So-called “surge pricing” is not the main thing to worry about with Uber. Investors who value the ethically challenged firm at an astonishing $40B have made a cynical (also ethically challenged) bet that “network effects” will permit the firm to basically own the 21st century successor to the taxi industry. Our main concern should be to ensure investors do not win that bet. In particular, public policy should focus on encouraging “multihoming”, where drivers advertise availability over several competing platforms (Uber, Lyft, Sidecar, etc.) simultaneously. Municipalities might also consider requiring that ride-sharing platforms support standard APIs that would enable Kayak-like metaplatforms to emerge. Or municipalities might offer such applications to the public directly. As usual, the question here is not “regulation” vs “deregulation”, but smart regulation to ensure a high-quality competitive marketplace. Fortunately, the right of municipalities to regulate transportation services is well established, so it should be straightforward for cities to impose conditions like nonexclusivity and publication of fares in standardized formats.

I don’t care all that much about Uber’s “surge pricing” — its practice of increasing its usual fare schedule by multiples during periods of high demand. I do, however, care about the damage done by a kind of idiot dogmatism that hijacks the name “economics”. Uber’s surge pricing may or may not serve Uber’s objectives of profit maximization and world domination. It may or may not increase “consumer welfare”. But it is not unambiguously a good practice, either from the perspective of the firm or as a matter of economic analysis. Its pricing practices impose tradeoffs that must be addressed with reference to actual, on-the-ground circumstances. Among prominent academic economists there may well be a (research-free) consensus that surge pricing promotes consumer welfare (ht Adam Ozimek), but that reflects the crude selection bias of the profession much more than actual analysis of the issue. The dogmatism which has arisen in support of Uber’s surge pricing is quite analogous to the case of urban rent regulation, a domain in which there is incredible heterogeneity across localities and nations, both of circumstance and policy, and a wide range of legitimate values that conflict and must be reconciled. (Here’s an interesting case in the news today, in Spain, ht Matt Yglesias.) Almost as a rite of passage, economists drone in every intro course that rent controls are bad. By preventing price signals from working their magicks, they prevent the explosion of real-estate supply that a truly free market would deliver. This is stated as uncontroversial fact even while economists who research and opine prominently on housing policy have endlessly documented that housing supply is not in fact price-elastic in the prosperous cities where rent controls are typically imposed. None of this is to say that rent controls are good or bad, or that non-price barriers to construction are good or bad. These are complex questions involving competing values textured by local circumstance. They deserve bespoke analysis, not pat dogma imposed by distant central planners economics professors.

Anyway, surge. The excellent Tim Lee grapples with the miserable dogmatism that surrounds the subject here:

The thing Lyft customers seem to hate the most about Uber is surge pricing. That’s when Uber automatically raises prices during periods of high demand…

The economic argument for surge pricing is impeccable: varying prices helps to balance supply and demand, ensuring that people who really need a ride can always get one. But businesses have to take customer preferences into account whether or not they’re rational. So it might make sense for Uber to adopt Lyft’s softer approach to demand-based pricing.

As in the case of rent control, the stereotyped economist’s case for surge pricing is based on a conjectured elasticity of supply. With higher prices, the reasoning goes, more drivers will hit the road, more customers will be served, and the world will be better off. And that’s a good case, as far as it goes. But it doesn’t go very far, without some empirical analysis. It doesn’t justify Uber’s actual practice of surge pricing, which is far from the transparent auction our stereotyped economist seems to imagine. It doesn’t account for the trade-offs imposed by price-rationing (as opposed to time- or lottery- rationing), both between customers and for the public at large.

First, how price elastic is driver supply? If we presume that Uber is a Walrasian auctioneer, a disinterested matchmaker of supply and demand, apparently supply is not very elastic. Uber surges prices by multiples, two, three, even four times “typical” pricing in periods of high demand. That’s extraordinary! If supply were in fact elastic, small increases in price would lead to large increases in supply. The supply-centered case for dynamic pricing is persuasive in direct proportion to actual elasticity of supply. Uber’s behavior suggests that the supply-based case is not so strong. Of course, we cannot make very strong inferences about driver supply from Uber’s behavior, because they are not in fact a disinterested Walrasian auctioneer. When Uber surges, it dramatically raises its own prices and earns a lot more money per ride, whether ride supply increases not at all, or whether it spikes so much that drivers end up competing heavily for riders and suffer long vacancies. As a profit maximizer, Uber’s incentives are to impose surges primarily as a function of demand, and say nice things about supply to con economists and journalists.

