Saturday, September 29, 2012

Steve Keen — Keen 2012 UMKC Reconciling MMT and MCT (video)


My presentation at the UMKC Post Keynesian conference in 2012 where I prove that, given endogenous money, effective demand is income plus the change in debt, and show that this is compatible with sectoral balances.
Steve Keen
Keen 2012 UMKC Reconciling MMT and MCT

159 comments:

Ramanan said...

Total hodgepodge.

First he screws the definition of aggregate demand. Second he changes the standard definition of "expenditure". (i.e., calls purchases of financial assets as expenditure).

This results in reconciliation of his own inconsistencies!

Anonymous said...

In complete agreement with Ramanan--this doesn't make any sense whatsoever.

beowulf said...

Ramanan, I'd love to see you drill down into his presentation on your site.

Dan Lynch said...

Made sense to me. He just took the static MMT models and used derivatives to create the dynamic models.

This would be a routine homework problem for a junior or senior level engineering student.

All models use simplifying assumptions to make them manageable, so I'm not saying Keen's model is perfect, if you wanted to pick nits. But is it accurate enough to give useful results ? It looked that way.

Anonymous said...

Well, of course he has changed the standard definitions of aggregate demand, and built financial transactions into the macro accounting! That's the whole point of his theory - to be able to describe problems of financial instability by building the financial sector into the model. The non-existence of a financial sector in standard macro is regarded as a major deficiency by the Minskyan tradition.

If you have some serious criticisms, fine. But simply noting these novel features is like saying that special relativity is all wet because it replaces the classical definition of momentum with the new concept of relativistic momentum.

Ramanan said...

"If you have some serious criticisms, fine. But simply noting these novel features is like saying that special relativity is all wet because it replaces the classical definition of momentum with the new concept of relativistic momentum."

Well, the presentation contains simple errors. Please don't compare it to Special Relativity!

One example: In the discussion around "discontinuities", Keen forgets that expenditure creates income!

He doesn't have to reinvent the wheel. Wynne Godley's approach is a good one.

In some of Wynne Godley's models, one has sectors purchasing financial assets by incurring liabilities. That's the right way to approach it.

In fact there is a Bank of England paper on this and how to study this empirically etc:

http://www.bankofengland.co.uk/publications/Documents/fsr/fs_paper10.pdf

which is based on Godley's approach.

That's the right way to approach it.

Ramanan said...

Beowulf,

It is difficult but can try. Will wait for his paper. I also wanted to look at his model which has some banking license as asset or something like that.

The reason it is difficult is that there is so much of new definitioning - without there being a dictionary from go from standard definitions to new ones. Finally Keen ends up talking to himself with his models. When not doing models, he is fun.

JKH said...

Steve Keen:

Always enjoy your presentations. Some comments which are basically repeats from several years ago, and more recently on your blog:

Your debt dynamic description is intuitively and empirically sensible, and that’s the important issue of substance, IMO.

The logical and definitional framework remains a bit awkward in my view. That's form only, but form can clarify.

Expenditure and income should be viewed as always equivalent at a fundamental level, ex post AND ex ante (on an ex ante scenario or expectational or planned basis).

The “discontinuity” here is about debt creation – not expenditure.

And it is a relative discontinuity, given that double entry accounting is always and everywhere a discrete sequence of “accounting events”.

Debt creation is a balance sheet “accounting event”.

Expenditure and income are income statement “accounting events”.

The two are separate.

There is an obvious intuitive and empirical causation and association between these separate things, as you describe.

But to suggest an equation or identity along the lines of your aggregate demand or effective demand definition is not ideal, IMO. It is a blend of accounting rigor with causative heuristics.

I’m not sure what the best terminology should be, but the point is that expenditure is always equal to income from a logical accounting standpoint. Ex ante projections, estimates, or scenarios of accounting outcomes don’t change the required equivalence. Nor does the insertion of a "discontinuous debt injection" which acts as a catalyst for increased expenditure. That increased expenditure must be matched by increased income under any consistent, coherent imagined scenario for the future.

Again, the “event” of debt creation is separate from expenditure and related income. The fact that debt creation becomes a catalyst for ex ante or future expenditure and income doesn’t negate this separateness; nor does it negate the equivalence between expenditure and resulting income, whether in the past, in the present, or in the future.

With all due respect, I don’t think Lebesgue integration is required to prove or disprove these logical accounting relationships.

The other point is that while the debt dynamics are intuitively reasonable and empirically obvious, casting the debt relationship as an equation is overkill, IMO. It is so because it is quite conceivable to achieve economic growth with higher money velocity instead of debt creation, at the margin. That again is a matter of empirical demonstration. So your debt dynamic is heuristically dominant, but theoretically impure, IMO.

Again, that’s all just a comment on form rather than substance. I’d tweak your definitional framework a bit.

Excellent presentation as usual though.

Clonal said...
This comment has been removed by the author.
Clonal said...

Matheus presentation was after Keen's - and can be found on his blog - he also had some thoughts on Lord Skidelsky's keynote address.

Dan Lynch said...

Re: "casting the debt relationship as an equation is overkill"

Steve's dynamic equation is overkill for you and I to use in our everyday thinking about economics -- we already have an intuitive understanding of debt and sectoral balances -- but it was necessary for Steve's computer program, since computers are not intuitive.

My background is in engineering, not accounting, so I can't comment on the accounting issues that people have raised. But Steve's use of derivatives to create a mathematical model of a dynamic system makes perfect sense to me. I did similar exercises routinely as an engineering student.

Almost all mathematical models are flawed by simplifying assumptions. The simplifications are necessary to keep the math manageable. As long as you state your assumptions and simplifications, and as long as the end result produces useful accuracy, it is considered a success.

That is not to say that someone may not be able to come up with a more accurate computer model.

Tom Hickey said...

Almost all mathematical models are flawed by simplifying assumptions. The simplifications are necessary to keep the math manageable. As long as you state your assumptions and simplifications, and as long as the end result produces useful accuracy, it is considered a success.

s I understand it, this is the difficult with modeling in social sciences, including economics and especially macro, which is now global in scope. It's similar to modeling atmospheric phenomena with the added complication of the human element.

My impression is that any adequate macroeconomic model is necessarily a computational nightmare involving complexity. Many economists have realized this an avoided the issue by simpler assumptions that sacrifice representational power to computational convenience. My conclusion is that until a model is developed that actually works to represent that which is purportedly model and confirmed by successful prediction, economists are whistling the wind and policy formulations based on flawed models are unlikely to be helpful.

On the other hand, there is already in existence a pretty standard accounting model with adequate "microfoundations" in that the data of aggregates in comprised of an immense number of individual transactions recorded on the books of firms, financial institutions, households, governments, etc.

While a huge amount of data is involved, a great deal of it is available already digitized so that the problem is not some much computational as data processing and reporting, which are much simpler and more tractable operations, using already established accounting rules and standard procedures.

So it seems to me like Godley's approach is the one to follow until a computational model addressing complexity proves superior.

Ramanan said...

"But Steve's use of derivatives to create a mathematical model of a dynamic system makes perfect sense to me. I did similar exercises routinely as an engineering student."

Dan (Lynch),

It is true that differential equations are powerful but they need to be used in a proper manner.

During the Krugman-Keen debate, there was a blog post here:

http://rwer.wordpress.com/2012/03/29/keen-krugman-and-national-accounting/

Keen is in a way right if he uses something other than "aggregate demand" as terminology.

What he says can be recast as an accounting identity in standard language. However that accounting identity cannot be taken as a behavioural hypothesis. Keen somehow mixes an identity with behaviour.

For example in Keen's model, it seems it is irrelevant how much of borrowing is for purchases of goods and services and how much for purchases of financial assets.



Tom Hickey said...

I should clarify my view about economic modeling. I am not opposed to any sort of modeling or any other kind of research. What I am opposed to is using inadequate justifications for policy formulation, especially when the justification is hyped beyond what it entails.

This is an issue with modeling based on computational convenience because the plethora of data or degree of complexity is intractable without simplification. There is also the issue of inadequate or "bad" data (typical where data is not collected and processed at a high standard) and also intentionally spurious data, e.g, due to misreporting and fraud.

There is the further issue of arguing based on computational convenience. A great deal of the neoliberal prejudice against government intrusion is based on this inconvenience. If exogenous influence is eliminated, modeling becomes much more tractable. So, the faux argument runs, at least implicitly, that government intervention should be minimized in order to "allow nature to take its course" iaw natural laws such as supply and demand as economic determinants of price, investment, and production (since we know how to handle this in our models).

The basic idea is that if exogenous forces like government are reduced, then the simple model become representational — so in the interest of better explanatory and predictive power, government should be minimized — for computations convenience. Similarly, finance. Modeling finance in economic models a problem, then make money a neutral veil. No matter than government cannot be eliminated or policy made constant, or that finance is a player in the game rather than a spectator.

If we are going to use something in policy formulation, then it should be used iaw its capabilities and limitations, rather than as an excuse for magical thinking. Just as magic was the bane of so-called primitive society, and dogma of religious society, math and science can be the bane of contemporary society when they are used beyond their limits. Then it becomes just another argument from authority rather than substance.

Finally, we have to admit the limitations of modeling in terms of the quality of data. GIGO. Moreover, data may not be as represented. Some economists try to account for the crisis as an "unforeseeable shock." Yet, there was ample warning of hidden data (off-book) and misrepresentation (fraud). The FBI warned of rampant fraud in the mortgage industry in 2005, for instance. Unless modeling can accomodate this the output will be garbage.

Magpie said...

@Ramanan @vimothy

While I am not convinced by Keen's presentation (I'll go into it in a moment), I find your criticism extremely unhelpful.

Similar goes to vimothy, with one additional consideration. This is taken from one of your responses (July 30, 2012 at 07:32 AM) to Nick Rowe's "Can you please read a first year textbook??"

"I used to be pretty invested in the idea that the economics mainstream was out to lunch. (...)

There were so many ways in which the mainstream was wrong, I found it kind of remarkable that the situation had ever reached the point it had. (...)

(...) If only they understood how banks work! Etc, etc. I was probably pretty insufferable.

"Gradually, though, I came to realise that—quelle surprise—the image of mainstream economics that I’d built up was a caricature. It turned out that I was mostly just completely ignorant of it. (...)"

That sounds pretty reasonable to me (and, to be honest, I am following Rowe's challenge).

Unfortunately, it seems all that good sense applies ONLY to criticism of naoclassical economics...

---------

But this is my criticism of Keen's presentation. He makes a valid distinction between ex ante and ex post expenditure (and aggregate demand).

But this distinction is not new. It's taken into account in the System of National Accounts (dealing with GDP): that's one of the functions of the item "inventory changes".

What I imagine Keen is trying to do is to link GDP accounts (the "equivalent" of a firm's income statement) to the National Balance Sheet accounts. In principle, this could be done (as it can be done with individual firms), except that National Balance Sheet accounts do not include financial assets/liabilities (it would be double counting: my liability is someone else's asset).

And, maybe it was me, but I can't see how he proposes to do that.

---------

This is Rowe's blog
http://worthwhile.typepad.com/worthwhile_canadian_initi/2012/07/can-you-please-read-a-first-year-textbook/comments/page/2/#comments

Ramanan said...

Magpie,

Not sure why unhelpful.

It is counterproductive to define aggregate demand as GDP plus change in debt.

Second for a closed economy

Income = Expenditure.

Third, in the discussion of discontinuities, (as I mention in the comment @ September 30, 2012 12:52 PM), Keen forgets that the expenditure creates income.

"except that National Balance Sheet accounts do not include financial assets/liabilities"

At a consolidated level, for a closed economy, it is zero. So,

Y = E

For a single sector it needn't be true but if one sums over all sectors it is true.

So while Keen is partly right saying for a single sector there is a term Net Lending (standard terminology, using Keen's language can be counterproductive), when summed over all sectors these cancel out which does not happen in the case of Keen.

It is actually in the case of an open economy that such a thing which looks like Keen's equations appear.

So a nation's GDP is different from its GNI and the difference is the increase in net indebtedness to foreigners. This is the sum of increase in net indebtedness of the private sector and the government sector to foreigners.

But open economy unlike the closed economy case of Keen.

Keen is basically trying to model the intuition that there can be debt-fuelled growth but both his accounting and his model need a lot of correction.

Magpie said...

Why it was unhelpful?

Your first comment, was 34 words long. This one I am replying to is 196. Doesn't this simple fact give you a hint?

More precisely, ask yourself: was there any attempt at a reasoning in those 34 words? Don't need to answer to me; answer to yourself.
---------
Now, let's go little by little.

"It is counterproductive to define aggregate demand as GDP plus change in debt."

I find it odd, too.
---------
"Second for a closed economy"
"Income = Expenditure."

If we are going to remark on the obvious, that's wrong:
Income = Expenditure = Value Added.
---------
"Third, in the discussion of discontinuities, (as I mention in the comment @ September 30, 2012 12:52 PM), Keen forgets that the expenditure creates income."

Frankly, that bit did not make much sense to me, either. But I don't think he forgets it: what I guess he is trying to say is that households' incomes fall short of their own ex post expenditure. The shortfall being covered by household debt. This is what the talk about ex ante, ex post seems to be heading to.
---------
" 'except that National Balance Sheet accounts do not include financial assets/liabilities'"

"At a consolidated level, for a closed economy, it is zero. So,
"Y = E"

No. Y, E are categories of the GDP accounts (flows). Financial assets/liabilities, if they were included, would be categories of the national balance sheet (stocks). They are not included in national balance sheets for the same reason intermediate consumption is not included in GDP: to include them would be double counting.

Incidentally, although excluding financial assets/liabilities in national balance sheets is the current standard practice, there were (are?) proposals to provide these statistics in a "national balance sheet", with the proviso that the double counting occurred.

More generally, what I understand Keen is trying to do is to explain GDP growth as partly funded by household debt, so he is not talking about the consolidated level.

To explain net operating profits for a firm with its leverage makes sense. The question is how do you extend this to the economy as a whole? That's not obvious to me, and I don't think Keen provided an answer.
---------
"For a single sector it needn't be true but if one sums over all sectors it is true."
"So while Keen is partly right saying for a single sector there is a term Net Lending (standard terminology, using Keen's language can be counterproductive), when summed over all sectors these cancel out which does not happen in the case of Keen."

I am not sure I follow you here, but I'll suppose you are talking about households. If I am correct in this assumption, that precisely appears to be what Keen is trying to do (see my "consolidated level" remark above). For him the key seems to be (and this is my interpretation only) how debt is distributed within a country. That, it seems to me, is why he does not use "Net Lending": he is talking about household debt.

Incidentally, who is indebted and who isn't within a national economy should be important, in my opinion.
---------
Like I said, I am not convinced. The talk was too short; many things were left floating around. One is left to try and make sense of what was said.

Ramanan said...

"Your first comment, was 34 words long. This one I am replying to is 196. Doesn't this simple fact give you a hint?"

Well, had my comment been 34196 words long, would it have made it sexier?

Just to give a background, many of us have been telling Steve Keen that his accounting is not so right.

Now some have a more positive tone but I was disappointed by the fact that this has become an issue to deserve a talk in a conference. So the tone.

