Featured Post

On Public Transit

On a couple of posts in the last two days I’ve mentioned Hernando de Soto’s important book The Other Path. I have two copies of this book, one with the original subtitle of “The Invisible Revolution in the Third World,” and the other with the later subtitle, “The Economic Answer to Terrorism.” I suspect the latter was the publisher’s attempt to profit off 9/11, but in fact de Soto is from Peru, home to the Shining Path, a violent communist insurgency, so there is a real homegrown element to that subtitle change. The book is now a quarter-century old, and deserves to be read by anyone interested in the interplay of politics and markets (as is his later book, The Mystery of Capital).

My specific reference was to de Soto’s chapter on public transportation in Lima, Peru. All I want to do in this post is quote extensively from that chapter, to give a sense of the point I’ve been making. Remember, this was written in 1989, so things may not be the same, but this is not so much about what Lima is, as about one example of how private, albeit illegal, transportation services can thrive and out-compete a state-provided service. It doesn’t mean I’m arguing for full privatization of bus, subway and commuter rail service in the U.S. Importantly, it seems to be the particularly poor quality of the state-provided service that makes it vulnerable to competition from the illegal transportation sector. I’m only saying that the argument that the private sector cannot do public transit may not be so solid as some instinctively think.

…91% of the 16,228 vehicles used for mass transit were being operated informally…Formal transport accounts for the remaining 9 percent of mass transit. Of this the state-owned National Urban Transport Corporation of Peru (ENATRU) provides 4 percent, and former cooperatives, worker-owned corporations, and the Lima Metropolitan Transport Corporation (TLMEPS) account for the remaining 5 percent.

Two brief notes: First, When de Soto uses the term “informal,” he means operating in black or gray markets; activities that are technically illegal, but to which government sometimes (but not always, and not consistently or reliably) turns a blind eye. Second, the “former cooperatives” and “worker-owned corporations” are previously informal transit services that managed over time to gain official authorization. There is a cost to balance against the gains of official authorization, as the transit operators have to accept controlled fares, and limits on routes serviced.

Informals provide a very important social function by meeting the transport needs of dwellers in informal settlements…Informal transport operators concentrate their services in popular areas, while the state corporation and other formal companies mostly serve traditional neighborhoods.

These informal settlements are also illegal. Like the transit services, they are sometimes ignored, occasionally treated with harsh reprisals, and sometimes–through years of diligent effort–made legal and brought into the formal sector. But being illegal, the state-run transport services tend not to serve them, lest it lend them an aura of legitimacy.

[I]nformal transport operators invade routes. Routes are not physical assets like land or the street, however, but intangible assets defined by the population’s movements and travel needs. A route is a unit consisting of different journeys between a first and last point…

The process is an economic calculation in which, primarily, informal transport operators try to evaluate different possible routes in order to decide which one to take over. To do this they must identify, at minimum, where there is potential demand, for which journeys there is inadequate service, and which new neighborhoods or settlements lack transportation…

Invasion is used not only to discover and appropriate original routes. It is also used by established informal transit operators to extend or modify the routes they already use. Such invasions are carried out not by individuals, but by groups, since an informal organization or committee must approve the invasion at a general meeting.

The actual routes, it seems, are “owned” by individuals who are part of that committee. De Soto emphasizes that these individuals can sell their rights to a route. In some organizations they can sell it to anyone, but if they sell it to someone outside the organization the purchaser must join, usually paying a sort of admission fee. Other organizations require the seller to offer it first to others already within the organization.

As the invaded route and the rights to it gradually increase in value, the informal operators begin to have incentive to organize, negotiate and deal with legal institutions. This requires them to set up organizations at basically two levels: first, committees of informal transport operators operating the same route; and second, different committees grouped together into unions, and, later, federations.

…There is an initial stage [to an invasion] in which, after invading a route independently, each informal transport operator runs a separate service, manages its hours and timetables and decides what fares to charge. The operator cannot remain independent indefinitely, however, and will have to organize with other transport operators covering the same route, for a route well chosen for its length and number of passengers increases in value and tempts an increasing number of invaders.

Organizing has a number of advantages. Operating a route in an orderly manner helps to reduce operating costs and ensures a regular service, which keeps passengers happy. Organizing also brings together a sufficient number of transport operators to negotiate with the authorities and preserve the route they have established. Third, it keeps out new invaders once there are sufficient vehicles to meet demand.

It’s worth noting that the negotiation with authorities takes the form of not only of legitimate negotiation, asking for official recognition, but of bribery and payoffs, not necessarily out of the operators’ fondness for corruption but because it’s the most efficient way to secure at least an unofficial toleration.

