Hacker News Reader

Comment by potatolicious | Hacker News Reader

What your driver was talking about may not be a bug/math error like stated here.

After rolling out guaranteed whole-ride pricing (as opposed to metering) Uber is allegedly testing algorithms to bid different prices for rides to both driver and passenger. i.e., passenger sees a fare of $X, while driver sees a fare of $Y, where presumably $Y < $X.

Passengers' fares can be "optimized" by estimating a specific passenger-trip's price ceiling, and driver payments can be "optimized" by estimating the driver-trip's price floor.

To be clear, I find this entire concept incredibly abhorrent, but apparently this is being trialed.

[1] https://nakedsecurity.sophos.com/2017/04/10/uber-showing-dri...

View on HN

Bloomberg's Matt Levine on this: https://www.bloomberg.com/view/articles/2017-05-22/fitbit-qu...

In a nutshell, uber is changing their business from charging a commission to acting more like a retailer: buy low (from drivers) and sell high (to passengers). The problem is that normally the two far ends of that transaction only ever interact with the middleman, but here they're literally sitting in the same car. Hard for both the buyer and seller not to feel (justifiably) ripped off when they can actually compare notes. Especially if even after their margin Uber is still taking a percentage of the driver's pay.

When I read about stuff like this and also how Uber is losing a lot of money, I have to wonder if maybe there's a fundamental problem with their business model.

Their value isn't in the driver business, because eventually they'll just have to pay for the marginal cost of damage to the car. If they can optimize pricing between locations better than others, they'll win in the driverless world. (This drives the value, not the spread between what they pay and collect)

That makes sense. I think that also explains why they started their own driverless car program. Whatever company makes the first viable driverless cars, they may decide that rather than selling the cars, they will own them all and operate a ride-sharing service with them. Uber would be screwed.

There is; they are hemorrhaging money, and only manage to stay afloat because of the billions of dollars being shoveled into the furnace by investors and nation-states. In a way, venture capital is subsidizing your rides.

They're not insane, though. They're buying market share and betting that they can pull self-driving cars into their fleets while they still have the market on lock. Without having to pay drivers or deal with their whiny attempts to enforce employment laws (against an app! Crazy,) they see themselves being able to print money hand over foot.

This is how you get away with being a dirtbag, in any situation. Move very quickly, and get to where you're going before anyone manages to claw you back far enough. Possession seems to be 100% of the law, when it comes to companies in this country.

Let's be clear that "they're buying market share and betting that they can pull self-driving cars into their fleets" thing is revisionist history. Nobody who invested billions in Uber years ago was told, "Our plan is to burn your money and win driverless cars." They didn't even have a driverless car program until about two years ago.

Uber thought that they had a viable business model based on human drivers, and that their markets would settle down into profitability once they established market share. They were wrong. They thought that they could successfully expand into a logistics business -- that was a big element of their redesign, which only landed 6 months ago, for god's sake. Now you never hear about logistics (because the only people who might conceivably want Uber-for-packages, Amazon, just went ahead and built the service in-house).

So now they're saying that they're a big bet on driverless cars.

I can't find the post, but there was a 'UBER finance guy' (self-claimed) on HN a few weeks back that 'spilled the beans' and quit on what Uber is actually: a sub-prime auto loan company. Again, this is my recollection, take it with salt. It goes like:

Driver (D): I want to drive for Uber!

Uber (U): Sure! Do you have a car that is safe and less than 5 years old? (I have no idea what the car reqs. are)

D: No

U: Ok, well, we can loan you one! Just sign this!

D: Great! I signed! What are the loan repayments like?

U: Well, you owe us, per the 3 year contract, $100/week (I have no idea what the actual loan terms are like, but 100 is a nice round number so lets use that)

D: Ok! That's not bad!

U: It gets better! We'll just auto-deduct the money you owe us from your driver's account, after that, all the money is yours! You don't have to worry about any of it, just drive for us, we'll take care of it.

D: Cool! What happens when I get sick or something and can't drive?

U: Well, then you still owe us the money, plus a 10% fee on that week's loan payment (again, I have no idea what the loan agreement is like, YMMV) that we'll automatically tack on to the next week's loan payment.

-A week goes by-

D: Oh man, the weather was great this week, not a lot of people needed me to drive them around! I only made $90!

U: Ok, well, per the loan agreement, you owe us $10 more then (100-90=10), plus the 10% late fee, making your next week's payment $100 + $10 + (0.1 x $100) = $120.

