Thursday, May 26, 2011

Orthodox vs MMT on Method

Those coming to MMT/Post Keynesian economics from the non- or at least less-academic world may not be aware of some of the methodological differences that exist between our approach and that of mainstream economics. These philosophical issues, though not nearly as exciting as talking about Federal Reserve policy or the organization of the financial market, truly lie at the heart of our disputes. One should, therefore, have at least have a passing familiarity with these in order to understand the rhetoric of the arguments (and how to win!).

Method involves how a particular school of thought views the nature of knowledge and the legitimate means of creating it. Say you are a biologist and want to understand more about the behavior of the field mouse. How should you do so? Do you imagine yourself as a mouse and decide what you would do in your daily life? Do you catch a field mouse and put it in your office, then write down what you observe? Do you set up cameras and try to capture images of the field mouse in its natural setting? Which of these would be most effective? These are the sort of questions that those who study method try to answer.

Volumes have been written on this subject, but I want to focus on a smaller issue. To put it into a context, imagine an argument:

Premise 1: All Texans own a cowboy hat.
Premise 2: Alex is a Texan.
Conclusion: Alex owns a cowboy hat.

This argument is valid if the conclusion is supported by the premises. Sure enough, that looks okay. Note that determining validity is a relatively (though not completely) objective process. Mathematics provides one means of determining validity, which is one of the reasons it’s so popular in economics. And while we can most certainly enter into some serious debates with Orthodox economists over validity, I think the next concept is more important because it’s more below the surface.

Whence come the premises? What is appropriate/acceptable in specifying them? This is a key methodological question. In mainstream economics, they have adopted the Cartesian tradition that argues that it makes more sense to base premises on reason than observation since the latter can be unreliable and biased by our preconceptions. If theory determines what you see, then using the latter to confirm the former is only a psuedoscientific exercise. True science involves basing the premises of your model on careful introspection and reflection. This is why Neoclassical microeconomics is based on concepts like, “all firms are short-term profit maximizers.” Did some economist do a study or survey to determine this? No. Instead, it just seems reasonable. More than that, they regard it as obviously true. How could any reasonable person disagree with this? The fact that studies show that firms are more concerned with sales and market share does not concern them. What people say and what they do are not the same. Besides, even if firms don’t try to short-term profit maximize, say Mainstream economists, they act as if they do. This is a critical methodological point. They believe that knowledge gained from introspection and the generation of “obviously true” premises is superior to that based on observation of the real world.

Post Keynesian economists come down on the opposite side of that issue. To them, premises that cannot be justified on the basis of something we can observe in the real world are questionable, at best. Of course, simplification is necessary for modeling, but the guide to what the non-simplified version of the premise would be is provided by observation. In addition, we believe it vital to be able to point to a real-world line of causation to justify our claims. Neoclassicals do not necessarily agree, thinking instead that analogies and thought experiments suffice–indeed, they see them as superior because they are supposedly uncontaminated.

And so this points to a deep divide between Post Keynesians/MMTers and orthodoxy. And it can be an especially frustrating one because it lies beneath the surface and leads to one side completely dismissing what the other views as “evidence.” For example, I ended up in a mini debate with a Monetarist on my Forbes.com blog. He was arguing that inflation is caused by an excess of money supply over money demand. I asked–on four separate occasions and in a very pointed manner–how can this be possible? How can the Fed raise money supply without the cooperation of the public? Can they make someone sell a government security or take out a loan? What mechanism does the Fed have at its disposal that allows it to force the money supply above money demand? I never received a direct answer. Instead, I got analogies: monetary policy is like an orchard or a fire place (Friedman’s classic mechanism being a helicopter). The Monetarist has now completely withdrawn from the conversation without ever clarifying this issue and yet at the same time stating, “If you cannot increase the money supply above money demand, you would never have inflation!” Hence, he believes this to be a critical point, but one he cannot justify in terms of real world events. He doesn’t think he needs to.

I won’t go into the philosophy of which position is correct other than to say that while I can see the merit of both approaches to some extent, ultimately reason is every bit as vulnerable to a priori contamination as observation and if the latter is not to be our ultimate guide to “are we getting it right?” then what is? How on earth can you put forward as your explanation of a key economic phenomenon a process which cannot occur in the real world? I just don’t see it. And I’m not sure they do either, in reality. I think it was very embarrassing to my detractor that he could not come up with an answer. In fact, at one point he stated, “Our postions are so diametrically opposed that I dońt think this conversation will lead ‘somewhere.’”

Quite right. I think economics should be relevant, and he doesn’t.

73 comments:

Anonymous said...

Theories can be evaluated by their predictive power. By this measure, mainstream economics should have been revised or discarded. The fact that it hasn't indicates there are ulterior motives for its continued existence.

NeilW said...

A classic case in point recently.

A solid argument that leveraged firms encourage price stickiness:

http://is.gd/CqACGS

which is then simply not backed up by the data

http://is.gd/CGXBkd

And that is the case across economics. Really sensible arguments that simply are not predictive, nor are they backed by any real world evidence.

MMT isn't immune. I'm still waiting for the evidence backing 'firms will quantity expand rather than price expand', plus evidence that floating exchange rates are stable.

Septeus7 said...

I also have a questions on floating exchange rate idea along the lines of Neil's observation.

MMT seems to assume a certain definition of sovereignty that assumes 3 things must exist for a currency to be Sovereign.

1. The central government has the power to levy taxes at the national level.

2. A central government is a fiat monopoly issuer of the public currency.

3. A floating exchanging rate.

However, I must think there are some undefined parameters I haven't seen addressed by MMT.

First, if I understand it correctly, a sovereign nation controls the value and issuance of it's currency and therefore a floating exchange rate is assumed necessary because in a system of floating exchange if you link you currency directly another currency then you can loss control of the value of your own currency because you no longer have exclusive control of the demand for the currency based on your nation's individual economic standing but the demand is now directly linked to a another countries policies.

Perhaps I am mistating the reasoning but it seems to depend on several assumptions. While a fixed partity is in a system of floating rate has obvious disadvantages however it seems that MMT ignores two issues.

1. In a fixed rate system where all the countries have partity agreements with each ala Bretton Woods I, there would be no room for currency speculation to cause rapid currency appreciation then follow by a rapid depreciation because the all the liquidity in the system is tied to a central institution such as the United States so attack on one currency was an attack on all the currencies and if someone started to try take advantage of the slight fluctations a nation could simply re-adjust the rate. In the "Predator State" Galbraith says "fixed exchange rates simulate to a degree, the security of operation with a single unit of account." I'm assuming Professor Galbraith knows what he is talking about and I have heard other make the same point.

How then is a floating exchange rate required for Sovereignity in a system another other than a floating exchange rate system. I seems that a nation could be perfectly Sovereign and perhaps even more so within in fixed rate system as long as it maintained the power of re-adjust on it's currency on a treaty basis rather a market one. Am I mistaken?

2. Assuming that a floating exchange rate system is how natural "law of the universe" (which I doubt) it would seem to me that a monetary sovereign would require more than just having a floating rate.

If a National Government was unable to place capital controls from influxes of foreign speculators because of a "free trade" agreement then couldn't it be possible so speculator to aquire and then dumb currency from circulation faster than any government could logically adjust tax policy and thus by manipulation of the velocity of the currency?

Essentially, doesn't a "free" floating system where capital controls over the lending windows and capital influxes mean that in the short term speculator could potentially act as competing currency issuers by reducing private sector holding denominated in the currency causing a panic sell off and using covert political leavage and disinformation campaigns to keep the Government from being able to respond with the poper policies?

I fail to see how a government without the ability to put on capital control could protect itself from what is essentially a pump and dumb.

I therefore thinking MMT needs to clarify that a monetary sovereign must also have the authority and ability to create and use capital control and a "Controlled" floating rate system is required not a "Free" floating rate system. Perhaps I am missing something and Mike might be able respond as I lack the technical knowledge of market function and monetary institutions to understand where my reason is wrong if it is wrong.

sparc5 said...

@Neil- can you give the context of 'firms will quantity expand rather than price expand'? On the surface it looks to me like it depends on the degree of monopoly the firm enjoys but I can't get too specific without knowing the context.

Floating exchange rates aren't stable, they float by design. Any currency is inherently risky, and those risks are usually manageable ones.

Tom Hickey said...

@ Laura.

Right. This is a signal that mainstream economics is a value-laden ideology masquerading as positive science.

Tom Hickey said...

@ Neil

1, "MMT isn't immune. I'm still waiting for the evidence backing 'firms will quantity expand rather than price expand',

The assumption is that in aggregate firms would chose to increase production rather than lose business through demand destruction, and that price increases result chiefly from increasing marginal costs. This presumes a reasonably competitive market, which may not be the case in many key market sectors. So I wonder about this, too. Seems that strong inflationary pressure can develop well before the point that MMT seems to presume.

2. "plus evidence that floating exchange rates are stable."

What would such evidence have to look like to satisfy you? I see a lot of worry out there about the wrong things. But suppose that the EZ comes undone, which is a distinct possibility, and many MMT proponent encourage this as a solution. What happens to the euro and how does that affect the currency market.

STF said...

"I'm still waiting for the evidence backing 'firms will quantity expand rather than price expand',"

See "asking about prices" by Blinder. also, 2 things--

First, it's not MMT, per se, but pretty well-grounded in PK models and empirical analysis. Also by Institutionalists (lots of research on cost-plus pricing there).

