” subletting their apartment when they were out of time.” Did you mean out of town?

  5. ChiGal in Carolina

    Great post. I confess to not reading the earlier ones cuz I made my mind up early on this and never have or would use Uber or Airbnb.

    What I especially appreciate is the extensive use of comments. I will be forwarding this to a number of folks to showcase what NC is all about – great critical thinking and a high-level discussion all around that just can’t be found anywhere else.

    Thanks, all!

  6. Larry

    I am suspect that Uber could ever drive professional car services out of business, especially at the burn rate of cash. It would be interesting to see some data about traditional taxi service numbers by employment or some other metric, so I’ll assume that number is hard to come by.

    This raises another question for me. When does Uber give up on its paltry autonomous car technology? This is what the hype machine focuses on now, the fact that Uber has unleashed supervised self driving cards in Pittsburgh. Lambert has well covered how far out autonomous auto tech really is, so would Uber really continue investments in this when their only path to profitability seems to be to destroy most competition and establish monopoly rents?

    1. Arizona Slim

      Supervised self-driving cars in Pittsburgh. The city of my birth.

      Also home to some of the steepest hills in America — try driving up a 20% or even a 37% grade. Or going downhill. Yikes.

      And don’t get me started on Pittsburgh’s abundant potholes. Just don’t.

      I have a feeling that this won’t end well.

    2. reslez

      Well, they aren’t actually self-driving because there’s a human person at the wheel ready to take over at a moment’s notice. The article I read stated the human monitor was visibly nervous and a human had to intervene at points. That doesn’t sound anywhere near “self-driving” to me. It sounds like vacuous hype.

      1. bob

        “there’s a human person at the wheel ready to take over at a moment’s notice”

        Take over RESPONSIBILITY at any point. In reality, today, the human never cedes responsibility.

        But “computers” never make mistakes.

        1. Arizona Slim

          I can’t wait to see what happens on Negley Hill on an icy January morning. Or any road leading up to Mount Washington.

      2. Carol Sterritt

        I have noticed the same “smart car expert” on various news shows bringing forward a “new consideration.” He goes on to explain that this new consideration revolves around the fact that it was necessary to “program” into smart cars the directive that passenger safety first, outside world safety second. (Or maybe not at all.) He made it seem like some soulful group of ethically serious saints undertook this decision. In fact, I think this is mere cover for the fact that the technology does not yet exist for road worthy smart cars.
        I am the survivor of a car-truck accident wherein a semi-trailer truck hit my car when the truck was doing fifty or so. It was a “no fault” accident. Our car was immobile behind a line of cars behind an earlier accident. The truck driver did nothing wrong. He was “test driving” a brand new truck right off thre assembly line. The brakes were faulty and they failed. At the last moment, he turned the truck’s wheel and jack knifed. This saved all four of us: my dad at the wheel, my mom in the passenger side and us two kids. Would a smart car know to jack knife if its brakes fail? Will a smart car immediately slow when the three year old straggling along behind his mom drops his teddy bear? You would slow and I would slow, as we know the kid is gonna turn and run our way to retrieve that teddy. But a smart car is going to run that kid down.

    3. Phil Martin

      Heard a lecture by a current lead engineer at Nissan who formerly was a Silicon Valley computer dude. He said driverless cars–the sort where you don’t need a pair of eyes watching the road/watching the computer–are still ten to twenty years away. And he added a point that’s often overlooked: engineers need to find a form of self-driving car that is “socially acceptable.” By that, he meant one where people felt safe and would choose the driverless vehicle over the one with a human at the wheel.

      He went on to say that one of the biggest problems was how to navigate a car through pedestrians using crosswalks at 4-way stops. Humans are too unpredictable. Moreover, cameras and sensors frequently mistake pedestrians for trees, signs, almost anything vertical. Yes, you can make an intelligent car, but for it to work, you need intelligent streets, intelligent maps, and much more intelligent sensors.

      We will most likely see driverless vehicles on major roads hauling freight long before we have passengers in such vehicles. Cargo doesn’t get anxious.

      Uber is grasping at straws.

  7. Damian

    Mr Horan – you have done a great job on this subject!

