The Philosophical Foundations of Mainstream Normative Economics

1. The Philosophical Foundations of Mainstream Normative Economics Daniel Hausman and Michael McPherson

2. An Example of Normative Economics at Work Lawrence Summers, chief of the World Bank in 1991: Just between you and me, shouldn’t the World Bank be encouraging more migration of the dirty industries to the LDC’s? I can think of three reasons: • The measurement of the costs of health-impairing pollution depends on the foregone earnings from increased morbidity and mortality. From this point of view a given amount of health-impairing pollution should be done in the country with the lowest cost, which will be the country with the lowest wages. • I’ve always thought that under populated countries in Africa are vastly under polluted; their air quality is probably vastly inefficiently high compared to Los Angeles or Mexico City. Only the lamentable facts that so much pollution is generated by non-tradable industries and that the unit transport costs of solid waste are so high prevent world-welfare enhancing trade in air pollution and waste. • The demand for a clean environment for aesthetic and health reasons is likely to have a very high income-elasticity. The concern over an agent that causes a one in a million change in the odds of prostate cancer is obviously going to be much higher in a country where people survive to get prostate cancer than in a country where under-5 mortality is 200 per thousand…The problem with the arguments against all of these proposals for more pollution in LDCs (intrinsic rights to certain goods, moral reasons, social concerns, lack of adequate markets, etc.) is that they could be turned around and used more or less effectively against every Bank proposal for liberalization. (quoted in The Economist, February 8, 1992, p. 66).

3. Summers’s Argument • For some amount of compensation C that the least agents in LDC will accept and the most agents in rich countries will offer, all rational individuals, whether in developed countries or in LDCs, would prefer to transfer pollution from a developed country to a LDC. • Whatever well-informed and rational individuals preferis what makesthem better off. • So, exporting pollution to LDCs from developed countries and paying compensation makes everyone better off. • One should adopt policies that make people better off. • So, one should adopt policies that shift pollution to LDCs and pay compensation.

4. Objections to Summers’s Argument • Encouraging dirty industries to migrate to LDCs might lead to more total pollution. Developed countries have stronger incentives, greater administrative capacity, and more resources to enforce pollution controls than do LDCs. • Even if people in both developed economies and LDCs would prefer to shift pollution to LDCs in exchange for appropriate compensation, the exchange may be unfair. Developed countries are exploiting the poverty of LDCs – which , in addition, they are often responsible for. It may not be right to make people better off if doing so involves injustice. Justice matters, too. • Summers’ analysis compares only one possible alternative to the status quo: that of shifting pollution to LDCs. But there may be other policies that would be better still. Notice Summers’ analysis depends on huge income disparities between rich and poor countries. • Satisfying preferences does not automatically increase welfare. People may prefer things that are bad for them. • Do economic costs and benefits capture what is morally relevant? Do rational and well-informed people have to accept market evaluations of the consequences of pollution?

5. Three Kinds of Individualism • Ontological Individualism maintains that only mental states and physical objects, including human beings, are real. Cultures, social institutions, and so forth are not real. • Explanatory Individualism or Methodological Individualism maintains that the fundamental laws should concern preferences, beliefs, and the choices of individual human beings. The view concedes that social entities and facts have causal consequences, but it insists that those consequences are mediated by the beliefs, preferences, and choices of individuals. • Ethical Individualism is the view that social entities are of no intrinsic moral importance. There is a moral reason to protect a culture, a religion, a state, a tribe or a corporation if and only if doing so is required by moral concern for individual human beings.

6. Theory of Rationality I • Humans are rational. The core idea is that explanations of individual choices often also justify those choices. The factors that cause choices also function as reasons for choices. • Rationality lies in the structure or form of choice and preferences, not in the content of what is preferred or chosen. • Individuals choose or act rationally if their actions are determined by their preferences, and their preferences are themselves rational. • Preferences are rational if they are complete and transitive. Completeness: for any two options x and y either x is preferred to y or y is preferred to xor one is indifferent between x and y. An agents preferences are transitive whenever an agent prefers x to z when it prefers x to y and y to z.

7. Theory of Rationality II • The theory of rationality that mainstream economics assumes is a normative theory of how people should behave, but not necessarily a moral theory. The norms of rationality are insensitive to the pursuit of good over evil. • The theory of rationality assumes that people are materially self-interested, they prefer more of a commodity to less of it, and more wealth to less wealth. • The theory of rationality takes self-interest and rationality as important components of explanation, but it does not need to assume that the principles are exceptionless laws or generalizations, such as laws of physics. The laws can be treated as having exceptions or qualifications, and agents can be taken to be attempting to approximate what the laws specify.