Suppose, then, that supply is not elastic. Is there any problem with Uber “charging what the market will bear”? Even for inelastically supplied goods, the stereotyped Econ 101 professor recommends price-rationing, as that should ensure that the scarce supply goes to those who most value it. Unfortunately, the argument for price-rationing (as opposed to lottery-rationing, or queue-rationing) of goods as being welfare-maximizing depends (at the very least) upon a rough equality of wealth so that interpersonal dollar values can stand in for interpersonal welfare comparisons. In an unequal society, price rationing ensures disproportionate access by the rich, even when they value a good or service relatively little. There is no solid case that price-rationing is optimal or even remotely a good idea when dispersion of purchasing power is very large. I’ve written about this, as has Matt Yglesias very recently. Matt Bruenig has two excellent posts relating this point to Uber specifically (as well as another post on ethical claims about Uber’s pricing). For a deep dive into how distributional concerns affect welfare-economic intuitions under perfectly orthodox economic analysis, I’ll recommend my own welfare economics series. It’s easy to write-off Uber controversializing as a masturbatory first-world problem among hipsters, rather than a pressing question of wealth and poverty. That’s a mistake. There’s little question that “app-mediated” car provision will soon replace conventional taxis, because it is a much higher quality product. Poor people are in fact one of the main clienteles of traditional taxis in the US, since nonpoor households typically own cars and use taxis primarily when traveling. As the industry transitions, poor people will be hit very immediately by whatever practices become standard. In an unequal society, distributional effects are a first-order concern.

Suppose you just don’t care about distribution and you favor price-rationing of scarce goods over alternative schemes full stop. Then you should still be troubled by Uber’s surges, because Uber itself is a cartel. The actual service providers are individual drivers. When Uber “surges”, it raises prices across its whole fleet of drivers. Yes, Uber faces competition, from traditional cabs, and (depending on the city) from other startups. But between perfect competition and monopoly, there are a lot of degrees of pricing power. In many cities, Uber already has a lot of pricing power, and that may increase over time, depending on how today’s competitive battles shake out. Like any potential monopolist, Uber’s incentives will be to “surge” to a price that is higher than the output-maximizing price that would obtain in a competitive market. There is no technical reason why Uber needs to be organized like a cartel. In fact, one of its competitors, Sidecar, allows each driver to set her own price, encouraging competition within the service. Like Sidecar, Uber claims to be a “platform”, and disavows any employment relationship with or liability for the actions of its drivers. Fine. It makes a market for independent contractors. Then why on earth do “free market economists” applaud when it forces those contractors to coordinate price increases? Why would antitrust laws even tolerate that?

Finally, we need to consider questions of economic calculation. In macroeconomics, we sometimes face tradeoffs between an increasing and unpredictably variable price-level and full employment. Wisely or not, our current policy is to stabilize the price level, even at short-term cost to output and employment, because stable prices enable longer-term economic calculation. That vague good, not visible on a supply/demand diagram, is deemed worth very large sacrifices. The same concern exists in a microeconomic context. If the “ride-sharing revolution” really takes hold, a lot of us will have decisions to make about whether to own a car or rely upon the Sidecars, Lyfts, and Ubers of the world to take us to work every day. To make those calculations, we will need something like predictable pricing. Commuting to our minimum wage jobs (average is over!) by Uber may be OK at standard pricing, but not so OK on a surge. In the desperate utopia of the “free-market economist”, there is always a solution to this problem. We can define futures markets on Uber trips, and so hedge our exposure to price volatility! In practice that is not so likely. For many people, time-uncertainty may be more tolerable than price-uncertainty in making future plans. If this weren’t the case, congestion pricing of roads would be much more popular than it is. Just as we leave home early now to account for the time we’ll spend parked on the expressway, we can summon a ride early to ensure we arrive on time even when there is no car immediately available.

It’s clear that in a lot of contexts, people have a strong preference for price-predictability over immediate access. The vast majority of services that we purchase and consume are not price-rationed in any fine-grained way. If your hairdresser or auto mechanic is busy, you get penciled in for next week. She doesn’t tell you she’ll fit you in tomorrow at double her usual rate. There are, as far as I know, no regulatory or technological impediments to more dynamic pricing schemes for everyday services. Even in the antediluvian, pre-app world, less routine sorts of service provision like hotels did price dynamically. People seem to tolerate dynamic prices of services they consume sporadically or as a discretionary luxury, but prefer price predictability and time uncertainty for services they consume routinely. You’d think economists of all people would “mark their beliefs to market”, but the stereotyped practitioners who define what Tim Lee calls “impeccable” economics are in fact wide-eyed utopians. They look past actual preferences that consumers express in purchasing behavior, and that providers reflect in pricing behavior, to a hypermarketized alternative reality where interactions are governed in a very fine-grained way by price-signals and market incentives. It’s not clear that very many humans actually want to live in their world. Lee expresses the incoherence of the “impeccable” economist very well when he writes, “businesses have to take customer preferences into account whether or not they’re rational.” In theory, of course, customer preferences can be inconsistent, but they can never be irrational. Economics as a discipline takes human preferences as given, and defines rationality as action that maximizes the degree to which those preferences are satisfied. But the “impeccable” economist so privileges stereotyped market mechanisms as analyzed in a deracinated fictional theoryworld that any preferences not consistent with means chosen a priori get deemed irrational. That way of thinking may be “impeccable”, but it is the opposite of good economics.

I don’t want to be too negative. As I said at the start, surge pricing per se is really not the major concern with Uber. Our efforts should be devoted to ensuring that no single price-coordinating “platform” dominates the nascent on-demand transportation industry. There is a solid case for using price to incentivize ride supply, or even to ration relatively fixed supply. Price-rationing may be welfare maximizing, among the options available to a firm like Uber. But there is also a solid case against, for preferring predictable pricing and lottery- or time-rationing. Even if we stipulate that price rationing is best, it’s hard to think of any consumer-welfare rationale for Uber-style fleet-wide surge pricing rather than a Sidecar-style competitive auction among drivers. Sidecar’s competitive provision is less prone to consumer-welfare-destructive monopoly rent extraction than Uber’s coordinated pricing. Sidecar’s system also permits heterogeneous strategies among drivers, allowing the market to decide and perhaps segment, as some users pay up for immediacy, while other users reward drivers who hew to stable prices by preferring them even when demand is slack.

