This chapter offers an economic framework that regulators might apply in addressing the tensions outlined in Chapter 3 between the taxi and limousine industries and companies that provide technology-enabled innovative mobility services. This economic framework sets the stage for subsequent chapters that focus on specific public policy concerns associated with TNC and taxi competition, as well as other aspects of innovative mobility services, including labor, safety, insurance, and equity. This chapter concentrates primarily on the rise of transportation network companies (TNCs) because these companies have, in many jurisdictions, upended the business model of conventional taxi and limousine services and generated considerable controversy. Moreover, the technologies on which TNCs rely appear to be disrupting the regulatory model that has long applied to taxis and limousines, raising questions about whether existing regulations require revision, an issue taken up across multiple regulatory dimensions in subsequent chapters.
Technology-enabled mobility services bring with them an array of economic impacts, both certain and speculative, small and large. The most straightforward of these impacts is increased mobility, and with it greater access to economic opportunities for users. Because travel is rarely undertaken simply for its own sake, most of its economic benefits flow from bringing people, goods, and services together in productive and satisfying ways. Mobility improvements generate
economic benefits by giving people and firms access to jobs, customers, friends, services, and opportunities. These benefits flow to individuals and firms by reducing the time, cost, or risk and uncertainty of access. The benefits also flow to society more broadly by lowering travel’s external costs—such as delay imposed on others, crashes, and emissions—and increasing its external benefits—such as lower unemployment and increased energy security.
The changes engendered by technology-enabled mobility can simultaneously confer both benefits and costs, and thus entail consequential distributional effects. New services can disrupt existing business models and regulatory regimes. Powerful new ideas and technology often offer consumers new services and potentially offer investors an opportunity for handsome returns, but in so doing, they can threaten the financial viability of incumbent firms and the livelihoods of the people they employ. Technology-enabled mobility services, and TNCs in particular, appear at this point to be prime examples of this phenomenon. To date, these new services have yielded a handful of successful companies; provided millions of rides to consumers willing to pay for them; and provided flexible work for more than 150,000 vehicle owners, most of whom work part-time (Hall and Krueger 2015). At the same time, the evidence suggests that these services also have undermined segments of the existing ride-for-hire industry, particularly the taxi industry, as discussed in Chapter 3. The taxi industry is an important source of both income and wealth for many people, and in many cities it has important social service responsibilities (e.g., Kwong 2014). With the advent of TNCs, large global firms have entered local ride-for-hire markets that have long been dominated by smaller local firms; these new firms have also become an important source of income and wealth for many people, although currently they typically have fewer government-mandated social service responsibilities. As discussed in Chapter 3, these local firms frequently are heavily regulated, almost always (at least to date) more so than the newcomer TNCs. One of the most important near-term impacts of technology-enabled mobility, therefore, is the fate of the taxi industry and its drivers.
Further, the economic benefits that flow from increases in personal mobility can also engender social costs. The great promise of
such services as car- and bikesharing systems and TNCs is that they offer many of the flexible mobility benefits of motor vehicle ownership without the high up-front costs of owning, insuring, and maintaining a car. So the benefits of expanded mobility among those without their own car may also bring increased energy consumption, congestion delays, and emissions, and with them increased external costs borne by society.
How these various private and societal costs and benefits will net out and who will “win” and “lose” as technology-enabled mobility services expand are critical questions, but difficult to answer with much certainty during this time of transition. Nonetheless, this chapter represents an attempt to describe systematically what is known about the likely and potential economic effects of these new services. As of 2015, TNCs were competing both with one another and with existing ride-for-hire businesses (including taxis and carsharing) for the existing pool of ride-for-hire customers. If this trend persists, TNCs are certain to have large economic impacts on the ride-for-hire market, but given their relatively small market share in overall metropolitan person travel, they may have a relatively small economic impact overall. If the ride-for-hire market changes little or grows only modestly, the new mobility services likely will have only minor effects on the number of trips taken and miles traveled in metropolitan areas. But while net effects on travel may be minor, if TNCs weaken or bankrupt many taxi services, then those without credit cards and smartphone access may find themselves with fewer mobility options than before.
