One-quarter of Uber's business happens in just five cities | Hacker News Reader

One-quarter of Uber's business happens in just five cities

slate.com

40 pointssmacktoward41 comments

2017 GDP Figures (Trillions USD)

  USA: 19.39
  NYC: 1.72
  LAX: 1.04
  CHI: 0.68
  DFW: 0.54
  DCA: 0.53

OMG, almost a quarter of the US's GDP (23%) comes from just five metropolitan areas!

Sure, and if any of those cities all the sudden decided to no longer produce domestic product, that would be a big risk to the US.

The GDP is a measure of all economic activity. I'd say there's a far greater risk that NYC might outlaw ride-sharing specifically, rather than, you know, suddenly decide to stop engaging in commerce altogether.

That is exactly his point.

I think he's trying to state how unlikely that scenario is (compared to how relatively easy it is to ban Uber)

He's saying: this is why Uber city-ridership as a Zipf law with alpha = X: the economy has a Zipf law with alpha = X.

You're saying: this value of alpha presents a concentration of risk for Uber. Yes, you're both saying things that aren't false.

DCA is surely about something other than real productive activity? (I guess people in DC might be making money, I thought they were spending it)

Perhaps, but if you want, you can replace it with SFO, which is roughly the same.

There is a surprisingly large real economy about the area.

The Washington DC MSA is, by population, the 6th largest MSA in the United States[1]. It also the 2nd most highly educated MSA in the US, by some measures[2]. The top three- and four of the top 6- richest counties in the country are part of the Washington DC MSA[3].

Given (2) and (3), the fact that it produces more than the fifth largest MSA by population in the country (Houston) should not surprise. The top four MSA's by population and by GDP are the same, in the same order- the one that is truly punching above its weight is the Bay Area, which is the 6th largest by GDP but the 12th by population.[4]

[1]: https://en.wikipedia.org/wiki/List_of_metropolitan_statistic... [2]: https://www.cnbc.com/2018/07/26/the-10-most-and-least-educat... [3]: https://en.wikipedia.org/wiki/List_of_highest-income_countie... [4]: https://en.wikipedia.org/wiki/List_of_U.S._metropolitan_area...

TIL. Thanks! I wondered if it was IPR income declared in the DC area but activity elsewhere but it looks like things happen beyond lobbying and CEO decision making there. I've only ever dealt with NFPs In the beltway.

Note that two different MSAs comprise the Bay Area: San Francisco and San Jose. Taken collectively, they'd be 3rd in GDP and 6th in population.

How is this a vulnerability? Shouldn't one expect something like a Pareto distribution to emerge here anyways?

If any one of those 5 cities decides to regulate more aggressively, it could be an immediate, material negative hit to Uber.

Pareto distributions are common in two-sided markets with geographic factors. BUT, when there's a reasonable possibility of extra regulation imposed at the regional level... that's a big risk.

Well then, the real need is to determine if Uber's competitors suffer a similar issue or not.

The last line in the article answers that pretty well:

"One thing the filing makes clear: It wouldn’t take a global movement to stress the company’s business model. A couple big-city mayors could do that all on their own. "

Their business model is unconstrained growth to funnel money into side projects. That can't persist forever. If they just stuck to their core business they could easily be profitable as taxi 2.0.

Is the net cost of these "side projects" greater than the $1.8B Uber is losing? I'm missing that in the S-1 if so.

it doesn't cost that much to pay off a mayor.

The "Government" is real cheap in Chicago.

I'm willing to bet the low hanging fruit if you're gonna bribe a mayor from that list is São Paulo.

Yeah, a children's book in every Uber.

It's not a vulnerability. It actually shows their revenue is pretty geographically diverse.

You can't trust financial news from propaganda outlets like Slate because they only care about their agenda.

What I'm interested in out of this information (but not interested enough to look it up myself): These five cities account for 1/4 of their business, but how does the population of these five cities compare to all of the cities they do business in? Is this just a straight-line distribution?

Basically, do these five cities account for 1/4 of the total population of the cities they do business in?

I don't know the answer to your question, but some factors that might skew the distribution could be how annoying and expensive it is to keep a car or park in each city, prevalence of recreational activity that encourages or discourages a car or truck, and the like.

According to https://en.wikipedia.org/wiki/List_of_urban_areas_by_populat... Tokyo and Seoul (not on the top five Uber list but have Uber presence) are both larger than any city on Uber's top 5 list. Sao Paulo (on Ubers list) is just barely larger than Mexico City (not on the list). Los Angeles (18th largest urban area by population) and London (35th largest urban area by population) are significantly smaller, and the Bay Area is all the way down at 61st, right behind Dallas-Fort Worth (not on Uber's list).

So no, those five cities do not make up a disproportionate amount of the total urban population that Uber covers, just a disproportionate amount of the riders.

These 5 cities are not directly related to Uber. Read again.

Looking at this another way, this one-quarter revenue coming from users in five cities probably is showing these users as the power users of the app so these are probably the least likely to suddenly give up Uber because they already have a habit and Uber is already such a big part of their life.

NYC has definitely always had a taxi "habit" despite their transit. I think London is similar. On the other hand Tokyo (which has one of the best transit systems) has lots if taxis as well, but I guess Über hasn't had much traction compared to local taxis (they've always been clean and offered good service).

About 3 years ago I've had issues multiple times in Tokyo with taxi drivers not speaking english at all, thus not being able to communicate the destination. I assume Uber would smooth over that aspect as there is literally no need for communication.

I don't doubt that, but foreigners who don't speak some Japanese is a real small percentage of the people who use taxi services. If this were a general problem, then I think Über would likely do better in a place like Tokyo.

Uber is not allowed to compete in Tokyo. When you use Uber, it just calls a normal licensed taxi. The advantage is presumably the fact that it shows your destination to the driver without you having to explain it, but for locals there's literally no advantage.

"I think London is similar."

Taxi culture gets boosted a bit when the city charges you £11.50/day to drive your own car.

Alternatively: one quarter of Uber's annual losses come from those cities :D

More seriously: if they increased the cost or reduce pay rate in just those cities they could have a significant profit impact while not impacting market acquisition in the rest of the world.

1. New York

2. Los Angeles

3. San Francisco

4. London

5. São Paulo

How did you get that ordering? LA is listed first in the S1.

Physical distance from commenter? :)

Not OP, but that’s the order presented in the article.

Why is this a surprise? Its part of the reason it was even invented in the first place...

Looking at that list, the biggest threat to Uber is good (and not overcrowded) public transportation.

Why are these 5 cities Uber’s top markets? Particularly Sao Paolo.

Reply on news.ycombinator.com