Second, we don't actually say this is necessarily the case--case in point is the JG, which sets a fixed wage to COUNTER the fact that bottlenecks exist in many markets and as an alternative to traditional Keynesian pump priming precisely because of problems associated with bottlenecks.

"plus evidence that floating exchange rates are stable."

Again, we don't say this. We say that regardless of fluctuations, the govt can still use a JG to create full employment.

Hugo Heden said...

The following is my understanding, as an MMT layperson.

Yes, there is that argument saying that firms, when facing a demand increase, are more likely to expand output, up until their capacity limit, than prices. They do this in order to not lose market share. This is not an MMT-claim per se, but actually a pretty standard mainstream idea.

Should there be slack in all sectors of the economy, ample research -- see STF:s comment -- show that unemployment could be decreased using standard Keynesian fiscal tools, "pump-priming", without too much inflation. At some point though, some sectors will reach bottlenecks, such that they will rise prices rather than increase output further. This will happen long before the point of full employment.

Note therefore that the claim that "in the context of a general aggregate demand increase, firms will expand quantity rather than prices" does NOT imply that FULL employment is reached before prices start to increase. That is NOT what is being claimed.

Of course, MMT argues that general aggregate demand management, Keynesian "pump priming", is a rather blunt tool. The idea with this indirect general demand stimulus is that it will to "trickle down" via spending multipliers, to eventually create new jobs.

Labor from all levels of the productivity spectrum is hired (sometimes the most technically proficient ones). The government competes for these workers in the market with the result of bidding up wages and prices, and taking away these workers from private sector activity.

A JG program could achieve full employment without added inflationary pressure. There is no indirect general demand stimulus intended to "trickle down" via spending multipliers - the unemployed are employed directly, hired from the "bottom". At the point of full employment, JG raises aggregate demand no further. This is all a result of automatic policy. From there, the Job Guarantee would in fact act to stabilize prices for various reasons elaborated on elsewhere.

Hugo Heden said...

Sorry, when I write JG I mean "Job Guarantee", the program prescribed by MMT.

vimothy said...

"They believe that knowledge gained from introspection and the generation of “obviously true” premises is superior to that based on observation of the real world."

Really? Because I can't imagine anyone agreeing to that statement. (Other than hardcore Austrians).

Where is your evidence that neoclassicals actually believe this?

Tom Hickey said...

I think that what John is saying is that this is actually what mainstream economists are doing, although they may not admit it or even realize it. Austrians are up front about it, as you note.

I am a professional philosopher, and I would agree with John broadly speaking. One thing that philosophers do is to uncover hidden assumption and unrecognized bias and influence. What has struck me about economics is that I see very little critique and almost no attention paid to what is relevant in the life and social sciences. Very little work has been done in the philosophy of economics to examine the foundations, either.

Basic principles like utility maximization and rational agency are presumed, for example, even though these are areas that have been controversial and disputed for thousands of years. Moreover, a lot of light ha been cast recently by scientific research. For the most part, economists seem to ignore this work. I would say at their peril, because the way these assumptions are stated seems to me to be simplistic, given the relevant literature in other fields. And I am not the only one who thinks this.

beowulf said...

Seems that strong inflationary pressure can develop well before the point that MMT seems to presume.

Agreed. That's why Bill Vickrey's plan for a anti-inflation market targeted gross markups instead of prices. A company could still meet its revenue targets (while avoiding the tax) by selling greater volume at lower prices. The increase in aggregate supply would reduce demand-push inflation while the tighter gross margins would reduce cost-push inflation. And just as a cap and trade carbon market has a functional equivalent in a carbon tax, you could set up a Vickrey ant-inflation tax instead.
http://traderscrucible.com/2011/04/21/yglesias-and-mmt/#comment-350

STF said...

Well said, Tom. There actually is a fairly thriving philosophy of science literature for economics, but it's a highly marginalized field, perhaps even more than PK. In fact, it's generally the heterodox groups that do philosophy of science, as by their nature they find neoclassical economics to be lacking in appropriate methodological foundations and not nearly as scientific as it thinks it is. Some of this can turn into a bit of bashing neoclassicals for the sake of doing it, or for the sake of being marginalized, but some of it is quite good.

STF said...

Good, beowulf, though I'm still trying to understand how anyone actually well read in MMT thinks that's what MMT says anyway.

If one is looking for precision, then read the research, and there it seems rather clear that we don't think that, otherwise we wouldn't find traditional Keynesian pump priming so problematic (for instance); we've always said that the traditional Keynesian approach creates bottlenecks well before true full employment is reached.

vimothy said...

So you have a proposition: Neoclassical economists believe that knowledge gained from introspection and the generation of “obviously true” premises is superior to that based on observation of the real world.

Did John Harvey do a study or survey to determine whether neoclassical economists do actually believe this? No. Instead, it just seems reasonable. More than that, he regards it as obviously true. So true all that’s required to establish it is a single anecdote.

Which is, if nothing else, rather ironic.

“I think that what John is saying is that this is actually what mainstream economists are doing, although they may not admit it or even realize it.”

Talk is cheap—isn’t that supposed to be the point? Where is the evidence to support this? I see a massive neoclassical applied literature. Econometrics modules make up at least a quarter of the taught sections of the MA and MSc programmes where I study. It’s certainly not self-evident that Harvey’s proposition is true.

Given the nature of the accusation—neoclassical economics doesn’t think real world evidence is important; MMT/PKE does—I think it’s reasonable to ask for some empirical proof. Otherwise, it looks a little bit like you’ve got the proposition back to front and this is just transference.

“What has struck me about economics is that I see very little critique and almost no attention paid to what is relevant in the life and social sciences.”

But you’ve said before that you don’t read any of the literature and you’re not interested in doing so, so why is it remarkable that you haven’t seen any of anything?

Things like utility maximisation and rationalism are modelling strategies. Are they meant to be ontological statements about reality? No. Are they simplifications? Yes, of course. Are they less realistic than the behavioural assumptions made in Godley and Lavoie’s models, which you championed elsewhere? No.

STF said...

Vimothy,

There's quite a bit of academic research to support the general thrust of John's point (Mirowski, Davis, Caldwell, Hands, Dow, McCloskey (kind of), etc., etc.). If this were an academic forum, obviously, some of it would have been cited, but John's been writing in the area of PK long enough he's already probably done that someewhere. (Unlike neoclassicals, critical evaluation of methodology is standard in PK, again, though, sometimes (though not in John's case) in a less helpful way that further marginalizes.)

Some of the research to this effect is even by neoclassicals. Larry Summers, for instance, did a well known paper back in the late 1980s arguing (with many citations) that econometric evidence appears to have almost no influence on the views of economists.

Tom Hickey said...

vim, I have been looking for mainstream philosophy of economics that examines foundations, and I have found practically nothing on it. I would appreciate it if you could point me in the right direction here.

Tom Hickey said...

"Econometrics modules make up at least a quarter of the taught sections of the MA and MSc programmes where I study."

It's assumptions, explicit and implicit, that are in question, as well as the meaning of key terminology. For example, are assumptions actually testable hypotheses or are they intuitional? Does meaning shift of key terminology shift, resulting in logical jumps? These are the kinds of questions that philosophy of economics asks.

John Harvey said...

*****PART ONE (apparently, my comment is too long!)*****

VIMOTHY: “Did John Harvey do a study or survey to determine whether neoclassical economists do actually believe this? No”

Actually, that’s incorrect. First off, I have a PhD in Neoclassical economics. Almost every single course I took as an undergraduate and graduate was in Neoclassical economics and my dissertation was a Barro and Grossman disequilibrium model adapted for the open economy. Furthermore, I’ve been teaching Neoclassical economics for 25 years. Not exclusively, of course, but one cannot avoid doing so in the current climate. And last, as STF suggested, I have, indeed, published papers discussing this (primarily on exchange rate theory). In my current research project, a book on different schools of thought in economics, methodology is a central concern. So you are arguing from a false premise. I did not simply assume this because it seemed reasonable.

VIMOTHY: “Given the nature of the accusation—neoclassical economics doesn’t think real world evidence is important; MMT/PKE does—I think it’s reasonable to ask for some empirical proof.”

Wow, where to start?! Obviously, I can’t review the entire body of Neoclassical literature here, so let me just give a quick example in my area of expertise, international economics. Take comparative advantage, the absolute core of Neclassical trade theory and one of the least controversial concepts in their tool kit. The explanation is based on a thought experiment, a world which does not now nor has ever existed: one in which all countries are in autarky and produce all goods and services domestically. The center piece of the explanation is that the autarkic opportunity cost ratios determine comparative advantage and what each nation exports. But because countries now do trade with each other, there is absolutely no way to test this. It’s a “as if.” Under conditions of complete specialization, we can never know how much output/hour labor could have produced in import industries. Even under incomplete specialization, it is impossible to know what productivity would have been at different levels of output. The fact that Neoclassical international trade theory immediately assumes away the world’s most critical problem, unemployment, creates even more problems. Why do they do so? Because under less-than-full employment, opportunity costs do not exist and so comparative advantage cannot be calculated!

I have barely scratched the surface here, but if you want more reading you can look here:

Prasch, Robert E. “Reassessing the Theory of Comparative Advantage,” Review of Political Economy, January 1996, v. 8, iss. 1, pp. 37-55.