Update History:

  • 30-Dec-2014, 11:15 p.m. EST: “to ensure that investors”; “over several competing platforms”; ” while the economists who research and opine most prominently on housing policy have endlessly documented the fact that”; “encouraging competition within the platform service“; “defines rationality as action that maximizes the degree to which those preferences are met satisfied“; “it’s hard to think of a any consumer-welfare rationale”
  • 31-Dec-2014, 10:15 a.m. EST: Added link to Dempsey paper, both as related academic work and as cite for claim that taxis significantly used by the poor.
  • 31-Dec-2014, 10:30 a.m. EST: Added link to third Matt Bruenig post on Uber.
  • 18-Jan-2015, 7:05 a.m. PST: “right rite of passage”, thanks Bob Jansen and commenter Bruce

21 Responses to “Surge!”

  1. Harald K writes:

    I had hoped you would write on Uber for a while. While there is much interesting here, I’m confused by some things. You write “Poor people are in fact one of the main clienteles of traditional taxis in the US”. Really??

    That is radically different than what it seems like in Norway, at least. Here, it’s rich people who take taxis, for much the same reason it’s rich people who have private chauffeurs. The poorer you are, the more you take trouble to arrange/use other means of transport when you don’t have a car available.

    True, taxis are presumably cheaper in the US. But cars are also far more expensive here in Norway, so how can it be so different?

    Also, you write about the nascent on-demand transportation industry. But what are traditional taxis, if not an on-demand transportation industry? I can order traditional taxis via phones, websites or apps already. I fail to see what new Uber brings to the table, besides “union”-busting with a cartel on top.

  2. Anon. writes:

    >the stereotyped Econ 101 professor recommends price-rationing, as that should ensure that the scarce supply goes to those who most value it. Unfortunately, the argument for price-rationing (as opposed to lottery-rationing, or queue-rationing) of goods as being welfare-maximizing depends (at the very least) upon a rough equality of wealth so that interpersonal dollar values can stand in for interpersonal welfare comparisons.

    This is a bait-and-switch. The first sentence clearly talks about value, which is unambiguous because of revealed preference. The second one suddenly transports us to the magical world of utils. It doesn’t matter if we can do welfare comparisons; value comparisons remain unaffected.

    Even worse, you ignore the welfare of the driver in the equation! (what do you think that says about you?) The driver is obviously likely going to be the poorest person in this little triangle you have constructed, so even in terms of welfare the surge pricing is unambiguously good.

    As for Uber’s pricing power and top-down approach you certainly have a point, but this is a field where the cost of entry is virtually zero. This market has none of the features of a natural monopoly, and the talk of network effects is mostly nonsense. So in the long term I don’t think this is a real issue.

  3. Nick Rowe writes:

    Three points about symmetry:

    1. Wot about the drivers? Their welfare matters too. Are they richer or poorer than their customers? Wot about their welfare in a surge?

    2. For every surge up, there must be a surge down, when price is below the mean price. If we ban surges up we ban surges down. Do we want to ban surges down?

    3. Let there be two different goods. Both have the same, perfectly inelastic supply. But good A has permanently higher demand than good B. Should we insist that both goods have the same price?

    Plus: It is a good question you ask about why people do seem to dislike time-varying prices. In the 1970’s, we thought that Azariades-Bailey-Gordon had the answer, in terms of simple risk-aversion. (They were trying to explain sticky wages, in terms of risk-averse workers, but it’s the same thing). But their models failed, and were abandoned by macroeconomists. Maybe it reflects a (rational) dislike of holdup costs in bilateral monopoly-monopsony?

  4. Anon2 writes:

    @Anon. 5:02 am

    You need to read more of Steve’s work — in particular the welfare economics series he links too.

    “The first sentence clearly talks about value, which is unambiguous because of revealed preference. The second one suddenly transports us to the magical world of utils.”

    I’m guessing you don’t have an economics background. The concept of utils was developed to address the problem that dollars don’t represent value. Utils clarify the problem (and likely impossibility) of interpersonal comparisons of value.

    The driver owns a car, so it is an extraordinarily strong and counterintuitive assumption that the driver is the poorest person in the triangle, but all would agree that the driver’s welfare must be taken into account.

    I think you need to study network effects. The whole point is that once most people have coordinated on a particular platform or product, that creates monopoly power, because of the difficulty of effecting a mass transition to a new platform — even one that is unarguably more efficient if the transition could be effected. Thus, the whole point of network effects is that the cost of entry is not measured by the costs of creating the platform, but the costs of coordinating a mass transition to using it.

  5. Lord writes:

    Your argument tends to undercut itself. Since the market is between those affluent enough not to limit their usage strictly to public transportation and walking and not affluent enough to have or rent their own transportation, their wealth distribution is probably not that significant. While any monopoly is undesirable, since the alternatives you provide already exist, it is hard to see why the customers won’t determine what is successful, or that one solution will be preferred since the uses vary so much between sporadic travel and daily commuting, but yes, let a multiplicity flower.