"Incidentally, although excluding financial assets/liabilities in national balance sheets is the current standard practice, there were (are?) proposals to provide these statistics in a "national balance sheet", with the proviso that the double counting occurred."

That's wrong.

First at a consolidated level, national balance sheets have claims on nonresidents and nonresidents' claims on residents.

Second, at a deconsolidated level, national balance sheets have sectoral claims on each other.

"That, it seems to me, is why he does not use "Net Lending": he is talking about household debt."

Keen has no track of whether he is talking of an individual sector or he is talking of the economy as a whole. He in fact mixes the two.

"No. Y, E are categories of the GDP accounts (flows). Financial assets/liabilities, if they were included, would be categories of the national balance sheet (stocks). They are not included in national balance sheets for the same reason intermediate consumption is not included in GDP: to include them would be double counting."

Now why mix so many things here? Why are you mixing stocks and flows?

"If we are going to remark on the obvious, that's wrong:
Income = Expenditure = Value Added."

Sorry what?

I think you are mixing up the "income=expenditure" approach which defines expenditure to include change in inventories (added to final purchases) so that income=expenditure.

Now, the point is that Keen himself is using this approach but claims that there is a correction which is change in debt. Just look at his equations, the change in debt is both private and public. If you actually know the numbers for the United States, you will see an order of magnitudes of difference between change in inventories (your point) versus the actual change in debt.

Anyways the point is that Keen has his own terminologies and own accounting. If you haven't realized it by now, there is no point in debating about this.

My understanding is that Keen somehow treats going into debt as a reduction of income.

If I borrow to consume these are two events. The first is borrowing and the second is consumption. The first event does not increase my net debt. Neither does it increase the net debt of the economy.

The second event increases my net indebtedness (if I am already in debt) but at the same time, provides income to someone else.

Keen takes into account my going into debt and misses out someone getting income as the result of my consumption.

paul meli said...

"My understanding is that Keen somehow treats going into debt as a reduction of income."

It is, as soon as payments are made. Loan payments subtract from income, or if you prefer, removes spending from the non-government.

The reduction in spending continues until the loan is paid off in an amount equal to the original spending plus an additional amount in accrued interest.

Some of the interest acrues to the non-government side of the bank to which the payments are made as income.

Tom Hickey said...

I don't think that it is quite correct to say that private debt repayment "reduces income" in that all expenditure is made from income, either 1) current, or 2) past (drawing down savings or selling assets purchased with past income, albeit in some cases inherited), or 3) future (private debt drawing future income forward).

Magpie said...

@ramanan

"Well, had my comment been 34196 words long, would it have made it sexier?"

No, but by mere chance perhaps you would have been able to say something intelligent.

Anonymous said...

Hi Magpie,

I’m sorry that you found my comment so unhelpful. It’s hard, when writing a comment, to take into account all of the possible audiences for it; so generally, I just write whatever comes to mind. As Alan Ginsberg said: “First thought, best thought.” Sometimes this results in very long comments, and sometimes it results in ones that are much shorter. Sometimes I post very few comments and sometimes I post lots in threads that stretch on into the hundreds. I tend to think that it’s the longer comments or comment threads that people find the most tedious, and everyoen’s time is limited anyway, so I try to keep mine “short and fat like Joe Pesci.” It’s not a perfect heuristic, nor do I always hit it squarely, but when I shoot, that’s what I aim for.

I’m afraid that I didn’t understand your reference to my comment at Nick Rowe’s blog, but I think it’s cool that you’re taking up his challenge. One of the best things about the internet is how easy it is to get your hands on free scanned textbooks, or very cheap second hand versions from Amazon Market Place. (My girlfriend, who has to live with an ever growing library of books she finds incredibly boring, might disagree with this.)

My issue with Keen’s “Law” is that he’s trying to solve a problem that isn’t actually a problem, in such a way that makes no sense. Yes, people make plans, and not all of these plans can be consistent with one another. No, this doesn’t mean that you should butcher the GDP identity. Yes, debt affects aggregate demand. No, aggregate demand is not “income plus the change in debt”. (Aggregate demand is nominal aggregate expenditure on final goods and services—allowing for international trade, by definition this is equal to aggregate income.) Yes, different levels of AD are possible. No, there is not one level before the fact and one level after. (Aggregate demand in the ADAS model is a curve.)

paul meli said...

Tom,

If the claim is made that credit increases or adds to income when issued, then it MUST be true that the repayments subtract from or decrease income.

I prefer (it's more accurate) to say that credit increases spending when issued and decreases spending when repaid.

Mathematically the two statements are identical, depending on ones definition of income.

Once again semantic judgements cloud the abstract issue, which is the one that matters.

Now, if you are saying that WHEN WE SPEAK TO LAYMEN we should be more careful in our wording, I agree completely, that's why I prefer using spending instead of Income, but that is not the issue wrt my comment previously.

Income is a placeholder for spending, it is one source of spending. but does not directly factor into economic activity.

Anonymous said...

If the claim is made that credit increases or adds to income when issued, then it MUST be true that the repayments subtract from or decrease income.

Interest constitutes income for the recipient. And credit doesn’t add to anyone's income unless someone draws down on it and spends. Thus, both the antecedent and the consequent in this argument are false. (Logically speaking, the argument is vacuously true.)

paul meli said...

vimothy,

"Interest constitutes income for the recipient."

Already stated this in another thread. It is also trivially obvious.

Here referring to principal. It may be a trivial observation (real-world relationships usually are) but it is fundamental and appears to be missed by most so-called knowledgable people.

"And credit doesn’t add to anyone's income unless someone draws down on it and spends."

…and I referred to this in the very comment you are criticising. The assumption is that the credit HAS BEEN ISSUED AS A LOAN, so an asset and liability now exists where there wasn't one previously. Much of the loans issued are spent immediatly on consumer goods when issued to consumers.

Apparently you didn't even bother to read and understand my comment before reacting.

Better criticism, please.

Anonymous said...

As ever, Paul, your comments are jumbles of noise.

What you wrote was,

If the claim is made that credit increases or adds to income when issued, then it MUST be true that the repayments subtract from or decrease income.

I prefer (it's more accurate) to say that credit increases spending when issued and decreases spending when repaid.

Mathematically the two statements are identical, depending on ones definition of income.


The two statements are not identical.

Re the first statement, a loan does not add to the income of the borrower. Repayment of principle does not add to income of the lender. That is standard. Payment of interest adds to the income of the lender, but does not subtract from the income of the borrower. That is also standard. Credit does not “add to income when it is issued”. If I borrow $1000 from my bank, no one’s income has increased. If I spend $500 of that loan on a new laptop, then (in the aggregate) my expenditure has contributed to someone’s income.

Re the second statement, credit increases spending when credit increases spending, not “when issued”. When debts (not credit) are repaid, purchasing power is redistributed, not reduced. What effect this has on the overall level of expenditure on the economy is an empirical question and cannot decided before the fact with such generality.

paul meli said...

vimothy,

Only to you and your ilk, vimothy.

You appear to only see words, not meaning or ideas.

"a loan does not add to the income of the borrower."

I didn't say that it did, I said "If one makes the claim that it does, then the opposite follows" logically.


"Repayment of principle does not add to income of the lender."

No kidding?

"but does not subtract from the income of the borrower"

You've got to be kidding. It subtracts from the borrowers spending, which I am repeating here for at least the third time. The net effect of the payment is a reduction in spending.

"If I spend $500 of that loan on a new laptop, then (in the aggregate) my expenditure has contributed to someone’s income. "

It increases the borrowers spending, when the borrower follows through and spends the loan, which after all is th epoint of borrowing. I have said that multiple times also.

"credit increases spending when credit increases spending, not “when issued”"

No Kidding? Only a moron would not understand the implication of my comment.

"When debts (not credit) are repaid, purchasing power is redistributed, not reduced.

Redistributed to whom? This is just plain wrong. When the debt is repaid, a liability is satisfied. The funds used to satisfy the debt come from income or savings. Income spent is normally spending, in this case it is nothing.

"What effect this has on the overall level of expenditure on the economy is an empirical question and cannot decided before the fact with such generality."

Really? This has been another edition of a discussion with a math-challenged individual.

Here's a generalization for you. If Income is used to pay down debt, that income does not add to the overall level of expenditure.




Tom Hickey said...

Let's get back on track rather than personally invested. That's not aimed at anyone in particular.

I thought JKH's comment was exemplary as a positive approach to offering constructive criticism. It's a good model to keep in mind.

The issues here are twofold:

First is the progress on Steve's modeling project to integrate MCT and MMT.

Scott communicated to me that this project is on target and the MMT economists are presently working not only with Steve but with others of similar mind to present a more coordinated PKE stance in contradistinction to the mainstream instead of continually bickering with each other over minor disagreements while the world is burning.

Secondly, what we have targeted as needed from MMT proponents is a narrative to present the MMT-MCT-PKE viewpoint to non-economists in a way that clear, concise and precise without being too technical for nono-economists to understand. I would say aiming for 12th grade competence is about right here. That means not departing from commonly understood meaning of terms. this is a particular challenge since economics and finance-accounting often use such terms technically with somewhat different denotation and connotation than ordinary language usage.

So for us, the issue is not so much over who is right but rather coming up with an approach that will advance policy needed to correct course before the people now in charge drive us over the cliff into the abyss.

Anonymous said...

I’m both amused and intrigued by the idea that others “of my ilk” are also reading this thread and silently disapproving of your comments. Speak, brothers!

I didn't say that it did, I said "If one makes the claim that it does, then the opposite follows" logically.

Anything at all can “follow logically”, because the antecedent is false. “If the moon is made of cheese, then loan repayments subtract from income”. Equally true, equally vacuous.

No kidding?

It’s not trivial for me to tell the difference between the statements that you are making that you know are false and are just throwing out there to keep us on our toes and the statements that you know are true and stand by.

You make one argment, call it A, and another, call it B. According to you, A and B are “mathematically equivalent”. Except you don’t believe that A tells us anything useful, you’re just kicking it around for no reason. So what does that say for argument B?

You've got to be kidding. It subtracts from the borrowers spending, which I am repeating here for at least the third time. The net effect of the payment is a reduction in spending.

I’m told that in theory you can repeat a thing an infinite number of times without it ever becoming true.

Notice that what I wrote was not “loan repayments do not subtract from the spending of the borrower”, but rather, “loan repayments do not subtract from the income of the borrower”. Whether loan repayments in fact reduce the spending of the borrower depends on a host of other things. What is the baseline comparison that you are making? What are you holding constant and what are you allowing to adjust?

For example, it depends on the use that the loan was put to. Imagine that I am renting a flat. Every month, I pay my landlord £1000 for the privilege. Then I decide to buy a house. I take out a mortgage that I repay at £1000 a month. My spending on other goods need not change at all. It might even go up, since I figure I can afford to save less. Or not. It’s hard to say, because people have minds of their own.

No Kidding? Only a moron would not understand the implication of my comment.

If what you mean is different from what you say, it’s not always easy for your listeners to parse.

Redistributed to whom? This is just plain wrong. When the debt is repaid, a liability is satisfied. The funds used to satisfy the debt come from income or savings. Income spent is normally spending, in this case it is nothing.

Say that I owe you a hundred bucks. When I repay you, my liability is extinguished, and you have a hundred bucks. Now you have the option of deciding whether to save that hundred bucks, or spend it, which is the same option I had when it was mine.

Here's a generalization for you. If Income is used to pay down debt, that income does not add to the overall level of expenditure.

Q: How do you know that? A: You don’t. If income is used to pay down debt, then the purchasing power of borrowers is reduced and the purchasing power of lenders is increased. That money goes somewhere, i.e., the system is consistent. What the recipients chose to do with it is an open question. (In equilibrium, aggregate expenditure and income are equal so if something is not spent or invested then it is not earnt, in an aggregate sense. That is, it doesn’t exist.)

paul meli said...

vimothy, can't you see I'm busy here?

Folks can disagree all they want, they just need to bring a credible argument to the table, which you haven't done.

Where you aren't just plain wrong you are missing the point altogether.

And, as always you appeal to authority (other peoples thinking) rather than use your own thinking.

You and I are done.

Ramanan said...

Magpie: "No, but by mere chance perhaps you would have been able to say something intelligent."

Ha!

JKH said...

Follow up:

I think my instinct is to revise it roughly as follows:

Look upward from sector balances and model the framework on the basis of a single, closed, global economy. The sector work can proceed later.

In that single economy, I’d say that expenditure equals income ex post.

In “modeling the future” or in constructing future scenarios or in looking at risk going forward, however you want to look at all that, I’d say that any future scenario on a discrete or probabilistic basis must follow ex post accounting rules as a matter of logical necessity in the future, even though the future considered now is ex ante.

So in that sense, future expenditure equals future income, both ex ante.

I’d then insert debt as a discontinuity at the point now considered to be the present – i.e. the dividing line between the past and the future.

And then I would model the effect of that debt in terms of its impact on something called aggregate demand or effective demand or whatever.

But that demand function should be specific to the duration of the future that is being modelled. And any such future is “closed” in the sense that future expenditure equals future income.

So the debt propels aggregate demand and an increase in future expenditure and income versus past expenditure and income. And if you like you can model that as future expenditure being equal to past income plus the marginal effect of the debt discontinuity on future expenditure and income. But that time split must be explicit. There should be no ambiguity in the sense of implying (falsely) that future expenditure won’t be equal to future income, merely due to a present debt discontinuity.

On the effect of debt, it seems to me that can happen in two ways. It can increase money supply if its bank debt. Or it can increase velocity of existing money if it is non-bank debt. You don’t really need a refined definition of money to see that.

On an equation of any sort, there are two complications. One is the complex empirical patterns of the different money supply and money velocity effects of debt as noted. The other is the secular trend in the horizontal compounding of the debt of financial firms. The creation of CDOs is a good example, where CDOs are debt created by portfolios of other debt. This horizontal compounding feature is a big aspect of gross debt trends over the past 30 years, I think. And that has to be considered in whatever reaction function you might be considering in an equation of the sort you model.

Unknown said...

"Say that I owe you a hundred bucks. When I repay you, my liability is extinguished, and you have a hundred bucks"

But in the case of a bank loan, that 'hundred bucks' is a liability of the bank, which also gets extinguished when the loan is repaid.

Anonymous said...

Paul,

Nowhere in my comment did I appeal to anyone else's thinking or authority.

In general, my belief is that reasonable people will disagree. That is, the exercise of reason results in disagreement. Unreasonable people also disagree, but when they do so it generally involves a lot more stamping and shouting and the spitting out of dummies.

One of the things that I like about this site is that it’s usually always possible to disagree with Tom (in particular, but it’s surely true of other regular commenters with whom I’m less familiar) in that rational fashion of two grown-ups who’ve come to their own conclusions about some matter and are able to discuss it without trading insults. With you it never seems possible. What you say is always obviously true and only a moron or a mathematically challenged individual could disagree.

The reason that I like to debate an issue with someone who takes a different view is that it is both a diverting intellectual exercise and it normally teaches me something, even if that something is only how to express my arguments in a more understandable way. (Alas, this Great Work is never completed.)

