Or: It’s The Velocity, Stupid.
I got quite a bit of blowback on my recent post suggesting that economists don’t understand accounting.
In response I give you Exhibit A: the almost-ubiquitous notion that more saving increases the supply of “loanable funds” — hence that more saving causes or at least allows more investment. (The absolute classic fallacy of the S=I accounting identity.)
On casual consideration, it seems like it would be right, right? You spend less than your income, so you have more money (stuffed in your mattress?), and you can lend it out.
Or more likely: you “put money in the bank” — deposit more than you withdraw — so the bank has more money; it can lend more.
It’s A Wonderful Life.
Here’s Mankiw in his textbook, saying exactly that:
Saving is the supply of loanable funds — households lend their saving to investors or deposit their saving in a bank that then loans the funds out.
But:
1. A little careful consideration shows that this casual consideration is logically incoherent — just plain wrong, by accounting identity.
And:
2. Economists are not supposed to be thinking, giving their sage advice, or corrupting our youth based on casual consideration.
Think about it:
You get $100,000 in wages. Your employers’ bank account is debited, and yours is credited. Your bank can lend against your higher balance; your employer’s bank can’t. Net zero.*
You spend $75,000. It’s transferred from your account to other people’s/businesses’ bank accounts. Their banks can lend more, yours can lend less.
Is the total stock of loanable funds affected by whether the money is on deposit at your bank, your employer’s bank, or the banks of people you bought stuff from? No.
Meantime, you don’t spend $25,000. You “save” it. The money sits there in your checking account. If the action of spending — transferring money from one account to another — doesn’t change the total stock, how could not transferring money do so? Your bank still has the money, which it can lend out. Other banks still don’t, and can’t.
It may help to think about this as if there was only one bank. (Which is not so far off. Bank deposits all consolidate back to accounts at the Fed.) Every person and business has an account. All the spending/transfers (or non-transfers, a.k.a. “saving”) just shift deposits between accounts, with no change in the (single) bank’s total deposits.
So the saving/spending mix has no effect on the stock of loanable funds. Shifting (or not shifting) those stocks around has no aggregate effect on the total stock.
But what about the flow — new loans from banks? Again: no.
Here’s a behavioral, rather than accounting-based assertion — not a controversial one, I think: In any period, banks in aggregate lend more — “print” more new money and deposit it in people’s/businesses’ accounts — because they think they can make money doing it at current interest rates. They think that for one primary reason: they are confidently optimistic about future prosperity — borrowers’ future income streams. If they’re less confidently optimistic they lend less, or ask for higher interest rates — which has the same effect: less lending.
Likewise borrowers: they borrow because they think future conditions will be good, and they’ll be able to service their loans at the asking rate out of strong income streams (and/including rising financial asset values).
Likewise spenders: they spend (that new) money because they think it will yield good returns from investment, and/or because they think they can consume today and be able to earn more money to pay for it (repay the loans) in the future.
So how does the saving/spending mix affect those expectations? Another behavioral assertion: Those expectations are set, to a great degree, by current conditions, because they’re the best predictor we’ve got. It’s difficult at best to predict future “shocks” that will change those conditions. Or as the Eight Ball says: “The future is … unclear.” Life is uncertain.
So how does a higher proportion of saving to spending affect current conditions?
It makes them worse. GDP is spending. Less spending (as a proportion of either income or wealth) means less economic activity. Less velocity. Less transactions. Less surplus from trade. Lower GDP. People, businesses, and banks, borrowers and lenders, are less prosperous, and less optimistic. So banks lend less, borrowers borrow less, and (in a potential downward spiral) spenders spend less.
Takeaway: An increased saving/spending proportion has no effect on the stock of loanable funds (it can’t), and it has only a second-order, expectation-driven behavioral effect on flows — it decreases them.
You really have to wonder sometimes where economists get this stuff that they put in their textbooks.
Nick Rowe attempted to save this conceptual situation recently in a comment posted hereabouts (emphasis mine):
Suppose there’s an increased demand for financial assets by households (a rightward shift in the demand curve). Will that increased demand lead to an increased quantity of investment by firms and an increased quantity of financial assets sold to households (a movement along a supply curve)? It may do. That depends on the model. It’s a behavioural question, not an accounting question.
His questions in the middle, and the last statement, are completely on the money. But his explanation begins right in the midst of the conceptual confusion, putting the modeling cart before the behavioral horse. The behavior doesn’t “depend on the model”; the model’s accuracy and usefulness depends on its assumed human response to incentives and constraints.
Or perhaps, rather, he’s climbed aboard the wrong behavioral horse — one that is wandering off rather aimlessly.
The “desire to save” is a conceptual representation, a mini-model, if you will, of one aspect of the economic situation. I’m suggesting that that construct is outside of, peripheral and irrelevant to, the behavioral chain of cause and effect.
People might want to save more/spend less in aggregate for various reasons:
• Times are tough — GDP and employment are weak — and they’re worried about future ability to consumption.
• Times are good, and they’re satisfying all their consumption desires.
• Rich people have a larger proportion of income and wealth, and their lower marginal propensity to consume drags down aggregate spending, relative to income and wealth.
Or some other scenario. (As Keynes said — not looking up the exact quote here — all economic activity is driven by the desire to consume.)
In the second scenario banks will want to lend more — but not because people and businesses (want to) save more. If that were true, banks would also want to lend more in the first scenario — which is completely contrary to what actually happens. (The results in the third scenario seem uncertain.)
Here’s a syllogism to make this widespread confusion clearer:
More investment results in more prosperity.
More saving results in more investment.
More pessimism (or less-confident optimism) results in more saving. (I think monetarists will stipulate to this.)
So more pessimism results in more prosperity!
Mankiw’s conceptual confusion is inevitable, and arises from two causes:
1. He’s starting with snaky (and conceptually confused) assumptions about the sources of human behavior, and:
2. New but related subject: He’s trying to think about flows (and get tender young minds to think about flows) using static, of-an-instant models like the standard S/D and IS-LM diagrams. The problem, when you’re trying to think about “supply,” is that a flow can’t exist in an instant (only stocks can); it’s a meaningless, impossible concept. And since stocks in our discussion here are unaffected by saving, he’s in a pickle, cause it’s all about flows. (And no: “comparative-static” methods don’t solve the conceptual confusion; they arguably only contribute to it because they impart the illusion of time and dynamism.)
The only way (that I know of) to model “flow supply” in a conceptually coherent way — or even think or talk about it really, which is mental modeling — is using a dynamic simulation model. Of late I’ve been quite taken with the power grid as a good metaphor for a dynamic model of the economy — one that I’ll expand and expound upon in a future post.
For now I’ll leave you with this: Clower/Burshaw on the difference between “stock supply” and “flow supply,” or peruse the literature here. Nick also talks quite a bit about the largely forgotten old 70s notion of “nominal” (roughly: “potential”) supply and demand — though mainly regarding money, not real goods.
I almost never see any consideration of these seemingly crucial concepts in economic discussions — much less cogent analysis, or incorporation of said concepts.
Which leads me to ask a question of economists:
Are you sure that you’re perfectly clear on what you mean when you use the words “supply” and “demand”?
Are we?
* I won’t even touch here on the widespread misconception among economists regarding bank lending, except to say that in practice bank lending is not constrained by deposits — banks lend most of their deposits then lend (much) more based on their excess capital (times X) — which, thus, is their effective constraint on lending. Not deposits.
Cross-posted at Angry Bear.
Comments
127 responses to “No: Saving Does Not Increase the Supply of Loanable Funds”
Steve,
I feel like I tend to defend mainstream economics when I’m commenting here, so, staying in that role, I’ll take a stab at defending Mankiw as well.
In fact, I’ve seen this sort of criticism of Mankiw before–some time ago, Bill Mitchell wrote a blog post in which he totally misconstrued a similar passage from the textbook.
When thinking about the loanable funds model, it can help to replace “funds” with “resources”. Put like that it’s very clear that saving does increase the supply of loanable “funds” (i.e. resources).
Think of the most simple single sector closed economy model you can. In any time period there is a basic choice facing society between consuming resources and investing them. In this model a decision not to consume (i.e. to save resources) is necessarily a decision to invest, and a decision to invest is necessarily a decision not to consume.
Hence, not consuming resources really does increase the total available to invest, and consuming resources really does decrease the total available to invest. That’s just how life is on Planet 3.
The desire to invest implies a desire to postpone consumption, but the reverse is not true; you can postpone consumption without investing by holding safe financial assets or paying down debt.
The desire to postpone consumption, without the desire to take risk, can only “fund” government spending, not private spending.
In the case of government guaranteed bank deposits, it’s like buying government bonds for the depositor, but like selling corporate bonds for the bank. The difference in yield is an unearned profit for the bank (does Mankiw mention this in his textbook? :-))
@vimothy
Yes, but no. Definitions of “save.”
We can “economist save” (increasing national savings) by creating more real assets, and consuming less of them. (Saving does not equal investment; saving is investment. And vice versa.)
We collectively save in a given period by producing assets that we don’t consume in that period. (By definition, these assets are investment goods — structures, hardware (equipment), software, and inventory. If they were consumed within the period, they’d be consumption goods.)
But that has nothing to do with individual’s and businesses’ decisions (even in aggregate) to “save” or spend. When you spend less, you’re keeping the “savings” in your bank account. When you spend more you’re transferring them to firms’ bank accounts. Individuals own firms, so…
Investment is economist-/national-saving. Not-spending (vernacular saving) does not do investment — the exact opposite in fact.
Very nicely put.
I would only defend the system here by saying that in return for those insurance payments, and regulations/restrictions on their activities, banks get a license to print money (with regulations and restrictions on that license.) Generally a (at least potentially) balanced deal.
But government bonds in general are another story, perhaps. Govt pays interest on risk-free holdings. Yes, that’s what the market wants/demands, but is govt obligated to provide it? Could just print dollar bills instead of t bills…
In return for distorting the market by giving free income to banks, government gets the ability to manage the market via OMOs.
Is it worth it? I dunno.
SR: In return for distorting the market by giving free income to banks, government gets the ability to manage the market via OMOs. Is it worth it? I dunno.
As Warren Mosler first pointed out in “Soft Currency Economics” (1994), the operational purpose of tsy security issuance under a non-convertible floating rate system (“soft currency”) is to drain excess reserve introduced by govt fiscal deficits. Tsy issuance is an interest rate management instrument.
However, if the cb pays interest on reserves equal to or greater than the target rate, then tsy issuance is not operationally necessary. Actually, the Fed is paying IOR now, obviating the operational need for tsy issuance.
Even though operationally unnecessary under the present monetary system, Tsy issuance is mandated politically, in that the Treasury cannot run an overdraft at the Fed.
What this means is that a political decision has been taken to provide a subsidy in form of interest on a default risk-free credit instrument.
Then the question becomes whether this subsidy is justified by some overarching public purpose that counterbalances the inefficiency incurred. Based on this some MMT economists, but not all, have called for ending this extraordinary subsidy for savers, which also creates fiscal drag, adding to the deficit reducing fiscal space for other more contributory uses, such as lowering taxes or increasing public investment.
