An Uber Impact | Hacker News Reader

An Uber Impact

blog.uber.com

71 pointsmp3jeep0161 comments

Politically speaking, it is very, very smart that Uber is doing these sorts of studies. As a policy wonk, I wish they were a bit more open about their datasets, analyses, and methodology, but I can see how that would just hand their opponents chances to twist it.

Re: jobs created, note that Uber (very specifically, is my guess) allows us to assume FTE when they say "jobs," but is probably more likely referring to the total number of driver profiles created and used at least once. A Daily/Monthly Driver metric would be more meaningful here.

I'd also be interested to see the impact on parking (especially at peak times) if people are using Uber over driving somewhere and leaving their car parked (idle/slack resource) there.

PS - the lack of a footnote attached to (what seems to be a copy/paste) 1 in the subheading is grating haha

I assume FTE = Full-time equivalent (http://en.wikipedia.org/wiki/Full-time_equivalent) in this context.

Yes, that's correct.

> but I can see how that would just hand their opponents chances to twist it.

if opponents can twist it why shouldn't i believe uber is twisting it to begin with?

I find this GoT quote to be fitting here—

Eddard Stark: Lord Baelish, perhaps I was wrong to distrust you. Petyr Baelish: Distrusting me was the wisest thing you've done since you climbed off your horse.

$90K in (gross) revenue (in NYC), before taking into account operating costs (fuel, lease expenses), depreciation, healthcare, self-employment taxes, etc.

It's off the cuff and I'm sure I'm missing a number of categories, while realizing that you can deduct a certain number as business expenses, but I'm fairly certain you're really (edit) not (/edit) seeing anything close to $90K in the end.

I'd also like to see how many hours on average someone making $90K in gross revenue in NYC is working.

I have no horse in this particular race, but taxi drivers also pay for fuel, their own health care, and I think generally self-employment taxes as well. So the actual difference is just the cost of operating the vehicle in wear and tear and depreciation. The latter of which is mostly relevant only if they purchase the car exclusively for this purpose.

So the $90k might be deceptive because it's not net, but it's a sensible way to compare to what traditional taxi drivers make given uncontrollable variability in car ownership expense.

20k jobs created? Is hired car use really going up that much? If I had to guess, I'd say a handful of new jobs are created while the rest are just taken from existing taxi companies.

Not that I think there's anything wrong with that. I just think the stat is pretty misleading.

Keep in mind that the number of taxis on the road - and thus the number of jobs - is limited in many cities to the number of "medallions" issued (which represents the number of taxis that can be operating in the city at the same time.)

In major cities like SF and NY, the number of drivers is almost certainly actually increasing, as Uber has basically rendered the medallion (and the protection it provides) somewhere near useless for many classes of fares. Some jobs are certainly being moved from black car and taxi companies who can't compete with Uber on overhead, but the taxi market has been highly regulated to the point of limiting job creation for many years.

Uber is a great service - head and shoulders above the entrenched taxi industry. Customers like them better, the drivers like them better; but they are lucky that Goldman Sachs had never created a medallion investment vehicle. If upper middle class doctors and lawyers had seen a drop in investment return you know Uber's "stealing" from the medallion holders would have been stopped.

"In Seattle the number of arrests for DUI to decrease by more than 10 percent." They better be careful here. If they start cutting significantly into state regulators revenue stream they will be stopped.

Uber's best bet is a "negotiated settlement" with local and state authorities to kick-back some of the cash they are diverting.

> I'd say a handful of new jobs are created while the rest are just taken

This is true of virtually all claimed job creation numbers.

I personally use Uber at least 20-50x as often as I would use taxis. That is, for each Uber trip I take, I would take only 1/20 to 1/50 of the same trips via taxi given the choice.

I think this accounts for the increased use. I would go out of my way to avoid taking a Taxi because they're so inconvenient and opaque.

I still don't plan to start using Uber extensively because I'm used to walking at this point, but the combination of transparent upfront pricing, accountability, ease of use, speed of pickup, and seemingly typically lower pricing is a compelling case versus zip car. Definitely an actual competitor with walking and public transit, whereas I see Taxis as only really a competitor with driving.

Funny way to arrange this comparison. To me it reads: for every Uber trip I take from my apartment in the Mission to my office in Soma, I would ride a taxi from 24th to 23rd street.

Same here. While I still consider a Taxi as something you only take when you're drunk and have to get home from a party, I've gotten used to taking an Uber or Sidecar for pretty much any trip that would require a muni or bart here in SF. It's gotten to the point that I am downright spoiled.

smells like bullshit

20k "jobs" / mo gives 240k jobs / year putting them in the top 20 largest private employers in the us [1] with just one year's claimed growth

edit: there may well have been a month where, for sufficiently lax definitions of "created" and "job" it is indeed true that "20,000 JOBS CREATED ON THE UBER PLATFORM EVERY MONTH" but I'm highly skeptical this claim is close to true in anything but the most generous sense

edit2: to their credit, they reported median income and not mean

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

These stats are mindblowing. Uber is available to a little less than half of all Americans and with an average pickup time of < 10 mins. I'm amazed they were able to accomplish this in 4 years. Travis Kalanick is an insane entrepreneur.

