Thursday, January 7, 2016

Steve Keen — Note To Joe Stiglitz: Banks Originate, Not Intermediate, And That’s Why Aggregate Demand Is Stuffed


Steve Keen schools Joe Stiglitz on how banks work.
Joe correctly notes that “the world faces a deficiency of aggregate demand”, and attributes this to both “growing inequality and a mindless wave of fiscal austerity”, neither of which I dispute. But then he adds that part of the problem is that “our banks … are not fit to fulfill their purpose” because “they have failed in their essential function of intermediation”….
I’m the last one to defend banks, but here Joe is quite wrong: the banks have very good reasons not to “fulfil their purpose” today, because that purpose is not what Joe thinks it is. Banks don’t “intermediate loans”, they “originate loans”, and they have every reason not to originate right now.
In effect, Joe is complaining that banks aren’t doing what economics textbooks say they should do. But those textbooks are profoundly wrong about the actual functioning of banks, and until the economics profession gets its head around this and why it matters, then the economy will be stuck in the Great Malaise that Joe is hoping to lift us out of.…
Not all economists are operating under wrong model bias, however.
For decades now, a handful of rebel economists have been disputing this—including me of course, but going back to Irving Fisher and even earlier, and including modern non-mainstream economists like Stephanie Kelton (who now advises Bernie Sanders), and University of Southampton Professor Richard Werner. Oh, and a guy named Hyman Minsky too, whom the mainstream ignored until the 2008 crisis. But the mainstream ignored us before the crisis, and continues to ignore us after it, because their “banks as intermediaries” model tells them that we are just spouting nonsense.
We’re not, of course: the ordinary public tends to get that, and even The Bank of England has come out and said that it’s the mainstream that is spouting nonsense, not the rebels. But the mainstream rejects our analysis out of hand, because their model tells them that it’s OK to do so.
This wouldn’t matter if we could ignore the mainstream of the economics profession, but we can’t, because they are the key individuals who influence the economic policies that are actually put in place by politicians.
Forbes
Note To Joe Stiglitz: Banks Originate, Not Intermediate, And That’s Why Aggregate Demand Is Stuffed
Steve Keen | Professor and Head Of School Of Economics, History & Politics, Kingston University, London

30 comments:

Ralph Musgrave said...

If banks don’t intermediate, i.e. if they don’t need savers’ money before lending, why have banks dished out hundreds of billions over the decades in the form of interest and dividends to savers (shareholders, bondholders, depositors etc) in order to attract savers’ money?

If a bank simply “originates”, i.e. creates money out of thin air and lends it out, it will run out of reserves. Each bank CAN ORIGINATE or do the latter “thin air” trick as long as it doesn’t engage in that practice to a significantly greater extent than other banks. However, the bulk of money for loans in any one year has to come from savers, i.e. from intermediation.

NeilW said...

"why have banks dished out hundreds of billions over the decades in the form of interest and dividends to savers (shareholders, bondholders, depositors etc) in order to attract savers’ money? "

Because marketing databases make money. Same reason that Facebook and Twitter 'give' their product away.

And deposits are cheaper capital than bonds or equity.

Essentially the reason deposits sit at the operating margin less than the lending rate is a dynamic function. Without that one of the banks could suck all the reserves to itself by offering a sliver over zero as interest and then make a killing lending the reserves back to the other banks. So what happens is that deposit rates rise until the 'lender of last resort' price neutralises the dynamic.

It would also be eliminated if the central bank simply ran an overdraft for the regulated banks and let some form of National Savings drain customer deposits to safety.

So the main reason that savers get any interest at all - and why loans are therefore more expensive than perhaps they should be - is because the central bank sets an interest rate and maintains its discount window too tight.

As ever the central bank sets the price of money.

Ralph Musgrave said...

"Because marketing databases make money"?? So why did banks hand out interest to savers in the days before computers and marketing databases, e.g. 100 years ago? And is a bank really going to hand out tens of thousands worth of interest / dividends to BIG shareholders or depositors so as to have them on a data base? I mean who ever banks for General Motors or Exxon and pays out millions by way of interest to them doesn't need a "database" to inform it of the existence of GM or Exxon.

Next two paras starting "Essentially" and "It would" are meaningless far as I'm concerned, but if anyone can explain what they mean, I'd be grateful.

Para starting "So the main..", I agree that most of the time CBs artificially inflate the rate of interest. I.e. I agree with Warren Mosler's claim that rates should stay at zero permanently.

NeilW said...

"Next two paras starting "Essentially" and "It would" are meaningless far as I'm concerned"

That's because you can't understand dynamic processes Ralph. If you did you'd know that full reserve is a fool's errand. But it's not uncommon for people to struggle to understand dynamics. In fact it is quite common. The majority of people do static analysis - including the vast majority of people who call themselves economists.