Robert also has an awesome article on Neoclassical method in general:

Prasch, Robert E. “The Origins of the A Priori Method in Classical Political Economy: A Reinterpretation,” Journal of Economic Issues, December 1996, v. 30, iss. 4, pp. 1105-25.

In terms of the PK approach to trade, Bruce Elmslie and William Milberg have been doing work on a theory that focuses on absolute advantage and puts technology and intraindustry trade at the center. No thought experiments or analogies that do not have real-world parallels.

John Harvey said...

*****PART TWO*****

In exchange rate theory, there’s too much to say here. I’ll point you to this article:

Harvey, John T., “Orthodox Approaches to Exchange Rate Determination: A Survey,” Journal of Post Keynesian Economics, Summer 1996, v. 18, iss. 4, pp. 567-83.

And then this book:

Harvey, John T. Currencies, Capital Flows and Crises: A post Keynesian analysis of exchange rate determination, Routledge 2009.

VIMOTHY: “Things like utility maximisation and rationalism are modelling strategies. Are they meant to be ontological statements about reality? No. Are they simplifications? Yes, of course. Are they less realistic than the behavioural assumptions made in Godley and Lavoie’s models, which you championed elsewhere? No.”

I strongly disagree. It’s a reasonable simplification when more realistic premises still lead you to the same general conclusion, but via a longer route. It’s poor modeling when it turns out that a) different premises lead to radically different conclusions and b) it does not take a great deal of effort to discover the nature of the premises that more properly ground the model is real-world behavior. Take for example this absolutely fantastic book that gives a PK micro theory:

Eichner, Alfred S. The Megacorp and Oligopoly: Foundations of Macro Dynamics, Cambridge U Press, 1976.

It’s a bit dated now (which is inevitable with a school of thought that insists that models be relevant to the current institutional structure of the economy), but it’s an awesome example of what micro theory would look like done properly.


In summary, I did not simply assume Neoclassical method. I was trained in it and have studied it for a quarter century. Second, examples are plentiful.

STF said...

"Things like utility maximisation and rationalism are modelling strategies."

If you insert "only" before "modelling strategies," then that would be more accurate depiction. Try getting a theoretical paper on behavior published in a top orthodox journal without them (or at least the first one--ad hoc adjustments to the assumed rationalism via "behavioral economics" are now permissable).

And the key methodological innovation of G/L-type models is incorporating NIPA/FoF consistency throughout. Yes, that's a highly empirical approach to modeling, even if there are admittedly other shortcomings for a rather new field. Such grounding in the actual method of "scorekeeping" of transactions in the real world doesn't happen in DSGE models--these also assume all private debts are paid off, though hip-hip-hooray for the better "behavioral assumptions" these models make. But, again, try saying any of that and getting published in a top orthodox journal.

STF said...

Meant to say "the only."

Seconded on Eichner's book. Well said, John.

vimothy said...

Scott,

I've read some of those guys actually. Mirowski is quite good fun. I used to work for a research project in maths education and, mostly because everybody else drew on Bourdieu a great deal and I like to be awkward, I developed an abiding interest in actor-network theory (developed by STS types like Latour, Callon, Law, etc) and social studies of finance/economics. I’m sure that you won’t have time to follow this up, but if you ever do, there is a great book on the sociology of finance and financial markets that came out of that literature: “An Engine Not a Camera” by Donald MacKenzie.

Of course (as I’m sure you’ll agree), “people agree with me” is not proof that your theory is correct. What specific papers are you thinking of?

Summers paper is interesting (I assume you’re referring to “The scientific illusion in Macroeconomics”). Here’s the abstract:

“It is argued that formal econometric work, where elaborate technique is used to apply theory to data or isolate the direction of causal relationships when they are not obvious a priori, virtually always fails. The only empirical research that has contributed to thinking about substantive issues and the development of economics is pragmatic empirical work, based on methodological principles directly opposed to those that have become fashionable in recent years.”

But that doesn’t seem like support for Harvey’s proposition to me: Summers implies that econometrics has contributed and continues to contribute to thinking about substantive issues and the development of economics. He implies that econometrics itself has developed. And he doesn’t think that one particular methodological development (relating to macroeconomics) is a good one, and so offers a critique.

TBH, OP reads like drive-by snark. “Neoclassical economists are complete idiots” is the not-so-subtle subtext.

vimothy said...

Hi John,

“Actually, that’s incorrect”

Okay.

“I have a PhD in Neoclassical economics... Furthermore, I’ve been teaching Neoclassical economics for 25 years. Not exclusively, of course, but one cannot avoid doing so in the current climate. And... I have... published papers discussing this (primarily on exchange rate theory).”

Fair enough, but what I actually asked was, “Did John Harvey do a study or survey to determine whether neoclassical economists do actually believe this?” A decently designed survey would be an authoritative way to settle this, I feel. But it doesn’t sound like you did. On the other hand, you have good reasons for making the assumptions you make. Don’t we all?

“Take comparative advantage... The explanation is based on a thought experiment, a world which does not now nor has ever existed”.

As is every model.

Take any field that you like. Is there an applied literature? If there is, people must think empirical work is important; otherwise, it wouldn’t occur. You can’t simply point to an example of inductive reasoning and say, “look, they use inductive reasoning; therefore, they don’t think observing the real world is necessary”.

“No thought experiments or analogies that do not have real-world parallels.”

What does that actually mean? Can you give me some examples?

Thanks for the refs, BTW.

John Harvey said...

VIMOTHY: Fair enough, but what I actually asked was, "Did John Harvey do a study or survey to determine whether neoclassical economists do actually believe this?"

JTH: Actually, you said more than that. You also said, that I believe this because it just seems reasonable and obviously true, implying that I was talking off the top of my head and had no scholarly basis my comments. My reply was intended to demonstrate that this was unfounded (as well as terribly condescending--I try to treat everyone with respect and feel I deserve the same in return).

I am, in fact, in the midst of the very sort of study you say would be an authoritative way to settle this (as part of my volume on different schools of thought in economics, forthcoming through Edward Elgar–at some point!). It wouldn’t actually settle it as there is no such thing as a definitive study, but that’s beside the point. First of all, one does not personally need to have done the research in a particular area to legitimately draw upon the insights of others. If that were necessary, every researcher would have to start at square one! Second, even if it were necessary to have done original research in order to be permitted to make observations, I might fairly ask have you done this? Are you published in economic method? I confess that I am not, but at least a) I am in the process of creating new research in this area and b) this required an extensive literature review before I put pen to paper. Scott offered some excellent citations and I highly recommend the Prasch article.


VIMOTHY: (quoting JTH) "Take comparative advantage... The explanation is based on a thought experiment, a world which does not now nor has ever existed"...As is every model.

JTH: Every model is a simplification (by definition). Every model does not, however, create alternate universes where, for example, money supplies are increased by helicopters, fully-functioning, pre-trade worlds exist, everyone has perfect foresight, etc. I already made this argument above and will simply repeat it here:

It’s a reasonable simplification when more realistic premises still lead you to the same general conclusion, but via a longer route. It’s poor modeling when it turns out that a) different premises lead to radically different conclusions and b) it does not take a great deal of effort to discover the nature of the premises that more properly ground the model is real-world behavior.

VIMOTHY: (quoting JTH) "No thought experiments or analogies that do not have real-world parallels."...What does that actually mean? Can you give me some examples?

JTH: I already gave several examples. Heck, just go back to the original post. In Friedman’s classic article, he says the money supply is increased by means of a helicopter. Okay, we all use simplifications. BUT, when the whole argument depends on that, one has to ask if it was truly a simplification? In point of fact, Friedman needs to say that to be able to argue that the supply of money can exceed the demand, making the former the cause of inflation (and implying that inflation is always and everywhere a monetary phenomenon). Now ask a Neoclassical economist how that can happen in the real world. How can the Fed force money on people? They can’t. Thus, when I made this argument on my Forbes.com blog, I got (from a monetarist) analogies, not actual mechanisms.

Why? Because we disagree on proper method. Both sides believe simplification is necessary, but one argues that it MUST BE rooted in something we can see in the real world.

STF said...

Vimothy . . . thanks for the fix on Summers--sounds like a critique of New Classicals to me. Read it almost 20 years ago, at which time they were mainstream macro. It does support John a bit, but certainly in a more narrow sense of that particular approach to macro (important as it is).

John . . . another thing on Friedman is that even the balance sheet mechanics of a helicopter drop are not possible from the central bank, since while a helicopter drop of money would create net financial assets for the non-govt sector, open market operations do not (unless the cb vastly overpays or buys pvt sector liabilities that are defaulted upon--which could happen, but is certainly not what Friedman had in mind, or so it would seem). In other words, central banks would not be able to do helicopter drops even if they wanted to--this is part of a bigger problem that many have in equating money and income.

STF said...

And thanks for the MacKenzie reference, Vimothy.

vimothy said...

“You also said, that I believe this because it just seems reasonable and obviously true, implying that I was talking off the top of my head and had no scholarly basis my comments. My reply was intended to demonstrate that this was unfounded (as well as terribly condescending--I try to treat everyone with respect and feel I deserve the same in return).”

And yet this exactly how you described neoclassical economists. I was quoting your original post verbatim:

“This is why Neoclassical microeconomics is based on concepts like, “all firms are short-term profit maximizers.” Did some economist do a study or survey to determine this? No. Instead, it just seems reasonable. More than that, they regard it as obviously true. How could any reasonable person disagree with this?”

