Nevertheless, this observation from a recent talk he gave is interesting:
One part of the discussion that I found quite notable was that, even after showing empirical evidence on the impact that rising and then falling private debt had on the economy both now and during the Great Depression, I couldn’t convince several of the academics in the audience of the importance of private debt: they kept coming back to “one person’s debt is another person’s asset, therefore the level of debt doesn’t matter”. They therefore argued vehemently that the distribution of debt was important, but its aggregate level was irrelevant.This is the core miscomprehension about debt that economists have. They see bank debt as the same as personal debt -- if I lend $10 to you, then I do not have $10 to spend myself, so the aggregate level of spending money available is unchanged.
However, bank lending does not work like this. If a bank lends $10, the $10 gets deposited in some other bank and is therefore available to lend again, limited only by capital requirements. In this sense, one person's asset is not another person's liability, and the total (gross) quantity of financial assets can expand and contract, while the total (net) quantity of financial assets remains the same.
I don't think that's the case at all. Steve has shown that a private sector credit circuit can work all by itself in his models. Which is very useful to know.
ReplyDeleteAnd the heuristic assumptions are clear - banks have to lend omnisciently and if the capital limits are not absolute but relative to bank equity you get a very unstable system (even with omniscient lending).
So far so good.
The reason he hasn't got a financial asset source in his models is because he hasn't done that bit yet. He's only modelled the Horizontal, not the Vertical.
He's now looking at the Vertical and you get a sense in his lectures that the nature of vertical is causing a touch of cognitive dissonance - which I think is captured beautifully in the phrase 'Central Bank debt is very strange debt indeed'.
See his lecture which contains the point (At ~17 to ~20 minutes where he briefly discusses the central bank and the government)
My problem with Steve's models is that he calls them double entry and then only has one entry or three in the rows of his double-entry tables.
'The clue's in the name, mate' :)
"This is the core miscomprehension about debt that economists have. They see bank debt as the same as personal debt..."
ReplyDeleteDunno. I think economists are an insecure bunch that like to grasp each other's straws, even when the other straws are statements like "macroeconomics in this original sense has succeeded: Its central problem of depression-prevention has been solved..."
All the idle chatter is about government debt and none is about private debt. Why that is, I don't know. But I know the problem has not been solved, and it will not be solved until we deal with the excessive accumulation of private-sector debt.
And then (after we solve the problem) we have to deal with its father, which is that we insist on expanding credit-use as the way to achieve growth.
Art
The reason they dont think private debt is a problem Art is that they view it as debt owed to each other as WS said. Ive seen Nick Rowe AND Paul Krugman state as much. They think if I am in debt that someone somewhere is in surplus and therefore that surplus is being invested and not causing a monetary imbalance. They fail to see that virtually all of us are indebted to the banking system, which is really a third party outside the economy so to speak. A banking systems surplus is of no more use than a govts budget surplus. Banks dont "spend" their surpluses. They exist only on their books as "outstanding loans" owed to them. But when we cant pay them back they freeze up.
ReplyDeleteMost academic economists do not understand banking. They think a bank is just an intermediary between savers and borrowers.
Greg: Exactly correct. There is a "loanable funds" market where banks take short term deposits, and lend them out as long term loans, x10 (which is a magic, and according to Austrians, criminal, money multiplier).
ReplyDeleteOf course, banking, and by extension the economy, doesn't work anything like this.
Neil, I just watched and listened to the section you recommended and that also highlighted Keen's non-acceptance of the MMT - he doesn't see it as a dynamic model and sees double entry accounting as static (seemingly making the same mistake he's critiquing Graziani for) which is the same argument that Bob Murphy and other Austrians have made. Bill Mitchell replied to this in his "MMT – an accounting-consistent, operationally-sound theoretical approach" post.
ReplyDeleteAnd yet he's attempting to use double entry as part of the dynamic model.
ReplyDeleteI have to say I struggle to see what fundamental difference there is between the two, and I've been looking and asking.