8. The Problem of Interpersonal Comparisons • Suppose A ranks x above y, and B ranks y above x. Given that the numbers assigned to x and y for both individuals are only ordinal rankings, how can we determine what state of affairs is better? • In order to compare A and B we need magnitudes to determine the degreeto whichx is betterthan y. But ordinal rankings are ranks thatcannot be added. 1st + 2nd = 3rd does not make any sense. 1 + 2 = 3 makes sense, since the numbers are cardinal. • A moral theory such as Utilitarianism can be the foundation of economic theory only if we can make sense of interpersonal comparisons of utility.

9. Pareto Criteria I • Economist Vilfredo Pareto was an Italian economist working on the concept of efficiency in economics. • Given an initial distribution of goods D, a change to a distinct distribution of goods Dn is a Pareto improvement, only if in Dn at least one person is better off, and no one is worse off. D1 is a Pareto improvement over D, since A is made better off and B is not made worse off. D2 is not a Pareto improvement over D, since A is made worse off even though B is made better off. D3 is a Pareto improvement over D, since both A and B are made better off and no one is made worse off.

10. Pareto Criteria II • If somebody prefers x to y and nobody prefers y to x, then x is Pareto superior to y. • D2 is Pareto superior to all other Dn, since A would prefer D2-4 to D1 & 5, and B would prefer D2 to all other Dn. D3 &4 are not Pareto superior to D2, since although A prefers themtoD1 &5, B and C do not prefer them to D2.

11. Pareto Criteria III • xis Pareto optimal or efficient if and only if there is no alternative that is Pareto superior to x. • D1 is Pareto optimal with respect to alternatives D2 and D3, since neither D2 or D3 are Pareto superior to D1. D2 is not Pareto superior to D1, since neither A nor B nor C prefer D2 to D1. D3 is not Pareto superior to D1, since although A prefers D3 to D1, neither B nor C prefer D3 to D1.

12. Pareto Criteria IV • X’s being Pareto optimal does not imply that X isPareto superior to the alternatives, even to non-optimal alternatives. • Suppose there are 6 units of bread to distribute between A and B, and both A and B prefer more units of bread to less units of bread. Consider the following distributions: • D1-3 are all Pareto efficient, since none of them waste bread. D4 is Pareto inefficient, since in it some bread goes to waste. However, D5 is Pareto efficient, but not Pareto superior to D4. In D1-3 A and B get as much if not more than they get in D4, since in D4 is inefficient because some bread goes to waste. D4 shows that an inefficient state of affairs can be Pareto superior to a efficient state of affairs. And if we add that a minimum of 2 units is required to avoid starvation, then D4 is morally required over D5.

13. Kaldor-Hicks • How do we compare policies when no policy isPareto superior toany other? If neither A nor B is Pareto superior to the other, how do we judge which policy to pursue? • Kaldor-Hicks: X has a greater capacity to satisfy preferences than Y if and only if X is a potential Pareto improvement over Y. X is a potential Pareto improvement over Y if there is some (not necessarily feasible) way of redistributing the goods available in X that makes X an actual Pareto improvement over Y. • For example, 7 units to A and 5 units to B is a potential Pareto improvement over the distribution of4 units to A and 6 to B, because it would be possible to redistribute the 12 units so as to achieve an actual Pareto improvement.

14. Compensation and Potential Pareto Improvements If X is a potential Pareto improvement over Y, then it is possible in some sense for the winners in a change from Y to X to compensate the losers. Whether the winners could compensate the losers is then operationalized in terms of willingness to pay. If the amount that the winners would be willing to pay to bring about a policy is larger than the amount the losers would need to be compensated to accept the policy, then the policy is a potential Pareto improvement over the status quo, and the policy purportedly brings about a more efficient state of affairs in which there is a net benefit – a greater capacity to satisfy preferences.

15. Criticisms of Cost-Benefit Analysis • Cost-benefit analysis uses “willingness to pay” rather than gains and losses in welfare. • Cost-benefit analysis uses the Pareto criteria, which is not sensitive to justice. • Cost-benefit analysis involves uncertainty coupled with preferences and willingness to pay. When people have mistaken beliefs about the constitution of the exhaust from the factory down the road, their willingness to pay to avoid breathing it will be an unreliable indicator of their true preferences, let alone the welfare consequences of the exhaust. People often do not know the consequences of alternatives and hence which alternative they would prefer if the did know the consequences. The problems of uncertainty are usually finessed by supposing that individuals possess subjective probability distributions over all the possible outcomes, but to suppose this involves extreme idealization; and there is little justification for respecting preferences based on largely fictitious probability distributions.