The socialisation of economists (crosspost from Crooked Timber)

I’m following up Henry Farrell’s post on the superiority or otherwise of economists, and Krugman’s piece, also bouncing off Fourcade et al, with a few observations of my own, that don’t amount to anything systematic. My perspective is a bit unusual, at least for the profession as it exists today. I didn’t go to graduate school, and I started out in an Australian civil service job in the low-status[^1] field of agricultural economics.

So, I have long experience as an outsider to the US-dominated global profession. But, largely due to one big piece of good luck early on (as well as the obligatory hard work and general ability), I’ve done pretty well and am now, in most respects, an insider, at least in the Australian context.

I’ve also followed an approach, strongly deprecated in US academia, of writing lots of papers on many different topics. has exposed me to a lot of different subfields of economics, each with their own status hierarchy and criteria for what counts as good work.

My conclusion from this experience is that the obvious problems of academic macroeconomics aren’t unique to that field (in fact, as Krugman points out, the sharp divisions within the field make it unusually open in some ways). In many parts of economics, the status rankings and evaluation criteria are driven much more by internal and aesthetic factors than by empirical validity or policy relevance.

There’s a big arbitrary element to all this. If the seminal (sic) paper by the recognised smartest guy (sic) in the field formulated the core problem in a particular way, his (sic) grad students will learn and propagate that way of doing things, and will vigorously resist any alternative approach, even a purely technical reformulation. They have, after all, a lot invested in the idea that they have learned something valuable that no one else really understands.

If you want to see the mentality at first hand, and have a strong stomach, take a dive into the cesspool that is Economics Job Market Rumors. I should mention that my distaste for EJMR is fully reciprocated. The idea that an understanding of economics combined with attention to actual events is more useful than mastery of a highly specialised bag of tricks is totally antithetical to the ethos of the site (unsurprisingly, they hate Krugman even more than they hate me). In a less extreme form, that’s true of the econ profession in general, as the Fourcade et al piece shows.

[^1]: At least in the international profession, dominated by US norms and values. Until recently in Australia, a large proportion of policy-oriented economists got their start in this field, and policy work, in turn has a status it lacks in the US.

74 thoughts on “The socialisation of economists (crosspost from Crooked Timber)

  1. @Ikonoclast

    Please, please, please, do not use the banner of Nobel Laureate when it comes to economics (or peace for that matter).

    This is just a self-serving publicity stunt by the ruling class.

  2. @Ivor

    I wasn’t quoting those two Nobel Laureate prize winners in a laudatory fashion. In fact, I equated them to used car salesmen who sell rustbuckets with a new coat of paint and sawdust in the sump.

    My final quote was from another thinker who pointed out the completely unempirical position of the two said N L winners.

    You needed to read all I posted but I can see in retrospect how you could lose patience early and not see where I was headed.

  3. Newtonian @15 p 1

    The article by Fourcade, Ollion, Algan, “The Superiority of Economists” is published as a discussion paper by the Max Planck Sciences Po Center on Coping with Instability in Market Societies.

    After reading the article several times, I could not find even a tangential link to the purpose of the Center. I am happy to be educated in this regard.

    Consider the following premise in the said article: “First, the theory of action that comes with economists’ analytical style is hardly compatible with the basic premise of much of the human sciences, namely that social processes shape individual preferences (rather than the other way round). In economics, by contrast, “de gustibus non est disputandum” (Stigler/Becker 1977): the action begins mainly when preferences are set. [1]

    How is one to interact with the social scientists, represented by the authors of the said article, without having to learn a new language, which, to tease the mind unnecessarily, bears the same name as the old one, namely English, or, alternatively, pretend each individual’s only “action” is to influence others’ (undefined) preferences and to be influenced by others’ (undefined) preferences – 24 hrs a day, 365 days per year??

    Obviously, it would seem to me, the question how are preferences (over?) formed is a different question to how people act, given their preferences for alternative physical objects and state contingent future payoffs (a primaty research area in economics).