    Yves – this format has been special on your site relative to the presentation on other subjects but also as someone who reads many sites may be unique in general.

    The doubling back in this edition to further penetrate the subject matter to dig deeper based on summarizing commentary and giving further information / explanation is great addition.

    The way to challenge this “Fake News” Obama Mime of alternative media is to show the NYT is the fake site just like Mr. Horan has done with the Uber comments from Malik.

    If this is done on a variety of subjects it will be deadly for the NYT propaganda and prove where the Fake News really lies – at the MSM. Irwin another NYT journalist is prime meat. The entire BS at the B.L.S. is waiting for discovery of the specific elements of the metrics used in the: Job Additions & Unemployment Rate – which Irwin extols each month as an Obama Tool.

    Good Luck and thank you !

    1. LA Mike

      I agree.

      For many sites today, the commenters make the site as much as the original content. (At Slate, the non-forum-spy commenters are superior to the content.) So to have a format wherein the writer takes several different comment-types and responds to them en masse is great.

      Good job, NC and Mr. Horan.

  8. Lune

    Fantastic summary! A few questions :

    Even if Uber succeeds in creating a worldwide monopoly, isn’t their valuation still too high? For all of ubers publicity, taxi services are a miniscule part of the global economy. Even airbnb has a higher potential given the total net worth of the hotel industry. If I were Kalanik, I would sell now because I doubt the validation would go Any higher even if he reaches world domination. But then again, maybe this is why he keeps spouting nonsense about replacing privately owned cars…

    For the Uber drivers: if you take a fare a long way from home, does the dispatch software give you precedence for a return fare back to your home turf? For example if you live in long Island and you take a customer into Manhattan, does Uber try to find you someone going back to long Island or is it completely random? For that matter do manual dispatchers in regular taxi companies try to do something similar and optimize your return trip?

    1. oho

      ‘isn’t their valuation still too high?’

      AMZN’s current/past valuation suggests that Mr. Market doesn’t care about GAAP valuation if you have the right story.

  9. Peepers

    There is a reader comment in the post that the formatting presents as Horan’s own voice under the “Did Uber make an ill-fated turn” section:

    Traditionally, they were called taxis, and the taxi operators were called “cabbies” and in the old days, when a cabbie gave a customer a rough ride, a call to the dispatcher by the customer would land the cabbie at the bottom of the call order list, waiting a loooong time for the next fare, and gasping for money to pay the daily rate. Cab companies even had their own communications infrastructure in the form of 100 foot tall aerials and radios in every car, so not even cell phones, never mind smart phones, but so what? That got the jawb done, which was pick up your fare and take them to where they want to go, the further the better.

    Go ahead and believe the whole Uber turd is worth 15 Nimitz class aircraft carriers, after they only invested in three and then sunk a few to gain “market share”. The Fed works in mysterious ways, and before you know it, the finance flim flam artists will have sold it to you.

  10. Arizona Slim

    I’ve been in rooms where Uber drivers have been in meetings with company representatives. Just about all the drivers are what I’d call down, out, and desperate. Many of them middle aged and elderly.

    The driver app gave them a lot of fits, and I couldn’t help thinking that it was designed that way. Sort of like the software that’s used for wage theft in other establishments.

    More than a few of them voiced displeasure at how they were being treated by Uber, and I saw plenty of evidence of that. To a man and woman, the company reps were condescending, and the drivers picked up on that right away.

    Most of them appeared to be trying Uber, and as soon as they realized that they weren’t going to be making any money, they bailed. I heard more than one “I quit!” rant.

  11. oho

    ‘, there is a curious subliterature in the economic journals as to whether taxi drivers actually fit the neoclassic model whereby market participants rigorously focus on revenue/profit maximization.’

    flat out no. Uber drivers (or humans) are not Homo perfecto economicus rationalitis.

    Example. Insurance coverage. Uber drivers have a non-odds of fatality/serious injury—especially given that for a ‘power driver’ a big bulk of miles driven occurs at the more dangerous overnight hours.

    It’s a fair bet that 95%+ of UberX drivers are severely under-insured based on actuarial risk (sadly an easy situation to fix if Uber acted like a risk mutualizer).

