Urban EconomicsAssignment #3: Land and SprawlNAME: ___________________________________This is the full chapter reading :::Chapter 7 Too Many Cars? Too Much Lawn? Too Much Blight?S ome people claim that cities in the United States have developed too sparsely, particularly in the years since World War II. Residential densities are too low to permit much use of mass transit, and automobile commuting encourages low densities. Americans would be better off, many commentators argue, if they lived closer together, worked closer together, and used more mass transit, as they used to do. The shorthand statement of this problem is that there is too much sprawl. In this chapter we will begin to analyze the complaints about sprawl. First, in a comparison of Tokyo and New York, we will see that U.S.-style cities are not inevitable; cities with a lot less sprawl work perfectly well. We will ask whether better transportation policies would make (or would have made) U.S. cities look more like Tokyo, and whether cities in developing countries should imitate Tokyo or Phoenix. Next, we will look at whether land taxes reduce sprawl, examining the effects of transportation and land taxes by applying the lessons of the preceding chapters. In section IV we will explore a different story about sprawl: the idea that U.S. cities are too sparsely populated not because transportation policies have been deficient but because externality problems, such as like crime, environmental degradation, and rowdyism, have not been handled well. If Newark were safer and if the fragile environments of the West Jersey highlands were better protected, more people would live in Newark and fewer in West Jersey McMansions. In section V, I digress somewhat from the topic of sprawl in order to explain the property rights approach to externalities. We will look at many externality issues in the rest of the booksome related to sprawl and some notand the property rights approach will prove to be a handy tool. To get the most from this chapter, you should be familiar with these concepts: externalities, Pareto imp rovement, potential Pareto imp rovement. You can find these terms in the Glossary. You will also need to understand open-city and closed-city cases. 145146 7 Too Many Cars? Too Much Lawn? Too Much Blight? I. Tokyo and New York The easiest way to argue that U.S. cities could be denser than they are today is to produce an example of a wealthy city that is considerably denser than any American city. Tokyo is such an example. Hatta and Okhawara (1994) makes the comparison. Even though New York is by far the largest, densest, and most transit-dependent metropolitan area in the United States, Tokyo is bigger, denser, and more transit-dependent. Both metropolitan areas cover about the same land area. About twenty-nine million people live in (metropolitan) Tokyo and thirteen million work there; sixteen million live in (metropolitan) New York and seven million work there. (Exactly what a particular metropolitan area consists of is nowhere uniformly defined; Hatta and Okhawara used definitions that tried to be as consistent as possible between the two cities.) Although Chiyoda ward in Tokyo is nowhere near as dense as Manhattan (possibly because the danger of earthquakes until recently inhibited high-rise construction), Tokyo stays at Brooklyn-like densities as far out as the distance from New York to the Jersey shore. The result is that, for the metropolitan area as a whole, Tokyo has about 5,900 residents per square mile, while New York has about 2,900. The Tokyo subway carries roughly twice as many passengers a year as the New York subway; Tokyo train service has five times as many regular riders as New York metropolitan-area train service. (New York mass transit has improved in quality and ridership has grown since the time that Hatta and Okhawara wrote, but it had been so far behind Tokyo that the ratios are little changed.) In 1980 only 3.2 percent of commuters to Chiyoda ward used cars; only 4.2 percent of commuters to the four central wards of Tokyo traveled by car. The volume of transit ridership in Tokyo is great enough that commuter lines are profitable without significant government subsidies, and trains run with astonishing frequency. During rush hour, Chuo-line trains arrive at Tokyo Station every three minutes (almost like an elevator); the New Have nline trains of Metro North arrive at Grand Central Terminal every twenty minutes. It is difficult to see how you could fit thirteen million jobs into some (probably expanded) version of the New York metropolitan area without either extraordinarily long commutes, huge traffic jams, or a mass transit and density pattern like Tokyos. If the economies of scale we talked about in Chapter 2 are important and extend into this range, then cities with only seven million jobs are not going to be as productive, per job, as cities with thirteen million. So New Yorks low density and poor transit system may be making it miss out on economies of scale that Tokyo is realizing. In that sense, New Yorks low density and poor mass transit are responsible for a major loss of productivityand the even lower density and poorer mass transit of other U.S. cities are responsible for much larger losses.147 II. Is Tokyo Better? II. Is Tokyo Better? One possible conclusion to be drawn from this exercise is that Tokyos planners have done a better job than New Yorks. But there is