I would give it up to their data science and machine learning team. They turned an idea into an executional reality and optimized it like no other.

I just heard Kevin Novak (@novakkm) speak @ Geekdom SF and he shared the approach he took along their journey.

Is this talk available online anywhere? I'd love to hear it, or at least a broad-strokes summary.

Here was the event page. http://www.meetup.com/SF-Data-Science/events/183290442/

No video up yet. I think it should be up soon. If I see it come up I will link it as a new post.

I'm living in one of the top cities in the Bay Area and they don't yet cover there.

Bay Area is not a country, not even a state, so saying "one of the top cities" doesn't make any sense.

My point being they are stating that their service is available to 137M people. The bay area is one of the top population centers in the US so I'm surprised they say they can get 40-50% coverage.

It's a state-sized region with close to 100 cities, so it kind of does make sense (to the extent that "one of the top cities" is meaningful anywhere).

Uber is pretty great, and I enjoy using it, but many of these statistics are pretty weak and out of context. As others have mentioned, using "income" to describe "revenue" is inaccurate. You can't compare making $90k driving an Uber to $90k at a desk job with a completely different set of expenses.

While the statistics used to determine the DUI rate in SF and Taxi crime in Chicago are beyond the levels of my knowledge and I can't verify their correctness, the calculations certainly seem a bit suspect. For example, as a pair city when doing a discontinuity test, they chose San Francisco, a city which notably, has Uber. The post doesn't mention a confidence interval or P value in a clear explicit way (outside of the table, with unexplained column headers).[1]

In the infographic, they claim these statistics as facts without mention of any underlying confidence intervals. I'm pro Uber, but this post seems to contain a lot of fuzzy math. It'd be interesting if someone with a statistics background could confirm or deny these suspicions...

[1] http://blog.uber.com/DUIratesdecline

For taxis, medallions are usually leased to drivers (few could afford one). This takes much of their revenue -- perhaps half or more surrendered to a medallion holding company just for the privilege of driving a taxi.

New York City presentation suggests a driver with $90k revenue has $41k expenses. [1]

[1] http://www.nyc.gov/html/tlc/downloads/pdf/presentation.pdf

They did give a P value, it's the final column labelled Pr (>|t|). The findings are significant.

They also gave the standard error, which you can use to compute your own confidence interval (95% would be about 2x, or ranging from no effect to a 20% reduction)

In this thread everyone discusses the actual numbers.

The numbers are not the interesting thing here - Uber seems to get battle-ready for lobby-work.

These numbers - correct or not - will be the ammunition their PR and lobby teams will use in the next months.

Absolutely correct, which is why it looks like an infographic, not a research paper.

But those of us here can parse the language quite closely and get some actual information out of it, as well as noting the use of shadows to hide and reveal.

I doubt the automotive pun was intended, but if it was, it's in poor taste, and if it wasn't, it's a nasty oversight - considering the last time people read a headline about an "Uber impact," it was because a little girl was killed. It's extremely important to consider such things in this kind of language.

When I first read the headline I thought it was a press release about an accident.

Me too. In fact, there was an accident last night in my neighborhood involving a black car and a drunk driver and I thought it was about that, despite Uber (kindly) chiming in on a local blog to say it wasn't one of their drivers. Thought this was some kind of horrible reversal.

> For example, the median income on uberX is more than $90,000/year/driver in New York and more than $74,000/year/driver in San Francisco.

That is impressive. Even if you detect fuel and maintenance and health care, it comes to around 50k a year or so. That's quite a bit given that I assume drivers can choose to work flexible hours and possibly not as stressed out (compared to software engineer with 70k salary).

I really wish they could give more data on how they reached their median income stats. I realize they have no incentive to release that. Curious how Lyft and Sidecar measure up as well.

I've conversed with numerous Sidecar drivers and they earn no more than $30/hr-- and that's before deducting expenses (gas, wear & tear, etc.). And most of them work part-time

Also, if an über driver makes $70k++ and a software engineer makes $100k++ base... that's kind of sad.

The Uber driver pays for his car's maintenance and fuel.

And the financing costs for the car to begin with, and in addition doesn't receive the health and fringe benefits of salaried employment.

I agree. That is sad. Hence why I ask for more data. I suspect it's not really painting the whole picture.

What's really significant is that the tech industry is slowly creating new job markets via 'app-driven services'. Doordash, grubhub, amazon, and google's instant delivery services are other examples that can contribute to this new trend.

The technology infrastructure for instant/convenient buying, selling, and delivery of goods and services has been built out. Now it can truly scale by staffing massive amounts of people to actually carry out the tasks surrounding it.

