Against Marginalist Pricing Theory: US Consumer Prices and Capacity Utilisation

Sticky-Prices

Marginalist economic theory tells us that when there is unemployment of capital resources prices should fall. Some marginalists like the New Keynesians and the neo-Keynesians will supplement this by saying that prices can tend to be ‘sticky’. Let us ignore these for a moment and come back to them in a moment. Let us first take the idea that prices should fall when there is unemployed plant and equipment.

First of all, some theory. The argument is extremely simple: if plant and equipment are unemployed then there is inadequate demand for the goods and services being produced. In marginalist theory firms should respond to this shortfall of demand by cutting prices. This will generate more demand at the lower price and the market should clear; i.e. the plant and equipment will be brought online once more.

If this argument is true then we should see prices fall when capacity utilisation falls. We should clear here: we are not saying that we should see inflation slow down when capacity utilisation falls. Rather we should see prices fall — i.e. we should not see inflation slow but rather we should see deflation.

During the extreme events of 2009-2010 we actually saw this relationship for a very brief moment in time. You can see how this relationship looks on the two graphs below. On the left is a scatter-plot diagram that maps the relationship between CPI inflation and capacity utilisation while on the right is a simple line graph depicting the same relationship. (The blue line is CPI and the red line is capacity utilisation).

CPI vs Capacity Utilisation 2009-2010

Without getting into the debate as to how far prices should theoretically fall for a given fall in capacity utilisation we can at least visualise how such a relationship should look. (In actual fact we should see far larger amount of price deflation to clear markets where capacity utilisation has fallen below 68% as we see above, but regardless…).

Now, when we turn the the larger aggregates of data we see something rather different altogether. Here is the data for the inflationary period between 1967-1981.

CPI vs Capacity Utilisation 1967-1981Not much a positive correlation here at all. In fact we appear to have something of a negative correlation (indicated by the downward-sloping trend-line). We could probably do some arbitrary ‘lagging’ here to get a better fit but as we shall see we would then have to alter the lags significantly in the next time period to get a similar fit. Besides, at no point in this period do prices actually fall.

But maybe given that the inflation in this period was rather high this was just a freak accident. So, let’s turn to the era of the ‘Great Moderation’.

CPI vs Capacity Utilisation 1982-2014

Here we have a slight positive correlation. But it is not very strong. Again, some lagging might do the trick. But a glance will confirm that the lags imposed on this time period would have to be very different from those applied in the last time period to make it fit. Frankly, I think any move in this direction would be the sort of statistical cheating that goes on so much in economics and should be dismissed well before it gets underway. Anyway, none of this should concern us too much lest we lose sight of the key argument.

What is most important, as we have said, is that we actually only see one period in which prices actually fall (2009-2010). Even when we do see the positive correlation that we are looking for in other periods — however slight that might be — it is only a manifestation of a slowing of inflation and not deflation.We can see this particularly clearly if we zoom in on the period of the 1981-1982 recession which, apart from the recession of the late-00s, is the largest on record after World War II.

CPI vs Capacity Utilisation 1981-1982

As we can see, we have a very tight positive correlation between inflation and capacity utilisation. But all we see is a fall in the rate of inflation. We never see actually see a fall in prices (i.e. deflation) and as we have said: for the marginalist argument to work prices have to actually fall when demand dries up and capacity utilisation falls.

Okay, but no one really believes that prices adjust in the short-run, right? As I said, the New Keynesians argue that prices are indeed ‘sticky’ and so we should not expect them to fall when demand for goods and services dries up. But my question then becomes: where is this ‘long-run’ period in which prices do fall in response to a shortfall in demand? If the long-run is merely a series of short-runs — and, indeed, what else can it be? — then when or where does it come about?

I suspect that it is not actually an entity that we can track in time-series data because it is not actually an entity that exists outside the minds of marginalist economists. When they say that prices adjust in the long-run they are saying nothing at all, or at least nothing that corresponds to anything that happens in the real economy. The New Keynesians hint that markets will clear up low capacity utilisations in the ‘long-run’ but this long-run never turns up in reality because it is a mystical and spectral entity. Indeed when we consult Wikipedia we see that the ‘long-run’ is actually a tautological entity.

