The Tyranny of Taxi Medallions | Hacker News Reader

The Tyranny of Taxi Medallions

blog.priceonomics.com

232 pointsajju216 comments

Structurally this practice is a lot like the preindustrial precursor of taxation, when governments would sell monopolies on the import or production of things in exchange for a lump sum.

When obsolete government practices persist, they tend to be pretty sinister.

Exactly. A majority of taxi drivers do not own medallions. But decades of being on a "wait list" for the right to buy one and FUD by taxi companies causes a lot of them to dislike ride sharing, when, in reality, it's a huge net win for them.

I never made the connection to selling monopoly import licenses, but that's exactly it.

Honest question, why sinister? What has happened?

Another poster is claiming the unregulated Amsterdam market has been taken over by organised crime. That actually sounds sinister.

It's sinister because it pretends to be altruistic while simultaneously serving the interest of a small faction.

Isn't it sinister because it's serving organised crime? Would you feel it similarly sinister if the market had been cornered by a startup (or three) with some silly name and a pile of VC debt?

When obsolete government practices persist, they tend to be pretty sinister.

Why is this true? Perhaps keeping these practices around, despite them being obsolete, implies ulterior motives?

It's something of a meaningless statement. A long-term government practice that is seen as good is not seen as obsolete.

long-term government practice that is seen as good

I've never heard of one, asides from the practices that define government itself (laws, courts, police, voting, congress etc), but these are more necessary than good.

Separation of Powers isn't necessary for government, but it is a good practice, for example. Would anyone serious call that obsolete?

I agree, it's a good practice, intended to prevent government from assuming a whole bunch of bad practices that (unopposed) it tends to assume. But to take your point to the extreme, democracy, courts, etc are also therefore not necessary functions of government.

I'd argue that they are necessary components of a good government. All the best parts of government limit the scope of government .. that says something.

So take voting. Just saying 'the vote' is way too simplistic. Allowing women to vote? Good, not obsolete. Allowing minorities to vote? Good, not obsolete. Electoral college method? Facilitates sinister behaviour, obsolete. Districting rules that allow for relatively easy gerrymandering? Obsolete, and definitively sinister. Promotional materials encouraging people to exercise their rights by voting? Good, not obsolete.

All the best parts of government limit the scope of government .. that says something.

No it doesn't. That's an overly simplistic slogan. Provision of universal healthcare is a clear and topical example.

Provision of universal healthcare is a clear and topical example.

Come on, you know that's political and there are arguments on both sides. For the record, I don't agree with you here.

I have no problem being simplistic, as long as I'm right. That being said, I take your criticism, and will rephrase part of what I said ..

Necessary components of government (such as many you mentioned) are the parts which define how the government works. These parts didn't start out perfect, and perfecting them is a good thing (i.e. women and minority voting).

Everything in government is political. Western democracy is a decent form of government, but it's not the best - benevolent dictatorship is the best; it just doesn't have checks against it turning non-benevolent.

The problem is where do you draw the line, because it has to be drawn somewhere. You claim that limiting the government's scope is good in all cases, yet this is clearly not true - if we were to take arrest powers away from police for example, society as a whole would suffer.

Limiting scope can be argued all the way down to anarchy, which is clearly not supportable at anything beyond the size of a small conclave.

I have no problem being simplistic, as long as I'm right.

The truth has never been simple in politics, and anyone who maintains that it is, is simply displaying contempt (open or otherwise) for your intellect. As a result, we can't rely on simplistic solutions for politics, and have to be cautious of things that try to make it one-size-fits-all.

I've brought this up in another thread, but over in the UK, due to a lack of property taxes, its possible to purchase a home, yet stil have to pay a yearly ground rent to a private party. The cost to own land in nothing, so there's little incentive to ever sell it, and instead lease it out. People may complain about property taxes in the USA, but it prevents this very practice from happening. It would be better for everyone if the medallion system switched over to a yearly taxation/bidding system. You would lose all these middleman renters.

I don't understand .. why would you purchase a home without the land it rests on? In a free market, perhaps there is some lower price where you would do this, but there is also a higher price where you could purchase the land as well.

Because that's the only type of home on the market!

Well, argumentum's point is that presumably there is some cost at which the ground holder would sell the land.

Like, if the ground rental brings in $10,000 per year to the owner of the title, and I came along and said, "I will pay your $1,000,000 for the title," well... I'm offering you one century's worth of ground-use fees up-front. Presumably all but the most irrational actors would sell their ground.

Of course, but that's nitpicking. The point is that the price is well out of reach of the average Tom.

Well, but why is it out of reach?

I mean, the same guy ("Tom") is going to be giving you money one way or the other. Either in bits and pieces as ground lease, or in lump-sum as purchase.

If you simply want more money than Tom can pay, well, you're out of luck. You don't get any money if you price your tenants out of the market.

If there is an amount of money that Tom can afford to pay in ground leases, then why can't he get a bank loan and make you an attractive offer on the purchase? My earlier example of 100 years rent was extreme to make a point. What if Tom offered you 20 years land rent in an up-front lump sum? Maybe not every land-owner would consider that a good deal, but it's hard to imagine that none of them would. The land-owner gets a present value that exceeds the future value of the rent to them, the bank gets interest, Tom gets synergistic value from now owning not just the house but the land.

I'm not sure adding £1,000,000 to the price of houses is a rational solution to ground rent.

I suspect you'll find the economics are such that most home owners wouldn't be able to afford to buy the land their house is on.

So what happens. Presumably there is a completely parallel market for land. Land owning elites presumably buy and sell land all the time and homes are bought and sold all the time as well. The two rarely intersect? Or maybe those that can buy land can easily buy the house as well? Very confusing. In US effectively the local government and the state "own" and you are just leasing it. There is not parallel land market unless jurisdiction lines get redrawn for some reason or a states secedes from the Union or something like that (you can imagine some negotiations then where the city say, this land with homes brings in $45M in property taxes/year, we wish to sell it for $1.3B cash...) of course this is all hypothetical...just to make a parallel.

Usually it's not a house, but a block of apartments, each of which are owned individually. It's usually cost prohibitive for someone who owns one apartment in a block of twenty to buy out the land for all twenty.

Once there are separate owners (for whatever reason it happened), it's not that easy to arrange that - since the land owner usually does not want to sell.

What if they do sell? Can land owners do whatever they like to the home owners? Like kick them out? Start drilling tunnels through the land, extract oil in the back yard?

Unscrupulous landlords will increase the rent.

Usually the home owner will have a lease that lasts for 99 years (or at least the person who first built the home had a lease for 99 years...).

Mortgage providers won't give you a mortgage term that ends after the lease does.

People buy Condo's all the time. Also, in the United States, you very often give up mineral rights. So you may own the "ground" your home is on, but you don't own whatever comes out of it.

Condominium owners usually own a proportional share of the land that the building rests upon and the common areas of the property.

That seems almost like a scam. Kind of like time shares.

Why would that be a scam? If anything, it protects the unit owners from things like the landlord selling the building and forcing everyone to move. You can't sell an entire condo building and land without clear title from every single unit owner.

Not really.

Some more info: http://en.wikipedia.org/wiki/Fractional_ownership

http://en.wikipedia.org/wiki/Concurrent_estate

Condos differ from time shares. In a condominium, you own 100% of a unit and everything inside the boundaries of that unit is your responsibility (repairs, damage, upgrades, etc). You also have exclusive access to that unit (nobody else can use it, unlike a time-share).

You own a fraction of the land that the building sits on, and a fraction of the "common areas" (lobby, swimming pool, parking lot, etc). You are responsible for contributing for the maintenance of those areas, or paying assessment fees to a management company that will take care of those areas.

Well, what if most houses in an economy were what we called apartments?

