Pieria: Introducing Post-Keynesian Economics

My newest article at Pieria provides an overview of the post-Keynesian theories of consumers, producers, money/banking and trade:

A common charge directed at heterodox economics is that it is defined as a negative and has little to offer in the way of an alternative to mainstream economics (at least, if we ignore the ‘extremes’ of Austrianism and Marxism). It’s true that heterodox economists, including myself, often spend more time criticising mainstream economics than we do offering alternative theories. Yet there is in fact a large amount of work on alternative theories of pricing, distribution, finance and trade. Below I will sketch out what is known as the ‘Post-Keynesian’ (PK) approach to economic theory….

The summary echoes what I’ve said before about the difference between mainstream and heterodox economics:

First, post-Keynesians tend to emphasise that key variables (wages, the rate of interest) are monetary, not real phenomena. This doesn’t mean the notion of the real is unimportant – far from it – but it does mean that it is often a poor starting point for analysis. Second, there is generally no special status accorded to particular variables. Consumers and producers are not ‘optimising’; trade between countries can be imbalanced for long periods of time; the economy can remain in a state of depressed demand and no adjustment of prices will save it. Third, there is a lot of emphasis on institutional considerations. Since prices, demand and trade depend somewhat on social norms and agreements, and since agents tend to fix their decisions for long periods of time to maintain a degree of certainty, different economic trends can persist based on historical path dependence, and there is no ‘one size fits all’ model.

I have to say that I’m not sure why post-Keynesians don’t spend more time on this stuff. I find the theories pretty comprehensive and quite obviously more grounded in reality than the neoclassical approach.

About these ads

, ,

  1. #1 by Roman P. on September 25, 2013 - 1:52 pm

    You wrote:
    “Furthermore, firms generally face at least constant – in fact, probably decreasing – returns to scale, so quantity is varied more readily than price (consumers also dislike frequent price changes).”
    Did you mean increasing returns to scale?

  2. #2 by Roman P. on September 25, 2013 - 2:09 pm

    I think that, overall, this article was hit and miss concerning its stated purpose: it mostly went “We think the world works thusly” without going into into theories of why and models of how. This doesn’t exactly disprove the notion that PK economics is just a rebuttal of neoclassical school. I guess the popular website for general public is not the right place for going into arcane stuff like Weintraub’s models, though.

    • #3 by Unlearningecon on September 25, 2013 - 2:19 pm

      Did you mean increasing returns to scale?

      Yes, whoops. That is not the first time I’ve written that phrase wrong – have fixed it.

      I think that, overall, this article was hit and miss concerning its stated purpose: it mostly went “We think the world works thusly” without going into into theories of why and models of how. This doesn’t exactly disprove the notion that PK economics is just a rebuttal of neoclassical school. I guess the popular website for general public is not the right place for going into arcane stuff like Weintraub’s models, though.

      Yeah, I mean it’s a ‘layman’s guide to post-Keynesian economics’, which is just meant to show that there are other ways you can model economic behaviour. It’s also true that it is difficult not to define post-Keynesianism in opposition to neoclassical economics, as I was eager to show how exactly the theories are different. Perhaps that’s a mistake.

      • #4 by Roman P. on September 25, 2013 - 5:13 pm

        This is a big problem for PK economics: convincing people that it is true. It is certainly exasperated by that some of its arguments consist of observations that the world works thusly and not thusly. It is trivially easy to convince a person who is bound by the laws of logic that a Euclidean triangle with equal sides will have angles of 60 degrees, but try convincing a fool that 2 + 2 does not equal five!
        An economist might be Gödel himself reborn, if he doesn’t check his axioms and the conformance of his ideas to reality (and institutional structures) he will end in failure. Garbage In, Garbage Out.

  3. #5 by J Sterling on September 25, 2013 - 2:26 pm

    “defined as a negative and has little to offer in the way of an alternative to mainstream economics (at least, if we ignore the ‘extremes’ of Austrianism and Marxism”

    I suspect that’s a bit of a Morton’s Fork: if your heterodoxy is not too far from orthodoxy, then they can say it’s defined by its relationship to the orthodoxy (like when you describe a restaurant as being a block from the town center, instead of placing it by its latitude and longitude). If your heterodoxy is sufficiently far from the orthodoxy to stand on its own as an independent description of reality, it’s “extreme”. Heads you lose.