By the 1950s, traditional transport [a combination of state-owned and private firms that were officially authorized and mostly predated the development of the state-owned transit firms–JH] had ceases to be profitable. In 1959, when the El Sol Transport Company closed its doors, an unexpected succession of bankruptcies left Lima almost without formal transport services. Byu 1960, 32 of the 42 private bus companies then in existence had ceased to be operated by their owners.

At least three factors underlay this situation. Declining profits, a result of strict fare controls, forced formal companies to postpone dealing with the depreciation of their vehicles, with the result that their fleets became ramshackle and obsolete. Second, informal transport, since it was not subject to state control, was better able to adapt to the growing demand created by Lima’s urban growth. Last, the increasing reserves which the formal companies were required to maintain to pay their workers’ social benefits ultimately condemned them to insolvency.

Informal operators stepped into the gap. Importantly, their profit-motivation did not induce them to provide poor service in ill-maintained machinery at prices people could not afford. Instead they constantly sought out routes that were not being sufficiently served, charged fares that people dependent on public transit could afford (even though it was more than the state-approved privates were allowed to offer), and continually upgraded their equipment, primarily meaning adding larger and larger buses to increase capacity. The general trend was from sedans to minibuses to Dodge D-300s, then, when the government ended its policy of keeping gasoline cheap, the D-500, a longer variant of the D-300, then the yet-again-larger D-800 and the Volvo Bb-57. Even then they could not keep up with demand, so eventually a Swedish firm that manufactured buses in Brazil offered them Marco Polo buses, which in contrast to the old Dodges, which were primarily repurposed or just a very old-style frame with a domestically built body, were “state of the art vehicles designed specifically for the service, with diesel turbo engines [and] an average capacity of 80 passengers…” (de Soto, p. 121).

In contrast, the legal operators were in a regulatory-induced race to the bottom.

The fact is that, over the years, the legal system has not allowed the tremendous business energies and entrepreneurial talents of formal and informals alike to be tapped. On the contrary, it seems to be designed to deal politically with only a very small number of interest groups, not with a widespread entrepreneurial class. Legal recognition of transport operators is not considered a right, but an agreement to be negotiated in exchange for the acceptance of political interference and the imposition of numerous burdens, including the control of fares.

This situation has completely altered the behavior of the transport operators and has shifted the competition in transit services from prices and quality to safety. Instead of competing by lowering fares or improving quality, the transport operators have been forced, over the years, to compete by reducing safety and paying the bribes requested by the authorities. One of the many privileges which transport operators have received in exchange for accepting government controls is, in fact, the relaxation of safety requirements by the authorities.

Again, this is not a an argument for privatizing urban transit services. It’s only a demonstration that a private transit service can be profitable and provide good customer service even in a competitive market.

[Note: In a comment to KatherineMW, I said this was going up tomorrow. Tod has asked me to go ahead and put it up today.]

Please do be so kind as to share this post.

53 thoughts on “On Public Transit

  1. 1. There are legitimate complaints about traffic and/or public transportation and then there are the complaints that people like to make because everyone likes complaining. How do you separate the the two? As far as I can tell almost no one likes their public transportation system and wishes it was better.

    2. Private mass transit is probably easier to do with buses and cars than trains. I don’t think anyone ever liked transporting people via train, we did it because it was the fastest mode for most of the 19th century. No one seemingly likes transporting people via air either but it is still done because it is fast.

    3. I’m guessing that the service in Lima, Peru was much much worse than any transit system in the U.S. and it also seemed to depend on density. It is interesting though that the private bus systems in the U.S. are more known for long-distance hauls like Greyhound and Peter Pan.


    • 1) I’ve liked most of the public transportation systems that I’ve used (as a resident, Victoria, Vancouver, and Ottawa; more briefly, Toronto, New York, and Paris). DC’s subway routes feel a little weirdly designed.

      There’s always ways I can think of for them to be improved, but I wouldn’t say any of them are bad.

      Recently Victoria has switched from full-size buses to mid-size vans/minibuses on some of the less-used neighbourhood routes, which I think is smart and economical.


      • All due respect, but you’ve been in really, really nice places for public transit.
        You’d probably feel way differently if you were in Harrisburg, PA and relying on public transit for, well, anything.


      • True, Kimmi. I’ve been living in places with governments who believe that 1) governments can do useful things; 2) public transit is one of those things; and 3) a service like public transit is worth funding and managing decently.

        There’s an amazing amount of overlap between politicians who think government is useless and politicians who deliver public services badly (or don’t deliver them at all).


  2. Lima did not have a subway line until the mid-aughts. Like other cities in the developing world, it focused on bus based transit because it’s cheaper to build and maintain than rail based transit.

    Private rail transit is an interesting proposition. It’s certainly possible to have private rail transit. It existed in the past and continues to exist in some countries, notably Japan, today. The real issue is that you can’t have a lot of competition in the same way that you can have with buses, taxis, or even airplanes because of the infrastructure. You can only build so many rail lines to a particular location before you have a mess.