    One of their ‘findings’ is that economists are more insulated than any others in their sample. Interestingly, they ‘find’ that there is more interrelationship between Finance and other ‘disciplines’ than between Economics and other ‘disciplines’. Nothing makes sense to me. Firstly, how do we get from “economists” to “Economics”? Second, if the said authors would realise that Finance is a sub-field of Economics, then one of their conclusions evaporates.

    And so on.

    To the best of my knowledge, Max Planck is a big name in scientific research. I am not sure Max Planck would be pleased to have his name associated with the said article.

    The said paper relates to the U.S.A. [1] Lets look a little further, for example to the 2014 (Bank of Sweden) Nobel Prize winner in Economics, Professor Jean Tirole (France). I suggest the said author’s presumptions about ‘economists’ is inconsistent with the content of Prof Tirole’s CV, linked to below:
    http://idei.fr/doc/cv/tirole_en_2014.pdf

    [1] I do not accept that all academic economists in the USA or even the majority would fit into the said author’s model.

  4. But Ernestine, whether preferences are treated as endogenous, or not, is clearly potentially of great significance to welfare economics, without which economics loses much of its point.

  5. Wylie, presumably the potentially great significance of treating preferences (defined over?) as endogenous (instead of treating them as given at a point in time) depends on the philosophical foundation of ‘the economy’ to be considered.

  6. @Ernestine Gross
    How so? So long as you base policy choice on net welfare comparisons, and you define welfare gains at the individual level in terms of preference satisfaction (i.e. ‘better off’ means an enlargement of the feasible consumption set) – which is what economic theory does, after all – then the question of whether or not enactment of a given policy (i.e. the reallocation that it implies) will *alter* preferences is highly germane. Economic analyses tend to choose an arbitrary and restricted set of arguments for the objective function, and assume that the preferences that function represents are both given and independent of allocations. It is by no means clear that the Fundamental Theorems of Welfare Economics hold in general when preferences are endogenous. Theoretically speaking, that’s rather a big deal.

  7. @Wylie Bradford

    It is not clear to me whether you are addressing social welfare functions or the first and second fundamental welfare theorems or both at the same time.

    My point was that at a theoretical level, economics is not independent of the philosophical base of ‘the economy’. For example, if you wish to have a theoretical model of a non-dictatorial society, then taking preferences as given is appropriate. Taking preferences as given does not mean individuals’ preferences are constant over time and it does not mean that the time profile of preference relations of each individual is not influenced by say new information, or at least partially interdependent. It just means the theoretician is not imposing prior restrictions on how these preference relations are formed on the individual level. Changing the set of feasible consumptions, say through a policy measure such as income redistribution, may well change the observable choices but not the preferences. Enlarging the set of feasible consumptions (technologically possible and budget feasible) may or may not have any effect on actual choices of any one or all individuals (eg if preferenes relations have a satiation point in the interior of the enlarged feasible consumption set(s).)

    In this context, I don’t understand your statement: “Economic analyses tend to choose an arbitrary and restricted set of arguments for the objective function, and assume that the preferences that function represents are both given and independent of allocations.”

    Now, suppose you wish to have a theoretical model of an economy where sociologists specify what preferences are (on, say the basis of consumer surveys, or whatever). (This would be a society where some people are allowed to impose their methodology of reaching conclusions on others). Call this endogneous (it is treated as part of the social norm). Obviously, the criterion for what consititutes a ‘welfare improvement’ would be different.

    Recently I’ve come across some work by neuro-scientists who study risk taking behaviour (an aspect of preferences) of individuals over their life span. Suppose, and this is really speculating as to what might come out of this, the output of this research program is then treated as endogenous to the society of interest. New theoretical research results might emerge (or not, given there very general notion of preferences). Would you not agree that following such a program presumes a society that accepts scientific methods on the philosophical level?

  8. @Ernestine Gross

    Ernestine,

    First of all I’m an economist, so the snide references to sociologists are otiose in my case 🙂

    Second, I rarely waste time talking about social welfare functions. After all, as economists we know that there has been a fairly robust argument to the effect that they are invalid constructions for 64 years now. Too bad that comparisons in those terms are at the heart of the theoretical basis of cost-benefit analysis, but, as we know, economists have a high tolerance level for incoherence.