    It’s also a fair bet that a big chunk of UberX drivers are judgement proof.

    It’s another fair bet that right now passengers are likely lucky that Uber will be willing to no-fault settle any losses from a ride. —that’ll probably change as Uber rein in its costs in the future.

    It you’re paranoid about tail risk like me :) —for heaven’s sake acknowledge your driver and take a few seconds to say hi and check her mental state (ie are they drowsy, really angry, distracted, intoxicated, etc).

    There’s zero mechanism on the app that prevents a driver from accepting a passenger if she is not in the right mental/physical state. just sayin’.

  12. Patrick

    I’ll fully accept that Uber has its eye on monopoly. But as an individual, I’m stuck in a bit of a tragedy of the commons. The rational choice right now is for me to take Uber drives and profit off their current large fare subsidy, even if it is a trap in the long run. In the event that they transform into a monopoly, my next best option is my personal vehicle, so the worst case doesn’t represent a huge cost.

    1. Dirk77

      If you weren’t going to pay for a regular cab if uber didn’t exist, then it doesn’t matter. In that sense, we should all take types of rides with uber that costs them the most subsidy-wise. Perhaps Horan can make suggestions?

  13. oho

    as a FYI. nearly everything that applies to Uber also applies to its main competitor Lyft—-just swap the corporate logos.

    Despite Lyft’s marketing that it’s the ‘Jet Blue’ of taxis. 100% pure PR-driven white/green-washing.

    only difference on Lyft is that you can tip your driver using your account’s credit card. and as such, Lyft’s platform, on the margin, attracts/retains drivers who are more customer service/’emotional intelligence’ aware.

    citation: anecdotal evidence talking to passengers as a driver on both platforms.

    and if you’re a passenger and have the means, tip your driver if/when appropriate. it’ll be greatly appreciated.

    1. Enon Zey

      I always tip service personnel in cash. That includes Uber drivers.

      Just anecdotal, but another reason Uber is eating the taxi companies alive – I know a nurse who works for a major metropolitan government owned hospital; she often has to call the local traditional taxi service to transport patients. They very rarely come on time.

      OTOH, my and hers experience with Uber is that the drivers arrive when the app says they will. Quickly.

  14. Vastydeep

    Venture Capital as we know it basically dates from the mid-1950’s – the transistor Nobel to Bardeen, Brattain and Shockley made the “traitorous 8” and the seminal Arthur Rock investment inevitable. Transistorization was going to spawn a trillion-dollar industry, and in February 2000, Worth magazine noted: “venture capital is an industry in which more than $20 billion in annual profits is produced by fewer than 3,000 workers.” *Everybody* made fantastic money, and that spawned the libertarian, master of the universe VC archetype we see today. Here practically 100% of success really was “just showing up.”

    “Sharing Economy” deals are a return to making money the old, pre-transistor way — through regulatory and informational arbitrage, followed by regulatory capture and then restraint of trade.

    If the Moore’s Law pipeline really has dried up, then we might expect to find arbitrage (rather than technological innovation) at the hearts of many of this round of unicorns.

  15. Tim Steele

    There is a fundamental misunderstanding of Uber’s business model in this series. Or the unit economics of a successful software company in the access economy in general that is scaling! That you are picking on Uber and Airbnb, speaks to your bravery sure, but also to your folly.

    For a far better analysis, please see: https://stratechry.com/2014/uber-fights/

    At least you have skin in the game! So when Uber goes public in the next year or two, and is not only profitable but appreciates in value, then we will have to revisit this post!! And when Airbnb reaches a valuation of over $50B, then we can really see who was right!

    1. Yves Smith Post author

      It appears that you have not read this series with any care. Uber is not a software company. The functions of its app are not distinctive, not hard to replicate, and not patent protected. Uber does not scale because it is providing $2 billion a year in direct subsidies to drivers. Its competitive advantage is a result of these subsidies, not any magical software.