an alternative, more benign interpretation of the U.S. experience. Tokyo may be more dense simply because the alternatives to Tokyo that the Japanese have are not as attractive as the alternatives to New York that Americans have. If Japanese people had alternatives as attractive as, say, Nebraska and California, you would need lawns and cars to entice them to work in Tokyo, just as you need lawns and cars to entice most Americans to work in New York. In 2000, even in the New York metropolitan area (the larger Consolidated Metropolitan Area), 71 percent of households had at least one car and almost 40 percent lived in single-family, detached homes (U.S. Bureau of the Census 2000). In the benign interpretation, the historical trend of declining density and declining mass transit use in the United States is a normal and natural process. Mills (1972) and Mieszkowski and Mills (1993) show that density and land-price gradients have been getting flatter in U.S. cities since about 1820; suburbanization is by no means a postWorld War II phenomenon. Continuing improvements in transportation alone would be sufficient to explain why these gradients have been flattening, but there have been many other contributing factors as well: frozen food, television, and air-conditioning, as I mentioned in Chapter 1; electricity, so that factories didnt need to be near f lowing water or some other power source; telephones, possibly, and rising incomegreenery and privacy are things that people are willing to spend more money on as they grow richer. Flattening density gradients and the eclipse of mass transit in the United States, in this view, are simply progress. Sprawl is a fact of modern life. The benign view does not argue that government actions played no role in the trend toward sprawl, that it sprang unaided from the pure workings of an untrammeled market. Clearly, governments in the United States could have stopped or reversed the trend at any time by taking one of any number of different actions: by banning automobiles, or stopping road construction, or paying for all high-rise construction, or not insuring mortgages, or not policing traffic at all, or enforcing a speed limit of fifteen miles per hour. Rather, the benign view of the U.S. trend is that government actions (and inactions) were more or less justified by cost-benefit analysis: that most road construction produced more in benefits than it absorbed in costs, that gas taxes reasonably meet the cost of air pollution, that larger transit subsidies would cost more than they would be worth, and so on. This view does not require that all relevant government policies have been wise, only that they not have been systematically unwise to such an extent that they created a vastly different density gradient.148 7 Too Many Cars? Too Much Lawn? Too Much Blight? I have already identified some relevant government policies that are probably unwise and probably affect density gradientsthe lack of congestion pricing, the lack of responsibility for accidents that auto insurance creates, the nontaxing of employer-provided parking, the silly structure of mass transit fares, the general undertaxing of land and consequent overtaxing of other commodities. The next few chapters will bring up more issues that affect densitymortgage deductibility, mortgage guarantees, average rather than marginal cost pricing of sewer and water service, zoning, crime, race. Many people will find current and historical policies in these areas deficient, too. But we will also look at other unwise policies that promote centralizationsubsidies to building rail transit, Western-style toilets, and many economic development activities, for instance. The United States (and Japan, too) subsidizes agriculture heavily, and these agricultural subsidies shift land out of urban use and into farming. They make cities too dense and too small. Probably the general drift of bad policy is to encourage low densities, but no one knows how much actual, measurable difference bad policies have made. To figure it out, you would first have to find optimal policies on all these areas and then figure out how U.S. cities would have grown if these policies had been in place. It would be a mammoth project, and no one has come close to doing it. Nor should anybody. If policies are bad, they are bad regardless of what they do to density gradients. The cities that are built today and will be built in the future, in the United States and in the rest of the world, will be built with and for new technologies. Lagos shouldnt end up looking like either New York or Tokyo or Phoenix. Currently, Chinese cars pollute more than American cars do; optimal density for Beijing is going to be different from optimal density for Los Angeles. Whether or not the policies of Robert Moses in the 1930s in New York were the right ones or the wrong ones depends in large measure on the technologies that were available thenhow people drove, shopped, worked, communicated with each other, and played. People in Lagos in decades to come wont be doing any of these things like 1930s New Yorkers did, and so figuring out how New York would look if Moses had followed better policies is not going to help policy makers in Lagos today. The important question is whether the policiestoward car safety, mass transit, housing, agriculture, sewers, and so onare the best they can be, not what a city looks like. If late twenty-first-century