In macroeconomics, the long run is the period when the general price level, contractual wage rates, and expectations adjust fully to the state of the economy, in contrast to the short run when these variables may not fully adjust.

The long-run is thus defined as being when the price level falls to soak up excess capacity. Now, when we examine the data this never appears to happen. Does this not then mean that, by the above definition, the long-run does not exist in the real-world? I would argue that this is indeed the case: the long-run as so defined does not exist in the real-world. It is either a tautological thought experiment or it is nothing at all.

So, why do the marginalists hold to such a tautological thought experiment? I believe that this is ideology plain and simple. It is simply a belief — not subject to empirical validation or invalidation (i.e. in Popperian terms: non-falsifiable and thus non-scientific) — that the marginalists believe because they want to believe in it. When you start to dig this deep you start to see what this particular wing of the profession actually are and what purpose they serve in society. But let us not address such issues here.

About pilkingtonphil

Philip Pilkington is a London-based economist and member of the Political Economy Research Group (PERG) at Kingston University. You can follow him on Twitter at @pilkingtonphil.
This entry was posted in Economic History, Economic Theory. Bookmark the permalink.

59 Responses to Against Marginalist Pricing Theory: US Consumer Prices and Capacity Utilisation

  1. NeilW says:

    It always amuses me that marginalises want micro foundations for all macro thoughts (as if they have some special insight on the aggregation function), but never seem to want micro foundations to the ‘long run’ arguments by referring to the short run (as if they can never know what the aggregation function is there).

    So they claim special clairvoyant powers in the field of aggregation, but not if that involves time.

  2. Is this really an “anti-marginalist” argument? I would argue that firms adjust their utilisation based on (expected) demand along fairly standard marginalist lines. If we assume diminishing marginal costs (scale economies) then an increase in utilisation supports lower prices under competition. Also, there are huge compositional factors not being considered here.

    • So you are therefore saying that properly conceived an increase in demand should, in mainstream economic theory, lead to a fall in price due to economies of scale? Do you really think that this is consistent with mainstream economics? Think about this hard before you give your answer. You are saying that when unemployment rises prices should increase. Is this what mainstream theory should say — i.e. that shortfalls in demand should cause price increases!?

  3. You are using utilisation (supply side decision) as a proxy for demand (demand side, obviously), and then arguing that a drop in demand is not leading to a fall in price. I personally think this is a long bow to draw.

  4. LK says:

    This is a great post.

    “where is this ‘long-run’ period in which prices do fall in response to a shortfall in demand?”

    Yes, you see Austrians making this claim quite often as well: that prices magically adjust in the long run. I hear this crazy rubbish all the time when I debate people at Robert Murphy’s blog.

    It’s nonsense, of course.

    The reason is the widespread prevalence of mark-up pricing (that is, price rigidity and production and quantity changes in response to demand changes) and what even mainstream economists have identified through their empirical research: price asymmetry in mark-pricing — a lot of businesses just shun price reductions, even when their total units costs fall. They prefer to maintain price and profit margins.

    • LK says:

      Sorry, I’ve got the HTML tags for italicize command wrong above.

    • Rob Rawlings says:

      LK,

      If mark-up pricing is so prevalent how do you explain the low but steady rate of inflation during the period that Pilkington covers in his post ? Who is raising their prices and why ?

      • Rob still doesn’t understand mark-up pricing. He thinks it means “prices do not move”. Which would mean that Post-Keynesians do not believe in inflation! Bizarre. Read the literature, please.

      • Rob Rawlings says:

        What is your problem, Phillip ? You have got to be in the running for the most obnoxious blogger on the internet.

        LK: These were genuine questions: In the models I am used to the CB targets inflation and increases the monetary base to achieve the target. Suppliers experience this as an increase in demand and respond with a combination of increased output and increased prices. If we are near to full employment then output cannot increase so prices will.

        This is easily explained in marginalist theory , but less easily (as far as I can see) in price+markup terms.

        I was just curious how cost+markup theorists explain the apparent ability of central banks to hit (pre-2008, anyway) inflation targets.