Evidence to date, however, suggests that the growth rate of the ride-for-hire market has been accelerating with the introduction of TNCs, perhaps substantially. If current trends continue, these services could have a much larger and more transformative effect on travel behavior and metropolitan transportation systems. For example, car- and bikesharing services, TNCs, and yet-to-emerge innovative shared mobility services could not only offer significant mobility benefits for those who lack access to or the ability to drive a private vehicle, but also encourage many more current drivers to own fewer cars and travel more by other modes. Should this occur, overall vehicle miles traveled (VMT) would likely fall, average vehicle occu-
pancy would likely climb, and some valuable land and capital currently spent on parking would be freed up for other purposes. These shared services could complement existing travel modes, particularly public transit, by functioning either as a first- or last-mile mode (that is, being used to travel to and from transit stops) or as a flexible backup when other modes—such as bus, train, carpool, or vanpool—could not adjust in response to unanticipated changes in travel or during times of limited or no service. If such travel markets expanded dramatically, innovative shared mobility services and traditional taxi services might well coexist comfortably, leaving intact the important social service roles played by taxis in many cities.
Uber and Lyft can be considered the most recent and most technologically evolved answer to the larger question of how to get a ride in someone else’s vehicle. The primary obstacles to getting such a ride, historically, have been problems of information, negotiation, and trust. In particular, those problems have involved asymmetric information and high transaction costs. Asymmetric information refers to situations in which not all knowledge relevant to a transaction is available to both parties (Akerlof 1970; Stiglitz 2001). Some information important to the buyer is known only to the seller, and vice versa. Transaction costs are simply what they sound like—the nonmonetary costs (in time, negotiation, energy, and so on) of arriving at an agreement (Coase 1937, 1960; Williamson 1979).
These problems are most evident when one considers the oldest approach to getting a ride from someone else: hitchhiking. Standing at a roadside and hoping to be picked up by a stranger involves high levels of uncertainty—about where and when to stand, about whether a vehicle will stop and after how long, about whether the driver who stops is going in the direction the passenger wishes to travel. Perhaps the greatest uncertainty relates to whether the passenger really wants to get into this strange vehicle (and for the driver, whether the passenger should be allowed in). Then there are questions about compensation: there are no standard fares for a hitched ride, so should the passenger contribute money for fuel, offer to drive, give the driver
cash, or some combination of these? In short, both driver and passenger must make important decisions while having little information. Efforts to gather more information may reduce uncertainty, but they cost time—the more the driver and passenger talk, the more time each of them loses by not being in motion.
The advent of dispatch taxi service solved a number of these problems. A centralized telephone number and a fleet of vehicles reduced the uncertainty associated with finding a ride: calling for a cab reduces much of the uncertainty associated with location, waiting, safety, and compensation. The taxi comes to the traveler, its destination is the traveler’s, and the driver works for an established company that can be penalized (legally, in reputation, or both) if something goes wrong. With rates often being regulated and certified-accurate taximeters in the vehicles, prices are transparent and require no negotiation (with the exception of tipping at the end of the ride). A similar situation holds for livery and limousine services, which a passenger also books ahead of time, usually for a set price.
Dispatch taxi and livery firms reduce information and negotiation costs by incurring large up-front costs. Dispatch service promises a vehicle that will come to the passenger within a reasonable amount of time or at a predetermined time. Delivering on such promises requires a fleet of easily identified vehicles (ideally outfitted with meters), a dispatcher to handle calls and direct drivers, and money spent to advertise. Most if not all of these costs are incurred before a single ride is given. These high barriers to entry into the taxi business give the dispatch taxi firms some qualities of a natural monopoly.
The high barriers to entry also make dispatch taxi service very different from the street-hail taxi business, whereby passengers can wave down a vehicle and get a ride. The street-hail business—if unregulated—entails very low barriers to entry. A person with a vehicle can drive along crowded streets and find riders. Street hailing also transfers some costs to passengers. Passengers cannot simply call for a cab but must do something more akin to hitchhiking by choosing a location where they are likely to find a vehicle, then hoping the vehicle slows and agrees to take them to their destination. Tension thus exists between low barriers to entry and high information costs, with much of the cost of arranging the ride devolving to the consumer.