Lagos had excellent policies and looked like twentieth-century Phoenix, that would be wonderful. If it had excellent policies and looked like fourth-century Rome, that would be wonderful, too. Cities are a way of helping people live their lives better; theyre not an end in themselves.149 IV. Flight from Blight III. Land Taxes One of the bad policies that most American cities pursue is that they undertax land and overtax improvements; this was one of the conclusions from Chapter 6. This also probably contributes to sprawl. This idea was expressed by Gaffney (1964, p. 132): Think how many miles of wires, ducts, pipes, and conduits there are in a skyscraper, not to mention stairs and elevators; all social overhead capital supplied at private expense, and taxed besides. If the same floor space were sprawled over a section of suburban prairie, the connecting wires et al. would not only be much longer, but supplied at public expense and/or heavily subsidized. We tax vertical transportation but subsidize horizontal. With such a large and systematic fiscal bias at work, is it surprising we get urban sprawl? Reducing taxes on structures and increasing them on land would lower the relative cost of structures (since the price of land to users would stay the same or go up a little) and so induce more building on the same amount of land. In an open-city model, cities would become both more dense and physically larger (since some construction in outlying areas would become profitable). Would this be desirable? It would be a potential Pareto improvement. The underuse of land taxes thus is reason to believe that cities are too sparse, not to believe that they cover too much ground. IV. Flight from Blight Theres an alternative story about U.S. cities, another explanation for why they might be too sparse and too big. Its called theflight-from-blight hypothesis. The idea is that policies have done a worse job than they should have in promoting health, safety, and morals in the older, more central parts of metropolitan areas. As a result, people who could afford to have moved to sparsely populated suburbs in greater numbers than they should have. Because crime is high in inner cities, for instance, people flee from them and build homes far away, in areas that should instead be left for fields and forests. Two kinds of everyday experiences give the flight-from-blight story some plausibility. The inner parts of many U.S. cities have lots of problems, and so do many of the people who live there. This is especially true for older cities in the Northeast and Midwest. More activity used to occur in these areas, and markets dont seem to be working right in them now. On the other hand, knocking down trees and tearing up farms to build housing developments and shopping malls on the outskirts of metropolitan areas has obvious environ150 7 Too Many Cars? Too Much Lawn? Too Much Blight? mental costs. If inner cities could be restored to their former glory, the thinking goes, the farms and forests could be spared and everyone (except maybe some greedy developers) would be better off. A generalized yearning for the old days, like this one, however, is not the same as a convincing demonstration of potential Pareto improvements. Several steps are missing. The first step is to show not just that externalities occur but that they are excessive. Cars are going to crash, influenza is going to spread, distraught lovers are going to murder each other, and thugs are going to rob gas stations these are facts of life that by themselves are of no interest to economists. The relevant question is not whether these events occur at all but whether some instances of these events could be avoided cheaply enough that the winners from avoiding them could compensate the losers. Thus to address flight from blight, we need to know not whether blight exists but whether too much blight exists. Second, we need to know where the excess blight is, assuming it is somewhere. The optimal level of traffic accidents, for instance, is almost certainly higher in more densely populated areas, where streets are narrower and lined with parked cars and children play on the sidewalks rather than in playgrounds or backyards. Or racial tensions may be greater when you can hear your neighbors toilet flushing and he has to listen to your childs birthday party. If this is the sort of blight that people are taking flight from, then it is surely no cause for worryand it may even be worth cutting down a few trees to facilitate their move. Third, we need to specify clearly what peoples alternatives are. Flight from blight makes little sense in the open-city model, for instance. In that model, if downtown Los Angeles is dirtier or more crime-ridden than it should be, there is no impact on the population of suburban Orange County or San Bernardino County. The people who would have lived or worked in downtown Los Angeles stay in Mexico or move to Denver instead. The only way to induce fewer people to move to Orange or San Bernardino County is to make those places less attractive, and making downtown Los Angeles nicer doesnt do that. Finally, even if excess blight is concentrated in the more densely populated parts of metropolitan areas, flight from blight is a serious problem only if the reduction in activity that excess blight causes is great. If excess blight causes little movement, then the densities and the dimensions of U.S. metropolitan