      • LK says:

        “Rob still doesn’t understand mark-up pricing. He thinks it means “prices do not move”.”

        Yes — odd indeed as he has supposedly read a lot of my posts on mark-up pricing.

        It just shows you how people don’t even make serious attempts to understand their opponents’ views before making stupid comments.

      • LK says:

        “If we are near to full employment then output cannot increase so prices will.”

        It is possible some prices will, but just as likely many prices won’t increase.

        How do we know this? A little thing called empirical reality: in a survey of 654 UK businesses, the business people were asked: what do you do when there is a boom in demand which cannot be met from stocks or inventories?

        Most UK firms said they (1) simply increase overtime of workers (as reported by 62% of firms), (2) hire more workers (12%), or (3) increase capacity (8%) to produce more output, rather than increase the price of their product.

        Only 12% said they would increase the price of their product (Hall, S., Walsh, M. and A. Yates. 2000. “Are UK Companies’ Prices Sticky?,” Oxford Economic Papers 52.3: 425–446, at p. 442).

      • Rob Rawlings says:

        I have no idea why you are asserting that I think markup pricing means that prices don’t move. I don’t and there is nothing I have said that would indicate that.

        I am simply trying to understand what (in cost+markup models) would cause a steady year-on-year 2% rate of inflation like we saw between 1988 and 2008. I am confused by LKs response where he appears to be claiming that businesses will not actually increase prices at full employment as this seems at odd to what actually happened.

      • LK says:

        (1) This is why debating with you is like talking to brick wall, Rob Rawlings.

        Did you even read what I wrote?

        It is possible some prices will, but just as likely many prices won’t increase, etc.”

        (2) “I am simply trying to understand what (in cost+markup models) would cause a steady year-on-year 2% rate of inflation like we saw between 1988 and 2008.”

        What caused it? Many factors. Most importantly, that most prices are mark-up prices and there is a price change asymmetry in how these prices are set: they are just raised much more frequently than lowered, often because businesses shun price reductions. Hence persistent inflation.

        Why was inflation so low? Because trade unions and wage growth has been restrained and a lot of production moved to the third world with lower costs, so many prices have fallen.

        Money supply growth is a necessary but not sufficient factor for sustained price inflation, of course: but people like you confuse the role of the central bank.

  5. Dantey says:

    Hi Philip,

    As an economics noob, I also inferred from your post that deflation is not the opposite of inflation. Am I correct?

  6. ivansml says:

    1) “Marginalist argument” you present is based on excess supply and disequilibrium in the goods market. Capacity utilization on the other hand measures the ratio between actual output and “sustainable maximum output-the greatest level of output a plant can maintain within the framework of a realistic work schedule, after factoring in normal downtime and assuming sufficient availability of inputs to operate the capital in place.”[1] Clearly these are different concepts. If anything, the basic marginalist model implies that competitive firm with excess capacity finds itself on a flat portion of its marginal cost curve, so its supply curve is horizontal and fluctuations in demand should have no effect on prices.

    2) So if this is not really a marginalist argument, what kind of argument is it? Economists who have actually studied the relationship [2] seem to think that it’s what Keynes claimed:

    The clearest early exposition of the relationship between the intensity with which resources are used in production and changes in the price level is provided in John Maynard Keynes’ General Theory of Employment, Interest, and Money. In his treatise, Keynes postulated that the price level was tied directly to the cost of production and that production costs, in turn, were linked to the intensity with which factors of production — labor and capital — were used. […] As demand increased, more and more industries would find themselves at full employment, and any further increase in demand would merely cause an increase in the prices they charged. Thus, as the economy as a whole got closer to fully employing labor and capital, prices would increase at an accelerated pace as aggregate demand increased. In other words, higher levels of capacity utilization would imply an increasingly higher price level.

    Isn’t it ironic?

    3) Long-run relationship can be tested by filtering out high-frequency movements from the data, or estimating cointegrating equations. Adjustment speeds can be studied by calculating impulse-responses and look at how fast variables return to their long-run values in response to shocks. Plenty of nontautological things to do, provided one is willing to learns some tools. Which doesn’t seem to be the case. But let us not address such issues here.