As discussed in Chapter 3, the low barriers to entry in the street-hail business make it subject to oversupply, resulting in congestion and competition for passengers that may extend to physical violence. Many drivers can end up competing with one another for customers in the relatively small number of areas that have large numbers of street-hail passengers (dense downtowns, airports, and so on). Oversupply in the taxi business was one of the original motivations behind taxi entry regulations.
Cities enacted these regulations to solve real problems. However, standard microeconomic theory suggests that many of the attributes of the dispatch industry—price controls, quantity controls, location controls, and a market structure that tends toward monopoly—create inefficiency and poor service (Demsetz 1982; McAfee et al. 2004). This inefficiency may be revealed in a number of ways, but the most common and most evident is shortages.1 People willing to pay for rides cannot get them, either because the supply of vehicles is limited or because drivers cannot charge the price at which they believe a trip is worthwhile, even if passengers are willing to pay it.
For example, taxi drivers subject to fare regulations may be unwilling to make long trips to places where return fares are unlikely because picking up a passenger in another jurisdiction is not permitted or there simply are few available passengers. Such a trip might be to a neighboring jurisdiction where the driver would be forced to deadhead back (e.g., Boston to Cambridge) or simply to an outlying neighborhood where most people have vehicles and few need taxis. These trips entail higher per mile costs for the driver (more of their miles are likely to be unpaid), but the price earned per mile remains unchanged. In principle, if the driver could charge more for these trips to compensate for the opportunity cost of deadheading and if
1 Shortages result in lost time, which is a pure loss (when time is lost, no one gains it). Monopolies also often result in a transfer of income from consumers to producers. Strictly speaking, this is not a loss (it is a transfer), but consumers often find it objectionable because the lower supply of rides drives up the per ride price. Because taxis are regulated, however, the extent to which this increase in producer surplus occurs will vary depending on the jurisdiction. Regulated fares make it difficult for taxi drivers to earn excess profits. In medallion cities, taxi owners might earn such windfalls, but this again would depend on whether and by how much the cities regulated the lease agreements between owners and drivers.
the customer were willing to pay more, both would be better off. Rate regulation, however, prevents this mutually beneficial exchange from occurring. Similarly, taxi drivers may be unwilling to work in hazardous conditions such as snowstorms or other emergencies because the job becomes more difficult but the pay no better.2 Again, some people wishing to travel in bad weather might be willing to pay a premium to do so, yet would legally lack that option. Lastly, there may be times when demand for taxis spikes, but quantity controls mean that too few taxis are available to meet that demand. In February 2015, winter storms shut down Boston’s subway network for two full days, and full service was not restored to the entire mass transit system for a month. As a result, thousands of people who normally did not drive were forced out onto the streets (which were largely clear) to look for rides. Many wanted taxis, but Boston has only about 1,800 cabs and could not satisfy the spike in demand. In such instances, shortages result. The shortage might manifest either as taxi companies turning people away or as long and unreliable waits for service—for example, people calling taxi services and being told a vehicle is coming, but then enduring long delays.
Other forms of shared mobility entail variants of the same problems. Traditional carpooling, for example, shares many of the information and coordination problems of hitchhiking, although on a less ad hoc basis. Carsharing services resolve some of these problems because a vehicle can be booked in advance, but travelers do need to be able to drive. Moreover, carsharing companies may also have high fixed costs since they need many vehicles and also need to locate (and relocate) them in prime locations.
This tension between information costs and entry barriers helps explain why TNCs such as Uber and Lyft have seen rapid growth and success—they have resolved this tension in a way other services have not been able to emulate. Certainly some TNCs’ advantages arise simply from lower levels of regulation: Uber and Lyft are not subject to price, location, or quantity controls and often face less costly requirements for vehicle liability insurance, driver background checks, and vehicle inspections. But these companies also have created a business
2 To be clear, some cities do allow increased fares or surcharges during snow emergencies.
model that sheds important nonregulatory up-front costs typically confronted by firms that sell prearranged rides. TNCs require that drivers provide the vehicle to be used in service—traditionally one of the largest up-front costs of the business. Their digital platform also allows direct coordination between drivers and passengers, eliminating the need for and cost of centralized dispatch services. Moreover, the app injects transparency into the customer-driver transaction: customers opening the apps on their phones see a map that prominently displays the location of nearby vehicles and provides an estimated wait time for pickup. Most apps also provide a fare estimate. Once the ride has been booked, the customer receives regular text message updates on the driver’s location and can track the vehicle’s progress on the map. The app, in other words, reduces both information costs between the firm’s passengers and drivers and fixed costs for the firm itself.