It’s also the case that most people change their minds about things, learn more and so on, as they go through life. If you have always understood economics and the economy much better than any economist because of whatever reason, then you are to be congratulated. For the rest of us, the long and arduous task of understanding starts at the bottom of a very tall hill and goes upwards. Have pity on your little brothers.

Tom Hickey said...

@ JKH

Way beyond my level of expertise but it sounds intuitively right to me.

1. Model the global economy as a closed economy and deconstruct from there. With global finance and the increasing share of multinationals, national economies are now "sectors" of the global economy.

2. Use standard accounting terminology and practice. We all need to be speaking the same language.

3. Acknowledge the distinction between bank and non-bank credit. Private debt is complicated and with derivatives intertwined exponentially so. Recognize the huge global implications that may be lurking.

4. When dealing with flows, specify periods, and observe SFC in modeling. should not need stating but apparently not all economists have gotten the message or follow it consistently.

I would add that is also necessary to recognize qualitative factors like best practices and being cognizant of potential "irregularities." A great deal of the present mess is due not only to private debt but also irregularities involved in its accumulation and distribution — as Bill Black and others have amply demonstrated and which many economists still don't get. It may be difficult to model this, but computational convenience is no reason to ignore its significance.

Anonymous said...

A paper with some interesting data is Schularick & Taylor (2009), "Credit booms gone bust: monetary policy, leverage cycles and financial crises, 1780-2008":

http://people.virginia.edu/~amt7u/papers/w15512.pdf

In particular, see the chart on page 10, which shows that credit in the post-war US grew much faster than money.

Matt Franko said...

Vim,

One time I challenged Tom on whether 2+2=4 was "true"....

Tom responded IIRC that (paraphrasing) that he probably thought it was true, but if somebody had a different idea, he would at least listen to what the person had to say.

I am still mulling this over... ;)

I think I know where Paul is coming from on this (my brain is probably similarly wired to Paul's) and it is extremely frustrating to see people discount or appear blind to what Bill Mitchell once described as "the tyanny of the math"... exteeeeemly frustrating you have no idea ....

Morality, higher order math modelling, economic sectarianism, etc... it all doesnt matter (to me) compared to some simple mathematical aspects of our monetary systems revealed thru MMT imo...

rsp,

Unknown said...

"But that time split must be explicit. There should be no ambiguity in the sense of implying (falsely) that future expenditure won’t be equal to future income."

Keen is explicit about that 'time split', and doesn't imply that future expenditure won't be equal to future income.

paul meli said...

vimothy, this is your first response to my comment:

"Interest constitutes income for the recipient. And credit doesn’t add to anyone's income unless someone draws down on it and spends. Thus, both the antecedent and the consequent in this argument are false. (Logically speaking, the argument is vacuously true.)"

Note the part in bold. I call that an insult, although merely a back-handed one. It was disrespectful. That set the tone for our conversation.

Tom in my wildest dreams would never make a comment like that even to an idiot.

This part:

"Interest constitutes income for the recipient. And credit doesn’t add to anyone's income unless someone draws down on it and spends"

…was implied from the get-go in my comment and frankly I don't see how my meaning it could have been misconstrued by a reasonable person. Why would someone take on debt if not to spend it? A significant subset of debt is spent on consumer spending immediately, stimulating demand.

Unless you just get a kick out of correcting someones sentence structure, this type of thing is a waste of time.

If you aren't sure what I or someone else means with a statement ask them to clarify it. Don't run with your misunderstanding and attack a strawman.

I comment here to discusss ideas, not words. Words and sentences are a cheap imitation of the "real thing™".

I prefer to assume that others are capable of abstract thoughts and the exchange of ideas based on them. I give everyone the benefit of the doubt until they prove unworthy.

Tom Hickey said...

Matt, One time I challenged Tom on whether 2+2=4 was "true"

What I likely was driving at is "assuming decimal arithmetic." Many people think it is true "intuitively." It's not. It's a matter of following rules.

A lot of the most contentious issues are over rules and following them, but most parties to the argument don't recognize this and resist it when pointed out. Logicians, mathematicians, lawyers and accountants get it right away tho.

Anonymous said...

Matt,

That's a neat story.

As you know, you can easily prove that 2+2=4, but you need to first make a set of assumptions about the natural numbers known as the Peano Axioms. Now, you could imagine a student who you were trying to teach this to say things like, "no, no, no, addition doesn't do that, it does this". Of course, you'd never get anywhere and it would be a waste of time for the both of you.

In your own private language words can mean whatever you want them to mean, Humpty Dumpty style. That's no good for communication, though.

paul meli said...

Tom,

"It's a matter of following rules."

Tom, from a different perspective, the "rules" have always been in place. They just "are" and always have existed.

Man didn't make the rules, he discovered them.

One can't "discover" something that doesn't already exist.

If man hadn't yet discovered the rules of aritmetic, they would still be.

Dan Lynch said...

Tom said "... observe SFC in modeling."

May I ask what is "SFC" ? :-)

While we are on the subject of communications and semantics, it's a good rule of thumb to avoid the use of acronyms and jargon, unless they are obvious to the audience, like "MMT." ;-)

I personally have no aptitude for accounting and am "turned off" by accounting language. Perhaps accounting terminology and rules are appropriate for beancounters, but not for creative thinkers.

On the other hand, I am comfortable with math. Math is the language of science and engineering. Engineers don't use double entry bookkeeping, they use math,plain english, and greek symbols. In fact, my technical writing professor discouraged the use of technical jargon, if it could be avoided.

However, as an engineering student, I was taught to list and define all my terms, symbols, and assumptions at the beginning of an exercise. Also, engineers almost always include a labeled drawing, even if the drawing is merely a "black box" with flows going in and out of the black box. Thus even though engineering terminology is not rigidly standardized like accounting terminology, it was always clearly spelled out so that the reader didn't have to guess at the meaning.

What I get out of the reaction to Keen's math is that people who are accustomed to thinking in static accounting terms had a hard time following it, while people who are fluent in math are OK with it.

That's not to say that there may not be bugs in Keen's work. That's why his "Minsky" program is still in beta.

I encourage those who have questions about Keen's assumptions to contact him and politely ask him to clarify. That's how peer review is supposed to work, isn't it ?



paul meli said...

Dan Lynch,

I'm not Tom but I spent the night in a Holiday Inn Express.

SFC = "stock flow consistency"

Tom Hickey said...

Man didn't make the rules, he discovered them.

This is an ongoing matter of debate in philosophy of logic, philosophy of mathematics, philosophy of science, anthropology, etc. Yours is one of the contending positions. The jury is still out.

Tom Hickey said...

@ Dan Lynch

Math may be the language of science but accounting is the language of money & banking, and finance and these spill into economics through monetary economics. Wynne Godley pioneered a new approach to macro modeling through stock-flow consistent models based on his monetary economics. MMT modeling is based on this.

The project of MMT is to integrate money & banking, finance, and economics into a single structure based on monetary economics using SFC modeling.

Steve Keen has agreed with this approach is integrating his approach to modeling with it. The recent UMKC conference was designed, at least in part, to further that collaboration.

Ramanan said...

"What I get out of the reaction to Keen's math is that people who are accustomed to thinking in static accounting terms had a hard time following it, while people who are fluent in math are OK with it."

Well, Keen's models are simple differential equations. There is no divine mathematics used.

"Perhaps accounting terminology and rules are appropriate for beancounters, but not for creative thinkers."

Dan, your comment sounds like you view accounting as just counting numbers or simply data entry. That is not really the case. It is actually a bit like "real analysis" of mathematics. Very rich.

Matt Franko said...

" Many people think it is true "intuitively." It's not. It's a matter of following rules."

This is part of my "mulling this over" Tom...

When I hear "2+2=4" I get an instant "visual" in my mind, like a picture, I would not describe my thought process as "rules based". This is hard for me to explain...

(FD via the old SATs: Math >700, Verbal <500)

Seems like brains wired heavily towards the math side "see" things differently... I think this is a large part of the current problem with govt officials who mostly have law degrees probably, so they are verbal/semantics vice mathematical....

So they hear from someone who they think is an "expert" like Greenspan going on Bloomberg just a few weeks ago saying "banks lend out the reserves" and they just FLAT OUT BELIEVE IT... having no tools in the box to use to truly test this statement as to it's accuracy... it's a math cognition "shortfall" they have imo...

rsp,

JKH said...

"That is not really the case. It is actually a bit like "real analysis" of mathematics. Very rich."

Much more like algebra than calculus, and algebra is a very deep area of pure mathematics.

Calculus is a fly on algebra's turd, but mainstream economists love the life style. It's why they avoid accounting outright, although they don't know that's the reason why.

Ramanan said...

Vim,

"Yes, people make plans, and not all of these plans can be consistent with one another. No, this doesn’t mean that you should butcher the GDP identity. Yes, debt affects aggregate demand. No, aggregate demand is not “income plus the change in debt”.

Good points. It is not possible always to write a detailed critique of something which is so obviously wrong to begin with.

It's ridiculous. Keen starts saying the approach "income=expenditure" is wrong but his definition of these terms is so wrong to begin with.

paul meli said...

"It is actually a bit like "real analysis" of mathematics."

This may be the result of a language barrier but accounting can only be simple arithmetic…adds and subtracts, not mathematics, which is a science. Accounting is bean-counting writ large. True, it is rules-based, but so is bean-counting.

There is no mathematics theory embedded in accounting. from Wikipedia:

"Accountancy is the process of communicating financial information about a business entity to users such as shareholders and managers.[1] The communication is generally in the form of financial statements that show in money terms the economic resources under the control of management; the art lies in selecting the information that is relevant to the user and is reliable.[2] The principles of accountancy are applied to business entities in three divisions of practical art, named accounting, bookkeeping, and auditing.[3]"

"Mathematics (from Greek μάθημα máthēma, “knowledge, study, learning”) is the abstract study of topics encompassing quantity,[2] structure,[3] space,[2] change,[4][5] and others;[6] it has no generally accepted definition.[7][8]
Mathematicians seek out patterns[9][10] and formulate new conjectures. Mathematicians resolve the truth or falsity of conjectures by mathematical proof."

Anonymous said...

Paul,

Your argument had the form P implies Q, or if P then Q. If we draw a truth table for this, we see that the argument is true everywhere except where P is true and Q is false.

In other words, if P is false, then the argument is true regardless of whether Q is true. Even if P and Q are both false, the argument is still true. Logicians and mathematicians call these types of arguments “vacuously true”.

My example of a vacuously true argument was,

"If the moon is made if cheese, then repayment of principle reduces income”.

Technically, this argument is true—but vacuously so, because the falsity of P provides no information about the truth of Q. “If Tony Blair is the King of Russia, then I am the greatest footballer of all time.” That’s great! Except Tony Blair isn’t the King of Russia and I’m not the greatest footballer of all time. (I think I’m probably up there, though.)

Here is a Wikipedia page on the subject: http://en.wikipedia.org/wiki/Vacuous_truth

This part:

"Interest constitutes income for the recipient. And credit doesn’t add to anyone's income unless someone draws down on it and spends"

…was implied from the get-go in my comment and frankly I don't see how my meaning it could have been misconstrued by a reasonable person. Why would someone take on debt if not to spend it? A significant subset of debt is spent on consumer spending immediately, stimulating demand.


Well, many things are possible, aren’t they? For one thing, debt and credit are not the same. Someone might want to be able to draw on some credit in the event of a future state of the world which might or might not occur. Someone might take out a loan and not spend all of it all at once.

The essence of what you are saying is that when people borrow, their purchasing power increases, so that as they repay their loan, their purchasing power should decrease. (Except you have tried to turn this rule of thumb into a law.)

The problem is that this is not necessarily true for the individual and not necessarily true on aggregate either, as I’ve tried to explain, even though intuitively it seems like it should be. The issue is what is being held constant and what is allowed to adjust. If you borrow to fund asset acquisition, then your income might (we hope) increase in excess of your repayment, or you might lower some other costs. Then, when you repay your creditor, they make choices as well.

Tom Hickey said...

paul, not mathematics, which is a science

Math is a branch of logic. Logical (mathematical) truth and falsity is syntactical and tautological rather than semantic and probabilistic. Math proof is demonstrated in terms of formation and transformation rules alone.

Scientific truth is ultimately semantic, arrived at through hypothesis-testing by experiment, although logic and math are used to get there. But consistency doesn't prove correspondence, only observation can confirm or disconfirm a scientific hypothesis or statement.

paul meli said...

vimothy,

"debt and credit are not the same."

OK you got me there, I should have written debt and not credit. Can we move on from there?

The implication should be clear to an abstract thinker that "sees" the embedded idea or pattern without reading a book-lenght description of it. It isn't that complicated, although you seem to be making it so.

I'm not going to go down any more of these semantic rabbit holes with you. If that's the way you are going to respond to my comments don't bother. I don't have the patience. Spmeone else can carry the hod.

I am interested in good-faith discussions over abstract and not-easily communicated issues while at the same time keeping the discussion as simple as possible. I don't want to play "gotcha". I interject to clarify things that look wrong to me.

There are several engineers and science-based commenters on the board that understand very well the points I am trying to make in general, as I tend to understand theirs.

When we disagree it doesn't take massive amounts of words to sort things out.

I am not trying to prove anything to you and my goal is not to "win".

I have a hard time letting things go that are mathematically false right out of the chute. With you I will try to make an exception.

paul meli said...

"paul, not mathematics, which is a science"

Tom, I realized I wrote something incorrect there, but let it go because this blog doesn't have an edit function after the post is published and I'm lazy. Within the little box we have to write in the beginning of the comment disappears pretty quickly.

The excerpt below from Wikipedia explains clearly what math is and isn't, so I figured folks would let it slide. I was wrong.

This is the kind of thing that is maddening to abstract-based thinkers. We don't need all of this nit-picking to understand what others mean. Semantic purity won't help us make the arguments to the general public anyway, so why be anal about it?

If one of us writes a little faux-paux but you still get the gist of what we are writing please, let it slide. If not, ask for a clarification.

Here, we are discussing ideas or theory not quantitative analysis, which is pointless if one doesn't have the theory right.

Since the ideas and theories aren't particularly complex, spelling and grammar-checking aren't particularly useful in the conveyance of ideas unless the audience isn't familiar with them in the first place.

I assume that here, most people are aware of the underlying theory of systems. I give everyone the benefit of the doubt. If not, they should ask and those of us that do will try to explain it to them.

You guys need to cut some slack.

Matt Franko said...

Hey Paul,

I think Warren understands "closed systems" from his latest:

"The answer begins with the absolute fact that government debt is equal to global ‘non government’
accumulations of euro financial assets. For any given ‘closed sector’ the euro is a traditional case of ‘inside money’, as with a ‘giro’ or ‘clearing house.’ The only way an agent could have net euro financial assets would be for another to be net borrowed."

And btw, I was having this conversation with 2 friends over the weekend, the one guy was a very smart computer science major and he was starting to get it while the other guy was being a bit obstinate and (I kid you not) the comp sci guy brought up the term "closed system" when he was trying to help me explain it to the other guy...

