1. 외부효과 ( 외부성 ) - 환경경제학 기초 및 응용 - Jonathan Gruber Public Finance and Public Policy ( 제 5 장, 제 6 장 ) Aaron S. Yelowitz - Copyright 2005 © Worth Publishers

2. Chapter 5 Externalities Problems and Solutions Jonathan Gruber Public Finance and Public Policy Aaron S. Yelowitz - Copyright 2005 © Worth Publishers

3. Introduction Externalities arise whenever the actions of one party make another party worse or better off, yet the first party neither bears the costs nor receives the benefits of doing so. As we will see, this represents a market failure for which government action could be appropriate and improve welfare.

4. Introduction Externalities can be negative or positive: Acid rain, global warming, pollution, or a neighbor’s loud music are all negative externalities. Research and development or asking good questions in class are positive externalities.

5. Introduction Consider global warming, a negative externality. Many scientists believe this warming trend is caused by human activity, namely the use of fossil fuels. These fuels, such as coal, oil, natural gas, and gasoline produce carbon dioxide that in turn traps heat from the sun in the earth’s atmosphere. Figure 1 Figure 1 shows the trend in warming over the last century.

6. Figure 1 This table shows the global temperature during the 20 th century. There has been a distinct trend upward in temperature

7. Introduction Although this warming trend has negative effects overall on society, the distributional consequences vary. In much of the United States, warmer temperatures will improve agricultural output and quality of life. In Bangladesh, which is near sea-level, much of the country will be flooded by rising sea levels. If you’re wondering why you should care about Bangladesh, then you have identified the market failure that arises from externalities. From your private perspective, you shouldn’t!

8. EXTERNALITY THEORY Externalities can either be negative or positive, and they can also arise on the supply side (production externalities) or the demand side (consumption externalities). A negative production externality is when a firm’s production reduces the well-being of others who are not compensated by the firm. A negative consumption externality is when an individual’s consumption reduces the well-being of others who are not compensated by the individual. The basic concepts in positive externalities mirror those in negative externalities.

9. Economics of Negative Production Externalities To understand the case of negative production externalities, consider the following example: A profit-maximizing steel firm, as a by-product of its production, dumps sludge into a river. The fishermen downstream are harmed by this activity, as the fish die and their profits fall. This is a negative production externalities because: Fishermen downstream are adversely affected. And they are not compensated for this harm. Figure 2 Figure 2 illustrates each party’s incentives in this situation.

10. Price of steel p1p1 p2p2 0Q2Q2 Q1Q1 This framework does not capture the harm done to the fishery, however. The steel firm sets PMB=PMC to find its privately optimal profit maximizing output, Q 1. Q STEEL D = PMB = SMB S=PMC SMC = PMC + MD MD Figure 2 Negative Production Externalities The socially optimal level of production is at Q 2, the intersection of SMC and SMB. The yellow triangle is the consumer and producer surplus at Q 1. The marginal damage curve (MD) represents the fishery’s harm per unit. The social marginal cost is the sum of PMC and MD, and represents the cost to society. The red triangle is the deadweight loss from the private production level. The steel firm overproduces from society’s viewpoint.

11. Economics of Negative Production Externalities The steel firm’s privately optimal production solves: This yields a quantity of steel Q 1 at a price of P 1.

12. Economics of Negative Production Externalities The steel firm’s emits pollution causing damage to the fishery. This is represented by the marginal damage curve. Ideally, the fishery prefers: This would yield zero steel production, which is obviously not in the steel firm’s best interests.

13. Economics of Negative Production Externalities The social marginal cost accounts for both the direct costs to the steel firm and the indirect harm to the fishery: We find the socially optimal quantity of steel Q 2 at a price of P 2, by solving:

14. Economics of Negative Production Externalities The socially optimal quantity entails less production of steel. By doing so, the steel firm would be worse off but the fishery would be better off. Graphically, this triangle in between the PMB and PMC curves from Q 2 to Q 1. The damage to the fishery is reduced as well. Graphically, this is the area under the MD curve from Q 2 to Q 1.

15. Economics of Negative Production Externalities The deadweight loss from the original production level Q 1 is graphically illustrated as the triangle in between the SMC and SMB curves from Q 2 to Q 1. Note that the SMB equals the PMB curve in this case.

16. Negative Consumption Externalities We now move on to negative consumption externalities. Consider the following example: A person at a restaurant smokes cigarettes. That smoking has a negative effect on your enjoyment of the restaurant meal. In this case, the consumption of a good reduces the well-being of someone else. Figure 3 Figure 3 illustrates each party’s incentives in the presence of a negative consumption externality.

17. Q CIGARETTES Price of cigarettes 0Q2Q2 D=PMB Q1Q1 p1p1 S=PMC=SMC SMB=PMB-MD MD p2p2 The yellow triangle is the surplus to the smokers (and producers) at Q 1. This framework does not capture the harm done to non-smokers, however. The smoker sets PMB=PMC to find his privately optimal quantity of cigarettes, Q 1. The MD curve represents the nonsmoker’s harm per pack of cigarettes. The social marginal benefit is the difference between PMB and MD. The socially optimal level of smoking is at Q 2, the intersection of SMC and SMB. The smoker consumes too many cigarettes from society’s viewpoint. The red triangle is the deadweight loss from the private production level. Figure 3 Negative Consumption Externalities

18. The smoker’s privately optimal quantity solves: This yields a quantity of cigarettes Q 1 at a price of P 1. The surplus is the same as before.