Perhaps it's my open source mindset, but I just want to reconcile the differences and get rid of the 'fork'.
http://www.asymptosis.com/winterspeak-is-right-economists-dont-understand-that-debt-matters-and-inflation-does-too.html
ReplyDeleteGreg: "They think if I am in debt that someone somewhere is in surplus and therefore that surplus is being invested"
ReplyDeleteThat's because Kuznets and company designed the national accounts to portray it that way, which allowed the NIPAs to ignore changes in both net and gross financial assets (eventually tracked in the flow of funds accounts).
If you take $10K out of the bank and buy 10 computers for your employees, in the NIPAs that is called both investment *and saving.*
Would you call that "saving"?
http://www.asymptosis.com/savings-equals-investment-equals-zero.html
It is in real terms. The problem is that 'saving' is a vastly overloaded word like 'inflation'. The first thing you have to do when using the word is work out which definition you are talking about.
ReplyDeleteThe term you mention is derived from the capitalisation process in accounting. Buying 10 computers stores up some 'computerness' for future accounting periods using only some of it in this one. So there is a stock of 'computerness' left which generalises to 'saving'.
Steve: Neil's exactly right. It's just an identity, so you need to be thoughtful in its interpretations.
ReplyDeleteAny real output produced in period X but not consumed in period X (and therefore available in period X+1) is "real investment". The nominal income earned in period X for the output produced in period X but not spent on that output must be available in period X+1. That is "nominal savings".
Neil and Winterspeak: "The problem is that 'saving' is a vastly overloaded word like 'inflation'. The first thing you have to do when using the word is work out which definition you are talking about."
ReplyDeleteRight -- along with "investment," "assets," and "capital" (among others). That's exactly the point of my post. The NIPAs take a perfectly good word -- investment -- and say that "saving" means exactly the same thing.
But investment is spending -- the exact opposite of saving (which -- if we're talking about money -- is ineluctably *not* spending).
If there's no money and you're a wheat farmer and only consume part of your wheat, yes, the remainder could be called saving. But in fact the wheat farmer consumes none of his wheat; he sells it all. We live in a financialized monetary economy.
This self-contradictory definition/identity allows/encourages economists to ignore the relationship between financial assets and real assets, and the very thing that Keen, quite sensibly, makes such a big deal of -- private debt issuance.
I'm just sayin: for the national accounts we should be using a Godley-like system. Notably in his key paper on stock-flow consistent accounting, the word "saving" does not appear.
I'm quite interested in the term "nominal saving." I find that it's absent from the mainstream textbooks. Is it a Godley term? (I'm waiting for the new edition to come out to get his book.)
Steve: Yup, the technical definition of these terms in the identity is different from how they are used more commonly. That creates confusion.
ReplyDeleteSo, to understand the identity, you need to understand how it defines its terms, and how these are different from the more common uses of those same words.
But to *really* understand the identity, you need to be able to think in real and nominal terms, with stock flow consistency, across two periods. My understanding of real vs nominal crystallized in the context of S=I.
I don't like Steve Keen because although he gets private debt issuance right, he gets public "debt issuance" wrong.
Left-over wheat is not "saving", it is "inventory". Saving is a purely nominal concept, and it simply means "not spending". It is not an action, it is the lack of an action.
"nominal savings" is a term I came up with, but I don't pretend I'm the first person to ever use it. It isn't technical, just precise.
""nominal savings" is a term I came up with, but I don't pretend I'm the first person to ever use it. It isn't technical, just precise."
ReplyDeleteHave you explained it? Haven't found it.
"I don't like Steve Keen because although he gets private debt issuance right, he gets public "debt issuance" wrong."
I can see what you're saying; he operates within the context of govt issuing debt and "saving."
What I've been wondering: in that context, shouldn't his definition of AD add/subtract PMC purchases/sales?
Also, you keep saying "I don't like Steve Keen." Don't you mean, "I disagree with some of Steve Keen's ideas"?
Hi Steve:
ReplyDeleteWhen I say "I don't like Steve Keen" what I mean is "Steve Keen is wrong". He's 50% right though, so I don't know why he cannot make this final leap, but I think that overall this makes him worse.
I talk about nominal savings in the context of S=I, as you really do need to be strict about real vs nominal, and action vs non-action to make it work.
As for your question, I'm not sure what you mean by AD or PMC.
"What I've been wondering: in that context, shouldn't his definition of AD add/subtract PMC purchases/sales?"
