Lecture April 9 Positive vs normative analysis Social choices Chapter 2 Positive vs Normative Analysis Positive economic analysis: observes and describes economic phenomena objectively. Normative economic analysis: identifies what policy should be pursued. Positive analysis tries to be value neutral (but values may `creep in') by doing objective descriptions of the world around us. Useful for policy makers, by answering questions like ow much lower will CO 2 emissions be if we raise the CO 2 tax by x %? What is the relationship between climate and profitability in agriculture. Normative analysis is useful in public policy, by answering questions like: What policy will be best for society? ow much CO 2 emissions should be accepted? ow much environmental protection should we have? Chapter 3 Making societal choices Environmental protection is an example of a question where we must make a collective decision, a social choice Everyone can decide by themself how much ice-cream to eat. Even if you only eat one I can choose to eat two. But since we live in the same environment we cannot make different decisions about the quality of the environment. So, how do we translate individual preferences into a social choice? Example 1: Air Pollution Control in Santiago. The most important source of pollution is emissions from busses and cars. The emissions from buses and cars can be reduced at a cost, which will affect the travelers. If buses has to reduce emissions they need to increase prices of tickets, this will merely harm the poor that go by bus. The rich may have a higher willingness to pay for an improved air quality. (But we may of course also use taxes on cars to solve the problem) Example 2 The California gnatcatcher, an endangered bird in coastal Southern California If widespread homebuilding is allowed in the bird's habitats, there is a risk of extinction. People also want to live there. If we make protection areas to protect the birds we cannot build that much houses. Who should decide? Residents of (coastal Southern) California? All residents of California? All residence of the world Future generations? Who should pay for the habitat preservation? 1
Example 3 The Three Gorges am, giving a huge water reservoir in the Yangtze river in China On the positive side: a lot of electricity is generated, reduced emissions from coal power plants. On the negative side: ancient cities are destroyed and spectacular canyons are lost We cannot produce electricity for those that prefer that and preserve the canyons for those that prefer that. An either or problem, We must make a social choice Biocentrism / Antropocentrism / Utilitarism An important distinction: Something has an instrumental value if it can serve as an instrument to achieve something useful Something can be totally useless and still have intrinsic value According to Biocentrism all living things have intrinsic value, regardless of their instrumental value. According to Anthropocentrism things have a value only if they are useful to human beings According to Utilitarianism it is the preferences of people that matter. If people thinks that a species has an intrinsic value it will have a value. Biocentrism / Antropocentrism / Utilitarism A problem with Biocentrism is that different species often has different interests in an environmental change. Climate change will benefit some species and hurt others. So Biocentrism has difficulties in deciding whether it is good or bad. From a utilitarian or anthropocentric perspective we can say that climate change is bad if human beings are among those species that looses. Some species like wolfs, lions and elephants probably have a negative instrumental value. Lions and wolfs eat our cattle and elephants destroy houses and gardens. But if people thinks that they have an intrinsic value their utility may be decreased if those species are extinct. And then the intrinsic value would end up in the utilitarian analyses as well. Precautionary Principle If there is a risk for a dangerous outcome of a project we should avoid that project. Minimizing bad outcomes rather than optimizing good outcomes. The principle may paralyze the decision process since a lot of projects have some very bad outcomes although unlikely to occur. Moreover if we would like to replace a chemical that we know is harmful with another that probably are less harmful but may be more harmful what should we do according to the precautionary principle? Should we use retardants to reduce the risk of fires or should we avoid them to reduce the risks of getting poisoned? What is the precautionary action? Sustainability A popular word with somewhat unclear definition. According to the Bruntland commission: evelopment that meets the needs of the present without compromising the ability of future generations to meet their own needs ow large are the needs of our generation? ow large are the needs of future generations? Economist have tried to make clearer definitions of the concept. Most of these definitions a based on an interpretation that future generations should have at least the same welfare or utility that we have. Sustainability A central issue, could natural resources be replaced by man made capital? Are natural capital and real capital substitutes? Maybe, since we have invented computers and internet we do not need newspapers anymore and do not need that many forests. Or are they complements? If all forest are dying we may not have much needs for big sawmills 2