  6. Alex Godofsky writes:

    If the “ride-sharing revolution” really takes hold, a lot of us will have decisions to make about whether to own a car or rely upon the Sidecars, Lyfts, and Ubers of the world to take us to work every day. To make those calculations, we will need something like predictable pricing. Commuting to our minimum wage jobs (average is over!) by Uber may be OK at standard pricing, but not so OK on a surge.

    In a world without surge pricing, rather than unpredictable pricing we get unpredictable availability. The latter is at least as problematic, if not more problematic for people who choose not to own a car.

  7. Brett writes:

    There is no solid case that price-rationing is optimal or even remotely a good idea when dispersion of purchasing power is very large.

    The lower purchasing power customers still present a potential market, and in a highly unequal society odds are they’ll present a very large potential market. There are a lot of products that were originally aimed affluent customers on a high-price, high-margin basis that were offered to larger bases of poorer customers – see organic produce.

    Price-rationing becomes more important the longer the period of time, because supply becomes more elastic over a longer period of time. That’s the problem with rent control – in the present, it’s nice for people who might otherwise face evictions and generally doesn’t hurt landlords much as long as they’re still making enough off rent to offset the costs. But it makes it much less attractive to build and add new “supply” that might be potentially slapped with rent control rules, particularly new density and bigger buildings.


    This market has none of the features of a natural monopoly, and the talk of network effects is mostly nonsense. So in the long term I don’t think this is a real issue.

    It’s not nonsense, but it’s neglecting to mention the way that network effects can help new firms as well. Network effects created MySpace, and then helped create Facebook which marginalized MySpace.

  8. Tom Davies writes:

    Steve, in my experience surges are not always integer multiples. At 2:30am on new years day in Newtown, Sydney, there was a 1.4x surge.

    Harald, in Sydney, at least, Uber brings far more than “union busting” (taxi drivers aren’t unionised). They bring transparency to when and whether a ride will actually come — when you order a taxi here, you are told that the ‘first available’ will be sent, but you have *no* way of finding out when that will be or whether any driver wants your fare. That UberX is c. 20% cheaper than a taxi, and the drivers are less smelly is secondary to that. Of course Norway may have a much better taxi system (I don’t recall any complaints in Knausgaard :-) ) but in Australia Uber is great (we don’t have Lyft here yet, as far as I know)

  9. Morgan Warstler writes:

    I think Steve, you have an emotional itch, a psychological discomfort with being cavalier towards the suffering of the poor.

    But then NO MATTER WHAT THE TOPIC, no matter what the policy, you try to scratch that same itch. The end result is that you lose the 30K view.

    Instead, I’d like you to come up with a “Yes let’s go FULL THROTTLE FREE MARKET CAPITALISM, but… and then have one SINGLE policy that violates it and scratches Steve’s itch until he literally says, “thanks, perfect, no more scratching necessary.”

    Now I’m not saying I’m definitely right, but I’d like you to prove me wrong by actually writing down a proposal…

    IF we did X,Y,Z – this market basket of safetynet this once omnibus plan, THEN, I’d stop worrying about my ith when I look at policy.

    Would it just be cash transfer? If not why? Are there things that people can’t handle for themselves if they have the cash to decide?

    Why give it to everyone and not just top everyone up?

    If in the end you prove that you need 50 different itc scratching policies I’d like you to show your work to get there.

  10. acarraro writes:

    I fully share the concern about network effects. We really don’t want Uber to become a monopolist. But this concern seems fairly minor at this point… There is a lot of competition in the space and the barriers to entry seem fairly small to me… Registering on a new app seems fairly easy to do and as long as the ensure Uber cannot require exclusivity without actually employing the drivers I think we will be okay.

    I don’t think surge pricing is a big problem personally. I doubt that surge pricing really affects the average price you pay for a ride. Are you against happy hours in bars? Or against congestion pricing? Moving prices to try and manage demand seems entirely sensible to me. I also doubt that taxis would ever be an efficient way to transport people to a minimum wage job. So I don’t think that’s a valid concern.

    Maybe the next competitor will try happy hour fares, with heavy discounts at some times (a little like sale I would guess, are you against sale as well?). Maybe it will be more popular. But is it really different?

  11. Foolish Jordan writes:

    I love that Uber uses surge pricing. We’ll finally get some empirical data on whether price rationing actually works outside financial markets. Let’s argue about whether the data is tainted by various network, monopoly or other effects once we have it. Uber might somehow get monopoly control of all the taxi markets in all the cities and cause untold harm to the poor and helpless … or they might not. If they do start causing all this harm, then we can pass some laws to undo it. I don’t see the need to panic today.