    The link you attempt to draw between the assumption of given preferences and a non-dictatorial stance on the part of the theoretician is simply spurious. The assumption of given preferences, fixed with regard to the problem under consideration, just makes things easier. A change increases an agent’s welfare if it moves him/her to a more preferred combination. That is the definition of being ‘better off’. The comparison between the before and after in this case must obviously be made with regard to one set of preferences. If the preferences were to change in response to the alteration in the agent’s circumstances, then clearly the judgement as to whether welfare had improved would be greatly complicated. Hence my reference to the Fundamental Theorems. In the case of the Second, preferences must be independent of the lump-sum transfers if we are to be sure that the result will hold. Note I don’t say that it definitely won’t hold if preferences are endogenous (the correct interpretation of that term being that they change in response to other variables in the problem) but that we can’t be sure that it will. That’s a fair summary of much of the literature on preference endogeneity IMO. That’s a messy result, raising the possibility of economists not being able to evaluate and compare outcomes. Hence, they respond by maintaining an assumption of preference exogeneity in order to yield clear results, despite the fairly obvious fact that preferences of actual agents are not exogenous in this way. You can call that being scientific if you like, but last time I checked the scientific method wasn’t based on expediency.

    As for the question of arguments in objective functions, economists routinely assume, for example, that preferences are defined over commodities only, and not at all over the organisation of work. Once again, this is expediency. If agents are treated as caring how things are produced as well as production is distributed in allocations then efficient allocations cannot be defined in isolation from the organisation of the labour process. Messy. Assuming away preferences over work means that consumption and production can be treated independently and clear results can be obtained. Of course, people do care very much about the organisation of work, so the restriction on utility functions is both arbitrary and false. And, yes, I am aware of the Arrow & Hahn type dodge of having agents demand occupation-specific leisure hours e.g baker-leisure, butcher-leisure etc. Since ‘occupations’ are not defined in the theory that is just another case of ordinary-language sleight of hand being used to deal with theoretical limitations. It does nothing to address the fundamental point.

  9. @Ernestine Gross
    Thanks for the comments.

    Apologies if my comments are to some extent stream of consciousness and I seem grumpy at economics/finance. 2008 has prejudiced me I think but I am still trying to put myself in the alternative shoes.

    On the matter of whether economics is a science of not I just stumbled on the following which may be of interest that relates to Bayes Nets where economists and scientists may find common ground.:

    CARTWRIGHT’S HUNTING CAUSES AND USING THEM CAUSAL PLURALISM AND THE LIMITS OF CAUSAL ANALYSIS$
    Kevin D. Hoover
    Review essay on Cartwright, N. (2007), Hunting Causes and Using Them:
    Approaches in Philosophy and Economics. Cambridge, UK: Cambridge
    University Press. xþ270pp. IBSN 978-0-521-86081-9 $85.

    I dont have Cartrights book but I do have some of her other papers and this review is consistent with the picture that emerges. It is fascinating here to see the subtleties here.

    Reading between the lines she is very respected philosopher of science and gives economists more credit that scientists might so is likely worth reading. But she also hates Bayes net related cause and effect development (I was looking for examples running counter to the Pearl narrative which seems to be better suited to hard science and engineering and medicine (human engineering?) where its adoption is going gangbusters.

  10. @Wylie Bradford

    Wylie,
    I assumed you are an economist currently at the MQ.

    Who made what snide reference to sociologist?

    “The link you attempt to draw between the assumption of given preferences and a non-dictatorial stance on the part of the theoretician is simply spurious.”

    Why? Please note, I did not talk about “the assumption of given preferences”. Furthermore, I didn’t talk about a non-dictatorial “stance on part of the theoretician”.

    “A change increases an agent’s welfare if it moves him/her to a more preferred combination.”
    A change in what? Combination of what?

    Your explanation as to why you refer to the 2 theorems, known as fundamental welfare theorems makes no sense to me. Would you please provide a reference of your source literature for the 2 fundamental welfare theorems as well as for ‘endogenous preferences’.