    • #6 by Unlearningecon on September 26, 2013 - 12:43 pm

      This is true, although personally I don’t regard the word “extreme” as an insult, even though many use it that way. As you can see, commenter luis has already charged that there’s little in PK economics that the mainstream can’t do. It’s been referred to as the “note and forget” approach – yeah, we can do lexiographic preferences, constant returns to scale, financial frictions etc, look at this paper – but the question of which behaviour is the more general rule, and which framework does the best job of modelling it, never seems to be answered. Instead we have a myriad of papers pushing the neoclassical framework as far as it will go, often in different directions.

  4. #7 by Magpie on September 27, 2013 - 1:07 am

    While I agree with Roman P (“I guess the popular website for general public is not the right place for going into arcane stuff like Weintraub’s models, though”), I am much more upbeat about the post.

    In fact, whatever the limitations imposed by the medium, as an overview is pretty great.

    Suggestion: elaborate on the “cost-plus” pricing rule in a separate article, perhaps in your own blog (BTW, the “cost-plus” link — 2nd paragraph after “Producer Theory” — does nothing). Maybe additional links would be useful, too.

    Cheers.

    • #8 by Unlearningecon on September 29, 2013 - 12:17 pm

      Sometimes JSTOR seems to bugger up, but I’ve done what I can to fix it and I’ve added a Wikipedia link, which is better than nothing.

      And yes, post-Keynesian pricing theory is interesting and I will probably look at it more in depth later. In the meantime, and if you haven’t seen it before, you might find Lord Keynes’ series on Lee’s book useful.

      • #9 by Magpie on October 7, 2013 - 11:31 pm

        I’ve been giving it a look.

        This is probably premature, but, so far, what I’ve seen in that particular blog about PK prices has left me underwhelmed, particularly the last blog, where Lavoie is quoted.

        I better wait until you have something.

        Cheers.

      • #10 by Unlearningecon on October 8, 2013 - 11:48 am

        I understand that it’s underwhelming from a theoretical perspective. I mean, as I’ve said before, I think economics should walk before it can run, and hence we’re better off looking at case studies etc. than trying to devise theories that are anywhere near as elaborate as the neoclassical theories.

  5. #11 by Luis Enrique on September 27, 2013 - 8:03 pm

    “charged that there’s little in PK economics that the mainstream can’t do”

    I merely pointed out that the conventional approach can capture preferences such that necessities are prioritised over fripperies. Of course it can. But actually that wasn’t very relevant, because you were making a much stronger claim (that I went on to try and show was not a very credible one) that even such a model would be wrong, because it would still involve the possibility of responding to relative price changes. So actually this was not a claim that the mainstream can do anything PK can – quite the opposite: the mainstream wouldn’t model things so people do not respond to relative price changes as you suggest. [personally I think you just dramatically over-state argument, a more careful claim might be that people may only respond slowly, and only to large price changes].

    I think “note and forget” is rich coming from you. I don’t know how many times I have corrected you when you claim economists do something they do not or don’t do something you say they do, without apparently having made any lasting impression on you. For instance I have told you before that rule of thumb mark-up pricing is actually covered by introductory mainstream micro where the size of the mark-up is related to the elasticity of demand. Here’s another example – in your article you firms generally face constant returns to scale, as if this is something mainstream economics neglects. But of course the ubiquitous Cobb Douglas production function is usually parameterized with constant returns to scale. I could go on to tell you all about the work economists have done with increasing returns, but you will probably dismiss this as just another one of a myriad of papers.

    And that’s what I want to ask you: as opposed to what? Of course economics exists in the form of research papers. And books. And university courses. I can show you books and courses about increasing returns to scale. And economists are constantly trying to work out “what description fits best” (i.e. estimating returns to scale in various setting) and explore the implications. What is it that you want to see? Where else are models with financial frictions supposed to exist? Not in introductory text books.

    I hope that isn’t too combative for you.

    • #12 by Unlearningecon on September 29, 2013 - 12:27 pm

      Well, again: you can be incredulous about the PK approach to consumer theory and say things like “hah! OF COURSE they would respond to relative prices changes!” But this is an empirical question and if the PK approach is at least a closer approximation than utility, as behavioural work (which you referenced, briefly) suggests, then it is the better one.