Also, in China, the people (gov) owns the land. We are only entitled to a 70 year lease on it. Oddly enough, this is applied to housing (apartments) also. But China also lacks a property tax and so has all the bad speculation behavior and rent-seeking that goes along with that.

I used to live in a city with a community land trust. You can buy the house and get a long-term lease on the land but the land trust (a non-profit organization) retains ownership of the land. They also have a right to buy the house back when you sell it, limiting you to 25% appreciation.

The intention is to keep house prices affordable. (Sort of an upscale trailer park, now that I think about it). If you had the means to buy elsewhere, you probably would.

http://en.wikipedia.org/wiki/Community_land_trust

Ground rent exists in the US as well. Baltimore has a lot of it, maybe Philly as well. A few years ago people were losing their houses over $30/yr bills from someone who found their great-great-grandpa had the ground rent title on whatever property.

"Baltimore's arcane system of ground rents, widely viewed as a harmless vestige of colonial law, is increasingly being used by some investors to seize homes or extract large fees from people who often are ignorant of the loosely regulated process, an investigation by The Sun has found."

In Hawaii it's pretty rare to be able to buy a property fee simple. Mostly it's long term leases.

We don't have yearly property taxes in Australia, just council rates which are usually around 0.2% of the house valuation (which is much lower than in the US AFAIK). However we don't have the split land/house purchasing like in the UK (or it's not very common at least) - I wonder why?

The UK is the type of place where every last square inch of land has been claimed and defended several times over. Australia likely has plenty of land that no one is in a hurry to claim.

The property market underwent a huge decade plus long boom in Australia, with land prices rising astronomically. There are all sorts of rules for land development in place, and limited availability for development in the major cities. It's not quite as bad as the UK, but it's not like we have huge chunks of land to spare.

I assume it's either a result of property law differences, or the sheer length of time land has been available to own in the UK. Would love to know more about it.

I can recommend the book "The Poor Had No Lawyers" for some background on the history of land ownership in Scotland:

http://www.amazon.co.uk/The-Poor-Had-No-Lawyers/dp/184158907...

It may be important to note that while Australia has a huge amount of land, most people live in the cities. Australia's urban population is 90%, versus the UK's 80% and the US's 82%. There are only a couple of large countries higher than Australia in this regard.

We have fooled not only the world into thinking we're all outback-dwellers, but also ourselves. What we like is tarmac and suburban strips.

http://en.wikipedia.org/wiki/Urbanization_by_country

We most certainly do have yearly property taxes in Australia. Your principal place of residence is exempt. Your holiday house, or your office building is not.

That is why residential property is almost entirely freehold not leasehold.

Aha, thanks for the clarification.

"However we don't have the split land/house purchasing like in the UK"

Of course you do [1], and it's even very common. My cursory Google search shows that in Victory, the state with the most private ownership, 40% of all land is lease hold. Other states even more! The whole city of Canberra is leasehold. That said, it's mostly leases of Crown land, so the economic incentives are a bit different from leasing land from a private owner.

[1] http://www.actpla.act.gov.au/topics/property_purchases/lease...

Hey, no fair taking away this part of my comment! "(or it's not very common at least)"

But when you say "40% of all land is lease hold" - are you talking all land now? I was talking about residential housing, and should have made it clearer I was talking about the domain I am used to, i.e. houses in major cities, i.e. in my case Brisbane, Sydney and Melbourne. Having a quick google around and it looks like in rural areas leasehold is more common.

Interesting to learn more about this stuff, thanks for the link.

Property taxes make a ton of sense for buildings which are not primary residences. If you don't live on a piece of land or operate a business there then the property taxes need to be high enough to discourage ownership and should promote highest and best use.

That and the fact that the US is 1/7th as densely populated as the UK (UK is 260 people per km, the US is 34/km). The US is #179 on the population density list country wise. If someone won't sell you a piece of land, there's a vast supply of other land to look at. The only place you'll run into a problem is in a few high density metros like NYC.

Those are pretty dishonest stats though considering the vast majority of that land is extremely rural and much of it is owned by the Federal government (the Feds own 28% of all land in the country). The UK doesn't have something like Alaska which is both huge and empty (Alaska alone is 7 times the size of the UK, but has just 730k people).

If the US ever had a land crunch, it could easily open up vast new tracks of land for settlement via the US Govt. It's very valuable to have that ability. The Feds are already beginning to sell off their hoard of property due to fiscal problems.

That's assuming anyone would want to live on the barren North Slope of Alaska. Just because it's land doesn't mean you can live there comfortably...

There's plenty of suitable land in the US that isn't barren frozen waste or barren desert.

I think there's definitely something to this, because land taxes aren't common in Australia, either, and yet almost all houses there are freehold house and land.

Doubtful. Land taxes are extremely common in Australia. All land is taxable, however exemptions exist for your own principle place of residence (ie home). If you own a holiday house or investment property then your are liable for land tax. That is the reason houses are primarily sold freehold as opposed to leasehold.

In practice, though, it's not that different to property taxes, just that the land is owned by some lord instead of the central government. The effect is just the same. Nobody really owns their land, whether leasehold or freehold. Ownership is really just a right to exclusive use in the best case.

As the old saying goes, if you want to know who really owns your land, stop paying taxes and find out.

The cornerstone of democratic philosophy is that there's a huge difference between a central democratic government and some lord.

In theory, yes. In practice, not so sure. As a little guy I still owe someone money even though I "own" the property. There is also mineral rights and such that make it complicated. Then zoning. At the end of the end of the day someone comes and slaps me for not paying them "rent".

Not to mention eminent domain - where if your property makes sense for a freeway, pipeline, or, in a new twist, a shopping mall if doing so will improve the "Public Good" - someone can basically just pay you what they calculate as fair market value and tell you to move.

The significant difference is that taxes are assessed on an impartial basis - usually a percentage of current market value. The classic problem with landlords is that tenants have no incentive to increase productivity when the landlord captures all of the gains by increasing rent. Even today, businesses often end up in situations where the landlord sees that they're doing well and jacks up the rent in the belief that moving to another location will be too disruptive. My municipal government can demand property tax, but it doesn't hike it every time I get a raise at work.

Referring to 'lord' here is actual lords, like the Lord of Westminster. When I lived there, I paid something like 55 pounds a year into the lords pocket for use of the land.

Landlord in your usage is a different terminology. In the UK a lord with lands either doesn't have the ability, or in practice, does not raise the land rents arbitrarily. Perhaps someone knowledgeable in UK properly law could weigh in.

As other have said, in theory it is different. In practice it isn't. A single property owner has about as much chance of swaying the central government as they have of swaying the earl of whatever.

There is no direct connection between being a landlord and being a Lord - you don't become a Lord by selling someone a leasehold neither are you required to be a Lord to do this (it's just a contract).

Of course, a number of famous landowners are Lords - notably the Duke of Westminster who owns vast amounts of London:

http://en.wikipedia.org/wiki/Duke_of_Westminster

Any country that actually believed that would have true allodial title.

I thought council tax was property tax.

"When obsolete government practices persist, they tend to be pretty sinister."

How might you propose to compensate people who, in good faith, bought taxi medallions (a legal practice)? (Or do you feel that's not necessary?)

Here in SF they are illegal to sell and must be issued to a person (not a company). People get on the waiting list, get their turn, then rent them to taxi companies for (I believe) ~$100K/yr. Many medallion owners have not driven a cab in years, or never have.

The article said it pretty well.

>So, medallion holders speculated in holding a government asset and lost. Some of these people are also taxi drivers or operators of taxi dispatch companies. Like Greek bondholders, they gambled and lost big.

"So, medallion holders speculated"

Don't agree with that at all that is what the article says. (It doesn't match up with my definition either based on my years of business experience either.)

Definition of "speculation" from investopedia:

"The act of trading in an asset, or conducting a financial transaction, that has a significant risk of losing most or all of the initial outlay, in expectation of a substantial gain."