    No brownie points for stating that the theoretical models for which the 2 fundamental welfare theorems are characterisations of the solution exclude ‘the organisation of work’ or that they do not include the term ‘occupation’. Firstly, Debreu’s 1959 ‘Theory of Value’ contains the 1st and 2nd fundamental welfare theorems. Debreu explicitly states that the theoretical model in this book excludes ‘industrial organisation’ (and it does not deal with the formation of prices). Second, I’d be very careful if I were you in criticising Arrow and Hahn for not having used the term ‘occupation’, you may end up being surprised as to how general the concept of a commodity is in the said models. Lately, we don’t hear much about ‘occupation’. Instead we hear about ‘jobs’. (The everyday language does not always correspond to concepts defined by suitable choice of mathematical objects.)

    I am surprised you are not mentioning the difficulty in getting people to reveal their preferences (see mechanism design).

    Strange, isn’t it, that the generic Pareto inefficiency result for competitive private ownership economies with incomplete markets is not (as yet) labelled the 3rd fundamental welfare theorem.

    To return to my point. Why is there a need to restrict oneself to a philosophical beliefs underlying theoretical models of a competitive private ownership economy? Isn’t it the case that by you ignoring my sentence of considering an society where people classified as sociologist provide input into ‘how the economy should work’, the link to the topic of this thread is lost, IMHO.

  11. @Newtownian

    Thank you for the reference.

    I don’t know how large the subset of all economists is who are frustrated with the institutional changes that resulted in the 2008 event, but I am pretty sure it is a non-trivial set. Incidentally, the joint work I most enjoyed was with an acoustics engineer and someone with an IT background. Communication was easy.

  12. Ernestine Gross :
    @Wylie Bradford
    Wylie,
    I assumed you are an economist currently at the MQ.

    That would be me 🙂

    Who made what snide reference to sociologist?

    I guess I took the reference to ” a theoretical model of an economy where sociologists specify what preferences are” as snide, following on from your snippy attitude to the Planck Institute paper.

    “The link you attempt to draw between the assumption of given preferences and a non-dictatorial stance on the part of the theoretician is simply spurious.”
    Why? Please note, I did not talk about “the assumption of given preferences”. Furthermore, I didn’t talk about a non-dictatorial “stance on part of the theoretician”.

    Leaving aside the fairly pointless diversionary squibbing about the precise words you used, you stated: “For example, if you wish to have a theoretical model of a non-dictatorial society, then taking preferences as given is appropriate”, and “the theoretician is not imposing prior restrictions on how these preference relations are formed on the individual level”. Furthermore, you characterised your ‘sociologists’ example as ” a society where some people are allowed to impose their methodology of reaching conclusions on others”. Putting all that together paints a pretty clear picture of a link being drawn between taking preferences as given and being (piously) non-dictatorial. As I said, a spurious link. An assumption that preferences are exogenous can be modified by treating them as a function of (or at least affected by) the other variables in the problem at hand. The alternative to assuming given preferences is not imposing a preference ordering from outside (i.e. abandoning a subjective theory of value). This is what you seem to be fixating on. Of course, economists do impose a lot from the outside. They have no idea what the preferences of actual agents are, and yet they assume functional forms for utility functions that imply properties for those underlying orderings. Naturally they do this, as with everything else, for tractability purposes.

    “A change increases an agent’s welfare if it moves him/her to a more preferred combination.”
    A change in what? Combination of what?

    I don’t know whether this is a stylistic trope of yours but I find it hard to believe that you don’t understand precisely what I mean. Imagine an agent with a utility function defined over whatever – a subset of commodities, all commodities, all commodities plus sundry other things – and the size of the feasible consumption set changes for some reason (e.g. a change in some prices). The agent is better off if after the change they are consuming a combination of whatever it is their utility function is defined over with a higher utility number than before the change. That is the definition of being ‘better off’. Clearly that requires comparison of combinations using the same utility function. Utility numbers from different functions being non-comparable, if then preferences change in response to whatever caused the change in the size of the consumption set, that particular definition of ‘better off’ is no longer applicable. That is why the issue of pereference endogeneity is important.