  12. […] Uber wants to maximize profitability, but both drivers and riders value predictability. (see Steve Randy Waldman’s excellent discussion). Economists sometimes (not always) have a blind spot in only considering efficiency in terms of […]

  13. […] Steve Randy Waldman’s excellent discussion). Economists sometimes (not always) have a blind spot in only considering efficiency in terms of […]

  14. […] Steve Randy Waldman’s excellent discussion). Economists sometimes (not always) have a blind spot in only considering efficiency in terms of […]

  15. […] Steve Randy Waldman’s excellent discussion). Economists sometimes (not always) have a blind spot in only considering efficiency in terms of […]

  16. […] Steve Randy Waldman’s excellent discussion). Economists sometimes (not always) have a blind spot in only considering efficiency in terms of […]

  17. Hubert Horan writes:

    While I don’t disagree with most of your specific points, I would like to challenge your two central claims:
    (A) “Uber’s surge pricing may or may not serve Uber’s objectives of profit maximization and world domination. It may or may not increase “consumer welfare”. But it is not unambiguously a good practice”. I disagree. Based on my knowledge of how variable pricing can increase efficiency and consumer welfare (from 30 years in aviation) and a strong basic understanding of urban transport economics (i.e., a very simple version of the “bespoke” analysis you requested). I would categorically state that Uber’s pricing approach is unambiguously a bad practice, and that Uber’s long term growth would materially worsen resource allocation and efficiency in the Urban Car Service Industry
    (B) I think your focus on “a kind of idiot dogmatism that hijacks the name “economics” seriously understates what is going on. The problem with the “Uber debate” is not that a few ideologically motivated “economics” writers have failed to provide empirical data supporting their claims. The problem is that the entire public discussion has been engineered to be utterly divorced from anything remotely similar to rational analysis, and has become totally impervious to any attempt to raise factual claims about financial valuations, urban transport demand, business startup dynamics, technological innovations, sustainable competitive advantage or any “complex questions involving competing values textured by local circumstance”. The investors and senior executives of Uber are playing a PR/propaganda game, designed to render you and anyone who reads internet sites like this totally impotent. I think the sins of the “economists” you mention are a trivial part of the problem.
    Let’s start with the Surge Pricing issues in your post, and show how simple comparisons of empirical evidence between taxi markets and pricing systems with better established airline markets and pricing systems leads to different conclusions than the ones you drew. One can immediately notice that absolutely none of “economics” authorities extolling Surge Pricing (or Uber in general) have any expertise in either urban transport economics or the variable pricing systems (airline, hotel, etc.) actually in use, and you won’t find a single article that attempts to estimate the supposed inefficiencies created by existing Urban Car Service pricing, or attempts to explain specifically how Uber’s Surge Pricing will eliminate these inefficiencies.
    1. Airline pricing is100% transparent and maximizes market information and welfare; Uber’s approach is designed to reduce and distort price information and to eliminate consumers contractual protections on pricing. Variable pricing works totally differently in a market like urban transport where consumers can only make last minute purchase decisions than in an advance purchase market like airlines or hotels. Airlines publish both their cheap restricted fares and their higher last minute fares in advance; customers know the exact fares on offer well in advance of making their travel plans and at any point in time the exact same fares are available to everyone. Passengers can calculate the exact savings possible if they shift to off-peak flights or more highly restricted tickets, and can readily compare all the price/schedule/flexibility options available in the marketplace. Frequent business travelers can quickly figure out what the going rates for last minute changes are. This high quality information not only helps consumers, but helps airlines respond to market and competitive shifts. If you buy a cheap ticket you have an enforceable contractual commitment—United can’t ruin your entire vacation by involuntarily kicking you off the flight you’d purchased because they’d underestimated last-minute full-fare demand (1) Uber’s entire business model is based on destroying the price certainty that today’s taxi industry provides and replacing it with a system where no one can have any advance knowledge of what a trip will cost. This non-transparency/unplannabilty is critical to Uber’s approach, and is a much better explanation of the consumer hostility you noted, than the mere existence of higher peak fares. If you are thinking of going out Saturday night, there’s no way to pin down Uber fares in advance so you could compare alternatives or adjust your schedule. (2) If it starts to rain unexpectedly, or it turns out there’s a big concert or convention in town, you might have to pay an extra $100 to get home. And when it starts raining, there’s nothing to stop Uber from cancelling all its recent $20 bookings in order to capture those $100 surge fares. Unlike the full-fare airline passenger, regular Uber riders face the risk that today’s tip costs 4-7X what the last several trips cost. Airline pricing not only gives consumers the information they need to make the purchase decision that works best for them, but maximizes revenue and welfare by segmenting demand based on price and schedule elasticity. Uber pricing is designed to eliminate the advance information about fares that today’s taxi’s provide so that Uber can impose increases at the last minute that would be nearly impossible to avoid, and fails to achieve the same segmentation benefits since it imposes the same (standard-low surge-high surge) fares on everyone (regardless of elasticity) without advance warning.