ReplyDeleteTypo. Shouldn't he add/subtract open market purchases and sales from his definition of aggregate demand? I think he's got govt defct/surplus already in there, and trade surplus/deficit, cause he's using GDP to start with.
??
Have you done a writeup of your concept of nominal savings? Link?
Thanks,
Steve
Hi Steve:
ReplyDeleteThanks for your clarification.
I don't think the Fed's open market purchases have any impact on supply or demand, and I think that's pretty obvious in both the data and the theory.
Under the various QE regimes, the Fed's balance sheet has expanded dramatically, and yet we don't see any impact on aggregate demand in the data.
In theory, why should the term structure of the Fed's assets matter? If the Fed owns $100 of assets, say, why does it matter if $10 of those are in 5 year Treasuries and $90 in 10 year Treasuries or the other way around?
As for "nominal savings" there's probably some S=I posts on this site where I talk about it in that context, but the search is a mess and I'm very disorganized. Ya gets what ya pay for!
Nevertheless, all I mean by nominal savings is nominal money that you earn, that you do not spend.
That's it! It's just a number in your bank account.
"Nevertheless, all I mean by nominal savings is nominal money that you earn, that you do not spend."
ReplyDeleteSo if I take $10K out of my bank account and *spend* it to buy computers for my employees, that is *not* (nominal) saving, right?
Even though according to the NIPAs, it is "saving."
If the Fed was not paying IOR. so excess reserves didn't exist (much or at all), do you think open market purchases and sales would affect AD?
ReplyDeleteHi Steve:
ReplyDeleteI haven't spent much time on decoding what NIPA is, so I don't want to speak to that, but if you draw down your bank account to buy some computers:
- There has been a real asset (inventory) transfer as the computers move from the seller to you
- There has been a nominal savings transfer as $ goes from numbers in your bank account, to numbers in the seller's bank account
The seller has booked income in this period and not spent it, so his nominal savings goes up.
you have dissaved in this period, so your nominal savings goes down (you have no income)
The Fed paying IOR does support AD as it's increase net financial assets in the non-govt sector, which is fiscal and expansionary.
ReplyDeleteAt the same time, since FFR is so low, fiscal transfer through interest on other Fed liabilities is low, which is contractionary.
Net net I think it's a wash
No, what I'm asking:
ReplyDelete*If* the Fed does not pay IOR.
*So* excess reserves basically don't exist.
*Then* does an FOM purchase increase (at least potential) AD?
Interested to hear your thoughts.
Another that I keep asking people: if IOR went to zero tomorrow, what do you think would happen?
Hi Steve:
ReplyDeleteReserves are just checking accounts, held as liabilities at the Fed and assets at the banks. Their primary purpose is check clearing within the banking system. The Fed also sets FFR by draining the reserves through open market operations, which makes the clearing price for the overnight interbank lending market tick up. That's it.
If the Fed buys a Treasury, then the quantity of reserves in the system go up. Then the market clearing price for reserves falls. Thus the FFR falls. All that's happening here is that the Fed is engaged in conventional monetary policy and is lowering interest rates. I do not think this is expansionary unless it triggers private credit expansion which would increase AD.
If there are excess reserves, then the FFR is zero and additional OPM has no impact regardless of whether there is IOR or not.
I have to admit I'm having trouble getting this whole thing. Every time I say private debt matters and the standard assets=liabilities story isn't true I am told that assets = liabilities is true by construction.
ReplyDeleteWhere can I get a simple explanation of why, even if this is 'true' by construction, it doesn't matter?
Private debt does matter, and assets=liabilities is true. But liabilities includes both non-equity liability and equity (liability).
ReplyDeleteRead "required readings" at moslereconomics.com
Unlearningecon:
ReplyDeleteWhen a bank issues a $100 loan, it puts money in someone's account (a liability for the person: the person owes the bank money), and has a credit on its own account.
They *net* to zero.
But there's and extra $100 out there, and there's an extra $100 IOW out there. *Gross* debt has increased, as has the money stock.
That hundred dollars, being spent, increases the quantity of money in circulation.
ReplyDeleteThe future interest cost of that hundred dollars adds to the accumulated cost of having money in circulation, a cost borne largely by borrowers in the private sector.