I think this term is important to include when explaining this to math/technical thinkers...

Rsp,

Anonymous said...

Paul,

This is not simply a semantic disagreement—I’ve criticised what you wrote in essence.

At the micro level, repayment of debt might lower someone’s spending, or it might not. It’s hard to tell because it depends on a bunch of other, counterfactual stuff. At the macro level, different levels and distributions of debt might result in different time paths of aggregate expenditure, or they might not matter. Again, it depends. That’s not very exciting. But it does have the distinct advantage of, I believe, being true.

paul meli said...

"Except you have tried to turn this rule of thumb into a law."

vimothy,

Rule of thumb? It's a mathematical reality, plain and simple. I don't know if that qualifies as a law or not.

I suppose the question I need to ask now is, under what scenario is that mathematical reality not true?

"The essence of what you are saying is that when people borrow, their purchasing power increases, so that as they repay their loan, their purchasing power should decrease…The problem is that this is not necessarily true for the individual and not necessarily true on aggregate either,"

This is the crux of our disagreement. Within a closed system it has to be true in the aggregate, although I haven't thought much about what is true for the individual. I am not a believer in micro. System dynamics is a macro domain.

The system of nominal dollars and bonds held by the public is closed, and the level cannot be altered within the system without some external intervention.

If you are making the argument that an increase in the value of real assets on a balnce sheet can shift the level of nominal (state issued) assets up or down in the aggregate then I have to say, no, that can't happen, it's impossible.

I am open to an argument why it isn't impossible though, just out of curiosity.

Let me make another argument along a similar line.

A business cannot succeed on sales to it's own employees alone. It follows (in my mind) that all businesses combined cannot succeed on sales to all of their employees combined.

Agree or disagree?

paul meli said...

Matt, I refer to closed system so often I sometimes feel like I come across as a parrot.

The thing that is so frustrating to me is that the implicatins of closed systems is so simple yet far-reaching and undermines so many assumptions in neoclassical economic theory, yet it is very difficult to convey because of it's utter simplicity.

How do we explain something so simple?

Obviously I haven't figured it out, been failing for over a year now.

Anonymous said...

Paul,

A very productive comment—I approve of it immensely.

If you are making the argument that an increase in the value of real assets on a balnce sheet can shift the level of nominal (state issued) assets up or down in the aggregate then I have to say, no, that can't happen, it's impossible.

I’m not sure that I quite understand this statement, so it’s definitely not the argument that I’m making. Let me try to translate it, slightly: it’s impossible for the (nominal) level of government liabilities held by the public to change without the government somewhere doing the changing. (For instance, issuing new money; retiring debt, etc.) That I agree with (although the public is also involved, which means that the terms on which those liabilities are held is not determined by the government alone).

The argument I’m making is that the reason the debt was incurred matters when trying to decide whether it lowers the individuals spending. We need to think about the counterfactuals. I might borrow some money and use it to build some prime office space in the business quarter downtown. Then, I earn a return from my investment as I rent it out to exciting start ups. Assuming that I’ve done this right, my return should exceed my costs. Now I can afford to spend more than I would have been able to in the absence of a loan, because in the absence of a loan I wouldn’t have been able to build the factory, and so my income would have been lower.

Borrowing doesn’t have to result in (economically) wasteful spending on consumption. It might result in you acquiring assets that raise your income. An obvious example of this is a student loan. I might borrow a bunch of money to go and study economics and econometrics at the LSE. Then, in the future, I can get a cushy job in asset management with some bulge bracket bank. Even though I have to repay my loan, my income is higher than it would have been otherwise, because now I don’t have to work in a call centre for the rest of my life. Or maybe I chose media studies and I’ll be returning to the call centre in short order. Like I said, it depends.

A business cannot succeed on sales to it's own employees alone. It follows (in my mind) that all businesses combined cannot succeed on sales to all of their employees combined.

Agree or disagree?


It depends what you mean by “employees”. Let’s say that our hypothetical firm employs labour and capital. Labour earns a fixed wage and capital earns the profits. Then the business can succeed on sales to the owners of capital and labour. This is basically a scaled down version of what happens at the macro level. Take the whole world. For the whole world, aggregate income equals aggregate expenditure. What people earn is exactly enough to purchase what people sell.

Ignacio said...

"Income", "expenditure", "aggregate demand" etc. are all a subset of semantic descriptions of a phenomenon.

What Keen is doing here, while maybe adding some confusion over academic jargon, is representing the dynamics of the phenomenon. That is, how the stocks and flows of these 'money things' behave and evolve over time. Getting obsessed over semantics and accuracy is being short-sighted and something usually academics and experts of a concrete field do, 'missing the point'. To everybody else the fundamental idea is what matters, while the presentation of the idea, conceptually (semantically) could be better; academics should contact the author about these 'rough' spots and compel him to fix them, but do you have an actual critique of the model? No, you don't, because it's accurately explaining the dynamic of how a real world phenomenon happens; this is what science does.

If the basic assumptions OF THE DYNAMIC hold, it does not matter if you describe them one way or an other (it's like if we should call an atom an atom or a banana, it's social convention), except for clarity, because the dynamic still is true.

The other 'accounting based' old models DON'T do that (including MMT), they just take snapshots at different times and tell you 'this is how it will look like at X, X+1, X+2, etc.). The model describes the dynamic of how you go from X to X+2.

Yeah it's not rocket science, but is good enough because it describes the evolution of that dynamic with the right mathematical methods to do that, through infinitesimally breaking these basic arithmetic accounting operations over time.

Now is a matter of adding complexity, one of the most difficult challenges will be how different financial assets mix and take over leverage and acquire/lose the quality of 'money things'.

P.S.1: Tom, there will be always qualitative problems. Your fellow 'Austrian' will tell you these models will never explain how these flows affect the real economy and the capital structure, that is: they are incapable of describing phenomenons like "malinvesting". And they are quite right.

Just because a model shows how a dynamic works, does it mean that when the dynamic changes from one state to other (is always changing) it's better or worse? We don't know if the economy is a better shape or not just by looking that. This is where things get really complex, but baby-steeps, because you can't fix 'micro' without a solid 'macro' safe environment and inducing a constant state of depression is no cure to anything as many 'austrians' imply.

P.S.2: Thanks for your posts JKH, very insightful.

Anonymous said...

Ignacio,

Other accounting based models are in fact dynamic. Have a look at Godley & Lavoie's textbook or some of the papers online at the Levy Institute’s site.

In general, in macro all of the models that are used are dynamic, so that there is nothing especially novel about the fact that Keen has written a dynamic model. It's either a good or bad model as the case may be.

Tom Hickey said...

Ignacio, Your fellow 'Austrian' will tell you these models will never explain how these flows affect the real economy and the capital structure, that is: they are incapable of describing phenomenons like "malinvesting". And they are quite right.

Actually, I was thinking in terms of Minsky, who is central to both MMT and Keen, as well as Michael Hudson. Others have noticed similarity between PKE and Austrian economics in this regard.

Oliver Davey said...

Schumpeter being the common Austrian denominator.

Ignacio said...

Vimothy,

Yes, a lot of neoclassical/mainstream macro models are dynamic, but are build based around certain assumptions that can be proved empirically wrong. Wasn't talking about these.

As for other 'accounting models' like 'functional finance', the actual modelling hasn't been done, what you have is descriptions of identities and snapshots. Keen model is actually an expansion on Godley (and Minsky), and takes these identities and formalize them to see how they behave through time as the different variables change.

As I said, is not rocket science, but if you have to compete with mainstream/neoclassical it has to be done, and it hasn't been done until now (and Keen is not the only one, there are various mathematical models popping based on functional finance).

Maybe we don't have the same definition of what constitutes a dynamic model or what 'dynamic' means.

Tom Hickey said...

yes, Minsky was a student of Schumpeter and Wray a student of Minsky. And Schumpeter was careful to dissociate himself from any school. Interestingly also, Minsky is not considered a Post Keynesian by some Post Keynesians and he expressly did not want to be. See Was Hyman Minsky a Post Keynesian?

Anonymous said...

Ignacio,

All economic schools use the same accounting identities drawn from the national accounts, etc. They’re not unique to Minksy or MMT. You’re right that they only give you snapshots of data at any one time, though.

By “accounting based models”, I meant post Keynesian models in the style of Godley. These models are certainly dynamic for any reasonable definition of the term. (They describe the evolution of the economy over time.) If you are unfamiliar with this approach, it’s pretty easy to locate a copy of Godley & Lavoie; you can also read many of Godley’s papers online at the Levy Institute.

Anonymous said...

If you have, or have access to, a copy of Eviews, model economies from Godley's textbook can be downloaded here:

http://gennaro.zezza.it/software/eviews/gl2006.php

Trixie said...

Ron Paul 2016!

Oliver Davey said...

2. Use standard accounting terminology and practice. We all need to be speaking the same language.

I know what you mean, but I don't quite agree with this.

The way I see it, bean counting is all about categorizing. And categorizing is framing. Globally, all stocks of financial assets and liabilities sum to 0. So the real meat lies in defining which sectioning of stocks and subsequently which measurements of flows between these stocks it makes sense to look at. Accounting standards are frameworks of sets. They are formed along well thought through, internally consistent, but nonetheless subjective lines. And it is these lines and a hierarchy among them (next to other things) that e.g. MMT, or Keen, and probably many other economists, are trying to realign or at least question with their respective work. For example, an MMT analysis will typically start with a distinction of 2 sector vertical vs. horizontal / government vs. rest of the world view. This doesn't per se question standard accounting terminology, but it certainly begs for new terms that reflect that division. The standard definition of 'saving' does not reflect that division. And 'net saving of financial assets by the non-government sector including business and foreigners' is truly unsexy. That's how the whole S=I discussion started. Or Frances Woolley from WCI may want a new term that measures finacial flows between men and women, on a global scale, even if there are no such contracts yet. SUch a term doesn't exist and it would be in conflict with the hierarchy of standard accounting that sets national boundaries. Why should a bunch of accountants be allowed to frame the world for us? But I do agree with Ramanan that Keen makes a mess of things. One never knows whose debt he's talking about. To use Neil Wilson's words: 'debt' the humpty dumpty term....

Unknown said...

aggregate demand = income + change in debt

I think the key is that "change in debt" can be either positive or negative, with a negative being net saving.

So income is what you just got and 'change in debt' is either how much is saved (net) or how much debt is created (net).

Expenditure is then the total of the above, which is equivalent to income.

And round again.

?

paul meli said...

"It depends what you mean by “employees”."

I mean employees as in it's workers. The people that work directly for the companies.

The bulk of dollars spent on consumption of consumer goods is spent by the wage-earner. Employees. Over 90% of wage earners do not accumulate a significant amount of savings. 65% accumulate none.

Corporations do not skim profits off of capital, unless you are referring to human capital. Profits accrue from wages entirely. Even if that weren't true, the end result is the same…the flow of fund moves in one direction only in the net.

There is no dynamic that will naturally move funds back to the cohort that will spend them.

Taxation is the only solution and if applied at an appropriate rate would suffice, but obviously that hasn't been the case over the past 30 or so years.

The accumulation of profits by a relative few removes a significant amount of dollars from the pool of funds available for spending in a continuous cycle from one year to the next.

Profits shift dollars from the many to the few, leaving less money for the many to buy the fews products.

Below a certain level of income (say the bottom 95%) most spending moves around that cohort like a school of fish and remains among them, less profits extracted by the few.

With these thoughts in mind, where do the funds come from to capitalize the consumer?

Anonymous said...

In that case, no, obviously not--unless profits and returns to other factors are zero.

By definition, what's required to purchase the output of the firm is its total revenue. Total revenue is the name given to the cost of the firm's output. If "employees" (i.e., labourers) do not earn that total revenue, then it is clear that they cannot possibly have purchased all of the output.

paul meli said...

vimothy,

Wouldn't it follow that government spending, targeted at the middle class and the poor is a necessary component of a thriving economy?

Anonymous said...

Paul,

I might agree with that statement taken alone, but I don't think you can extract it as a logical consequence of your argument.

paul meli said...

vimothy,

There's a few steps omitted. It goes like this…

• assume the system is closed (balanced budget and trade)

• Net spending flows accrue to the top of the income spectrum (as the tax code is currently arranged).

• Under the balanced budget scenario funds continue to be re-distributed from the top to the bottom creating government induced flow, adding to spending.

• This government-induced flow is not enough to counteract the net upward flow towards the top (obviously).

• The pool of funds available for spending (a stock) declines as a result, except for one reality still on the table…

• Private debt fills the gap…temporarilly.

For how long? It's hard to say because we have been running deficits consistently over the past 30 years, the same timeframe over which private debt became a major economic stimulus.

Even with expanding deficits the system has locked up. There's not enough spending to keep everyone employed.

• Eventually a ceiling is reached where the ability of most borrowers to expand debt is limited by their ability to service the debt. Their income has not risen at the same rate as their debt burden.

• What happens now?

If we don't expand deficit spending (or have a debt jubilee) the excess debt will have to be slowly (a decade? two decades?) wound down through distribution at the current rate. Much misery.

The funds can only come from re-distribution, until credit gets some headroom, but further increasing debt to move the economy is just going from the frying pan back into the fire.

Mainstream economics theory doesn't seem to get this. Their theory is that expectations, etc. will create the necessary additional flows. As value is added somehow this will generate funds at the bottom.

We should expect the opposite, we've established that those kinds of flows net to the top.

Flows require funds. More spending flows from a large pool than a smaller one, but the pool keeps getting smaller.

From where will the funds come?

And then in the real world we also have a trade deficit…

Tom Hickey said...

paul, Warren has consistently approached this from the perspective of income being spent ("consumed") or not spent ("saved") in a period. In a closed systme, to the degree that income is saved, goods able to be supplied using available resources ("production") are not sold and build up as "unplanned inventory" during the period. This is a signal to lower prices or cut production. But if prices are cut so will wages be cut, and the situation will not be remedied if savings rate is constant or increases. If production is cut back then employment increases, wages fall in aggregate and reduced aggregate income results in less effective demand and less sales, which sends a signal not to produce.

Moreover, private debt remains nominally the same, and debt service becomes an issue with lower income to service it.

This is the way that MMT economists approach the system in terms of effective demand and demand leakage. This is the basic "Keynesian" approach.

As far as thinking about NFA held by non-govt in aggregate, take all assets held by non-govt in the unit of account (currency) and look to the relationship between asset and liability. If the asset and liability are both in non-govt, then the net is zero in aggregate. If the asset is held by non-govt, and the liability is that of the govt, then this is included in non-govt NFA.

paul meli said...

Tom,

Exactly. I think the ideas expressed in my comment above are completely in harmony with this, just from a different perspective.

Think of me as the proverbial visitor from Mars, seeing what's going on with no pre-conceived worldview (at least as far as economics is concerned).

Anonymous said...

Paul,

Some questions:

Net spending flows accrue to the top of the income spectrum (as the tax code is currently arranged).

What do you mean by “net spending flows”? What do you mean when you say that they “accrue to the top of the income spectrum”? (By definition, people at the top of the income distribution earn a greater share of aggregate income than people at the bottom.)