19. Negative Consumption Externalities The smoker’s consumption causes damage to the other restaurant patrons. They would prefer: This would yield zero cigarette smoking, which is detrimental to the smoker.

20. Negative Consumption Externalities The social marginal benefit accounts for both the direct benefit to the smoker and the indirect harm to the other patrons: We find the socially optimal quantity of cigarettes Q 2 at a price of P 2, by solving:

21. Negative Consumption Externalities The socially optimal quantity entails less smoking. By doing so, the cigarette smoker is worse off, but the other patrons are better off. The surplus to the smoker (and tobacco companies) falls. Graphically, this is the triangle in between the PMB and PMC curves from Q 2 to Q 1. The harm to other restaurant patrons is reduced as well. Graphically, this is the area under the MD curve from Q 2 to Q 1.

22. Negative Consumption Externalities The deadweight loss from the original consumption level Q 1 is illustrated graphically as the triangle in between the SMC and SMB curves from Q 2 to Q 1. Note that the SMC equals the PMC curve in this case.

23. The Externality of SUVs Consider a real-life example: the use of sport utility vehicles (SUVs). They create three sorts of externalities: Environmental externalities: They consume a lot of gasoline and create more pollution. Wear and tear on roads: SUV drivers do not bear the costs that result from their vehicles. Safety externalities: When SUVs are in accidents, the other drivers are often more severely injured. Application

24. Positive Externalities Positive externalities can occur in production or consumption. A positive production externality is when a firm’s production increases the well-being of others, but the firm is not compensated by those others. Research and development is a production externality. A positive consumption externality is when an individual’s consumption increases the well-being of others, but the individual is not compensated by those others. Nice landscaping could be a consumption externality.

25. Positive Externalities Let’s consider positive production externalities. Consider the following example: A policeman buys donuts near your home. As a consequence, the neighbors are safer because of the policeman’s continued presence. In this case, the production of donuts increases the well-being of the neighbors. Figure 4 Figure 4 illustrates each party’s incentives in the presence of a positive production externality.

26. Q DONUTS Price of donuts 0Q2Q2 D = PMB = SMB Q1Q1 p1p1 S = PMC SMC = PMC - EMB EMB p2p2 The social marginal cost subtracts EMB from PMC. The socially optimal level of donuts is at Q 2, the intersection of SMC and SMB. This framework does not capture the benefit to the neighbors, however. The yellow triangle is the consumer and producer surplus at Q 1. The donut shop sets PMB = PMC to find its privately optimal profit maximizing output, Q 1. The external marginal benefit (EMB) represents the neighbor’s benefit. The donut shop underproduces from society’s viewpoint. The red triangle is the deadweight loss from the private production level. Figure 4 Positive Production Externalities

27. Positive Externalities The donut shop’s privately optimal production solves: This yields a quantity of donuts Q 1 at a price of P 1.

28. Positive Externalities The shop creates positive externalities to the neighbors through the presence of police. This is represented by the external marginal benefit. Ideally, the neighbors prefer: This would yield much more donut production, which is obviously not in shop’s best interests.

29. Positive Externalities The social marginal cost accounts for both the direct costs to the donut shop and the indirect benefit to the neighbors: We find the socially optimal quantity of donuts Q 2 at a price of P 2, by solving:

30. Positive Externalities The socially optimal quantity entails more production of donuts. By doing so, the donut shop would be worse off but the neighbors would be better off. The consumer and producer surplus fall. Graphically, this triangle is between the PMC and PMB curves from Q 1 to Q 2. The benefit to the neighbors is increased as well. It goes up. Graphically, this is the area under the EMB curve from Q 1 to Q 2.

31. Positive Externalities The deadweight loss from the original donut production level Q 1 is graphically illustrated by the triangle in between the SMB and SMC curves from Q 1 to Q 2. Note that the SMB equals the PMB curve in this case.

32. Positive Externalities Finally, there can be positive consumption externalities. A neighbor’s improved landscape is a good example of this. The graphical analysis is similar to negative consumption externalities, except that the SMB curve shifts outward, not inward.

33. Positive Externalities The theory shows that when a negative externality is present, the private market will produce too much of the good, creating deadweight loss. When a positive externality is present, the private market produces too little of the good, again creating deadweight loss.

34. The Solution (Coase Theorem) The Coase Theorem: When there are well-defined property rights and costless bargaining, then negotiations between the parties will bring about the socially efficient level. Thus, the role of government intervention may be very limited—that of simply enforcing property rights.

35. The Solution (Coase Theorem) Consider the Coase Theorem in the context of the negative production externality example from before. Give the fishermen property rights over the amount of steel production. Figure 5 Figure 5 illustrates this scenario.

