No, Uber didn't fail anyone. To the extent there's a failure somewhere, it's the monopoly the municipality holds on taxi medallions, which artificially limits supply of taxis overall.

Municipalities and states hold "monopolies" on licenses of all sorts; the argument that says medallions are bad public policy because they're monopolistic applies equally to many other forms of licensure, for instance zoning.

While this is true, what makes taxi medallions so unique is the price discovery. This is one of those rare things where you can see, quite unambiguously, the direct effect of the monopolistic behavior. Other things like zoning tend to be very complex and ambiguous, but taxi medallions are simple. The effect? NYC taxi medallion prices have grown by 29% annually over 30 years. As someone in the investment world, that might be the single best performing asset class I can think of, all for the privilege of taking someone from point A to point B. I noticed airlines don't need medallions - how has that worked out?

http://www.theatlantic.com/business/archive/2011/10/the-amaz...

Did you just suggest that it's easier to start a commercial airline than it is to get a taxi medallion? Have you ever paid attention to the drama about who gets what gates at which airports? Or watched local political battles over the addition of new runways at an airport?

No. One's supply limited by the government, the other's is not. At the end of the day they both just transport people from point A to point B. Owners of planes have gone bankrupt 1,000 times since 1980, while owners of taxis have the single best returning asset I can think of. Do you really think government meddling (or medallioning, heh heh) hasn't been about 85% responsible for this otherwise incomprehensible result?

The supply of available airport gates is absolutely limited by governments.

Well this is going nowhere. I don't know why I try sometimes.

> Do you really think government meddling (or medallioning, heh heh) hasn't been about 85% responsible for this otherwise incomprehensible result?

Its 100% responsible, from one perspective, because "bankruptcy" is one form of government "meddling" in economic markets.

Where do you get the 29% annual growth from? The article you linked to states 1,900% increase over 30 years. That works out to about 10.5% annual growth. That is still a decent increase, but no longer that extra-ordinary.

> NYC taxi medallion prices have grown by 29% annually over 30 years.

I agree that NYC has not increased the number of medallions in a manner commensurate with its growth since 1970. But this is like referring to the sale prices of commercial condos - you don't need to own a storefront in order to have a retail store (most business owners don't). Likewise, you don't need to own a medallion, just have access to one (ie, rent it, which is what most taxi drivers in NYC do).

Yes, yes it does.

If Uber could help usher in the end of some of the other government monopolies, we'd be looking at some huge steps forward.

I think it's time to reconsider a lot of state licensing regimes. But the fact that they're licensing regimes isn't what makes them bad. Licensing regimes are a perfectly reasonable way for the state to control behavior that involves negative externalities. I think almost everyone agrees that governments shouldn't allow factories to be created in residential areas, or allow drivers to drive without liability insurance.

Careful: ancaps, hardly an Internet endangered species, definitely don't agree with this.

And we don't necessarily have to take every single view seriously. The vast majority of people don't want a massive polluting factory built next to their single-family home. The fact that a few people disagree doesn't mean that every project get bogged down in a discussion of the legitimacy of zoning as a concept.

Ancaps simply have more efficient ways of keeping the polluting factories away from homes than existing zoning processes can achieve.

>I think almost everyone agrees that governments shouldn't allow factories to be created in residential areas, or allow drivers to drive without liability insurance.

Zoning is not needed for the former.

I think it only makes sense to argue on a case by case basis. Saying one unreasonable practice is ok because there are other unreasonable practices is hardly constructive.

For example there is a fixed limit on available land, therefore it makes sense to regulate its use for the common good. The problem with licensing taxis isn't that they are licensed, it's that the number of licenses is fixed arbitrarily without regard to the actual need for taxis.

So you're saying there is a valid public policy case for limiting the number of taxis, and you just think San Francisco got the number wrong? (For what it's worth: that's what I think too).

>the argument that says medallions are bad public policy because they're monopolistic applies equally to many other forms of licensure, for instance zoning.

True; many of those other forms of licensure are also bad.

Zoning is not a fair comparison. Land use has very different kinds of externalities than for-hire transportation. If I build a munitions plant next to your house that impacts you in ways that my getting paid to drive someone from A to B does not.

Two responses.

First, land use has different externalities, but for-hire cars still have obvious externalities. And because of the civic role that taxis have assumed over the last 50 years, decisions about the management of the taxi market have still more externalities.

Second, my point wasn't that land use and taxis have the same issues, but rather that the logic that taxi medallions were bad public policy because "monopoly" was flimsy. Which, it is.

> for-hire cars still have obvious externalities

Like what? Note that, for example, "They take up space on the road and pollute the air" doesn't count because not-for-hire cars do the exact same thing. It has to be an externality that is caused by the fact that the ride is for-hire. There may be such externalities, but I think they are far from obvious.