      Mark up pricing is indeed introduced in introductory textbooks, but of course it is argued that it’s just the same as equating MC=MR if firms use a mark up which is a function of the PED. And no evidence I’ve seen suggests they do. So why bother clinging to the approach? I’m painting a picture of a firm which has a somewhat arbitrary, constant mark up and spare capacity which varies according to demand, which is always at least somewhat unpredictable. This is simply at odds with the marginalist idea that firms squeeze out every possible drop of profit.

      And yes, C-D can have any level of returns to scale, and I don’t know where I’ve claimed otherwise. Of course C-D also has its own problems.

      What is it that you want to see? Where else are models with financial frictions supposed to exist? Not in introductory text books.

      The question is ultimately whether we start with the microfounded, optimising agent framework and move on to incorporate various frictions or what have you, or whether we start with a different approach that is less mathematically precise but more grounded in observed behaviour. It takes A LOT of tweaking for the optimising approach to become even loosely consistent with one or two facets of the observed behaviour of agents/economies, whereas eg basic stock-flow consistent models can model both stagflation and financial crises.

    • #13 by Boatwright on September 29, 2013 - 3:26 pm

      LE:

      PK theory starting from the computational impossibility of calculating optimal utility, posits that consumers follow simple rules of thumb, follow fashion, self-limit a choice set, etc. — all of which do not fit well into the basic NC principle that economic choices are fundamentally rational decisions. Quoting UL: “Consumers aren’t really ‘optimizing’ anything; they are at best ‘satisficing’.”

      It seems NC tries to answer this problem by adding theory elements that prioritize necessities, etc.. I fail to see how these attempts to tune-up NC theory answer the fundamental problem of the shopping cart’s computational impossibility. When comparing the price of a Picasso to the cost of a loaf to a hungry man, the essential element of NC consumer theory, that consumers will generally optimize utility, loses all meaning. Perhaps we need to look elsewhere for answers in models that look for emergent patterns that can be modeled from actual non-rational (as distinct from irrational) human behavior.

  6. #14 by Ramanan on September 28, 2013 - 12:39 am

    Nice you mention Verdoorn Law.

    An important aspect of PKE is the “reverse causality” (right causality!) and the Kaldor-Verdoorn Law is my favourite in this – especially because of its connection to the balance of payments constraint and growth story.

    • #15 by Unlearningecon on September 29, 2013 - 12:28 pm

      Reverse causality seems to be a frequent problems with things like this. The purported ‘wealth effect’ of asset prices springs to mind.

  7. #16 by Luis Enrique on September 30, 2013 - 12:14 pm

    come on now. you know the mc=mr profit maximisation isn’t claimed to be “what firms actually do”- that’s why when introducing rule of thumb pricing in textbooks, economists write something along the lines that firms do not generally have given, known, cost and demand curves, hence need for a rule of thumb. We are talking about highly stylised abstractions. How many times have I had to make that point? 

     Why bother clinging to the profits max approach? For one things, because it tells us something about what profit maximisation means. Imagine if economists had not built these models:

    Q: so how do firms choose prices?
    A: oh, they just slap on an arbitrary mark up
    Q: and if they wanted to increase profits, what should they do?
    A: dunno
    Q. Right. So do we know how close actually existing firms come to maximising profits?
    A: no, and we have no way of knowing because we haven’t worked out what that would look like.

    And for another thing, it gives us an idea of why firms choose different markups. Do you really think that the difference in the mark ups Apple and HTC achieve on their mobile phones has nothing to do with their elasticites of demand? Ferrari and Nissan? I think economics should be able to say something about what might influence the choice of markup. And it can. 

    And maybe you could write to Ed Milliband telling him not to worry about profiteering energy firms because firms just apply a mark up to costs so any price increases simply reflect input price changes. Meanwhile as a neoclassical economist I could agree that oligopolists trying to max profits might be ripping us off. (although I’d also say price caps are a dumb idea, but that’s another story). 

     On whether the idea that rental expenditure does not respond to relative price changes, would evidence from a randomised experiment in which some households were given rent subsidies to change the relative price of accommodation settle it? 
    http://hanushek.stanford.edu/sites/default/files/publications/Hanushek%2BQuigley%201980%20REStat%2062%283%29_0.pdf
    That’s that settled then.