@Tom Hickey “What this means is that a political decision has been taken to provide a subsidy in form of interest on a default risk-free credit instrument.”
But isn’t paying IORs equally a subsidy?
Yes, we’re doing both now, but (haven’t thought this through) absent bonds, would IORs (have to) be higher?
But isn’t paying IORs equally a subsidy?
If the cb wishes to set a target rate above zero, then govt is providing the funding for the ability to set rates, so it more of an operational expense than a subsidy to savers.
The question is whether this expense for the ability to set the rate above zero justifiable. And is it appropriate for a smal group of unelected and unaccountable technocrats to control the monetary policy of a country. Some see this as anti-democratic political and a command system economically.
Therefore some MMT economists, Warren Mosler for one, recommends that the cb set the overnight rate to zero, which is “the natural rate” in Mosler’s terms, and leave the adjustments to the economy to fiscal policy, which is arrived at democratically by temporary office-holders who are periodically accountable to the public at the voting booth.
MMT economists argue that fiscal policy is economically superior to monetary policy anyway, since it can be tightly targeted whereas monetary policy is blunt instrument in which there are inevitably winner and losers, i.e, savers and borrowers. Why should the cb be using monetary policy to target inflation using a buffer stock of unemployed as a policy tool, which is what it dones now. Why should the cb favor savers or borrowers in controlling for inflation when monetary policy is rather inefficient and ineffective in comparison with the sectoral balance approach and functional finance.
Steve,
I dunno, I think it may be simpler than you’re making it out to be.
Economists don’t think that all saving is national or that saving is investment. They just think that, for the whole economy, forgetting about foreign saving, saving is equal to investment.
Now, there’s no particular need to take a postition on whether increased saving *causes* increased investment. It might, of course, or it might affect the trade balance, say. Or it might be the case that the causality is working the other direction. In practice, it seems to me that it’s more likely that the causality is running in *both* directions.
Investment does need to be financed, and financed by savings. You can’t pull that finance out of thin air just because of some institutional feature of the banking system. If domestic saving is less than domestic investment, then there must be a net inflow of foreign capital. In other words, domestic investment must be financed by some combination of domestic and foreign saving.
If a bank makes a loan to a business to finance investment, then that’s a use of loanable funds (ultimate lender–households). If a business sells a bond to finance investment, then that’s another use of loanable funds (same lender). If a business makes the decision to finance out of retained earnings, that’s another use of loanable funds–it’s own loanable funds, which would have otherwise accrued to the household sector as profit (so, same lender again).
@vimothy
“You can’t pull that finance out of thin air just because of some institutional feature of the banking system.”
If that institutional feature is the license to print money, it seems like you can “pull finance out of thin air” — pretty much by definition.
“Investment does need to be financed, and financed by savings.”
As I understand things, no. It can be financed by new money printing — new net bank lending.
If the new money is used for investment, it *creates* economists’ savings — unconsumed real assets.
Again, simplified, stylized:
Banks (always) lend about 90% of their deposits, keeping the rest in reserve so transactions clear. You could call that 90% the “stock” of loanable funds. It’s unaffected in aggregate by which bank accounts it’s sitting in, whether deposits shift between accounts.
Banks also, in addition, lend some multiple of their capital. That’s money printing, and it’s the flow of new loans. Money turnover between accounts (saving/spending proportion) has no *direct* effect on banks’ willingness to make those new loans.
If people save and *also* pay down loans, that could increase banks’ willingness. But that’s a different proportion — the percent of savings held vs. used to pay off loans.
That ratio would seem to have a direct effect on banks’ willingness to lend.
Steve,
When a bank creates a consumer loan, for example, the banking sector increases its assets and its liabilties, so it’s just a wash. Increasing savings is not equivalent to bank sector credit growth. In particular, I don’t think that you should call bank deposits, “the stock of loanable funds”. Call the stock of loanable funds the stock of loanable funds and bank depostis bank deposits. Much less confusing that way!
If a bank creates a loan, which is then used to finance a new business investment, then either the domestic public end up holding the asset via the intermediation of the banking sector (so that S = I), or foreigners do (so that S + F = I, where F is foreign capital). There is no third option in which investment is financed by something other than savings.
@vimothy
“When a bank creates a consumer loan, for example, the banking sector increases its assets and its liabilties, so it’s just a wash.”
Yeah and you can say the same about the household sector. But it has more money to spend today.
“Increasing savings is not equivalent to bank sector credit growth.”
Not sure what you mean by “equivalent,” but bank sector credit growth — to the extent that the new money lent/printed is used for investment spending, purchasing real goods that are not consumed in the period — increases savings. (Because savings is investment. And vice versa.)
“Call the stock of loanable funds the stock of loanable funds”
Define “stock of loanable funds” for me.
I’m not at all sure such a thing exists, in any meaningful or useful sense. There’s:
o Bank deposits, which in actuality are always all lent out, so are no longer “loanable,” and
o Banks’ licenses/ability/willingness to lend more than that against their capital (times X), which seems hard to characterize as a “stock of loanable funds.”
??
Steve,
Stock of loanable funds := savings.
In my own opinion, the problem with this stuff is not that doesn’t make sense, but rather that it’s complicated. In fact it does make sense, it just takes a while to get an intuitive feel for how it all fits together. A couple of years ago, my thinking was exactly in line with where yours is now. (I hope that doesn’t sound condescending–if anything, it’s a reflection of my own limitations that it’s taken me so long to come to any sort of putative understanding).
When a bank makes a loan–in any context–the assets and liabilities of the banking system increase by the same amount, which means that the assets and the liabilities of the non-bank sector (i.e., the consolidated sector of everyone who is not a bank) also increase by the same amount, since the non-bank sector owns the bank sector.
The simple fact of bank lending cannot by itself increase saving, since bank lending represents simultaneous saving–asset is acquired by owners of bank–, and dissaving–liability is acquired by customers of bank. On aggregate the dissaving costumers of the bank are its saving owners, so the whole thing is a wash.
In order for the whole thing not to wash out, the borrowers need to take that loan and use it to finance new investment. But say instead that they buy second hand capital equipment. Then, there is another asset swap with no change to net assets, no new aggregate saving, and no new investment.
If you think about it, it’s obvious from the nature of financial assets and double entry bookkeeping why it’s impossible for society taken as a whole to increase its savings or net wealth simply by lending money to itself.
Hmm, I think I sound like a bit of an ass there. Obviously, I could be completely wrong about all of this. 🙂
@vimothy
“Stock of loanable funds := savings.”
Okay then define “savings.” As a stock. This is not a measure/term that exists in the national accounts.
Are you talking about monetary savings (maybe net financial assets?), or the stock of fixed assets?
In the singular, saving (S, the flow) is in fixed assets and inventory. We as a nation don’t have more money because of investment/saving (same thing); we have more unconsumed real assets.
“it’s impossible for society taken as a whole to increase its savings or net wealth simply by lending money to itself”
You just shifted terms: “net wealth” (define please) vs. “savings.”
Assume for a moment:
1. All net lending is spent on investment.
2. Consumption quantity remains the same.
Period A: Net lending positive (more lent than retired). More saving/investing than there would have been absent the lending. If more is spent on investment, saving must be greater. S=I.
Period B: Net lending negative (more retired than lent). Less saving/investing than there would have been otherwise.
I think this is what you mean. The loans have to be paid off in some future period, so net zero.
The proportion of investment/saving to consumption is different in the two periods. There’s more “saving” in Period A (and more Y, because I+C=Y).
But that’s not because people chose to save more/spend less. It’s because they chose to borrow more in Period A and spend it on investment (aka saving).
In Period B they chose to borrow less, and spent less on investment. Less saving. Same amount of consumption. Less Y.
@Asymptosis
IOW, people’s desire to borrow/lend/payback/call loans is not determined by people’s desire to save, because that desire to save can have many possible causes. (See my three bulleted scenarios above.)
Those causes can *also* affect the borrowing/lending inclination, but the direction of the effect can be opposite to that of the saving inclination.
Individuals/businesses can desire and achieve less saving (more consumption spending), and at the same time desire (and achieve) more investment/economist-saving.
They can achieve this by borrowing, and using the new money for for investment spending.
Seems to me if we think of money “spent” into the economy (government spending) and money “lent” into the economy (through bank lending) as separate entities (which they are) it’s pretty easy to keep track of. Money “spent” into the economy is permanenent (until taxed), money “lent” into the economy is temporary (exists until paid back). For the moment I am ignoring the external account.
These monies co-mingle and can’t be differentiated at the micro level (a dollar is a dollar) but credit dollars always carry an attached liability, which can be observed at the aggregate or macro level. If these liabilities are written off as losses at the Fed level (jubilee) the dollars would become NFA dollars.
this seems obvious and I only mention it because I can’t quite put my finger on what the point actually being discussed here is. The discussion seems peripheral to the accounting that matters.
Or maybe I’m just dense.
Plus, why doesn’t the stock of loanable funds fall somewhere between a banks reserves and infinity?
Steve,
Aggregate savings and net wealth (people talk about “national net wealth”) are equivalent terms–two ways of representing the same phenomenon. For an individual, to “save” is to increase his or her net worth. The same is also true at a sectoral and aggregate level, so that, for example, for the household sector to save is for it to increase its net worth / net assets / savings. (In any period t, the flow of saving is equal to the change in the stock of saving, or S(t) – S(t-1)).
In your model, you seem to have investment driving saving in terms of causality, which could be right or wrong. But why assume from the outset that one is driving the other? It seems more likely that they are both endogenous. (And I think you could say the same for output and investment).
It’s follows from the identity of saving and investment that more investment equals more saving. (And also, of course, that more saving equals more investment).
But it’s also the case that if there was no “market” for loanable funds where whatever agent in the economy has income that they don’t want to consume can lend to investors who want to finance capital formation, then there would be no investment.
If there is no market, there’s no transaction and there’s nothing to observe. That would be “out of equilibrium” behaviour. Given that investment takes place, there is always a flow of saving–though not necessarily domestic saving. But that doesn’t mean that one causes the other. A market needs two sides for trade to take place.
“Aggregate savings and net wealth (people talk about “national net wealthâ€) are equivalent terms”…
No, they aren’t. Aggregate savings means cash or cash equivalents. “National net wealth” is a nebulous term that includes cash and cash equivalents but beyond that could mean anything, depending on how someone values their real assets. Unless you count real estate and stuff like that as “savings”.
Vimothy,
“The simple fact of bank lending cannot by itself increase saving, since bank lending represents simultaneous saving–asset is acquired by owners of bank–, and dissaving–liability is acquired by customers of bank. On aggregate the dissaving costumers of the bank are its saving owners, so the whole thing is a wash.”