    [1] http://www.federalreserve.gov/RELEASES/G17/CapNotes.htm
    [2] http://www.philadelphiafed.org/research-and-data/publications/business-review/2005/q2/Q2_05_Dotsey_Stark.pdf

    • (1) I can use unemployment if you’d like. I promise I will get the same results. Don’t be obfuscatory. You know that prices do not decline when demand falls.

      (2) This is AS-AD, not Keynes. Keynes did not believe prices adjusted when demand fell. Nor did he think that this would soak up excess capacity.

      (3) Okay, where do the price falls show up in this long-run? It looks to me that if I filter the CPI I will still never see these price adjustments… I am getting the sense that you are not tackling the above argument head on… perhaps because you know I’m right?

  7. Rob Rawlings says:

    (WordPress seems to be eating my comments, apologies if I am in fact posting duplicates)

    As we can see, we have a very tight positive correlation between inflation and capacity utilisation. But all we see is a fall in the rate of inflation. We never see actually see a fall in prices (i.e. deflation) and as we have said: for the marginalist argument to work prices have to actually fall when demand dries up and capacity utilisation falls.”

    Aren’t you ignoring the fact that the central bank was targeting inflation during this period ? In other words whenever capacity utilization showed a tendency to fall, then this would be be reflected in a fall in the inflation rate. The CB would then reduce its target interest rate and bring the inflation rate back towards target. Once you factor this in, then that chart shows exactly what you would expect if prices were set along the lines described by marginalist theory (with price stickiness factored in), and in parallel the monetary authorities carried out counter-cyclical policy.

    • CB intervention works through demand stimulation. Catch up, Rob. Think your arguments through.

      We are trying to show what happens to prices when demand falls.

      • Rob Rawlings says:

        You say “CB intervention works through demand stimulation”. But that’s exactly the point ! Its this demand stimulation that prevents the inflation rate falling further or going negative.

        I believe it is you who has not thought through your arguments.

        Your chart shows that when capacity utilization falls (ie demand falls) the rate of inflation falls. However the rate of inflation rarely goes negative.

        The could be because either
        1) marginalist theory is wrong
        or
        2) something else happens to prevent deflation that fits into the marginalist model

        I have suggested that this “something else” is counter-cyclical monetary policy. You respond by stating the obvious (“CB intervention works through demand stimulation”). This is simply dodging the issue.

      • No Rob. You’re not getting it. The CU measure takes into account the Fed intervention. Come on, man, think about this.

      • Rob Rawlings says:

        “The CU measure takes into account the Fed intervention”

        At any point in time CU will take into account current levels of Fed intervention. But if the CB is targeting inflation and , and low CU causes inflation to drop below target, then the level of intervention will increase, and (other things equal) CU will increase as a direct result.

        BTW: your statement “for the marginalist argument to work prices have to actually fall when demand dries up and capacity utilisation falls.” shows a disturbing ignorance of both monetary and marginalist theory.

      • The last comment makes no sense. And if you think it does then you’re beyond help.

    • Pontus says:

      Rob,

      You’re obviously right here. No marginalist theory would predict that prices would fall relative to zero. Price would fall relative to the counterfactual. What is the counterfactual? If the CB is an automaton, then the counterfactual would be a growing trend of prices. What we then would expect to see is a slowdown in inflation, not deflation. If the CB has an inflation target, the situation gets even hairier: With a competent (or omnipotent) enough CB, there wouldn’t be a slowdown in inflation at all. The truth is somewhere in between.

      Anyway, CPI is a bad measure here. I believe the PPI is better. And the theory performs quite well I must say.

      http://research.stlouisfed.org/fred2/graph/?g=Fer

      • (1) No. Marginalist theory says that consumer prices will FALL when demand for consumer goods FALLS. The evidence is clear. They do not. Case closed.

        (2) PPI is not a kosher measure because when capacity utilisation falls demand for investment and intermediate goods falls as machines are turned off and inventory builds. We’re looking to see if low consumer demand leads to lower consumer prices. It does not. So market obviously cannot clear.