    And really … No you didn’t “claim otherwise” about CD production functions. You did, however, appear to claim that the idea firms generally have constant or increasing returns to scale was some sort of PK insight. Hence me pointing out it is most common to assume constant returns. 

    the more I think about this “firms apply arbitrary mark-up” idea, the more entertaining it gets. It implies we can shut down the competition commission, for example. In the standard model as firms gain market power their elasticity of demand changes and the mark-up profit-maximizers would would choose rises. But if firms are just slapping on a habitual or arbitrary mark-ups unrelated to demand elasticities, these concerns disappear.

    meanwhile, you write “no evidence I’ve seen suggests” mark-ups are related to elasticities of demand. And how familiar are you with the empirical evidence?

    Here’s a nice set of introductory lecture notes on pricing by Pindyck, which mentions some empirical cases as it goes along (esp drugs and razor pricing)

    http://web.mit.edu/rpindyck/www/Courses/Pricing.pdf

    here’s an excerpt from a very sophisticated attempt to estimate pricing behaviour in the auto industry

    http://pages.stern.nyu.edu/~rslee/teaching/io/papers.demand/Berry%20Levinsohn%20Pakes%20(2004%20JPE)%20-%20Differentiated%20Produce%20Demand%20Systems%20from%20a%20Combination%20of%20Micro%20and%20Macro%20Data%20-%20The%20New%20Car%20Market.pdf

    ” … the pattern of the elasticities seems be fairly robust across our estimates of and accords well with industry reports (especially to reports circa 1993). Semi-elasticities decrease in
    price, and given price, vans (both mini and full-sized), pickups, SUVs and, to a lesser extent, sports cars have noticeably smaller elasticities than other vehicles. This goes a long way toward explaining reports of high markups for these vehicles”

    you might find appendix C here instructive too

    http://dev3.cepr.org/meets/wkcn/6/6696/papers/Cassiman_markupsv5.pdf

    here’s something on how mark-ups respond to variations in aggregate demand

    http://www.bankofengland.co.uk/publications/Documents/quarterlybulletin/qb080203.pdf

    it took me about 5 minutes to find this lot.

    i know you don’t find my comments productive, personally I think it would be productive if you digested this stuff and amended your beliefs accordingly.

    • #17 by Unlearningecon on September 30, 2013 - 1:42 pm

      The more I read your comments on this matter, the more I’m convinced you either need to improve your reading comprehension or learn to use the principle of charity.

      I used the expression “somewhat arbitrary” to refer to the fact that the mark up is often quite a neat, round number, nothing more. You could not read my passage in the article on the mark up and think that I think it is meaningless and we don’t need competition policy – at least, not if you were at all interested in interpreting me correctly. Obviously, a firm with more market power is, on average, going to use a higher mark up than one without. In the short term, however, firms are more likely to vary their sales as price changes are clumsy and disruptive for both sides.

      Those lecture notes merely “suppose” that we work for a company and asks how we might price using economist’s theories. There is no attempt to discern whether or not firms actually do this.The consumer theory is beyond my pay grade but it seems to do something similar: assume economist’s theories, then interpret the data through this lens. In any case, as I said: within similar categories of goods, like cars, we’d probably expect to see something closer to the neoclassical model. So that paper shows very little that I disagree with, at least for the purposes of this debate.

      Re: that rental paper, PKs say that such a thing would change consumption patterns because it changes income. Lavoie explicitly says in his paper on consumer theory that the PK position is that income efffects dominate substitution effects. So nothing in there contradicts my point.

      personally I think it would be productive if you digested this stuff and amended your beliefs accordingly.

      Personally I think it would be helpful if you didn’t respond to my arguments by googling for 5 minutes and throwing papers at me that are only vaguely related to your point. I could sit here and go on at you about how you haven’t read the post-Keynesian approach to xyz fully, but I don’t.

      We are talking about highly stylised abstractions. How many times have I had to make that point?

      And how many times to I have to make the point that these are useless and there are better approaches? At Boatwright points out, people cannot optimise. It is computationally impossible. And since we often do know the actual decision rules they use, why not go with them instead?

  8. #18 by Luis Enrique on September 30, 2013 - 3:30 pm

    mainstream models suggest that the mark-up that maximizes profits is related to the elasticity of demand.

    you think those models are useless, and say you’ve seen “no evidence” to suggest firms choose mark-ups in that way.