I may get around to creating a simulation at some point that demonstrates the processes in action. It's generally easier to see things when they are visual and are in motion and of course modern computers make that easy. It's just finding the time.

Matt Franko said...

Ralph maybe this way...

Lets look at any business (in financial terms) as "building a balance sheet"...

So in this case a bank...

Banks are licensed to credit accounts at a multiple of capital... so think of a balance sheet... you have assets (loans for a bank) on the left and liabilities and capital on the right..

So lets say we get 10M together and form a bank ok we have 10M of capital on the right... we are allowed to loan at 13x so then we go out and make 130M of home equity loans... ok lets say we stop right there and examine the balance sheet: we have 130M on the left (loan assets) and 10M of capital on the right... BUT our balance sheet doesnt balance...

Soooo we have to figure out a way to establish liabilities on the right of 120M so our balance sheet can balance.. what is the least expensive way to for us to establish these liabilities if we are a bank? Deposits.

Hence you see banks advertising for deposits because they look at deposits as least cost way of establishing the necessary liabilities so their balance sheet will balance... otherwise they may have to issue bonds or other preferred securities, etc that have a higher rate of interest on the liabilities and we would make "less money!"....

Matt Franko said...

iow the choice is either pay for advertising to obtain deposits or pay a bunch of lawyers to write up some securities to issue...

Many banks think the advertising route is cheaper/easier....

NeilW said...

Matt,

the problem with that is that it doesn't explain the 'what else are they going to do with bank deposits other than keep them in a bank' issue.

The actual price of deposits is zero because every loan creates a deposit and every piece of capital and equity used to be a deposit. So why does the price rise above zero? It can't all be explain by people walking around with suitcases full of cash - since that doesn't happen.

There is no static reason. You have to go to dynamics to see the reason. It's the reason that base rates have any effect at all.

Matt Franko said...

Stiglitz (I assume Jewish) is just applying his textbook OT Jewish Temple 101 operations of "borrow > default > forgive" vice what we of the nations (are at least supposed to) use which is "issue > redeem" or as Keen here can see it "originate > redeem"...

eg. Siglitz was advocating "debt forgiveness" for Greece last year... then its right back to "borrow > default > yada > yada >...."

the Lord to the "borrow > default > forgive" house of Israel: "The poor you will always have..." Mat 26:11

we will always have the "poor" (meaning some among us lacking means of subsistence) as long as policy continues to be established within this OT "borrow > default > forgive" paradigm..

No JG or BIG or whatever policy we may choose to eliminate those made "poor" among us ("the poor we will always have") as long as this OT "borrow > default > forgive" paradigm remains dominant and operative within TPTB...



Matt Franko said...

Neil I agree with you wrt the systemic view... Ralph just tweeted out "why would a bank be seen advertising for deposits?" and I needed more than 140 characters...

iow A bank, ie a single bank within the overall system could choose to not pursue deposits and instead mainly use the issuance of securities to establish its liabilities... deposits would be left mainly to go to other banks in the overall system who preferred deposits and all the retail ops that goes with that...

NeilW said...

"ie a single bank within the overall system could choose to not pursue deposits and instead mainly use the issuance of securities to establish its liabilities"

Even that is funny because you're still processing deposits. You receive the reserves and deposits from elsewhere, and then delete the deposit to replace it with a security.

The reserves are then depleted when the deposit created by the loan is transferred elsewhere.

The funding cycle is then about maintaining the reserves buffer so you have the lowest cost liquidity.

Again not something you can have a handle on unless you view it dynamically.

Tom Hickey said...

Banks are licensed to credit accounts at a multiple of capital... so think of a balance sheet... you have assets (loans for a bank) on the left and liabilities and capital on the right..

So lets say we get 10M together and form a bank ok we have 10M of capital on the right... we are allowed to loan at 13x so then we go out and make 130M of home equity loans... ok lets say we stop right there and examine the balance sheet: we have 130M on the left (loan assets) and 10M of capital on the right... BUT our balance sheet doesnt balance...


Matt, the deposits that the loans created are recorded as liabilities. The books do balance and if all transactions occur intrabank, No problem. Those liabilities just get shifted around into different customer accounts but bank liabilities are unchanged.

This is true of the banking system as a whole.

The issues arise owing to interbank transactions and that is where the situation becomes dynamic systemically. That is, it becomes a mutual "balancing act." The purpose of a CB is to ensure that the system stays in balance. Otherwise, banks may have to raise capital on an emergency basis as happen in 1907 when JP Morgan himself got the bankers to set up to the place. That led to the FRA of 1913.

Matt Franko said...