This government-induced flow is not enough to counteract the net upward flow towards the top (obviously).

What do you mean by “net upwards flow towards the top”? Are you saying that govt redistribution of income is not complete, i.e. there are some progressive policies but still income inequality?

The pool of funds available for spending (a stock) declines as a result, except for one reality still on the table…

What do you mean by “pool of funds”? Savings? (Private savings? National savings?) Declines as a result of… what? (Income inequality?)

Private debt fills the gap…temporarilly.

Fills what gap?

Are you saying something like: income inequality has somehow caused people to constantly dissave or borrow more to increase their consumption, which is unsustainable?

Eventually a ceiling is reached where the ability of most borrowers to expand debt is limited by their ability to service the debt. Their income has not risen at the same rate as their debt burden.

If borrowers borrow as much as they are able to, and then stop, why’s that such a big deal?
If we don't expand deficit spending (or have a debt jubilee) the excess debt will have to be slowly (a decade? two decades?) wound down through distribution at the current rate. Much misery.

What do you mean by excess debt? What does deficit spending have to do with it?

paul meli said...

What do you mean by “net spending flows”?

Ok…divide the closed system into two groups, companies that earn profits and everyone else, the cohort they sell their products to. Wage-earners and their dependents.

The net flow of funds is to profits (savings) that don't get spent or re-invested in future production; away from demand.

paul meli said...

"Fills what gap?



Are you saying something like: income inequality has somehow caused people to constantly dissave or borrow more to increase their consumption, which is unsustainable? "


Yes.

"

If borrowers borrow as much as they are able to, and then stop, why’s that such a big deal?
"

If the expansion of credit stops suddenly that portion of spending does also. What happens to overall spending if debt-related spending is a significant part of it and stops?

What will take it's place?

Further, income diverted to debt service is removed from spending and by extension demand.

Anonymous said...

Paul,


Ok…divide the closed system into two groups, companies that earn profits and everyone else, the cohort they sell their products to. Wage-earners and their dependents.

The net flow of funds is to profits (savings) that don't get spent or re-invested in future production; away from demand.


In what sense is the flow of funds “net”? Net of what?

Think of an economy with one firm. That firm employs labour and rent capital. It sells its output and raises revenue that is split three ways: wages for labour, rent for capital and profits for the owners. Any one of those can be spent or saved. Profits do not have to be saved and wages do not have to be spent.

Say that of the total income of the economy, a constant fraction is saved. Then, by definition, nothing is being lost from demand. Aggregate demand is expenditure on all components of GDP, not just consumption. Given that the economy is saving an amount of its income by assumption, we must have some investment or some lending to foreigners. If saving causes income to fall because it removes funds from circulation, then saving is impossible on aggregate. Either people spend all their income or the economy shrinks away to nothing. But that is empirically false. Aggregate saving is completely normal in economies all over the world.

Tom is wrong about the basic Keynesian model. The issue here is not that saving is a demand leakage. The issue is that in equilibrium saving must equal investment. (Investment is a component of demand.) If actual saving exceeds planned investment, actual consumption expenditure will fall below the level expected by firms. This will cause inventories to rise as goods go unsold. Because for Keynes interest rates are determined by liquidity demand, interest rates will not adjust in response to the increased supply of savings and stimulate more business investment. Therefore, we get a recession and income falls until it returns saving and investment to a new equilibrium level.

“Are you saying something like: income inequality has somehow caused people to constantly dissave or borrow more to increase their consumption, which is unsustainable? "

Yes.


In that case, it would be much simpler, in my view, to just say it. It’s good to try to come up with theories that can explain the general case, but I don’t see how you can posit this outcome as a logical consequence of the fact that firms earn profits.

Oliver Davey said...

Vimothy


I think it's best to start with the lowest common denominator in this discussion between MMT, Keen, Circuitists, Minskyans etc. vs. what is often summarised by the former as 'the neoclassical approach'.

I personally see the origin of the discussion in 2 factors: a) the acceptance of the credit theory of money and with it, the endogeneity of the money supply and b) the demand led approach for explaining cycles. A further, specifically MMTfocus is the superposition of the state theory of money on to the above mentioned credit theory (I'll get to that later).

In a nutshell, the credit theory of money says: all money things originate as asset / liability pairs when a payment is made on credit. And, subsequently, both a monetary asset and a liability of equal value disappear when the debt is settled / principle is paid. System wide (regardless of the boundary), all new financial assets come with an equal liability.

The demand led business cycle says (to me) that it is current demand that informs business on how to invest and produce in future and not current investment that determines future demand. The equilibrium system you describe is static in that it assumes a fixed amount of funds which clearly violates the endogeneity story from above. I say, not even in theory can you have it both ways, and in practice endogenous money is a fact.

The MMT / state theory of money then posits that money is not only an accounting artefact, but is also (necessarily) introduced and controlled by an authority which, in most modern nations, is equal to the state. Mosler says it this way: money is a simple government monopoly, in which it controls the price (through interest rates for banks and through prices effectively paid by government) and floats the amount of assets according to demand for them at the given price. Think of it as a pyramid of asset / liabilities in which base money forms the tip, followed by bank credit, followed by non-bank credit and so on. Due to the construction of the payments system and the fiscal / monetary nexus, governments and, by extention, banks are not liquidity/credit constrained in the same way as members of the class 'non-government' are. Wrt to the specific currency being examined, this is true also of foreign governments...

Oliver Davey said...

...To my knowledge, this constuction of monetary hierarchy is informed by a stylised fact and subsequently infered causal relation, that says: for an expansion of real GDP to be stable, it must be accompanied by a net financial surplus of the non-government sector. Such a surplus can either be achieved through a positive trade balance, in absence of such, a government deficit. Since both are endogenously determined by the economy, the result is called 'the net saving desire of the non-government sector'. Further contraints are that not all countries can simultaneously have positive trade balance and that the external position is difficult to control, at least in the short run. Thus, the government budget should be, indeed due to its endogenous nature, always ends up being anyway, more or less accomodative of net saving desire.

So, the fact that there has been a private credit bubble leads MMT to the following conclusions: effective demand is roughly a function of the financial assets produced by the credit expansion times the propensity of the recipients to spend it on goods and services. Since CPI has been well under control since the monetarists took over, the case for reducing overall demand cannot be made.The capability of the economy to service the debt (the liability side of private credit), on the other hand, is a function of its distribution, as well as, of course, the quality of the investments made and / or the growth rate of the asset bubble :-). In any case, the quality of investments cannot be changed an after the fact. So, the generic MMT response is to redistribute finanical assets from top to bottom while doing the opposite with the liabilites, with the overall effect of more 'net financial assets' available for the non government sector. This is opposed to, say the generic Austrian response that focuses only on reducing the amount of gross liabilities over all sectors with no regard for the asset side of the equation (liquidationism). The standard / monetarist informed response, as reflected in say Bernanke, ist to try reignite the credit cycle without a redistribution of the liability side on to the government books (loose money & fiscal consolidation). Europeans are wed to the idea of an export led strategy within a tight fiscal & tight monetary policy environment.

Anonymous said...

Hi Oliver,

My feeling is that what you call the credit theory of money is not something that anyone would disagree with necessarily. The same is also true for endogenous money. The dominant approach in macroeconomics thinks of monetary policy in terms of policy rules for interest rates—the determination of the money supply is left to the public. (Mike Woodford even wrote a famous paper on controlling inflation in an economy without money).

The short story that I told about excess saving was just meant to explain the basic Keynesian model, which Tom referred to in an earlier comment. I was not suggesting that this is the only or even the best way to understand business cycles. In this model, excess saving causes a demand-led recession. (Excess saving means that consumption expenditure is lower than expected. Profits fall. Businesses cut back.) If we used the theory of loanable funds, then we might expect that the interest rate would fall as saving rose above its equilibrium level causing investment supply to increase and savings supply to decline, clearing the market and preventing any kind of recession. However, Keynes used a different theory of interest rate determination, known as the theory of liquidity preference. According to this theory, the interest rate is not a function of the supply of savings.

Often, I see comments that suggest that saving per se lowers demand. This is an easy intuitive leap to make, but it is not quite correct. For the whole economy, by definition, saving is equal to investment. Investment demand contributes to aggregate demand. Therefore, as long as investment is positive, savings will be positive as well. Moreover, capital depreciates, so that unless investment equals the replacement rate at least, the capital stock will be vanishing over time. (Imagine that aggregate savings were zero. Consumption would be very high in the short run because no output is being invested. In the long run, however, consumption would fall to very low levels because eventually the capital stock would fall away to nothing.)

Consequently, at any given point in time for any given economy, we are likely to see a positive saving rate for the whole economy (assuming the economy is not totally dysfunctional). This doesn’t mean that demand is “leaking” out of the economy. If it did then all these economies with positive saving rates would be constantly in recession, and we’d all be screwed because some level of saving is necessary to maintain the capital stock. (If the population is growing, this is a moving target.)

The issue for Keynes was whether there was anything that coordinates saving and investment. Classical economists thought that this was achieved by the interest rate, which acts to bring the supply and demand for funds into equilibrium. Keynes disagreed. For Keynes, the economy could start in equilibrium (with S = I), but then a shock could shift the saving supply function such that income shrinks to return the system to equilibrium rather than this being affected by the interest rate on loans with no negative impact on income, as n the classical story.

paul meli said...

"So, the generic MMT response is to redistribute finanical assets from top to bottom"

It's the generic response period, not just for MMT.

We have always (since progressive taxation was implemented) taxed and spent as a means to induce flow.

Tom Hickey said...

Good summary, vimothy. That is the view of Keynes as I understand it. He was in "orthodox" economist in so far as buying into long-run equilibrium, i.e., over time economies tend to equilibrium states. It seems to me that a difference between Keynes and neoclassicals of the time was that neoclassicals assumed return to equilibrium after a shock and Keynes assumed multiple equilibriums.

Here is where the successors to Keynes parted company. Old Keynesians (Samuelson, Krugman) and New Keynesians (Mankiw) continue to presuppose long-run equilibrium. The discussion then become one of single or multiple equilibriums (See some recent posts of Nick Rowe). These are "orthodox" views.

Post Keynesians reject the assumption of equilibrium in a dynamic economy as too static a model to represent reality. They contend that DSGE is untenable and needs to be replaced with a new dynamic model. This is a heterodox view. (See the recent Krugman-Keen "debate").

Anonymous said...

Oliver,

I just had a quick check of posts at Billy Blog, and you might find this pair of posts on aggregate demand helpful:

http://bilbo.economicoutlook.net/blog/?p=21048
http://bilbo.economicoutlook.net/blog/?p=21099

I expect that Tom has posted them elsewhere on this site already, but they tell more or less the same basic story in more depth, with some helpful diagrams. (In particular, note the section on equilibrium income in part 5 and the section on unplanned investment in part 6).

Tom Hickey said...

BTW, vimothy, unplanned inventory is counted as investment, and so S= I does not not necessarily imply an identity of saving and fixed capital. That's the fly in the ointment.

The other fly in the ointment, as Marx observed, is deterioration and obsolescence of fixed capital (as you allude to). Without replacement, the rate of profit falls.

Add to this the falsity of the loanable funds doctrine in an endogenous monetary system with a lender of last resort, and the neoclassical assumptions that underlie the neoliberal position don't hold as represented. Which is a reason that "expansionary fiscal austerity" doomed to fail.

Anonymous said...

Tom,

That’s right. I referred to that in my original explanation:

“If actual saving exceeds planned investment, actual consumption expenditure will fall below the level expected by firms. This will cause inventories to rise as goods go unsold.”

This reflects the fact that S = I is both an ex post identity that is always satisfied and an equilibrium condition.

It also provides us with a useful economic indicator: inventory changes lead output over the cycle. (Bill Mitchell discusses this in the second link I just posted.)

BTW, I’m not sure how far you will get in economic analysis if you don’t have any concept of equilibrium. (Mitchell obviously does.) Equilibrium is just short hand for, “let’s imagine that everything is stable so that we can isolate the effects of whatever change we’re interested in looking at”. It’s a heuristic. It enables discussion. For instance, say that I want to know what will happen if there’s a sudden and unanticipated rise in the saving rate. In reality, everything is in flux, but in theory, I would like to look at this one particular thing. So it makes sense to start the model from a situation in which actual and expected saving is equal, then change it.

paul meli said...

vimothy,

"In what sense is the flow of funds “net”? Net of what?


…
In that case, it would be much simpler, in my view, to just say it…

…I don’t see how you can posit this outcome as a logical consequence of the fact that firms earn profits."


The quantity of funds within a closed economy is fixed; (I-S)=0…

The economy consists of workers earning wages plus their dependents, totalling some 240 million people plus about 40 million retirees, leaving some 35 million in a position to earn non-financial profits. The exact number isn't important.

However funds are distributed initially it should be obvious that the net flow of funds through successful investment will be from the 280 million consumers to the 35 million investors (the economic form of investment).

The actual distribution of funds is skewed heavily towards a much smaller cohort consisting of maybe less than 3 million people.

Yes, I know there are exceptions to the case as I have laid it out but the number of exceptions will be neglegible.

No arrangements in place currently will reverse the dynamic of a shrinking pie for the cohort that must support aggregate demand in order for business to succeed.

Oliver Davey said...

Vimothy

Thanks, very educative.

My feeling is that what you call the credit theory of money is not something that anyone would disagree with necessarily. The same is also true for endogenous money. The dominant approach in macroeconomics thinks of monetary policy in terms of policy rules for interest rates—the determination of the money supply is left to the public.

I think PKers/circuitists lament that both theories may be ostensibly widely accepted but are never reflected in the models used. An inconsistency argument.

Often, I see comments that suggest that saving per se lowers demand. This is an easy intuitive leap to make, but it is not quite correct. For the whole economy, by definition, saving is equal to investment. Investment demand contributes to aggregate demand. Therefore, as long as investment is positive, savings will be positive as well. Moreover, capital depreciates, so that unless investment equals the replacement rate at least, the capital stock will be vanishing over time. (Imagine that aggregate savings were zero. Consumption would be very high in the short run because no output is being invested. In the long run, however, consumption would fall to very low levels because eventually the capital stock would fall away to nothing.)


There was a long debate here, among other places, about common misconceptions by MMT informed lay economists regarding the meaning of the word saving / savings. It all started with the tautologic formula brought up by JKH that says S=I+(S-I) (look it up at MMR if you feel inclined to blow your mind). The essence for me was that there is such a thing as a desire to save financial assets net of investment, or to save 'risk free', if you like. With the effect that S cannot be I if that desire is met.There is a residue that must come from somewhere. And that somewhere must be either another currency zone (funded via exports) or your friendly neighbourhood government (funded via deficits). I think it is fully compatible with the depreciation story though, only framed slightly differently. In fact, I believe Joseph Laliberté had a post to that extent at 'Fictional Reserve Barking'.

The issue for Keynes was whether there was anything that coordinates saving and investment. Classical economists thought that this was achieved by the interest rate, which acts to bring the supply and demand for funds into equilibrium. Keynes disagreed. For Keynes, the economy could start in equilibrium (with S = I), but then a shock could shift the saving supply function such that income shrinks to return the system to equilibrium rather than this being affected by the interest rate on loans with no negative impact on income, as n the classical story.