In the OP you are making a claim about the nature of neoclassical arguments—that they are metaphysical—and about the nature of PK arguments—that they are more solidly grounded in the real world. My criticism relates to how this argument stacks up on its own terms. If PK arguments are better grounded in the data, and this is a PK argument, then show us how it is grounded in the data. It was meant to reflect your own challenge to the monetarist commenter at your blog.

“It wouldn’t actually settle it as there is no such thing as a definitive study, but that’s beside the point.”

If you were to survey a representative group of neoclassical economists and ask them the question and they mostly said yes, I would consider myself refuted, at any rate.

“one does not personally need to have done the research in a particular area to legitimately draw upon the insights of others”

Didn’t mean to imply that you did, sorry.

“Every model is a simplification (by definition). Every model does not, however, create alternate universes where, for example, money supplies are increased by helicopters, fully-functioning, pre-trade worlds exist, everyone has perfect foresight, etc.”

Every model necessarily creates an alternative universe where to all intents and purposes the modeller is God: nothing exists unless it is assumed into existence. Reality is realistic. Models are just a bunch of equations. You can spin a bunch of equations however you like. I don’t know the work on PK micro that you cited but I’m familiar with Godley and Lavoie’s stuff. They make equally unrealistic assumptions.

“It’s poor modeling when it turns out that a) different premises lead to radically different conclusions and b) it does not take a great deal of effort to discover the nature of the premises that more properly ground the model is real-world behavior.”

Who would disagree with this?

“I already gave several examples.”

I meant an example where a PK researcher does work that illustrates the distinction you’re making. It was in response to this:

“Bruce Elmslie and William Milberg have been doing work on a theory that focuses on absolute advantage and puts technology and intraindustry trade at the center. No thought experiments or analogies that do not have real-world parallels.”

Isn’t that what everyone is going for? So how is it actually different in practice?

“Now ask a Neoclassical economist how that can happen in the real world.”

A testable proposition, eh! Actually, I’ve already done this: I asked our lecturer in monetary theory, and he said that the Fed can’t directly control any of the broader monetary aggregates, which is why they don’t bother trying. He also told me that there was no stable velocity multiplier. In fact he explicitly contradicted all four of the assumptions that you wrote about in your Forbes article.

This is a good example of what I’m having a problem with. You make an empirical claim and offer anecdotal evidence, but it goes against my own experience.

vimothy said...

Scott,

"sounds like a critique of New Classicals".

Agreed.

Also, can't recommend MacKenzie enough.

John Harvey said...

*****PART ONE*****

VIMOTHY: “My criticism relates to how this argument stacks up on its own terms.”

JTH: That’s not true. I am happy to discuss that, but your main criticism was that I had no rational, scholarly basis for my claims, that it was simply “drive-by snark.” You clearly state this, this is not inference on my part: “Did John Harvey do a study or survey to determine whether neoclassical economists do actually believe this? No.” This was both part of your opening statement and it was a continuing theme. So, to pretend that this whole discussion has been an intellectual discussion of method is disingenuous. I was, in essence, being called a hack. I did not do this to you, nor did I turn the tables and ask if you had done a specific study on this issue that would allow you to refute my claims. It’s not necessary to have done so in order to contribute to such a debate (something you have now openly admitted, although my “hack” status still appears to in question).


VIMOTHY: “If PK arguments are better grounded in the data, and this is a PK argument, then show us how it is grounded in the data. It was meant to reflect your own challenge to the monetarist commenter at your blog.”

JTH: It’s not a question of arguments better grounded in data, but being concerned that premises approximate processes that can actually happen in the real world. Read here for example:

http://blogs.forbes.com/johntharvey/2011/05/14/money-growth-does-not-cause-inflation/

It offers two arguments about inflation, the second one allowing for the manner in which central banks actually operate and which does not assume constant velocity or full employment. This is the point.


VIMOTHY: (quoting JTH) “It’s poor modeling when it turns out that a) different premises lead to radically different conclusions and b) it does not take a great deal of effort to discover the nature of the premises that more properly ground the model is real-world behavior.” (Now Vimothy): Who would disagree with this?

JTH: NEOCLASSICISM!!!!!!!! Maybe you are not getting my central point. This is a philosophical disagreement. Neoclassical micro, for example, self consciously adopts axiomatic “obviously true” premises rather than basing behavioral assumptions on observation. Why, are they lazy? No, they think that’s a superior method. Observation, in the Cartesian tradition, is biased and unreliable. Like Neo in The Matrix, we have to rely on introspection and self reflection. That is the only reliable means of developing premises. This is the Neoclassical view. Hence, even if a premise is both critical and out of synch with what we observe in the real world, it’s beside the point. Assuming a helicopter is, in that tradition, a superior means of understanding the world. Even if the world doesn’t operate that way, it acts as if it does. This isn’t viewed as a weakness within Neoclassicism, it’s seen as a strength. I don’t agree, and neither do the other Post Keynesians. You, however, seem to be arguing that they don’t do that in the first place, thus implicitly accepting the PK view of methodology but then defending the Neoclassical one.

John Harvey said...

*****PART TWO*****

VIMOTHY: (quoting JTH) “I already gave several examples.” (Now Vimothy) I meant an example where a PK researcher does work that illustrates the distinction you’re making. It was in response to this: (quoting JTH) “Bruce Elmslie and William Milberg have been doing work on a theory that focuses on absolute advantage and puts technology and intraindustry trade at the center. No thought experiments or analogies that do not have real-world parallels.” (Now Vimothy) Isn’t that what everyone is going for? So how is it actually different in practice?

JTH: No, it’s not, as I argued above. Neoclassicism’s adopted method says relying on observation is untrustworthy. We can see how it can be dangerous, but what other reference is superior? None, say PKs.


VIMOTHY: (quoting JTH) “Now ask a Neoclassical economist how that can happen in the real world.” (Now Vimothy) A testable proposition, eh! Actually, I’ve already done this: I asked our lecturer in monetary theory, and he said that the Fed can’t directly control any of the broader monetary aggregates, which is why they don’t bother trying. He also told me that there was no stable velocity multiplier. In fact he explicitly contradicted all four of the assumptions that you wrote about in your Forbes article.

JTH: This is actually an excellent example of what I’m talking about. Did you ask if he believes that excess monetary growth causes inflation? Because with the above explanation, it can’t. And yet, that’s the Neoclassical theory. So they know very well that in terms of real world mechanisms, it’s impossible. But their method allows them to say this at the end: “Yes, but the real world acts as if it were possible.”

vimothy said...

John,

I think my meaning was clear. You said in your article,

“This is why Neoclassical microeconomics is based on concepts like, “all firms are short-term profit maximizers.” Did some economist do a study or survey to determine this? No. Instead, it just seems reasonable. More than that, they regard it as obviously true. How could any reasonable person disagree with this?”

I simply turned this around and asked it of you. If you find the implication offensive, well, maybe you shouldn’t be so quick to make it of others.

You go on,

“I was, in essence, being called a hack.”

They aren’t my words, are they—they’re yours! That was, in fact, the point. Was this too literary a move for an economics blog? You describe a large number of people as hacks, because they don’t do X. I ask, Do you, ahem, do X? You respond, Wait, what? Are you suggesting that I’m some sort of hack?! How dare you!

“Nor did I turn the tables and ask if you had done a specific study on this issue that would allow you to refute my claims”

Feel free to do this. I don’t see that it alters my argument in the slightest.

“It’s not necessary to have done so in order to contribute to such a debate (something you have now openly admitted”

I have never denied this.

“It’s not a question of arguments better grounded in data, but being concerned that premises approximate processes that can actually happen in the real world.”

How do you know that your premises approximate processes that can actually happen in the real world if you do not look at what actually happens in the real world?

“It offers two arguments about inflation, the second one allowing for the manner in which central banks actually operate and which does not assume constant velocity or full employment. This is the point.”

But what does this actually prove?

“NEOCLASSICISM!!!!!!!! Maybe you are not getting my central point. This is a philosophical disagreement.”

No, I think I understand what your argument is. The problem is that what you’re saying about neoclassical economics is quite plainly pejorative. When I described you in the same terms, you became highly offended.

“Neoclassical micro, for example, self consciously adopts axiomatic “obviously true” premises rather than basing behavioral assumptions on observation.”

Really? Don’t you also use axioms when you build models? I think what you mean here is, “our models are better.”

“No, it’s not, as I argued above. Neoclassicism’s adopted method says relying on observation is untrustworthy. We can see how it can be dangerous, but what other reference is superior? None, say PKs.”

You have theory and you have data. Neoclassicals have the same. I don’t see what the actual difference is in practice.

“Did you ask if he believes that excess monetary growth causes inflation?”

I already know the answer to that one.

“Because with the above explanation, it can’t.”

Well, I disagree, of course.

“And yet, that’s the Neoclassical theory.”

It’s not though is it? It’s just one observation. It’s a theory.

vimothy said...

This is why Neoclassical microeconomics is based on concepts like, “all firms are short-term profit maximizers.” Did some economist do a study or survey to determine this? No. Instead, it just seems reasonable. More than that, they regard it as obviously true. How could any reasonable person disagree with this?