Strong sustainability ecological perspective The value of natural capital most be preserved to future generations. Weak sustainaility neoclasical perspective The total value of capital must be preserved but natural capital can be replaced by real capital. The Aral Sea Irrigation in Uzbekistan and Turkmenistan, in former Soviet Union, made the Aral Sea a desert. Weak sustainability - Neoclassical perspective. o the increased harvest in Uzbekistan and Turkmenistan worth more or at least the same as the reduction in fishing etc in the Aral Sea it was a sustainable action to destroy the sea. The problem is reduced to a problem of compensation between the farmers in Uzbekistan and Turkmenistan and the fishermen of the Aral Sea. Strong sustainability - ecological perspective. The ecosystems have changed largely. Will probably not be constant with constant natural capital. Is irrigated farming land, natural capital or real capital? Anthropocentrism or Biocentrism? o the Aral Sea have an intrinsic value? Social Choice from Individual Values / Individual Preferences. To make a societal choice the whole society must agree. Two important questions: 1. ow do we know the preferences of the individuals 2. ow to aggregate the preferences together. Sustainable development A popular word with very unclear meaning 3
The Utility Function FIGURE 3.1 Possible set of indifference curves for an individual who values the environment more at higher levels of material consumption. Assume an economy with N individuals. Individual i:s consumption of the market good is: where =1,, Air quality is denoted with e (Since everyone uses the same air we do not need any index.) Each individual has her own utility function. Normally we assume that we only care about our own consumption. The utility functions of each individual are in that case: =,e =,e Assume the following utility functions: e e Assume the following utility functions: And that we choose between the following alternatives. =,, = 2,2,3 =,, = 2,4,2 And that we choose between the following alternatives. =,, = 2,2,3 =,, = 3,3,2 Utility of individual 1 6 4 Utility of individual 2 6 Utility of individual 1 2, 3, =2,7 3, 2, =2,3 Utility of individual 2 2, 3, =2,3 3, 2, =2,7 Individual 1 would prefer and individual 2 prefer (since individual 2 is the one that pays for the higher air quality) Individual 1 would prefer and individual 2 prefer (since individual 2 prefer consumption before air quality) The Pareto Criterion Unanimity / Pareto Criterion : if all individuals prefer one alternative, then society should do that too. If everybody is at least as well of with (compared to ), and at least one person is better of with than is preferred to Formal definition: If > for at least one individual for all individuals then for the group as a whole is Pareto preferred to If just one individual prefers and all the others prefer a then is not Pareto preferred to. Assume the following utility functions: And that we choose between the following alternatives. =,, = 2,2,3 =,, = 3,3,2 Utility of individual 1 2, 3, =2,7 3, 2, =2,3 Utility of individual 2 2, 3, =2,3 3, 2, =2,7 None of the alternatives is Pareto preferred. 4
Assume the following utility functions: And that we choose between the following alternatives. =,, = 2,2,3 =,, = 3,3,2 Utility of individual 1 2 3=6 3 2=6 Utility of individual 2 2, 3, =2,3 3, 2, =2,7 Is Pareto preferred to FIGURE 3.2 Utility possibilities for two-person society. If we start at point w the blue area gives all possible Pareto improvements. If we start at point x the blue area gives all possible Pareto improvements. What allocations that are supported by the Pareto criterion will depend upon the starting point. On the utility possibility frontier there are no possible Pareto improvement. If there are no possible Pareto improvements the situation is Pareto efficient. The Pareto criterion has a conservative bias, a bias towards status quo since most often at least one person in the world would dislike a specific proposal of a change. The aldor icks compensation principle or the potential Pareto criteria Assume the utility functions : And the alternatives: e =,, = 2,2,3 e =,, = 2,6,2 Utility of individual 1 6 4 Utility of individual 2 6 12 The Pareto criterion cannot rank the alternatives. But assume that the individuals negotiate of a side payment: Given that alternative 2 is chosen individual 2 gives 2 units of the consumption good to individual 1. In that way a third outcome is constructed that is Paretopreferred. =,, = 4,4,2 The aldor icks compensation principle or the potential Pareto criteria Assume the utility functions : And the alternatives: e =,, = 2,2,3 e =,, = 2,6,2 Utility of individual 1 6 4 Utility of individual 2 6 12 is a potential Pareto improvement compared to since it is possible to compensate the losers so that everyone become winners. aldor icks criterion prefers even if no compensations actually occur. If the benefits of the winners are larger then the cost of the losers the project has a positive net benefit. This is the criteria of the cost benefit analysis which is the dominating view among economist on how to make decisions on the societal level. Majority voting and super majority Majority Rule: If a majority prefers to, then `society prefers. Weakness: does not take into account that some individuals may strongly oppose (while the others may be only weakly in favour of ). That is, voting does not take into account the intensity of the preferences. For important changes there is sometimes a requirement of a super majority. For example, there may be a requirement of 2/3 of the votes in the parliament to change the constitution. Majority voting at two alternatives at a time In the political process at the parliaments a common procedure is to vote at two alternatives at the time where one alternative is eliminated in each voting. Usually the chairman decides the order which turns out to be very important. A B C B C A A C B 5