    Your explanation as to why you refer to the 2 theorems, known as fundamental welfare theorems makes no sense to me. Would you please provide a reference of your source literature for the 2 fundamental welfare theorems as well as for ‘endogenous preferences’.

    It follows pretty straightforwardly from the above. Try thinking about it using the way we teach this stuff to undergrads (e.g. 2 agents, 2 commodities, Edgeworth Box). If the indifference curves of the agents went wandering all over the place any time the endowment point was shifted it would complicate the demonstration of the point that any desired Pareto-efficient allocation can be supported as a price equilibrium (or quasi-eq) by making the appropriate transfers.

    No brownie points for stating that the theoretical models for which the 2 fundamental welfare theorems are characterisations of the solution exclude ‘the organisation of work’ or that they do not include the term ‘occupation’. Firstly, Debreu’s 1959 ‘Theory of Value’ contains the 1st and 2nd fundamental welfare theorems. Debreu explicitly states that the theoretical model in this book excludes ‘industrial organisation’ (and it does not deal with the formation of prices).

    Ah, the Debreu defence – ‘You can’t criticise Debreu for ignoring X – he explicitly leaves X out!’ It is a key contender for the most pointless work in the history of economics. Of course, even if he did include the organisation of work or industrial organisation (two different things btw) it would be beside the point given his explicit adherence to formalism as a mathematical philosophy means that Theory of Value isn’t *about* anything at all. It’s just a bunch of strings.

    Second, I’d be very careful if I were you in criticising Arrow and Hahn for not having used the term ‘occupation’, you may end up being surprised as to how general the concept of a commodity is in the said models. Lately, we don’t hear much about ‘occupation’. Instead we hear about ‘jobs’. (The everyday language does not always correspond to concepts defined by suitable choice of mathematical objects.)

    Again beside the point. Having agents demanding job-specific leisure does not address the question of the relationship between work organisation and welfare (and the implications for the meaning of ‘efficiency’).

    I am surprised you are not mentioning the difficulty in getting people to reveal their preferences (see mechanism design).

    Why? Such problems would only be compounded in the presence of preference endogeneity.

    Strange, isn’t it, that the generic Pareto inefficiency result for competitive private ownership economies with incomplete markets is not (as yet) labelled the 3rd fundamental welfare theorem.

    Outcomes like that are not popular with a lot of economists.

    To return to my point. Why is there a need to restrict oneself to a philosophical beliefs underlying theoretical models of a competitive private ownership economy? Isn’t it the case that by you ignoring my sentence of considering an society where people classified as sociologist provide input into ‘how the economy should work’, the link to the topic of this thread is lost, IMHO.

    You’re barking up the wrong tree. The question is not how do agents come to have the preferences that they do (which I agree is too hard), but what are the implications for economic analysis of preferences that *change* in response to changes in variables in economic problems (i.e. preferences that cease to be data).

  13. @Wylie Bradford

    Wylie,

    1. Thank you for spelling out how you reached the conclusion of one of my statement being spurios. Your conclusion follows from your perceptions.

    2. You didn’t give a reference to the original source literature on which your understanding of the two so-called fundamental welfare theorems is based. Instead you refer to the Edgeworth-Box diagram. In your example, you confuse the case of two individuals being unable to make a decision at all (like the woman in the comedy Open All Hours) or they don’t know even for an stance of time their endowments (this is what you describe) with the case you are allegedly interested in, namely, when their preferences change as a consequence of the cause of a change in their feasible consumption set. (How about you consider a finite but multiperiod economy to at least clarify this issue.)

    3. I asked for a reference on endogenous preferences. This is a sincere request.

    4. I ignore your paragraph about Debreu.

    5. What exactly is your point – a movable goal post?

    6. EG: “I am surprised you are not mentioning the difficulty in getting people to reveal their preferences (see mechanism design).”

    WB: “Why? Such problems would only be compounded in the presence of preference endogeneity.”

    Yes it would. This is why I mentioned it. Since you have been so strong on criticising others for “expendiency”, it was surely interesting to get your answer to my question. And what do we get?