Soon enough, that hundred will find its way into someone's savings. The quantity of circulating money will thus decrease, and the hundred will have to be replaced.
Replacement borrowing yet again increases the cost of having money in circulation. And it may well be that the *same* hundred dollars is loaned out again, though interest on the first borrowing is still being paid.
Winterspeak: I've been wondering a lot lately about the concept of "money in circulation." Doesn't really make sense to me, a "stock" of money "in circulation" seems to confuse flows and stocks.
ReplyDeleteAnother way to think about it: there's a big pool/stock of "money," stored in all sorts of financial instruments -- checking accounts, stocks, bonds, etc.
In the course of a year (or whatever period) a certain amount of that money transits through the real economy to purchase real goods and services. Each transaction goes, almost instantly, from the financial pool back into the financial pool -- from one person's/business's account to another.
That stock of money is never "in circulation" -- it's always in some financial instrument. But it does *circulate.*
Money also circulates within the financial system/economy (these flows actually dwarf the flows through the real economy), but what matters for the well being of the economy is how much circulates (briefly, instantaneously) through the real economy. Those purchases and sales are where the surplus is realized.
Make sense?
"That stock of money is never "in circulation" -- it's always in some financial instrument. "
ReplyDeleteNot necessarily. If it flows into a bank as repayment for a loan it disappears, much as it disappears when paid as taxation.
So a flow can end up in a 'black hole'.
"If it flows into a bank as repayment for a loan it disappears"
ReplyDeleteRight. Private debt issuance and retirement expands and contracts the pool of "money"/financial securities.
But that doesn't mean there's a *stock* of money "in circulation," distinct from the stock of financial assets.
If there's a disproportionate stock of money held in liquid forms -- checking accounts, for instance -- that means the account holders *can* easily spend it (transfer it to others' accounts) to purchase real-economy goods and services. But that doesn't mean they will, or that their reason for holding it in that form (instead of bonds or stocks, for instance) is because they're raring to spend. They may just prefer the liquidity and (due to deposit insurance etc.) lack of risk.
As for myself, there is the money which comes to me as income, which must be circulating if it comes to me. And there is the money I spend, which also must be circulating, and the money I save (which is *NOT* circulating).
ReplyDeleteSurely, Steve, you do not mean to suggest that I cannot receive income, nor spend, simply because of some confusion of stock and flow?
"In circulation" is not a place. It is a description. It it "circulates" then it is "in circulation". Far as I can see.
The pool of money that people expect to spend is in circulation. The pool of money that people hope not to dip into and spend is in savings and is not circulating.
If I take a dollar out of savings and spend it, it increases M1 money but not M2. It increases money in circulation.
Happy Thanksgiving.
hi all:
ReplyDeleteTake two periods: 1 and 2.
In period 1, everyone has a certain amount of money. In period 2, everyone has a certain amount of money which is different.
So, if you look at period 1, nothing is flowing. Same with period 2. But between 1 and 2, there has been some flow.
Maybe Steve is thinking about how close something is to "cash", and that a Treasury bill is money taken out of circulation, while wallet cash is very much in circulation?
In this view, if the Government did not buy bonds, then all that money would just sit in checking accounts, and therefore leap out as people spend it.
It is this view that says the Government no longer issuing debt (thus leaving money in checking accounts) is inflationary.
Arthur:
ReplyDeleteI'm saying it's only "in circulation" in the instant that it is transferred to you as income, or from you (or more generally between accounts). The rest of the time it's in somebody's account.
(Related: the concept of "spending out of income" doesn't make sense except as a kind of useful vernacular shorthand. When somebody hands you a five-dollar bill, you can't spend out of that instantaneous transfer. It makes no sense. You can only spend out of the stock that you then have in your hand.)
An account is a stock. It's increased and decreased by (instantaneous) flows -- circulation. This is decidedly Godley-esque, isn't it?
So there's no way to really say that a certain quantity of money is "in circulation" at a given time -- or not in any meaningful way (how many electrons are in transit during this nanosecond [or picosecond]?).
What Winterspeak said. Except:
"In this view, if the Government did not buy bonds, then all that money would just sit in checking accounts, and therefore leap out as people spend it."
