Seminar 6

Welfare, efficiency and fairness

Readings

  • Kaplow, L. and Shavell, S. (2002). Fairness versus Welfare. Harvard University Press, Cambridge, Mass, ch 1 and 2
  • Coleman, J. L. (2003). The grounds of welfare. The Yale Law Journal.

7 comments on “Seminar 6

  1. giorgiomonti says:

    you might also want to have a look at these two reactions to the Shavell and Kaplow appraoch:

    Farber ‘What (If Anything) Can Economics Say about Equity?'(2003) 101(6) Michigan Law Review 1791
    Hadfield Feminism, Fairness, and Welfare: An Invitation to Feminist Law and Economics (2005) Annual Review of Law and Social Sciences 285 (you can find a copy here: http://law.bepress.com/usclwps-lewps/art40)

  2. Tim says:

    My feeling about the Kaplow and Shavell book chapters is that they approach law-making from a paternalistic angle. Whether or not a law is satisfactory depends on the welfare enhancing features of it. Laws should then be measured against their capability to enhance the well-being of those who are subject to the law. What those who are subject to the law think about fairness does not matter, as taking this into account is possibly welfare-diminishing.
    However, apart from an argument drawn from democratic principles, it seems that a legal order in which law-making should be based upon the well-being of all, seems to be open for manipulation. After all, it seems to me that there might be a problem of accountability – one can have an idea how the notion “well-being” should be defined, but others may disagree. It follows that the benchmark to which law is tested can vary but shall never be incorrect (unless manifestly beyond all reason). I have the feeling that fairness is a more objective benchmark – even though people may disagree. But I have the feeling that it is easier to come to a common understanding of what is fair, and that this is more easy to monitor by third parties (e.g. by courts).

  3. Irina says:

    While Kaplow & Shavell’s (KS) claim that welfare economics should be the normative approach to the evaluation of legal rules, Coleman argues critically that their argument is flawed.

    For example, Coleman criticizes that KS take as their starting point a tautology with regard to the definition of the notion of fairness, which KS do acknowledge. In fact, I agree with the criticism, as I found that KS derive their main argument from an extremely narrow definition of fairness and find themselves in trouble of always having to explain and clarify their argument with regard to their particular notion of fairness. While this might be a methodologically valid approach, it does weaken their normative argument. However, I do not agree with Coleman’s statement that the tautology used is not informative. I found the distinction between welfare and fairness, however sorted, eye-opening, given the unreflected use of both those terms in every-day politics and law-making.

    I cannot subscribe to the other criticism raised by Coleman. I do not share his contention that KS diminish the use of the concept of fairness when assessing legal rules as such. All KS are claiming is that “fairness” should not be used as an evaluative principle when choosing legal rules. So, their point refers to the regulatory law-making choices, that is when adopting laws. They, in my point of view, convincingly argue that the contexts in which to apply fairness and welfare economics as parameters of evaluation are different: while fairness refers to social norms that prescribe or condemn certain behavior, legal rules are used precisely for cases in which those social regulations cannot control some undesirable behavior. So, for KS – admittedly based on their narrow definition of fairness – fairness is used in a more inter-social context, whereas welfare economics are supposed to be the normative guideline for regulators. This means that that the two parameters are simply not used to assess the same situation. Here, I think, Coleman treats KS’ argument for welfare considerations narrower than it is.

  4. Argyri says:

    – What is an individual’s well-being?
    Although I find the overall argument of welfare versus fairness appealing, it seems to me that in order to accept the theory presented by Kaplow and Shavell one has to agree that there is one, unquestionable, or rather objective measure of an individual’s well-being. Which reminds me of the Socratic or Platonic idea of one good which at the same time seems to be precisely what the authors do not want or criticize (if I understand it correctly they are positioning themselves against the deontological theories, as also explained by Coleman).
    Against these theories however, there is the school of the moral relativists who believe that there is no objective truth or good. Likewise one could argue that there is no objective individual (or collective) well-being in the sense that welfare economics of Kaplow and Shavell try to achieve.

    They define the notion of well-being as incorporating in a positive way everything that an individual might value and an individual’s well-being as reflecting in a negative way harms to his or her person and property, costs and inconveniences, and anything else that the individual might find distasteful. Still this definition doesn’t give solution to the problem of aggregation. They do accept this on p. 26, but it seems that they do find it imperative to complete the definition proposing a measure of aggregation. This is, I think, a crucial gap that places them in the end in the same position to the deontological theories they might oppose to. I think that well-being might be clearer in situations such as road traffic (absence of accidents) but less clear in many other paradigms that I didn’t see touched upon in these two chapters of the book. And if well-being is left calculated by each individual without a serious proposal regarding aggregation, then what could possibly point to a policy? Which brings me to my second question of who defines well-being:

    – Who defines it?
    The authors, trying to explain what they mean by well-being (that an individual is better off), note: “We note, however, that our assumption that well-being is unambiguous is one of convenience; if individuals do not understand how situations affect their well- being, our argument may be applied to individuals’ actual well-being—what they would prefer if they correctly understood how they would be affected— rather than to individuals’ well-being as reflected in their mistaken preferences.” (p. 23). However, if that is the case I really do not see much difference between their theory and that of Plato’s who wants the Philosopher-Kings knowing what is best for the people because they are wiser. In this sense I agree with Tim’s point that they are in essence being quite paternalistic.