This representation is incomplete. Both the bank and the customer acquire an asset and liability; the bank acquires a loan (asset) and deposit (liability), whereas the customer acquires a deposit (asset) and debt (liability). It’s not split between the two. It’s a wash from the perspective of net financial assets for the private sector, but a net new deposit is in fact created. Why isn’t this considered “loanable funds?” I can use that new deposit to buy a treasury bond or to invest in a new enterprise or my business etc etc. And the ability for banks to create those loans is only limited by capital requirements; in a theoretical model we could assume those away, and at any point in time, banks would have unlimited ability to loan. This gives a very different flavor than loanable funds, which suggests a fixed pool of funds to do things with.
“If you think about it, it’s obvious from the nature of financial assets and double entry bookkeeping why it’s impossible for society taken as a whole to increase its savings or net wealth simply by lending money to itself.”
Correct, except when the govt deficit spends (and if we are defining society as the non govt sector).
@vimothy “you seem to have investment driving saving in terms of causality”
No! The production of lasting real goods is the ultimate source of cap-S economist-saving. Net new lending spurs — incentivizes, purchases — more production of those assets. Net retirements discourages it (withdraws money for purchases).
DisSaving in the economist sense is consumption of long-term goods — of inventory (produced circa last year), and of fixed assets (produced in all preceding years) through use, etc.
See Net National Product: GDP – consumption of fixed capital.
“It’s follows from the identity of saving and investment that more investment equals more saving. (And also, of course, that more saving equals more investment).”
Can’t resist wisecracking: that is a rather tautologically tautological statement. 😉
I really think it helps to think, instead of S equals I, S is I.
Or: S in the NIPAs does not represent an actual flow (I does); it’s an accounting construct, in JKH’s words a “temporary accounting repository†that’s necessary to resolve sources and uses of funds — basically to reconcile the NIPAs with the Fed FOFs.
Paulie46’s reply shows the problem with using vernacular terms that aren’t clearly defined. Simon Kuznets said that the “true wealth of the nation” is the stock of (unconsumed) fixed assets.
I would add that there is a huge stock of unaccounted real assets that aren’t “fixed” — knowledge, skills, ideas, organizational capital, etc., not to mention the population’s ability to work in the future. So health. Also, social capital like mutual trust.
It’s a very open question whether or not those intangible and unmeasured but very real assets are (accurately) represented by the market value of financial assets at any given moment, which value significantly exceeds the value of fixed assets.
So “national wealth” can be a very tricky concept.
@wh10
“This gives a very different flavor than loanable funds, which suggests a fixed pool of funds to do things with.”
Exactly. It’s hard to characterize the banks’ willingness/ability to lend (theoretically, absent capital constraints, in unlimited amounts) as a “stock.”
wh10,
Why isn’t this considered “loanable funds?â€
This is a good question. The reason is that we are not interested in all the borrowing and lending going on in the economy. We are interested in a macroeconomic model of national saving and aggregate investment. Given that S = I, why should S = I take on any one particular value, as opposed to another?
@vimothy
“Given that S = I, why should S = I take on any one particular value, as opposed to another?”
Right again. Why do people invest?
*Not* because they have a desire to spend less (save more).
Because investment *is* spending — on fixed assets.
@vimothy
The wise crack is that’s why neoclassicals failed to see the financial crisis coming ;).
But more seriously, any activity you might deem “I” can be funded from a bank loan. But bank loans are not restricted by fixed loanable funds.
Alternatively, a statement like Mankiw’s (Saving is the supply of loanable funds — households lend their saving to investors or deposit their saving in a bank that then loans the funds out) suggests there is fixed pool, because in his model, lent money comes from other people’s saving. It’s a wash, like you say. But that model is incorrect.
@wh10
And in that statement, it’s pretty clear Mankiw isn’t even referring to real resources. He is quite clearly suggesting that he’s talking about deposits – nominal financial assets – as comprising loanable funds. And once you admit deposits comprise loanable funds, then the concept of ‘loanable funds’ breaks down, because there is no theoretical limit to deposits in a monetary system like the US’s.
@wh10
To be clear, I agree, it IS a wash from the perspective of nominal financial assets, but it’s not a wash in terms of spendable deposits.
BTW does it matter if we clarify whether or not the economy is at full capacity?
For what I was trying to put across, at least, no. The quantity of lending is not a function of “available funds.” It’s a function of current conditions and resultant expectations.
Steve,
For the whole economy, in a closed economy model, the stock of savings is equal to the capital stock. In open economy model, the stock savings is domestic capital plus or minus net foreign assets.
For the whole economy, then, “dissaving” is depreciation of capital and sale of assets to the foreign sector.
That S = I is an accounting tautology. It doesn’t imply a particular causal relationship. We need to use economic theory to establish causality. S is not the same as I. If S were the same as I, then S would always equal I. However, S does not always equal I. Therefore, S cannot be the same as I.
I would add that there is a huge stock of unaccounted real assets that aren’t “fixed†— knowledge, skills, ideas, organizational capital, etc…. So “national wealth†can be a very tricky concept.
All good points. People do try to value things like human and intangible capital though when determining national net wealth. You might like to read this paper, “The U.S. Balance Sheet: What Is It and What Does It Tell Us?”, Keith Carlson, 1991, which has some details:
http://research.stlouisfed.org/publications/review/91/09/Balance_Sep_Oct1991.pdf
@vimothy “S does not always equal I.”
Before I reply, I want to be clear what you mean by this. It doesn’t?
“S = I is an accounting tautology.”
I agree completely! Because S is I, saying that they’re equal is a tautology.
On the open/closed thing, I totally get it. We’re talking about the difference between “national” saving(s) and “domestic” saving(s). Each is what it is.
Probably a lot of the confusion over S and I is due to there not being a convenient identity sign on the standard keyboard, so the equal sign is used instead. The accounting identity states that S is identical to I, not that S equals I. Some programming languages use “==” for identity and “=” for equality, for instance. Programmers know they they need to keep this straight. People using S and I don’t always pay careful attention to it, and that can result in ambiguity and confusion.
Identity states that two signs signify the same thing, e. g., the morning star and the evening star both signify Venus. Equality states that the value of two expressions is the same. One could say that S is identical to I is similar to the morning star and evening star in that S is the financial symbol from the household side for the same actual that I is the financial symbol from the firm side.
The identity just tells us that two variables take on identically equal values (where we’re assuming autarky).
The market for loanable funds is a model of S = I. In this model you have the supply side, S, and the demand side, I. Given that S = I, what does S = I equal, and why?
@Asymptosis
Before I reply, I want to be clear what you mean by this. It doesn’t?
No. In our discussion so far, we’ve dropped a term from the identity, implicitly assuming a closed economy or a balanced flow of lending to the outside world. If there is no change to net foreign assets during the period, then S will equal I. Otherwise, the country will be a net “importer” or “exporter” of savings.
I agree completely! Because S is I
S is not I. In autarky, S is identically equal to I and national saving is always equal to aggregate investment.
“In autarky, S is identically equal to I and national saving is always equal to aggregate investment.”
How do you define “aggregate investment”?. I know what national savings is (it is quantifiable) but I’m curious how it relates to “aggregate investment”.
Paulie,
Aggregate investment is the sum of public and private investment (where “investment” is economics-speak for production of new capital).
@vimothy
Let me put in another way. “equals” states a relationship that is not true necessarily but contingently. This is how scientific hypotheses expressed as mathematical equations work. No amount of looking at equations alone can determine whether they are true when interpreted semantically to formulate a hypothesis. Their truth is said to be synthetic or empirical.
“is identical to” means that the expression is necessarily true in that it is tautologous. If one understands the meaning of the symbols one can determine that the expression is true analytically without looking beyond the symbols.
@Tom Hickey
Right–it’s just a tautology. We can define national saving such that its value is always equal to investment.
You can’t can’t get, “if we increase investment, saving will increase” from a tautology. After the fact, S = I (or S + CA = I) by definition, but we need a model to think about what will actually happen when people try to increase their investment, and whether or not this will lead to increased saving.
I’m not sure what’s going on here with all this S and I business, but it’s clear as day that Mankiw is talking about loanable funds as financial assets (see my comments 27-30), which causes the loanable funds theory to self-destruct.
Here’s a particularly pertinent passage from a recently published working paper (http://www.peri.umass.edu/fileadmin/pdf/working_papers/working_papers_251-300/WP279.pdf)
“We find the French-Italian PK circuit approach particularly useful for driving home the point that the finance for spending must come from somewhere. (Graziani 1990) Most recognize that to finance a purchase one needs to use income, to sell an asset, or to borrow. At the individual level that is certainly true. Yet, the “finance†that comes from income flows as well as the receipts from sales of assets also must come from somewhere–and an “infinite regress†is not logically compelling. The typical neoclassical deus ex machina source of finance is saving–but if saving is in financial form it must have been generated by someone else’s spending, another infinite regress. Hence, when the circuitiste begins with a bank loan to finance purchase of commodities (to be used to produce commodities) all logical problems are resolved.
Spending and creation of “money†in the form of a bank deposit are linked. It is best to think of these as balance sheet entries: the bank accepts the IOU of the borrower and credits her demand deposit; theborrower’s IOU is offset by the credit to her deposit. Spending then simply shifts the demand deposit to a seller. Money is created “endogenously†to finance spending. Later, when loans are repaid, the demand deposit as well as the borrower’s IOU are debited–money is destroyed. There is no magic involved, no “manna from heavenâ€, no separation of the “real†(say, IS curve) from the “monetary†(LM curve). As Clower (1965) would remark, money buys goods and goods buy money but goods do not buy goods. Barter is ruled out as one must first obtain money–from income flows, asset sales, or
borrowing–before spending. And the money must get created with an initiating purchase.”
wh10,
Mankiw is emphatically not referring to financial assets, he is not referring to money and he is not referring to deposits. He is referring to saving and investment. Aggregate savings are not increased simply because a bank makes a loan, because the loan represents simultaneous saving and dissaving that nets to zero when you aggregate over the whole economy (assuming no foreign sector).
(Financial assets only contribute to net assets (i.e. savings) if the liability is held by the foreign sector.)
The elasticity of loan supply is thus irrelevant, as far as I can see.
@vimothy
Vimothy,
“Saving is the supply of loanable funds — households lend their saving to investors or deposit their saving in a bank that then loans the funds out.”
He says ‘deposit their saving in a bank.’ He is EMPHATICALLY referring to financial assets. He even says the word “deposit.”
Also, can you clarify your use of the words ‘saving’ and ‘dissaving?’ I like asset and liability better. I don’t like dissaving, because it implies someone else has to forgo spending in order for the loan to be made, which is NOT what happens when a bank makes a loan.
@wh10
Sorry, mis-typed there. Let me restate, based upon your initial use of ‘saving’ and ‘dissaving’:
“The simple fact of bank lending cannot by itself increase saving, since bank lending represents simultaneous saving–asset is acquired by owners of bank–, and dissaving–liability is acquired by customers of bank. On aggregate the dissaving costumers of the bank are its saving owners, so the whole thing is a wash.”