Tom yes I agree system wide they balance .... but Ralph was asking why a bank would advertise for deposits ..... they need the lowest cost liabilities of some sort (which might be deposits for a specific bank) was my answer...

These bank people would probably want to do away with deposits entirely as long as they could still be allowed to leverage 13x... my bank doesnt even give out change for the parking meters unless you have an account there... so the people who need to park here have to go in to all the merchants here for change for the parking meters.... they suck....

Ralph Musgrave said...

Neil,

If full reserve is a "fool's errand" how come at least four economics Nobel laureates also favored full reserve (Milton Friedman, Merton Miller and Maurice Allais, and James Tobin). Are they all fools?

Matt,

Banks do not need to “find” something to put on the liability side of the balance sheets in order to balance loans. Reason is that when a loan of $X is made something automatically goes on EACH SIDE of the balance sheet: the loan on the asset side and the money granted to the borrower on the other side.

The reason banks have to get hold of money to (approximately) balance loans is that (as I said above) they run out of reserves otherwise. When a bank makes an $Y loan, at least 90% of that on average gets deposited on OTHER banks, and those banks will want genuine base money / Fed issued money / reserves in settlement.

They can of course borrow reserves, either from the central bank or from other banks, but no bank if it wants to stay afloat wants to go too far down that path. Access to those loans may dry up, as they did with Northern Rock.


Random said...

"If full reserve is a "fool's errand" how come at least four economics Nobel laureates also favored full reserve (Milton Friedman, Merton Miller and Maurice Allais, and James Tobin). Are they all fools?"

Correct. They are "fools" on this issue. An appeal to authority is no argument.

Matt Franko said...

Ralph a bank probably doesnt even have to make a loan...

You could probably get 10M together and go out and buy 130M of high yield bonds instead of making loans.... offset the bond assets with some bonds you issue yourself that yield much less than the high yield you hold... collect the NIM.... no deposits other than those in transit...

Matt Franko said...

Ralph here:

http://www.federalreserve.gov/releases/h8/current/

over 25% of US bank credit is used to just simply buy securities (over $3T..)

go out and issue a bunch of corporate bonds at 2% and buy MBS at 4% leveraged 13x, make 26% gross NIM.... no deposits required...

Or seek deposits at 1% instead of issuing bonds at 2% and make 39% gross NIM...

Good work if you can get it....

MRW said...

Ralph,

You wrote:

"When a bank makes an $Y loan, at least 90% of that on average gets deposited on OTHER banks, and those banks will want genuine base money / Fed issued money / reserves in settlement."

When a customer at Bank A transfers his deposit or cash at Bank B, the reserves go with it. Automatically. I don't have a link, but this is standard bank practice. (Scott Fullwiler also wrote about this...again, no link.) If Bank A doesn't have enough reserves to cover the transfer, then it has to go buy some.

Bank settle their reserve requirements on loans with the Fed a few weeks after the loan is made. It's not instant.

Tom Hickey said...

A great deal of interbank settlement now occurs through clearinghouses rather than at the central bank.

Reserves are only needed to settle after netting.

As lender of last resort, the cb stands ready to ensure that all accounts settle in the payments system but generally banks needing reserve balances at the close of period borrow at the policy rate from banks that have excess reserve balances or repo in the money market or at the cb in order to avoid the penalty rate the cb levies at the discount window on banks that come up short on balances at the end of the period.

MRW said...

Ralph, it only costs a bank $2 in reserves to buy a $100 treasury security. No link, but it's in my library somewhere.

NeilW said...

"Are they all fools?"

Yes. Because of course the 'Nobel' is a chocolate nobel and worth about as much as the chocolate coins my children like.

Services to Swedish neoliberal central banking is not something that should be celebrated.

You really should stop doing this appeal to authority business. You should realise by now I don't accept that authority.

Their beliefs are simply wrong systemically. All those people still believe banks are intermediaries. They are not as Professor Richard Werner has actually proved doing science. (Did you see what I did there?)

NeilW said...

"When a bank makes an $Y loan, at least 90% of that on average gets deposited on OTHER banks, "

You made that up didn't you. And it doesn't actually matter to the way the system works.

"and those banks will want genuine base money / Fed issued money / reserves in settlement."

Nope. Dynamically incorrect again.

Without a central bank, the target bank must take the place of the transferring depositor in the source bank *or the transfer will never take place* (clearing fails due to insufficient liquidity). That is the target bank lending to the source bank in return for getting the deposit transferred.

With a central bank all that happens is that the central bank acts as a clearinghouse for the transactions. The net result at the end of the day is that some target bank ends up lending to the source bank - simply mirroring and balancing where customers have transferred their deposits during the day. The central bank essentially tries to get itself out of the loop at the end of every day. In other words all the central bank has done is inject intraday liquidity into the market to ensure that the market clears properly.