I think it's worth quoting some neoclassical micro here. I’m copying this from The Theory of Industrial Organization by Jean Tirole, a very mainstream neoclassical micro textbook:

“The profit-maximization hypothesis. It is a postulate of this book, and of most economic theory, that firms maximize expected profits. There is, however, a widespread feeling that in practice their managers have other objectives (e.g. maximizing the firm’s size and growth and the prerequisites of the managerial position). This section presents arguments for and against the profit maximization hypothesis. It also discusses the power of current industrial organization theory in the presence of non-profit-maximizing firms...”

John Harvey said...

*****PART ONE*****

VIMOTHY: “I think my meaning was clear. You said in your article,

(Quoting JTH) “This is why Neoclassical microeconomics is based on concepts like, “all firms are short-term profit maximizers.” Did some economist do a study or survey to determine this? No. Instead, it just seems reasonable. More than that, they regard it as obviously true. How could any reasonable person disagree with this?”

(Now Vimothy) I simply turned this around and asked it of you. If you find the implication offensive, well, maybe you shouldn’t be so quick to make it of others.”

You go on,

(Quoting JTH) “I was, in essence, being called a hack.”

(Now Vimothy) They aren’t my words, are they—they’re yours!

JTH: Indeed, those were not your words. I never said they were, which is why my post says “in essence.” But your intent was clear. And to put the above back into context, I wrote this because you implied that our exchange was focused on some sort of scholarly debate. I disagreed, reminding you that your original main point was that I was unqualified to discuss this issue (i.e., a hack). It’s quite clear from the above posts. I am not denying that I was arguing that Neoclassical economics is poorly done. However, this was a result of a quarter century as a professional economist, during which this has been a major side interest of mine and, of late, a major research interest. Hence, this was not, as you said, something of which I quickly accused people. I did years of research before I arrived at that conclusion. How much of my work did you review before arriving at yours? More than this Mike Norman post???

********************************

VIMOTHY: (Quoting JTH) “Nor did I turn the tables and ask if you had done a specific study on this issue that would allow you to refute my claims” (Now Vimothy) Feel free to do this. I don’t see that it alters my argument in the slightest.

JTH: Of course it would. It would mark you as unqualified to make such a statement about my work. I don’t agree with this, but by your logic that would be a valid concusion.

********************************

VIMOTHY: (Quoting JTH) “It’s not necessary to have done so in order to contribute to such a debate (something you have now openly admitted” (Now Vimothy) I have never denied this.

JTH: Yes you did: “Did John Harvey do a study or survey to determine whether neoclassical economists do actually believe this? No.” And: “Fair enough, but what I actually asked was, “Did John Harvey do a study or survey to determine whether neoclassical economists do actually believe this?” A decently designed survey would be an authoritative way to settle this, I feel. But it doesn’t sound like you did.” However, I accept now that you do not believe this.

John Harvey said...

*****PART TWO*****

VIMOTHY: (Quoting JTH) “It’s not a question of arguments better grounded in data, but being concerned that premises approximate processes that can actually happen in the real world.” (Now Vimothy) How do you know that your premises approximate processes that can actually happen in the real world if you do not look at what actually happens in the real world?

JTH: I totally agree with you.

********************************

VIMOTHY: (Quoting JTH) “It offers two arguments about inflation, the second one allowing for the manner in which central banks actually operate and which does not assume constant velocity or full employment. This is the point.” (Now Vimothy) But what does this actually prove?

JTH: That Neoclassicals are aware of these inconsistencies but they don’t bother them.

********************************

VIMOTHY: (Quoting JTH) “NEOCLASSICISM!!!!!!!! Maybe you are not getting my central point. This is a philosophical disagreement.” (Now Vimothy) No, I think I understand what your argument is. The problem is that what you’re saying about neoclassical economics is quite plainly pejorative. When I described you in the same terms, you became highly offended.

JTH: Of course, because I come from a school of thought that doesn’t agree with that approach. Note that Neoclassicism doesn’t see a problem with a priori deductivism. To them, being accused of that is not an insult. It’s like saying to a Texan, “Ha ha, you believe that the best barbecue is beef!!!” Yea, that’s right. But to a Tennessean, it’s not. It’s pork. At any rate, you were saying more than that. You were saying I was unqualified, and then when I showed otherwise you never made mention of it again. I actually found that more offensive, but whatever.

********************************

VIMOTHY: (Quoting JTH) “Neoclassical micro, for example, self consciously adopts axiomatic “obviously true” premises rather than basing behavioral assumptions on observation.” (Now Vimothy) Really? Don’t you also use axioms when you build models?

JTH: Correct. Neoclassicism has consciously adopted an a priori, deductivist, natural-law approach to the world. In Post Keynesianism, the emphasis is placed on being able to draw premises from historical and institutional detail and to specify the line of causation at work. Can this always be done? No, but the point is that legitimacy for a premise is a function of how closely it mirrors at real-world phenomenon. That is not the case in Neoclassicism.

********************************

VIMOTHY: (Quoting JTH) “Did you ask if he believes that excess monetary growth causes inflation?” (Now Vimothy) I already know the answer to that one. (Quoting JTH) “Because with the above explanation, it can’t.” (Now Vimothy) Well, I disagree, of course.

JTH: PLEASE tell me how this is possible!!!!!!!!!!!!!!! I would love to hear the explanation.

********************************

SUMMARY: In rereading all this, Vimothy, I’m beginning to think that this is the problem: you actually don’t agree with the a priori, deductivist method. Maybe this is why it bothers you that I say that Neoclassicism uses such an approach. But that they do is not controversial. Neoclassical methodologists have no problem with this (again, the Prasch article is an awesome introduction to this). Maybe at the end of the day, however, you are actually a Post Keynesian at heart? I’m not being sarcastic here. I would welcome it.

vimothy said...

"Your original main point was that I was unqualified to discuss this issue"

This was never stated or implied in any of my comments. Frankly, and since you in fact brought them up, I don't see how your qualifications are relevant. Impressive though they may be.

John Harvey said...

Man, I’m sorry, but you did and very clearly. You state that I have no legitimate basis to offer an opinion on this issue if I haven’t done a study on it myself. However, if this is meant to be an apology or a statement that you didn’t intend to say what you did, fair enough. I totally agree that it should never have been an issue in the first place.

vimothy said...

No John, I did not.

Please quote the specific bit of text that you're referring to.

John Harvey said...

I've cited the initial comment over and over, but okay:

"Did John Harvey do a study or survey to determine whether neoclassical economists do actually believe this? No."

Note the "no" after the first sentence. If your intent wasn't to imply that I therefore had no scholarly basis for my comments, then I'm not sure what it was.

While it continues to bother me that you would make this a core part of your argument and then ignore it once shown evidence that it was false, but whatever. It should never have been an issue in the first place. The real issue is whether or not the a priori deductivist method used by Neoclassicism is superior to the Critical Realism of the Post Keynesians. My answer is obvious. I wrote the post because I assumed that many people might be unaware of this methodological difference.

vimothy said...

"Note the "no" after the first sentence."

Yes, note the "no".

This,

“Did John Harvey do a study or survey to determine whether neoclassical economists do actually believe this? No.”

Is me quoting you and replacing “some economist” with “John Harvey” and “this” with “whether neoclassical economists do actually believe this”:

“Did some economist do a study or survey to determine this? No.”

Which is in the original post in the top of the page.

I think this is quite clear. I even go on to quote you some more:

"Instead, it just seems reasonable. More than that, he regards it as obviously true."

Here, I replace "they" with "he":

"Instead, it just seems reasonable. More than that, they regard it as obviously true"

Those are your words. You find them offensive!

It certainly doesn't bother me of you have or haven't. Although, as I said, if you have surveyed a bunch of neoclassical economists and got them to agree that observing the real world isn't important, I'll eat my hat.

John Harvey said...

The original source of the language was not in question. Rather, the issue is (as I have explained over and over) the point you were trying to make. It’s quite clear that you are saying that if I haven’t done research in the area, then I have nothing worthwhile to say about it (although you did not apply this standard to yourself). I am, in fact, pretty darn well read on it and have a book under contract where method is a central focus.

Second, as already explained above, of course I would take offense to having my research characterized this way–I don’t agree with that method. But, again as mentioned above, what I found more offensive was the lack of acknowledgment on your part that one of the main premises of your argument (i.e., that I just thought this seemed like it made sense and I had not done any research on it) was false.

Third, I have yet to see you answer my question on how the money supply can rise above the demand for money in a modern financial system. I would truly be very interested in hearing this.


Again, through all this I’m wondering if the real issue is that, like me, you are not really a fan of the a priori/deductivist method. Hence, like me, you are taking it as a negative; but, unlike me, you are objecting to seeing Neoclassicism associated with it. In fact, that this positivist approach is their preferred means of doing research is not actually controversial.

vimothy said...

That was never the point. I literally could not care less if you have done a study. I mean, I'm totally indifferent. I have never, ever given any thought to the question before this very moment. It is of no interest to me whatsoever.

"It’s quite clear that you are saying that if I haven’t done research in the area, then I have nothing worthwhile to say about it (although you did not apply this standard to yourself)"

The criteria is yours. The words are yours!

Now, you can claim that I meant something else. You are entitled to believe whatever you choose, but of course I know what I meant because I wrote it. And I didn't mean that.

JakeS said...

Moving back towards the subject at hand, I would rather like to hear how "too much money" can cause inflation if the money supply is endogenous.

Because that strikes me as a topsy-turvy view of cause and effect.

(I can not even see how low interest rates can cause inflation. I *can* see how *lax lending standards among private banks* can cause (asset price) inflation, but I cannot see how low interest rates incentivises poor diligence among private banks.)