As I said, above, your use of the terms ‘save’ and ‘dissave’ here make no sense. The same thing happens for both the bank and the borrower. Both the bank and the customer acquire an asset and liability; the bank acquires a loan (asset) and deposit (liability), whereas the customer acquires a deposit (asset) and debt (liability). It’s not split between the two.
Thus, besides the fact that I don’t like the use of the word ‘save’ since no one has to save income in order for a loan to be made, I see no reason why one or the other party in a loan transaction is the ‘saver’ while the other is the ‘dissaver.’ They both acquire an asset and liability.
@Tom:
Thanks for this. The nature of identity helps us think about what these accounting “things” are and how we can use them to think clearly about economics. Saving in particular is not like Income, Consumption, and Investment, all of which are constructs representing actual flows in the economy. Saving represents a non-flow, and in that sense is a more artificial construct, one that is useful in thinking about sources and uses of Funds. Since we have the perfectly good I already, which is identical with it, S is only useful in that accounting context — not in the context of money flows going into some undefined stock called “loanable funds.”
@wh10: Great quote. Fullwiler, Kelton, Wray, Clower, at least, seem to agree with me on the circular nature of money, hence the impossibility of “saving” (not spending) somehow increasing the stock of “loanable funds.”
This all relates back to the Cambridge Capital Controversy, of course, and the Limey’s successful demonstration that the neoclassical error is one of composition. One person can have more money due to saving; people in aggregate cannot.
@vimothy “You can’t can’t get, “if we increase investment, saving will increase”
Right. Ditto the reverse.
@wh10
BTW this conversation is rather amusing because here we have a case of someone defending neoclassicals using the concept of aggregate savings, which MMTers get criticized about using all the time.
Vimothy, you said: “(Financial assets only contribute to net assets (i.e. savings) if the liability is held by the foreign sector.)â€
You’re forgetting about the government sector. Govt spending ALSO increases private domestic sector net financial assets. It is the govt that then holds the liability. (This is MMT thing I was referring to above.)
Vimothy, you also said: “In order for the whole thing not to wash out, the borrowers need to take that loan and use it to finance new investment. But say instead that they buy second hand capital equipment. Then, there is another asset swap with no change to net assets, no new aggregate saving, and no new investment.â€
Vimothy, it seems you are confusing a lot of stuff here and being very ambiguous about your use of the terms ‘wash out.’ Let’s be very clear, double-entry accounting always holds, regardless of whether the investment is new or not. Everything always ‘washes out’ from that perspective. Getting to the point, why does it have to be second hand equipment? Why can’t it finance new investment? Of course it can! There’s your new investment. Assets and liabilities still balance on the books, but ‘new investment’ still occurs.
Steve, is there a way we can use S = I + (S-I) here? It seems bank loans can increase “I,†whereas, thinking about sector balances, net financial asset inflows from the govt or foreign sector can increase (S-I). According to Vimothy, what Mankiw really means is that loanable funds = S. But that really doesn’t make a difference, since we know that there isn’t a ‘stock limit’ to I or (S-I). But this is why I asked if it matters whether or not the economy is at full capacity, because at that point, govt spending or new loans aren’t going to finance the creation of new real resources. Steve, in response to that you said, “For what I was trying to put across, at least, no. The quantity of lending is not a function of “available funds.†It’s a function of current conditions and resultant expectations.” And I guess that makes sense, since ‘full capacity’ will feed into your function of ‘current conditions and resultant expectations.’
@vimothy
You (and Mankiw) are confusing monetary with real, individual with national.
Individuals save in the form of financial assets (some of which are *claims* on fixed assets — stock certificates, ownerships deeds, etc.).
The nation saves in the form of unconsumed fixed assets (actually, real assets, but only fixed assets are counted in the NIPAs).
This is the fundamental error of composition that the Cambridge Capital Controversy demonstrated in spades — a controversy that the Limeys won in a knockout but that neoclassicals refuse to acknowledge.
Again: S does not represent an actual flow (it can’t, because it’s nonspending by households). It’s a temporary accounting receptacle — a conceptual construct — used to think about and account for flows of funds.
It’s the artificial conceptual device by which the gap between real (NIPAs) and monetary (FOFs) is bridged.
@wh10 “no one has to save income in order for a loan to be made”
I think this cuts pretty much to the crux.
If we were talking about wheat, it would be different — lending requires prior saving. But for money — at least once banks have lent all (90% of) their deposits (which in practice they always have), then the only constraint on lending is banks’ willingness and ability within the constraints of the regulatory structure and available capital. That willingness, and (related) the availability of capital, are constrained by the interaction of expectations and interest rates.
Which makes me wonder: is it useful to think about the “supply” of available bank capital? Or does that notion suffer from the same conceptual falsities as the supply of “loanable funds”? I think the latter.
@Asymptosis
Steve, I’ve been meaning to critique your comment that ‘banks lend deposits.’ Given your knowledge of all of this, I would be surprised if you weren’t aware that MMTers would not agree with the way you state this, so I am curious why you choose to state it that way. From an accounting perspective, the act of lending never involves the lending out of deposits. It’s always just an issue of regulatory requirements.
Good question in your second paragraph. First, for it to be a question, it seems we have to assume there are capital requirements. At the macro level, we know NFA = equity = capital can only come from NFA inflows from the govt or foreign sector. We know govt spending isn’t limited by loanable funds. If govt spending is the source of bank capital, then it seems loanable funds would also not apply to bank capital. But can bank capital also be sourced from private sector bank loan creation, which doesn’t create NFA=equity=capital for the private sector as a whole?
wh10,
Mankiw is not referring to financial assets. He’s referring to the flow of fixed capital formation, i.e., “investment”, i.e. all the new factories and capital equipment that has been built in the period.
In his heuristic, the reader is asked to imagine that households are the suppliers of funds: to banks, who onlend the funds to firms as pure intermediaries; and directly to firms. The chronology would be (1), households save some portion of their income, (2), households make a choice about how they hold that saving, whether as bank deposit liabilities or as bonds from firms.
What we are modelling is saving and investment. We are not modelling credit creation or the financial sector per se.
To clarify, like Mankiw, I’m using “saving” in the conventional way: income net of consumption expenditure.
As I said, above, your use of the terms ‘save’ and ‘dissave’ here make no sense.
Let’s imagine that the banking sector is a single bank, which is a pure intermediary (i.e. no capital), which makes loans and holds deposits. Then, the liabilities of the bank are the assets of the non-bank sector, and the assets of the bank are the liabilities of the non-bank sector. Since bank assets equal bank liabilities, it is impossible for the non-bank sector to increase its savings via the simple act of bank lending.
Steve,
Right. Ditto the reverse.
Absolutely.
However, S is not identical to I. In certain very specific contexts, the value of S and the value of I are equal (identically equal). But in general, S does not equal I (because of international capital flows/absorption/trade imbalances/etc).
But consider this analogy: in equilibrium, supply equals demand and we can write, S = D, or S(.) = D(.). Since there is nothing to observe out of equilibrium, S will always equal D. But that does not imply that supply is demand. Clearly, supply and demand are conceptually distinct phenomena.
@wh10 “your comment that ‘banks lend deposits.’ Given your knowledge of all of this, I would be surprised if you weren’t aware that MMTers would not agree with the way you state this”
Yeah I’ve been aware of that, but haven’t taken the time to carefully track down MMT thinking. Off the top of my head, I think it has to do with how you think and talk about the accounting.
I don’t think anyone would want to claim that banks don’t lend out deposits. They clearly do. But at the margin — which is always in practice far beyond the point where all deposits have been lent — capital requirements are always the constraint on lending.
I’d be interested to hear “mainstream MMT” critiques of this thinking. In particular, does it misrepresent things in a way that confuses our understanding of how monetary economies work? I don’t think it does. At least, it seems like it helps me think clearly, avoiding logical contradictions.
@vimothy
“In his heuristic, the reader is asked to imagine that households are the suppliers of funds: to banks, who onlend the funds to firms as pure intermediaries; ”
Ok, so it seems, in his heuristic, banks as they exist in the real world do not exist (because they never work that way in the real world). Instead, banks are some imaginary entity that take deposits and actually loan those out. The problem is that I know of no planet on which banks actually work this way, so its applicability to the real is extremely questionable.
You made a point about a bank loan financing second-hand capital, but why can’t it finance net new investment? Why can’t it finance new factories and capital equipment? Of course it can.
“Let’s imagine that the banking sector is a single bank, which is a pure intermediary (i.e. no capital), which makes loans and holds deposits. Then, the liabilities of the bank are the assets of the non-bank sector, and the assets of the bank are the liabilities of the non-bank sector. Since bank assets equal bank liabilities, it is impossible for the non-bank sector to increase its savings via the simple act of bank lending.”
Vimothy, you are really confused here. Whether or not the bank is a pure intermediary or functions as banks actually do in the real world, and whether or not we impose capital requirements, assets and liabilities always balance. With that out of the way, the ‘pure intermediary’ model you are literally asking me to imagine is something which simply doesn’t exist on planet earth. Banks just don’t work that way. How can this be a useful model of the world if it’s incomplete and inaccurate?
Please explain to me how new investment occurs in your mind. And then, explain to me, *in the real world using the way banks actually make loans,* how a bank loan cannot finance net new investment. Here’s how I see it: Bank A gives a loan to Tom for $100. Bank A has a $100 loan asset and $100 deposit liability. Tom as a $100 deposit asset and a $100 loan liability. Tom then pays Mary to build a new factory. Boom – new investment. Assets and liabilities still balance. They always do, it has nothing to do with whether investment is new or not. New investment doesn’t allow you to break the laws of accounting.
wh10,
You’re forgetting about the government sector.
I’m not forgetting. In fact, I’m deliberately aggregating over both private and government sectors to get national saving, which is the sum of private and govt saving, and aggregate investment, which is the sum of private and govt investment, as I explained upthread.
Net govt borrowing will increase non-govt net financial assets, by definition. But it cannot by itself increase national savings for the same reasons simple bank lending cannot increase savings. If we ignore the foreign sector, then the change in net financial assets of the non-govt equals the change in liabilities of the govt and it all nets out to zero when we consider the economy as a whole.
Vimothy, it seems you are confusing a lot of stuff here
Lol. Always a possibility. On the other hand, who is more likely to be right–you, or economic theory, which has been developed over hundreds of years?
Everything always ‘washes out’ from that perspective.
Only financial assets wash out from this perspective. The capital stock does not wash out. That’s why it’s possible to have S > 0, and why it’s meaningful to talk about national savings / national net wealth / net assets / etc.
Getting to the point, why does it have to be second hand equipment?
It does not have to be second hand equipment. If it is, then it has no effect on S = I.
BTW, it’s not true that there is no limit to the amount of investment that can take place in an economy. In a theoretical model, of course you can set investment to wherever you want, and even have it go shooting off into infinity. But in reality, that is not possible and there is a relationship between output and investment.
@Asymptosis
“I don’t think anyone would want to claim that banks don’t lend out deposits. They clearly do.”