D: Oh man! I better hope that the weather is bad so people want to take an Uber that I am driving!

-Another week goes by-

D: The weather was really rainy, but even still, I only got notifications to pick up $90 worth of people! Not again!

U: Well, now you owe us $120 - $90 = $30 plus the 10% late fee on the $120, making your next week's payment $100 + $30 + (0.1 x $120) = $142.

D: Well, darn, I thought a lot of people wanted to take Ubers! All my friends take them! But I don't get enough notifications to pick people up from Uber!

U: We know. We are the people that are giving you a loan and simultaneously are the ones that control how much of that loan you can pay back because we are the ones that tell you how many people you pick up and we also set the 'surge' rates and regular rates that we tell you. Also, we have something of a history of acting in morally reprehensible way without shame.

D: Wait, what did you say?!

U: Nothing! Nothing at all! Get back to driving! That car we bought isn't going to pay itself off, you are!

I'm as negative as the next guy on Uber, and that's a great narrative, but they are losing vast quantities of money. Any dastardly deed that they do has to square itself with the fact that they are losing vast quantities of money.

A predatory auto-loan company is not a $70B business.

No, but the loan sideline allows to classify the drivers as contractors.

And there's externalities. No properly run bank would issue a car loan to an Uber driver, they'd run the math for the applicant and find it isn't going to work out. But Uber with its infinite pockets does not have to consider profitability, and eventually the driver has to declare bankruptcy and becomes a burden on the public pocket. It's a disgrace.

True, but you can make a ton of money when you design a system where you have no labor costs (drivers make no net income). It's not $70 billion, but it is on the order of billions carried out for a decade. Perhaps that is why Uber is so ... nasty. Even with their scummy tactics and greyball and whatnot, they still can't make ends meet.

Also, what is stopping them, or anyone else, from running the same scam with other large capital expenditures? AirBnB could do this with houses, DoorDash could do this with heavy industrial kitchen equipment, etc. It's evil as hell, but Wall Street backed VCs have no issues finding people to do it.

It's almost like transport is a mature industry, with lots of competition, decades if not centuries of research on cost reduction, and running on thin margins!

I said this before, Uber should seriously consider closing down its operation in other countries. It is expanding virtually exponentially (they have a huge engineering team) for the type of company they are running. They need to focus on the U.S. market. They are losing way too much money everywhere combiend.

Microsoft did this with the Xbox One (almost all the extra features they advertised were US only) and it completely destroyed their market share, making the PS4 sell 2-to-1. WhatsApp went the other way (the US doesn't matter, you want the global market) and it got them a $19 billion sticker price. The US is a market, not the market.

That's a very bold statement to make with zero knowledge of the financials.

Why not just lease... Are the drivers really falling for this en masse?

Leases tend to be low mileage which is not conducive to taxi driving. Yes lots of people do this.

As far as I know, Uber car loans are fairly different from typical subprime car leases in that you can get out of them with little or no penalty.

Thanks for the data! I had heard the lease terms were for 3 years, no exceptions and no getting out of them. Still, they may change from location to location and over time. As always, read the contract!

My original source for the cancellation terms was an Uber driver, and I just looked on Uber's web site, which more or less seems to confirm this:

  "Flexible leases from Xchange Leasing, LLC are potentially up 
   to 36 months with the flexibility to return the vehicle as 
   early as 30 days after the first payment due date with a    
   payment of a $250 disposition fee. Not all applicants will 
   qualify for this program. Certain items present on your 
   credit report, and other information, may cause you to be 
   excluded from this offer, including without limitation, a 
   repossessed vehicle within the last 12 months, a pending 
   bankruptcy, or a discharged bankruptcy within the last 12 
   months."

https://www.uber.com/drive/vehicle-solutions/leasing/

https://news.ycombinator.com/item?id=13764994 -> https://www.reddit.com/r/cars/comments/5wrdk2/car_that_uber_...

It's their xchangleasing leasing program. It's basically a sub prime car loan like from the housing bubble days.

if Uber cuts it's rates they still have same payment but less fare money each month. They did this to a lot of people and it's well known within the company that a lot of Uber black drivers were screwed when this happened.

Bingo! Thank you for the links!

Honestly though, years ago when I heard on the radio that google gave Uber 1 billion in funding I loudly exclaimed to everyone in my tiny office "ah... they want to invest in self driving cars through Uber!"