- Jake

John Harvey said...

Vimothy, you have told me over and over what you did not mean--what did you mean????

And I again am anxious to hear how (as Jake said) a Ms>Md in an endogenous money world.

vimothy said...

What I mean is that in practice you are not that far from neoclassical economists. You seem to find this offensive.

"And I again am anxious to hear how (as Jake said) a Ms>Md in an endogenous money world."

Isn't everything ultimately endogenous? And yet, here we are.

Say that there is a negative shock to money demand. As long as the supply does not adjust instantaneously, there will be some period during which the private sector is holding "excess" money. How will it get rid of it?

Or let's say there is a positive shock to money demand and supply doesn't adjust instantaneously. You personally can increase your holdings of money by either not spending your income or by selling something. If everyone does this there will be aggregate expenditure effects.

While we're sharing, I'm intrigued as to what you think causes inflation if not growth in the money supply? ("Inflation causes money supply growth" is not an answer).

John Harvey said...

VIMOTHY: What I mean is that in practice you are not that far from neoclassical economists. You seem to find this offensive.

JTH: How can that be true if your statement that I had done no research was false? We are again left with your original argument being based on an unwarranted premise, and yet you have not acknowledged this.


With respect to money supply, you are still avoiding the question. It’s really very simple: what mechanism does the fed have that allows them to raise the money supply above demand? That’s all I’m asking. I would be delighted to answer your question when you have answered mine (in fact, it’s the subject of my next Forbes.com post).

One more time to be absolutely, crystal clear: by what specific, real-world process does the central bank increase money supply above money demand?

vimothy said...

"How can that be true if your statement that I had done no research was false?"

I never made that statement.

"With respect to money supply, you are still avoiding the question."

That's a different question. Jakes question was "how does "too much money" cause inflation given that the money supply is endogenous?"

Your question here is, how does the central bank control the money supply, given that the money supply is endogenous. My answer is that it does not control the money supply. There's no direct link between Fed OMO and broad money growth. I think this is mostly captured in the term "endogenous".

John Harvey said...

I’m going to drop the whole thing about the research since you have obviously somehow convinced yourself that you did not type what I can read in the posts above.

However, you write: “Your question here is, how does the central bank control the money supply, given that the money supply is endogenous.”

No, that’s not my question. I wrote it out twice so there would be no misunderstanding. But, I’ll write it again: by what specific, real-world process does the central bank increase money supply above money demand?

That and nothing else is my question.

vimothy said...

"I’m going to drop the whole thing about the research since you have obviously somehow convinced yourself that you did not type what I can read in the posts above."

Erm, no John, obviously I typed what you can read in the comments above, but the words do not mean what you think they mean.

You have a remarkable talent for misreading posts. You've managed to avoid engaging with anything substantive that I've written and spent most of the conversation banging on about this, despite the fact that I’ve clarified it several times. No, I’m not suggesting that you haven’t done any research. No, I don’t actually care if you have done a survey. It's particularly ironic given that you yourself wrote the words you object to so vehemently. You brought the survey up in the first place, remember?

“Did some economist do a study or survey to determine this? No.”

Of course, neoclassical economists just make it up. It’s what they do. On the other hand, PK economists make sure that they can point to real world processes. Right. The fact that you are a PK economist and this distinction makes PK economists look good relative to neoclassical economists is unrelated to the distinction itself, I’m sure.

You wrote in the OP that,

Neoclassical economists "believe that knowledge gained from introspection and the generation of “obviously true” premises is superior to that based on observation of the real world."

Can you point to something in the real world to back this up? You don’t in the OP. Did you, in fact, do a survey to verify it? No. So, quite regardless of anything else you may or may not have done (apparently, it was a major side interest of yours for 25 years, before you decided that "Neoclassical economics is poorly done"—thanks for that) and regardless of its truth or falsity, that's something you share with the straw man microeconomist that you wrote about, who didn’t do a survey either and who doesn't know for certain whether or not firms in the real world maximize profits. There is a sense in which you are both alike.

Again, your claim was that neoclassical arguments have property X, and PK arguments have property Y, where Y > X. I have suggested that your claim is itself a PK argument, and that it shares some of property X, which is a contradiction. For your part you have stated at length how offended you are that anyone should have the temerity to suggest that one of your arguments has the property X. You are actually offended! I mean, really. It’s too much.

"No, that’s not my question. I wrote it out twice so there would be no misunderstanding. But, I’ll write it again: by what specific, real-world process does the central bank increase money supply above money demand?"

Well, I already answered it. The central bank does not increase the "money supply" at all. The central bank engages in OMOs, which determine a policy rate and the stock of outside money. (Excepting some unconventional operations like QE, which can affect broad money directly). The money supply in terms of the stock of liquid bank deposits is endogenously determined by the private sector.

Other things are also endogenously determined by the private sector. For instance, the supply of potato chips. How does the central bank increase the supply of potato chips so that PCs>PCd? Can it ever be the case that PCs>PCd? How, if potato chips are endogenous? Crazy world, ain’t she.

John Harvey said...

Vimothy, let me point out that I am not the only person reading this exchange who believes that you are now changing your mind about what you said. So if I have a talent for misreading posts, apparently I am not alone. I have answered your questions completely and directly, I just don’t think you want to believe that you are wrong. Okay. Nothing else I can do on that front.

With respect to answering my question, no you most certainly have not. I asked a very, very simple question (in fact, I’ve asked it 8 times now, including the exchange with the monetarist at Forbes.com, and have yet to see response): by what specific, real-world process does the central bank increase money supply above money demand?

You say: “The central bank does not increase the "money supply" at all. The central bank engages in OMOs, which determine a policy rate and the stock of outside money. (Excepting some unconventional operations like QE, which can affect broad money directly). The money supply in terms of the stock of liquid bank deposits is endogenously determined by the private sector. “

So, are you saying that it can’t? If so, please say so–give me a direct answer to my question: “Actually, the central bank cannot act to increase the supply of money above the demand.” Although, if that’s the answer, them I’m awfully curious about how the Neoclassical story of Ms>Md actually works. And THAT has been the point all along, the fact that Neoclassicism is perfectly willing to accept as a major premise something that is not just a simplification. It specifies a process that cannot occur in the real world.

So, for the ninth time: by what specific, real-world process does the central bank increase money supply above money demand?

Unless I’m missing something here, the answer has to be either, “It happens like this...” or “It can’t happen.” Which one is it?????

vimothy said...

"Vimothy, let me point out that I am not the only person reading this exchange who believes that you are now changing your mind about what you said"

Another solid argument. Far be it from me to tell you what I did or didn't mean.

I certainly agree that we've done this to death now.

"With respect to answering my question, no you most certainly have not."

I have answered it several times.

If the central bank does not control the supply of bank deposits then by definition it can't directly change the supply of bank deposits. What else is there to say?

Things that the central bank can do can affect liquid bank deposit supply and demand functions, but there is no simple way that the central bank can control an aggregate like M4. This is why monetarism was abandoned as a practical endeavour decades ago.

Indeed, in my own neoclassical monetary theory classes, this is exactly what we were taught.

In other words, you've claimed that neoclassicals believe the central bank can do X. I've claimed that they don't believe that at all. You have responded by saying, "For the nth time, tell me how the central bank can do X!"

It's a false-flag / false dichotomy / straw man type argument. Nobody believes that the central bank can directly control the money supply. If they did, we would see central banks attempting to directly control the money supply. However, we do not.

JakeS said...

"Say that there is a negative shock to money demand."

Has a negative shock to money demand ever actually been observed in the real world? Where and when?

Positive shocks to money demand have been observed, and institutionalists have a cause-and-effect explanation for how those comes about (see, e.g. Hyman Minsky, Stabilizing an Unstable Economy). But off the top of my head, I can't think of any historical example of a negative shock to money demand.

"As long as the supply does not adjust instantaneously, there will be some period during which the private sector is holding "excess" money. How will it get rid of it?"

What is the proposed mechanism behind this postulated negative shock to money demand? It can't be an inverse Minsky Moment, because the cause-and-effect explanation of Minsky Moments involves fundamental asymmetries between expanding and contracting credit.

If the mechanism is reduced savings, then the money supply *will* change instantaneously to accommodate it. The cause of inflation (if in fact there is any) under this scenario is greater aggregate demand from reduced savings under constant or increasing short-term income. Money supply has nothing to do with it.

"While we're sharing, I'm intrigued as to what you think causes inflation if not growth in the money supply?"

The proximate cause is that vendors raise prices.

The ultimate cause can be lots of different things.

The price increase can be due to increased market power by sellers, in which case you will find that sellers increase their markup.

The price increase can be due to unresolved power struggles in the political economy - the price-wage spiral is the canonical example here.

The price increase can be due to real scarcities that the political economy is poorly equipped to deal with. This will be characterised by constant or declining margins and wages at the same time as inflation (stagflation).

The price increase can be due to an excess of demand over what the political economy is capable of satisfying.

Inflation can be caused by the industrial planning system planning for inflation, because moderate inflation increases the stability and power of the industrial planning system (by reducing real debt burdens and by making nominal growth easier).

And so on and so forth.

It's essentially a historical question, not a predictive one. And as long as inflation is below the double digits, it is a historical question of very little practical relevance for economic planning.

The central bank cannot, as a practical matter, increase inflation. It can refrain from suppressing inflation. But the only way in which the central bank can suppress inflation in the first place is by generating idle capacity in the productive economy by creating an artificial minimum required return on investment.