TNCs also can reduce shortages by raising prices in response to high demand. Uber’s “surge pricing” and Lyft’s “prime time” exemplify this approach: when demand is high, passengers are charged some multiple of normal fare rates—for instance, a surge rate of 2 means rides normally costing $5 will cost $10 instead. The surge is designed to lure drivers who are not driving (or who are working in other neighborhoods) to places where demand is high. While taxis can sometimes increase prices—for example, a $15 surcharge is added to each trip during snow emergencies in Washington, D.C.3—TNCs have much more flexibility with their prices in response to changing levels of demand for service.
TNCs can employ surge pricing effectively for four reasons. First and most obviously, they are not subject to formal price controls, so they can legally raise or lower their fares in response to changing conditions in nearly real time. Second, the dispersed and flexible nature of TNC employment means that firms have a large pool of capital and labor (cars and people) to call on at any given time. Third, there is no legal limit on the number of TNC vehicles that can be on the road at any given time. And fourth, drivers are not restricted to picking up passengers in particular municipalities. Thus it is not
just the absence of price controls that allows surge pricing to work; the absence of quantity and location controls and the fact that TNC drivers do not work assigned shifts also are crucial. After all, high prices cannot prevent shortages if regulations prevent new supply from reaching those places where demand is increasing. More specifically, raising prices without increasing supply can reduce shortages only by encouraging some people not to take vehicle trips, whereas raising prices and allowing those increases to attract new supply can reduce shortages by providing more trips. For example, Uber used surge4 pricing when Boston’s transit service suffered delays and cancellations. The company could call on more than 10,000 drivers in the city and its surrounding areas, and these vehicles could arrive from outside the city to give rides. Providing these rides was worthwhile for drivers because Uber charged up to three times its normal fares during the transit shutdown (Szaniszlo 2015).
Surge pricing has been controversial. Critics have argued that price spikes disadvantage lower-income people or that they violate laws against price gouging. New York state officials criticized Uber for using surge pricing during Hurricane Sandy, for instance, and later signed an agreement with the company that would restrict the use of such pricing during states of emergency (New York State Attorney General 2014). TNCs have agreed to limit the use of surge pricing in other cities as well.
From an economic perspective, however, surge pricing is best practice. Economists generally frown on price controls of any sort, and many view the concept of price gouging (and laws prohibiting it) with suspicion. Although it may appear unfair to charge passengers more during emergencies, it may also be unfair to expect drivers to provide service during emergencies without extra compensation. Moreover, customers knowingly and freely pay surge rates: the higher rate is displayed as soon as the phone app is opened, and passengers must explicitly consent to the increased rate by typing it into their mobile device. Thus while riders almost certainly do not like high prices, and little evidence suggests they are being deceived into paying them, the
4 Drivers throughout the taxi industry can also choose when and how much they work, but since fares are typically regulated, taxi prices cannot surge pricing when demand increases.
variation in prices adds a degree of unpredictability to the service, as does the unpredictability of response time in taxi service.
One rejoinder to this logic is that during emergencies, consumers pay more only because they have little choice, and that surge pricing is therefore a form of coercion. The validity of this argument hinges on the most likely counterfactual—what would happen in the absence of a surge. Certainly if the most probable alternative to a surge price is the same number of rides at a lower price, then companies may be coercing riders. But this scenario is unlikely: the more likely counterfactual to a price surge that delivers 100 rides for double the normal rate is not 100 rides at a lower rate, but fewer rides. And if this is the case, then the surge gives people choices; in its absence, some people who could choose between paying and waiting will be left with no option but to wait.
Nor is there much reason to believe that low-income people would benefit from lower prices if those low prices were also accompanied by shortages in service. A price control only regulates the price of the ride; it says nothing about who gets the ride. Thus while price controls enhance equality, they may do so at the expense of quality: during times of high demand, instead of making rides more accessible to everyone, price controls simply may make rides equally inaccessible by subjecting everyone to a shortage. From an economic perspective, concerns about the inability of low-income people to afford transportation should ideally be addressed by programs that explicitly target low-income travelers, not by laws that lower prices across the board.