So yeah, there were plenty of people who were looking at that. I think they just didn't know how expensive that long con would wind up being

So, to be clear, you think that when Google gave Uber 250M in funding (not 1B), in 2013, when Google had without-a-doubt the most advanced driverless car program in the world, and Uber was roughly two years away from beginning any driverless car program, Google made its investment in order to get into driverless cars?

Huh.

People in the investment community were for sure talking about Uber and driverless cars in 2013. It was clear at the time Uber was going to have to make a buy vs. build decision. But it wasn't out of the question that Uber would become an early adopter and major customer of Google self-driving vehicles.

https://techcrunch.com/2013/08/25/uberauto/

https://www.quora.com/Uber-in-2013-How-can-Uber-survive-driv...

https://www.forbes.com/sites/roberthof/2013/08/23/as-google-...

I agree that one possible synergy at the time between Google and Uber was that Uber might use Google's driverless car technology.

That's very different from the revisionist rationale that Uber was always themselves making a long term play for driverless vehicles, or that Google got into Uber because Google wanted to invest in driverless vehicles.

The narrative at the time was that Uber was going to become a dominant rides-for-hire business (a real business, not a gaping wound hoping that technology would catch up to them) and at least a player in logistics, and then when driverless cars came along, everything would be even better for Uber.

You'll note that as that narrative was increasingly pounded into the dust, Google never invested further in Uber and relations between them have at this point obviously completely soured.

Uber's first massive round was in August 2013. By that time, everyone knew driverless cars were going to be critically important in the long term and Uber would have to make a build vs. buy decision. Whether Uber was ferrying passengers or goods didn't really impact this.

But yes, the investment thesis has always depended on passenger taxi being a decent business for Uber in the midterm.

I don't think that we have a ton of disagreement here. I agree that in 2013, people were looking ahead to a future of driverless cars, and that it was part of the calculation on Uber. I think at that point the assumption was that Uber would probably buy, not build (they did not, after all, start a serious build program for another couple of years), but Uber (I think wisely) was probably on the fence themselves for a while until the technology got clearer.

The claim at the time was that Uber would ferry both passengers and goods. And we agree that the value Uber was claiming was dependent on being a decent business.

I think that what happened since was they were forced into the "build" side of the build/buy decision when it became increasingly clear that they didn't have a decent business in passenger taxi, and they weren't getting traction in logistics. At that point, if they bought driverless cars, you would be forgiven for asking what value exactly Uber was bringing to any business. Now they have to bank on their differentiator being that they have unique access to valuable driverless car tech, which still strikes me as grossly unlikely.

* They didn't even have a driverless car program until about two years ago.*

Now everybody has one. 33 companies have registered with the California DMV to test. Five more accidents are listed for March and April. Cruise has three; they were rear-ended twice while stopped, and were clipped once by a pickup trying to drive around a double-parked mobile crane. Waymo was rear-ended once, and was forced off the road into a curb once by a vehicle drifting out of lane.

This is progress. Cruise is getting hit when it's not their fault, which is way ahead of their earlier driving record. We're seeing more unusual situations, where the other driver did something weird. Nobody was hurt in any of these.

The auto industry is indicating that the 2020 model year is when self-driving starts to ship. They have to be public about that, because the parts suppliers need to get their production lines ready to make the components in quantity. It's not going to be just Uber. It's going to be all the big automakers.

> against an app!

Should that in some way magically exempt them from hard won (with blood [1]) labor protections?

1. https://en.m.wikipedia.org/wiki/Haymarket_affair

It's astounding to me that they can't make it work based on the convenience of the service. I understand the strategy of wanting to buy market share quickly, and to bet on self-driving cars. That they can't compete on the convenience of the service at a fair price and playing by the rules of the districts they operate in is just astonishingly bad management.

I'd argue they did - I was around San Francisco when Uber was getting off the ground, and the initial service is exactly that: convenient, high-quality, provided professional drivers with commercial insurance.

It was also expensive, and nowadays known as UberBlack, and is a market many times smaller than Uber's ambitions.

UberBlack was a great product with unit economics that would allow for a high-quality product that was a win-win for both drivers and Uber. But it was not a big enough product to justify Uber's stratospheric valuation - hence UberX.

I think we're running into the fundamental truth here that providing high-quality taxi rides at such cut-rate prices is not possible, at least not when a human is at the wheel.

Cutting corners and playing dirty are IMO a manifestation of this fundamental problem.

Absolutely agree. Another part of the problem is the need to become a unicorn at any cost.