        (3) Even in the PPI it is clear that prices barely decline at all in many periods. What we see when capacity utilisation falls is disinflation in many periods (exceptions are 2001 and 2009 recession). But the theory would predict DEFLATION. Even here the theory doesn’t hold up very well.

      • Rob Rawlings says:

        “No. Marginalist theory says that consumer prices will FALL when demand for consumer goods FALLS. The evidence is clear. They do not. Case closed.”

        This is total rubbish Phillip. You are disregarding monetary factors. Aggregate demand may be falling at the same time that the value of money is also falling. The net result may be that nominal prices continue to rise even while AD falls. There are many reasons why the value of money may fall, including that the CB is targeting a non-zero inflation rate.

      • Your entire comment is a tautology that doesn’t seem to realise that a fall in the value of money has effects on aggregate demand (duh!). If the value of money falls then aggregate demand also falls. And this is registered in any measure of aggregate demand. Again, duh! So when you say:

        Aggregate demand may be falling at the same time that the value of money is also falling.

        You are actually saying:

        Aggregate demand may be falling at the same time that aggregate demand is falling.

        It’s actually rather hilarious if you read it back in that light and it confirms what I have long suspected: you have a very watery grasp of basic economic concepts. I think Pontus registered this confusion too. But then he manifests similar idiocies on a regular basis. You people, I swear! Ivansml didn’t fall into the idiot trap. But he’s long proved to be the most able defender of marginalist economics on here.

        Basically you’re mistaking “inflation” as a causal explanation of the data when rather it has to be an effect that requires explanation. Perhaps invoking well-timed supply shocks that hold prices up despite demand shortfalls every time there is a recession might work? I’m not sure if any rational person would believe such a deus ex machina. And why then would the PPI not register this supply-shock… hmmm. But you’re welcome to the explanation. It’s on the house.

      • Rob Rawlings says:

        “If the value of money falls then aggregate demand also falls”

        You realize you are saying that inflation causes AD to fall, right ?

        “Basically you’re mistaking “inflation” as a causal explanation of the data when rather it has to be an effect that requires explanation”

        I’m saying that in a regime where the CB is targeting inflation its is kind of weird to draw any conclusions about marginal price theory from the absence of a fall in nominal prices.
        I’m saying that the whole premise of your post is seriously flawed.

      • You realize you are saying that inflation causes AD to fall, right?

        Ceteris paribus it obviously does. If wages are 100 and prices go from 100 to 200 real demand falls. Duh. Maybe you should read up on the Pigou effect or, I dunno, learn actual economic theory rather than whatever Austrian rot you pick up online.

        I’m saying that in a regime where the CB is targeting inflation its is kind of weird to draw any conclusions about marginal price theory from the absence of a fall in nominal prices.

        Not remotely.

        I’m saying that the whole premise of your post is seriously flawed.

        And I’m saying you’ve showed time and again that you can’t string together a coherent argument.

      • pontus says:

        (1) No it does not. Link or STFU. Marginalist theory says that consumer prices will fall relative to the counterfactual path that would have emerged in the absence of the fall in demand. Even the most standard DAD-SAS model predicts this (which is just AD-AS in differences). You’re barking up the wrong tree (again and again and again).
        http://en.wikipedia.org/wiki/DAD%E2%80%93SAS_model

        (2) I brought up PPI as I thought you were talking about demand for investment goods. Then this is the relevant price. If you’re talking about aggregate demand more broadly I suppose CPI inflation will do.

        (3) Again, no theory predicts inflation. Link, citation, or whatever to substantiate this. Or it did not happen.

      • Regarding the DAD-SAS what it predicts all depends on how you calibrate the model. It is well-recognised that a DAD-SAS can generate deflation with a substantial demand shock. (See here). The theory does not ‘predict’ anything outside of how you calibrate it. It also assumes stable inflation/growth paths in the ‘long-run’ which is a concept that I criticise in the piece (over your head, I assume…).