36. Q STEEL Price of steel 0Q2Q2 D = PMB SMB Q1Q1 p1p1 S = PMC SMC = PMC + MD MD p2p2 But there is room to bargain. The steel firm gets a lot of surplus from the first unit. 12 This bargaining process will continue until the socially efficient level. There is still room to bargain. The steel firm gets a bit less surplus from the second unit. Thus, it is possible for the steel firm to “bribe” the fishery in order to produce the first unit. The reason is because any steel production makes the fishery worse off. Thus, it is possible for the steel firm to “bribe” the fishery in order to produce the next unit. If the fishery had property rights, it would initially impose zero steel production. While the fishery suffers only a modest amount of damage. While the fishery suffers the same damage as from the first unit. Figure 5 Negative Production Externalities and Bargaining The gain to society is this area, the difference between (PMB - PMC) and MD for the second unit. The gain to society is this area, the difference between (PMB - PMC) and MD for the first unit.

37. The Solution (Coase theorem) Through a process of bargaining, the steel firm will bribe the fishery to arrive at Q 2, the socially optimal level. After that point, the MD exceeds (PMB - PMC), so the steel firm cannot come up with a large enough bribe to expand production further.

38. The Solution (Coase Theorem) Another implication of the Coase Theorem is that the efficient solution does not depend on which party is assigned the property rights, as long as someone is assigned them. The direction in which the bribes go does depend on the assignment, however. Now, let’s give the property rights to the steel firm over the amount of steel production. Figure 6 Figure 6 illustrates this scenario.

39. Figure 6 Negative Production Externalities and Bargaining Q STEEL Price of steel 0Q2Q2 D=PMB=SMB Q1Q1 p1p1 S = PMC SMC = PMC + MD MD p2p2 The fishery gets a lot of surplus from cutting back steel production by one unit. This level of production maximizes the consumer and producer surplus. If the steel firm had property rights, it would initially choose Q 1. While the steel firm suffers a larger loss in profits. The gain to society is this area, the difference between MD and (PMB-PMC) by cutting back 1 unit. While the steel firm suffers only a modest loss in profits. The gain to society is this area, the difference between MD and (PMB - PMC) by cutting another unit. This bargaining process will continue until the socially efficient level. Thus, it is possible for the fishery to “bribe” the steel firm to cut back another unit. Thus, it is possible for the fishery to “bribe” the steel firm to cut back. The fishery gets the same surplus as cutting back from the first unit.

40. The Solution (Coase Theorem) Figure 6 Figure 6 shows that even though the bargaining process is somewhat different, the socially efficient quantity of Q 2 is achieved.

41. Problems with Coasian Solutions There are several problems with the Coase Theorem, however. The assignment problem The holdout problem The free rider problem Transaction costs and negotiating problems

42. Problems with Coasian Solutions The “assignment problem” relates to two issues: It can be difficult to truly assign blame. It is hard to value the marginal damage in reality.

43. Problems with Coasian Solutions The “holdout problem” arises when the property rights in question are held by more than one party. The shared property rights give each party power over all others. This could lead to a breakdown in negotiations.

44. Problems with Coasian Solutions The “free rider” problem is that when an investment has a personal cost but a common benefit, individuals will underinvest. For example, if the steel firm were assigned property rights and you are the last (of many) fishermen to pay, the bribe is larger than the marginal damage to you personally.

45. Problems with Coasian Solutions Finally, it is hard to negotiate when there are large numbers of individuals on one or both sides.

46. Problems with Coasian Solutions In summary, the Coase Theorem is provocative, but perhaps not terribly relevant to many of the most pressing environmental problems.

47. PUBLIC-SECTOR REMEDIES FOR EXTERNALITIES Coasian solutions are insufficient to deal with large scale externalities. Public policy makes use of three types of remedies to address negative externalities: Corrective taxation Subsidies Regulation

48. Corrective Taxation The government can impose a “Pigouvian” tax on the steel firm, which lower its output and reduces deadweight loss. If the per-unit tax equals the marginal damage at the socially optimal quantity, the firm will cut back to that point. Figure 7 Figure 7 illustrates such a tax.

49. Q STEEL Price of steel 0Q2Q2 D = PMB = SMB Q1Q1 p1p1 S=PMC SMC=PMC+MD p2p2 The steel firm initially produces at Q 1, the intersection of PMC and PMB. Imposing a tax shifts the PMC curve upward and reduces steel production. S=PMC+tax Imposing a tax equal to the MD shifts the PMC curve such that it equals SMC. The socially optimal level of production, Q 2, then maximizes profits. Figure 7 Pigouvian Tax

50. Corrective Taxation The Pigouvian tax essentially shifts the private marginal cost. The firm cuts back output, which is a good thing when there is a negative externality.

51. Corrective Taxation The steel firm’s privately optimal production solves: When the tax equals MD, this becomes: But this last equation is simply the one used to determine the efficient level of production.

52. Subsidies The government can impose a “Pigouvian” subsidy on producers of positive externalities, which increases its output. If the subsidy equals the external marginal benefit at the socially optimal quantity, the firm will increase production to that point. Figure 8 Figure 8 illustrates such a subsidy.

53. Q DONUTS Price of donuts 0Q2Q2 D = PMB = SMB Q1Q1 p1p1 S = PMC SMC=PMC-EMB p2p2 The donut shop initially chooses Q 1, maximizing its profits. Providing a subsidy shifts the PMC curve downward. The socially optimal level of donuts, Q 2, is achieved by such a subsidy. Providing a subsidy equal to EMB shifts the PMC curve downward to SMC. Figure 8 Pigouvian Subsidy

54. Subsidies The subsidy also shifts the private marginal cost. The firm cuts expand output, which is a good thing when there is a positive externality.

