Home > Uncategorized > An early rebuttal of the ‘representative consumer’ – and what it teaches us

An early rebuttal of the ‘representative consumer’ – and what it teaches us

Around 1900 John Bates Clark introduced the mythical ‘representative consumer’, the idea that you can model the sector households as if it is one person, as well as the idea that the ‘social utility’ (his phrase) experienced by this entity is the ultimate standard of social welfare (emphasis added):

>”If each man could measure the usefulness of an article by the effort that it costs him to get it, and if he could attain a fixed unit of effort, he could state the utility of a number of different articles in a sum total. Similarly, if all society acts in reality as one man, it makes such measurements of all commodities, and the trouble arising from the fact that there are many measurers disappear. A market secures this result, for society acts as a unit—like an individual buyer (chapter XXIV.14).

Interestingly, the economist Charles E. Persons, citing the last sentence of the quote above and explicitly attacking Clark, rebutted this view of the world already in 1913, among other things using data on inequality in the UK which in all probability are, recently, also used by Piketty (didn’t check this, though):

“The ultimate standard of value, then, for modern society, does not exist as a positive measure. That it does not is due to the presence of a large degree of inequality. In such a society, either the utility standard or the disutility standard must include incommensurable quantities, or (perhaps better stated) qualities. The problem is insoluble …. One cannot equate and unify either the pains or the pleasures of rich, well-to-do, and poor. We cannot find a positive measure of value in a society with such classes. The ultimate word declares only that with a given concentration of wealth, the society discounts the pains of the poor in a certain degree. Likewise in such a society there is a corresponding over-estimate of the sacrifices of the rich. Again, in such society the utilities enjoyed by the various classes are measured by various standards. Great pleasures for the poor count little; slight pleasures for the rich count much. We must add to the formula: “value depends on scarcity and utility,” the statement “each of these is conditioned by the existence of more or less of inequality.”

At first sight, this sounds depressing: a neoclassical economist invoking a mythical entity and a critic who points out the obvious flaws and inconsistencies – one hundred years ago. We seem to be running around in circles.

At second sight, however, we did move on. Clark as well as Persons was searching for an ultimate standard of value. Neoclassical economists did since not really progress beyond the ideas of Clark and still assume that the sector households acts as an individual buyer and still assume social welfare (the Euler equations and Samuelsonian shorthand they nowadays use do not make a fundamental difference, in my view). The critics, however, developed a whole array of methods to conceptualize and measure social welfare. Look here for 41 ‘headline well being’ metrics for the UK which, however, still exclude estimates of inequality (there is a poverty metric which might be used as a very crude inequality index). But look here for Eurostat data, published today, on ‘quality of life’ indicators which do contain data on inequality. In the end, ‘social welfare’ turns out to be a multi-dimensional thing which is not captured by relative market prices and which is difficult to optimize. Persons was right: Clark was wrong.

Thanks to Marko for providing links.

  1. originalsandwichman
    March 21, 2014 at 5:08 pm

    Very nice. See also the appreciation of Persons’s argument by J. Mayer in “Pseudo-scientific method in economics,” Econometrica: Journal of the Econometric Society, 1933 and “Pseudo-Scientific Economic Doctrine–Continued,” Philosophy of Science, 1936.

    • originalsandwichman
      March 21, 2014 at 5:13 pm

      J. Mayer:

      “Faulty generalizations about early society, together with the ‘alter- native’ sophistry, also helped materially in building up the structure of subjective utility analysis, the most important form of which concerned itself with the supposed homogeneity and comparability of subjective utilities and disutilities, which modern psychology and C. E. Persons’ penetrating analysis appear to have finally made ready for the scrap heap, even though some economists here and there still drag out the rusty pieces and give them a new polish. The pseudo-scientific character of these concepts may be briefly summarized as follows: They assume a general rationality in human behavior which does not exist except in an exceedingly narrow sphere; they posit a refined system of barter which is quite out of harmony with the facts of the market; they substitute artificial abstractions (utility and disutility) for the actual pecuniary logic (profits and losses) which is at the bot- tom of a money economy; they continue to be based upon a purely hedonistic view of human nature, long since exploded, merely substituting utility-disutility for pleasure-pain with similar connotations and applications; their analogies (statics, normality, and the organismic concept) are quite at variance with reality, the facts of evolution and institutional change, and modern researches in anthropology and sociology.”

      • originalsandwichman
        March 21, 2014 at 5:38 pm

        nb: The “‘alternative’ sophistry” Mayer refers to is usually referred to currently as opportunity cost.

  2. March 21, 2014 at 5:16 pm

    Merijn –

    Thanks for the [undeserved] link credit, but I think it was Marko who actually provided what you cited.

    As an aside:
    As to your Q (elsewhere) in re: whether Persons is an unrecognized Veblen or just another economic scribbler…

    I suspect he ran in some of the same circles as not only Veblen, but also John R. Commons, Richard T. Ely, Edwin E. Witte et al, and was heavily influenced by [among others] Henry Dunning Macleod (whose credit theory of money in turn directly inspired Alfred Mitchell-Inness’ chartalist views – e.g., see “What Is Money?’, Banking Law Journal 1913 – which Keynes himself reviewed approvingly).

