Discussion questions on Hausman and McPherson, Robbins, and Kaldor on Normative Economics

1. What are preferences? To what extent are they defined by the axioms governing preferences?

2. What does it mean to claim that rationality requires that preferences be complete and transitive? Is it true? How does the claim about rationality link up with claims about the the features of people's actual preferences? What, if anything, do the claims about rationality and actuality have to do with claims about morality?

3. What, if anything, are the connections between self-interest the model of rationality that economists adhere to?

4. Robbins maintains that there is no possible evidence the bears on interpersonal comparisons of utility (preference satisfaction). Is he right? Can you think of any evidence at all that might give one reason to think that giving something to person P makes a larger different in the extent to which P's preferences are satisfied than giving something to Q?

5. Robbins maintains that not only that there is no evidence that bears on interpersonal utility (welfare) comparison, but in addition he maintains that interpersonal welfare comparisons are normative -- expressions of value judgments -- rather than descriptive. What are the relations between the two theses (a) that no evidence bears on interpersonal utility comparisons and (b) that interpersonal utility comparisons are evaluative rather than descriptive? Why does Robbins think that interpersonal utility comparisons are normative or evaluative rather than descriptive? Is he right to think so?

6. Robbins maintains that the field public finance (as it was practiced in the 1930s) assumes that if individuals are in similar circumstances, then the extent to which their preferences are satisfied is similar. That assumption, he maintains, is a normative claim. So what follows from it is not part of positive economics. So what? How does Robbins' thesis bear on the possibility and character of normative economics?

7. Kaldor agrees with Robbins concerning the scientific status of interpersonal comparisons. Yet he draws very different conclusions. Why?

8. Harrod and Robbins argue that free trade in grains increases total well-being only if the increase in well-being in the rest of society is larger than the decrease in well-being of landlords, and such a conclusion requires interpersonal comparison of utility, which, Robbins maintains, is impossible or an expression of values. Kaldor disagrees. He maintains that economists can show that free trade in grains is advantageous without making any interpersonal utility comparisons. How is this possible?

9. An economic state of affairs S is called "a potential Pareto improvement" over another economic state of affairs R if and only if it is possible (in principle) to carry out a redistribution in S that results in a state of affairs S' that is an (actual) Pareto improvement over R. (S' is an actual Pareto improvement over R if and only if at least one person prefers S' to R, and nobody prefers R to S'.) Why should we care whether S is or is not a potential Pareto improvement over R? How does the notion of a potential Pareto improvement link up to Kaldor's ideas?