> the logic that taxi medallions were bad public policy because "monopoly" was flimsy

It's not the medallions per se that are the bad public policy, it's artificially limiting the supply (and then compounding the problem by making them transferrable). That's what the OP (almost surely) meant by "monopoly". Zoning also artificially constrains the supply of land available for certain uses, but that is clearly justifiable because of the (actually obvious) externalities of land use. The externalities of paying someone for a ride are far less clear.

>Like what? Note that, for example, "They take up space on the road and pollute the air" doesn't count because not-for-hire cars do the exact same thing. It has to be an externality that is caused by the fact that the ride is for-hire. There may be such externalities, but I think they are far from obvious.

More specifically, the externality argument about taxis is that they take up space while hunting for fares, while private cars only drive when actually being used to get somewhere.

I agree though, that the point is irrelevant to Uber, since it's strictly a by-request car service that is not hailed from the street, and so doesn't cause the congestion that typical fare hunters would.

Not necessarily a failure. One of the purposes of medallions is to control the number of taxis on the road. It's a conscious choice, and the fallout is accepted in exchange for less taxi congestion.

On the other hand, the limit on taxi medallions protects a common resource, room on the streets. Maybe the balance isn't right, but you're going to need some artificial limitation or else you're going to have a classic tragedy of the commons.

There does not need to be an "artificial limitation" to protect every 'commons'.[1] Many have made the case that services like Uber and Lyft actually free up street space, and allow more people to get to work faster, because less parking spaces are required, and these services usually allow ride-sharing, which is banned (or effectively prevented) by many cities which license taxis and taxi companies.

[1] http://en.wikipedia.org/wiki/Elinor_Ostrom

Thanks for the link to Elinor Ostrom.

It seems like what she was saying is that if the people who own and use the common resource have the ability to create rules and enforce them, then they can avoid a tragedy of the commons. I'd totally agree with that, you need some way to govern the resource.

About freeing up street space, I'd be curious to see some kind of statistics about how many lyft drivers were headed that way anyway and just pick up passengers vs. how many are providing a taxi-like service. Is there a way in the lyft or uber software as a driver to say that you're only looking for passengers going from one location to another location?

It seems like slugging is more likely to provide those benefits.

Using ownership of a resource to give an incentive for preservation and long-term thinking has its roots at least as far back as Armen A. Alchian.[1] Ostrom's achievement was finding that many cultures had alternative solutions, and that these solutions were very effective.

I am sorry to say that I do not have the statistics that you are looking for.

[1] http://www.econlib.org/library/Enc/PropertyRights.html

This isn't necessarily the case though. I am in no way going to argue that the "tragedy of the commons" is a real phenomenon. But their is a natural counterbalance to the streets filling up with taxis.

People operate taxis to make money. If the number of taxis becomes such that the roads fill with taxis, then the number or taxis should lower itself for two reasons. A.) If the taxis can't get anywhere either then people won't pay for them. B.) The ability of any single taxi to turn a positive profit will approach 0.

So I'm a bit confused by your phrasing. Are you saying that the tragedy of the commons isn't something that actually happens?

I agree with your point that there are natural pressures that would keep the number of taxis down if the artificial limitation was removed. It's just that I don't think that the natural equilibrium point financially is going to be the optimal level in terms of getting people where they want to go efficiently. In my mind, the entire transportation system is useful to provide a service to individuals and other sectors of the economy, and if it's not doing that well it doesn't matter much if the people working in transportation are at a natural equilibrium.

I'm also not saying that the medallion system is the right way to handle the situation. A taxi or car on the road uses up some common, limited resource (space on the road). Maybe you'd have more luck charging for that resource more directly, like a congestion charge?

I think you misunderstand what surge pricing _is_. Part of the reason Uber can't recruit more drivers is that the price isn't worth it for them; and for all the times where they don't _need_ more drivers, this is completely fine. When there is a sudden need for more drivers, they can quickly recruit more by raising the prices...this is exactly what surge pricing is.

Modeless is actually in agreement with you, as I understand it. The distinction they're making is a bit of a fine one, but they're basically saying Uber ought to allocate all of the increase from the surge pricing to drivers rather than taking a cut. By not taking a cut Uber would more efficiently incentivize drivers to get out and take rides, and more rapidly affect an increase in supply and the corresponding price rebalancing. This would also have the happy side effect of making surge pricing wholly defensible, since Uber sees no benefit from it.

Whether or not that is feasible under Uber's business model is, of course, a totally different question.

What incentive is there for Uber to do this? Remember, they're a for-profit company, not a public service. If Uber didn't take a surge pricing cut, then the only reason they have to use surge pricing is to maximize the number of rides on the road. And that's certainly something they could do, as more rides == more money. But they'd be leaving money on the table that way, and as a for-profit company, there's very little incentive to do that. You stated that it would make surge pricing more defensible, which is basically the only reasonable incentive I can think of, but obviously Uber believes that it's in their interests to make more money than to have a stronger moral defense for surge pricing.