Post Keynesians have a different interpretation of Keynes' own words, I believe. The deus ex machina problem with the shocks is covered by Minsky, while the problem of the liquidity trap is 'solved' with functional finance approach of extending monetary policy with fiscal means to accomodate any given liquidity preference. In fact, interest rates can be taken out of the equation entirely, at least as thought experiment.

Tom Hickey said...

PKE does not deny equilibrium at any point in time (snapshot) only a knowable equilibrium state or states in the future. A system is always in equilibrium statically in that equation (n the case of monetary economies, identities) have to be satisfied. What PKE holds is that there is no determinable long-run general equilibrium state or multiple states as points of stability that can be determined a priori. That is to say, economies are dynamically complex and always in flux, with variable degrees of stability. Stability of any state does not imply either continuance of that state or return to it after disruption.

One why of looking is closed-ended, and the other is open-ended. This is a significant methodological difference. The difference is between a "natural" system that rights itself as long as it is not interfered with, and a system always in flux that can be controlled through knowledgeable and skillful application of intelligence.

This is very important at critical turning points because a chief human propensity is to assume that the future will strong resemble the past. That can be radically wrong at turning points when an entirely different dynamic takes over.

There are two "danger" areas involved, 1) matters that are knowable but not being taken into consideration and 2) matters that are unknowable in principle, at least far enough in advance to take into account in a timely way.

One view is to get out of the way and not interfere with market dynamics and everything will be OK due to operation of natural laws, and the other is to use knowledge and skill to rescue a ship that is taking on water (saving excessively).

Heterodox economists are waving red flags about the first, e.g., based on PKE and Minsky, and environmental scientists about the second, e.g., based on energy science, ecology, and climate science. I am actually most concerned about the latter, since it is species-existential.

paul meli said...

vimothy said…

"For the whole economy, by definition, saving is equal to investment."

Only when (I-S=0), but this says nothing about the level of either component, only the net difference.

"Investment demand contributes to aggregate demand"

I described why this isn't true above. Investment removes more demand than it creates.

"Therefore, as long as investment is positive, savings will be positive as well."

I'm not so sure about this. Savings can flow towards investment. This reduces savings but increases investment. Personally I haven't figured out the real-world ramifications of (I-S) or whether it is anything more than a placeholder in the sectoral balances. All of the other parameters of the S/B are knowns.

Tom Hickey said...

@ Oliver

S = I applies only to a closed economy, which is what I took the thread to be about at this point of developing a simple model.

In a closed economy

Income = Expenditure

Uses of Income = Uses of Expenditure

C + S = C + I.

Simplifying,

S = I

Different an open economy with exogenous input, where S does not necessarily equal I.

Bill Mitchell, Sunday, June 19, 2011 at 8:29

Dear Steve Roth (at 2011/06/19 at 1:17)

You asked:

If S=I — another inviolable, given identity — doesn’t that balance always equal zero?

Only in a simple macroeconomic model with no government sector and no trade (external) sector.

Then S must always equal I although desired saving might not equal planned investment. In that case, unplanned investment (via undesired inventory run-down or accumulation) will occur and firms will alter production and employment levels.

In a closed economy with government sector the following must hold:
(S – I) + (T – G) = 0

And then adding the external sector gives you:
(S – I) + (T – G) – (X – M) = 0

Hope that helps.

best wishes

bill

Anonymous said...

"S" and "I" are letters that we use to stand in for variables within a macroeconomic model. "S" doesn't have to stand in for "private saving". It also often stands in for national, or economy-wide saving.

We must have S = I for any closed economy. That is, total saving for the entire economy must equal total investment. If that economy consists of one single sector, then both the economy and the sector will have S = I at all times. If that economy consists of several sectors, then saving needn't equal investment for individual sectors, but it must still hold at all times for the economy as a whole.

You can easily see this just by extending the basic principles in Mitchell's reply above.

C + I = C + S

[Cp + Cg] + [Ip + Ig] = C + S

The government or the private sector can either consume or invest. Hence national saving is equal to private and public investment.

In practice, the only closed economy we get to see is the entire world. For the global economy, saving must equal investment.

paul meli said...

"In practice, the only closed economy we get to see is the entire world."

I think you misunderstand the meaning of the term "closed system".

We can define a closed-system boundary anywhere arbitrarily. A skilled analyst will choose a boundary that makes the analysis the least complex with the fewest unknowns.

For analysis one can define the overall system or any subsystem(s) within no matter how big or small as desired. We can define an individual agent as a closed system. A glass of water. Any container.

We can look at the entire world economy as a whole (currently difficult and indeterminate) or we can look at parts of it and then look at the interactions between the parts.

(I-S)=0 is the part we have been discussing. If we don't fully understand that part of what use will moving on to a more complex system serve?

Anonymous said...

I don't misunderstand the term, "closed economy", though.

Anonymous said...

Paul

The quantity of funds within a closed economy is fixed; (I-S)=0…

Not sure what you mean by this. The quantity of funds in any economy—closed or otherwise—is not fixed. Available funds (presumably you’re thinking of savings) can be increased because the economy constantly produces output that generates new income. This new income can be saved, adding to the existing stock for use in the future.

However funds are distributed initially it should be obvious that the net flow of funds through successful investment will be from the 280 million consumers to the 35 million investors (the economic form of investment).

Why is it obvious? I can see that rich people find it easy to capture institutions that serve to keep them rich, but I don’t see what this has to do with saving and investment or profits.

No arrangements in place currently will reverse the dynamic of a shrinking pie for the cohort that must support aggregate demand in order for business to succeed.

Let’s say you’re right. Why should the government step in to make business succeed? Won’t that cause the pie to continue to shrink for the majority, while ensuring business owners still have plenty to eat?

Only when (I-S=0), but this says nothing about the level of either component, only the net difference.

Right, it desn’t tell you anything about the level of saving, just that it’s going to equal the level of investment.

I described why this isn't true above. Investment removes more demand than it creates.

What do you mean, “investment removes more demand than it creates”? How does investment remove demand? How does investment create demand?

I'm not so sure about this. Savings can flow towards investment. This reduces savings but increases investment

What do you mean? Savings are a source of finance for investment. If savings finance investment, how are savings reduced?

paul meli said...

"I don't misunderstand the term, "closed economy", though."

It seemed to me from your comment that you did, sorry if i misconstrued.

The only requirement for a closed economy is the boundary definition. When we say (I-S)=0 we have removed the government money-creation machine, the external sector and the Federal Reserve Banking banking system from the equation.

Anonymous said...

In macroeconomics, an open economy is one that has a foreign sector, whereas a closed economy does not.

See, e.g.: http://en.wikipedia.org/wiki/Open_economy

paul meli said...

"What do you mean, “investment removes more demand than it creates”? How does investment remove demand? …"

The act of investment doesn't. Once the investment is capitalized and has succeeded (earned a profit) there are fewer funds in the remaining subsector.

"…How does investment create demand?"

It doesn't. Not in the net anyway.

"What do you mean? Savings are a source of finance for investment. If savings finance investment, how are savings reduced?"

If savings (a stock) are a source of funds for investment (a flow), doesn't it follow that when the savings are invested the savings stock becomes smaller and flow is increased?

Investment is a form of spending. Until someone down the chain removes those funds from spending again.

"Savings are a source of finance for investment"

Savings CAN BE a source, but isn't necessarily. The source of funds for investment can also be credit, where the funds are created out of thin air.

Tom Hickey said...

Probably simplest to think of it as C = nominal value of consumption. C is on both sides since what is produced on the LHS is consumed on the RHS with funds payed for production of C as income. S is the residual of income not spent on C and I as what is produced that this not consumed, essentially capital goods and inventory, planned and unplanned. So "S" as "saving" may differ from what "saving" means in ordinary language usage, and the same with "I" for "investment." Just as "positive" and "negative" have different meanings wrt to electricity and ordinary usage.

Boils down to maintaining the identity of income and expenditure in a period, reflecting that the books balance. And there's no causality associated with accounting identities a priori.

paul meli said...

"In macroeconomics, an open economy is one that has a foreign sector, whereas a closed economy does not."

That's why I re-defined it for you…it appeared you were thinking in those terms.

I'm looking at this from a perspective of universal system behavior, not macroeconomics. A math-based approach.

I have never been secretive about this. MMT is based on a similar approach, which may be why neoclassicals have problems with the way some terms are used within MMT analysis.

Besides, macroeconomics as an economic discipline is not a science. It doesn't seem right that it should co-opt terminology that has always been in place in math and science and is well understood by the community.

That said, if one is going to do a mathematical and system analysis of an economic system (I'm trying) the terms used must be as defined in that context.

It shouldn't matter as long as you are aware of my meaning.

Anonymous said...

"Closed economy" has a specific meaning in economics. If you ask Bill Mitchell or Randy Wray what "closed economy" means, they will give you the same answer that I did, because they are economists. No one is co-opting the term. If engineers wish to use it to mean something else entirely then they are free to do so.

paul meli said...

vimothy,

Bill Mitchell uses the term both ways and he knows which is which.

The terminolgy itself is meaningless without context, What is important is an understanding of the underlying abstraction in a dsicussion.

"Closed system" has a specific meaning. That's how I am using it as applied to an economic system.

I don't think it would be fair to take a particular known concept off the table just because someone involved isn't familiar with it. Anyway, now you are.

In the future I will try to be more careful when I mix metaphors.

Anonymous said...

Paul,

The act of investment doesn't. Once the investment is capitalized and has succeeded (earned a profit) there are fewer funds in the remaining subsector.

I’m not sure that I understand this. When savings are used to finance investment, those savings don’t disappear. Imagine that I buy bonds from Apple, and that Apple uses the funds acquired to build a new factory. My savings are now in the form of bonds.

If savings (a stock) are a source of funds for investment (a flow), doesn't it follow that when the savings are invested the savings stock becomes smaller and flow is increased?

Okay, I think I understand where you’re coming from.

The stock of savings is already tied up in capital from previous periods. To finance new investment (that is, new capital), new savings are required. This would be problematic if the pool of savings was fixed at some prior level, but this is not the case. The economy constantly generates income that can be saved and used to finance new investment. And that’s precisely what we see in practice.

Savings CAN BE a source, but isn't necessarily. The source of funds for investment can also be credit, where the funds are created out of thin air.

It’s certainly true that investment can be financed by credit. However, even if the funds are created “out of thin air”, someone must be on the asset side of that arrangement, which means that they are saving.

Anonymous said...

It's the term "closed economy" that has the specific meaning in macro. (I've searched at Billyblog, but the only instances that I can find refer to standard closed economy macro). If you want to use "closed system" at the same time, I don't think that that would be particularly confusing.

Oliver Davey said...

Available funds (presumably you’re thinking of savings) can be increased because the economy constantly produces output that generates new income.

You're mixing stocks (funds) with flows (income). The stock of funds does not increase with new production (unless you assume factors are payed with credit).

Anonymous said...

Define "funds".

Savings (stock) can be increased by saving (flow) out of current income (flow). (That's where those savings came from in the first place.) Flows adjust stocks.

Tom Hickey said...

"Money is not a commodity." It is neither produced nor consumed, and it is not created by the circular flow of production and consumption of goods although it enters into it. Money also influences this circular flow so it is not neutral to it as merely a bookkeeping function.

In a monetary economic system, money must come from outside the circular flow of consumption and production and, in fact, it is created either by credit or from govt injection through purchases, transfers, interest payments etc. in excess of taxes.

So a monetary economy w/o govt is exclusively a credit economy in which some may consume in excess of income but not everyone, since one person's saving (asset) is another's borrowing (liability), i.e., there can be no net saving in aggregate. Since assets must equal liabilities, in aggregate financial assets net to zero.

IThere can only be aggregate net save nominally if the net amount comes from outside.

If a govt is introduced that is the monopoly supplier of the currency, aggregate nominal net saving of non-govt in the govt's currency can only come from the govt's net provision of the currency.

Then this system can then be considered as a closed system with two separate levels of money creation, one inside ("private"), i.e., credit that always nets to zero, and the other outside ("public"), i.e., the govt fiscal balance that can increase or decrease the net financial assets of the private sector.

Oliver Davey said...

"Money is not a commodity."

This is exactly what I mean. A couple of posts further up he accepts the credit theory of money and then he pulls a 180° and argues along commodity money lines.

There can only be aggregate net save nominally if the net amount comes from outside

I think the main contention is the definition of 'outside'. Sure, a systems engineer can define inside and outside wherever he or she likes. But that doesn't make that boundary meaningful.

I would say all money, whether bank or government created, is a form of credit that arises with a specific payment. The difference is on the liability side, in that private debt is an individual phenomenon whereas the national debt (whether incurred directly through government deficits, or indirectly through a trade deficit) is a, well national, i.e. socialized thing. I'd say it's a difference in degree and thus probably meaning, but not necessarily in kind. MMT tries too hard to make it a difference in kind through its use of language, I find.

geerussell said...

I'd say it's a difference in degree and thus probably meaning, but not necessarily in kind. MMT tries too hard to make it a difference in kind through its use of language, I find.

Solvency constrained vs not solvency constrained is a genuine difference in kind.

paul meli said...

"It's the term "closed economy" that has the specific meaning in macro. (I've searched at Billyblog, but the only instances that I can find refer to standard closed economy macro). If you want to use "closed system" at the same time, I don't think that that would be particularly confusing."

I understand your criticism and I apologize for not being more clear.

I think if you read back over my comments you will see most of the time I've used the term "closed system". Every now and then I use the term "closed economy" to make the connection that I mean the economic system.

When I use the term "closed system" wrt the economic system I don't know what else to call it, but if I say I'm talking about (I-S)= 0 that should clear it up.

FYI, if I want to include the external account I can define the boundary to include that and it will still be a closed system, but with an additional sub-sector.

Ahh,language.

Tom Hickey said...

Oliver, here we are building up a model from a simple closed system w/o govt, and then adding govt. Haven't added an external sector, which further complicates the system's sectoral relationships.

We began with a monetary economy but could have begun with a non-monetary barter economy and added money.

I would expect that a beginning course in macro would start this way.

Of course, beginning with a non-monetary barter system implies nothing about the actual facts of history. It just shows in terms of simple systems how stocks and flows work both actually and nominally.

This would then build to adding the financial components, both private and public, to show how money is added to a non-monetary system.

The gap between a non-monetary barter exchange system a monetary credit-based system can be illustrated with non-monetary credit-based system to develop the distinction between spot exchange and delayed exchange based on trust.

paul meli said...

"I think the main contention is the definition of 'outside'. Sure, a systems engineer can define inside and outside wherever he or she likes. But that doesn't make that boundary meaningful."

The boundary condition is quite meaningful if it's properly defined. Most complex systems would be indeterminate if we couldn't analyze the parts.

It's a tool of convenience because we know the limitations of closed systems. Once a boundary is defined many out-of-paradigm claims are eliminated from possibility immediately. We don't get dragged down rabbit-holes that don't go anywhere.

In the case of an monetary-based economic system it makes it clear that economic activity cannot create it's own spending and growth. It must be applied from a source external to the system.