    Furthermore, I found the analysis of social norms (pp. 62 et. seq) that they made was too simple and that they would need to dig in more into philosophical or moral roots of the notion of fairness, or better the notion of justice.

    Nevertheless, I think they did make several fair points in their overall analysis. I found interesting that they calculate the welfare economic importance of fairness as depending on what individuals’ tastes happen to be, seen from an empirical point of view (pp.21-22). Also I agree with their criticism that under notions of fairness, legal rules would still be evaluated (ex-post) based on factors that are independent of individuals’ well-being (p. 51) – whatever this well-being might mean. Furthermore, I found their critique to notions of fairness interesting (pp. 45 et seq.), especially to nonconsequentialist notions of fairness. Last, their thoughts on distribution of income I think are also, to me, noteworthy.

    Overall, being myself critical to Kaplow and Shavell’s argument as I read it, I then found the criticism to this book from Coleman quite an enjoyable read, especially as he points out strongly to the big gaps in their argument, or rather the lack of argumentative support of their bold normative claim.

  5. Irina Domurath says:

    While Kaplow & Shavell’s (KS) claim that welfare economics should be the normative approach to the evaluation of legal rules, Coleman argues critically that their argument is flawed.

    For example, Coleman criticizes that KS take as their starting point a tautology with regard to the definition of the notion of fairness, which KS do acknowledge. In fact, I agree with the criticism, as I found that KS derive their main argument from an extremely narrow definition of fairness and find themselves in trouble of always having to explain and clarify their argument with regard to their particular notion of fairness. While this might be a methodologically valid approach, it does weaken their normative argument. However, I do not agree with Coleman’s statement that the tautology used is not informative. I found the distinction between welfare and fairness, however sorted, eye-opening, given the unreflected use of both those terms in every-day politics and law-making.

    I cannot subscribe to the other criticism raised by Coleman. I do not share his contention that KS diminish the use of the concept of fairness when assessing legal rules as such. All KS are claiming is that “fairness” should not be used as an evaluative principle when choosing legal rules. So, their point refers to the regulatory law-making choices, that is when adopting laws. They, in my point of view, convincingly argue that the contexts in which to apply fairness and welfare economics as parameters of evaluation are different: while fairness refers to social norms that prescribe or condemn certain behavior, legal rules are used precisely for cases in which those social regulations cannot control some undesirable behavior. So, for KS – admittedly based on their narrow definition of fairness – fairness is used in a more inter-social context, whereas welfare economics are supposed to be the normative guideline for regulators. This means that that the two parameters are simply not used to assess the same situation. Here, I think, Coleman treats KS’ argument for welfare considerations narrower than it is.

  6. Haukur says:

    I started by reading the book chapters and soon started to doubt where they were going with the argument. For the first they seemed to define their welfare standard as including the concept of fairness, but excluding few instances of fairness that were to be considered as intuitive mistakes. My immediate thought was that perhaps their calculation of welfare did not include some important elements that intuition might point towards through the sense of fairness. Why should they be right in what is best for humans and evolution wrong? I would, at least in a long race, put my money on evolution.

    This doubt was confirmed when they made the utilitarian argument that the pursuit of fairness could reduce everyones welfare and is accordingly the wrong standard to pursue. This argument, just like any other utilitarian argument, neglects the fact that in order for there to be any efficiency payoffs to distribute, a unity of cooperation must be maintained. Thus a egalitarian distribution of payoffs is better then no payoffs at all which will result from disunity. If utilitarian distribution leads to disunity the hypothetical welfare enhancement is never realisable in real terms and thus the second best option of fair distribution that maintains unity is preferable. This is an insight evolution seems to have right, but the Professors left out in their model. The argument is not any different whether it is about distribution of efficiency gain or loss. Disunity is always the nuclear option which justifies choosing fair solution over utilitarian solution.

    Coleman’s article tears the argument of the book to pieces. Not much to say about it. I mostly agree with his criticism.

  7. Marios says:

    Here comes my belated reply:

    What I find most problematic with K&S’s approach is their agnostic stance to the method of aggregation (eg. Justice a la Rawls, utilitarians, preference satisfaction etc). I understand that they see the two issues as separated (and that it suits their argument), but at the same time I think they are taking the easy way out. They are basically saying that we should measure well-being, but not telling us anything about how we should go about doing it. If they think fairness is relevant to the extent it relates to individuals’ well-being, isn’t that already a normative argument about how the aggregation and distribution should be done? If this had been an argument about religion the equivalent would have been to say that they are sure there is a (one) God, but they don’t really know which one is the true one. Well, to a person wanting to follow a God (rather than being interested in the phenomenon of religion) it will be of no avail to learn that there is one, if they cannot be told which one is the right one.

    Time and again their argument seems to be based on narrow definitions and ideal circumstances. It reminds me of studying Physics at school and being constantly frustrated by the fact that you had to assume all these things for any of the formulae to work (no friction, no gravity, etc). So I would have liked to have seen more elaboration on what is “fairness”, whose “well-being” is counted (only rational individuals, self-interested, benevolent, do their values matter, what about animals (eg. Singer), incapacitated, the dead, should their welfare count?).

    Finally, I would have liked to have seen some discussion of incommensurability and perfectionism (Raz), but maybe they take this up later on in the book.

    I haven’t managed to read the critique by Coleman, but judging by the others’ reactions I should probably do that.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s