Based on history there is no reason to believe that the purchase of a taxi medallion could result reasonably in losing "most or all" of the initial outlay. (Unless of course you decided to buy one while ignoring the current events going on of course).

http://www.investopedia.com/terms/s/speculation.asp

Of course you can make any investment or purchase speculative depending on how much of a chance you want to take.

If you decide to get in on an up and coming area in real estate (a run down area that you hope will change over the course of many years) that could easily be seen as speculation. Otoh, buying a condo in a well established neighborhood that has a history would not be seen as speculation which is not the same as saying that you bought because you felt that the value could double and if that doesn't happen you may feel you have "lost" something (the high gain you expected).

Even in the case of eminent domain taking of property for public good the property owners are compensated somewhat fairly for the loss of value of their property. The government doesn't come along and say "hey we need this area for a road sorry guys to bad it happened to you".

I would go further to say that these medallion owners would band together and at least attempt to do whatever they could legally to find a way to prevent something like this from happening if they could. Why? There is enough money involved to pay lawyers to muck up the process or delay it (just like with eminent domain, people even if there is compensation, will do that.)

The comparison with eminent domain is interesting. If the government abolished taxi medallions, are they really taking something from those who own them? The medallion is basically the right to operate a taxi. If the government abolished medallions, then they'd still have the right to operate a taxi, it's just that a lot of other people would too. Obviously they lose a major competitive edge, but strictly speaking, the government hasn't taken any property here.

The best comparison I can think of involving real property would be owning a private tract of land in a valuable area that's almost all owned and unused by the government. One day, the government decides to sell its land. The value of your land plummets. It's still valuable due to whatever intrinsic wealth it has, but the new glut of supply makes it much less valuable. Is the government obligated to compensate you for that loss of value when they sell their land? I'd think not.

> Based on history there is no reason to believe that the purchase of a taxi medallion could result reasonably in losing "most or all" of the initial outlay.

This is always the case before an asset loses value for the first time.

> "The act of trading in an asset, or conducting a financial transaction, that has a significant risk of losing most or all of the initial outlay, in expectation of a substantial gain."

Based on recent history no. But maybe based on longer history? Imagine a pandemic that wipes our a large # of population, now taxi medallions for whatever reason are not that important so they drop in value.

Medallion system is sustain by the laws. Laws can change. The bet is that the laws own't change drastically but that is just a "speculation".

Government is nice enough to offer a compensation in eminent domain, but they don't have to. What if they don't? What would you do? Sue them? You can't. Gather a militia and march towards the White House or the Senate building?

If you had asked someone 20-30 years or so ago. "Do you imagine government one day will be able to store and read all your communications, telephone, mail?" Most people would have bet it would never happen because of Free Speech and it was just something we made fun of other corrupt and evil countries of doing. Yet here we are today. If you'd somehow made a financial bet against it, it would be a losing bet.

I think that you're a little too generous about government exercise of eminent domain.

I wrote an article about a New Jersey city which offered $49,000 for a three-bedroom rowhome, and when residents wouldn't take it, ripped up sidewalks and stopped collecting trash. http://www.huffingtonpost.com/2012/08/01/new-jersey-developm...

Outright seizure gets headlines, but the government can harm your property through regulatory takings as well. So crazy zoning can make your property value decline, and it's not well-established practice to compensate you for your loss. See Richard Epstein's work on this.

The Supreme Court will decide this year on a potentially really important case, Koontz v. St. Johns River Water Management District. A landowner wanted to develop some property, but the local water protection agency denied the permit because he wouldn't pay for mitigation that was totally offsite. This stuff can be borderline extortion.

How might you propose to compensate people who, in good faith, bought taxi medallions

I'm sure the following question was asked many times to abolitionists:

How might you propose to compensate people who, in good faith, bought other people

It seems as disturbing to me as it apparently does to you, but it was recently in the news that the UK actually did a buyout when they abolished slavery. http://www.independent.co.uk/news/uk/home-news/britains-colo...

It's a tricky situation, philosophically. This may have helped the UK abolish slavery sooner (a good outcome), but its morally and legally dubious.

This was suggested many times during the lead-up to the Civil War, but (I think rightly) Lincoln as well as the more hard-core abolitionists viewed this strategy as contradicting their position that Slavery was a crime against Natural Law.

Lincoln's strategy required the forced labor (and for some, subsequent deaths) of tens of thousands of people. The irony of conscription wasn't lost on Americans at the time either, and the military drafts were extremely unpopular on both sides.

Lincoln's strategy itself was "required* because slave-owners and slave-owning states would not give up their slaves without a fight. Blame them, not Lincoln.

Lincoln himself was well aware of the costs that would be imposed by war, but chose this course of action due to the absurdly perverse nature of slavery.

http://www.bartleby.com/124/pres32.html

I don't doubt that Lincoln personally resolved his internal ethical dilemmas, but I do doubt that this would be any consolation to me if I were forced to fight in an army.

As satisfying as it might seem to just shoot the slaveholding bastards instead, that ends up being even worse in humanitarian terms.

You are aware, that the first slave owner in Virginia, who went to court over his slave, was a black man, right?

How does that change anything I wrote? White slaveholders held white slaves in ancient Rome but it would have been justified to kill them all as well, too.

"How might you propose to compensate people who, in good faith, bought other people"

Seriously? You're making that comparison?

I'm comparing the logic, not the gravity of the situations.

You had pointed out (rightly) that purchasing and holding medallions is legal, and those who bought them did so with the understanding that they would profit. At the same time, they must have observed that the very existence of the medallion system limited the freedom of others. Some might have even genuinely viewed this as altruistic.

All that was true of slaveowners and slaves.

Driving taxis isn't a crime against humanity.

Your argument is that requiring medallions to drive a taxi (which I agree has become an inefficient and slightly corrupt situation) is in any way comparable to slavery is an insult to anybody who's ever read any history.

> Driving taxis isn't a crime against humanity

The problem here isn't some people "driving taxis", its that these people, and organizations associated with them, lobbied for and bought into a system that prevents, by force of government, anyone else from driving taxis.

And of course, even that is not a crime against humanity. If it was, I'd be advocating the use of military intervention if it was serious enough, such as in the case of mass slavery or genocide. All I'm saying is that if/when we get rid of this system, we don't need to compensate medallion owners.

That's a valid argument (I think some token compensation is fair, but can totally see both sides).

But if you interlaced every other word with 'fuck' and questioning the sexuality of other posters, you couldn't expect to be judged on the merits of your argument. Personally (others may disagree), I see totally unnecessary comparisons to slavery the same way.

Slavery is soon going to need to be added as an additional clause to Godwin's Law.

It's like godwin's law squared. Maybe because I'm american or maybe because the hitler comparison is cliche, but I'm more insulted by spurious comparison to a slave-owner than I am to Hitler.

You have every right to be offended or insulted.

But it makes no difference to the validity of the argument presented.

Well the argument is clearly invalid. Slave owners??

But there's a difference between lousy analogies that further debate through clarification and lousy analogies that lead to everyone telling each other fuck you.

I had no intention of offending you, or the poster, and if that is how it was taken, I apologize.

The poster was concerned about potential losses incurred by existing medallion owners if the law changed. The comparison was made to point out that laws are often wrong, sometimes more seriously wrong (as with slavery or genocide), sometimes less wrong, but wrong nonetheless.

Individuals or classes who benefit from bad laws should not get an affirmative right to recover losses incurred when these laws are changed. In the most serious cases like slavery or genocide, they may even be punished ex post facto for following such laws. I'm obviously not advocating that here, only that the law be changed through the democratic process, despite the potential effects on existing medallion owners.

I prefer making the comparison to dodging it.