    I was responding to the passage ” [mainstream theory says] firms use a mark up which is a function of the PED. And no evidence I’ve seen suggests they do….I’m painting a picture of a firm which has a somewhat arbitrary, constant mark up”

    I provide some evidence for you: a bunch of papers showing how mark-ups are related to the elasticity of demand. [you must have read those lectures notes too quickly: there are some claims that actual pricing decisions match the theory]

    you think I am throwing papers at you only “vaguely related” to the point?

    I now discover you used the words “somewhat arbitrary” to mean they pick a round number. I thought you meant their choice is not a function of the elasticity of demand but is chosen in some other “somewhat arbitrary” fashion. You think I was wilfully misunderstanding you.

    You also describe firms mark-ups as “constant” – now again, that would suggest to me you don’t think firms adjust their mark-ups as their market power changes. We have a competition commission because we think firms may change their mark-ups as they gain market power. Or am I now wilfully misunderstanding what you mean by “constant”?

    You say it’s obvious firms with more market power will choose higher mark-ups. Well yes, that’s bang in line with the mainstream model, firms use a mark up which is a function of the PED, which you reject. But now apparently agree with.

    I could go on … but we hit diminishing returns a long time ago didn’t we.

    ” I could sit here and go on at you about how you haven’t read the post-Keynesian approach to xyz fully”

    yes for example you could have linked to something showing how PK says firms choose mark-ups that differs from relating the mark-up to the elasticity of demand. that would be helpful. In apparent contrast to you, I think having my attention drawn to new information is a good thing.

    “At Boatwright points out, people cannot optimise. It is computationally impossible”

    you’re doing it again – mathematical abstraction != what people actually do. Do you honestly think economists believe people compute optimal decisions?

    ” And since we often do know the actual decision rules they use,”

    whoa there! we know the actual decision rules people actually use to take economic decisions??!!

    could you maybe sketch the pricing decision rule of a mobile phone manufacturer?

    you know, I hound you like this because I think you get a lot wrong about economics that I want to correct and I think I might change your mind.

    I’m not going to, am I.

    • #19 by Unlearningecon on October 2, 2013 - 12:42 pm

      You need to learn to understand to interpret someone’s point as a whole rather than zeroing in on a particular thing. This “somewhat arbitrary” debate is worthless and tiring, but just to hammer the point home I’ll quote myself:

      This mark up can depend on any number of things: market power; norms and the firm’s history; the demands of shareholders, and can even be somewhat arbitrary (eg in many cases it is a round number).

      Later I repeat the term “somewhat arbitrary”. How should you interpret the use of this term? In the same way I’ve used it previously, or in a new, stronger sense?

      And “constant” means constant for relatively long periods of time, to gain a degree of predictability for a particular project or business plan. Obviously, firms will review and change their prices intermittently, but they’d prefer not to, as if their product is important (like energy), and particularly if they increase the price too much, there is likely to be a big hoo-ha from consumers and politicians. Firms are generally mroe interested in expanding profits through new markets than just increasing prices.

      And here’s what I’m objecting to with the PED point: firms do not use cost-plus pricing in such a way that it is equivalent to MC=MR pricing, as mainstream textbooks try to make out. Firms will consider their market power but from what I’ve seen in eg Lee there is no real estimation of the PED (in fact in one place the notion of a downward sloping demand curve is disputed for certain industries). The basic fact is this: firms very rarely use neoclassical tools. They don’t use them because they are too abstract and neat, and are simply useless in an uncertain and changing world with various institutional constraints. Hence, we are better off looking at the firms than the theories.

      you’re doing it again – mathematical abstraction != what people actually do. Do you honestly think economists believe people compute optimal decisions?

      Unfortunately I’m well aware of the ‘as if‘ argument and consider it ascientific. Yes, OK: if we had nothing better, we might stick to marginalism. But psychological traits and advances in behavioural economics, as well as the aforementioned studies of the firm, mean that often we do know what people actually do, and hence our models can improve over optimisation.

      could you maybe sketch the pricing decision rule of a mobile phone manufacturer?

      That would require case studies of mobile firms, the kind of thing I’d be very happy for intro economics textbooks to use.

      you know, I hound you like this because I think you get a lot wrong about economics that I want to correct and I think I might change your mind.