And I am not wholly certain why it should be the central bank's job to make business plans that are profitable on their merits unprofitable simply by policy fiat. That strikes me as a fairly crude and dumb way to control inflation, compared to addressing the underlying causes (such as unresolved power struggles or strategic resource bottlenecks).

- Jake

John Harvey said...

VIMOTHY: In other words, you've claimed that neoclassicals believe the central bank can do X. I've claimed that they don't believe that at all. You have responded by saying, "For the nth time, tell me how the central bank can do X!"

Vimothy, you are completely misunderstanding the argument. I never said that Neoclassicals do believe that the central bank has complete control of an exogenous money supply. I said that they need to to make their explanation work, making their argument inconsistent. In fact, we went over this very ground before. Here is our earlier exchange:

“VIMOTHY: (quoting JTH) “Now ask a Neoclassical economist how that can happen in the real world.” (Now Vimothy) A testable proposition, eh! Actually, I’ve already done this: I asked our lecturer in monetary theory, and he said that the Fed can’t directly control any of the broader monetary aggregates, which is why they don’t bother trying. He also told me that there was no stable velocity multiplier. In fact he explicitly contradicted all four of the assumptions that you wrote about in your Forbes article.

JTH: This is actually an excellent example of what I’m talking about. Did you ask if he believes that excess monetary growth causes inflation? Because with the above explanation, it can’t. And yet, that’s the Neoclassical theory. So they know very well that in terms of real world mechanisms, it’s impossible. But their method allows them to say this at the end: “Yes, but the real world acts as if it were possible.”“

How can the Neoclassical explanation of inflation, based on Ms>Md, survive in an endogenous-money world? And yet they do.

John Harvey said...

Nice post, Jake!

vimothy said...

I have already responded to that:

Say that there is a negative shock to money demand. As long as the supply does not adjust instantaneously, there will be some period during which the private sector is holding "excess" money. How will it get rid of it?

Or let's say there is a positive shock to money demand and supply doesn't adjust instantaneously. You personally can increase your holdings of money by either not spending your income or by selling something. If everyone does this there will be aggregate expenditure effects.


[That's what Jake is referring to in his post at May 30, 2011 12:21 PM]

Whereupon you responded, ignoring my explanation,

"For the nth time, tell me how they do X!"

One more time to be absolutely, crystal clear: by what specific, real-world process does the central bank increase money supply above money demand?

So, to recap: Supply and demand for liquid bank deposit is not always at equilibrium.

Think about it in terms of supply and demand for any good. Call this good "potato chips". Now, the supply of potato chips is endogenously determined by private supply and demand. Does this imply that this market is always in equilibrium? Hell no. Say there is a shock to supply or demand--there you go. And this despite its obvious endogeneity.

John Harvey said...

So no real-world mechanism exists to allow the central bank to create the Ms>Md that Neoclassicism says causes inflation?

I didn't think so.

vimothy said...

Jake,

"Has a negative shock to money demand ever actually been observed in the real world? Where and when?"

Imagine a situation in which the market receives some information regarding the ability of the government of a country to pay its debts. All of sudden, people say, "By Toutatis! I have all these money claims against that government. I need to get rid."

Something like that would do it, I think. I'm sure I could come up with other scenarios that might explain why demand for a type of claim against a particular type of entity might shift.

"The proximate cause [of inflation] is that vendors raise prices."

Yes, but that's what inflation is, so that's a circular argument. We want to understand why prices rise. "Because they rise" is not much help. We know already that they do.

"The ultimate cause can be lots of different things.

Seems to boil down to “because they rise” again (price-wage spirals; excess demand) or relative price confusion (supply shocks, market power).

I’m not sure what you mean by this, though: “Inflation can be caused by the industrial planning system planning for inflation”.

What is the industrial planning system? How does it cause inflation?

“And as long as inflation is below the double digits, it is a historical question of very little practical relevance for economic planning.”

That doesn’t seem like a particularly responsible attitude for a policy maker to take, if they don’t actually know why inflation has been low.

vimothy said...

"So no real-world mechanism exists to allow the central bank to create the Ms>Md that Neoclassicism says causes inflation?"

How many millions of times?

No, the central bank cannot exogenously set the supply of inside money wherever it likes.

No, the fact that inflation is a monetary phenomenon does not rely on the ability of the central bank to exogenously set the supply of inside money wherever it likes.

It's a straw-man argument.

JakeS said...

"Say that there is a negative shock to money demand."

No, don't just say that. Specify the mechanics of this postulated "shock to money demand."

"So, to recap: Supply and demand for liquid bank deposit is not always at equilibrium."

Of course they are not "at equilibrium" - the financial system is fundamentally not an equilibrium system (more appropriate analogies would be biological homeostasis, or electrical grid load-balancing).

But what you meant to say with "disequilibrium" is actually "failing to clear." And you need a mechanism for such clearing failures that is capable of causing inflation. There exists a both plausible and observed mechanism for financial clearing failures that cause deflation (Fisher-Minsky debt-deflation spirals). But I am not aware of any proposed (let alone observed) causal mechanism for inflation-causing clearing failures.

"Think about it in terms of supply and demand for any good."

Money is not a good. It is a representation of power relationships. It turns out that this distinction is not vital to the present argument, but it is sufficiently important to be worth mentioning anyway.

"Call this good "potato chips". Now, the supply of potato chips is endogenously determined by private supply and demand. Does this imply that this market is always in equilibrium? Hell no. Say there is a shock to supply or demand--there you go. And this despite its obvious endogeneity."

However, it is easy to specify the causal mechanism behind shocks to potato chip supply or demand (crop failure, trade agreements, economic depression), and how they interact with the mechanics of production and distribution of potato chips to cause price adjustments and clearing failures.

So what's the mechanism by which you get an inflation-causing clearing failure in the money markets?

If you don't have a plausible cause-and-effect story, you're doing tooth fairy science.

- Jake

vimothy said...

"No, don't just say that. Specify the mechanics of this postulated "shock to money demand.""

I just did:

Imagine a situation in which the market receives some information regarding the ability of the government of a country to pay its debts. All of sudden, people say, "By Toutatis! I have all these money claims against that government. I need to get rid."

Something like that would do it, I think. I'm sure I could come up with other scenarios that might explain why demand for a type of claim against a particular type of entity might shift.

vimothy said...

"Of course they are not "at equilibrium"

The issue is (or was), "How can Ms > Md given endogeneity of M?"

If Ms=Md then the market for M can be said to be at equilibrium. In other words, if the it cannot be the case that Ms > Md, then either Ms < Md or Ms=Md. Ignoring the former (since the endogeneity argument applies to it equally), the claim that Ms cannot exceed Md is equivalent to the claim that the market for M is always in equilibrium, where equilibrium means Ms=Md.

I'm not making any grandiose statements about the nature of reality here.

the financial system is fundamentally not an equilibrium system (more appropriate analogies would be biological homeostasis, or electrical grid load-balancing).

You can use whatever metaphors you like to describe its behaviour.

But what you meant to say with "disequilibrium" is actually "failing to clear."

What I meant to say was what I did say. If there is excess supply, then given p and a demand function, the public holds a quantity q greater than desired.

If people hold more money than they want, they will try to get rid of it, and this will contribute to aggregate expenditure. Steve Waldman has been developing a model that you might find interesting here: http://www.interfluidity.com/v2/1699.html#comments

If you don't have a plausible cause-and-effect story, you're doing tooth fairy science.

Orilly?

JakeS said...

"Imagine a situation in which the market receives some information regarding the ability of the government of a country to pay its debts.

[...]

Something like that would do it, I think."

Create inflation?

No, that is not the historical experience. Historically, that creates deflation. See, e.g., Greece and Indochina.

(And the reason it does that is that what you're actually observing here is a positive shock to hard currency demand in the target country, to pay down hard currency obligations that its residents cannot roll over.)

If the target country attempts to maintain an overvalued currency peg, the resulting economic collapse may cause a currency collapse with (if the target country has a structural import dependency and a negative foreign primary balance) accompanying hyperinflation. But that is a very different story from the one you were telling.

"Seems to boil down to “because they rise” again (price-wage spirals; excess demand)"

Uh, no.

Price-wage spirals involve an explicit cause-and-effect story. It's just not a cause-and-effect story about supply and demand, because price-wage spirals have nothing to do with supply and demand and everything to do with power relationships in the industrial planning system.

During price-wage spirals, prices rise because both firms and labour have the power to fix prices on their own products, and are in a political conflict over the distribution of added value. Firms increase prices to capture a larger share of value added, and labour demands higher wages to restore their share of the value added. Because firms coordinate prices within but not across sectors, the Nash equilibrium of that process is (potentially accelerating) across-the-board price increases.

[TBC]

JakeS said...

"or relative price confusion (supply shocks, market power)."

No, this isn't "relative price confusion."

In the context of downward nominal price and wage rigidity, strategic resource bottlenecks will result in general price increases, because relative prices tend to adjust by some prices going up rather than some going up and others down.

And in case you were wondering, there are excellent fundamental reasons for nominal downward price and wage rigidity in the credit economy. Namely that interest and amortisation expenses have strong nominal downward rigidity.

If I make oil and you make cars and we are both geared 2:1 (fairly conservative gearing), and the real cost of oil increases, then it is much easier for us to make the relative price adjustment by me raising nominal prices than by you lowering nominal prices. Because me raising nominal prices screws the bank without costing me anything, while you lowering prices screws you directly.