55. Subsidies The donut shop’s production solves: When the subsidy equals EMB, this becomes: But this last equation is simply the one used to determine the efficient level of production.

56. Regulation Finally, the government can impose quantity regulation, rather than relying on the price mechanism. Figure 9 For example, return to the steel firm in Figure 9.

57. Q STEEL Price of steel 0Q2Q2 D = PMB = SMB Q1Q1 p1p1 S = PMC SMC = PMC + MD p2p2 The firm has an incentive to produce Q 1. Yet the government could simply require it to produce no more than Q 2. Figure 9 Quantity Regulation

58. Regulation In an ideal world, Pigouvian taxation and quantity regulation give identical policy outcomes. In practice, there are complications that may make taxes a more effective means of addressing externalities.

59. DISTINCTIONS BETWEEN THE PRICE AND QUANTITY APPROACHES TO ADDRESSING EXTERNALITIES The key goal is, for any reduction in pollution, to find the least-cost means of achieving that reduction. One approach could simply be to reduce output. Another approach would be to adopt pollution- reduction technology.

60. DISTINCTIONS BETWEEN THE PRICE AND QUANTITY APPROACHES TO ADDRESSING EXTERNALITIES The models we have relied on so far have examined reductions in output. Thus, we will modify this. Our basic model now examines pollution reduction, rather than say, steel production. Figure 10 Figure 10 illustrates its features.

61. Figure 10 Model of Pollution Reduction On its own, the steel company would set Q R =0 and Q Steel =Q 1. QRQR PRPR 0 MD = SMB R*R* S=PMC=SMC D = PMB S=PMC While it faces increasing marginal costs from reducing its pollution level. While the benefit of pollution reduction is zero the firm, society benefits by MD. The good that is being created is “pollution reduction.” Since it pays for the pollution reduction, the SMC is the same as PMC. Pollution reduction has a price associated with it. The steel firm’s private marginal benefit from pollution reduction is zero. Such an action maximizes its profits. The optimal level of pollution reduction is therefore R *. R Full At some level of pollution reduction, the firm has achieved full pollution reduction. More pollution P*P* P Full 0 Thus, the x-axis also measures pollution levels as we move toward the origin.

62. DISTINCTIONS BETWEEN THE PRICE AND QUANTITY APPROACHES TO ADDRESSING EXTERNALITIES Figure 10 As Figure 10 shows, the private market outcome is zero pollution reduction, while the socially efficient level is higher. In the figure, the optimal tax would simply be MD– the firm would reduce pollution levels to R *, because its MC is less than the tax up until that point, but no further. Quantity regulation is even simpler–just mandate pollution reduction of R *.

63. DISTINCTIONS BETWEEN THE PRICE AND QUANTITY APPROACHES TO ADDRESSING EXTERNALITIES Assume now there are two firms, with different technologies for reducing pollution. Assume firm “A” is more efficient than firm “B” at such reduction. Figure 11 Figure 11 illustrates the situation.

64. QRQR PRPR 0 MD=SMB R*R* S = PMC A + PMC B = SMC Firm B has relatively inefficient pollution reduction technology. PMC B PMC A PMC B PMC A For any given output level, PMC B >PMC A. While Firm A’s is more efficient. The SMB curve is the same as before. RA,RBRA,RB Quantity regulation in this way is clearly inefficient, since Firm B is “worse” at reducing pollution. If, instead, we got more reduction from Firm A, we could lower the total social cost. RARA RBRB The efficient level of pollution reduction is the same as before. To get the total marginal cost, we sum horizontally. Efficient regulation is where the marginal cost of pollution reduction for each firm equals SMB. Quantity regulation could involve equal reductions in pollution by both firms, such that R 1 + R 2 = R *. Imposing a Pigouvian tax equal to MD induces these levels of output. Figure 11 Two Firms Emit Pollution

65. DISTINCTIONS BETWEEN THE PRICE AND QUANTITY APPROACHES TO ADDRESSING EXTERNALITIES Figure 11 Figure 11 shows that price regulation through taxes is more efficient than is quantity regulation. A final option is quantity regulation with tradable permits. Idea is to: Issue permits that allow firms to pollute And allow firms to trade the permits

66. DISTINCTIONS BETWEEN THE PRICE AND QUANTITY APPROACHES TO ADDRESSING EXTERNALITIES As in the previous figure, initially the permits might be assigned as quantity regulation was assigned. This means that initially R A = R B. But now Firm B has an interest in buying some of Firm A’s permits, since reducing its emissions costs PMC B (>PMC A ). Both sides could be made better off by Firm A selling a permit to Firm B, and then Firm A simply reducing its pollution level. This trading process continue until PMC B =PMC A.

67. DISTINCTIONS BETWEEN THE PRICE AND QUANTITY APPROACHES TO ADDRESSING EXTERNALITIES Finally, the government may not always know with certainty how costly it is for a firm to reduce its pollution levels. Figure 12 Figure 12 shows the case when the social marginal benefit is “locally flat.”

