The Sharing Economy: Using the internet improves efficiency over 1940s technology SHOCK!!

Disruptive Change in the Taxi Business: The Case of Uber


In most cities, the taxi industry is highly regulated and utilizes technology developed in the 1940s. Ride sharing services such as Uber and Lyft, which use modern internet-based mobile technology to connect passengers and drivers, have begun to compete with traditional taxis. This paper examines the efficiency of ride sharing services vis-a-vis taxis by comparing the capacity utilization rate of UberX drivers with that of traditional taxi drivers in five cities. The capacity utilization rate is measured by  the fraction of time a driver has a fare-paying passenger in the car while he or she is working, and by the share of total miles that drivers log in which a passenger is in their car. The main conclusion is that, in most cities with data available, UberX drivers spend a significantly higher fraction of their time, and drive a substantially higher share of miles, with a passenger in their car than do taxi drivers. Four factors likely contribute to the higher capacity utilization rate of UberX drivers: 1) Uber’s more efficient driver-passenger matching technology; 2)the larger scale of Uber than taxi companies; 3) inefficient taxi regulations; and 4) Uber’s flexible labor supply model and surge pricing more closely match supply with demand throughout the day.

by Judd Cramer, Alan B. Krueger. Download from this link.

This entry was posted in Economics and public policy, Innovation. Bookmark the permalink.

11 Responses to The Sharing Economy: Using the internet improves efficiency over 1940s technology SHOCK!!

  1. Jim says:

    Can we please stop using the term ‘sharing economy’? Uber, Luft, Air BnB etc. do not provide for a ‘sharing economy’. They don’t facilitate ‘sharing’, they facilitate transactions. It is only called the ‘sharing economy’ to make the the wanna-be-bohemians that rule the IT universe feel better about their capitalist tendencies.

    What these services actually do is provide a low transaction cost platform to match the owners of underutilised private capital (their car, their house etc.), with someone who is prepared to pay them to use it. No ‘sharing’ involved.

    These services are absolutely great, but give them another name that actually reflects what they are (perhaps PCUS (Private Capital Utilisation Services)). That should go down well with the bohemian IT crowd.

  2. LJS says:

    Beaten to the punch by Jim. Uber doesn’t refer to itself as ride sharing anymore, that was just part of the initial PR campaign. The number of people who truly shared a ride using Uber is probably low single digits.

    Without wanting to sound “anti-disruption” and attract too much millenial scorn, and with my cynics hat on, some of these services appear pretty border line exploitative. In the case of Uber it seems to very much vary from jurisdiction to jurisdiction (the ACT appears to be ahead of the curve in quickly legislating to allow established taxis to compete with Uber before the lower prices doom them,) but Uber and similar seem to be vigorously probing the edges of contractor/employee law, and hoping people suck at math when it comes to the true cost of running a car service.

    From the consumer side it’s cheap(er, usually,) and more convenient so I imagine that’s as far as most people get.

    An interesting question is if and how a distributed work force who have signed individual arbitration contracts gets organised, if they decide/realise they’ve been duped?

  3. Pappinbarra Fox says:

    My boss was UBERsmashed the other day. Paid $96 for a 3km ride durig peak hour, original estimate offered was $30.

    So UBER cheaper – I don’t think so but hey what do I know I take the train or bus everywhere.

  4. Nicholas Gruen says:

    Yes, sharing economy is a propaganda term, but collaborative consumption was worse. Still we get stuck with names that stick! That’s a feature of language – and a bug! Regarding the claim “The number of people who truly shared a ride using Uber is probably low single digits”, that’s presumably true of people just popping into town and picking someone up on the way. But Uber Pool is the default in San Francisco. You hop in a car, and pick up someone on the way. Often interesting people to talk to if you want to. And the price – super cheap – about twice or three times the cost of public transport.

    There’s little doubt that Uber has a robber baron mentality, but the thing is, most of what it does is good. The taxi industry has never given a damn about its workforce and Uber doesn’t either. But what’s there not to like about increasing demand for it?

    The evidence I’ve seen is that the users of these services are lower income than the users of their competition (though with Uber that’s probably a function of age). The really big gain for Uber in my mind is the flexibility of work – a huge deal. I’d say about half the Uber rides I take are driven by someone in their spare time. I asked one driver “how much do you do it for the money, how much for something to do”. He said 70/30.