What's also really significant is that the majority of these jobs are minimum-wage, minimum-training, completely-replaceable jobs. What these people are doing is essentially an interface between software and reality - they are human robots.

Just a temporary hold over until we have proper automated delivery machinery. And we did just get that self driving car article.

Well, here's some food for thought on all those new "job markets" created:

http://www.fastcompany.com/3027355/pixel-and-dimed-on-not-ge...

Uber truly is quite a leap forward in transportation.

But I am currently perfectly fine with them running into all the regulatory roadblocks. They have so far shown zero interest in working with the various constituents to figure out what the right structure is for their services. It's easy to say that taxis suck and regulation squashes innovation. But some of this stuff exists for good reason and until Uber acknowledges that, I'm OK with hurdles they are facing.

43% of Americans are now covered by Uber. Seems like they've hit most of the densely populated areas. I'm sure it will take much longer than that to hit the less populated areas, if ever. For most cities (read: not large and dense), Uber will probably not work.

Uber just came to my college town (San Luis Obispo) and their rates are absurdly high compared to anything else. Even if Uber drivers were less than 10 minutes away, I would wait the extra 10 for a cab and save myself the ~$20.

It's interesting to look at these figures while keeping in mind Google's self-driving car project and their large investment in Uber. In 10 years Uber will be able to replace all those jobs they've created with self-driving cars.

90k is not much in NYC when you take into account how much a driver will need to spend to maintain a car that does livery all day long in NYC.

Does this mean more people are taking Taxis?

That, I think is the key question and no data is given. IMHO, I really doubt Uber has expanded the market, but instead taken a share of it, so 'creating' 20,000 jobs is at the expense of X thousand others.

That's a widespread fear among Uber detractors, as seen in action outside of US in the past few weeks.

I think without a third-party assessment of reliable data on the market every opinion on the subject is based on FUD. Unfortunately, at this time we seem to be missing both reliable data and an "objective" assessment.

Anecdotally I take way way more Lyft's and Uber's now than I did taxi's 3 years ago with very little change in my expendable income. FYI, im in LA.

My guess is yes. Uber makes it so easy to quickly and painlessly get a ride that I find myself using the service much more often than I would a normal Taxi in the past. From casual observation I find this to be true for the majority of my friends and even family members as well.

Normal Taxi services are a pain in the ass to book, not connected to my credit card (so you have to go through the whole guilt trip for not having cash, takes longer, etc), and for whatever reason I always had bad experiences with them at least half the time. None of these things are issues with Uber, leading to much more usage on my part. It's just so damn easy, and even satisfying, to tap a few times and have a car magically show up.

I'm surprised taxi services are so far behind here. Why can't I request a taxi in a single click, select the vehicle, charge it to my account, see the location of the taxi on a map and the ETA until it arrives? Why can't I see the route it'll take, or modify the route myself and set a waypoint?

I'm sure some taxi services offer a couple of these features, but none of the ones that I've used. I think one had online booking, but it was broken and I had to find a phone, and then I got no answer, and then I was on hold for 5 minutes.

Why can't I set a custom alert, so I get notified when the taxi is 2 minutes away? That way I don't have to keep looking out the window, or spend 15 minutes sitting on a curb, wondering if they forgot, and I'm going to be late.

If taxis are losing business to Uber, it's their own fault, and lack of innovation.

> I'm surprised taxi services are so far behind here. Why can't I request a taxi in a single click, select the vehicle, charge it to my account, see the location of the taxi on a map and the ETA until it arrives

You can, it's called Uber. It's why it exists :)

And in Chicago at least, you can summon a cab with Uber just like you can a black car or SUV.

In most cities can't you just walk out on the street, stick out your arm and have a taxi at least as quickly as it takes to book and wait for one through the Uber app? I guess it's much more useful when the streets aren't filled with cabs like in the suburbs or at night.

I'm in Ireland, and while I don't use Uber (too expensive) I keep telling this story to explain why I use Hailo a lot: Around 3am one Saturday after a family party my parents were worried about getting a taxi, because they had heard lots of horror stories of driver stealing things etc, but were prepared to catch one from the taxi rank across the road. I pulled up Hailo, was able to book (and know the driver's name a face), and have the taxi arrive before they could cross the road (in a suburban town). And since I was paying for it by card they just got in and went home and I didn't have to worry about a thing. These things sold me completely on the system.

I live in a pretty central part of Chicago, and have to walk at least 3-4 blocks to get somewhere where hailing a cab within 10 minutes is a reasonable expectation.

i still dont get how it has 'created' jobs? i get how uber is convenient/practical for a customer; how is it thaat beneficial for the taxi driver?

I guess this is actual disruption - lots of people are already complaining about loosing jobs :).

The taxi drivers in Brussels, Berlin and Paris are complaining the most!

Reply on news.ycombinator.com