    7. EG: “Strange, isn’t it, that the generic Pareto inefficiency result for competitive private ownership economies with incomplete markets is not (as yet) labelled the 3rd fundamental welfare theorem.”

    WB: “Outcomes like that are not popular with a lot of economists.”

    The “outcome”, as you so appropriately label it, is the outcome of a research program which takes its origin in the theoretical work of Arrow-Debreu-Hahn-Radner- ………. …….. all of whom you have no time for. I assume you are providing your impression or perception on what other economists think of the said result rather than some data.

    8. Have you read the article by Fourcade et al to which I initially responded?

  14. @Ernestine Gross

    I think we’ve exhausted the seam here.

    1. No comment.

    2. Really, it’s verging on silly to characterise my remarks as confusion over whether agents know their preferences or their endowments. My point relates to my earlier comment (unremarked on by you), to wit: ” Imagine an agent with a utility function defined over whatever – a subset of commodities, all commodities, all commodities plus sundry other things – and the size of the feasible consumption set changes for some reason (e.g. a change in some prices). The agent is better off if after the change they are consuming a combination of whatever it is their utility function is defined over with a higher utility number than before the change. That is the definition of being ‘better off’. Clearly that requires comparison of combinations using the same utility function. Utility numbers from different functions being non-comparable, if then preferences change in response to whatever caused the change in the size of the consumption set, that particular definition of ‘better off’ is no longer applicable. That is why the issue of preference endogeneity is important.”

    Clearly such considerations are to the fore when considering lump-sum changes in endowments, as per the second fundamental theorem. Handwaving about the type of economy involved is beside the point as the issue at stake is what is to be meant by ‘better off’ when doing comparative statics – it is a definitional issue. Since the definition of Pareto efficiency is inextricably bound up with the notion of ‘better off’ the significance of the potential implications of preference endogeneity (properly understood) is, and should be obvious.

    3. I could dig some up, I suppose. It has been some time. Then again, there’s always ECONLIT 😉

    4. I’m not surprised. Kind of like the way the economics profession at large has effectively ignored Debreu. Sure he’s famous and all, but what exactly would be different about economics in general today (and what substantive gains would have been lost) if the development of the Walrasian program petered out with Wald? I know the answers that GE devotees give – the GE literature wouldn’t exist. In their minds that’s all that counts. In reality though (and especially for applied work) the demonstration that certain axioms and certain string manipulation rules produce certain strings (theorems) is about as relevant as doing sudoku. Especially when these strings aren’t about anything.

    5. No. Just reiterating that the choice of arguments in objective functions by economists is typically arbitrary and leaves out some things that are evidently of great importance to actual agents. You effectively acknowledged that in invoking what I called the ‘Debreu defence’.

    6. I don’t get your point here. It wasn’t necessary for me to mention it. In any situation in which welfare gains/losses are defined in relation to preference satisfaction, and the evaluation of policy alternatives based on the consideration of such welfare changes (e.g. as in economics), the question of preference revelation arises and is potentially problematic. It doesn’t matter whether those preferences are regarded as exogenous and fixed or endogenous and variable. The latter case may be more complicated (given the existence of two interfering factors) but the issues are not directly linked.

    7. I wasn’t dissing the outcome, I was alluding to those economists who reflexively reject any suggestion whatever that calls into question the optimality of competition. Unfortunately, there’s quite a lot of them out there. That’s data BTW, based on observation of people and literature . over a number of years. I’m aware of the literature and it isn’t true that I have no time for anyone involved. I’m a big fan of Arrow, and I spruik him to students at every opportunity (undergrads today having no idea who he is, sadly. Samuelson either).

    8. Yes. I found it an interesting piece, actually. I think your criticisms were well wide of the mark, frankly.

    Anyway, enjoy Christmas and have a happy New Year.

