Hacker News Reader

Comment by emodendroket | Hacker News Reader

The overhead is not as large as you're imagining. http://www.nakedcapitalism.com/2016/11/can-uber-ever-deliver...

> By contrast, in the hundred years since the first motorized taxi, there has been no evidence of significant scale economies in the urban car service industry. That explains why successful operators never expanded to other cities and why there was no natural tendency towards concentration in individual markets. Drivers, vehicles and fuel account for 85% of urban car service costs. None of these costs decline significantly as companies grow. As the P&L data above demonstrates, Uber has not discovered a magical new way to drive down unit costs.

View on HN

That explains why successful operators never expanded to other cities

Sweden's largest taxi company currently operates in 50+ cities in Sweden, Norway and Denmark. So "never" is probably the wrong word.

Sweden famously deregulated their taxi industry. American taxi regulations are awful--I can't compare them to other countries, but they're terrible in an absolute sense, and they vary widely by city. In New York, they even change depending on where you are and what color the taxi is (yellow ones and green ones have different rules). So all these city-by-city variations become a big diseconomy of scale.

The reason for the US' horrible state of affairs: corruption. Ever since regulation of the taxi industry started, the 'players' in each city decided to use this artificial constraint to enrich themselves. Medallions (the right to drive a cab in a city) were doled out to the politically connected, who, in turn, rented them out to the actual workers (drivers) for a tidy sum. It was a great scam.

If there's one positive thing about Uber, it's that they brought an end to this medallion monopoly.

Come on; that's too simple. Unregulated gypsy cabs present a real risk to people.

The medallion system helps create gypsy cabs. If starting a legal and licensed taxi service just required filing the same type of paperwork as any other similar company, and becoming a regulated taxi driver was more like becoming a truck driver you'd probably see a lot less gypsy cabs.

In NYC, that's exactly how it works, just pass a simple test and do some paperwork, and you can drive a cab. Medallions only restrict cabs that are permitted to take street hails in lower manhattan -- as long as you don't do that, there's no limitation on numbers of cars.

Uber and related apps are a bit overrated in terms of their "innovation."

You see, Uber-like "radio dispatch" services have existed in the NYC area for decades, but they used regular phone voice calls, not apps, which was slightly less convenient, I guess, but barely. It was not really qualitatively different IMHO (you waited about the same amount of time, and have about the same amount of certainty about the fare and whether they'll show up.). In fact, it could arguably be easier, as you didn't have to sign up for an account, download an app, set up payment method ahead of time, you just dialed a phone number

The only real innovation ridesharing apps have brought IMHO, are carpooling options like UberPOOL, Via, etc. This was not as practical to do before apps that could automate the dispatching and matching involved

I suspect that you have never actually used a phone-dispatch car service in NYC. You have no visibility into their ETA. Call back and ask... the answer is always the same "5 minutes".

I've used it tons of times. I grew up around here. You're right, you don't have much visibility into the ETA. I THOUGHT that was an advantage of Uber, but it's not -- I've sometimes called Uber and they blatantly lie, bait-and-switching about ETAs (eg, 3 min before I click, then it becomes 12min). Also, dealing with the ETA is easy. Wait 6 minutes, if they don't show up, call a different car service. The problem will take car of itself.

The "best" part about Uber is they penalize you for waiting too long to cancel your Uber so if the guy takes three times as long as promised and you cancel then you are charged the cancelation fee.

I'd argue there's also an interest in not filling the streets with excess cab capacity.

Yep. San Francisco has been utterly inundated. The last estimate I saw was ~48,000 more vehicles from the app services. Most of those are additional, because they are coming in to the city to drive (e.g., the drivers don't live here).

Important to keep in mind that that is not only a lot of cars for SF, but also that they aren't just driving a bit and parking. They're driving and driving (and driving and driving), when they're not blocking traffic picking up or dropping off.

In urban areas there are good reasons for restrictions on numbers.

Absolutely agree. Since Uber and friends, congestion has become worse in NYC.

However, a simple toll would be fairer, as it would treat all cars equally, regardless of whether they are or are not taxis

I don't think it's "fairer" to treat personal and commercial traffic the same.

Having to comply with tons of different rules on a per-state or even per-municipality basis is characteristic of the entire US system, though.

American taxi regulations are awful

But that is a quirk of the current American taxi system, not something inherent to the industry.

It's actually pretty widespread. If you think they're bad here, taxis are even more entrenched in Europe.

One imagines the cost of complying with labor regulation in those countries makes for better economies of scale. I agree that "seldom in the U.S." is more precise, though.

The population of these countries are comparable to a Metropolitan area in the U.S.

It's basically correct in the US case.

robot driven electric powered vehicles that operate 24/7 (minus maintenance and recharge time) are a major way to gain efficiencies in drivers/vehicles/fuel. the nature of an automated ride hailing service, also, gains efficiencies in densely populated areas. it doesn't quite reach economy-of-scale level stuff the way a manufacturing business does, but there are definite efficiencies that Uber (and their competitors) are aggressively pursuing.

The parent comment on which this thread is based is arguing that Uber's elimination of top-heavy management in traditional taxi companies is so great that they should be able to profit even if they never achieve self-driving cars. Clearly if they can entirely eliminate the cost of labor it complicates the question.

That seems like a strange conclusion considering Uber doesn't even pay for the vehicles and fuel.

UberPool is one pointed example of scale economics working. Another would be Uber's brand encouraging more riders and the ability to scale into smaller markets that taxis normally couldn't service.

> That seems like a strange conclusion considering Uber doesn't even pay for the vehicles and fuel.

Well, they do, indirectly, in that that is part of the money the driver is paid. Yes, one way they've improved the numbers is squeezing their drivers more, but they can't do that forever. They're already encountering a lot of pushback.

I question the claim that Uber is serving a bunch of small markets that traditional livery services do not. Do you have an example?

Not OP, but I can say from personal observation/experience that while not 'small', Uber has increased taxi use in the general sense in Texas cities that are so car-centric and-or sprawl-like that traditional taxi services were never really popular (Houston, Austin, Dallas, etc) - having more drivers available in an ad-hoc manner has made the dispatch time much more reasonable (minutes vs a hit-or-miss scheduling ordeal of ~1h) and so using taxis (e.g. Uber) as a viable means of transportation, especially for things like social events, is much more common than it was before. I suspect the same is likely in similar places where 'everyone has a car' and the geography is very spread out.

That's the kind of place I live but there are multiple livery services operating here too.

For example, Boca Raton in Florida is a wealthy area that had no taxi service until Uber. You literally had to get a limousine to go somewhere. Now its simple and cheap to get an Uber.

When you say a "limousine" are you not just describing a livery service where you call a dispatcher and they send a car?

The point is that you can get a $5-10 Uber ride instead of before where you'd spend $100+ on a limousine ride that you need to arrange well ahead of time. It wasn't a realistic option before except for special occasions.

> the ability to scale into smaller markets that taxis normally couldn't service.

Go on.

Mountain towns in Colorado, for example. Uber and Lyft operate in the major mountain towns where there is no taxi service at all.

Simple economics argues you are wrong about the costs aspect of things here. Do you think drivers are simply doing public service? Perhaps they have special gas pumps that are 30% cheaper?

Reply on news.ycombinator.com