Wait, how can you be so sure? Where does the accounting of bank lending ever suggest that? The accounting of a bank loan is: +loan asset and +deposit liability for the bank, and +deposit and +loan liability for the borrower. No previous deposits are involved or ‘lent out.’ Where in the accounting is a ‘deposit’ ever lent out? Now, if the bank needs to acquire reserves, due to reserve requirements, then they go borrow them from the interbank market or go to the discount window, or maybe the Fed has to give them reserves in exchange for treasuries if the FFR starts to creep up. But again, no deposits are lent out here. Amongst other things, deposits may be cheaper sources of loan financing then borrowing reserves to meet reserve requirements; that’s one main motivation for attracting depositors.
“I’d be interested to hear “mainstream MMT†critiques of this thinking. In particular, does it misrepresent things in a way that confuses our understanding of how monetary economies work? I don’t think it does. At least, it seems like it helps me think clearly, avoiding logical contradictions.”
So I’d say it does misrepresent things, both from a pure accounting perspective and from an implication perspective.
@vimothy
“I’m not forgetting. In fact, I’m deliberately aggregating over both private and government sectors to get national saving, which is the sum of private and govt saving, and aggregate investment, which is the sum of private and govt investment, as I explained upthread.”
Ok fine.
“Net govt borrowing will increase non-govt net financial assets, by definition. But it cannot by itself increase national savings for the same reasons simple bank lending cannot increase savings. If we ignore the foreign sector, then the change in net financial assets of the non-govt equals the change in liabilities of the govt and it all nets out to zero when we consider the economy as a whole.”
Ok. Fine….
“It does not have to be second hand equipment. If it is, then it has no effect on S = I.”
AND IF IT ISN’T????? Then a bank loan drove net new investment for the private domestic sector???????
wh10,
Ok, so it seems, in his heuristic, banks as they exist in the real world do not exist (because they never work that way in the real world).
We made no assumptions about the behaviour of banks–all we did was simplify their capital structure. Even in the real world, if we ignore the foreign sector, we can still say that banks’ customers and owners are the same people on aggregate.
You made a point about a bank loan financing second-hand capital, but why can’t it finance net new investment?
Well, obviously it can. Nothing that I’ve written here implies otherwise. If the bank loan finances new investment, then there will be a corresponding real flow of saving (assuming autarky). Banks are not important in this model, which is a model of macroeconomic saving and investment.
Vimothy, you are really confused here.
Again, huh! I guess I’m really not doing well here. On the other hand, I do remember something called an “equity equation”: total assets – minus total liabilities = equity. Confused or not, I’m just setting the right hand side to zero.
How can this be a useful model of the world if it’s incomplete and inaccurate?
A very good question. In response, I’d like to ask you to give your own model of saving and investment.
“Lol. Always a possibility. On the other hand, who is more likely to be right–you, or economic theory, which has been developed over hundreds of years?”
I find this comment interesting – an appeal to authority to diminish someone else’s argument…
It reminds me of an old adage regarding sports – “practice makes perfect”. Turns out this is not precisely true – as someone once pointed out to me – it should be “perfect practice makes perfect”.
It does one no good to improve one’s skills doing the wrong thing.
wh10,
AND IF IT ISN’T????? Then a bank loan drove net new investment for the private domestic sector???????
If the capital is new, then it appears in I, since I just sums up all the investment that has occurred over that period. The bank loan corresponds to a saving flow, so it appears in S. S = I.
paulie46,
You are really confused here.
@vimothy “Let’s imagine that the banking sector is a single bank, which is a pure intermediary (i.e. no capital), which makes loans and holds deposits.”
Beautiful! This is exactly the thought experiment I bruited recently somewhere hereabouts, and that you just queried me on.
There is also another bank that only has capital — no deposits — and lends against that capital times X.
In practice the two are combined. And the arithmetic of calculating regulatory ratios also combines the two, incorporating deposits, reserves, capital, and total loans outstanding in those ratios.
This makes perfect sense, but it’s also confusing if you’re trying to understand conceptually what drives bank lending.
I think the two-bank model really helps clarify that. The first bank is always fully lent (or, 90%), and the available lending funds for that bank are unaffected by the velocity/turnover of transfers (spending aka not saving) among its depositors’ accounts.
Lending by the second bank is affected by velocity, but not directly — only to the extent that higher/increasing velocity (aka economic activity) stimulates higher expectations in both borrowers and lenders, and (at even more removes) all the game theory of monetary policy, fiscal policy, etc.
The effect of more saving — less spending, less GDP, lower expectations — on the second bank is to reduce lending.
Another way I recently thought of for thinking about this: don’t think about something’s effect on “loanable funds”; it’s kind of useless and causes logical errors. Think about something’s effect on lending.
@Asymptosis
“I’d be interested to hear “mainstream MMT†critiques of this thinking. In particular, does it misrepresent things in a way that confuses our understanding of how monetary economies work? I don’t think it does. At least, it seems like it helps me think clearly, avoiding logical contradictions.”
Steve, I’m with wh10 here, I do think it misrepresents things.
In the MMT world, the money for loans is created out of thin air, and when paid back disappears into the ether. This makes sense to me. Using $11 of a savers deposit to make a $100 loan doesn’t.
The savers deposit at the bank is still there and the saver can have it on demand.
In what way is the savers deposit being used? How does this work in banking systems with no reserve requirement (like Canada)?
@vimothy
Confused how, can you be specific?
@vimothy
I need to go do work, but quickly-
“Well, obviously it can. Nothing that I’ve written here implies otherwise. If the bank loan finances new investment, then there will be a corresponding real flow of saving (assuming autarky).”
Okay, and if bank lending is theoretically unrestricted since bank loans create deposits out of thin air, then there is theoretically no limit to new investment and savings that might be financed by those bank loans. And so in this model, the concept of ‘loanable funds’ for the private domestic sector makes no sense since it implies there are pre-existing funds, in a fixed amount, waiting to be lent out. And who cares how you aggregate sectors? The laws of accounting hold everywhere.
@Asymptosis
Steve, I have definitely seen Fullwiler say that you should not think of reserves as capital, because they are not. I need to learn more about the specifics, but at a high level, capital = asset – liabilities. If we think about a customer making a deposit, that deposit will create a reserve asset but also a deposit liability. There is no capital created from that. The ‘reserves’ you read about in the calculation of capital requirements are not those banks keep in Fed accounts, as far as I understand.
paulie46,
The “appeal to authority”, such as it was, was in response to wh10’s claim that I am really confused about this stuff. Of course, as I also wrote in my “appeal to authority”, maybe I am really confused about this stuff. Since you seem to think so, perhaps you could explain how…?
@wh10
I think you’re saying that the reality of how accounting is done in the banking system means that my way of understanding that system doesn’t make sense. But I don’t think you’re right.
Lemme think about this:
Could I open a bank that just takes deposits and lends 90% of them — borrow short, lend long — keeping 10% on reserve at the fed? (What would my capital requirements be? Very small, I think?)
For my bank, the following would not be true:
“No previous deposits are involved or ‘lent out.’”
My bank would have no effect on total deposits or lending — I’d just be holding/lending deposits that could be held/lent by other banks, which are lending against capital times X — far beyond 90% of deposits.
Another way to think about it: suppose banks didn’t have the license to print money; they could only lend their deposits. (Only the first/my type of bank existed.) There would be lending, no?
“I’d say it does misrepresent things, both from a pure accounting perspective.”
“From a pure accounting perspective” — how banks actually do their accounting — you’re absolutely correct.
But I don’t think that means my conceptual representation is problematic.
“and from an implication perspective”
I don’t think so. My representation still leaves us in a world where, both theoretically and in practice:
A. Saving does not create loanable funds, and
B. Capital is the constraint on lending (because all deposits are — or can be seen as being — lent out at all times).
??
My bank would not affect the amount of lending, because the constraint on lending is at the margin
@wh10
Steve, I should probably learn before I type, but I see that you might mean that reserves will be factored into ‘risk weighted assets.’ I can’t find any sources that suggest this. I have seen that cash and cash equivalents are given a 0 weighting in calculating risk weighted assets.
@Asymptosis
I think I want to change the preceding:
“all deposits must (from an accounting perspective) be seen as being lent out at all times”
@vimothy
My mistake, apparently I misunderstood your intent. For that I apologize.
@wh10 “I see that you might mean that reserves will be factored into ‘risk weighted assets.’”
‘kay do this for me:
A bank has $100 in top-tier capital and no deposits. How much can it lend?
It receives $100 in deposits. How much can it lend?
Can it lend 90 more dollars after the deposit? My conceptual model says yes. Is that wrong?
@paulie46 “Using $11 of a savers deposit to make a $100 loan doesn’t.”
That’s exactly not what I said. I said use $100 of deposits to create $90 in loans.
wh10,
there is theoretically no limit to new investment and savings that might be financed by those bank loans.
There certainly are theoretical limits to saving and investment. Saving and investment are both real activities. Even if we assume that bank sector credit supply is infinitely elastic, the supply of savings is limited by real output and the demand for savings is limited by real investment. If we want to think about expanding the supply of credit, we cannot simply find the supply of saving or investment via some simple linear transformation of the change in bank lending.
Moreover, there’s nothing forcing creditors of banks to be the located in the domestic economy, and nothing forcing the debtors of banks to be investors in the domestic economy.
the concept of ‘loanable funds’ for the private domestic sector makes no sense since it implies there are pre-existing funds, in a fixed amount, waiting to be lent out
The funds originate from the total flow of saving for the economy: S = Y – C. Clearly, S is not fixed, but depends on Y and C.
@Asymptosis
“@paulie46 “Using $11 of a savers deposit to make a $100 loan doesn’t.â€
That’s exactly not what I said. I said use $100 of deposits to create $90 in loans.”
Sorry, It appears like my comment is out of sync with the discussion – I said that earlier before you said this:
“A bank has $100 in top-tier capital and no deposits. How much can it lend?
It receives $100 in deposits. How much can it lend?”
That said, a bank can lend $900 with $100 in deposits (roughly).
paulie46, No worries and no need to apologise.
@vimothy “give your own model of saving and investment”
As a nation, we save — create “savings” — by creating fixed assets (actually real assets, but only fixed are counted).
People/businesses have three choices at any moment:
• Spend on consumption. This depletes national savings by consuming inventory, and fixed stock (through use) to refill the inventory.
• Spend on investment. This increases national savings by increasing inventories and the stock of fixed assets.
• Save/don’t spend. This non-action has no effect on national savings.
@paulie46 “That said, a bank can lend $900 with $100 in deposits (roughly).”
Really?? Absent any capital? I don’t think that’s right. Aren’t you expressing the whole fractional-reserve banking confusion, in a nutshell?
@Asymptosis
Or did you type an extra zero after $90 by accident?
Steve,
I’ve been having a bit of trouble figuring out how your example works!
If a bank is all equity on the liability side of its balance sheet, taking no deposits, when it makes a loan, what actually happens?
It sounds like fractional reserve banking but that isn’t the dynamic. Loans do create deposits in the banking system, but the level of reserves don’t limit a banks ability to lend. That’s another discussion.