Why prefer having a huge valuation (based on thin air) when you can have a more modest but totally profit generating business?

This is a question that many in the Valley should answer honestly.

There is. Uber is predicated on brazen lawlessness and squeezing workers.

Thanks for your link. I hadn't heard of this.

So there will basically be no transparency to anyone involved in the transaction except Uber if this is adopted?

The company seems to be guided by a culture of manipulation - manipulate the drivers with erroneous information, the riders with surge pricing, regulatory agencies with different screen views and now the possible manipulation of the bid ask spread.

Uber just seems rotten to the core.

Information asymmetry at its best, not that there is anything wrong with it.

If Uber charges X for trip, and pays the driver Y for the service, it's obviously a taxi company. As such, I expect authorities to enforce the licensing laws, especially that the public support of Uber starts to diminish.

I completely agree with your points.

It would not be abhorrent if it were the only source of income for Uber, which would basically be earning the bid-ask spread.

Adding fees on top of that, though...

Except they also penalize drivers for not accepting many rides. So if you want to continue driving for Uber, you have to "sell" your driving services at some point, even if the price isn't good for you.

Indeed, and that's an extension of "if it were the only source of income for Uber". If Uber proposed too high a spread, it's Uber that should be penalized.

IOW it's not the spread mechanism that's wrong, it's everything on top of it.

As well they should, otherwise it'd be impossible to get ubers from certain places or at certain times.

Then Uber should pay drivers more in certain places at certain times.

What is it, free market only works for Uber but not for the drivers?

>To be clear, I find this entire concept incredibly abhorrent

It's not clear to me why it's abhorrent. Doesn't every company optimize their vendors and customers separately?

>To be clear, I find this entire concept incredibly abhorrent

why?

Some people have moral frameworks that always make seeking profits immoral. I'm not sure if that's what OP is saying, or merely that showing the driver and the passenger different rates and subtracting the difference (this is an approximation, not exactly what the post said) is shady and bad.

From previous HN discussions my impression is that the primary issue is with this charge being implicit. Uber already takes its money from drivers, and both drivers and users are participating in the program based on the assumption that what user is charged = what driver gets, minus Uber fee. But then it turns out that what user is charged more than Uber tells the driver (and from what it subtracts its fee), while the company also pockets the difference. This feels wrong.

Effective and fair markets rely on transparency, while this is the exact opposite - the deliberate creation of massive information asymmetry.

Worse even, the information asymmetry isn't meant to favor one side of the transaction over the other (say, passengers over drivers), but is meant to favor the exchange (Uber) itself.

Imagine if you went to your stock broker to buy 10 shares of XYZ. Your broker tells you that the market price is $100 per share, but in reality found a seller at $80/share and is pocketing the difference, without disclosing this to either party in the transaction.

Besides the ethical shadiness of this kind of action, functioning markets are predicated on buyers and sellers having sufficient information, and the exchange behaving in consistent and transparent ways. This sort of double-dealing undermines it.

Another poster snidely speculates that I may operate on a moral framework that finds profit-seeking intrinsically objectionable. This seems like a straw man that's intended to erase the difference between unfair market practices and fair market practices.

> Imagine if you went to your stock broker to buy 10 shares of XYZ. Your broker tells you that the market price is $100 per share, but in reality found a seller at $80/share and is pocketing the difference, without disclosing this to either party in the transaction.

That's a poor analogy, because the stock is an exchange value, but the car ride is a use value.

A better analogy would be buying a pear, and getting angry that the fruit stand bought the pear really cheaply, without passing the savings on to you.

So by this they're just throwing "let's see if we can run an arbitrage on the concept of price sensitivity itself" against a wall and see who will fund it?

Yup. Wall Street will like everything about this, so the success or fail of it will be measured by 'can you survive if ONLY Wall Street likes you'.

It would be fun to compile a corpus of economics and business books, distill them down to their major points, then throw them into a dack.com style permutation generator. Optionally, post every one as an "Ask HN," or whatever could be called "Etsy for angel funding."

Isn't this the future? Machine learning for everything? Just wait until some algorithm says 50:50 culture and diversity ratios aren't optimal [1].

[1]:https://m.xkcd.com/1838/

> abhorrent

Is this not a business trying to maximize profit for itself? If you find this abhorrent than you must also despise capitalism because that is all this is. I may not personally like it as a consumer of their product but I find this to be a clever way to maximize profits for a middle man.

Reply on news.ycombinator.com