68. Figure 12 Model with Uncertainty and Locally “Flat” Benefits QRQR PRPR 0 MD = SMB R1R1 PMC 1 First, assume the SMB is downward sloping, but fairly flat. R Full More pollution P Full 0 This could be the case for global warming, for example. In addition, imagine that the government’s best guess of costs is PMC 1. But it is possible for the firm’s costs to be PMC 2. PMC 2 Regulation mandates R 1. If, instead, the government levied a tax, it would equal MD at Q R = R 1. Suppose the true costs are PMC 2. Then there is large deadweight loss. This results in a much smaller DWL, and much less pollution reduction. R3R3

69. DISTINCTIONS BETWEEN THE PRICE AND QUANTITY APPROACHES TO ADDRESSING EXTERNALITIES Figure 13 Figure 13 shows the case when the social marginal benefit is “locally steep.”

70. Figure 13 Model with Uncertainty and Locally “Steep” Benefits QRQR PRPR 0 MD = SMB R1R1 PMC 1 R Full More pollution P Full 0 In addition, imagine that the government’s best guess of costs is PMC 1. But it is possible for the firm’s costs to be PMC 2. PMC 2 Regulation mandates R 1. If, instead, the government levied a tax, it would equal MD at Q R = R 1. This results in a larger DWL, and much less pollution reduction. R3R3 First, assume the SMB is downward sloping, and fairly steep. This could be the case for nuclear leakage, for example. Suppose the true costs are PMC 2. Then there is small deadweight loss.

71. DISTINCTIONS BETWEEN THE PRICE AND QUANTITY APPROACHES TO ADDRESSING EXTERNALITIES These figures show the implications for choice of quantity regulation versus corrective taxes. The key issue is whether the government wants to get the amount of pollution reduction correct, or to minimize firm costs. Quantity regulation assures the desired level of pollution reduction. When it is important to get the right level (such as with nuclear leakage), this instrument works well. However, corrective taxation protects firms against large cost overruns.

72. Recap of Externalities: Problems and Solutions Externality theory Private-sector solutions Public-sector solutions Distinctions between price and quantity approaches to addressing externalities

73. Chapter 6 Externalities in Action: Environmental and Health Externalities Jonathan Gruber Public Finance and Public Policy Aaron S. Yelowitz - Copyright 2005 © Worth Publishers

74. Introduction This lesson will review, in detail, some real life externalities related to the environment and health. Along the way, it also reviews some recent empirical evidence.

75. Introduction The four major areas reviewed are: Acid rain Global warming Smoking Other behaviors

76. ACID RAIN When sulfur dioxide (SO 2 ) and nitrogen oxides (NO X ) are released into the atmosphere, they form sulfuric and nitric acids, respectively. These acids may fall back to earth hundreds of miles away from their original source, known as acid rain. Majority of acid rain in North America caused by SO 2, the majority of which comes from coal-fired power plants concentrated in the Ohio River Valley.

77. The Damage of Acid Rain Acid rain is a negative production externality. It causes damage by: Making lakes more acidic. Eroding forests. Causing damage to property ($5 billion/year). Reducing visibility. Leading to adverse health outcomes.

78. History of Acid Rain Regulation Regulation of acid rain 1970 Clean Air Act 1990 amendments and emissions trading

79. History of Acid Rain Regulation 1970 Clean Air Act set maximum standards for various substances, including SO 2. It set New Source Performance Standards (NSPS) for any new power plant, forcing the plant to either reduce emissions or install scrubbers.

80. History of Acid Rain Regulation Thus, the 1970s Clean Air Act relied on regulation, not corrective taxation or tradable permits. It also created a loophole by excluding older plants from the NSPS. Utilities had a greater incentive to run older, dirtier plants. This is one of the hazards of “partial policy reform.”

81. History of Acid Rain Regulation 1990 Clean Air Act Amendments mandated a reduction of more than 50% of the level of SO 2 nationwide, and included all plants. It offered an SO 2 allowance system that granted plants permits to emit SO 2 in limited quantities, based on their historical fuel utilization.

82. History of Acid Rain Regulation Plants were allowed to buy, sell, or save their allowances. The allowances involved very few restrictions– trading could occur anywhere within the United States, with no approval or review, and the frequency and mechanism of trading were unlimited.

83. History of Acid Rain Regulation The 1990 amendments and emissions trading drew opposition from two very diverse groups: Those opposed on economic grounds, like utilities and coal miners. An industry study predicted the full cost of the regulations to be up to $7.4 billion, with a loss of up to 4 million jobs. It was also opposed by environmentalists. They opposed the 1990 amendments on the grounds that they created a “market for vice and virtue.”

84. History of Acid Rain Regulation Estimates suggest that emissions trading significantly lowered the costs of the 1990 amendments. Over the 1995-2007 period, costs were lowered from $35 billion to $15 billion. Thus, trading has worked to greatly improve the efficiency of regulation. Even environmentalists are now more sympathetic to emissions trading, because it reduces economic opposition. “In less than a decade, emissions trading has gone from being a pariah among policymakers to being a star.” -- Daniel Ellerman, expert on acid rain regulations

85. Estimating the Adverse Health Effects of Particulates Although there is a large empirical literature relating pollution to health outcomes, most studies are not all that convincing. Relate adult mortality in an area to the level of particulates. Areas with more particulates may differ in many other ways from areas with less particulates in ways that affect mortality, such as job safety. Empirical Evidence

86. Estimating the Adverse Health Effects of Particulates Recent empirical work by Chay and Greenstone (2003) is much more convincing. The Clean Air Act of 1970 created a quasi-experiment. Counties were classified as “attainment” or “non- attainment” based on whether their pollution levels were below a threshold. “Non-attainment” counties were subject to state regulations, while “attainment” counties just below the threshold were not subject to these regulations. Empirical Evidence