Example olstad, Chapter 3, Exercise 3 3 alternative iking (), ayaking (), looking for odos () Boris > > Maggie > > William > > A Social Welfare Function A Social Welfare Function is a societal version of the individuals utility function. It may be a function of the individual utility functions: A common case: = + +...+ This is an example of a utilitarian SWF (sometimes you may give different weights to different individuals) Utilitarian SWFW= where 0 Rawls argued that the most important was to maximize utility for those individuals that are worse off. Rawlsian SWF W=min ( ) We may illustrate the SWF with indifference curves Utilitarian Rawlsian Egalitarian Egalitharian SWF W= ( ) FIGURE 3.4 Social indifference curves. Assume the utility functions : And the alternatives: e =,, = 2,2,3 e =,, = 2,4,2 Utility of individual 1 6 4 Utility of individual 2 6 A rawlsian SWF would prefer a 1 A utilitarian SWF with equal weights for all individuals would consider the alternatives to be indifferent. According to the Pareto criterion Z is not preferred to X But according to this SWF Z is preferred to X (Probably also according to aldor icks, the potential Pareto criterion) 6
A social welfare function gives us a ranking of all alternatives from the best to the worst. But who should decide what the social welfare function look like? The social welfare function is a theoretical concept that is not very useful in reality when we should make public decisions. Arrows impossibility theorem ifferent ways of construction SWF from individual utility function would give different outcomes. Is there a preferred procedure to construct a social welfare function? A perfect mechanism of social decision-making? Arrow put up a couple of reasonable requirement of such a mechanism and found that it was impossible to make a decision rule that would fulfil all these requirements. Arrows requirements (Axioms): Completeness: we should be able to compare all alternatives Unanimity: if everyone prefers a to b than the SWF should prefer a to b. (The Pareto criterion) No dictatorship: No one should always get their way, no ones preferences should be exactly the same as the preferences of the society. Universality. Any possible individual ranking of the alternative is permissible. Transitivity: If a is social preferred to b and b is social preferred to c then a should be social preferred to c. Independence of irrelevant alternatives. The social ranking between a and b should only depend upon the individuals ranking of a and b regardless of other alternatives. Criticism of the utilitarian perspective. Utility functions may change: Criticism of the utilitarian perspective. Who should decide: In the utilitarian perspective we take utility functions as given and argue over how to aggregate them together. In real decision making processes a lot of effort are spent on changing people preferences Who should decide who should have the right to participate in the decision making? Especially complicated with future generations. Criticism of the utilitarian perspective. Utilitarian perspective versus a rights perspective: In the utilitarian perspective we can allow our self to harm others if our utility increases more than their utility is decreases. In the legal system a lot of things are illegal without any consideration of utility. We use a rights perspective. If you beat someone up, saying that you had more fun than what the victim suffered is probably no a viable argument in the courtroom. Even if you are hungry you are not allowed to steal bread. o we have the right to use cars and thus destroy agricultural land in Bangladesh just by arguing that our willingness to pay for cars are higher than their willingness to pay for bread? Should we use a rights perspective or a utilitarian perspective for environmental issues? Example 1: Air Pollution Control in Santiago. The most important source of pollution is emissions from busses and cars. The emissions from buses and cars can be reduced at a cost, which will affect the travelers. If buses has to reduce emissions they need to increase prices of tickets, this will merely harm the poor that go by bus. The rich may have a higher willingness to pay for an improved air quality. Economists tend to measure the revenue from improved environment as aggregated willingness to pay. The rich may have a higher willingness to pay for increased air quality then the cost of the poor for increased price of bus tickets. A cost benefit analysis will probably say that it is a good thing to reduce emissions from buses. But the poor will probably win a majority vote. 7
What decision is supported by different decision making procedures? With the Pareto criterion probably no change is possible. Someone would dislike stricter environmental standards, some other will dislike weaker regulation. The existing regulation will prevail, a bias toward status quo. With compensation to those that dislike a change a change may be possible, but there is huge risk of freeriding. If those that say that they dislike the change get some money maybe a lot of people say that they do not like it. With aldor icks criteria, the potential Pareto criterion we may get a significant reduction of emissions. (The cost benefit analysis) With majority voting the poor may vote against stricter environmental standards, (and thus more expensive bus tickets) But the outcome not only depend on what decision making procedure we use, individual preferences will also matter. With strong preferences for clean air stricter environmental policy will more likely pass the majority voting and will more likely give a positive net benefit in the cost benefit analysis.