  15. @Wylie Bradford

    The seam (your term) was exhausted with your first post to me, although I couldn’t be sure until 2 rounds later. Repetition does not constitute an argument and certainly not a proof. I shall now focus on the crucial points only.

    a) The so called second fundamental welfare theorem has nothing to do with comparative static or with cost/benefit analysis. (At best, the latter are dealing only with a projection on a sub-space of the space on which preferences and technolgies are defined.)

    b) My point about your description of the problem, using the 2-dimensional Edgworth Box diagram stands. It should be obvious that you cannot treat what you allude to , namely a change in preferences as a function of a change in some parameter value somewhere in the decision problem, because you need more than two dimensions (see definition of a commodity in Arrow or Debreu or both).

    c). The first and second fundamental welfare theorems are more than 50 years old. They are still useful. For example, the current policy proposal of setting the unemployment benefit for young people (age range is defined in the policy proposal) to zero without offering government employment and given empirical evidence that there is excess supply in the so-called ‘labour market’, strongly violates the minimum wealth condition of the theoretical models of so-called ‘market economies’. All those whose parents cannot make a private wealth transfer are ‘moved to’ (to use your hand-waving language) to a no-survival point. Their set of ‘feasible consumptions’ is a zero (for all dimensions). No cost/benefit analysis is required to reach this conclusion and no comparative static is required. The policy proposal is fundamentally inconsistent with the philosophical base of the theoretical model for which the said welfare theorems constitute a property of the solution to the model.

    d) Whether or not the question of how work is organised is a separate one from ‘Industrial Organisation’ or merely one question which belongs to I.O. is moot. I don’t care one way or the other. But, I agree that the question of how work is organised (‘managed’) is of importance to people. (If you teach management students, you could set an assignment asking how they would apply the notion of Pareto efficiency when allocating tasks in their particular area. From experience, such an assignment throws light on who manages for the sake of excercising power and who is interested in achieving ‘productivity’ which is, as far as possible, consistent with the preferences of ‘labour’.)

    d) Now you talk about empirical relevance. Fine. It should be obvious that if you want to empirically investigate whether preferences change in response to a change in a parameter value somewhere in the ‘system’, then you first have to tackle the problem of eliciting the true preferences of the subjects in question regarding their next period preferences (ie after, say a policy change). Things have moved on during the past 50 years. No wonder you couldn’t find a reference to substantiate your assertions or allusions or whatever.

    Anyway, your provocations provided a nice diversion from the Christmas preparations.

    With best wishes for the holiday season and the New Year.

  16. The first and second fundamental welfare theorems are more than 50 years old. They are still useful. For example, the current policy proposal of setting the unemployment benefit for young people […] Their set of ‘feasible consumptions’ is a zero (for all dimensions). No cost/benefit analysis is required to reach this conclusion and no comparative static is required.

    I’m not certain you need the first and second fundamental welfare theorems to come to this conclusion; I mean, I don’t really have the faintest clue what these theorems say and I still managed to come to this conclusion fairly quickly; I don’t think that this is unusual.

    If they are useful, what do you use them for?

  17. @Ernestine Gross
    Well, may be one more. Ignoring points made doesn’t make an argument either. If you can show that the notion of ‘better off’ used in welfare economics doesn’t require constancy of preferences (not forever, but certainly in the context of the change in the agent’s environment that is being contemplated) then feel fee to do so. I have no interests in proofs, per se, as they have zero empirical content so far as economics is concerned.

    a) I didn’t say that the second fundamental theorem is related to cost-benefit analysis. I said that cost-benefit analysis is ultimately based, theoretically, on social welfare functions (which are invalid constructions).

    b) Fine. Let it be more than 2 dimensions. What stands is that the definition of Pareto efficiency depends on the notion of ‘better off’ which in turn is defined with reference to constant preferences. Although, it isn’t immediately clear why agents need to be aware that they have preferences, or what they are. We assume that agents’ choice behaviour conforms to a certain pattern, which is consistent with stability of preferences. Kagel, Battalio and Green (1995) claimed to show that rats behaved in a manner consistent with consumer choice theory in an experimental setting. Presumably we can agree that rats don’t have awareness of their preferences.