areas are fundamentally all right, even if a few jobs or residences are in the wrong places. On this last question, economists have done a substantial amount of empirical research, and their conclusions have generally not supported the flightfrom-blight hypothesis. Mieszkowski and Mills (1993, p. 140) find that the timing of suburbanization argues against a large role for flight from blight:151 V. Negotiation and Property Rights The most rapid pace of suburbanizatio noccurredbetween 1920 and 1950, a period not known for racial tensions, school desegregation, rising local taxes and high crime rates. In fact, between 1940 and 1950, national crime rates fell. After reviewing a number of cross-section studies that tried to find correlations between blight and flight, they conclude that cities with more crime, worse schools, and higher taxes dont lose either jobs or residents faster. The cross-section studies do show that metropolitan areas with a higher percentage of African Americans in the central city decentralize faster. This f inding would strongly support flight from blight if African Americans were blight, but that is ridiculous. African Americans are some of the most educated, talented, productive, and accomplished people the world has ever known. This finding in the cross-section studies tells us that race relations affect spatial location, but it doesnt tell us what to do about race relations. That is a topic I will discuss in detail in Chapters 11 and 12. The cross-section studies are hardly definitive evidence against the flightfrom-blight story, however, because they do not measure the effect of excess blight; they just measure blight. Finding out how much blight is excessive and where its located is a much more difficult task, one best accomplished by looking at each possible source of blight in some depth. That is the strategy I will follow in this book. This piecemeal strategy has a great advantage over trying to put together a more overarching approach to sprawl: it makes for better policy recommendations. Whats wrong with sprawl (if anything) is that it precludes certain potential Pareto improvements. If we can identify those potential Pareto improvements directly and figure out what to do about them, then we really dont have to spend any time thinking about sprawl, defining it, or measuring it. For instance, if metropolitan areas are too sparsely populated because inner-city schools are not as good as they should be, then the appropriate policy is to improve inner-city schools, not to build subways or preserve farmland. Lousy schools are a problem no matter how much sprawl they cause or dont cause. V. Negotiation and Property Rights Policies toward externalities are going to be a major theme in the rest of this book, just as they were in the preceding chapters on transportation. In those chapters, all you needed to know about externalities was that if you make people bear the full cost of their actions, all will be well (or efficient, at least). For much of the following text, that basic knowledge will not be enough. A more sophisticated understanding of externalities will be required. The possibility of negotiation is one area in which greater sophistication would help. In the discussions of traffic policy, negotiation was not feasible:152 7 Too Many Cars? Too Much Lawn? Too Much Blight? you cant roll down your window in the Lincoln Tunnel queue and shout out an offer to all the drivers behind you that you will let them pass you if they collectively pay you enough. Sometimes, though, negotiation is feasible, and when it is, policy prescriptions are different. Consider, for example, two people who will be in a small room together for a while. One is a smoker who would like to smoke; the other is a nonsmoker who would be annoyed if the smoker smoked. Suppose they can talk with each other. Then the nonsmoker will try to bribe the smoker not to smoke. If smoking causes the nonsmoker more harm than its worth to the smoker, the nonsmoker will succeed and the smoker wont smoke. If not, the smoker will smoke. In either case, after they have talked, no potential Pareto improvement is possible: if the smoker isnt smoking, its because smoking causes more harm than benefit; if the smoker is smoking, its because the opposite is true. The same result will occur in a society where smokers need permission to smoke, so long as negotiation is possible. (Perhaps the nonsmokers can call the smoking police to stop the smokers, but can also refrain from doing so.) In this case it is the smoker who will take the initiative and try to bribe the nonsmoker. If smoking causes more harm to the nonsmoker than its worth to the smoker, the nonsmoker will refuse the smokers entreaties and there will be no smoking. If not, the bribe will be accepted and the smoker will smoke. Thus, in terms of efficiency, so long as they can talk it doesnt matter who needs permission from whom: the smoker will smoke if smoking causes more benefit than harm, and will not smoke if the opposite is the case. Negotiation will lead to efficiency. This result is known as the Coase theorem, after Ronald Coase who published the first account of it in 1960. How does the Coase theorem modify the way we should think about public policy toward externalities? Suppose the smoker and nonsmoker negotiated, the smoker paid the nonsmoker for permission to smoke, and then lit up. If a public-spirited