    • The Pacific Electric Red Cars here in LA were interesting. They were privately constructed, but collapsed due to several factors, notably the inability to extend the rights of way for lack of eminent domain, and the love affair with cars.
      There are many myths around them- that they were beloved (they weren’t) and that the oil and tire companies killed them off (not true).


      • The Pacific Electric was built in the early 20th century so like many other tram and interurban systems, it was built and run by a private corporation. Incidentally, the main reason Samuel Huntington built the Pacific Electric was to further his real estate development business. Like the private Japanese rail companies of today, the main source of profits was never from getting people around but from real estate. A lot of other tram systems were run as subsidiaries of the local electric company. The local electric companies also built and ran amusement parks as a way to encourage people to ride their trams on weekends.

        During the mid-1940s, there was a plan to municipalize the Pacific Electric and use it as the basis of a mass transit system for LA county, This plan was scraped in favor of focusing on the nascent car culture. When LA county began to rebuild its rail system during the 1980s, I’m sure many of the officials involved were really angry that LA didn’t municipalize the Pacific Electric when it had a chance to. It would have saved LA a lot of money, time, and traffic jams.


  3. Damn, I saw your references and noted them, so I’m glad you wrote this post since I didn’t wanna have to read the whole damn book to get what you were talking about.

    I’m off to read the post!


    • Well, I read it and I agree. One thing that hit me during the reading is that if gummint initiates a social service because private folks won’t or haven’t, we get a different set of justifications than if gummint tries to control services that would happen in any event. I think lots of liberals tend to think that these and similar types of services wouldn’t exist without government impetus behind them, but I think commonsense as well as the quotations you supply show that those conclusions aren’t correct.

      Sensible regulation is one thing. Gummint control is another.

      Good post.


  4. Is it necessarily going to be the case that a governmental agency will be as dramatically ossified in its decision-making, and as captured by corruption, as we read described in Lima? It’s easy to imagine that private enterprise can usually be more supple in its thought and more responsive to user demand. If we’re talking about bus routes, the cost to change, augment, or add them is not prohibitive and a sufficiently well-run governmental agency could (in theory) respond to the city’s growth.

    Rail transit is a different animal, because you have to build stationary rail lines. Whether public or private, they don’t get moved easily once built. Especially if they’re underground. At that point, they become points of developmental efflorescence: ready access to a Tube station has a measurable and often dramatic impact on the price of a flat in London, for instance, attracting wealthier residents and higher-end retail businesses.


      • Yeah, the Chinatown buses are interesting. You can read more here:


        “Because of their low fares, the services became popular among non-Chinese customers as well. Between 1997 and 2007, Chinatown buses took 60% of Greyhound Lines’ market share in the northeast United States.”

        The “Shutdowns” header also lists a couple others that have run into trouble, usually for safety violations but they’ve also run into opposition by doing their pickups / drop offs curbside rather than at stations.


      • “One business got shut down because its buses were unsafe, but a bunch of others just like it had already popped up and are operating just fine, therefore private buses are technically illegal” is an unconvincing argument.


      • Mike,
        I dunno, but after the disasterous “every single private line went out of business”… I could see the county outlawing private buses (under the theory that “this was a mistake and the taxpayer had to pay for it”)


      • Eh, that’s not surprising, but still doesn’t make private buses technically illegal when they’re thriving.

        I’ve taken Megabus to Houston. It’s kinda weird just being dropped off at a mall. Though I remember taking Greyhound from Kentucky to Tennessee, or from Dallas to Austin, long ago, and the small town stops were just at gas stations and country stores.


      • Mike,

        I think that depends on whether we’re talking about inter-city buses or buses for in-city transit.

        If you’re talking about inter-city buses that cross state lines, that’s obviously in the federal government’s domain, and the answer is clearly that it’s legal (Greyhound, charter coach services, Chinatown bus).

        If you’re talking about inter-city buses that run just within a state, like the Green Tortoise between SF and LA, then presumably it’s a state-level issue, but since states don’t normally provide that kind of service themselves, I’d be surprised if any outlawed private firms doing it.

        The real question, l think, would be in urban transit. I can imagine cities banning it, but I don’t know if any/many actually have. The problematic aspect for a private firm would probably be how it manages stops and stays compliant. If it blocks traffic by stopping at particular street corners, it may create a backlash and run afoul of local traffic enforcement. If it tries using the public bus stops, it creates the kind of backlash Google bus runs into, and is probably illegal to do anyway. So a municipality might be able to effectively outlaw a private urban transit bus without officially outlawing it. But if they used minibuses and picked up/dropped off people in parking lots (and let’s say they worked out deals with the parking lot owners)? My off-the-cuff guess is that then they’d fall under the municipality’s taxicab/limo regulations, which vary from place to place.