      I’m not going to, am I.

      I change my mind all the time. My critique of utility on the grounds that it assumes preferences are fixed was partially answered by endogenous preferences. My opinion of econophysics has steadily declined, while my opinion of evolutionary economics has done the opposite. There are a few areas where I’ve realised Steve Keen makes mistakes (such as his conflation of partial/general equilibrium and his misquote from Mas-Collel). But you’ll find it very hard to convince me that, overall, mainstream consumer and producer theory are worthwhile.

      • #20 by Luis Enrique on October 3, 2013 - 2:20 pm

        ffs, I just explained how I’d misinterpreted you. Obviously I’d forgotten your use of “somewhat arbitrary” in an earlier article.

        here are some distinct claims:

        1. This is what profit maximization looks like in theory
        2. Prices in practise are reasonably close to what theory predicts
        3. Firms actually choose prices by direct application of theory

        even if only 1. is true, theory is worthwhile. I provided you with some papers suggesting 2 is true, regarding the relationship between markups and PED. Previously you had seen “no evidence” for this. I have no argument with the claim that few firms do 3 because, as you say, they face a far more complex problem (you can find far more complex dynamic pricing theory too, but let’s not go there).

        But I am sorry to disabuse you, despite many interesting insights from behavioural economics etc. we do not actually know what rules people use to take economic decisions. .

        So if you want to make progress on various questions, you need a tractable model that will not be a photo-realistic picture of what firms “actually do”, but a useful abstraction. You started by writing about how PK can make positive contributions. I hope that’s correct, If so you will find PK theorists making use of unrealistic models in the sense that are not the “tools firms actually use” but might be realistic in the “mark-ups do seem to be related to the PED” sense.

      • #21 by Unlearningecon on October 4, 2013 - 7:27 pm

        The thing is, “stylised facts” about firm behaviour (and consumer behaviour) are consistent with a wide range of theories- for instance, monopolies having higher prices verges on trivial. Since we don’t have controlled experiments in economics, I feel the best thing we can do to know what’s really going on is shine a light on internal mechanics. The best approach to business economics would be similar to that taken in business school: industries and departments are looked at differently, with case studies and so forth, and theories where they are relevant. I am happy to have neoclassical theories included in this in areas where they are applicable, such as agriculture and auctions.

        The problem now is that there is no such debate to speak of. It’s neoclassical or it’s nothing, and I just don’t see how neoclassical economics has earned this status as the ‘default’ approach into which every revelation must be either annexed or discarded. Why not just accept competing models, including the neoclassical, post-Keynesian, behavioural, neurological, evolutionary and whatever other approaches, and have the debate out in the open?

  9. #22 by Luis Enrique on October 7, 2013 - 8:22 pm

    So, we have a very simple theory that tells us why firms with more monopoly power (lower PED) will choose higher markups, because thats how to max profits. This fits the data pretty well. You are not impressed and say other theories could explain the data. Can you give me an example of such a theory, that provide another reason why we observe this correlation between markups and PED?

    Whilst case studies can be interesting, economists need tractable abstract models of price setting that they can put to work in models that require a price setting component.

    • #23 by Unlearningecon on October 8, 2013 - 12:06 pm

      Yes, the neoclassical theory is consistent with that fact; however, I am concerned that it is not consistent with others, such as the idea that firms maximise profits.

      Can you give me an example of such a theory, that provide another reason why we observe this correlation between markups and PED?

      Based on the fact that most businesses don’t actually consider or estimate the PED, Kalecki theorised in terms of the “degree of monopoly”, which was largely based on industry concentration: firms would price what they could get away with, base don the possible behaviour of other firms in the industry.

      • #24 by Luis Enrique on October 8, 2013 - 1:19 pm

        I would say that most firms want to increase their profits, evaluate new products in terms of their profitability and take decisions to increase or protect profits. The theoretical objective of maximizing profits is merely a neat, precise way of putting that into a model. So when you say firms do not maximize profits, I think you are missing the point.

        I would like you to explain why firms with more monopoly power (lower PED) will choose higher markups. Yes, the degree of monopoly has something to do with industry concentration (number of competitors) but WHY will firms “price what they could get away with”?