Other than that the CB operates like all other central clearinghouses. It reduces the overall liquidity the members require to get to a cleared position since transactions can be 'netted off' at the end of the day.

Positive central bank reserves held by banks at the end of the day clearing are generally a result of net spending into the system by the government or regulatory requirements (Reserves are essentially the banks being forced to lend to the government sector).

What happens when the market gets nervous of bank asset losses and the end of day bank-to-bank lending starts to break down is that the central bank steps in as 'lender of last resort' (yep that's where the term comes from) and starts lending to banks that are solvent but illiquid while accepting the reserve deposits of banks that are nervous of lending in the interbank market. It then imposes losses on both sides so that they come to their senses.

MRW said...

Ralph, this may seem wonkish but Fullwiler writes complicated well. He's clear.

Modern Central Bank Operations—The General Principles
Scott T. Fullwiler
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1658232

Fullwiler lists 10 principles. Principle 1: Reserve balances are held for only two purposes: payment settlement and (where applicable) meeting reserve requirements. Reserve balances do not “fund” loans or otherwise aid the creation of outside money.

Nick Edmonds said...

http://www.concertedaction.com/2016/01/07/the-phrase-financial-intermediary-in-national-accounts/

I would add that I don't think there's anything in Stiglitz's article to suggest that he doesn't know that banks originate.

The actions of banks matter. Understanding that they are intermediaries is rather essential to understanding how and why their actions matter.

Matt Franko said...

Nick here from the Stiglitz link:

"Between long-term savers (for example, sovereign wealth funds and those saving for retirement) and long-term investment in infrastructure stands our short-sighted and dysfunctional financial sector."

So he sees savers as the "lenders" and the investors as the "borrowers" with the banks as intermediary between these 2.... he's asserting causation between the 2 terms in the S=I accounting identity... this is like saying "a balance sheet is the intermediary between assets and liabilities/capital..." or something like that... doesnt make any functional sense...

Nick Edmonds said...

Matt,

In balance sheet terms, banks are indeed between those "savers" and those "investors". The savers hold claims on the banks; the banks hold claims on the investors. Stiglitz is saying this matters because it means that how banks behave effects the terms on which funds are provided to the investors. I certainly don't think he suggesting that S causes I.

Tom Hickey said...

Fullwiler lists 10 principles. Principle 1: Reserve balances are held for only two purposes: payment settlement and (where applicable) meeting reserve requirements. Reserve balances do not “fund” loans or otherwise aid the creation of outside money.

This is why I prefer "settlement balances" to "reserve balances." Rectification of names aka "calling a spade a spade."

Tom Hickey said...

@ Nick Edmonds

Ramanan agrees.

The Phrase “Financial Intermediary” In National Accounts

Calgacus argues similarly for "printing money."

Tom Hickey said...

My point above, if it’s not clear, is that we can get hung up on the letter and miss the spirit. Meaning is determined by context. The same term can have different meanings in different contexts. There is no harm in heterodoxy using customary terms while clarifying them.

I don't think that heterodox economists have to revise the vocabulary of economics, and it may be unwise to make the attempt in many cases. The object is to set the record straight by providing understanding of proper use versus misuse.

Matt Franko said...

Ramanan puts too much stock in the NIA schemes... accountants arent the ones who run things... they keep track of things...

Stiglitz is obviously saying much of the problems we are having is due to "banks not doing their job".... and to him "their job" is to foment loans as an intermediary between unrelated/unaffiliated savers and investors...

Nick loans dont happen just because "the money is there".... the borrowers need to identify increased revenue flow into their business/entity that are more than adequate to pay the new principle and interest in order to increase their credit lines...

Banks loan against a stock (their capital) BUT the criteria they use to determine whether to do the loan is based on flows... no increased flows then no increased credit lines...

Were in some pretty serious austerity these days with barely positive leading USD flow from the govt sector... so of course banks are not going to increase lending very much there is very little increase in leading USD flow from the govt sector...

Stiglitz looks at fiscal and bank lending as "two equals" here... they are NOT equal under a numismatic system like we are running currently where the govt "issues" or "originates" all USDs either directly or via their subordinate fiscal agents at the banks...

Stiglitz and Chritendumb right along shoulder to shoulder with him is stuck in the Old Testament paradigm of "borrow > default > forgive" ... today we are running under a New Testament paradigm of "issue > redeem"...

Nick Edmonds said...

Tom - I'd generally agree. We can have much more productive discussions if we stick to common terminology and there's no reason it should prevent us making the points we want. Calling something an intermediary, doesn't imply anything about how active or passive it is.

Matt - I really don't think Stiglitz is saying that banks are no more than brokers and I'd be surprised if that was his view, but maybe we're just interpreting his words differently.