TNCs, in sum, can offer more reliable rides (as a result of more flexible prices), and often lower fares (a result of having more vehicles), relative to traditional for-hire mobility services. In some markets, TNCs also may draw drivers away from the taxi industry. And once both customers and drivers have established loyalty to TNCs, that loyalty is globally transferable. Uber is a global brand, and Lyft operates across the United States; the presence of such widespread branding reduces the transaction costs of arriving in a new city. Travelers and movers need not learn which taxi companies are or are not reliable, nor need they worry (in the case of international travel) about having the correct currency to get a ride or (for business travelers) remembering to get a receipt. They can simply switch on
their phone and arrange a ride, and have a digital receipt emailed to them. Similarly, a person earning money by driving for a TNC has portable skills and can drive professionally (with the aid of navigation software) wherever that TNC platform is supported.
TNCs as Monopolists
Regulators might worry about whether TNCs could transition from injecting new competition into closed markets to becoming monopolists themselves. Some aspects of the current TNC market suggest this possibility. There are currently two dominant TNCs, and one (Uber) is much larger than the other (Lyft). Because TNCs are network companies, they are more efficient if they penetrate more of the market. It is therefore a natural concern that one or two large companies might come to dominate the ride-for-hire industry in many markets. What would happen, for example, if Uber and Lyft began colluding and raising prices or if Uber were to buy Lyft and become the lone dominant TNC?
One can never answer such questions with certainty, of course. But one can consider a few factors. First, active collusion—creating a cartel to fix prices—is illegal. Some industries that have one or two large players are able to engage in what is called tacit collusion: keeping prices high without explicitly agreeing to do so (Green and Porter 1984; Ivaldi et al. 2003). The hallmark of these industries, however, is that they compete on quantity when it is extremely difficult to add new supply. One classic example is makers of large jet airplanes (Krugman and Wells 2013). Commercial airplane manufacturing is dominated by Boeing and Airbus, and it is easy for each company to know how many airplanes the other will produce in a given year. As a result, both firms can set prices high because they know the other firm has little incentive to undersell them (lowering prices is advantageous only when it allows a firm to win new business, but winning new business is pointless if the firm cannot produce more airplanes). When this condition of sticky supply does not hold—when supply can be quickly added—tacit collusion becomes impossible, and even firms in oligopolistic markets will compete
on price. Coke and Pepsi, for example, dominate the American soft drink market, but ramping up soft drink production is fairly easy, so rather than collude, the two companies ferociously compete.
TNCs resemble Coke and Pepsi more than Boeing and Airbus. Adding supply is very easy with TNCs. New drivers are easily recruited, and drivers can work for multiple firms. Lyft and Uber currently engage in price battles for precisely this reason. The ease of recruiting drivers points to a larger idea: the same factor that has allowed TNCs to undermine taxi firms—low barriers to entry—should, in theory, also prevent TNCs from becoming monopolies themselves. If one TNC completely took over a market and ramped up its prices, a competing platform could enter the market at relatively low cost and, through lower prices, take some of that TNC’s business.
Nonetheless, freedom from market dominance by one TNC is by no means guaranteed. One could imagine commercial practices that would facilitate dominance, such as ubiquitous service coverage that consistently gave one TNC response time advantages over competitors, a dominant TNC that required drivers to work exclusively for that company, or the preloading of smartphones with the app of a single TNC. For example, as of 2015, Uber is partnered with Google to promote Uber as a travel option in the Google Maps app. Barriers to entry can also be created by governments. If a dominant TNC successfully lobbied for regulations to protect its market (for instance, if it supported legislation saying that no new TNCs could be licensed unless they had a minimum number of drivers), that TNC might well become a monopoly. Of course, if new entrants simply entered markets without seeking permission from regulators, as TNCs currently do, even government barriers to entry might not be effective.