No:
1. It would only leap out if people *chose* to spend it (for instance if they were spurred to do so because inflation was causing it to lose buying power).
2. At (effectively) the same instant it leaps out, it leaps back in to somebody else's account.
So no:
"It is this view that says the Government no longer issuing debt (thus leaving money in checking accounts) is inflationary."
Money just sitting in checking accounts -- not circulating between them -- can be totally sterile, not affect the real economy or inflation. Same for money sitting in bonds, or any other financial storage vehicle.
In the simplest understanding, inflation is caused by demand (the sum of all those instantaneous transactions) banging against (potential) supply. Prices rise. If that spending doesn't happen -- the money doesn't circulate between accounts -- inflation doesn't happen. No matter how much liquid money is lying around.
So I guess I'm saying that it's all about real-economy transaction volume, unrelated to any measure of "liquid" stocks. (M0 or anything else.) People spend money not because they have it in liquid form, but because they want to spend it.
I've been wrestling with this for a while, so any other thoughts welcome.
@Steve Roth: I think I see exactly what you mean, and where there may be some confusion.
ReplyDelete"Money in circulation" sounds like the stock of money that is currently in circulation.
-- "In our little village, in total there is $100 in savings, locked in time deposits, and $50 in circulation"
-- "Yes, but I'm going to dis-save a bit soon and add another $5 to the money in circulation".
-- "Cool! That should really get spending going!"
I agree that this is confused. Hmm. The following would be better perhaps? (Using the term 'money circulation' in stead of 'money in circulation'):
-- "In our little village, in total there is a money supply of $150".
(These guys has a fixed money supply for some reason.)
-- "Yeah, I know. And last year, there was $100 never touched, locked in time deposits. Only $50 were used for transactions".
-- "Oh really, that little!? But that's pretty irrelevant though, isn't it? How big was the 'money circulation'?"
-- "During the whole year? Um, about $10.000. Lots of spending going on. (These ten $5-notes are getting pretty rugged)."
Hugo: 'zactly.
ReplyDeleteAdd:
"Since all us village people felt like we had plenty of money stocked away for our future needs (given our incomes, spending desires, and predictions of future same), we were ready to spend, wanted to buy $11,000 worth of food. We would have except the producers only produced $10,000 worth. Since people were lining up demanding to buy their food, they raised prices. Who wouldn't? This year we'll be able to turn over the $11K a year we're willing to spend."
Also: not just locked time deposits, other less liquid assets. Think how easy it is to convert stocks and bonds to "cash." Push of a button. True, everyone can't convert all of it at once, but lots can convert lots. (Say, if they're nervous about those assets declining in value.) The stock of financial assets ("money") is circa four times GDP. Stock and bond prices decline because of the selloff. But presumably all the money goes into more liquid assets/"cash", because every sale is a purchase.
The total quantity of "money" in the village doesn't decline. (Unless people *also* pay off debt in aggregate, the govt runs a surplus, and/or net imports is positive.)
(Does more money being in more-liquid form mean people in aggregate will spend more of it, more times per year? No. There may be *zero correlation*. In this case they're holding it not because they want to buy more -- QTC -- but because they're nervous and want the safety of storing it in a vehicle that won't decline in dollar value.)
So this supports the idea that the Fed converting bonds to cash or vice versa has no *direct* effect on people's spending proclivities -- they have the same amount of "money," whatever its form -- or hence on inflation. It only affects interest rates (which could affect people's proclivities, but only secondhand, a much weaker lever).
ReplyDeleteSteve, Hugo: I have been figuring things out for myself for a long time and it is certainly possible that I don't have every last thing right. However, I think you are making a distinction that serves no purpose other than to undermine my point of view.
ReplyDeleteSteve, you write: I'm saying it's only "in circulation" in the instant that it is transferred to you as income, or from you (or more generally between accounts). The rest of the time it's in somebody's account.
I think you are being too literal. The money that goes into my wallet on payday and is gone by the end of the week is "in circulation" I say. So is the money that goes into my checking account and is used to pay the bills by the end of the month -- or the end of the year, even, for those annual bills.
As Winterspeak said, "wallet cash is very much in circulation".