analyst then came on the scene and were naive about the Coase theorem, she would observe the nonsmoker wheezing and coughing, the smoker puffing, and conclude that a tax on smoking could produce a potential Pareto improvement, but she would be wrong because she missed out on the negotiations. If the parties have already negotiated, nothing more can produce a potential Pareto improvement. If a tax is imposed and it stops the smoker from smoking, the gain to the nonsmoker is less than the loss to the smoker. Government intervention isnt needed and doesnt help when the parties have negotiated; just looking at a physical relationship that appears to be an externality doesnt tell you everything you need to know. That doesnt mean that government intervention can never produce a potential Pareto improvement. Sometimesas in the case of traffic and air pollution, for instancenegotiation is impossible. The Coase theorem means that, in arguing for a government intervention, you have to show not only that153 V. Negotiation and Property Rights what someone does affects someone else, but also that the parties cannot negotiate themselves out of the problem. The Coase theorem also suggests another type of policy, aside from taxes, subsidies, or tolls, to alleviate the problems caused by externalitieshelping the parties negotiate. The smoker and the nonsmoker could negotiate only because we imagined there were clear rules about who could agree to what and what would happen in the absence of an agreement and that there was some way to enforce an agreement. Without clear guidelines, negotiation cant take place. One way to alleviate externalities, therefore, is to provide such guidelines. This is generally a job for governments. This idea of encouraging negotiation by straightening out ambiguities is called the property-rights approach to externalities. Property is the branch of law concerned with who gets to do what with what, and so establishing property rights is just another way of saying writing down clear rules for negotiating. Notice that ownership is not a simple, absolute concept; writing down clear rules for negotiating can take a long time, and there are many different ways of doing it. My owning the Coca-Cola trademark may or may not imply that I can keep someone else from using the URL www.coca-cola.com. My owning a cigarette may or may not permit me to smoke it in an elevator. My owning a mountain bike may or may not permit me to ride it in Central Park. My owning a tire iron may or may not (it probably does not) permit me to beat you over the head with it. Lawyers sometimes refer to property as a bundle of sticks. By that they mean that ownership is really a collection of rights. What we call owning a tire iron consists of the right to keep other people from using it (except the government, and even the government can use it only under certain conditions), the right to carry it in your trunk, the right to change your own tires with it, the right to prop open a door you own with it, the right to display it in your living room, the right to melt it down and recast the metal into little statues of yourself, the right to throw it at rabbits and snakes (real or imaginary) in your backyard, and so on. These are the sticks that are in the bundle. Owning a tire iron, in most of the United States today, does not include the right to use it to remove other peoples tires unless they say you can, the right to attach it perpendicularly to your front bumper, the right to throw it at imaginary rabbits on a crowded sidewalk, the right to melt it down and recast the metal into little statues of Mickey Mouse and sell them on the street, the right to have it displayed in the Metropolitan Museum of Art, and so on. These are some of the sticks not in the bundle labeled owning a tire iron. The Coase theorem says that, so far as efficiency is concerned, if people can negotiate, it doesnt matter which sticks are in which bundle. But every stick has to be assigned unambiguously to one and only one bundle, and its154 7 Too Many Cars? Too Much Lawn? Too Much Blight? the governments job to assign them. Does owning a tire iron give you the right to melt it down and sell statues that almost look like Mickey Mouse? Can you put your tire iron on your dashboard and drive with a small child in the car? Can you throw it at a snake in a national forest? I dont know. Even such a prosaic and well-worn concept as owning a tire iron has ambiguities, and changes frequently. In more technologically dynamic arenas like the Internet, organ transplants, and the deep seabed, sticks that are in no bundle are constantly turning up. Efficient negotiation can take place only after those sticks have found a home in some bundle. Finally, while the Coase theorem indicates that no matter what sticks are in which bundle, negotiation will produce an efficient outcome, it doesnt say that nobody should care about what sticks are in which bundle. If you own the Coca-Cola trademark, it sure makes a lot of difference to you whether or not that means you also own the URL www.coca-cola.com. Considerations like perceived justice and fairness are important in designing property rights. (In fact, one interpretation of the Coase theorem is that if negotiating opportunities are good enough, justice and fairness are the only things you need to worry about in assigning property rights, because efficiency will take care of