        You have claimed that basic mainstream pricing theory is useless. I claim it can explain observed pricing behaviour, which looks like a good reason to call it useful and insightful to me. You have said alternative theories can equally well explain observed pricing behaviour. I would like you to give me an example.

        “firms would price what they could get away with, based on the possible behaviour of other firms in the industry.” does not explain anything. How are other firms going to behave? What does “get away with” mean? A firm with monopoly power could certainly “get away with” charging a lower price such that it makes less profit – why don’t they? They could also get away with charging a higher price, such that demand as low and again they make less profit. Why don’t they? What are they trying to achieve when they choose higher mark-ups when the PED is lower?

        p.s. I realise you only gave me a very brief description of Kalecki’s ideas: I am asking you to expand on that.

      • #25 by Unlearningecon on October 10, 2013 - 2:50 pm

        The “maximisation of profits”, as you put it, merely becomes a catch-all into which we can rationalise any behaviour. Focusing on growth? Well, they’re maximising profits in the long term. What is the problem with a plethora of different models, some of which focus on profit maximisation, some which don’t?

        I would like you to explain why firms with more monopoly power (lower PED) will choose higher markups.

        You’re assuming your conclusion by defining “monopoly power” as a lower PED.

        Take table 7 of this paper. It presents evidence that firm’s prices do not really respond to demand shocks. Kalecki’s theory proposes that instead, the ‘markup’ will depend on market structure and industry concentration levels, which is what is meant by the “degree of monopoly”. Like a lot of heterodox work, there is limited empirical evidence available (though both papers refer to some), but ultimately it’s a question of whether the “degree of monopoly” or the PED are better approximations.

        What’s more, it’s worth noting that your Friedman-esque instrumental approach, focusing on ‘prediction’, is not the be all and end all of science. After all, many challengers to the status quo, like the idea that the earth revolves around the sun, initially had inferior predictions, but are still a more accurate description of the way the world works.

        And again: the data in those lecture notes are very loose. For example:

        …this implies a profit-maximizing price in the range of $6.00 to $8.00. When Gillette introduced the Mach 3 in 1998, they priced it to sell at retail for $6.49 to $6.99.

        This is not a very precise measurement. Now, I understand that: precise measurements are difficult in social science. But for heaven’s sake, why not actually discuss what went on inside Gillete?

      • #26 by Luis Enrique on October 14, 2013 - 1:20 pm

        “.. “maximisation of profits”, as you put it, merely becomes a catch-all into which we can rationalise any behaviour.”

        no it cannot. For instance it cannot rationalize a decision that raises growth and wipes out profits.

        “You’re assuming your conclusion by defining “monopoly power” as a lower PED”.

        what conclusion have I assumed?

        let’s recap

        me: standard theory says mark up function of PED
        you: I have seen no evidence for that
        me: provides evidence
        you: yes but other theories are consistent with that data too
        me: what theories please?

        you haven’t EXPLAINED anything yet.

        “What’s more, it’s worth noting that your Friedman-esque instrumental approach, focusing on ‘prediction’, is not the be all and end all of science.”

        yes I know that, which is why I still find value in theories with poor predictive power. I thought you were the one who wanted economics more empirical

  10. #27 by rf on October 12, 2013 - 6:22 pm

    A minor quibble, but calculating optimal utility is not “computationally impossible” (undecidable). Rather, it is NP-Hard, which means that verifying a given solution is feasible but that probably no algorithm exists whose run time doesn’t blow up exponentially as the number of goods increases. (I say “probably” because this is the famous open question of whether P = NP).

    The way computer scientists deal with these kinds of problems is usually by applying heuristics to prune decision trees – reject any bundle with 500 oranges immediately and don’t bother checking whether the agent might prefer 2 or 3 bananas. Often, finding the optimal solution isn’t even important, only finding a good solution, or even any solution.

    In other words, we use rules of thumb and satisficing.

    • #28 by Unlearningecon on October 13, 2013 - 1:51 pm

      Thanks for your comment – I agree that “computationally impossible” might be a misnomer; perhaps “computationally unfeasible” would have been more accurate.

      What I was trying to get at with “impossible” was that calculating, say, the optimal bundle of goods from a supermarket, might take longer than the life of the universe.

Follow

Get every new post delivered to your Inbox.

Join 948 other followers