That said, a bank can lend nearly 10 times it’s deposits (or capital requirement). I didn’t think that idea was controversial.
Otherwise are you implying we have 100% reserve requirement?
@Asymptosis
Steve,
I do think you are thinking about it incorrectly conceptually, but I have little experience calculate actual capital reqs etc, so I am learning in the moment.
Let’s say the rule is that capital must be at least 8% of total risk-weighted assets. Let’s say loans are risk weighted 1X. So $100 of capital gives us the ability to create $1250 loans max. When the bank receives $100 in deposits, that doesn’t change this calculation. The new reserves that come along with the deposit do not supply the bank with additional capital; they equal the liability of the deposit. Furthermore, bank lending *is not constrained by deposits or reserves.* Banks acquire the reserves after the fact, in the interbank market or from the Fed, if necessary.
Am I following your example properly?
@vimothy
Vimothy, my sense is we keep talking past each other. In the real world, of course there are limits. it doesn’t change my points, but I am going to give up now because clearly the internet just doesn’t work for these types of discussions.
@wh10
Steve, I have been trying to find a Fullwiler piece on this. Never have actually seen this one, but it gets at these points. http://www.neweconomicperspectives.org/2009/06/dont-fear-rise-in-feds-reserve-balances.html
So to be clear:
“A bank has $100 in top-tier capital and no deposits. How much can it lend?”
Under my dummy rules, $1250.
“It receives $100 in deposits. How much can it lend?”
Same.
“Can it lend 90 more dollars after the deposit?”
No. The deposits don’t change capital, and the bank doesn’t need or use deposits to lend anyways.
I realize before you were trying to create a bank that does lend deposits, but I don’t really understand how that works and how it is applicable to the real world. I would implore you to try to do the accounting for such a bank loan. If I am still not getting what you are saying, that might help me understand better.
@wh10
Doesn’t the loan create an asset on the banks balance sheet? They get an income stream in excess of principal that creates value from money they loaned from thin air.
@wh10 “When the bank receives $100 in deposits, that doesn’t change this calculation.”
Okay, wow, if that’s true I am indeed totally wrong.
Here’s how I imagine the accounting:
$100 deposit received. LHS Cash asset, RHS Deposit liability
$90 lent. – LHS Cash asset, + LHS loan asset
Assuming the Fed lets them do this — borrow short, lend long, with a buffer reserve — the accounting doesn’t seem wrong…?
Suppose a bank has $100 in capital and $1250 in outstanding loans.
I come along and deposit a million dollars at the corner branch.
That really doesn’t increase their ability to lend?
@paulie46
Right, the max $1250 would be assets on the bank’s balance sheet (I called them ‘risk weighted assets’ for the purposes of capital reqs). *If* they net profit off of these loans, that could create additional capital for that bank, allowing them to loan even more.
Steve,
As a nation, we save — create “savings†— by creating fixed assets (actually real assets, but only fixed are counted).
What do you mean by fixed assets and what do you mean by real assets?
In an open economy saving is not necessarily equal to investment; that is, we can also save as an aggregate by acquiring net foreign assets.
Spend on consumption. This depletes national savings
Consumption spending only depletes savings if it occurs in excess of income.
Spend on investment. This increases national savings
What if the investment is funded by foreign capital? In that case, national savings do not increase.
Save/don’t spend. This non-action has no effect on national savings.
Why? In what sense?
@Asymptosis
I really need to stop. This is really fun to do, but I have a day job!
“Okay, wow, if that’s true I am indeed totally wrong.â€
That’s how I understand it. Fullwiler in the past has said if you think of reserves as capital, you’re going to get banking totally wrong. *I believe* this is the capacity in which he meant that.
“Suppose a bank has $100 in capital and $1250 in outstanding loans. I come along and deposit a million dollars at the corner branch. That really doesn’t increase their ability to lend?”
***Holding everything else constant*** (see next paragraph), from the way you are thinking about bank capacity for loan creation, theoretically, it wouldn’t. Why and how would it? Show me how it increases the bank’s capital. It’s the same reason why QE doesn’t increase bank capacity for lending. Those excess reserves don’t give banks any more fire power to loan (read the Fullwiler piece). We know it doesn’t change capital, and we know that banks don’t need deposits in order to make loans – they can always borrow the reserves/cash from the interbank market/Fed.
***However, you might question if this allows the bank to lower the rate on its loans, and thus attract more borrowers (thereby increasing the amount it can loan), since it has reserves that it doesn’t have to borrow at the FFR. A long time ago, when I had a much poorer understanding of this stuff, I asked Fullwiler this. His response: “Excess reserves that are in excess of what’s desired as a buffer against overdrafts can always be (1) loaned at the fed funds rate (pre-Lehman), or (2) held to receive IOR at the fed funds rate (current situation). Given that, it makes no sense that they would reduce the rate charged.†(What Fullwiler means, I think, is that the FFR is always the opportunity cost driving the base level interest rate on loans.) “It is true, though, that if a bank has a lower cost of liabilities (i.e., can easily acquire more deposits as it expands its balance sheet), it can make a loan at a lower rate. This has nothing to do with the ER position, though, since when a bank makes a loan, it normally expects the deposits created by the loan to be withdrawn–so it’s more about whether the bank can acquire more deposits as it expands its assets. Also, the potential rate reduction even in this case would only be marginally so since the rate should be risk-weighted, and if the rate is set significantly below what competitors set, then all the riskier folks come to the bank with the lower rate. Capital gets used up more quickly as riskier loans are made, and regulators watch a bit more closely.â€
“$90 lent. – LHS Cash asset, + LHS loan asset”
Let me try this two ways.
First, let’s assume there are no capital requirements. Why would the bank stop at $90? Its deposits are not restricting its ability to make loans… It can always acquire those reserves/cash in the interbank market/Fed at the prevailing FFR. I am assuming this is a real-world bank; I am not sure why it’s useful to think of these imaginary banks that are just intermediaries – it would then cease to be a bank as we know it.
Second, let’s assume there are capital requirements and do the accounting in chronological sequence. First, after the loan is made, it’s +LHS loan asset, + RHS deposit. If the borrower withdraws, it’s –LHS cash, – RHS deposit. Ok, so on net, it’s –LHS cash, +LHS asset, as you say. BUT, the bank is still bounded by capital requirements; it’s receiving of the $100 in deposits doesn’t change the $1250 calculation and thus doesn’t give it greater allowance for loans. Furthermore, if those $100 deposits were excess reserves at the time they were received, the bank would loan them in the interbank market, otherwise they are foregoing the FFR they could be earning (the opportunity cost). There’s no point to them keeping the $90 to make future loans, since they could acquire them at the FFR anyways. You’ll notice, prior to QE and IOR, banks had near 0 excess reserves; keeping excess reserves beyond what they wanted as a buffer was a cost to them.
As far as borrowing short, lending long, not sure if this is relevant to your specific point, but Warren posted this the other day, with challenge and clarification in the comments. “Bank’s aren’t allowed to take what’s called ‘interest rate risk’ by borrowing short and lending long. It’s the first thing the regulators and supervisors look for.” http://moslereconomics.com/2012/02/01/another-gross-error/
@wh10
Steve, to clarify, despite the points in my “***” paragraph, I still don’t see how the deposit of $1M would allow the bank to loan more in and of itself. It seems to me, for the bank to loan more assuming it’s at the max of $1250 loans, it needs more capital. Maybe I have the formula for capital requirements wrong, but I don’t think so… Perhaps the $1M deposit allows the bank to make a profit off of various fees/services, and this increases capital, but the $1M itself doesn’t directly change loan capacity AFAIK.
@wh10 “First, let’s assume there are no capital requirements. Why would the bank stop at $90?”
Because for any bank with more than $11.5 million in deposits, there are *reserve* requirements.
http://www.federalreserve.gov/monetarypolicy/reservereq.htm#table1
Steve,
I wouldn’t describe the above as a model of saving and investment, BTW. It might motivate a model, but it is not the same as one.
Assume a closed economy; consequently, S = I at all times. If we increase the value of I, then it follows by accounting identity that S is also increased. But that’s just restating the tautology. What we want to know is, what does it mean for I to increase? And what effect will the increase in investment–whatever that means–have on saving?
That’s what your model needs to address.
@Asymptosis
Steve, but the point is that the bank will *acquire the reserves in the interbank market or from the Fed after they make the loan.* They don’t need reserves to make loans!
That bank, assuming no capital requirements, can theoretically make much more than $90 in loans. It could make $1000 in loans, or whatever. It just needs to make sure it acquires 10% of deposits in the form of reserves from the interbank market.
@wh10
Great link to the working paper at #41. I’ve now read it through and I think it has finally (in my mind) crystallized the arguments between the MMT community and (pretty much) everyone else in the world. It isn’t really about math although that is part of it.
Until reading the paper I had been having a hard time understanding the logic of their arguments but I think I see it now.
Not that I agree with it because from where I sit he (Fiebiger) seems to be solving the wrong problem.
@paulie46
I skimmed it the other day. Fiebiger’s writing is so poor, it’s hard for to me follow. His rejoinder also seems to completely ignore the MMTer’s response. In any case, the Lavoie critique is better, and there is a good thread on it at Heteconomist, with Fullwiler’s reply (which was folded into this working paper).
wh10,
Fair enough. Incidentally, I don’t agree that we are talking past each other. It seems to me that we are disagreeing pretty straightforwardly about the coherence of the loanable funds model. And FWIW, and I’ve no idea if this is anything to do with anything, I don’t mind being described as confused. I consider that banter and all in the spirit of a good internet dust up.
Here’s a stab at clarifying one issue:
Take a four sector economy, with, private, govt, financial and foreign sectors. The financial sector balance sheet is M = L (deposits equals loans), which implies d(elta)M = dL. By assumption, the financial sector lends to all other sectors but borrows only from the private sector (I’ll write subscripts on L when this happens). Obviously, it’s not really possible to write out everything in the comments here. I’ll just focus in on a couple of accounts.
How does the private sector acquire assets during any given period? Gross private sector asset acquisition can be written as I_p + dB_p + dM, where I_p is private investment, dB_p is net new lending to the government and dM is new deposits.
How does the private sector fund the acquisition of those assets? Sources of finance can be written as S_p + dL_pg + dFB, where S_p is private saving, dL_pb is loans from the banking sector, and dFB is net foreign borrowing.
The final two terms in the second identity represent borrowing from other sectors. The private sector will acquire increased net assets to the extent that it maximises the proportion of its financing represented by private saving. (The flow of) private saving is akin to the change in the private sector’s net worth.
@vimothy
I am but a poor self-taught MMT student (albeit an old one) but in my view the sectoral balances equation is a representation of the flow of NOMINAL funds between the three sectors (the banking system is in the non-government sector). So when one talks about assets and liabilities within this system one is talking about FINANCIAL assets only i.e. dollars and treasury securities only.
The flow of these NFA’s affects the real economy but it seems to me there is no “hard” connection between real and nominal. The effect is more like induction, a soft, non-linear relationship (non-linear because the transfer function is unknown.)