    2. ChrisPacific

      You’ve offered no evidence at all to support your point of view beyond the link. I did a quick read through the content at the link and discovered that it was similarly light on evidence. It (a) is largely focused on how Uber can defeat competitors like Lyft, (b) makes no attempt to explain how Uber can remain competitive with traditional taxi services when it stops subsidizing customers at 60 cents on the dollar, (c ) contains a number of embedded assumptions that have been demonstrated to be fallacies in this series (“ride sharing”) with no attempt to justify them, and (d) places complete and blind faith in a plausibility argument of a winner-take-all dynamic based on network effects without any attempt to justify it based on evidence. For example, if the winner-take-all dynamic based on service quality is so strong then I’d love to know why existing taxi services hadn’t already consolidated into a single monopoly provider before Uber came along.

      I’m not sure why I’m bothering to respond, since posts like yours were a dime a dozen during the dot com bubble and the authors were generally deaf to any counterarguments, but the fact that you have found your way to NC makes me think there might be some hope for you. I encourage you to stick around and learn to distinguish between evidence-based reasoning and hype.

  16. etherist

    Look for a Yellow Cab hailing app in the app store fir your phone. Marvel at the reviews describing buggy software, poor UI, and terrible service from the actual cab drivers and dispatchers . Really, all the reviews that are more than five words long all suck!

    I hate cabs in my city. Except coming home from the airport. Every time I’ve tried to request one by phone, they’ve been late. Every time.

    Can we have a non exploitative Uber? I would be OK paying twice what Uber normally costs, especially if there’s no tipping.

  17. thebarnacle

    I once had a gold-medal winning US Olympic runner drive me in an Uber in SF. He was now a 70 year-old man without many job options. It seemed surreal to me. (I looked him up after to see if it was true, and the pictures indicated it was him)

  18. Tim

    This was just sent out by CB Insights this afternoon. It not only reinforces Mr. Horan’s premise entirely it also points out that the idea that self-driving cars will somehow bail Uber out is completely antithetical to it’s original pitch to investors.

    Also, as an aside, I rode back on the train from NY to Boston a few weeks ago with a 27 year-old Uber interface programmer (a Stanford EE) who was so full of herself she was about to burst. Couldn’t shut up about the vineyards and mansions she was going to buy when she was able to cash her stock options. She also finished off the better part of a quart of Jack Daniels she had in her bag in three hours which all indicated to me that maybe she wasn’t so sure about the Uber dream herself.

    If you believe this, this and this, Uber’s strategy makes total sense

    So everyone keeps talking about Uber’s desire to move to autonomous vehicles because when that happens, the company becomes this crazy cash and profit generating volcano.

    And Uber has convinced folks of this narrative. They acquired an AI company this week called Geometric Intelligence which should help them advance these ambitions. And have generally been getting slightly more active doing acquisitions and investments.

    But when (and if) this autonomous future happens, doesn’t that break the two-sided network model that makes Uber so compelling? Uber, which has raised $12.5B+, is said to be “capital efficient” cuz it sits between drivers (who incur the costs for the cars, insurance, gas, parking, traffic tickets, etc) and passengers who just summon it on their mobile phone. They just make sure supply & demand are right and then bam – mo’ money, mo’ money, mo’ money.

    When autonomous happens, won’t they have to buy and maintain a bunch of these autonomous vehicles (gas, insurance, cost of vehicle, etc) and also have real estate to park them? When I hear people talk about this future, it’s generally a bunch of hand wavy nonsense built on a whole lot of assumptions:
    • car ownership in the future will look different
    • they’ll be so efficient as a result of driverless that it offsets these costs
    • the marginal cost of a new rider will be blah blah blah
    Then, of course, let’s not forget that driverless vehicles in cities which is where Uber does a lot of biz are a big unknown.

    So all of you smart folks on this newsletter — what am I missing? Why is autonomous so good for Uber?

  19. oho

    ‘Why is autonomous so good for Uber?’

    presumably individuals would buy the autonomous cars (presumably embedded w/Uber IP) then use the Uber platform to rent out their robocar. ie:

    You take your Robocar to the office. After you get dropped off, your Robocar Ubers itself during the workday. Goes offline fetches the kids. Picks you up, takes you home.

    No way, any sane person at Uber HQ would imagine that Uber would own/operate a fleet of Uber robocars.

    1. Lune

      Yep. This is their plan. People would buy driverless cars for themselves, and release them to uber when they’re not using them.