I think you need to go back to the premise here a bit. The scarcity of taxi medallions was not a feature of their original purpose (to protect riders from unsafe drivers), but a corruption of it (to provide risk-limited profit to the existing holders of them).

I would expect that the profit stream off a medallion is effectively greater than that of any other risk-limited or risk-free investment in a government asset (bonds and such) or they wouldn't be in such great demand even at their vastly inflated prices.

So is, perhaps, the decades of extra profit not compensation enough? Maybe recent buyers have some claim to have had the value of the instrument disrupted, but anyone who's been leaching this system for decades should frankly be happy that the government (and by extension society as a whole) propped them up for so long.

At any rate I expect there would be a gradual expansion of the number of medallions on the market until it was an unrestricted license just like any other, at which point trading or transferring them would become illegal. This would provide some ability for holders to get some of their value back as it lowered.

I don't think there's a vested property right in them, but even if there is, there isn't in their scarcity. The government could issue medallions to all comers instead of abolishing the existing one.

Immediate compensation could be to allow "face value plus 5%" of the medallion to be used to pay any local/city taxes, with say 7 or 10 years given to use the credit.

The IRS rules http://www.irs.gov/publications/p535/ch08.html state you have to amortize/depreciate the license over 15 years (180 months). So, any licenses that are 16 or more years old, have already been written off in full, so, a lesser amount would be needed to compensate the owners.

See also: liquor licenses, various types of zoning permits, etc.

In certain states and cities with especially draconian licensing systems and quotas, the aftermarkets for permits and licenses tend to get ridiculous. Not quite as ridiculous as taxi medallions, but pretty ridiculous nonetheless. (Liquor licenses in certain parts of California have broken the $200,000 mark at auction, for instance).

In healthcare many states require "Certificates of Need", before building any healthcare facilities or even buying equipment. I think it is cost control by granting monopolies. If you eliminate competition, the providers will be able to operate more efficiently and lower costs... I kid you not, this is the explanation.

"governments would sell monopolies on the import or production of things in exchange for a lump sum"

Still happens: http://www.smh.com.au/nsw/port-leases-garner-5-billion-windf...

Apart from all the obvious externalities of cab driving operations, the extremely high probability of fraud, and the fact that policing cabs is extraordinarily expensive, there's another problem a lot of these analyses miss about why cabs are regulated the way they are.

In many large cities, the taxi system is a fundamental component of the transport infrastructure. There are locations in most cities that are not efficiently or safely reachable via bus. Visitors to cities also tend to need point-to-point transit; it takes some know-how and a lot of flexibility to get from an airport to a business meeting in a near suburb, for instance. Taxicabs are effectively a privatized component of metro transport systems.

Cities therefore have a powerful interest in making sure that whatever else happens, there is a strata of taxi service available to all residents of the city at some predictable rate.

I don't know what taxis you ride in, but i routinely ride in medallion taxis that either use GPS or their iphone to figure out where to go. I've also had some claim that they're at the destination until I pull out my phone and point to them on the map where to go.

If you want to regulate taxis, you can still do that - but there is no reason to limit the supply of taxis.

The guy you're replying to just gave you a reason to limit the supply of taxis: regulating them is expensive.

That's not a reason to limit the supply of taxis. That's a reason to charge a fee for a taxi license that defrays the cost of regulating them.

Sounds like a reason to stop regulating taxis to me. If some sort of independent "certification" is needed, there's no reason the free market can't provide such a service. In fact, it such a service is actually valuable, it's almost inevitable that it would emerge.

> there's no reason the free market can't provide such a service

sure there is. here's a bunch of reasons: http://en.wikipedia.org/wiki/Market_failure

Let me rephrase that, then:

> there's no particular reason the free market can't provide such a service, but there's no guarantee that it will.

Still, I stand by my assertion that, in general, if such a service is legitimately valued and necessary, a provider will emerge. No, a free market does not guarantee an optimal outcome for every individual, not does it guarantee that well get what we wish for. But free markets are still generally better than non-free ones. Pointing out that there are pathologies in full fledged laissez-faire doesn't count for much, unless one contends - and can prove - that (one or more of) the other options has no pathologies of it's own.

As the other commenter pointed out, the problem is primarily with the supply limit on taxis. Not limiting the supply of cabs artificially would result in more supply for whoever needs a ride, arguably lower the cost of a ride and help with more revenue for policing.

Making sure that taxis pick up in poor neighborhoods is an important, tough, and long standing problem. It would be disingenuous to argue that the "market" will fix it on it's own but then laws requiring cab drivers stop for all customers are routinely flouted, so it is not the case that regulation can fix this problem either.

The problem with supply limits is that once they are in place, everyone involved, other than the customers, has a vested interest in keeping the limits where they are for decades regardless of an increase in demand.

Traffic is something to consider - that might be one reason a customer would want to limit taxis.

Definitely, although it's not clear that ride sharing will increase traffic for sure. Cabs and Ridesharing are essentially car multiplexing. Absent an on-demand ride, most people would drive into the city, and then park there.

Increasing the number of taxis might decrease the number of cars on the street, especially in cities where public transport is not a viable option. Allowing more taxis certainly would decrease the demand for parking in already-congested central areas.

Although I admit that you could use the arguments you put forth to make a case for regulation of the taxi industry I think you'd have a hard time justifying the existing medallion system based solely on those reasons.

The medallion system (like the car dealership system) is made to protect the incumbents. Period. There was no a priori argument that this was going to be good for the people of the city or that every citizen could expect a predictable rate and level of service. In fact, the complete opposite was true -- these regulations were born out of period with a glut of taxi labor and falling rates.

What you are saying is fine, but that's not why taxi's were regulated, at least in New York. The regulations go back to the depression and were put in place specifically to limit competition in order to keep prices high and protect employment of existing cab drivers. The history is right on the NYC gov page:

http://www.nyc.gov/html/media/totweb/taxioftomorrow_history_...

quoting from the NYC site:

"Widespread poverty prompted many New Yorkers to opt for less-expensive forms of transportation, decreasing the demand for taxis. This put many companies out of business and caused many cabdrivers to lose their jobs. The situation was made worse by the tactics of “wildcat” (unlicensed) taxis who used what some considered to be “underhanded tactics,” such as drastically lowering fares, to get more business."

As you wrote: "The situation was made worse by the tactics of “wildcat” (unlicensed) taxis who used what some considered to be “underhanded tactics,” such as drastically lowering fares, to get more business."

Protecting a critical industry from market failure is more than just "protecting employment"

Why would you say that it would lead to market failure? It would have led to more cabs, lower fares, and more jobs (albeit at lower wages).

BTW - if it wasn't clear, I didn't write that sentence. I copied and pasted it from the nyc.gov site I linked. I'll edit my post to make it more clear. But I think it important that the regulators themselves are being honest about why the regulations came to be.

> The situation was made worse by the tactics of “wildcat” (unlicensed) taxis who used what some considered to be “underhanded tactics,” such as drastically lowering fares, to get more business.

Lowering fares to get more business? Oh my. What good could possibly come of such a thing.

If they lower fares by not getting insurance and not maintaining their vehicles it's probably a bad thing.

Wouldn't requiring them to carry insurance and maintain their vehicles still be better than government-granted monopolies?

Yes, but it's considerably harder to enforce.

The medallion is what puts teeth on the enforcement of those rules. It's very easy for the owner of a fly-by-night cab company to get new drivers back on the street, perhaps operating under a wholly new LLC with new livery. It's not easy for that owner to get new medallions.

That's just treating the medallion as a deposit. If that's the point then why not use an actual deposit and not artificially limit the number that are issued? Or just treat the cab itself as collateral and seize it for any violation that would have caused forfeiture of a medallion? It may not be worth several hundred thousand dollars, but it's certainly worth enough to have to take it seriously.

see: Fung Wah buses on the East Coast

> "who used what some considered to be “underhanded tactics,” such as drastically lowering fares, to get more business."