    I replied by explaining my own motives for asking the question which were that I can imagine fancying Uber driving for a few hours every week. It would be kind of fun, driving around, meeting and talking to people. Staying away from people who spew in your car on Friday night. Anyway, I said to him I’m fairly busy, so couldn’t find the time, but if someone were to give me the time, I’d gladly do it. There are a lot of jobs people do that aren’t like that.

  5. Fyodor says:

    Nick, you should mention that Krueger has been paid by Uber to produce previous research. It doesn’t mean that this particular piece is corporate agitprop but, you know.

    On the work itself, it’s hard to critique in detail as it’s behind a paywall. Personally, I think the conclusions are bullshit.

    The principal economic innovation provided by Uber is not “using the Internet”. Its matching system adds little beyond the dispatch systems already used by taxi companies.

    No, Uber’s chief “innovations” are regulatory arbitrage – i.e. operating a taxi service without paying for licenses or complying with regulations – and shifting the fixed cost of running vehicles onto the driver. There is nothing proprietary or revolutionary about their “technology”.

    Leaving aside the fact that Uber is still running at a massive loss, both of these mean that the supply curve for Uber services is radically different from regular taxi companies, who are burdened with numerous regulations (including price-fixing). Most importantly, as you point out most drivers are only part-time, not shift-workers. That has profound implications for productivity – i.e. drivers are much more likely to operate when they can utilise their capacity (both their time and their own vehicle) most efficiently. Only a study that accounts for these important differences can even attempt to compare efficiency.

    For a bit more context on Uber, Krueger, the Democrat Party and (dis)organised labour, this is a good read.

  6. Nicholas Gruen says:

    For this week only you can download the paper from here.

    As for it being bullshit, I realise you want to unburden yourself of views that you feel strongly about – and which I agree with. But feeling one thing strongly does not exclude other facts about the world. If someone like Alan B. Krueger says that Uber drivers’ cars have higher occupancy rates than taxis, I’m going to believe him until there’s a good reason not to.

  7. LJS says:

    Nick, I’ll defer to you on how well Uber Pool works, though I maintain that ride sharing as Uber initially offered is distinct from fare splitting, which is what Uber Pool is, and that the semantics are important despite the water lapping at my ankles :)

    I guess I’m a bit on the fence in that taxi operations and their regulators have ended up unsympathetic targets – sometimes through collusion and corruption, sometimes through inertia – so a bit of a shake up is a good thing, just not sure Uber is the answer. At least I hope not. Lyft seems to take a less confrontational approach with regulators, or at least they seem to be involved in fewer disputes!

    Re: part time Uber driving, my understanding is that to actually make a living you have to work part time, driving mostly during surge pricing because outside of that you lose money, though I’m certainly open to evidence to the contrary. I’m sure it varies from place to place depending on how quick on their feet the regulators are etc.

    Having similar concerns to Fyodor I wonder what the service will look like down the track once the hopefuls have gone broke, the taxis folded, and no one will pick you up from the burbs at midday because there’s no money in it. I struggle to shake the cynicism, so I hope I’m wrong!

  8. John walker says:

    We use Uber much more in Canberra than in Sydney. Taxis in Canberra , for whatever reason, were pretty bad, often dirty, slow to turn up with a really infuriating booking system and unreliable. I.e its not really about the price, rather it’s about : the Uber cars turn up in a few minutes and are new and clean ( similar situation on the Gold Coast.)

    In comparison in Sydney taxis are mostly OK and, as others have noted ,Uber can be expensive (and hard to figure) .

  9. Nicholas Gruen says:

    It may get more expensive in the burbs if they’re being cross-subsidised – that’s fine with me. Not sure why they require welfare.

    And on the model, the new technologies enable lots of different models – and we can choose – though there are natural monopolies in the network. That’s why we shouldn’t ratify the TPP. We’ll need to clean up whatever mess gets made and we don’t want Uber running off to ISDS (well that and other reasons for not signing the TPP.)

  10. pablo says:

    I agree with the public transport bias. Maybe if all uber vehicles were electric via solar or wind power, the ‘bohemian IT crowd’ would gladly pay a little extra to cover the uber-ETS. Now there’s a sales pitch!

Comments are closed.