      The problem is, who wants to risk damaging their private car with riders you don’t know and will never see? At least now Uber drivers have a minimal option to refuse a passenger or kick them out if they become too dangerous.

      And the first time a passenger hops in your car at 4:30pm and takes a 50 mile journey, stranding you at your office until 7, or the first time your car returns from a Saturday night of driving drunken revelers and you find vomit in the seats, is likely the last time you or your friends will ever let your car drive for Uber.

  20. rusti

    How incredibly annoying that the author clips a comment of mine from another thread and posts it in a way to make it look as though I’m a shill for Uber. Even more annoying considering I enjoyed this series and linked it to others.

    In order to efficient serve mass market demand, real world transportation companies require vehicles that can be intensively utilized 7 days a week. The car in your driveway cannot do that. It might be able to handle a couple taxi rides on Saturday night, but Uber (or any other taxi operator) needs a huge fleet of dedicated vehicles and drivers to serve the other 98% of demand.

    The sentence you clipped out was in a thread specifically about the perspective of automakers, who I was saying are advertising their intention to become “mobility” companies by virtue of meeting the requirement of having a huge fleet of dedicated vehicles. I didn’t opine on whether they would actually be able to accomplish this in practice other than saying Uber probably can’t, and I find it absolutely insulting that you clipped my comment and then made essentially the same point while calling it “nonsense” when I said it.

    Here is my comment in its entirety:

    The idea is that there will be fleet operators who save money on economies of scale. Most cars are parked something like 98% of the time, representing a huge lost opportunity cost. Car owners are generally pretty lousy about maintenance, which fleet operators could be in a much better position to handle and can presumably benefit from having lots of vehicles of the same model. In big urban centers parking (both at home and around the city) are major costs that can be reduced with on-call transport. These are the sorts of reasons you’ll see in auto manufacturers like Ford’s digressions about becoming “mobility” companies rather than vehicle manufacturers.

    As someone who primarily commutes by bicycle and doesn’t own a car, I’m not wholly opposed to that line of thinking for urban areas but think that it should only be supplementary to robust public transport.

    and from later in the thread:

    It’s not too difficult of an engineering problem, but putting the infrastructure in place and making a business case to reach a critical mass of users is far from trivial. I think the author makes a compelling case that Uber isn’t doing the latter.

  21. kuda

    I can verify the point asserted under “Do taxi drivers focus on peak demand in order to maximize earnings?”. I’ve been a cabbie for 9 years, in a city where Uber started operating relatively recently (around 2 years ago), & work for the most popular (pre Uber) taxi company in town. Before Uber was here we had enough business that I mostly worked until my take home money would be around $20/hr. The cabs are leased for 12 hours, & I’d start at 4pm & typically hit that goal after around 8 hours (on weeknights). It’s pretty easy to do the mental math when it’s just a flat fee to the company plus predictable fuel costs, but I’d gross ~ $275 & be off by midnight, & avoid having to deal with the customers who stayed at the bar til the last minute & are desperate, impatient & hammered, or the sitting around waiting in a ghost town that happens after everything’s closed.
    The thing is there’s two ways to look at it: If you project those numbers out over a full 12 hour shift vs 8 hours (as my girlfriend likes to do) then over 90% of that money is going straight in my pocket & is literally an extra $100, it’s a neoclassical no-brainer. But in reality, that $20/hr only jumps to around $24, & totally depends on whether you maintain a semi busy pace during those hours. Not worth it when I wasn’t desperate, especially when you can expect to go back the next night & repeat.

  22. T

    Great set of articles, thank you.

    If anyone is interested in what an established non – subsidized ride sharing service looks like, google ‘BlaBlaCar’ and search for rides in it’s longest running market, France.

    If your start or end point is not an urban centre or airport, your options are drastically limited as there are no ‘incidental’ drivers offering service (I.E. drivers who will be covering this route with or without you).

    In addition, the natural market price for the ride is only enough to subsidize the cost of the trip for the driver, not enough to turn a profit or replace full time employment.

    This is what trying to use Uber in 5 – 10 years will look like for rural users.

Comments are closed.