"I’m not sure what you mean by this, though: “Inflation can be caused by the industrial planning system planning for inflation”."

Large, capital-intensive industrial firms benefit from moderate inflation, because it reduces the power of banks to obtain concession and makes it easier to satisfy shareholders' growth requirements (which are formulated in nominal terms).

So if large, capital-intensive industrial firms are the dominant power in the political economy, they have both the means (oligopolistic price-fixing), motive (greater independence from banks and shareholders) and political air cover (from rented politicians and bought pundits) to ensure moderate, planned inflation.

"That doesn’t seem like a particularly responsible attitude for a policy maker to take, if they don’t actually know why inflation has been low."

Why not?

Inflation is fairly harmless until it passes the double digits. As long as the money held for ordinary transactions does not depreciate appreciably during the time it is held in the course of ordinary transactions, inflation is of very little concern to the productive economy.

Or, to put it in simpler terms, any inflation below 10 % p.a. is a tax on lazy money, and I fail to see why policymakers should care about the well-being of lazy money.

- Jake

vimothy said...

"No, that is not the historical experience. Historically, that creates deflation."

So let me get this straight: govts whose ability to service their debts collapses tend to experience an increase in the real value of the money they issue? Are you sure?

JakeS said...

"So let me get this straight: govts whose ability to service their debts collapses tend to experience an increase in the real value of the money they issue?"

No, of course not.

Because the premise is false. Governments can by definition never experience a collapse in their ability to serve debts denominated in the currency they themselves issue. Governments can only experience inability to service debts in hard currency - that is, currency that other people issue.

So right out of the gate, your premise is nonsense.

Now, what actually happens during a currency crisis is:

1) Country accumulates hard currency debts.

2) Country loses access to hard currency credit.

3a) Country decides to defend currency against depreciation.

3b) Country raises interest rates and/or engages in Austerity(TM) to defend against depreciation. This is deflationary (that, in fact, is the whole point of the exercise - it would not work to temporarily defend the exchange rate if it were not deflationary).

3c) Country's economy goes down the crapper due to the deflationary effects of interest rate increases and Austerity(TM).

3d) Country calls the IMF. The IMF does what the IMF does best: Turns the disaster into a catastrophe.

4) Country gives in to the inevitable and permits currency to depreciate.

5) Country experiences imported inflation from deteriorating terms of trade.

(3a-d are optional, but frequently observed.)

For an example of the whole chain of events, see Indochina in the '90s. Russia skipped over the deflation bit and went straight to the catastrophe merchants at the IMF (with predictable results), while Greece has not yet dropped its overvalued currency peg, and so is still in the deflation phase.

The inflation here is caused by deteriorating terms of trade and has nothing to do with domestic supply and demand except inasmuch as it touches upon the foreign balance.

- Jake

John Harvey said...

VIMOTHY: "No, the fact that inflation is a monetary phenomenon does not rely on the ability of the central bank to exogenously set the supply of inside money wherever it likes."

Then what does it rely on.

vimothy said...

"Because the premise is false. Governments can by definition never experience a collapse in their ability to serve debts denominated in the currency they themselves issue"

Er, wat? Obviously, a government can experience a collapse in its ability to aquire real assets via taxation.

But it's irrelevant in any case. Even if we're in your MMT derived alterna-world where the govt faces no real constraint (I'm pretty sure that this is not actually an MMT position, BTW), all that it would take would be a market belief that the govt is going to default, and people would try to economise on their holdings of its money. I.e., this could happen based purely on animal spirits and without any basis in fundamentals.

You asked for a plausible mechanism. I gave you one.

JakeS said...

"Er, wat? Obviously, a government can experience a collapse in its ability to aquire real assets via taxation."

But that is irrelevant to its ability to pay soft-currency debts.

Yes, a collapse of the real, productive economy can cause inflation. But as a practical matter, economic collapse is usually associated with (and caused by) deflation.

"But it's irrelevant in any case. Even if we're in your MMT derived alterna-world where the govt faces no real constraint (I'm pretty sure that this is not actually an MMT position, BTW),"

No, the MMT position is that the government faces only real (and political) constraints - which means that soft currency debt (which is only nominal, not real) is never a constraint on government behaviour in and of itself.

"all that it would take would be a market belief that the govt is going to default, and people would try to economise on their holdings of its money."

Uh, no.

Sovereigns do not typically default on cash. They occasionally default on bonds, and "the market" may believe that they will do so based on animal spirits.

But, well, that doesn't matter. If people shift out of government bonds into cash, then the CB will just supply that cash and then absorb it again through the support rate on excess reserves.

If people want to shift out of the currency altogether, then you do get inflation. But that's from deteriorating terms of trade, not from an excess of soft currency. (This, by the way, is an empirically testable claim - if the price increases on imported goods lead price increases on local goods, the inflation is imported; if the reverse, the inflation is local.)

- Jake

vimothy said...

"But that is irrelevant to its ability to pay soft-currency debts."

No it isn't. Of course, the govt can issue as much currency as it likes. So what? If I hold its debt, this is not something that fills me with confidence about the govt's ability to honour its obligations in real terms.

"Yes, a collapse of the real, productive economy can cause inflation"

How?

"Sovereigns do not typically default on cash."

Who said anything about defaulting on cash? The issue is its real value. If the govt cannot honour its debts in real terms then it must create new money to do so by definition. Indeed, you seem to champion this as a positive feature! Given that fact, if govt cannot cover its debts in real terms, people will expect the value of money to fall and hold less of it.

"They occasionally default on bonds, and "the market" may believe that they will do so based on animal spirits."

Occasionally? For the period 1970-2000, we averaged 4 sovereign defaults a year. See e.g. De Paoli and Saporta (2006), ‘Output costs of sovereign defaults: some empirical estimates’.

John Harvey said...

VIMOTHY: "No, the fact that inflation is a monetary phenomenon does not rely on the ability of the central bank to exogenously set the supply of inside money wherever it likes."

Then what does it rely on.

JakeS said...

"If I hold its debt, this is not something that fills me with confidence about the govt's ability to honour its obligations in real terms."

But that does not allow you, as a vendor, to raise your prices. Nor for that matter do you, as a vendor, particularly care.

A vendor can only raise his prices when more people want to buy his goods than his competitors can reliably supply.

""Yes, a collapse of the real, productive economy can cause inflation"

How?"

Typically through an inability to procure strategically important imports, or honour hard-currency obligations.

"Given that fact, if govt cannot cover its debts in real terms, people will expect the value of money to fall and hold less of it."

Eh, no.

First, you are implying that people will divest themselves massively of government securities if the interest rate drops from 5 % to 0 %. This is simply not observed in the real world.

And that is the only effect of the government unilaterally deciding to not roll over bonds. The nominal interest rate drops to zero. In fact, it doesn't even have to do that, because the CB can pay a support rate.

There is nothing magical about cash that distinguishes it from a zero per cent interest bond in terms of contributing to aggregate demand. If having bond interest rate drop from 5 % to 2½ % does not cause bondholders to go on a massive spending binge, there is little reason to believe that having bond interest rates drop from 2½ % to zero will cause them to do so either.

Further, the bulk of all money is bank money. Since government money can crowd out bank money, you'll get demand-pull inflation from the real resources that the government procures with its spending (and from the Keynesian multiplier effect) long before you'll get a loss of value due to excess liquidity, unless you are spending the money in highly peculiar (and criminally stupid) ways.

If I have a € 100 business plan that is remunerative in real terms and I have € 20, then I need to borrow € 80 from the bank. If the government pays me € 10, I only need to borrow € 70 from the bank, but the end result is still that I invest € 100. The share of bank money in the total money supply drops, but my contribution to aggregate demand remains the same: € 100. And since only demand can drive inflation (remember, the vendor only sees his sales volume at the price he charges - he cannot raise prices just because he is an inflation chickenhawk), those € 10 of government payments do not create any inflationary pressure. The real goods and services that the government bought from me with those € 10 can create inflationary pressure, but only if the economy is operating at capacity.

The government paying me € 10 ticks off the banksters, of course, because they accrue money and power from controlling my ability to invest. But ticking off the banksters is not a problem for the productive economy.

Running a deficit today will not create inflation tomorrow. It may create inflation today, if the economy is operating at capacity, but there's simply no mechanism for it to create inflation tomorrow.

"For the period 1970-2000, we averaged 4 sovereign defaults a year."

On hard or soft currency debts?

Nobody disputes that countries can be insolvent in hard currency. But that is hardly relevant to the present discussion.

- Jake

Deus-DJ said...

Vimothy: I hope you've had fun trolling here, but some of us know that you're simply trying to become a defender of an indefensible neoclassical economics against a group of people who deem you rational in your conversations due to whatever intelligence you may exhibit. But you are nothing but a troll, and nothing but a neoclassical hack(of the worst variety, too). Get your sorry ass out of here, if you want to troll about something try not to make it about economics.

Daniel Conceicao said...

Wait, Vimothy, don't go yet. I'm still curious for the answer to John Harvey's question. It should be entertaining...

"VIMOTHY: "No, the fact that inflation is a monetary phenomenon does not rely on the ability of the central bank to exogenously set the supply of inside money wherever it likes."

John Harvey understandably asked:

"Then what does it rely on." (????)

And you never answered. Why?