Now maybe I’m out in left field but the discussion here is not (?) real vs nominal but some of the arguments are mixing the two universes. I think those realms are parallel and separate.
@vimothy
Vimothy – You’re probably not confused about your model, but there is a chance you are confused about my model. This happens with Nick Rowe a lot. He doesn’t engage readers on their own terms with the models they bring to the table. He always brings it back to the model he believes in. “Look at this way.” I can’t fault him for that, it’s totally understandable. You do the same thing, and I do the same thing. In that sense, I think we are talking past each other, because we aren’t directly addressing each other’s points. There seems to be some disagreement regarding the implications of double-entry accounting and sector conglomeration.
I leave you with this, as it’s how I see it:
“First is the accounting logic of real-world transactions. Every transaction in a real-world economy affects financial statements of those engaged, and if an economic theory or a posited model is not consistent with how real-world financial statements are affected, then the theory is inapplicable. A typical example used by MMT’ers is a framework used in mainstream economics, the so-called loanable funds market. It posits a demand for loanable funds and a supply of loanable funds available for the macroeconomy, and contains classic supply-demand curve assumptions from goods markets, that higher prices (in this case interest rates) will elicit more “supply” (as in investors will divert more funds from other uses, such as risky venture investments, and make them available for lending). This model is simply inapplicable to our current monetary system in which empirical studies have demonstrated that banks create loans “out of thin air†without the requirement of prior reserve balances or deposits to “fund†the loan’s creation. Completely contrary to the loanable funds model, in fact, the vast majority of bank liabilities have been created by banks simply growing their balance sheets through loans and asset purchases. Similarly, there are macroeconomic accounting identities, such as the often-cited sector financial balances equation in which the domestic private sector’s net saving of financial assets is by definition equal to the government sector’s deficit and the current account balance (see here, here, and here for further discussion). MMT’ers understand very well that an accurate understanding of accounting is not in itself a theory.”
http://www.modernmt.net/download/MMT%20Primer.pdf
@wh10
BTW, Vimothy, this is what is motivating me to get a graduate degree in Economics, so I can give myself a fair chance of understanding where the mainstream is coming from. The undergraduate version obviously didn’t convince me enough, given that I have found these heterodox theories more robust and logical relative to the real world. It’s really hard to break out of one mold and to see the other side. Perhaps a couple of years ago I would have seen it the way you do; right now, it’s really hard to. I think I would need some intense, isolated time with the material to be more conversant.
@vimothy
‘kay. I was under the misapprehension that the required reserves were sufficient to lend (90% of) the $1 million, when they are in fact only *necessary* (and borrowable). Sufficient capital is also necessary.
But returning to the spur for this mini-thread:
Suppose this bank gets the million-dollar deposit, sez “yipee!” and acquires circa $80K in capital. +LHS cash, +RHS equity.
It can then lend (circa 90% of) the million dollars, right? -LHS cash, +LHS loan asset.
So is it wrong in that case to say that it’s “lending [a portion of] its deposits”?
But that aside:
“Banks aren’t allowed to take what’s called ‘interest rate risk’ by borrowing short and lending long.”
Wow. That’s huge. It’s one thing to debunk the fractional-reserve notion. Getting rid of this one will take decades or centuries.
Help me with one more thing:
I transfer a million dollars from Bank A to Bank B. What does Bank B do with that money? (Assume a pre-’08, pre-IOR regime for simplicity.)
@wh10
That last one should have been directed to you, not vimothy.
@vimothy: “What do you mean by fixed assets and what do you mean by real assets?”
FAs are the ones counted in the NIPAs — structures, equipment, and software.
Real assets are much harder to define much less quantify. Skills, knowledge, ideas, “organizational capital,” etc. But in total, they’re the things that allow us to produce stuff in the future. FAs being an (important) part of that “stock.”
paulie,
The flow of these NFA’s affects the real economy but it seems to me there is no “hard†connection between real and nominal.
That’s quite an important insight. Given that you are self-taught you must have good intuition. Most schools of thought hold that there is no particular thing tying real variables to nominal variables–hence, macroeconomic theory. In my view, at best MMT is not very interested in this distinction, which is quite problematic from a theoretical point of view.
The sector financial balances can be given in real terms and in nominal terms. For example, there is a real flow of NFA and a nominal flow of NFA. Given our set of identities, they must hold ex post at all times in real and nominal terms.
The sector balances are just residual terms that are created by the inter-sectoral flows of income and expenditure.
The flow of funds (real or nominal), are not identical to the flows of inter-sectoral lending and borrowing (the cross-sector income-expenditure residuals). The flow of funds are something which happens in addition to current income flows–imagine that them overlaying the income-expenditure flows like a palimpsest. (JKH is obviously your man for this kind of thing).
I gave an example above where I showed a hypothetical decomposition for private sector gross asset acquisition and financing flows. The upshot was basically that private sector assert acquisition is equal to private saving plus private borrowing. At the same time, private sector income is equal to private saving plus private consumption plus taxes plus interest payments to foreign creditors.
@All:
This Steve Randy Waldman statement in a comment on his latest post raised my eyebrows:
“Regulators pay attention to retail deposits as a marker of the stability of bank funding,”
Thoughts on deposits removing individual banks’ “constraints” on lending, given that?
wh10,
In that sense, I think we are talking past each other, because we aren’t directly addressing each other’s points.
Could be. On the other hand, I think I understand where you’re coming from to some extent, because I remember thinking similar things, and even arguing them to some of my lecturers.
The disagreement about accounting that you refer to might relate to the way I’ve treated banks so far. So let me try to shed some light on that. Way upthread, you wrote,
Both the bank and the customer acquire an asset and liability; the bank acquires a loan (asset) and deposit (liability), whereas the customer acquires a deposit (asset) and debt (liability). It’s not split between the two.
Notice here that the bank’s liability is the same as the customer’s asset, i.e. the deposit. If you forget about the deposit for a moment, what you have as that the customer owes the bank some money–in other words, the customer’s liability is the bank’s asset. But, on aggregate, the customers own the bank, since the bank is ultimately borrowing from and lending to the same group of people. So the whole thing is a wash. That’s why MMTers talk about net financial assets. Borrowing and lending within a sector cannot generate any net financial assets for that sector. You need to lend to another sector, such as the government or the foreign sector, to do that.
vimothy,
…In my view, at best MMT is not very interested in this distinction, which is quite problematic from a theoretical point of view.”
What is important about MMT is that insights gained from observing nominal flows can rule out many policy proposals that don’t pass the math test so that one can focus on the policies that can work or at least lead to the desired result. An illustration is the quote by Fulwiller posted by wh10 yesterday at #49.
When you understand (believe?) that we can’t “run out of money”, that public debt and private debt are distinctly different in relationship to the economy, and that “printing money” alone does not correlate with inflation (so far) it opens up a wide policy space beyond that deemed prudent with neoclassical thinking.
From an engineering (my) point of view MMT is more like science than other economic schools of thought. It is a systems approach rather than an approach that models some parts of a black box to try to predict its output.
The 2nd Law of Thermodynamics rules out perpetual motion machines.
paulie46,
Well, I’ve probably come to almost the exact opposite conclusion: MMT is mostly black box. The bits that aren’t black box are quite interesting, though.
Everybody else working who does large scale macro modelling already takes the idea of accounting consistency seriously. The proposition that the aggregate accounts don’t have to balance is not a feature of mainstream theory. Of course, the budget constraints derived from the sector accounts can help to define the “feasible set” of the policy maker’s choice problem. That’s a very natural way of approaching things from a mainstream point of view.
And I think everybody else in the discipline realises that we can’t simply “run-out” of money. (What would that even mean)? That’s what is implied by the nominality of of money. It’s just an arbitrary number with no intrinsic value. What we’re most interested in from a welfare perspective is the behaviour of real variables–and our interested in nominal variables stems from that.
The correlation between money supply growth and inflation is actually very well established. For example, see here: http://minneapolisfed.org/research/qr/qr1931.html
McCandless, George T. and Warren E. Weber, “Some Monetary Facts,†Federal Reserve Bank of Minneapolis Quarterly Review, 1995
For a review of some of the evidence. IIRC, the authors find the correlation between money growth and inflation for several definitions of money in a large sample of countries (the population?) to be almost unity (with some sub-sample variation), and the correlation between money and output and inflation and output to be almost zero in both cases.
“Borrowing and lending within a sector cannot generate any net financial assets for that sector. You need to lend to another sector, such as the government or the foreign sector, to do that.”
Banks straddle the government and non-government. All of the bank’s “real-world transactions” (not a good description, sorry) such as employee salaries, expenses and interest earned occur in the non-government sector but the money that banks issue thru lending is a liability of the Fed.
The only sector where “net financial assets” has any meaning in real-world terms is the non-government sector.
If we define the closed system in terms of dollars then the foreign sector can be considered (for the moment) as part of the non-government sector to the extent it holds dollars or T-securities. The math is the same.
Government obligations are the equivalent of keeping score – accounting only, nothing real.
It is mathematically impossible for the government to borrow “funds” from the private sector and increase net dollar assets. We can only increase the stock of treasury securities through borrowing from the non-government sector. Subsequent deficit spending can only replace the funds that were removed to purchase the treasuries. In the case of a new country with a new fiat system introduced, obviously no borrowing to “seed ” the system would be possible.
Still, the government must have a way to create money (as net dollar assets) without “borrowing” from the public. 1/3rd of the National Debtâ„¢ is held by the Fed, 2/3’s by the public.
vimothy,
“The correlation between money supply growth and inflation is actually very well established. For example, see here: http://minneapolisfed.org/research/qr/qr1931.html”
I should have been more specific, I was referring to the US in particular…
Don’t know much about the dynamics of the rest of the world other than those countries with a monetary system similar to ours (Canada, Australia, Japan, UK) for which I expect my statement would hold true. Still, don’t want to get off on a tangent.
paulie46,
The only “money” that is a liability of the Fed is base money, which is notes and coin and reserves. Base money held by banks is held as an asset. The “money” that the banks create is deposit money, which is a liability of the banking sector.
Bank money != central bank money (that’s a “not equals” sign, BTW).
Net financial assets just means net financial assets. It has meaning in any context where we’re comparing assets and liabilities.
It is mathematically impossible for the government to borrow “funds†from the private sector and increase net dollar assets.
The government increases the net financial assets of the non-government sector precisely by borrowing from it.
Don’t know much about the dynamics of the rest of the world other than those countries with a monetary system similar to ours (Canada, Australia, Japan, UK) for which I expect my statement would hold true. Still, don’t want to get off on a tangent.
I think the authors address that through a sub-sample analysis. In any case, there is a huge literature on this if you are interested in finding out if and where your statement does hold true. But I agree that this is tangential.
“It is mathematically impossible for the government to borrow “funds†from the private sector and increase net dollar assets.”
“The government increases the net financial assets of the non-government sector precisely by borrowing from it.”
These two statements are mutually exclusive, so one of us has to be wrong.