        Your version of argument is very typical of the current generation of quote-unquote ‘economists’ that universities are churning out. You think that showing me a (neo-Keynesian) model (that assumes sticky prices because its neo-Keynesian and not pure marginalist you dolt!) that can generate the process shown in the data is proof that marginalism is right. Well, you can calibrate such models to generate a lot of things. That’s not the point. If that were the point then you could never be either right or wrong. Because any time I show you data you can just fiddle with your model to produce the result. That may make you feel very special altogether (“look mom, my model looks just like the real thing!”) but the rest of us aren’t fooled.

        You really must learn to distinguish between an economic argument and throwing around models that can be calibrated in a multitude of different ways to produce any number of different results. That way you might be able to actually put forward economic arguments and we won’t have a repeat of the PKSG incident. Just a thought.

        NOTE: Here is how you would tinker with the model to produce whatever result you please.

      • pontus says:

        Correction last part: No theory predicts DEFLATION.

      • Rob Rawlings says:

        Historically it is easy to list many examples of economies where an increase in the inflation rate leads to a boost in AD. Can you provide any examples where a change in the inflation rate and AD are negative correlated, which you claim should be the norm.

      • pontus says:

        Seriously Phil, you must have some cognitive disabilities.

        I have said over and over again that marginalist theory does NOT say that “[…] consumer prices will FALL when demand for consumer goods FALLS”, or in particular that, according to the theory, “[…] we should not see inflation slow but rather we should see deflation.” In fact, these are your quotes and they are all wrong (as you just admitted). I also told you to provide a link where your claim that “we should not see inflation slow but rather we should see deflation” or STFU. You did neither. Shame on you.

        As for the DAD-SAS model being not *pure marginalist you dolt* is a red-herring. First of all, a “pure marginalist” model is your own weird construct which you seem to use as a silly straw man (wouldn’t expect nothing else from you when cornered; being intellectually honest is not really your trademark). And second, there is simply no marginalist model that would predict what you claim: demand shock –> deflation. Again, cough up the evidence or stop putting words in other peoples’ mouths. However, I find it somewhat cute that you had to google some lecture notes to actually get this point.

        As for the argument that the DAD-SAS model is too versatile to avoid any type of criticism is just plain stupid. It does predict a deceleration in the price level with respect to a demand shock, but not deflation as the only outcome (plus a multitude of other things which are largely confirmed by the data).

        But this whole discussion does make me curious: I can only draw the conclusion that post-keynesian thinking actually preclude deflation as the result of a severe enough demand shock. If that is so, please let me know and I’ll debunk that asap. If not, please tell how you’re going to get out of this mess.

        As for the PKSG: Why don’t you suggest to give a presentation of your own. Perhaps about your “General Theory of Prices” (you couldn’t come up any less modest title, I assume)? I can actually put your name forward if you wish. It would be a real treat to be the discussant.

      • Philip says:

        Marginalist economic theory tells us that when there is unemployment of capital resources prices should fall. Some marginalists like the New Keynesians and the neo-Keynesians will supplement this by saying that prices can tend to be ‘sticky’. Let us ignore these for a moment and come back to them in a moment.

        Pontus says:

        Dur Phil, you are so wrong about marginalist theory! Take a look at this neo-Keynesian/New Keynesian model which, in assuming that prices are sticky in the short-run, will totally blow your argument away!

        I wouldn’t want you as a discussant because you can’t discuss. Years have modeling have resulted in you not being able to listen to arguments properly. You don’t digest. When you consume an argument you just puke it back up and leave the rest of us to clean up the mess. It’s embarrassing frankly.

      • pontus says:

        Phil,

        The issue is that what I’m saying is true in every marginalist monetary model, sticky prices or not.

        But again, please quench my curiosity: Do I understand you correctly that Post-Keynesian economics predicts that there will NEVER be deflation in response to a severe enough demand shock? Because either you will have to stand on that foot, or you will have to admit that PK theory is a vacuous as the marginalist you’re attacking. Which one is it?

      • Strange that you provided a neo-Keynesian/New Keynesian model when I expressly said that I wasn’t dealing with this then. Hmmm… one would be forgiven for thinking that most of the time you don’t even know what other people are saying. This probably goes a long way toward accounting for the fact that beyond manipulating models you often don’t have the least clue what you’re talking about. What’s that phrase? Those who can, do; those who can’t, teach. Seems apt.