87. Estimating the Adverse Health Effects of Particulates Clearly, if we examined the 3,000-plus counties in the United States, the non-attainment counties look a lot different than the attainment counties. But if we examine counties very close to the arbitrary threshold, those attainment and non- attainment counties are much more similar. Figure 1 Figure 1 shows the effects of regulation on pollution levels. Empirical Evidence

88. Figure 1 Effect of regulation on pollution levels Nonattainment Annual Mean Regulation Threshold Attainment 196919701971197219731974 50 60 70 80 90 100 110 Trends in Mean TSPs Concentrations, by 1972 Nonattainment Status Mean of Average Daily Readings (Micrograms per Cubic Meter) TSP stands for “total suspended particulates” and is a measure of pollution emissions. Nonattainment Annual Mean Regulation Threshold Attainment There was no obvious trend for either group, however. While attainment counties had lower emissions. The Clean Air Act created a regulation threshold for non- attainment counties. While emissions for the attainment “control” counties showed no obvious trend. The Clean Air Act became effective in 1971, which creates “treatment” and “control” groups. We can examine pollution levels before the law change … Non-attainment counties had higher emissions than the threshold before 1972. After the Clean Air Act, emissions for the non-attainment counties trended down. And after the law change. This law change creates a convincing “difference-in- differences” approach.

89. Estimating the Adverse Health Effects of Particulates Chay and Greenstone go on to examine the infant mortality rate in the different counties over time. The infant mortality rate is the share of newborns who die before their first birthday. They find that infant mortality declined substantially from the regulation-induced emissions. Each 10% decline in emissions led to a 5% decline in the infant mortality rate. They find that 1,300 fewer infants died in 1972 as a result of the Clean Air Act. Empirical Evidence

90. Has the Clean Air Act Been a Success? The overall success of the Clean Air Act is much harder to determine. The regulations were costly. In its first 15 years, the Clean Air Act cost: 600,000 jobs. $75 billion in output. They did result in benefits, too. Health improvements, such as reductions in infant mortality. Burtraw, et al. (1997) estimate that the health benefits alone exceed the cost of reduction by a factor of seven, once the lower-cost trading regime was implemented.

91. GLOBAL WARMING The earth is heated by solar radiation that passes through our atmosphere and warms the earth’s surface. A large portion of the heat is trapped by certain gasses in the earth’s atmosphere, which reflect the heat back toward the earth again. This is known as the greenhouse effect.

92. GLOBAL WARMING The concentration of greenhouse gasses like carbon dioxide and methane has increased due to human activity. Using fossil fuels like coal, oil, and natural gas produce carbon dioxide and contribute to this effect.

93. GLOBAL WARMING The surface temperatures have increased by more than 1 degree Fahrenheit in the past 30 years. Projections for the next 100 years suggest an unprecedented increase by as much as 6-10 degrees. This could have very bad consequences for the environment. Figure 2 Figure 2 shows carbon dioxide emissions by country.

94. Table 2 The U.S. is currently responsible for nearly 25% of the planet’s carbon dioxide emissions. Japan contributes only 5% of annual emissions. Developing counties like China and India emit large quantities of greenhouse gasses.

95. GLOBAL WARMING Global warming is truly a global problem. Carbon emissions in Boston and Bangkok have the same effect on the global environment. The stock, not the flow, of carbon dioxide cause the warming. Thus, it takes a long time to undo the damage. Global warming is a thus a complicated externality involving many nations and many generations of emitters.

96. The Kyoto Treaty The goal of the Kyoto treaty in 1997 was to reduce the emissions of greenhouse gasses to 5% below their 1990 levels. United States and Russia have not signed on; many other of the 38 industrialized nations have, however.

97. The Kyoto Treaty For the United States, the Kyoto treaty would: Mean reducing emissions in 2010 by roughly 30% With a present discounted cost of $1,100,000,000,000. That’s 1.1 trillion. The United States would bear 90% of the total world cost, even though it contributes only 25% of annual greenhouse gas emissions.

98. Can Trading Make Kyoto More Cost-Effective? Kyoto treaty introduced international emissions trading. Under the Kyoto treaty, the industrialized signatories are allowed to trade emissions rights among themselves, as long as the total emissions goals are met. There are tremendous differences across developed nations in terms of meeting these goals, for two reasons: Slow growth in some countries: Relatively easy for a country like Russia to meet its goal. Estimates suggest that emissions trading (say, from Russia to United States) could lower the cost of the treaty by 75%. Environmentally conscious growth: Other countries, like Japan, tend to use more gas and nuclear-powered production. Figure 3 Figure 3 shows the benefits of international emissions trading.

99. Figure 3 The benefits of trading Yet the treaty calls for the U.S. to reduce emissions a lot. Carbon Reduction (millions of metric tons) Price of carbon reduction 0630 It is fairly expensive for the U.S. to reduce its emissions. S US SRSR $210 $20 440190 It is easier for Russia and others to reduce their emissions. And the requirements are lower, too. The total cost to the U.S. is 440x$210. The overall cost, with no emissions trading, is $96 billion. STST $50 With emissions trading, the supply curve is summed horizontally. The cost of worldwide emission reduction is $50 per ton with S T. 590 The total cost to Russia and others is 190x$20. The U.S. buys 400 permits (440-40). The overall cost, with emissions trading, is $32 billion. 40

100. Can Trading Make Kyoto More Cost-Effective? By distributing the reduction from the high-cost United States to the low-cost other nations, the total price is significantly lowered. It is lowered by about a factor of four.