    2. Variable airline pricing maximizes capacity utilization because of the specific nature of airline demand; Uber markets will not allow variable pricing to balance supply/demand or achieve other efficiency gains. A significant portion of long-distance airline demand is not locked into specific travel times, and (since there’s excellent advance info about big discounts) can be induced to shift travel away from more highly demanded flights. Pre-deregulation aviation had huge demand peaks (Fri/Sun night, summer) and poor off-peak capacity utilization; although it took decades for the full benefits to be realized, airline revenue management systems drove massive efficiency gains by shifting flexible, price-sensitive travelers to empty off-peak seats, significantly increasing total revenue by capturing demand all along the price elasticity curve, and maximizing the productivity of airplanes, pilots, airport gates and other longer-term costs. Both the dramatically different nature of urban transport demand and the limitations of its business model will prevent Uber from achieving any remotely similar efficiency improvements. Peak demand for intra-urban travel is either driven by weather or by fixed-in-time commitments. As you correctly noted, Urban Car Service an important part of every urban transport system and demand is dominated by working class people taking non-discretionary trips where public transport isn’t an option (urgent trips when it rains, remote locations at odd hours, shopping/family trips by non-car owners, etc.). Taxi use by wealthier people has even less schedule elasticity (airport trips, restaurants/clubs/concerts). Airline passengers can easily shift the trip to Orlando from August to October or shift a Friday night return to Saturday morning. But taxi riders will not shift their Saturday night dinner date to Tuesday morning, and will not refuse doctor’s appointments or overtime work opportunities that might hit an Uber price surge. It is hard to think of any segment of taxi demand that could readily be induced to change their basic time-of-day/day-of-week pattern by Surge Pricing, especially since there’s no way to determine what Uber fares will be at any point in the future. The “economists” defending Surge Pricing portray Uber as a heroic innovator that should be applauded for bringing variable pricing to the backward taxi industry, but are willfully indifferent to how taxi demand actually works, and to the fact that urban transport has been using peak/off-peak pricing for a century. Had they looked at the data, it would have been obvious how these demand fundamentals sharply limit efficiency gains. Rush hour is still rush hour, and adding capacity to serve it is staggeringly expensive because it is poorly utilized the other 20 hours of the day. The impact of these demand factors on capacity utilization are a major reason why no large-scale urban transport provider anywhere on the planet can operate on a private sector/laissez-faire profitable basis. The true cost of meeting peak demand (whether the morning rush on the Long Island Railroad or a rainy Saturday night in tourist season for taxicabs) is so high that a service operated on strict libertarian wetdream principles would be strictly limited to the 1% in the peak and would destroy most of the economic activity that relies on urban mobility throughout the rest of the week. The approaches that have best addressed these issues (combining non-price peak rationalization and some entry barriers to artificially boost the revenue base) may deeply offend those libertarians, but they are solidly grounded in verifiable, real-world evidence about demand and costs.
    3. Uber’s basic operating/business model eliminates the basic mechanism by which other transport companies optimize capacity/revenue. The central economic challenge of any transport company is tailoring capacity costs (supply) against a revenue potential (demand) that is inconsistent, uncertain and volatile. All large transport companies are organized so they can focus on this central challenge, with direct ownership and control of the capital assets that provide capacity, and integrated management functions tasked with optimizing capacity/revenue tradeoffs. Every airline tightly links its capacity (fleet/staffing) planning, scheduling, maintenance, pricing, and revenue management functions; profitability is a function of how the trade-offs between these functions are managed against the demand/competitive situation; none can be evaluated independently of the others. Any reasonably run Yellow Cab company will similarly centralize and coordinate the vehicle acquisition, maintenance, insurance, driver rostering, pricing and trip scheduling functions based on its detailed knowledge of the local market. Uber’s business model includes two radical departures from this longstanding approach: (a) It abdicates all control over the capital assets (vehicles) that are the heart of the business and (b) eliminates all management functions designed to proactively optimize the matching of capacity costs (supply) against local market revenue potential (demand). Many observers have noted that the Uber business model (combining the owner of a scheduling/pricing software program with tens of thousands of “independent driver/contractors”) is designed to make Uber’s balance sheet look better (it outsourcing both labor (hours spent driving), and capital costs and risks to its “contractors”), to eliminate consumer redress against Uber for accidents and service failures, and to maximize Uber’s leverage over those “contractors” (although no one has analyzed the potential to reduce driver costs below the industry’s already pathetically low levels). But none of Uber’s anti-empirical ‘economist” fanboys have explained how Uber can optimize the capacity/revenue tradeoff, if it doesn’t control any capacity, and all economic decisions about capital acquisition, and when that capacity will be offered to the market have been delegated to tens of thousands of outsiders. Nor have they explained how those tens of thousands of isolated individuals, currently working in very low wage jobs, can acquire vehicles, insurance and maintenance at lower cost than the fleet managers at Yellow Cab, and that its generic scheduling/software can better match capacity with demand than Yellow Cab dispatchers with local market knowledge. When Tim Lee claims that “Surge pricing gets more drivers on the road” and other “economists” claim that driver supply elasticities need price surge triggers to optimize the operation, it simply shows that they have been too lazy to understand how “capital” and “labor” happen to work in the Uber business model. As you mentioned fare surges are totally independent of whether capacity actually increases or how surge revenue is actually distributed; if drivers offered twice as much (very expensive) peak capacity next Saturday night, Uber fares would fall instead of rising to cover the higher cost. Hypothetically, a