      • pontus says:

        The DAD-SAS in the flexible price setting is given by a vertical SAS curve (that is, when output is always equal to its potential). The result is still the same. This is becoming disgraceful.

    • Thorstein says:

      I think you are right on this Rob,

      “The could be because either
      1) marginalist theory is wrong
      or
      2) something else happens to prevent deflation that fits into the marginalist model”

      But to me, it just shows the problem of stating the relation as holding under the “everything else being equal”. Or at least, the problem that if a marginalist claim is in difficulty, there is always the possibility of invoking that not everything has remained equal during the period. So, Phil is entirely true when he insists that it is non-falsifiable. Or, I prefer to say : nearly impossible to falsify when confronted with someone with very entrenched marginalist preconceptions.

      • No. The CU measure takes into account Fed interventions. I’m surprised that people would think otherwise. Fed interventions work through demand stimulation. CU measures demand. The idea is to see if falling demand leads to falling prices as the marginalist argument says. It clearly does not. Case closed.

      • Thorstein says:

        “No. The CU measure takes into account Fed interventions.”

        Yes, sorry you are right. My example was misleading, but what about the argument?

        If “The idea is to see if falling demand leads to falling prices as the marginalist argument says.”, I was saying that marginalists add the ceteris paribus clause. So the argument, at least as I learned it at University, is not that prices will fall when demand falls, but that prices will fall when demand falls, everything else being equal. In a sense, they are trying to make a rough causal relationship. So, is a typical marginalist right to point out that you did not falsify that “causal” relationships because you did not control for all other possible factors in your analysis? In short, rejecting your quick and incomplete correlation analysis, which stays at the level of events and so unable to falsify what they consider as a deeper “causal” claim.

        I’m not supporting marginalism, just that I came across exactly the same debate in the past and that typically, confronted to these kind of data and correlation, some “exogeneous” supply shocks are invoked in order to explain prices behaviour and to save the initial claim.

      • From a methodological point-of-view you are correct. Marginalists can always invoke ceteris paribus assumptions. They are not engaged in science in the Popperian sense. This has long been noted. I sat on a panel with the former president of the British society for the philosophy of science and he made this point.

        But even in this case it is quite unclear what the “disturbing” factor that is supposedly causing the price dynamics in the above correlations is. Rob and Pontus are trying to define it above but they’re talking nonsense. They’re invoking inflation. Then they are saying that this inflation is caused by CB intervention. But the CB intervention is working through demand stimulation (lower interest rates). When this is pointed out Rob starts confusing a falling value of money with some sort of causal argument. At this point you get to what I’ve long suspected: marginalists cannot actually make coherent economic arguments. They can only construct woolly thought experiments.

  8. Be A Debaser says:

    The wikipedia quote is hilarious, like something from an Onion article on the state of modern Economics.

  9. YC says:

    If this argument is true then we should see prices fall when capacity utilisation falls. We should clear here: we are not saying that we should see inflation slow down when capacity utilisation falls. Rather we should see prices fall — i.e. we should not see inflation slow but rather we should see deflation.

    In a hypothetical barter economy, perhaps. In a monetary economy with CB that targets inflation, no.

    Now ignoring this simple aggregation proclems and about the perhaps, prices are indeed sticky. This is an empirical fact and there are ample of reasons for it. DSGE models sticky prices via imperfect competition plus menu costs while other models, IMO the more important ones, explain price stickiness via adverse selection.
    About “margininalism”, no idea what that is supposed to be. You think that it means that capital utilization and prices should be positively correlated, well, no economist I know entertains such a position. In the short run prices are rigid so a decline of capital utilization does not neccessarily lead to falling prices.

    If “marginialism” is supposed to mean that agents maximize some objective function, well, this is just common sensical unless you think that it is literally true. If you think that agents maximize some objective function but not all the time, i.e. they tolerate small deviations from the optimum as they are not computers but people who use simply heuristics, well, then you arrive at Yellen & Akerlof’s concept of near-rationality (if I remember correctly they wrote 2 or 3 papers about it in the eighties) which, not incidentally, can explain price stickiness. I always welcome such interdisciplinary approaches, especially when they explain basic macro issues.