101. Developing country participation The Kyoto treaty does not including developing nations, yet they will produce more than half the world’s emissions by 2030. Cost to developing countries for complying with the treaty is 10 times smaller than for developed countries because developing nations do not need to “retrofit” their industrial base. Including these counties lowers the cost of reducing emissions by another factor of four.

102. Developing country participation Including both developing counties and tradable permits lowers the cost to 1/16 th of the cost without those flexibilities. Resistance among developing counties, however. Solution to including developing nations will likely involve significant international transfers.

103. THE ECONOMICS OF SMOKING Figure 4 As Figure 4 shows, the number of cigarettes smoked has declined substantially in the United States over time.

104. Figure 4

105. THE ECONOMICS OF SMOKING Figure 5 Yet, as Figure 5 shows, smoking still causes more than 440,000 deaths each year, four times as much as AIDS, motor vehicle accidents, homicide, and suicide combined. By 2030, 10 million persons will die annually from smoking-related disease, and it will be the leading cause of death throughout the world.

106. Figure 5

107. THE ECONOMICS OF SMOKING What is the role for government intervention in the case of a decision like smoking? We will go through several arguments: Smoking is bad for you. Smoking is addictive. It generates negative externalities to the health system, workplace, and fire departments. It generates positive externalities to the Social Security and Medicare system. It creates negative externalities to other family members through secondhand smoke.

108. THE ECONOMICS OF SMOKING Smoking is bad for you. In standard utility-maximization, any damage individuals do to themselves from dangerous activities results from a rational tradeoff of benefits against potential costs.

109. THE ECONOMICS OF SMOKING Smoking is addictive. “Rational addicts” understand that each cigarette that they smoke today increases their addiction. Smokers consider not only the cost of today’s pack of cigarettes, but the cost of all additional future packs that will now be purchased because their addiction is deepened. Smokers also understand that smoking doesn’t just reduce health through the current cigarette, but all future cigarettes that will be consumed because of the addiction. With this model, smoking remains a rational choice.

110. The Externalities of Smoking Smoking generates negative externalities due to higher health costs. Smoking-related disease increases U.S. medical care costs by $75 billion annually, 5% of the total. If insurance companies can make actuarial adjustments, they simply charge smokers higher rates. Such adjustments internalize the medical cost externality from smoking. In a simple model, there are no health externalities because smokers pay for the high medical costs through higher premiums.

111. The Externalities of Smoking Actuarial adjustments are often not the case with employer health insurance, however. In this case, the externality is financial, not physical. This is an externality because the social marginal benefit from an individual’s cigarette consumption is below the private marginal benefit–the individual’s coworkers have to pay higher premiums. In addition to higher health costs to the private sector, individuals who receive government insurance exert a negative externality onto taxpayers. The same is true of the uninsured (smokers and non- smokers alike)–they exert negative externalities onto medical providers, who pass along the costs to consumers.

112. The Economics of Smoking Smoking generates negative externalities due to lower workplace productivity and more frequent absences. Firms may be able to adjust wages to compensate for this type of problem. If workers’ wages adjust to compensate for their lower expected productivity, then the externality is internalized, akin to the adjustments in health premiums.

113. The Economics of Smoking Smoking generates negative externalities due to fires, mostly due to falling asleep with a burning cigarette. To the extent the smoker only damages himself and his own property, there is not an externality. But if the fire spreads to other properties, there is an externality. Also costs to fire departments and insurance companies that may not be fully internalized.

114. The Externalities of Smoking Smoking generates positive externalities to taxpayers due to the early deaths of smokers and lower payouts for some social insurance programs. Often contribute payroll tax for Social Security and Medicare during working life, but smokers may not be alive to collect benefits when they are elderly. This is known as the “death benefit.”

115. The Economics of Smoking Smoking generates negative externalities (mostly to other family members) through secondhand smoke. Considerable medical uncertainty about the damages done from this. Moreover, if the smoker maximizes family utility rather than individual utility, he rationally trades off the benefits to himself versus the harm to his family. Evidence suggest family utility maximization is incomplete, however.

116. The Economics of Smoking Taken together, the external costs of smoking are roughly 40¢ per pack of cigarettes in 2003 dollars. Estimates of external costs of secondhand smoke vary widely, from 1¢ to $1.16 per pack. The average federal plus state cigarette tax is over $1 per pack.

117. Should We Care Only About Externalities, or Do “Internalities” Matter Also? Traditional economics approach cares only about externalities that smokers impose on others. Model ignores some key features of the smoking decision that may motivate government intervention. Youth smoking decisions Inability of adults to quit

118. Should We Care Only About Externalities, or Do “Internalities” Matter Also? Youth smoking decisions: 75% of adults who smoke began smoking before their 19 th birthday. Sometime between when we are born and become adults, we start making rational decisions, but it is not clear when this happens. Thus, youths may be unable to make rational tradeoffs between current benefits of smoking and future costs.