itself.) VI. Conclusion Up until a century ago, moving to a large city meant that you were more likely to die soon; if you didnt die right away, your children were more likely to die at birth, and if they survived, they were more likely to be stunted and sickly. Population density was at the root of this public health disaster, and until these conditions were ameliorated, no city could survive with many more than a million people in it. Improvements in medicine, nutrition, and transportation, along with rising incomes, have combined over the past century to make cities healthier places to live, healthier even than rural areas. As human diseases mutate and evolve, however, there is no guarantee that this happy state will continue indefinitely. Lowering the population density in U.S. cities was an important process for improving urban health a hundred years agoit created positive externalities then. More recently, though, some have argued that the process has gone too far. Essentially, two arguments have been advanced for why sprawl is excessive in the United States. One emphasizes transportation: policies have been too friendly to cars and too unfriendly to mass transit. The other argument emphasizes externalities in the older central parts of metropolitan areas: because these factorssources of urban blighthave not been controlled as well as they should be, too many people have fled to the suburbs.155 Questions Certainly there are many reasons to think that reducing sprawl might be a potential Pareto improvement. For instance, Americans drive too much and cause too much congestion, pollution, and damage from traffic accidents. Land taxes are underused. Parking is subsidized. Correcting these problems would result in significant potential Pareto improvements, even if it didnt reduce sprawl much. So would reducing excessive central-city blight. Other distortions, such as farm subsidies, work in the opposite directionthey make cities too small and too dense. The major reason U.S. cities are larger and less dense today than they were a hundred years ago is that people are richer and technology is better. The change was not produced totally through some conspiracy or by a mistake. The prescription to relieve excessive sprawl is identical to the prescription for the problems that make sprawl excessive. Making potential Pareto improvements has some value; reducing sprawl per se does not. We need to look at sources of urban blight one by one. This will require a more sophisticated approach to externalities than was used in the transportation chapters. To think seriously about externalities, you have to consider possibilities for negotiation and you have to understand the various ways property rights can be put together. Like externalities, property is a much more complex idea than it first appears. Questions 1. Tetracycline is an antibiotic that has quite a few different uses. It can relieve the symptoms of some comparatively minor ailments (acne and conjunctivitis, for instance), as well as cure some serious diseases (typhus and pneumonia, for example). The bacteria that tetracycline acts ag ainst live short lives and multiply rapidly. Occasionally a mutant that is resistant to tetracycline pops up. The more often tetracycline is used in the world, the higher the proportion of resistant bacteria becomes, and so the less effective tetracycline is. Eventually tetracycline will lose enough effectiveness that people will have to turn to other antibiotics, which will be more expensive and less effective. a. Explain the externality involved in tetracycline use. Outline a potential Pareto improvement. b. Describe a Pigouvian policy for realizing this potential Pareto improvement. c. Describe a Coasian policy for realizing this potential Pareto improvement.156 7 Too Many Cars? Too Much Lawn? Too Much Blight? 2. Cybersquatting is a practice in which ordinary people acquire e-mail addresses and Web site addresses (URLs) with names that are popularly associated with famous companies. The companies then often have to buy the names back from the squatters. In the short run, what does the Coase theorem imply about public policy toward cybersquatting? Should the Internet be allowed to develop along Coasian lines? 3. The total benefit (measured in dollars) that a musician gets from making music is B m = 25 ln D, where D denotes the decibels involved. A sleeper lives next door. The total benefit that she gets is B s = 20 ln (20 D ), where 20 is the loudest sound technically possible. a. Find the level of sound that maximizes total benefits for musician and sleeper. b. Suppose no ordinance about noise exists. What will happen if the two can bargain freely? Is this efficient? c. The city adopts an ordinance forbidding noise levels above 5 decibels. Being drawn and quartered is the penalty for violating the ordinance, but the city has hired no one to enforce the ordinance. It will respond only to complaints. What will happen? Is this efficient? d. The city revises the ordinance. The penalty for violating the ordinance is now $1 for each decibel over 5. Enforcement is by inspectors who are perfectly effective, omnipresent, and absolutely incorruptible. Complaints have no standingthe ordinance will be enforced whether or not the neighbors complain. What will happen? Will this be efficient? How does this compare with the situation in part c? If you were on the city council, would you support the rev