        I’m sure they’d need some kind of permits. And in any city where the cabs and limos are cartelized I’m sure they’d push back politically.


      • Right. I imagine that in most places, a full line of local buses run by a private company would simply have too many hurdles to jump over. City-run bus systems are able to get around things like taking up sidewalk space and even private property by utilizing existing laws related to public right of way that might be more difficult, or at least significantly more expensive, for a private company.

        Of course, city transit services can only survive in most places because they are heavily subsidized by city, state, and federal dollars. In order to make transit profitable, you’d probably have to charge enough to eliminate most current transit customers.


      • Of course, city transit services can only survive in most places because they are heavily subsidized by city, state, and federal dollars. In order to make transit profitable, you’d probably have to charge enough to eliminate most current transit customers.

        It’s like you didn’t even read the OP, or read it, but don’t think you actually have to explain why it’s different here.


      • You ignored “most places.” Dude, you read people who disagree with you so uncharitably that I’m not sure it’s possible to have a conversation with you until three cycles in, when you settle into actually discussing what was said. Just calm down, man, and assume your interlocutors are not idiots or demons.

        Lima is awesome. South America has, in the last decade or so, produced some of the best transit stories in the world, in part because things are really cheap down there, so building and operating costs are significantly less than they are in the U.S. or most of Europe, say. Curitiba may have the best bus system in the world, and that bus system has radically altered the city itself (it’s publicly funded for the most part, I believe), and most people have never even heard of Curitiba unless they’re into transit issues. I doubt there’s a city in this country that could build a bus system as good as Curitiba’s, because building costs alone would likely be well into the billions with all the infrastructure building that would have to be built out of whole cloth. Eugene may be the closest we have in the U.S., and its system is so great because it can be so limited and still be highly functional for people who live and work in Eugene.

        In the U.S., however, at least in the smaller cities (like, say, Austin, or Minneapolis, or Omaha, or Columbus, and so on), the operating costs — whether a system is publicly or privately run — are high for buses, and much higher for trains, particularly if they have to cross bridges. And that’s just the operating costs. The system-building costs, in the U.S., are even higher, and again, a private company will have to either make deals with a municipality or spend a lot of time and money building its infrastructure. Not going to be worth it, for local service.


      • You ignored “most places.”

        No, Chris, I didn’t. I said you had to explain why [edit: as you did this time], not just assert it.

        Dude, you read people who disagree with you so uncharitably

        I love irony.


      • The only reason Austin’s not as flooded as Venice is a series of fairly extensive flood control measures. We could just tear those down, and next thing you know we’re the hottest tourist destination for couples looking for a romantic boat ride piloted by a guy with a funny accent and a giant stick on this side of the Atlantic.


      • Eugene may be the closest we have in the U.S.,

        Eugene? Oregon? Then it must have been dramatically transformed in the last 14 years, because as of 2000, when I was riding it, the only thing it was close to was the bus system of Fort Wayne, Indiana.


      • James, yeah, they have one of two or three true Bus Rapid Transit systems in the U.S. I believe it opened in ’07 or ’08. It’s a great, great bus, with its own lanes and wonderful covered stops. I dream about it every time I save 2-3 minutes on one of Austin’s “rapid” buses after baking in direct afternoon sunlight at one of the “covered” stops. It’s only one line, I think, but one, maybe two more and Eugene will pretty much be set.

        Seattle has a great BRT system as well, I hear, though I haven’t ridden it.


      • To tell you how bad Austin’s “rapid” system is, it’s not uncommon on the main route (the 801, for those who might be visiting soon) for the “local” bus, that is the non-rapid bus, to pass the “rapid” bus in traffic. I have a friend who frequently boasts about her ability to make it from Austin’s midtown (the Triangle) to campus on her bike faster than the “rapid” bus can make the same trip.


    • Is it necessarily going to be the case that a governmental agency will be as dramatically ossified in its decision-making, and as captured by corruption, as we read described in Lima?

      Good lord, I hope not, and I think the evidence suggests not…unless we’re talking about San Francisco municipal government.


  5. I don’t think anyone ever argued that private services could not act as a transit network. The question is whether those private services are able to provide that service in a manner that benefits both riders *and* drivers.

    It is clear that the existence of multiple independent suppliers is critical to a market that operates in a manner that best serves both supplier and customer interests. If there’s only one supplier–either the government or Uber–then what you get ends up focused on the supplier’s wants at the expense of the customers’.


    • To rephrase you are arguing against the exploitative potential of monopolies and cartels?

      If so I agree. Uber certainly would not be a monopoly as anyone can legally compete with that idea or create another way to skin that cat.


Comments are closed.