119. Should We Care Only About Externalities, or Do “Internalities” Matter Also? Suggestive that teens who smoke may not account for the long-run implications of addiction; they are overly optimistic. Among high school seniors who smoked more than 1 pack per day: Among those who said they would not be smoking 5 years later, 74% still were. Among those who said they would be smoking 5 years later, 72% still were.

120. Should We Care Only About Externalities, or Do “Internalities” Matter Also? Many adults who smoke would like to quit, but are unable to do so. 80% of adults who smoke express a desire to quit, but many fewer actually succeed. Over 80% of smokers try to quit in a typical year, and the average smoker tries every 8.5 months. 54% of serious attempts to quit fail within one week.

121. Should We Care Only About Externalities or Do “Internalities” Matter Also? Smokers may face a self-control problem, an inability to carry out optimal strategies for consumption. Economic theory assumes that individuals not only can maximize their utility, but carry out these optimal plans. Psychology contradicts this assumption: individuals are often unable to carry out long-term plans when there are short-term costs of doing so.

122. Should We Care Only About Externalities or Do “Internalities” Matter Also? Examples include: Smoking Retirement savings Diet and/or exercise.

123. Should We Care Only About Externalities or Do “Internalities” Matter Also? Evidence for self-control problems includes laboratory experiments and real-life institutional features. In experiments, individuals reveal they are willing to be patient in the future, but impatient in the present. For example, in one experiment, individuals preferred a check for $100 today versus one for $200 in the future. Yet the same people preferred a check for $200 eight years from now versus a check for $100 six years from now.

124. Should We Care Only About Externalities, or Do “Internalities” Matter Also? The real-life evidence comes from the demand for commitment devices, used to fight off their bad instincts. Smokers regularly set up systems to restrain themselves, such as betting with others or telling people about their decision. “Christmas clubs” that are low-interest, forced saving. Expensive memberships at health clubs to commit themselves to exercise.

125. Implications for government policy The lack of self-control and irrationalities have implications for government policy. For smokers who make mistakes (rather than deliberate choices), the damage from smoking could be viewed as an internality–the damage people do to themselves through adverse behavior. The internality justifies government intervention in the same way an externality does. The market failure is the cost imposed on a person’s long- run well-being generated by his or her short-run impatience or irrationality.

126. Implications for government policy Recall that the external costs of smoking were roughly 40¢ per pack. The internal costs are much larger: Smoker lives about 6 fewer years than a nonsmoker. A year of life is valued by economists at about $200,000. With this estimate, the internal costs of smoking are about $35 per pack, roughly 100 times larger than the external costs.

127. Implications for government policy Government has several policy tools for addressing “internalities” Information/outreach campaigns–these have reduced the smoking rate a lot over the last 30 years. Reducing access to cigarettes for teenagers. Taxation–elasticity of demand for cigarettes is around -0.5, and higher for youth smokers.

128. The Economics of Other Addictive Behaviors Other health behaviors can be analyzed, too. Alcohol consumption Illicit drugs Obesity

129. The Economics of Other Addictive Behaviors: Drinking Alcohol consumption is an interesting alternative to smoking. Major externality associated with alcohol is drunk driving, with over 17,000 persons per year killed and 500,000 injured. The external cost is roughly $120 billion. Drunk drivers may lose their license, go to jail, or see insurance premiums rise, but are unlikely to bear the full costs of their action.

130. Drinking It is possible that alcohol consumption leads to domestic violence, in which it creates another externality. The Center for Disease Control reports that “Alcohol use is frequently associated with violence between intimate partners.” There are serious questions about causality, however. http://www.cdc.gov/ncipc/factsheets/ipvfacts.htm

131. Drinking Alcohol taxes vary between 9 to 24¢ per ounce of ethanol, while the external costs are calculated to be around 80¢ per ounce. “Internalities” from alcohol consumption are certainly much smaller. Small quantities of alcohol may be good for long-run health. Only small share of drinkers do damage to themselves.

132. Drinking Role for government in drinking is more difficult, because negative externalities arise from small share of drinking that results in drunk driving. In principle, optimal policy would target drunk driving, with steeper fines and penalties. But hard to realistically raise the fines high enough. On the other hand, taxing alcohol consumption is a very blunt instrument. It lowers drinking too much among those who aren’t going to drive drunk.

133. Illicit Drugs Government heavily regulates illicit drugs, such as marijuana, cocaine, ecstasy, and heroin; often they are illegal to consume. For some drugs, essentially no more externalities than with cigarettes.

134. Illicit Drugs Milton Friedman wrote: “The harm to us from the addiction of others arises almost wholly from the fact that drugs are illegal. A recent committee of the American Bar Association estimated that addicts commit one-third to one-half of all street crime in the U.S. Legalize drugs, and street crime will drop dramatically.” Nonetheless, the government appears to have concluded that individuals are not making the right long-term decisions for themselves with respect to illegal drugs.

135. Obesity The prevalence of obesity has risen dramatically in the United States. The fraction of adults classified as obese has risen from 12% in 1960 to 28% in 1999. External and internal costs of obesity may exceed either alcohol or cigarettes. Using tax policy is difficult, however, because of the complicated relationship between food consumption and health.

136. Recap of Externalities in Action: Environmental and Health Externalities Acid rain Global warming The economics of smoking The economics of other addictive behaviors