The Earth is a closed system, it can't grow on it's own. It requires energy from the Sun. Otherwise it would be a frozen, barren planet.

If it could it would be like a perpetual-motion machine and we know that isn't possible.

paul meli said...

"I’m not sure that I understand this. When savings are used to finance investment, those savings don’t disappear. Imagine that I buy bonds from Apple, and that Apple uses the funds acquired to build a new factory. My savings are now in the form of bonds."

When you buy a bond from Apple you acquire a "placeholder" for your savings. To you it appears to be savings but it isn't by definition.

Apple takes your savings funds and spends it on plant, labor, resources used to generate goods and services.

Your savings is now spending again until someone else saves it.

Your Apple bond does not "remove" any funds from the economy.

When dollars are swapped for Treasuries, they are effectively "sequestered". Selling a Treasury to another agent just swaps his spending for yours.

When dollars are spent from Income to repay debt, those funds are "sequestered" permanently.

Tom Hickey said...

FYI, if I want to include the external account I can define the boundary

Yes, the external sector cannot create the currency, and if foreign banks operating in the US with Fed access are given this franchise, it is within the USD system. So it is possible to define the dollar zone as a closed system.

Anonymous said...

Paul,

So an open economy is a closed system that includes a domestic economy and a foreign sector. It's closed, not at the level of "the economy", but at the level of "the economy plus the foreign sector"?

Say something more about what you mean by a closed system. I'm never sure if I understand what you mean by it. In what sense is it closed? Does it have a precise definition?

paul meli said...

vimothy,

From Wikipedia:

This is the part relevant to the discussion, the complete definition is more broad…

"…For a simple system, with only one type of particle (atom or molecule), a closed system amounts to a constant number of particles.…"

Substitute "dollars", "euros", "pounds", etc. (for atom or molecule) as appropriate.

Hey, I said it was simple. :-)

But extremely powerful if you think of the implications.

Anonymous said...

Paul,

The thing is, is that you shouldn’t associate the stock of dollars, or the stock of government liabilities more generally, with the stock of available funds Available funds are independent of government debt, which is only one type of asset among many. The economy needs saving because it needs investment. That’s the vehicle for saving—capital. No problem with demand for or supply of saving arises as a necessary effect of the fact that only the government can issue government liabilities. The economy can generate funds and put them to use all at the same time.

Anonymous said...

Paul,

When you buy a bond from Apple you acquire a "placeholder" for your savings. To you it appears to be savings but it isn't by definition.

Your savings are your asset—the bond—, which is an obligation of Apple to you.

Apple takes your savings funds and spends it on plant, labor, resources used to generate goods and services.

So you save, Apples borrows your savings and uses them to buy new capital. That flow of new spending on capital is “investment”. Investment contributes to aggregate expenditure and therefore to aggregate income. Apple’s new plant becomes income for the suppliers of that plant. That’s why investment is part of aggregate demand.

Your savings is now spending again until someone else saves it.

I would put that the other way around, but basically, yeah.

Your Apple bond does not "remove" any funds from the economy.

Right. It didn’t remove any funds. It put them to use.

paul meli said...

"The thing is, is that you shouldn’t associate the stock of dollars, or the stock of government liabilities more generally, with the stock of available funds"

I don't. The only stock of funds I care about is the stock in consumers hands for spending. That has to be there for the system to function. If they come from private debt, the system will be unsustainable, as we see from observing natural systems.

I don't confuse the funds in consumers hands as a "stock" except in the sense that that is the subset of the money supply that is actually the "cause" of economic activity. When it is spent.

Investment doesn't put those funds in those hands in the aggregate. Investment only shifts existing funds around.

Progressive taxation puts those funds in the hands of consumers. Deficit spending accounts for leakages.

If we look at the economy as a "closed system" we see that no amount of bond-buying, investment, accounting tricks, etc. can increase the level of funds available, it can only shift them around.

The financial industry has fiqured out how to squeeze funds out of the external sources and acquire the asset side of the transaction while leaving the public with the liability part.

Thus we have what we have.

Anonymous said...

That's a problem with income and income inequality. It's not a problem that arises from saving and investment.

Tom Hickey said...

Right. It didn’t remove any funds. It put them to use.

True of primary investment and debt, which create savings vehicles, e.g., stocks and bonds, that get exchanged subsequently based on trading and portfolio preference. Most financial "investment" is just exchanging savings vehicles rather than resulting in new primary investment in non-financial capital, although it provides a market that enable new primary investment and more savings vehicles resulting from it.

While this is obvious to anyone who understands what's actually happening, many people don't understand the different meaning of "investment" wrt primary, secondary, and derivatives, and they think that all "investment" is the same.

Tom Hickey said...

Stock is the only thing that is actual. Flows are just changes in stock.The size of a stock at any point is dependent on the flows that affect it, in or out.

In a closed system all the stocks and flows are within the boundaries of the systems. Thus it is possible to describe the system in terms of those stocks and flows and how they interact with each others, on the basic assumption that nothing gets in or out.

According to the conservation law the universe is a closed system wrt energy, energy being neither created nor destroyed. The entire system is based on analysis of energy flows wrt various stocks. Similarly, an atom can be analyzed wrt constituent energy states and their transformation, e.g., electrons orbiting nucleus.

Similarly an electrical circuit with its own energy source, e.g., a flashlight with its own battery, with two potential states, on and off. Of course, in actuality the battery will discharge over time and with use, but we don't need to take that into consideration in the model.

Similarly with a closed economic system with a pairs of and consumer-producers that barter (Crusoe and Friday). Or with a representative consumer and a representative producer and a single money source (bank).

From these simple models of closed systems it is possible to build out to great complexity, which may be more representational of the events whose behavior are actually interested in.

Oliver Davey said...

Tom: Oliver, here we are building up a model from a simple closed system w/o govt, and then adding govt. Haven't added an external sector, which further complicates the system's sectoral relationships.

Yes, I understand that and also agree it's a good model to start with. But one could construct a narrative with a different starting point. There are boundaries within that system that are very meaningful, too. E.g. the boundaries between business / non business, or between finance / non finance, or national boundaries. I'm saying MMT language is not open to a choice in starting points. It's a good, powerful model to begin with, but this strength is also its weakness.


Vimothy: Your savings are your asset—the bond—, which is an obligation of Apple to you.

You're right, of course, in that primary investment creates a new financial obligation. The MMT point is that it takes another kind of funds, namely savings of government/bank liabilities, to buy those bonds in the first place - and also to retire them. A stock which the investment / production of real goods has no influence on. One could say, bonds are derivatives of money, because they are further down in the pyramid of assets / liabilities. And unlike bank credit, which is a derivative of base money, bonds are not guaranteed to trade at par wiht currency, nor can they be used to make payments. They are thus not money proper. They are however savings.

Oliver Davey said...

It is neither produced nor consumed, and it is not created by the circular flow of production and consumption of goods although it enters into it. Money also influences this circular flow so it is not neutral to it as merely a bookkeeping function.

I like this view:

it is worth exploring the very payment that gives rise to this double-entry. In the case in
point, Post Keynesians and Circuitists agree that firms have to finance their expenditure on the
factor market by obtaining a loan from banks. As Lavoie put it, ‘[t]hese flows of credit then reappear
as deposits on the liability side of the balance sheet of banks when firms use these loans to remunerate
their factors of production’ (Lavoie, 1984, p. 774). In fact, this is the loans to deposits causality put
to the fore by both strands of thought, first spelled out by Withers back in 1909 (see Realfonzo,
1998, ch. 6).
Now, to focus exclusively on the point at issue in this section, one has to consider that in the
income-generating process depicted in Table 1, the money creation process carried out by the bank
only provides the economy with the number of money units asked for by the firm (on the assumption
that the firm’s creditworthiness satisfies the benchmark set by the banker). To state it clearly, it is the
remuneration of labour that gives a purchasing power to money, which, as such, is a mere numerical
form of no value whatsoever. Were it not for the monetisation of the production process, banks
would be unable to create purchasing power on their own. So, bank deposits are a ‘liquid, multilaterally
accepted asset’ (Chick, 2000, p. 131), because they are the organic result (that is, a stock magnitude)
of two intimately related actions (or flows): (1) creation, on the monetary side, of the numerical
form of payments (money proper) by the banking system, and (2) production, on the real side, of
physical output (money’s worth) by the non-bank public, that is, firms and workers taken together.8
So, the flow of money and the flow of production are complementary aspects of the same (incomegenerating)
process. ‘From the beginning, banking and productive systems thus contribute to the
determination of a unique macroeconomic structure’ (Cencini, 1997, p. 276).
In sum, from this point of view money as such is a flow, whose result is a stock

http://www.csbancari.ch/pubblicazioni/RMElab/rossi.pdf

Oliver Davey said...
This comment has been removed by the author.
Oliver Davey said...

In sum, from this point of view money as such is a flow, whose result is a stock (of liquid wealth)
in the form of bank deposits. Contrary to the ‘cloakroom theory of banking’ à la Cannan (1921),
6
bank deposits are not a financial asset sui generis, originating in some ‘central mystery of modern
banking’ (Chick, 2000, p. 131). According to the theory of money emissions, bank deposits are the
alter ego of physical output, and come to light as soon as the latter is monetised via the remuneration
of wage-earners by firms. The purchasing power of bank deposits has therefore nothing to do with
the agents’ trust and confidence in the banking system. In this framework, let us emphasise it, money
balances are net worth because they are output – before final consumption of the latter takes place
on the goods market.9 Then, when output is sold on the market for produced goods, an equivalent
(some would say identical) sum of bank deposits are destroyed, since deposit holders transform a
liquid store of wealth into a physical value-in-use, or, to put it in the phraseology of Fama (1980),
they exchange a monetary form of wealth for a real form. This exchange, taking place on the product
market, destroys a sum of bank deposits equal to the amount of money wages adding up to the
production cost of output sold.

Anonymous said...

So, one of the things that we’ve been discussing is whether investment somehow removes funds or removes demand from the economy. (Remember that investment also has a specific meaning in macroeconomics, which is the flow of expenditure on new capital.) There’s no difference in principle between getting your income from a company that makes final consumption goods and from a company that makes final investment goods. This means that there’s no difference from the point of view of the aggregate income of the economy between consumption expenditure and investment expenditure. If 50 percent of GDP were invested, then saving for the economy would be 50 percent of GDP and income would still be 100 percent.

Anonymous said...

Oliver,

You're right, of course, in that primary investment creates a new financial obligation. The MMT point is that it takes another kind of funds, namely savings of government/bank liabilities, to buy those bonds in the first place - and also to retire them.

It’s easy to confuse this argument. What’s needed from a system wide point of view is that the banking system in toto has sufficient reserve balances to enable payment and settlement to take place. But this is simply the basic reliance, always present, of the banking system on outside money as a means of final settlement. No central bank to make markets and provide a clearing function, no banking system—at least as currently imagined.

Does this mean that if the government doesn’t deficit spend, I can’t buy a bond? No! As long as the banking system has sufficient reserves to function (this can be a quantity that is arbitrarily small relative to the size of the economy), then the path of government deficits is completely irrelevant. The banking system will continue to operate regardless of the government’s tax and spend plans.

Anonymous said...

By the way, your comment reminded me of Perry Mehrling's "natural hierarchy of money":

http://ineteconomics.org/sites/inet.civicactions.net/files/Lec%2002--The%20Natural%20Hierarchy%20of%20Money.pdf

Oliver Davey said...

There’s no difference in principle between getting your income from a company that makes final consumption goods and from a company that makes final investment goods. This means that there’s no difference from the point of view of the aggregate income of the economy between consumption expenditure and investment expenditure. If 50 percent of GDP were invested, then saving for the economy would be 50 percent of GDP and income would still be 100 percent.

Except for that part of income that is spent neither on consumption nor on investment, i.e. not at all. And the reserves for this residue must also come from somewhere. That's where government comes in via the CB and the settlement system.

Does this mean that if the government doesn’t deficit spend, I can’t buy a bond? No!

In theory, no. But the stylized fact remains that such a credit economy is unstable due to the necessity of a constant growth / stable velocity for the credit constrained individuals to service the debt that comes with it.

Anonymous said...

No part of aggregate income is not "spent", because, by identity,

Aggregate Income = Aggregate Expenditure

All income originates as someone, somewhere's expenditure.

But the stylized fact remains that such a credit economy is unstable due to the necessity of a constant growth / stable velocity for the credit constrained individuals to service the debt that comes with it.

I don't see why this should mean that government bonds are needed for private individuals to enter into a private debt contract.

My point was that, just because the economy needs a certain amount of outside money for final settlement, this doesn't mean that it needs a certain amount of deficit spending or government debt. They are separate issues.

Oliver Davey said...

No part of aggregate income is not "spent", because, by identity,

Aggregate Income = Aggregate Expenditure


Money is first spent into existence through the credit creation process. It is then recorded as income for the recipient until it is again spent. The final recipient does not spend it. So his income is the balancing item to the original expenditure, leaving income = expenditure, but stock of money unequal before and after period. The other possibility is that the final recipient uses it to discharge the orginal debt, in which case it is again income for the bank while simultaneously destroying the money. This is the implication of the credit theory of money. Your explanation never addresses how money is created or destroyed.

paul meli said...

"So, one of the things that we’ve been discussing is whether investment somehow removes funds or removes demand from the economy."

Investment (once successfully capitalized) re-distributes funds from worker/consumers to businesses, a much smaller cohort. This is a continuous process, a flow. This removes demand from the economy.

It follows that the cohort of spenders have less and less of the funds necessary for spending after each investment cycle. The economy is a function of the size of the spending flow, not levels of dollars. They are related, but not like one would think.

The flow as described is a decaying process within the economy. It can be counteracted in one of two ways:

• Put money in the consumer/worker cohorts pockets through deficit spending.

• Make the tax code more progressive. Raise taxes on the rich, lower taxes on the workers.

(a debt jubilee is a special case of the first option)

Monetary policy has no bullets here.

"I don't see why this should mean that government bonds are needed for private individuals to enter into a private debt contract."

Government bonds aren't "needed". They are a convenience.

"My point was that, just because the economy needs a certain amount of outside money for final settlement, this doesn't mean that it needs a certain amount of deficit spending or government debt. They are separate issues."

Sure it does. There is no other source of outside spending, unless you are referring to net exports (surplus), an unlikely scenario, and in any case an option that would be impossible (literally) for every country to take advantage of at the same time.

Someone has to run deficits.

Government debt is a misnomer, it is an accounting abstraction. It isn't a debt that is liable to be repaid, nor should it be.

When the U.S. government creates debt, it is creating an asset, a bond in the non-government. The government is borrowing from the non-government right pocket and spending into the left pocket, figuratively speaking, and the level of net assets in the non-government increases as a result, because the government creates an equal amount of new spending along with the bond.

When this transaction is viewed wrt closed system principles it becomes clear that what appears to be happening is not. The "loan from the system" cannot increase the level of it's own assets. This can only come from an injection.

This is true of any case, be it private bond issue, stock issue, privately-issued money (Bitcoin), etc.