    c) This is really the corker. The minimum wealth conditions (whether they be expressed as assumptions of interiority, desirable commodities/productive labour, irreducibility, resource-relatedness etc) are there to support the proof of existence of a competitive equilibrium i.e. that certain strings are derivable from certain axioms using certain manipulation rules. They, like your beloved A-D framework, have no empirical content. To argue against a policy proposal on the grounds that it is inconsistent with these conditions is bizarre. Firstly, it is question-begging on a cosmic scale – what argument do you have to back up the implicit premise here that the A-D framework is empirically descriptive? I can’t see much evidence that Debreu thought it was (given his explicit commitment to Formalism). Replace ‘price’ throughout Theory of Value with ‘bunyip’ and ‘commodity’ with ‘jabberwocky’ and all the strings are still derivable and no more meaningful. If it has no empirical content, why should we care at all whether existence proofs are jeopardised? Have we been in a competitive equilibrium up until now, and this policy will mean that one will no longer exist? So what? If we haven’t been in a competitive equilibrium up until now, then who cares? You can see the effect on unbiased observers of this conflation of purely logical exercises with empirical policy assessment in Collin’s reaction. His closing question is certainly to the point.

    As for the idea that the minimum wealth conditions represent a philosophical base of the model (i.e. that equality of some kind is fundamentally important), that brings to mind J.B. Clark’s justly derided ethical interpretation of marginal productivity theory. The conditions are there purely for technical reasons, they are not a political manifesto. Koopmans, in his Three Essays, discusses the starvation problem, and reports a conversation with Debreu in which the latter pointed out that the life plan formed by Arrow-Debreu consumers in response to the price system includes a specification of longevity based on resource holdings and the ability to work. Thus, survival need only be guaranteed into the first period. Subsequently, in Koopmans’s words, “the amount and initial distribution of resources and skills in the population [will] determine the pace and extent, if any, of starvation” (p63). So, yeah, all those assumptions are there because they really care about people.

    d) You agree that work organisation is important to actual agents. You pointed out that Debreu explicitly excludes such considerations. So where does that leave the welfare conclusions, and how do you accommodate this obvious lacuna in your beloved framework?

    e) Now you’re just conflating the determination of whether preferences have actually changed in a given instance with the issue of whether they should be *assumed* to be constant for theoretical purposes.

  18. @Collin Street

    1. You reaching the same conclusion, without knowledge of the theoretical models to which the first and second fundamental welfare theorems are characterisations of the solution to the model, as I have, using this knowledge, is, I suggest akin to a relative of a deceased having reached the same conclusion as that of a medical practitioner, namely the person is dead, even though the said relative does not have knowledge of medicine.

    2. It would be a great worry if a theoretical model of an economy, modelled as economic aspects a social system, would lead to all implications making no sense to the people in the economy. Your comment clearly shows that this is not the case.

    3. Another example of usefulness is that those who argue for ‘freedom of choice’ and ignoring the minimum wealth constraint are people who either doen’t know what they are talking about or they want to pull the wool over others’ eyes.

    Best wishes for Christmas and the New Year.

  19. @Ernestine Gross
    If, of course, what doctors did was scrabble together some axioms about the human body and then start generating theorems, all the while maintaining that, in Debreu’s words, “the theory is…logically entirely disconnected from its interpretations”.

    That you would invite comparison between a pretty effective inductive empirical discipline such as medicine and the otiose logical scribblings of the Neo-Walrasians is a pretty good example of the self-regarding faux superiority of economists which prompted this thread in the first place.

  20. @Wylie Bradford

    You are so far the only self-declared “economist” among those who commend on this thread and you are doing an excellent job in portraying yourself as a self-regarding faux superior.

  21. Lol. Yes, because I was the one taking MaxPo to task for besmirching their good name by deviating from the scientific method and publishing a paper by a sociologist commenting on economics. No, wait, that wasn’t me, was it? This thread concerned economists’ tendency to regard themselves as superior to other disciplines in the social sciences/humanities. I haven’t referred to any other disciplines at all, nor do I consider myself superior to anyone else on the basis of my educational background alone.

    I merely disagreed with you on some matters to do with economics. If that (and my mean comments about the Neo-Walrasian program) upsets you so, then perhaps you should invest in acquiring some coping skills.

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