Lower fares is not the only consequence of this type of free market. In NYC we've seen a rise of intercity buses competing at cut-rate prices, and one such bus company recently and spectacularly failed every single safety inspection performed on their vehicles. They also have an accident rate several times that of their less price-sensitive consumers (Greyhound, Amtrak, etc).

Free markets often have substantial downsides, it's dishonest to pretend that regulation is strictly driven by parasitic behavior. There are also substantial upsides to commoditization - i.e., gains in efficiency when every consumer does not have to engage in complex analysis of the product on every sale. There is a loss of market confidence when you have to go through a mound of paperwork just to determine which intercity bus line is safe to ride.

When a market is over-regulated, such regulation often have legitimate and still-relevant roots. Those who cry for complete deregulation would be wise to figure them out first. Regulations may run amok, but they very often still sit on top of a legitimate need.

You say, "it's dishonest to pretend that regulation is strictly driven by parasitic behavior," but I was just quoting directly from the nyc.gov site that explains that's exactly why those regulations came to be.

You can argue separately for safety regulations, but that's not why NY started to regulate cabs - it was protecting an existing industry from competition. Also, put yourself in the 1935 mindset - was there even auto insurances? were there seat belts? The risk profile people lived with was much different.

> it was protecting an existing industry from competition

Otherwise known as "keeping people employed at a living wage." When the market decides that human labor is worth nothing, it's irresponsible not to contradict it.

I will agree that medallions should have gone away after the crisis, but it wasn't "parasitic" at the time.

>Otherwise known as "keeping people employed at a living wage."

Or, stated differently, keeping some people employed at above-market wages and others unemployed with no wages at all.

>When the market decides that human labor is worth nothing, it's irresponsible not to contradict it.

If that ever happens we'll know regulation is needed.

You can also spin this as "keeping people who need a ride poor by raising the price of the good they depend on", or even "preventing the poor from working by making it uneconomical for them to get to work".

If you're going to argue in favor of breaking competition, you absolutely have to have a better argument than "it's good for the sellers." There's two sides to every economic transaction, and I see no evidence that cab-medallion-owners are drastically more worthy people than cab-riders.

In many of these cities, it is impossible to flag a cab at any price during peak hours. It would be far more efficient for the government to top up the earnings of low-earners than to manipulate the supply of cabs.

> "Also, put yourself in the 1935 mindset - was there even auto insurances? were there seat belts? The risk profile people lived with was much different."

Indeed, and that's precisely the start of much industry regulation. Not just taxis, but railways, boats, and all other forms of shared transport. Transportation safety was a Big Deal in the early 20th century, and the market did differentiate itself on that point - but there was also little regulation or oversight to ensure that claims of improved safety were actually real (see the most infamous example in the Titanic).

Mandated seat belts, car insurance, brake mechanisms (e.g., trains), speeds, inspections. All of these were borne out of necessity. The lack of safety in the 1935-life is why we are living with so many regulations today.

Our history seems to go: new technology, a lot of people are killed by new technology, regulation to improve safety of technology.

The "let's deregulate everything!" angle is just winding back the clock. To before the regulations became onerous and excessive, but also to before the regulations saved a lot of lives.

Regardless of origins, I think there's still an important role for city governments to manage the number of cars on the street. NPR Planet Money did a story about the effect of NYC's planned +2,000 cabs (in additional to the existing 13,200) will have on the city: shorter wait times, but longer drive times for everyone. If the story is correct, more deregulation with more cars on the street without other measures (such as congestion pricing) would be bad for the public. NYC traffic is already bad enough.

http://www.npr.org/blogs/money/2012/07/31/157477611/does-new...

It's far from obvious that demand has outstripped supply. The population of Manhattan was significantly higher in 1937 than it is now 1.9 million vs. 1.6 million. On a population basis it would seem that we would need fewer rather than more cabs.

What drives the demand for cabs is the the price of a ride versus the alternative. If the regulated fare is low relative to subways & buses and the overall price level, cabs will be in high demand. If the fare is high, demand for cabs will decline.

You really can't compare the use-per-capita of cabs between 1937 and 2013...

Why not? An ellipsis isn't an argument.

I thought calling attention to it would be enough. Life, wealth distribution, transportation modes, etc. are very different now than they were 80 years ago. Treating it as a roughly apples to apples comparison is ludicrous.

> It's far from obvious that demand has outstripped supply. The population of Manhattan was significantly higher in 1937 than it is now 1.9 million vs. 1.6 million. On a population basis it would seem that we would need fewer rather than more cabs.

Wrong population number, in the case of Manhattan where real estate prices are so high that 'number of people officially living there' becomes the wrong metric, and the right one begins to look like 'number of people physically present on the island each day', which is closer to 4 million people: http://wagner.nyu.edu/rudincenter/publications/dynamic_pop_m... (Significantly higher than 1.6 million!) And these commuters are also much more likely to need transportation - like taxis - for obvious reasons.

You miss the point -- it's not about the number -- it's about the change since 1937. No where in the reference you provide is there any evidence that this number has grown since 1937.

http://www.nypost.com/p/news/local/subways_on_roll_fDYxgwu92...

From the Apr 10th NY Post: "More than 1.65 billion straphangers used the system in 2012 — numbers that haven’t been seen since 1946 when ridership was more than 2 billion, according to MTA statistics released yesterday"

Has the relative income of the residents changed since then? I.e. if the average income in 1937 was on par with the national average and now its 200x that, perhaps there's greater demand for a luxury service such as cabs from a greater slice of the population.

> Many New Yorkers then worked six days a week and many of today’s free-transfer points required a second fare.

What I don't get is why a city as dense as NYC supports any transportation method that puts a few individuals above the public good. Cars in a dense urban area epitomize the tragedy of the commons. Buses and delivery trucks are the only vehicles that must have access to the street. Everyone else should be walking, taking public transportation, or bicycling.

Not having to wait on a street corner for a crosswalk signal or only doing so sparingly would save NYCs on average more time than the occasional taxi ride. On top of that, the absence of cabs would set many minds on the problem of moving all New Yorkers around the city as fast as possible. If you're rich and you think that your time is valuable, then vote for spending on public transportation innovations that benefit everyone equally.

If you eliminate cars and time truck deliveries well, you can engineer a transportation system that runs like a well engineered clock with many complications. Cars are responsible for most of the non-detrminism in a public transportation system.

Taxis are not the problem. Private cars are the problem, and the city does a pretty good job keeping them out of Manhattan as well, with tolls on many of the bridges (any bridge into Manhattan connected to a highway is tolled; local street bridges are not) and a 19% parking tax.

But there are certain situations where you need a taxi and the bus or subway won't do. Think the elderly, families with small children, etc.

I would love to see a city mandate something the size of the Lit Motors vehicle for the cases you mentioned. However for many of the cases you mentioned, taxis still aren't the solution. There are plenty of families and handicapped people that get by with public transportation. They can use things such as scooters, wheelchairs, motorized wheelchairs and strollers. Technology has advanced to the point where a motorized wheelchair or scooter is no longer a rare luxury.

While I can see the argument for taxies, I;m totally in agreement about private cars and especially parking. Parking has no business in the modern urban dense city.

> What I don't get is why a city as dense as NYC supports any transportation method that puts a few individuals above the public good.

Well, for one thing, NYC is answerable to NYS. Lots of people commute into NYC from its suburbs, they like to drive, and they do vote.

There's a solution for that. Grand Central Station. Thousands upon thousands of people commute to NYC by train everyday, leaving their cars at the suburban train station parking lot.

Are you suggesting that there will be not be an increase in the number of cabs in NYC if more medallions are opened up?

This is why it's so annoying that Albany vetoed the congestion pricing plan. Congestion pricing would keep vehicle traffic down to acceptable levels without discriminating based on type of vehicle.