Lessons of Deregulation
The experience with taxi deregulation 30-plus years ago (reviewed in Chapter 3) is informative for current consideration of the role and possible effects of rapidly expanding TNCs. Problems resulting from deregulation were focused on taxi stand and street-hail markets that became oversupplied by newly licensed independent
drivers. TNCs, by contrast, serve the dispatch segment of the market, where they have the same incentive as traditional taxi dispatch operations to balance supply and demand of trips. Predictions that de facto deregulation of entry for TNCs would have the same effect as the earlier move to open up entry for taxis need to address this difference.
At the same time, the experience with deregulation points to the important interdependence of market segments. Taxi drivers tend to serve multiple markets: taxi stands downtown and at the airport, dispatch trips throughout the city. In so doing, they gain efficiencies from, for example, being able to respond to a dispatch call in an outlying residential neighborhood after dropping off a passenger whose trip started at an airport taxi stand. Without this mix of trips, drivers spend precious time and fuel deadheading back to more lucrative areas.
TNCs in the current regulatory regime have an advantage over taxis insofar as they have better opportunities to avoid deadheading when crossing jurisdictional boundaries within a metropolitan area. As discussed in Chapter 3, in many regions, a taxi dropping off a passenger outside of that taxi’s home jurisdiction cannot then pick up a passenger in that same jurisdiction. For TNCs, by comparison, a metropolitan area is a unified market, with no restrictions on dropping off and picking up passengers. One way for policy makers at the regional level to equalize competition between TNCs and taxis would be to allow taxis crossing jurisdictional boundaries the same ability to pick up passengers. But geographic restrictions on taxis evolved for a reason.
If TNC and taxi drivers had information on the demand for rides at a given time in a given area but limited or no information on the supply of vehicles at that time in that area (which aptly describes the current situation for street-hail cabs), removing geographic pickup restrictions might result in both TNCs and taxis (with TNC-like apps) clustering in areas with high demand while awaiting requests for rides. Less densely developed areas thus would be left with little service that could respond within a short period of time. However, if the apps provided the drivers with real-time information on places where the ratio of drivers to customers was high and where it
was low, then the incentive for drivers to cluster and saturate high-demand locations irrespective of vehicle supply (which is common with street-hail cabs) would be considerably diminished.
To address problems with taxis clustering in the absence of real-time information on both the demand for service and the supply of vehicles, taxi-regulating jurisdictions have adopted regulations that limit when and where cabs may pick up customers. New York City, Los Angeles, and Las Vegas, for example, have created zones and restrict certain cabs to picking up in particular zones. Anaheim–Orange County and Seattle–King County have achieved similar effects by different means. Whether such geographic time and place pickup restrictions might need to be extended to TNCs, be limited to street-hail cabs, or be dropped altogether for taxis remains to be seen. This example, however, illustrates the complexities and tradeoffs faced by regulators in attempting to level the competitive playing field between TNCs and taxis.
Securing a ride in someone else’s vehicle typically involves balancing information and negotiation costs against barriers to entry for firms providing rides, which before smartphone app–enabled TNCs involved the high cost of entering taxi dispatch markets. Of the many technology-enabled mobility services that have emerged to date, TNCs have resolved this tension most successfully because their apps not only provide greater transparency for drivers and passengers, but also reduce the up-front costs for the firms themselves. In the short run, this innovation, combined with a typically lighter regulatory burden, allows TNCs to offer, at least in some instances, service and pricing superior to those of traditional taxis while incurring fewer of the regulation-associated costs that taxis must bear, potentially eliminating social costs in the process, as described in subsequent chapters. Moreover, the lighter regulation of TNCs facilitates the development of national and international companies with economies of scale and brand identities that provide a competitive edge over locally regulated for-hire firms and a motivation to protect the market image of the multinational brand.
Probably the most obvious and least controversial result of the rise of TNCs is that many people now avail themselves of their services, suggesting that these firms offer real improvements in mobility. The more difficult near-term question is what the rise of TNCs means for the taxi industry economically and for the current set of regulations to which it is subject. TNCs offer strong efficiency benefits, but cities have come to rely on taxi operators to advance a number of social service and equity goals as well—goals whose attainment is for the most part cross-subsidized directly by taxi customers or indirectly by cab or medallion owners. Taxis often ferry riders with disabilities, and they also are an important means of mobility for low-income people unable to purchase or insure private vehicles. If TNCs harm or (at the extreme) eliminate the taxi industry in some markets, people with disabilities and low incomes could find themselves with fewer means of getting around in the absence of some public policy intervention. These issues are considered more fully in Chapter 8.