  10. Rob Rawlings says:

    This isn’t very complicated Phillip.

    – Your post claims that the failure to observe falling prices when CU falls (which you take as a proxy for AD) disproves marginal theory.
    – However all that marginal theory would claim is that (other things equal) a fall in AD will lead to a fall in the rate of inflation. (In extreme cases this may , but does not necessarily have to, lead to negative inflation.)
    – A falling rate of inflation when CU falls is exactly what we see in the last chart you show

    Far from disproving marginal theory your analyze provides support for it !

    To assert as you do that “for the marginalist argument to work prices have to actually fall when demand dries up and capacity utilisation falls.” is simply wrong. You and your readers should be aware that you are attacking a straw man here.

    • If demand falls then employment falls. In marginalist theory — which involves the assumption that markets clear — it is assumed that wages and prices will DECLINE to reach a new equilibrium. If prices do not fall then the economy will not tend toward a full employment market-clearing equilibrium. I’m sick of explaining macro 101 to you and that AS-AD models are neo-Keynesian (i.e. short-run unemployment equilibrium and thus not what I was addressing in the above post) to Pontus. This blog comments section used to be high quality. It has turned into a playground in the last six months.

    • Rob Rawlings says:

      Well , I’m starting to think that your problem may be that you never got much beyond Econ 101, which is why you seem to struggle with even slightly advanced topics like the fact that a slow down in the rate of inflation can be similar in its effects to an actual fall in nominal prices as far as many economic theories are concerned. Your last comment shows a lack of understanding of something quite simple (real v nominal wages and prices).

      I doubt I’ll become a regular reader of your blog based on this post, and the uncivil manner you seem to cultivate – so enjoy your playground.

      • philippe101 says:

        Phil,

        “In marginalist theory — which involves the assumption that markets clear — it is assumed that wages and prices will DECLINE to reach a new equilibrium”

        Don’t you mean real prices and wages, rather than nominal prices and wages?

      • pontus says:

        Philippe, there is no such thing as “real prices”. The real price level is always and everywhere equal to one, and the rest are just relative prices.

      • You’re clueless Rob. You always have been.

  11. A H says:

    Weird that none of the defenders of orthodoxy have noted the real reason mainstream theory says prices don’t decline in mainstream models is rational expectations. A temporary change in capcity utilization in a marginal world with rat-ex would not neccessarily lead to falling prices, since the agents in their infintie wisdom would see into the future and adjust prices accordingly.

    So I think Phil is wrong to say that this is knock out evidence against contempary marginalism. But of course Rational Expectations are ridicoulus in themselves.

  12. Lincoln's economics says:

    Corret me if I’m wrong on marginalism, which as I understand Keynes did not fully reject. In short, marginalism amounts to a reversed causality of the classical school. In the classical school (Smith-Ricardo-Marx) wages drove the final price of the good, though flawed in terms of a subsistence wage (story for another day). In marginalism the logic is reversed: a market price is established which in turn which imputes a wage. If this is the case, then I would like to hear a marginalist explain how the market price of an automobile consisting of 14K parts comes about. A consumer has absolutely no concept of what the final cost of production of a car should be, nor should he. His demand and emotional desires are irrelevant to price formation. The problem with marginalism as I see it that it amounts to barter with money (QTM) sprinkled on top to achieve a price level. This is the reverse of reality. Money is nothing more than a definition of a base labor unit, by decree through the minimum wage. Wages are not just sticky, they are rigid for a reason: to track productivity gains. It is a necessity. If wages were flexible, then an inferior competitor cut use a pay cut to gain an advantage over a superior competitor (superior productivity). This would defeat the entire growth process of an industrial economy. In the end, the workers who build the car will always have the purchasing power to buy it in a non-marginalist model. A true cost of production model (modified classical school) makes it so. Quantity adjustments, not price adjustment rule the roots in such a model.

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