Still though it should be recognized that the medallion system leads to massive returns for a tiny minority of existing medallion holders who are basically just extracting rents. The system could have easily been set up so as to distribute the returns to riders, or drivers, or taxpayers.

"but that's not why taxi's were regulated, at least in New York."

Why does it matter why they were first regulated? It matters just as much why they keep being regulated because the reason can change or morph over time. And even if regulated for the wrong reasons it doesn't mean that in the end there aren't any benefits.

I'm not sure why it matters what the original impetus behind medallions was. Maybe that means we should have a different regulatory regime more suited to ensuring that every resident of a major metro area has access to safe, efficient taxi transport and less suited towards ensuring a living wage for cab drivers. That's fine. But it's not an argument for no regulation at all.

There may be a compelling argument for no regulation. I'm just saying, that's not it.

I'm just gonna throw this out there, "underhanded tactics" in the depression era were probably quite a bit more underhanded (no scare quotes) than undercutting the competition on price.

Policing cabs is vastly cheaper than the cost of a medallion.

https://news.ycombinator.com/item?id=1650659

Cities therefore have a powerful interest in making sure that whatever else happens, there is a strata of taxi service available to all residents of the city at some predictable rate.

Limiting the supply of taxes reduces the likelihood that this will occur.

Not that NY has anything like the "universal service" you are hinting at, taxes rarely go above 125th st.

Taxis are required to service all destinations in manhattan.

But yeah, originations are a problem.

They are required to service all of NYC, but will do almost anything not to.

Correction, taxis are required to take you to "any destination in NYC, Westchester, Nassau, or Newark Airport;"

Source: http://www.nyc.gov/html/tlc/html/passenger/taxicab_rights.sh...

They often stop with their doors locked, roll down the window, and ask where you are going.

Then any system which fulfills the following criteria should work:

- good (safe, knowledgeable, rested) taxi drivers - taxis in good condition - a sufficient number of taxis on the road - a reasonable price to the customer - and a system to handle complaints

It seems to me that medallions actually fail on some of these points. Instead, it would be worthwhile to:

- license drivers, not medallions. Non-transferable right to operate. - license taxis with regular inspections - use technology to track cabs - have the city run a centralized, mostly web-and-touchtone based dispatch system - and adopt suitable regulations to avoid feast-or-famine at the airport, downtown, and that socio-economic region that the cabs tend to avoid

license drivers, not medallions. Non-transferable right to operate. - license taxis with regular inspections - use technology to track cabs - have the city run a centralized, mostly web-and-touchtone based dispatch system

Which is exactly the model within which Uber et al operate: livery drivers and webapps.

None of these things are reasons for a limited number of medallions.

To be clear: a medallion system says, "In San Francisco, we will have 680 cabs, no more."

That is separate from the question of whether you license cab drivers or charge for the license.

You could say, "In order to drive a cab in this city, you must pass an area knowledge test, pass a more-onerous driving test, you must re-pass these tests every two years, and we will revoke your license if you're involved in an accident. And in order to defray the costs of administering this program, we will charge you $1,000 for that license. And you must use this pricing scheme for fares." But if you were willing to license every person who passes those tests and pays that fee, without regard for how many people have already passed the tests and paid the fee, you aren't operating a traditional medallion system.

Don't forget the mandate that a certain number of taxis need to be wheelchair accessible. I'm not disabled, but I would like to see both SideCar, Lift, Uber and InstantCab offer wheelchair accessible community cars.

" the taxi system is a fundamental component of the transport infrastructure."

Startups, VC's and angels don't want to consider any points like that. They only care about how they can get a part of the pie (like any business.) VC and angels are business people who are motivate by profit (full disclosure: so am I). Startups are started by people who simply think everything from the old way of doing things is bad and needs to be overturned and not to trust anyone over 30 as the old saying went.

Along these lines I'm waiting for startups to try and "disrupt" the liquor control system (and liquor licenses) that are in place in various communities across the country for good reason even though legacy owners profit from that "restriction of supply" as well.

Liquor controls are an example of a genuinely obsolete regulatory regime, as are distillery licenses.

On the other hand, as we can plainly see, late-night permits and ABC licenses in California are a matter of open conflict between residential and commercial parties.

I'm still looking for a regulation omni-apologist who can explain to me why it is rational to restrict the supply of cabs so that it is impossible to flag one during peak hours in many cities. It seems the taxis monopolies benefit and the politicians they support benefit, but the vast majority of us lose.

That's why uber exists and is highly profitable. Unregulated cabs at any price are better than regulated cabs that might as well not exist. And it doesn't hurt that uber has innovated on an industry that has been stagnant for a long time, like most regulated industries.

Medallions exist for a very good reason: they prevent a tragedy of the commons from creating gridlock and pollution in major cities by limiting the number of cars circling around doing nothing on city streets.

Most of the time, cabs just drive around waiting for a fare. This is because cabs are often hailed when they are seen, but even if a cab is summoned by a phone call (or smartphone) the closet cab to the fare is chosen, and it's hard to get close to fares without driving around, because parking the cab downtown is expensive and slow.

So, cabs use the city's public streets as mobile parking lots, hovering close to potential fares, snarling traffic, and emitting greenhouse gasses until they snag one. Increase the number of medallions, and you have a corresponding increase in the amount of traffic in a city as cabs wander around fishing. Cab drivers don't care if they're slowing everyone else down. They just want to earn a living.

So it's better to have fewer cabs busy 100% of the time than more cabs which aren't always busy. Thus, medallions are scarce.

Recently NYC decided not to issue more medallions, because the increased availability of cabs would be more then offset by increased time to get from point A to point B due to more traffic congestion.

This is a strong argument for regulating this new breed of unlicensed cab services. They must not be permitted to loiter around in traffic hoping to be the car closest to the next fare. They must either have a medallion or be parked when not actively on a call.

"they prevent a tragedy of the commons from creating gridlock and pollution in major cities by limiting the number of cars circling around doing nothing on city streets."

Why would a free market not prevent this? if there are excess taxis, taxi-driving is not profitable. If taxi-driving is profitable and there is still gridlock, that is a different problem. Avoiding gridlock by creating an artificial shortage of transport is not an appropriate solution IMO.

What about other industries that need the streets? What you're saying is akin to saying "Who cares if the creek is poisoned downstream, I need to dump into it to maximise profits!"

The reason there is a tragedy of the commons is that not _all_ the costs are internal to the driver. Sure, if traffic is slow enough it becomes unprofitable for everyone, but before that point there's a situation where it's profitable to you but you're causing so much costs for other drivers it's a net bad for society.

Simple: the number of cabs at which taxi driving is no longer profitable is far, far more than the optimal number of cabs.

The optimal number of cabs balances cab availability with traffic such that the time taken to go from point A to point B is minimal. This takes into account the time it takes to hail the cab (waiting around if there is none) and the time it takes for the cab to fight though traffic to point B. An additional constraint is that the number of cabs shouldn't prevent miserable gridlock for other drivers.

The number of cabs required before cab driving becomes unprofitable is a function of the number of people willing to work for meager wages, often less than minimum wage (in a big city, this number can be considered unlimited, especially if the job is part-time), the cost of gas, car, and insurance, and the market fare. Assuming the amortized cost of a car is fifty cents a mile, there is enough profit in cab driving for less privileged workers, even if the cab is idle most of the time.

Most importantly, traffic congestion in a big city is non-linear: it only takes a few extra cars on the road to create major gridlock.

> Simple: the number of cabs at which taxi driving is no longer profitable is far, far more than the optimal number of cabs.

Do you have any evidence of this claim? As with anything, there isn't some magical shut-off point where everyone is selling something and then no one is after the price increases another dollar.