During this time of uncertainty and transition in mobility services, the ultimate impact of TNCs on the taxi industry is unknown. Perhaps the two services will coexist to the extent that regulators equalize their regulatory treatment to some degree. It is not clear, however, whether or how to make taxis more competitive with TNCs. The taxi industry is responding to the competition from TNCs by beginning to adopt technologies that match passengers with drivers. On the other hand, regulating TNCs more heavily could undermine many of their price and responsiveness advantages (if, for instance, controls on surge pricing were applied to the currently variable TNC fares). Reducing some or many of the pricing and supply regulations imposed on taxis might be a better approach to leveling the taxi–TNC playing field, and some regulating authorities have recently moved in this direction. Lessons from taxi deregulation suggest that allowing TNCs to compete in the street-hail market would lead to oversupply and excess competition; thus, the street-hail business does not appear to be a promising area for lighter regulation. Opening up competition in the dispatch market, by contrast, appears to hold promise for consumers. However, a less regulated taxi industry
competing with TNCs would be very different from what currently exists in those jurisdictions where medallion holders (for example) are guaranteed a good deal of heavily regulated market power in exchange for agreeing to provide equity-based social services. If TNCs continue to erode the demand for taxis in for-hire transportation, regulating authorities may have a more difficult time attaching conditions to taxi licenses for purposes of geographic coverage and service for passengers with disabilities.
Akerlof, G. 1970. The Market for “Lemons”: Quality Uncertainty and the Market Mechanism. Quarterly Journal of Economics, Vol. 84, pp. 353–374.
Coase, R. 1937. The Nature of the Firm. Economica, Vol. 4, No. 16, pp. 386–405.
Coase, R. 1960. The Problem of Social Cost. Journal of Law and Economics, Vol. 3, pp. 1–44.
Demsetz, H. 1982. Barriers to Entry. American Economic Review, Vol. 72, No. 1, pp. 47–57.
Green, E., and R. Porter. 1984. Noncooperative Collusion Under Imperfect Price Information. Econometrica, Vol. 52, No. 1, pp. 87–91.
Hall, J., and A. Krueger. 2015. An Analysis of the Labor Market for Uber’s Driver-Partners in the United States. Princeton University Industrial Relations Section Working Paper 587. Jan. http://dataspace.princeton.edu/jspui/bitstream/88435/dsp010z708z67d/5/587.pdf.
Ivaldi, M., B. Jullien, P. Rey, P. Seabright, and J. Tirole. 2003. The Economics of Tacit Collusion. http://academico.direito-rio.fgv.br/ccmw/images/5/5e/Tirole_-_The_economics_of_tacit_collusion.pdf.
Krugman, P., and R. Wells. 2013. Microeconomics. Worth Publishing, New York.
Kwong, J. 2014. Report Says SF Taxis Suffering Greatly. San Francisco Examiner, Sept. 16. http://archives.sfexaminer.com/sanfrancisco/report-says-sftaxis-suffering-greatly/Content?oid=2899618.
McAfee, R. P., H. M. Mialon, and M. A. Williams. 2004. What Is a Barrier to Entry? The American Economic Review—Papers and Proceedings, Vol. 94, No. 2, pp. 461–465.
New York State Attorney General. 2014. New York A.G. Schneiderman Announces Agreement with Uber to Cap Pricing During Emergencies and Natural Disasters. Press Release, July 8. http://ag.ny.gov/press-release/ag-schneiderman-announces-agreement-uber-cap-pricing-duringemergencies-and-natural.
Stiglitz, J. 2001. Information and the Change in the Paradigm in Economics. Nobel Prize Lecture, Dec. 8.
Szaniszlo, M. 2015. Uber, Taxis Fare Well Thanks to Snowfall, MBTA Delays. Boston Herald, Feb. 19. http://www.bostonherald.com/business/business_markets/2015/02/uber_taxis_fare_well_thanks_to_snowfall_mbta_delays.
Williamson, O. 1979. Transaction-Cost Economics: The Governance of Contractual Relations. Journal of Law and Economics, Vol. 22, No. 2, pp. 233–261.