Uber-type app could help manage the (highly elastic) supply of short-term labor in isolation (imagine a similar app for day workers on construction sites) but there is absolutely no relationship between short-term fare surges and the (very inelastic) supply of the expensive capital dedicated to peak Saturday Night demand. American Airlines and Yellow Cab do not have these problems because they control all aspects of the capacity/revenue equation and strictly focus on maximizing the bottom line. Uber’s model is based on tens of thousands of people focused on optimizing tens of thousands of smaller equations, which cannot possibly produce an efficient overall result. Uber’s “we are just a passive middleman” model destroys the customer-operator link that is central to purchase decisions in every other transport business, and makes it impossible to achieve high levels of service reliability. Uber’s drivers are like Ebay sellers; they are free to offer their services to the market, but are under no obligation to do so. But unlike highly discretionary Ebay purchased, urban transport users depend on capacity being dependably available, day after day. There is little reason to believe that Uber’s scheduling app can dependably ensure the same service levels that traditional Yellow Cab approached have achieved, and much less reason to believe that it provide that service more efficiently than Yellow Cab, or at costs that will allow both Uber’s investors and “contractors” to earn strong returns.
    4. Discussions of using price mechanisms to optimize supply and demand are fundamentally absurd given Uber’s clear ambition to eventually achieve market dominance and market power over consumers and workers. Your post clearly recognizes that Uber is not a “disinterested Walrasian auctioneer”, that its business model features a platform-driven cartel approach and other features fundamentally incompatible with any discussion of welfare-maximizing resource allocation. But any discussion of “innovative” pricing or driver supply mechanisms needs to start from Uber’s clear intention to achieve market power over consumers and workers. Any other assumption is totally incompatible with its $50 billion financial valuation, the investment approach of its high-profile investors (3) and the history of the high-growth companies that Uber claims to be emulating. In an analysis of Uber’s valuation, NYU Finance Professor Aswath Damodaran said the only way he could justify a valuation as high as $6 billion was if Uber destroyed most existing providers and became the primary taxi provider in every city across the entire planet while maintaining the 20% margins it currently gets in SF and NY. (4) This is a situation where it is totally inappropriate to step back and “let the market decide how big Uber becomes”. Uber should be judged on the basis of its actual business model and its investors’ actual objectives. If Uber was unable to grow much beyond a niche position providing towncar services for Brooklyn hipsters and Silicon Valley Tech Workers, its investors would consider it one of the biggest failures since the dot-com era. Like many other high-growth start-ups, there is no evidence that it could earn sustainable profits if it fell well short of its longer-term growth ambitions. It is absolutely inconceivable that these investors aren’t explicitly pursuing total industry disruption and domination, and the ability to exploit artificial power over consumers and suppliers. Uber’s domination/market power objectives also illuminate a number of specific issues. It highlights the complete contradiction between claims that Surge Pricing is designed to calibrate competitive demand elasticities against competitive driver supply elasticities. The highly opaque/unplannable aspects of Uber Surge Pricing generates considerable consumer hostility, but with market dominance these features would it much easier to exploit artificial pricing power. Uber’s hyper-aggressive public behavior is directly tied to its strategic growth/dominance objectives; its journalist harassment and open flouting of court orders is central to this approach and cannot be dismissed as one-off bad behavior by poorly socialized tech startup guys.
    5. There is no evidence suggesting Uber will create any long-range economic benefits, no supporters have even attempted to outline a potential economic/financial justification for Uber and those supporters have completely blown off any criticisms based on financial, economic or industry data. There is absolutely no rational evidence supporting the claims that Uber will be significantly more efficient than existing suppliers, can grow rapidly by profitably providing more/cheaper service than existing providers, or would increase economic/consumer welfare above current levels. (5) All evidence suggests that a mature Uber will have materially higher vehicle ownership, maintenance, insurance and financing costs than existing operators. Uber will need higher driver costs for many years in order to achieve rapid growth, and these costs must compensate drivers for their capital costs and risks. Uber would appear to be at a huge competitive disadvantage because it makes no attempt to proactively balance capacity costs with demand and has no experience or knowledge of the diverse markets it is entering. Uber is not introducing any significant technological or process innovation aside from its scheduling/pricing software, and there is no evidence that its software creates any significant advantage since it is substantially less sophisticated than variable pricing systems in other industries (airlines, hotels), it will not materially reduce operating costs, and because Uber fails to link scheduling and pricing with capacity/capital planning, as other transport companies do. There is absolutely no evidence of Ebay/Facebook type “network effects” in taxicabs or any other urban transport industry. There is absolutely no evidence that Uber’s business plan could possibly make money within a strongly competitive Urban Car Service industry, and substantial evidence that its investors are explicitly pursuing a market dominance strategy. Other well-publicized startups may have raised competitive/profitability issues but could readily point to sources of competitive advantage; Uber appears to be the unique case of a major startup that raises many economic issues (platform cartelization, deliberate trashing on consumer protections regarding prices and liability insurance; dependence on exploiting market dominance, etc.) without any evidence of any major offsetting economic benefits. (6) The problem with Uber’s economic defenders is not that they have dismissed non-price peak rationing approaches on ideological grounds, or the lack of market-specific evidence in their claims on behalf of Surge Pricing or their failure to notice the cartel risks of Uber’s price coordination platform. The problem is the complete absence of any type of legitimate business/economic/financial/urban