Your analysis strikes me as too simplistic. It ignores opportunity costs and the possibility that it might decrease personal car ownership outright if a viable taxi system were to exist.

At any rate, even if it's taken as fact it doesn't justify a fixed number of medallions revised once every few decades. A price set on them, adjusted regularly, would achieve the same cause and be less limiting and more responsive to growth (something cities tend to do a lot of between the issuance of new medallions).

Creating artificial scarcity does nothing but enable some people to profit with little risk.

The market fare is also affected by the number of cabs - if the number of cabs available significantly exceeds the demand for cabs, the market fare will fall. This means that each additional cab on the road is a double-whammy against the profitability of all cabs - it reduces the market fare and it increases the average idle time.

If you want to limit congestion, price congestion.

Don't limit taxis, the right to drive people, the right to drive a car ending with a particular license plate number, or the amount of fuel available (all have been tried to reduce traffic) -- just make the roads themselves cost more when there are more cars on them, and make the roads cost less when there aren't.

And here I thought you were going to propose a solution? The implementation of "make the roads themselves cost more" is empirically non-trivial.

Lets get everyone on Fastrak and see how they like it :P

http://www.tfl.gov.uk/roadusers/congestioncharging/

These new systems exclusively use smartphones to hail cabs. Cabs don't need to circle to find fares. This is much more efficient than the medallion system

I find this a very weak argument. You're essentially saying that we have to limit car usage because too many cars cause congestion.

Congestion has an explicit cost (wasted time) even if it is not priced in with some artifical time-of-use payment.

Ultimately people will get onto the road one way or another if they need to get somewhere at a set time. So by limiting cab licences, all you are doing is limiting the mix of cabs vs other cars. You can't honestly say that, at 5pm on a Friday afternoon, people are going to not catch a flight to the airport because there aren't enough cabs. They'll either order a town car (a different type of cab) or they will get a lift. Limiting the number of taxis is a very poor proxy for limiting congestion.

The issue of empty cabs driving around is spurious because it is explicitly fixed by the technology. If that's what the purpose of limited taxi licensing really was, there would be a different solution in place than limiting the overall number of cabs.

NYC might have said they're trying to limit congestion, but I think that's a very naive view of the situation where vested interests are clearly benefitting from a government-granted monopoly. Even if you rule out corruption, the status quo doesn't affect the people choosing the medallion numbers, so I doubt they would see a need for change.

The idea is not to reduce cab counts to the point where people call towncars instead; that's likely a worse outcome, because the towncar spends only ~1/3 of it's time full: the other 2/3rds are spent driving to and from the towncar's parking space.

Ideally cabs would spend near zero time driving around empty. Today this is solved by medallions, imperfectly, but tomorrow this can be solved by technology.

What can't be solved by technology is the number of cabs trolling around. You can have perfect GPS based dispatch, with your next fare always ready to get in your cab just as you're dropping off the last fare, and there would still be huge problems, because there's nothing to stop this ideal set of 100% in-use cabs from being diluted into five times as many 20% in use cabs. People will take the closest cab and drivers will drive around to make sure they're the closest cab.

You can charge a per-mile tax on empty taxis, but that just raises cab fares for citizens (or forces caps) for no good reason. Just set the optimal number of cabs by fiat.

In some respects, this is a natural monopoly.

The process that would stop 100% in-use cabs from being diluted into fives times as many 20% in-use cabs is that the 20% in-use cabs would all be losing money.

If congestion is the problem, then directly charge for that - London has for example implemented a time-based congestion charge for the city centre.

>Just set the optimal number of cabs by fiat.

There is no person, committee or even algorithm that can set the correct supply better than a freely trading market. This has been known for a long time.

>In some respects, this is a natural monopoly.

No, it's an artificial monopoly. What we are talking about here is exchanging rides for money. That is the market - paying someone to drive you on a point-to-point basis.

If the market were left to it's own devices, and the only regulations were around traceability, security and the usual laws against blackmail etc, then you'd have a floating market with different prices per time of day and per trip. This would actually work against congestion by pricing it in.

Instead we have a central government committee deciding supply by whatever they use, and fixing prices. The natural outcome of fixed supply and fixed prices is sub-standard service and a shortage of available product. Which is exactly what cab users everywhere experience.

I disagree: the number of cabs in circulation is a complex function of the market, but the ideal number of cabs is straightforward to calculate but won't be found by the market.

Why not? Given the choice between an optimal number of cabs (fully utilized) making a larger profit and an excessive number of cabs (underutilized) making a tinier profit, the market will always choose the latter. Markets will tend to attract competitors and thus supply until profit margins drop to near zero. In both cases prices are the same, revenue is the same, but in the latter case the pie has more slices, increasing congestion. There is less profit per taxi in the latter case, especially if prices drop with increased competition, but as long as there is at least an iota of profit, the number of cabs will not decrease to the most efficient level.

Taxing congestion itself is a good idea generally, and helps increase road efficiency, but this won't result in the optimal number of cabs, either. It will result in increased prices which will cut demand and thus supply, but given this new demand, it won't magically remove competion such that fewer cabs are servicing more passengers, relatively speaking. The number of cabs will still remain at the maximum number possible while still remaining profitable. For any given cab, the amount of time spent idling is still the same--too high.

While that may be a benefit of medallions (or not), it's almost certainly not, historically, why they exist.

I can't speak for all medallion programs, but NYC medallions were created to solve a surplus of taxis and associated congestion, low wages, and poor service, at least according to the NYTimes:

http://www.nytimes.com/1996/05/11/nyregion/medallion-limits-...

As far as I can tell that article implies the main goal of medallion limits was to increase prices by decreasing competition. Which is exactly what I would have expected.

"Medallions exist for a very good reason: they prevent a tragedy of the commons from creating gridlock and pollution in major cities by limiting the number of cars circling around doing nothing on city streets."

You're missing a huge piece of the equation: politics and corruption. Medallions were currency used to do favors to politicians, power-brokers, etc. It was a cheap way for someone to make some cash by renting it out. The government created the monopoly and scarcity; then those in power used it to make money. A story that's repeated all over the place.

You can't say with a straight face that if 11,000 taxis are good enough for 1930s NYC that 13,000 taxis are good enough for 2013 NYC.

I drove a taxi as a summer job in college in my hometown (small city near Boston). I thought I would share a few observations.

In our city, there were three companies that owned the medallions. An unwritten rule: The city was divided into territories, and we weren't supposed to pick up in the other territories unless we were flagged down. If someone from another territory called our company, they were told to call one of the others and arrange for the pickup.

75% of our business was call-ins or pre-arranged pickups (students, travellers, sometimes people with revoked licenses). Another 20% was pickups at the taxi stands outside the subways stations or hotels. There were a very small number of people flagging us down on the street. In Boston, there are many more street pickups. Note that this was all before the cellphone era.

We also couldn't pick up in Boston, unless someone called us in advance. Even if we were flagged while driving through the city (for instance, coming back from a drop-off), we had to ignore. This was by law, to prevent cabs from Cambridge or other towns from taking business away from the Boston cabs (and Boston cabs from taking business from Cambridge and the 'burbs, etc.). It made airport runs a bit of a risk -- while the fare was higher, since it was almost impossible to be able to pick someone up after dropping off someone at Logan, we were basically losing the opportunity to take other passengers for 30 minutes (or longer, depending on the traffic).

I should also add that rates in Boston were cheaper than other nearby cities and towns. This would make some passengers angry -- "it only cost me $X to come from the airport. Why is it $X + $20 now?"

We got a cut of proceeds off the meter, plus tips. Part-timers had to take the worst shifts, and the worst cars -- old LTDs and Ford Fairmonts that often had major mechanical problems. It was a poor way to make extra money -- unless you had a busy shift (rush hour, or rainy days) there was lots of down-time, which meant fewer receipts.