So I guess the real question is, for those of you who believe Uber should behave this way (and, w4, I recognize that you're merely explaining modeless's argument, not claiming you agree with it), do you believe Uber should be expected to behave this way, given that they are a for-profit company, or do you believe that it is merely the *right thing to do (for some ambiguously-defined moralistic definition of "right")? I suspect that what you really believe is that it's what you personally think should happen, because it would result in lower surge prices for the same supply, and you would like to pay less money. And that's not a very compelling argument.

Well, I mean, I do see the argument that it's a more efficient outcome.

Imagine for a moment that surge pricing were kept the same. The only difference is that, now, Uber takes no cut whatsoever. This would result in greater profits for Uber's drivers during a surge, better incentivizing drivers to get out on the street and start accepting fares. This increase in the incentive would, one can assume, result in an even greater increase in supply during a price surge than is produced under the current system, and more effectively address the supply and demand imbalance that creates the need for surge pricing in the first place.

Now, does Uber have a moral obligation to do this? Fuck no. However, it's certainly disingenuous to suggest that surge pricing is just all about addressing supply and demand imbalances.

(I'm not saying you say that's what it's all about; you don't. That is, however, what wutbrodo and a lot of other posters seem to be suggesting).

EDIT: As an aside, Uber's incentive (besides defensibility) for not taking a cut of the surge is clear: by making the incentive for drivers to get on the road as powerful as possible, Uber would all but guarantee that its users can get a ride at any time, any day, in any conditions. That would solidify Uber's primacy over other, less reliable transportation options. However, given that their product is already so superior to the alternatives, it doesn't seem to be the case that they need to look at doing anything like that for the time being.

I Shouldve added a quote since it's a bit unclear, but I was responding to the claim that surge pricing represent a "failure" go recruit enough drivers on ubers part (and thus that it's not fair for them to make more in those situations).

It is a 'please get out of bed' payment, to try and manage shortages of drivers.

If they could do this without tacking it onto price, by making an explicit 'please get out of bed' payment from uber to a fresh driver on their first fare during a shortage, they may be able to create the same economic incentive to the driver without being seen to price gouge. Of course, this would involve raising prices slightly overall to pay for it, but you get rid of the swings in price.

At first glance you seem to have a reasonable point, but it actually doesn't work out. Surge pricing doesn't merely raise supply to meet demand, it also suppresses demand to meet supply. In the latter case, there are fewer rides going around than there would with an infinite supply, so if Uber doesn't take a cut of the surge pricing themselves, then they're missing out on money they would have had without the suppressed demand.

Perhaps you think they deserve to make less money, but as wutbrodo pointed out, the lack of supply at these times does not mean Uber has failed at recruiting drivers. After all, having extra drivers sitting around idle is not helpful to anyone (not to Uber, not to the drivers themselves, and not to the passengers). The only mechanism Uber has for putting more drivers on the road when demand increases is to adjust prices. Theoretically they could merely reduce their cut to put more drivers on the road without increasing the cost to passengers, but there's no incentive for them to do that.

Ultimately, what you said would make sense if drivers were actually Uber employees that operated at the direction of Uber. In that scenario, Uber could order more drivers onto the road without surge pricing. But that's not the case, drivers are voluntary contractors that make the decision themselves about whether to start working. The only tool Uber has to influence how many drivers start working is to increase their pay, and the only way to do that* is to turn on surge pricing. And since surge pricing acts to reduce demand in addition to increasing supply, the only way Uber itself has an incentive to use surge pricing is if Uber itself gets more money that way (which means taking a cut of the surge-boosted price instead of the original price).

I'll grant you that if Uber was a regulated service operating on behalf of the public, then perhaps it makes sense to penalize them for not adequately meeting demand. And in such a scenario, surge pricing can't be justified anyway, so the natural penalty of not filling as many rides as there is demand for would suffice. But Uber is a for-profit company, offering what is essentially a luxury good, and the only arguments that really have any bearing on Uber's behavior are economic ones.

it's possible that there are times when it makes sense to do that, e.g. when surge pricing is already on and raising it further will suppress demand enough that they won't make as much money. It's possible that reducing their cut could increase driver supply enough to make up for the difference (due to more rides happening). But I don't know if this is ever true in practice.

*short of reducing their cut, as mentioned previously, but there's no incentive to do this instead of surge pricing

From an economics perspective there was money left on the table an Uber is taking it, and that's the end of it. However, economics will also tell you that this doesn't seem to be an equilibrium situation, and therefore competitors will start entering the market until Uber's profits are razor thin, unless the regulators get in the way and screw the whole pricing scheme up.

Theres no point recruiting drivers for 1-2 hour surges.

But there is a point in incentivizing a larger percentage of the total number of drivers in a given area into accepting rides during 1-2 hour surges.

Start an Uber competitor that only gives surge pricing charges to drivers. Your rates should be lower.

As long as monopoly abuses are prevented, this problem will work itself out.

And that's caused by lack of competition.

Wow.