First, the government increases the “net financial assets” by swapping treasuries for the existing dollars (x). At this point there are no more dollars (unencumbered) in existence. All that exists is net financial assets in the form of treasuries.The government then “keystrokes” new dollars into the non-government sector by crediting bank accounts. The final position is 2x in NFA’s of which there are x treasuries and x dollars. The next go-round there is 3x NFA’s of which 2x is treasuries. NFA dollar assets can never increase.
By analogy, what you are saying is…
Start with a swimming pool with one pail of water in it. The pool represents the closed system and the water the financial assets. (Imagine each drop of water is a dollar if it helps)
Then dip the pail of water out of the pool and pour it back in the other end.
Repeat over and over agin until the pool fills up.
This is called bootstrapping or “pulling one’s self up by the bootstraps”.
It is mathematically and physically impossible, the pool can only be filled from an external source (outside the closed system). The government (Fed/Treasury) must create the new dollar assets that are removed and swapped for Treasuries. These assets do not come from the private sector – they are entered into the system from thin air.
I can demonstrate the same thing with a simple algebraic proof but you get the picture.
@vimothy: “The correlation between money supply growth and inflation is actually very well established.”
Right. But the causation is decidedly not. From a Mark Thoma posting, see cite therein:
“This high correlation does not, however, have any implications for causality. … An alternative possibility, equally consistent with the high correlation, is that other factors generate inflation, and central banks allow the growth rate of money to adjust.”
http://economistsview.typepad.com/economistsview/2007/04/thomas_palley_t.html
Isn’t this basically the MMT view? That the Fed just accomodates the banks’ need for money by issuing reserves?
So the need for money would arise at the same rate as inflation (which is presumably all caused by demand banging against supply)? I’m pretty fuzzy here…
Steve,
Causation is a different and more technically demanding matter. We have been discussing correlations here, because paulie46 claimed that money growth does not correlate with inflation, which is false.
The authors of the study I cited try to account for the selection bias effect you identify by taking a large sample with variation across policy regimes. (In standard regression analysis, conditioning on a set of covariates solves this selection bias problem–but I don’t know much about their methodology.)
They discuss this issue briefly:
We hope that the range of policy rules in our cross section of countries is so broad that the correlations we observe are independent of the policy rules. Even if all central
banks were following a constant money growth rule, we doubt that they’d all be following the same one. That’s true for feedback rules too. So, by using a large cross section
of countries, we hope our correlations will be free of policy rule influences.
Independence of correlations from policy rules is important because we want the correlations we find to be useful for determining whether causal relationships exist.
While correlations are not direct evidence of causality, they do lend support to causal hypotheses that yield predictions consistent with the correlations. Consider, for example, the hypothesis that a monetary policy with a higher growth rate of money will result in a higher inflation rate than a policy with a lower rate of growth in an otherwise identical
economy. That hypothesis would be supported (though by no means conclusively) by observed positive correlations between money growth and inflation.
One last thing–according to the study, there is no correlation between money growth and output growth and no correlation between inflation and output growth (so money does not grow faster when the economy is growing faster, for example, which you seem to suggest in your final para).
@Steve
“I got quite a bit of blowback on my recent post suggesting that economists don’t understand accounting.”
This is the first sentence of your original post. I would suggest an amendment…
“I got quite a bit of blowback on my recent post suggesting that economists don’t understand…” closed systems
paulie46,
These two statements are mutually exclusive, so one of us has to be wrong.
There’s more to it than that. This is actually a point of agreement between my argument and MMT’s argument, so if you are right and I am wrong, MMTers must be wrong as well.
In your comment, which is not easy to parse, you seem to be assuming a fixed supply of “dollars”–but there is no fixed supply of anything. Dollar deposits are certainly not fixed. Base money is not fixed either.
The supply of money is not the same as the supply of financial assets, and the supply of financial assets is not the same as the supply of net financial assets in the MMT sense.
So,
First, the government increases the “net financial assets†by swapping treasuries for the existing dollars
When the government sells a bond, it is borrowing from the non-government and the bond sale increases the non-government sector’s holdings of net financial assets.
If the government is simply swapping existing treasuries for dollars, then it is the central bank and it is not increasing the supply of net financial assets, just changing the composition of what already exists.
At this point there are no more dollars (unencumbered) in existence.
What do you mean “the are no more dollars in existence”?
The government then “keystrokes†new dollars into the non-government sector by crediting bank accounts
What do you mean “keystrokes new dollars into existence”–I thought the government had just obtained some dollars by swapping them for treasuries?
The final position is 2x in NFA’s of which there are x treasuries and x dollars.
No, the final position for the non-govt is the same amount of dollar deposits and x new NFA.
NFA dollar assets can never increase.
But this completely contradicts everything you’ve just written, to say nothing of basic laws of accounting and the accounting identities that MMT holds dear. If NFA cannot increase, then how did the non-govt sector end up with x new NFA as the result of the govt’s net expenditure / bond issuance?
By analogy, what you are saying is…
That isn’t anything like what I’m saying!
Your analogy fails because it’s a bad analogy. It’s not that economists don’t understand closed systems–you’re not describing the monetary system we actually have.
@vimothy
“In your comment, which is not easy to parse, you seem to be assuming a fixed supply of “dollarsâ€â€“but there is no fixed supply of anything. Dollar deposits are certainly not fixed. Base money is not fixed either.”
Dollar deposits and base money remain fixed until the government (Fed/Treasury) changes them.
The non-government sector is a closed system. At any point in time the number of dollars and/or treasuries that exist within it is finite. It is limited to what the government has created thru that point in time. By definition in a closed system the quantity (number) of those assets cannot be changed by any event occurring within the system. The assets can only be re-distributed. The non-government sector is not allowed to create money. So unless there is an injection from an exogenous source (the Fed/Treasury) or a leakage (tax imposed) the quantities cannot change.
This is not controversial. The misunderstanding is typical of conventional economic thinking.
Now, if the PTB decide that the private sector should be allowed to create state money then we have a different system. But the arithmetic will be the same – money will have to be created out of thin air, There just won’t be any net financial assets for the economy as a whole – in order for someone to win someone else will have to lose (within the system).
“…you’re not describing the monetary system we actually have.”
It is an apt description of the monetary system we have – institutional arrangements can make money creation more difficult or stop it altogether but money creation occurs outside of the private economy. The pool can’t fill itself.
I should amend the above…
Dollars available for dollar deposits and base money remain fixed until the government (Fed/Treasury) changes them.
paulie46,
Dollar deposits and base money remain fixed until the government (Fed/Treasury) changes them.
This is not true, and not something that I’ve ever heard an MMTer say.
The Fed controls only the supply of base money–in practice this supply is endogenously determined by public demand for cash and bank demand for reserves at a given interest rate target.
The supply of bank deposits is wholly endogenous and generated by the private sector not by the Fed. “Endogenous money” is a feature of MMT, and one I happen to agree with.
The non-government sector is a closed system.
What do you mean by “closed system”?
At any point in time the number of dollars and/or treasuries that exist within it is finite.
That’s obvious though–right? What point are you making here?
By definition in a closed system the quantity (number) of those assets cannot be changed by any event occurring within the system.
From basic accounting and the way we’ve partitioned the economy, the non-govt sector can only acquire net financial assets by lending to the govt sector.
The assets can only be re-distributed. The non-government sector is not allowed to create money.
The non-government sector is not allowed to create base money. The other types of money, it has no trouble creating, which is why the vast majority of the money stock is non-government money.
So unless there is an injection from an exogenous source (the Fed/Treasury) or a leakage (tax imposed) the quantities cannot change.
You’re confusing money with net financial assets. Quantities of private sector or “inside” money can take on any values, within reason. What cannot happen is the non-government sector increasing or decreasing its nominal claims on the government (i.e. “NFA”) without the government running either a deficit or surplus respectively.
This is not controversial.
The treasury can supply “net financial assets” to the non-government by running deficits. The central bank can supply base money to the non-government sector by swapping some of that stock of NFA for currency and reserves. The non-government sector can create money and financial assets all on its own. What it cannot do is create base money or new NET financial assets. NFA != base money. Base money != bank deposits. Bank deposits != NFA.
money creation occurs outside of the private economy
Nope–“outside money” creation (i.e., the creation of base money) occurs outside of the private economy; “inside money” creation (i.e., the creation of deposit money, etc) occurs inside the private economy. (All MMT’ers recognise this fact, AFAIK–e.g., google “MMT horizontal money”). I have no idea how you square this notion with the existence of banks. Surely you must have had a bank account at some point. Was it with a government agency or a private business?
“From basic accounting and the way we’ve partitioned the economy, the non-govt sector can only acquire net financial assets by lending to the govt sector.”
This is where you create a perpetual-motion machine
“inside money†creation (i.e., the creation of deposit money, etc) occurs inside the private economy
You are confusing financial assets with NET financial assets.
Credit money (a financial asset) comes attached with an offsetting liability – i.e an NFA is split into money and anti-money and freed to go about their merry way separately. The anti-money (liability) must eventually be extinguished satisfying the borrowers liability to the bank and simultaneously the bank’s liability to the Fed.
No NFA’s were harmed in this operation.
“You’re confusing money with net financial assets.”
My comments have been intended exclusively re the accounting of NFA’s within a closed system. “Money” is a much more broad definition and can be conflated with many things, depending on your definition. I got sloppy.
Seems like the fundamental disagreements here are definitional so we haven’t actually gotten anywhere, while simultaneously hijacking the thread. I take responsibility for improperly using the term “money” and the thread hijack and respectfully bow out.
[…] I’ve been coming to the same conclusions about NGDP and velocity. Cf. the subtitle to this […]
[…] As we saw quite clearly in The Great Keen-Krugman Debate, Paul is a firm believer in that model (and obviously Hempton is too). I and many others (notably Keen) have suggested that the model is ridiculous and nonsensical on its face. […]
@vimothy
vim, first “funds” does not mean “resources.” If he meant “resources” he should have said “resources” and distinguished between financial and non-financial resources and explained exactly what financial resources he means.
Secondly, it is simple algebra to show in a close economy of two sectors, households and firms, that saving = investment. So? That says nothing of interest about a more complex open economy.
Loanable funds means what it says. A stock of money available for saving or investments in which interest rates determine the balance. That is simply not the case in an open modern monetary economy.
Loanable funds and S = I are nice simple models for Econ 101. Students should be told clearly that this model does not apply to the real world. It’s just an aid to thinking about how various economic variables function in an imaginary world that is purposefully simplified. If Mankiw or any other textbook writer does that, fine.
@vimothy
“From basic accounting and the way we’ve partitioned the economy, the non-govt sector can only acquire net financial assets by lending to the govt sector.”
Let’s see the accounting. How do you get net positive out of a system in which all DBA entries MUST net to zero?
[…] goes right back to the crazy “loanable funds” notion, that savers “fund” borrowers, and that people spending less (saving more) means other […]
[…] goes right back to the crazy “loanable funds” notion, that savers “fund” borrowers, and that people spending less (saving more) means other […]