Normally I am all in favor of making distinctions. But the distinction you make between in circulation at the moment of transaction and is a distinction I cannot make.
Call it "spending money" if you prefer, instead of "money in circulation". The two are altogether equivalent in my view.
Make that:
ReplyDeleteNormally I am all in favor of making distinctions. But the distinction you make between in circulation at the moment of transaction and "in a place that I put money I plan to spend" is a distinction I cannot make.
"a distinction that serves no purpose other than to undermine my point of view."
ReplyDeleteNo, really. Truth told, your point of view (on this) hasn't had any effect on my thinking. (At least, yet...)
The reason I think the distinction is important is because the form one's "money" is in seems rather immaterial to whether or not one will spend it in a given year.
And the concept of a stock of money "in circulation" is widely associated (viz QTM) with the opposite view, a view that doesn't make sense to me.
If you have a bunch of water in glasses in your fridge, will you drink more water than you would with the water sitting in a pitcher, so you have to pour a glass before drinking?
I don't think so. If you want to spend and your money's in some other form than a checking deposit, you'll sell some bonds or stocks, then spend. It's one big cookie jar. (though only one kind of cookie can be traded for other things, its easy to convert one kind of cookie to another)
I've also always found financial advisors' distinction between principal and income to be nonsensical or at least misbegotten. One's a flow, the other's a stock. Once you've received income, it's part of your principal! Its history is immaterial; only the future matters.
IOW, it doesn't matter how that $5 bill ended up in your hand. (It's "sunk" income.) You've got it. Now: are you going to spend it, or hold on to it? Depends on your desires, on how many other $5 bills you have (and other things that can be converted into $5 bills), and your predictions for the future of the world -- including your predicted future receipts of $5 bills.
ReplyDeleteI think the "stock of money in circulation" = "quantity of liquid. tradeable financial assets" = "proclivity to spend" idea confuses people about these relationships. You don't spend *because* your money is stored in liquid form. Again viz QTM.
@The Arthurian
ReplyDeleteSorry, didn't mean to argue against you. Not even sure what the debate is about -- just happened to see one of Steve's comments.
There is probably a good and crisp definition of "money in circulation". M2 or something? And then the question is how relevant that measure is?
I believe MMTers emphasize the importance of spending (which is related to aggregate demand), rather than various money stock measures? Spending is a flow -- money per unit of time. (Right?)
And among the various financial stock measures, I think they emphasize Net Financial Assets (rather than M2 etc).
Regarding the spending flow -- the circular flow of production and income -- it is often talked about leakages and injections. Income (during some period) that has not been re-spent on consumption (during the same period) is a leakage.
I'm not sure whether such a leakage primarily should be thought of as a decrease in the spending flow (i.e the circular flow of production and income)? Or rather as a decrease in the stock of "money in circulation"?
Confusing stuff, this. (Actually, all of the above may be totally wrong. I'm tired.)
Ah. The issue was how changing debt levels affect spending. (Perhaps more of a Keen area than MMT?)
ReplyDeleteOk, I sort of get the intuition there. If Net Financial Assets don't increase as would have been needed (to support a growing economy), what happens then?
A "solution" would be increasing private debt levels instead (for a while, until it breaks down).
Loans create deposits. So new loans can create new liquidity as needed to support increased desired spending.
Again, not sure if we should primarily think of this as a new deposit being added to the stock of "money in circulation"?
Or if the new spending (from the loan) is rather an injection into the circular spending flow?
Hugo: There is probably a good and crisp definition of "money in circulation". M2 or something?
ReplyDeleteFRED says "M1 includes funds that are readily accessible for spending."
http://research.stlouisfed.org/fred2/series/M1SL
FRED says "M2 includes a broader set of financial assets held principally by households. M2 consists of M1 plus: (1) savings deposits ... (2) small-denomination time deposits ... and (3) balances in retail money market mutual funds (MMMFs)."
http://research.stlouisfed.org/fred2/series/M2SL
I reduce those good definitions to these crisp ones: M1 is spending money. M2 is M1 plus money in savings. Nonstandard? Perhaps. Easy to grasp? It works for me.
And as far as whether the definitions are relevant, my motto is Never challenge the Fed.
Good motto!
ReplyDelete