They all must be capitalized through deficit spending (in the aggregate) otherwise some will win, but others will lose an equal amount. Zero sum. The closed sytem cannot gain without external input.

The government is simply injecting additional net assets into the system that the system cannot provide for itself.

The system cannot create net assets unless you include the "value of real assets." The "value" of real assets exists only in the minds of economic participants.

"Value" is meaningless wrt the underlying system. There is no transfer function, although there is an abstract relationship.

Value can disappear overnight. The actual components of the closed system (dollars) can not.

Oliver Davey said...

They all must be capitalized through deficit spending (in the aggregate) otherwise some will win, but others will lose an equal amount. Zero sum. The closed sytem cannot gain without external input.

Government = the people = inside. The liability incurred by a government deficit is your's and mine just as with bank credit. It's the different debt dynamics due to respective, legally determined, positions on the 'pyramid of assets & liabilities' that distiguishes governments & (central) banks from us mere mortals.

Anonymous said...

[Long comment]

Paul,

Investment (once successfully capitalized) re-distributes funds from worker/consumers to businesses, a much smaller cohort. This is a continuous process, a flow. This removes demand from the economy.

Investment is expenditure on new capital equipment. There’s no difference in principle between expenditure on consumption and expenditure on investment, as far as employees of firms producing this output are concerned. Any expenditure must result in someone receiving that expenditure in payment for some good or service. Whatever the expenditure is on (government consumption, private investment, anything), it becomes somebody’s income.

Earlier you agreed that this was the case:

“Investment is a form of spending.”

“Apple takes your savings funds and spends it on plant, labor, resources used to generate goods and services.”

“Your Apple bond does not "remove" any funds from the economy.”

Imagine an economy with only two firms. One firm makes cupcakes. One firm makes machines that make cupcakes. Everyone in this country is really into cupcakes. It’s a national obsession. Three quarters of all spending is on cupcakes, the rest on machines to make the damn things. When someone buys a cupcake, the cupcake making business gets paid. From this revenue it distributes wages to its workers and profits to its owners. When someone buys a cupcake-making machine, the cupcake-making machine making business gets paid. From this revenue it distributes wages to its workers and profits to its owners—exactly the same as the cupcake making business. It doesn’t matter, from the point of view of income, whether “demand” is for investment or consumption goods, because it all results in someone getting paid, which means that someone earns an income.

Remember: buying a bunch of Apple bonds is not “investment”. Buying stocks and bonds is “saving”. (Okay, technically I suppose that’s more like portfolio choice, but you get the picture.) Expenditure on new capital equipment is investment.

It follows that the cohort of spenders have less and less of the funds necessary for spending after each investment cycle.

What is an “investment cycle”?

Who is the “cohort of spenders”?

Do you mean that investment causes average wages in the economy to be held down?

The economy is a function of the size of the spending flow, not levels of dollars.

I think that’s correct to an extent. The size of the economy is definitely not a function of the size of the level of dollars (because then the government could make the economy arbitrarily large just by increasing the supply of outside money). The size of the economy is exactly equal to the amount of spending. That’s what this means:

Aggregate Income = Aggregate Expenditure

The flow as described is a decaying process within the economy. It can be counteracted in one of two ways:

What flow? The flow of expenditure? By definition, the flow of expenditure equals the flow of income. Generally, it grows over time. In the US it has never been higher.

It can be counteracted in one of two ways:

• Put money in the consumer/worker cohorts pockets through deficit spending.

• Make the tax code more progressive. Raise taxes on the rich, lower taxes on the workers.


We’re running a bunch of arguments together here. Just because a certain amount of income is saved and a certain amount of output is invested, that doesn’t mean that we end up with income inequality or stagnant incomes. Just because we end up with stagnant incomes, doesn’t mean that the economy is going to break and GDP will decay to zero. The economy will continue to work fine, it will just be increasingly crappy for the majority—but that has nothing to do with the size of investment expenditure relative to aggregate demand.

Anonymous said...

[Cont.]


Government bonds aren't "needed". They are a convenience.

How? In what sense? Convenient for what?

Sure it does. There is no other source of outside spending, unless you are referring to net exports (surplus), an unlikely scenario, and in any case an option that would be impossible (literally) for every country to take advantage of at the same time.

What do you mean “outside spending”? Spending within the private sector from another sector? Why would that be necessary? What does it have to do with my ability to buy a bond from Apple?

Someone has to run deficits.

Why? For who?

Government debt is a misnomer, it is an accounting abstraction. It isn't a debt that is liable to be repaid, nor should it be.

What’s the point in the private sector acquiring it, then, if it’s just an accounting fiction that will never be repaid?

When the U.S. government creates debt, it is creating an asset, a bond in the non-government.

You just said it was a fiction. If the government isn’t liable for anything, in what sense can its liabilities be assets?

The government is borrowing from the non-government right pocket and spending into the left pocket, figuratively speaking, and the level of net assets in the non-government increases as a result, because the government creates an equal amount of new spending along with the bond.

This is just a description for any borrowing anywhere in the economy. When I borrow, I borrow from the rest of the economy, and spend it. Then the rest of the economy gets my spending plus my liability.

So why is the government so special? The only difference seems to be that I have to repay my debts whereas the government has no intention of doing so.

When this transaction is viewed wrt closed system principles it becomes clear that what appears to be happening is not. The "loan from the system" cannot increase the level of it's own assets. This can only come from an injection.

Only government liabilities can come from the government. Only Apple liabilities can come from Apple. Only Paul liabilities can come from Paul. The economy can and does generate assets continuously, regardless of government borrowing. Tangible assets are the result of investment. Investment is occurring continuously in any major economy. Gross financial assets grow continuously. Net financial assets require for obvious reasons someone “outside”. For instance, the household sector lends to the corporate sector. The domestic private sector lends to the government. The economy as a whole lends to another country.

This is true of any case, be it private bond issue, stock issue, privately-issued money (Bitcoin), etc.

They all must be capitalized through deficit spending (in the aggregate) otherwise some will win, but others will lose an equal amount. Zero sum. The closed system cannot gain without external input.


What do you mean “capitalised through deficit spending”? Why will some lose and some win? Lose and win what?

There isn’t a fixed quantity of funds in existence that only the government can add to. Is that what you’re assuming here?

Anonymous said...

[Cont.]

The government is simply injecting additional net assets into the system that the system cannot provide for itself.

By definition, only the government can provide government liabilities. Only the government can lend to the non-government. That’s just a function of the way we’ve divided up the economy for the purposes of discussion. Only Paul can supply net financial assets to the non-Paul economy. So what?

The system cannot create net assets unless you include the "value of real assets." The "value" of real assets exists only in the minds of economic participants.

Take a set. For any arbitrary partition of this set, the only supplier of NFA to either subset is the other subset. That’s just the definition of NFA again.

“Net assets” for a given sector are its assets less its liabilities. This will come out as its tangible assets plus its net lending to other sectors. If you want to think about the domestic private sector then it has tangible capital plus net lending to govt plus net lending to foreign sector (either of the latter two can be negative)

"Value" is meaningless wrt the underlying system. There is no transfer function, although there is an abstract relationship. .

What do you mean? What is a transfer function? What abstract relationship?

Value can disappear overnight. The actual components of the closed system (dollars) can not.

What do you mean? When you say value, what are you referring to? How can it disappear? When you say that dollars can’t disappear overnight, do you simply mean that the only way the supply of dollars can change (in nominal terms) is if the government changes them (in the same way that the only way the supply of Apple bonds can change is if Apple changes them)?

Tom Hickey said...

vimothy Aggregate Income = Aggregate Expenditure

All income originates as someone, somewhere's expenditure.


Yes. What this implies is that in a closed system w/o govt, no net saving is possible.

In a closed system wi govt., govt expenditure provides NFA to nongovt, enabling nongovt net saving in aggregate.

Anonymous said...

Tom,

Yes. What this implies is that in a closed system w/o govt, no net saving is possible.

No net saving is possible for the whole economy--that's the definition of net saving again. Net saving is still possible for sub-sectors of that economy. For instance, the household sector could run a surplus and the corporate sector a deficit.

Tom Hickey said...

No net saving is possible for the whole economy--that's the definition of net saving again. Net saving is still possible for sub-sectors of that economy. For instance, the household sector could run a surplus and the corporate sector a deficit.

Yes. That's what my next sentence states. I though the connection was obvious, but apparently it was not.

Anonymous said...

Tom, What I'm trying to say is: What does it matter if only the government can supply net financial assets to the non-government? Why is that such a big deal? It doesn't look like a very interesting statement.

Anonymous said...

Within the private sector, only the corporate sector can supply net financial assets to the household sector. But so what?

Anonymous said...

Within the household sector, only arbitrary subsector A can supply NFA to arbitrary subsector not-A. Therefore, arbitrary subsector A should run continuous deficits to meet arbitrary subsector not-A's NFA demand.

Tom Hickey said...

vimothy, the importance is the essence of the MMT approach to sectoral balances and functional finance, i.e., the heart of MMT as a policy science.

The MMT contention is that changing non-govt saving desire is a given and increasing saving desire results in demand leakage that results in potentially significant economic consequences in that the sectoral balance identities must hold through the changes.

What this implies is that an increase in non-govt saving desire (combined domestic private and external balance) can reduce effective demand to the degree that what the economy can produce given available resources will not be sold (unplanned inventory will build at FE).

This results in sub-optimal performance (cutting back inventory build up) and idling of resources (laying off workers), one manifestation of which is a rising unemployment rate and another an increasing output gap.

Govt can offset an increase in non-govt saving desire that results in lagging effective demand by increasing its fiscal deficit correspondingly to the non-govt surplus in order to accomodate the saving without reducing effective demand, thereby maintaining optimal economic performance evidenced by optimal output (no gap), full employment (less transitional) and price stability (moderate affect on price level).

Tom Hickey said...

Within the private sector, only the corporate sector can supply net financial assets to the household sector. But so what?

W/o outside money, the net is zero, so the net saving is offset by net dissaving. Factors associated with relationship of net saving and net dissaving have economic consequences, chiefly wrt the sustainability of private debt indebtedness but also wrt social, political and economic consequences of inequality of wealth, financial and non-financial.

Anonymous said...

Tom,


That (i.e. your comment at October 5, 2012 1:08 PM) sounds like standard Keynesian reasoning. If there is an unanticipated increase in saving, the government should run a deficit to prevent the economy from contracting.

What’s uniquely MMT about this argument?

What Paul seems to be saying is something more: that the economy needs a constant supply of new NFA in order to survive, as the Earth needs light from the sun.

Anonymous said...

W/o outside money, the net is zero, so the net saving is offset by net dissaving. Factors associated with relationship of net saving and net dissaving have economic consequences, chiefly wrt the sustainability of private debt indebtedness but also wrt social, political and economic consequences of inequality of wealth, financial and non-financial.

With any financial asset, the net is zero. If one sector net lends to another sector, then one sector has positive net lending, and the other negative. Then when you consolidate the two sectors, that net lending disappears. It doesn’t matter if we’re talking about inside money, outside money, or any other kind of financial asset.

When we consolidate government and private sector, the national economy has net foreign assets, but lending to the government nets to zero, just like the way borrowing flows within the private cancel out when we consolidate household and corporate sectors.

Tom Hickey said...

That (i.e. your comment at October 5, 2012 1:08 PM) sounds like standard Keynesian reasoning. If there is an unanticipated increase in saving, the government should run a deficit to prevent the economy from contracting.

What’s uniquely MMT about this argument?


Yes. "Standard" Keynesian, but not all "standard" Kenynesian views are the same. PKE, including MMT rejects Samuelson's bastard Keynesianism and New Keynesianism, which is what the mainstream now associates with "Keynesianism." MMT is in the PKE tradition of heterodox economics, i.e, no neoclassical synthesis.

The MMT take is a combination of Godley's SFC approach to sectoral balances, Abba Lerner's functional finance, Knapp's chartalism, and Minksy's financial approach and an adaptation of MInsky's JG, built on a PKE foundation.

Anonymous said...

Okay, but it's not an argument that says that the government must run continuous deficits in order to satisfy something called "net saving desires". It's an argument that says that fiscal policy should intervene in demand-led recessions.

Tom Hickey said...

With any financial asset, the net is zero. If one sector net lends to another sector, then one sector has positive net lending, and the other negative. Then when you consolidate the two sectors, that net lending disappears. It doesn’t matter if we’re talking about inside money, outside money, or any other kind of financial asset.

True but who is at risk in case of inability to pay is a big deal, as the GFC revealed when govt took massive amounts of non-performing liabilities on its balance sheet and provided unlimited liquidity. Otherwise the financial sector would have collapsed and the US economy would have spun into depression, probably dragging the global economy with it.

When the private sector's saving-debt balance is such that debtors get overextended, then the system is in danger of debt-deflation. Absent an increase in exports and more even income distribution, then only govt can provide the offset.

Such an ad hoc offset, MMT would call a "bad" deficit, although a necessary one to prevent further damage.

A "good" deficit is one that automatically adjusts for changing non-govt saving as a preventative measure by employing policy tools like variable tax rate, automatic stabilization and buffer stock of employed.

Tom Hickey said...

Okay, but it's not an argument that says that the government must run continuous deficits in order to satisfy something called "net saving desires". It's an argument that says that fiscal policy should intervene in demand-led recessions.

That's "standard" Keynesian thinking, which MMT considers sub-standard in light of subsequent developments, as well as a return to what Keynes actually said.

MMT holds that using the sectoral balance approach and appropriate economic policy, changes in non-govt saving desire can be neutralized by policy tools based on functional finance that operate automatically.

Oliver Davey said...

re: No part of aggregate income is not "spent", because, by identity,

Aggregate Income = Aggregate Expenditure
second attempt:

Wrong way around. All expenditure is someone's income, but that income need not be spent. The implication is absurd.

And i'd add to Tom's last comment that, orthogonal to the issue about automatic stabilizers, MMT does assert that deficits are the norm and that there is no need, other than to satisfy mathematical models, for any entity, including the government, to balance its budget over any however defined cycle. And it is through their study of central bank operations that they can prove the stability of such a proposition under certain conditions that are usually referred to as currency sovereignty.

Anonymous said...

Oliver,

That's an identity. It's true by definition. It doesn't matter which way round you write it. AE = Y and Y = AE are equivalent statements.

Oliver Davey said...

you said, all income is spent. it is, but not by the income receiver. do you spend all you get? hardly...

Unknown said...

@Ramanan

Ignoring the foreign sector, the only net money in an economy comes from public sector deficits.
But this doesn't mean that the "money supply" at any particular time is equal to of comprised solely of government debt.
The vast bulk of the money supply is comprised of bank credit.

Agreed?

So even though the private sector balance of net financial assets must be zero, at any one time there may be a positive balance of outstanding credit.
And it is this outstanding credit that adds to aggregate demand when viewed at a specific time.

For example, the public sector runs a balanced budget, leaving our economy with zero net financial assets.
A bank then makes a $100 loan which is immediately spent.
So will the private sector has a zero net balance, the money supply is $100.
When viewed this specific time
demand = income + change in debt
100 = 0 + 100

Would you agree with this?