There is a flipside however, for those who think deregulation is the answer.

In Amsterdam, the taxi market has been completely deregulated. You do not want to get a taxi there unless you order it from a trusted source, and even then it's ridiculously expensive.

And you don't want to try to disrupt that market unless you have good organized crime connections...

yeah, i'd be interested in comparisons of urban transportation across the globe. beyond taxis, an analysis with comparisons like underground use vs. taxi etc would be a great follow up to this article.

That seems more like a failure with "regulating" organized crime than a call for regulating taxi cabs.

what it means is you changed one monopoly for another.

>And you don't want to try to disrupt that market unless you have good organized crime connections...

How much of that is caused by decriminalization of drug use without decriminalization of large scale drug supply?

If organized crime is involved (or similar), then it's a failure of government.

Taxi drivers significantly under report earnings for tax purposes. They always claim close to minimum wage. You'll find the same in every cash business.

Probably, but I can guarantee you that the vast majority don't make a lot of money.

Presumably they're making more than their next best alternative.

I once asked a cab driver how much he made (in SF), and after humming and hawing, he implied that he made close to $90K/year, mostly tax-free.

Take it for what it's worth (sample size == 1), but there are a lot of people who would love to be driving a cab in SF. Cab drivers who rent the cabs pay $300 for an 8-hour shift on weekends (Friday evening - Saturday morning), and still manage to make enough money to make it worth their while.

Taxis (in New York at least) are definitely not cash businesses these days. A lot pay by card.

"In the last half decade, two trends conspired to end the taxi medallion regime. First, people are more comfortable with trusting strangers."

This isn't true. When I was growing up, hitch hiking was still very common.

I'd argue that people used to trust strangers to a radically higher degree before the prohibition based crime wave (the one that makes swaths of every city in the US more dangerous than Afghanistan) that began in the 1960s.

I agree. But as the article goes on to point out, we are not more trusting of complete strangers, but rather strangers who've been vetted by many AirBnb reviews, background checked by Uber, etc. Technology has changed the definition of a 'stranger'.

>The reason taxi drivers have to pay for the right to work is that they need access to a taxi medallion to do their job. A medallion is a permit issued by the government that is required to drive a cab in most cities in America. If you don’t use the medallion yourself, you can rent it out to other drivers on your own or, more commonly, through a taxi company. Taxi companies that rent out access to the medallions have immense economic power over the drivers. If you’re not willing to basically become an indentured servant to get medallion access, well, you’re out of luck.

I don't see how the medallion itself (a permit issued by the government) is the problem here. The problem is that the taxi drivers are not allowed to get it directly for themselves, but have to rent it. And another problem is that this "indentured servant" renting is allowed to take place.

Fix that, not the requirement to have a permit to drive a cab (which is useful, as it acts as a filter for all kinds of scum drivers if enforced properly).

Taxi drivers are allowed to get medallions. But since the supply is so low, they usually cost a few hundred thousand dollars. So people rent instead.

Fixing that by outlawing medallion-rental is like trying to make housing costs cheaper by outlawing apartments. A much better solution would be to eliminate medallions, which could be done without removing the licensing system filter.

>I don't see how the medallion itself (a permit issued by the government)

Well... now you're playing word games. A medallion, as commonly understood, is a very specific type of government-issued permit, with special properties (like being transferable and severely limited in supply).

People are bringing up "market forces", etc.

The solution that I think would work in SF is as follows. Charge a flat rate (say, $50K/year; though a Dutch Auction might be reasonable too) for leasing out a permit to qualified drivers. You qualify drivers by running background checks, and testing them on the knowledge of the city. Pass a law that makes illegally running a cab a serious offence. This will automatically solve the "tragedy of the commons" or "too many cabs" problem, because if there are too many cabs, they won't be able to make the $50K/year needed to break even. If 2000 cabs (a number 20% larger than the current fleet size) enroll in this, you're looking at an additional $100M/year.

To run the program, have a small enforcement wing of SFMTA (instead of the current gigantic bureaucracy that micromanages everything). Offer up an open API so people can build apps that track cabs, hail cabs, etc.

The medallion system is 19th century. We need a 21st century system.

The medallion system is certainly far from optimal and a different regulatory regime that allows for innovation while still addressing concerns around impact on traffic (public roads=classic tragedy of the commons case) and passenger safety would be widely welcome. It could certainly improve the customer experience on many axes (quality, availability, timeliness, cost, etc) though I'm skeptical it would do much to improve the lot of the average driver. Taxi driving is a relatively low skill field, so there wouldn't be much incentive for these new ride sharing companies to treat drivers any more fairly than they are currently treated, since drivers will need to work for whichever companies have the largest market share of potential fares.

This article makes one awesome point for regulation that would be beneficial: exchange of data on passengers and drivers. Even if opt-in (for privacy reasons) this would be awesome. As a customer, I would love for my reputation on Lyft to be usable on SideCar and vice versa. On top of that, I'd love to be able to call a car from both companies simultaneously and whichever company gets met a driver first wins my money.

Requesting a car needs to look more like the financial system where your broker puts out a buy order and a broker representing the driver takes that buy order.

>After the customer leaves the car, there is no record of their behavior in the taxi.

Really? I haven't looked hard for them in NYC Yellow Cabs, but every livery car I've hired has had a security camera inside.

By the time they get around to de-regulating the medallion system, self-driving cars will be available, which will put taxi drivers out of business. (Unless they block that too.)

Self-driving taxis will put themselves out of business once they either a) crash, killing/injuring someone, b) are regularly used for illegal activities, or c) deliver someone into a dangerous situation.

If that doesn't happen, the existing taxi industry will surely get into the game to make one of those happen, tout-de-suite.

Taxis seem to keep their status by being a very focused interest group. In NYC there are never enough. Why not just sell more medallions? I don't buy the "There's too much traffic" argument. Or why not sell regional ones? Try getting a taxi in Harlem. The city's approach has been to look the other way on private cars. I would rather see Bloomberg allow disrupters to break the system. What's he have to lose? He can't get elected anyway...

In NYC there are never enough.

Depends very much on where you are in the city. Aside from shift changeover periods, I'd say that there are plenty of cabs in Manhattan. In the outer boroughs, less so- Bloomberg proposed a plan to allow 'green cabs' that could pick up passengers in the outer boroughs only. It was swiftly beaten in the courts by the taxi industry.

I wish all the cab-alternative companies using unlicensed drivers would get together and form some kind of industry association. Standardize a background check, safety check on the vehicle, incident reporting, etc., through this independent non-profit. Essentially they could neutralize the moral argument for licensing ("safety! think of the children") at minimal cost, then get back to innovating.

Does anyone here actually use Priceonomics to look up prices on things? I find it almost completely useless, is it just me?

Great article. I'm interested as to what the effects would be on the taxi industry from these ride sharing apps combined with self driving cars. I think it would likely result in a further more efficient transport system - less people having to own cars, a decreased need for car parking.

annoyingly this link goes to the home page, not the actual article -- otherwise a good read.

Fixed. Sorry about that.

In 20 years all taxi drivers will be unemployed due to self-driving cars. So who cares?

Robots don't need health insurance and never strike.

An excellent current and historical comparison can be made to the ambulance medallion system. Oh wait, there isn't one...

Odd, no mention of driverless cars, which are sure to wipe out the taxi industry entirely.

The last sentence of the article:

> Ride-sharing apps should be allowed to take over the market - until they too are disrupted by self-driving cars. And so it goes.

There are a large number of technical and political hurdles before driverless taxis will be possible. We will need autonomous vehicles on the road for many years for legislation to adapt and for people to become comfortable with the driverless concept from a safety and liability standpoint. Still, I'm with you that it is inevitable, and very exciting.

I'm cashing in all my bitcoins and investing in NY taxi medallions

Reply on news.ycombinator.com