    I find Persons’ case fascinating/interesting, and am only now beginning to do some research… if/when I can compile enough a clear enough picture, I may well take you up on the idea of posting a Wikipedia entry on him.

    • merijnknibbe
      March 21, 2014 at 7:48 pm

      Oops, link credit changed… and at least for me this little journey of discovery is already quite enlightening!

  3. March 25, 2014 at 11:39 am

    I think i have to disagree (which, as a heretic, is my general-equilibrium determined utility in this representative consumer society (whose national anthem is the stirring “Give Me conveniance, or Give Me Death” (dead kennedy’s, 80’s band)). There are many maps from an n-dimensional system (eg n people) to a one dimensional one, or from n heterogeneous individuals to n identical, and/or indistinguishable ones. These are done for mathematical conveniance meaning not as bad approximations —- people choosing to write down equations they can solve, rather than equations which represent the problem, but as a better notation. (Eg in relativity theory people often use ‘moving rest frames’ and then convert back to stationary ones to make the math eaiser; and they often set all constantrs equal to 1 and later on plug in real values so you don’t end up with an apparent word salad). (Such transofromations are not , of course, always possible without approximations; see for example inverse problems, doob on potential theory, or things like splines, WAT, or O&W.).

    However there is Absolutely no reason why a ‘representative consumer’ cannot be constructed (represented by a utility function’) who is sensitive to, amnd adverse to (or in favor of) inequality. Its no different than saying individuals don’t care about relative amopunts of food, air, and water, or salt, sugar, and protein they consume

    A swociety viewed as a single superorganism, or input output device is as easily cponceivable as a single object as a combination of atoms or chemicals or different people. (which is to say, not really tractable, or easily). A utility or soicial welfare function which is n-dimensional is to me not theoretically all that removed from one which is one dimensional.

    (But i guess alot of people, possibly those without much math interest, might seem to think ‘all math can deal with is straight lines, as shown by Descartes, as we leasrned in school and college.. . But I propose a unique , all brand new heterodox radical paradigm shift, in which we start including some power and product terms in our equations, i.e. introducing nonlinearities. . I can forsee entire academic departments being routed as those who cpould only write y=ax + b are replaced by new experts highly trained in the emerging technical fields who can write y=ax**2+xy +b, and theorize that if they get an NSF grant they can find that the equation y=x***3 exists, and also that if we let n’=3+1, n”=n’+1 etc then we can show functions y(n(‘))=x**n(‘) exist for all n(‘) if and only if NSF grant (n(‘)) eixsts.. It costs money to get the benefits from research. Their PhD students can then go on to discover the quadratic formula, galois theory, invest in technology for numerical research such as calculators, mathematica, etc.)

    If your representative input output system is malfunctioninjg, some of its parts are broken. How can you tell if it is malfunctioning, overheating,etc.? One could give it a blog, and then see how its rankings compare to other blogs, or give it an SAT or personallity test, look up its profile in the DSM to see if it has add or is on pcp, deserves the 3rd degree or should take the 5th, or drink one with some coke. One could ask god WWJD (what would jesus do) or Dawkins ‘what is the optimal, most fit strategy of of the genes for this particular input output device’? Does it matter if the rhino goes extinct, or if they are so plentiful we have no spare bedrooms to put them up in?

    Then one can ask ‘is there some space’ which makes the ,metric in some sense ‘natural’. Eg, you can write general relative theory approximately as either a system in either curved or flat space time, so that matter is either externalized or internalized. Most think (using occam’s razor) the curved version is more elegant. (Similarily quantum theory can be written as a set of coupled classical equations, but most people see no use in that representation and use the standard quantum (complex valued) equations. Thyere are these speculators into ‘transhumanism’ (eg some former GM manager type and hedge fund manager in one of those tiny mideast countries) who think humans and their society are simnply transitional to some higher form (possiby as envisioned by Vonnegut in sirens of titan). However,it can be conjectured that he’s a deluded lunatic.

    (Also, one can posit that actually all metrics actually can be more or less turned into each other, as white light can be resolved into a spectrum, which can be resolved further. Or a seemingly uniform flow of colored liquid can be resolved into chaotically mixing indivudal colors by considering its full set of poincare slices. The whole being the sum of the parts, and the whole being the parts of the parts.
    ‘a wo/man on a deserted island disocvered a track of footprints on the beach and followed them, only to discover they were his own’—–a eddingtion)

    • davetaylor1
      March 25, 2014 at 10:42 pm

      I’m not quite sure what you are disagreeing with, ishi, but I very much enjoyed your rant. Perhaps a little more attention to capitals and typos? But seriously, there is food for thought in both spectrum analysis being into a time-power series of two-dimensional (i.e orthgonal sine and cosine) waves and white light being reproduced on a tv by just three dots of fixed colour. Maybe that hidden agenda amid all that variety of economic interests can be modelled by a PID servo?

  1. No trackbacks yet.

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 )

Google photo

You are commenting using your Google 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 )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.