transport evidence or analysis in every published pro-Uber article. The problem is that no Uber supporter makes any attempt to explain how Uber could actually make sustainable profits, nor any attempt to define the specific sources of competitive advantage, nor any attempt to demonstrate a plausible link between the magnitude of those advantages and profits and Uber’s hyper-aggressive growth and market share objectives, nor any attempt to analyze whether a Uber-dominated industry would utilize society’s resources better than today’s. Uber’s “economist” fanboys don’t even acknowledge that these are legitimate questions that people might want answered. The problem is a complete, categorical rejection of the ways startup business investments and urban transport economics have traditionally been analyzed and a complete, categorical rejection of the use of empirical evidence in that analysis. You said you “don’t want to be too negative” because you seemed to be looking narrowly at a handful of Tim Lee type internet posts, while ignoring the larger context in which those posts were created. I think the empirical evidence fully supports a much stronger, blunter negative conclusion.
    6. Uber’s only “innovation” is its adaption of PR/propaganda techniques to “sell” its IPO and to block serious public/media scrutiny based on hard evidence.. Without dwelling too much on it here, you’ve assumed that Uber and its supporters are putting forward a business case supporting the returns its investors hope to achieve based on data and logic. In fact they have mounted a massive PR/propaganda campaign in support of their investors’ financial objectives. It is similar to PR/propaganda campaigns you are familiar with from national politics (in support of unsuccessful wars, to limit health care reform or fiscal policy options) or from finance and similarly politicized industries. Those campaigns destroy the possibility of serious fact-based discussion by using emotional and ideological signaling to structure the “debate” around a battle between tribal loyalists and enemies, so that the media fixates on unresolvable blue team/red team and he said/she said disputes, and ignores all the substantive issues that might erode public support for whatever the elite/corporate interests are trying to achieve. Uber represents the first major introduction of these techniques to a venture capital/investment banker pursuit of IPO wealth in an area completely outside any partisan political context. Every past startup had PR support, but Uber is the first case where it supporters make no effort to put forth efficiency/competitive/economic justifications (as no plausible case can be made), and the business and tech media have made no effort to analyze company/industry specific economics. Remember that success for Uber’s investors and senior managers has nothing to do with the creation of true economic value and a sustainably competitive company. All they have to do is to extract enough cash from the IPO to provide strong returns for the venture capitalists and staggering wealth for the senior managers.
    Uber’s strategy has been to create the appearance of a big brass band that will inevitably march from startup to IPO to eventual industry dominance, use emotive/tribal signaling to ensure the diehard loyalty of its most important (venture capital/libertarian/tech industry/business press) constituents, while isolating anyone raising objections (local regulators, Sarah Lacy) as the Luddite pawns of the diabolical taxi medallion cartel who are blocking Uber’s heroic innovations that will vastly reduce the cost of taxi service for struggling working class people throughout our major cities. The articles defending the economic advantages of Surge Pricing have absolutely nothing to do with the economics of Surge Pricing. Those writers are merely demonstrating their support for Uber’s Big Brass Band, and their tribal/ideological affinity for Uber’s allies. The willful indifference these articles display to the actual workings of the Urban Car System industry variable pricing systems is a feature, not a bug. I cannot rationally explain why so many commentators have cheered on Uber’s Big Brass Band so enthusiastically, given the actual process here where Uber’s deep pockets will allow them to bankrupt hundreds of more efficient existing companies, putting thousands of their workers out of work, in support of the efforts of Goldman, Blackrock, TPG, to extract most of the industry’s economic value into their own pockets. When you criticize their poor analysis, or note Uber’s “ethically challenged” management, or assert the right of municipalities to regulate transport services, or note how the welfare impacts of Surge pricing worsen with increased inequality, you are playing directly into their hands. They can show their allies that people fighting Uber are your tribal enemies—weak-kneed technocrats who could never stomach the dirty work and thus could never build a Real Business, who think government regulation is the answer to everything, and just want to transfer your wealth to poor people. You are responding to Uber’s 21st Century PR/propaganda campaign with 20th Century evidence and arguments. This Won’t Work.
    PS—I hope this is taken in the spirit of someone who has appreciated your blog for many years. I have absolutely no link with any Urban Car Service Industry company, investor or regulator
    (1) I recognize that the recent radical reductions in airline competition have created sustainable pricing power, but this is strictly due to consolidation, and has nothing to do with airline revenue management (variable pricing) systems, which are the issue here.
    (2) Although Uber’s approach precludes it, it is still possible to have both complete advance fare transparency and variable (peak/off-peak) fares, as illustrated by Sydney’s approach (h/t Pando’s Michael Carney)
    (3) list of major investors see:
    (5) many of Uber’s “economist” supporters cite Uber’s current ability to beat Yellow Cab current fares and service, ignoring the massive subsidies supporting Uber’s startup (and that every major high-growth startup incurred major operating losses throughout its startup
    (6) for a longer version see my TPM piece

  18. aaaa writes:

    You are incorrect about price stability. At the micro level, stable prices benefit price-insensitive consumers at the expense of price-sensitive ones. Think coupons and end-of-season sales.

  19. bruce writes:

    right rite of passage”.

  20. My brother proposed I’d this way web page. Your dog had been altogether proper. This particular article basically created the evening mutuo prima casa bnl. You can not look at the best way lots occasion I needed put in with this details! Many thanks!

  21. Chikago writes:

    Very well-written article, you certainly got a style. I’d like to offer you to join our team of college essay writers for one of the following online essay services, I’ve presented here: