Comments on: How the taxi-medallion bubble might burst A slice of lime in the soda Sun, 26 Oct 2014 19:05:02 +0000 hourly 1 By: OneOfTheSheep Tue, 24 Jan 2012 23:49:45 +0000 @Komanoff,

Your explanation makes sense. Thank you for your detailed response!

By: Komanoff Tue, 24 Jan 2012 21:57:32 +0000 @OneOfTheSheep — I composed the analysis that motivated and underlies Felix’s post.

What I’m calling journey time is the sum of in-cab travel time and hail time. With more taxis, the former goes up while the latter goes down by roughly the same amount (a coincidence). Journey time thus goes down almost imperceptibly, causing demand for taxis to rise, also almost imperceptibly (the 0.2% you cited, or intended to).

Medallion taxi fares are 100% regulated, so you’re right that they won’t drop despite the increased competition. I also agree with your sense that wheelchair accessibility will add little to demand.

What *will* change is the percent of taxi time spent cruising (fareless), from 35% now, to 40%.

You can see this and lots more in my spreadsheet model. Please download my report (via the “hard math” link in Felix’s post). The link to the model is under References, in the back.

By: OneOfTheSheep Tue, 24 Jan 2012 06:51:06 +0000 Maybe I’m missing something here. If fare rates do not drop due to increased competition, demand for taxis should remain unchanged. A few people unsuccessful in hailing a cab before might now “get lucky”, or a few more in wheel chairs may become new taxi customers.

Days ago on another Reuters commentary on this subject I responded (in part): “The number of “hired trips” is determined by the number of riders, not cabs; so there should be NO effect on traffic from the same number of “fares”.

Your data bears out what I said. You show taxi trips per day (riders/fares) of 472,200 increasing to 472,000, or only 0.02%.

We need to know taxi miles driven per shift. If a given taxi is one of those that gets in the taxi queue at the airport until their”turn, goes out, and returns to do the same thing the queue is longer, “dead time” longer, and time “in traffic” less. If a given taxi is “on the prowl” in a fare-rich area, they are “in traffic” their whole shift even though they get fewer riders/fares in a shift. What will be the “mix” of these “new” 2,000 medallions?

You show a journey time reduction from 19.2 to 19.1 minutes. No “added congestion” evident here. But you also claim a purported drop in travel speed of 12.1 % (from 9.5 MPH to 8.4 MPH). These are obviously figures at odds with each other. What am I missing?

By: SteveHamlin Mon, 23 Jan 2012 15:16:12 +0000 @JohnHaskell – Felix isn’t suggesting that everyone use private cars instead of taxis, he’s merely explaining that the congestion effect from adding 2,000 taxis (in Manhattan/B/Q and driving 24-7) is similar to adding 80,000 private cars (driving in, parking for 9 hours, driving out).

If you were to replace X number of taxis with exactly X number of private cars, then yes, street congestion would go down. And then taxi fares would go up (or demand would drastically outstrip supply resulting in long waits, which has the same economic effect if time = money), black cars and gypsy cabs would flourish, parking garage rates would rise, subway ridership would go up and current patterns of mobility would change.

Note that you could only limit private cars in order to enforce the congestion decrease with strict governmental laws/regulations. You’d need a parallel medallion system (“right to enter Manhattan”) for private cars. Otherwise your second paragraph would come true (fewer taxis, many more privates, increased congestion).

By: johnhhaskell Sat, 21 Jan 2012 10:16:54 +0000 If taxis spend 40 times as much time in congested areas as private cars do, why don’t we get rid of all the NYC taxis and have everyone drive in their private cars! There, congestion problem solved.

Oh wait a second, cities which push their residents to use cabs (e.g. Hong Kong) have much less congestion than cities which don’t (e.g. Moscow). So maybe we need to think a little bit harder.

By: bryanX Sat, 21 Jan 2012 06:03:33 +0000 Wow. This might make cab drivers angry, causing them to start beeping, yelling, and driving reckless all the time.

By: TSTS Sat, 21 Jan 2012 03:50:06 +0000 Felix: “And of course the income for the driver is going to fall more than 19%, because the medallion owners are going to be very reluctant to drop the amount they charge the drivers per shift. All of which implies to me that if there’s a medallion-price bubble.”

This statement seems contradictory because the value of a medallion depends on the amount charged to drivers per shift, not the driver’s income. So if the drivers’ income drops by more than 19%, it would seem the value of the medallion would drop by less than that.

But more likely, the income of the drivers would drop by less than the income of the medallion owners as there is a lower bound to what they are willing to work for and we are probably close to that bound already. (Otherwise, what prevents the owners from raising the charge to the drivers now?) And this does not take into account costs such as fuel and maintenance that are also significant and that will drop by less than 19%.

So overall, even a slight decrease in trips could result in the value of a medallion crashing by a much larger amount.