Saturday, January 28, 2012

Steve Keen — Economics in the Age of Deleveraging


Non-economists might expect professional economists to pay great heed to these indicators—after all, surely private debt affects the economy? However, the dominant approach to economics—known as “Neoclassical Economics” —ignores them completely, on the a priori grounds that the aggregate level of private debt doesn’t matter: only its distribution can have macroeconomic impacts.
The argument is that a rise in debt merely indicates a transfer of spending power from a saver to a borrower. The debtor’s spending power rises, but saver’s spending power also falls, so in the aggregate there will only be a macroeconomic effect if there is a very large difference in behaviour between the saver and borrower.
Therefore only the distribution of debt matters, not its level or rate of change.
US Federal Reserve Chairman Ben Bernanke provided precisely this rationale to explain why neoclassical economists ignored Irving Fisher’s “debt-deflation” explanation of the Great Depression (Fisher 1933), and he also asserted that the differences in behaviour between saver and borrower could not be large enough to explain the Great Depression....
Not only New Classicals think this, but also New Keynesians.
Similarly, Nobel Prize winner Paul Krugman argued recently that the aggregate level of private debt was not a factor in the GFC: only its distribution could be. He therefore developed a model in which the distribution of debt, rather than its level, was the causal factor...
Steve shows goes on to show why this is wrong thinking.

Read it at DebtWatch
By Steve Keen

29 comments:

Joe said...

From reading Keen's work, I get that banks don't lend your deposit to someone else. When you get a loan, that credit money is created on the spot and reserves are only use to settle balances. So what limits how much credit a bank can create?

Clonal said...

The bank is limited not by reserve requirements, but by capital requirements. Losses come out of the capitalization. In other words, if there are large losses, the bank has to look for new investment, but who will invest in a loss making bank?

Tom Hickey said...

Bank lending is constrained by capital requirements and risk. When banks lend, they put capital at risk.

Ryan Harris said...

The corollary is that savings don't necessarily get transmitted to another participant in the economy. Loans and Savings aren't coupled in the same way they were when reserve requirements limited lending. Plentiful excess reserves limit opportunities for interbank lending. That has broad implications that weren't taught in econ 101.

Matt Franko said...

Joe,

this is probably over simplifying (you have to find JKH in the heterodox blogosphere if you want to go deep):

Think of the banks balance sheet.

Assets on the Left, Capital and Liabilities on the right.

Bank owners say put in 10M, that is the banks capital and is on the right side.

Regulators dictate then the ratio of that capital to the assets a bank can accumulate on the left side of the BS. Say that reg ratio is 10:1.

That means our bank can position assets of 100M on the left side of the BS. To a bank, loans are assets. so our bank can make loans up to 100M. (This is why it is good to be a banker; you only put in 10M and you can control 100M simply by computer key strokes).

What is missing at this point is the 90M of liabilities needed to make your balance sheet "balance".

The liabilites are the deposits you have to attract. So you can build your banks balance sheet by putting in 10M capital, writing 100M of loans, and then attracting 90M of deposits.

So you put in 10. Then write 100M of loans at 5%, this gives you 5M gross interest revenues. maybe today you pay 1% in deposit interest, this costs you 1M. So you net 900k in net interest. 4.1M return on only 10M invested is a 41% return on capital invested, not bad.

Resp,

PS If I ever hit the Mega Millions lottery for 200M, the first thing I'm going to do the next day is apply for a bank license ;)

Matt Franko said...

Joe I screwed the last paragraph up. here is the last paragraph corrected:

should be:

"So you put in 10. Then write 100M of loans at 5%, this gives you 5M gross interest revenues. maybe today you pay 1% in deposit interest, this costs you 900k (1% on 90M deposits). So you net 4.1M in net interest. 4.1M return on only 10M invested is a 41% return on capital invested, not bad."

Sorry, Rsp

wh10 said...

Matt, if the bank wrote $100M in loans, that means there are $100 in deposits. So the 1% interest is paid on $100M in deposits, which costs $1M. The net is $4M.

The invested capital is on the right side as equity and on the left side as something or other. Let's say it's cash. But it's double entry, so it has to balance.

So LHS = 10M (cash) + 100M (deposits)
RHS = 100M (loans)

Equity = Assets - Liabilities = 110 - 100 = 10

wh10 said...

Sorry- that was a bit confusing. Capital = assets - liabilities. The banker owners invest capital, which is an asset.

wh10 said...

So I should have written


Assets = 10M (cash) + 100M (deposits)
Liabilities = 100M (loans)
Equity = 10M (owner equity)

LHS = Assets
RHS = L + E

Unforgiven said...

So, are reserves just a stopgap till banks can attract enough deposits?

wh10 said...

Unforgiven, deposits create reserves. The money a bank receives from a deposit becomes reserves. Acquiring reserves by attracting depositors, though, may be cheaper than having to borrow reserves in the interbank market, so that's the motivation to attract depositors.

Matt Franko said...

wh10,

We're getting deeper into it here and I can quickly get in over my head...

Youre probably right when you say the 100M loans create 100M deposits. The balances that the original investors put in I believe you also correct show up on both sides, but I believe it is capital on the Right, perhaps it is "cash" aka RBs on the left.

But I believe in bank accounting, LOANS are assets and DEPOSITS are liabilities. reserves are on the left (assets?), and capital is on the right.

At this point there are better people to explain this than I....

Resp,

Clonal said...

Matt,

When a bank makes you a loan of $100K, they normally don't hand you the loan amount in "dollar bills" They open an account for you, and credit that account with the money. So the bank creates an asset (the loan) with a value of $100K and at the same time they create a liability (the bank deposit) with a value od $100K. As the bank deposit get drawn down, and they run short on the reserve requirement, they either borrow at the interbank rate, or they try and attract depositors. When the Fed is running a ZIRP policy, the bank has no need for deposits.

Matt Franko said...

Heres a link to BofAs BS, go to page 33

http://thomson.mobular.net/thomson/7/3171/4426/

Resp,

Clonal said...

I should have said "The bank has no need for interest seeking depositors"

Matt Franko said...

Clonal,

agree/correct the two processes, loan origination and attracting deposits appear to happen simultaneously.

But I believe JKH has opined that these functions are somewhat not directly related to each other...

like a Loan Officer wouldnt call a compliance manager to find out if the bank had "room" to make new loans. If a loan has been made I'd assume others in compliance dept would take note of it and perhaps seek to attract addl deposits if necessary.

Resp

Matt Franko said...

Clonal,

"When the Fed is running a ZIRP policy, the bank has no need for deposits."


Then how could a bank build the right side of its BS if there were no deposits which make up liabilities?

Resp,

Clonal said...

Matt,

The bank is borrowing money and that goes on the liabilities side of the balance sheet.

When you deposit money in your bank account, you are loaning the bank money (whether or not the bank pays any interest on it)

So the bank has the choice of going to a depositor or th the Fed for making sure that it is meeting its reserve requirement. At ZIRP, going to interbank borrowing is far easier!

Clonal said...

And of course it is much much easier for the bank if the Fed buys up all your bad loans and you don't have to take a loss for your bad decisions -- time to give yourself a big pat on the back and an even larger bonus!

Ralph Musgrave said...

Why do we allow private banks to create credit / money at all? Absent private banks, anyone wanting to invest has to endure the pain of forgone consumption so as to invest. In contrast, private banks create “savings” out of thin air and lend them out. Sounds too good to be true. And it is.

Spending those fake savings raises demand, which means government has to rein in demand so as to avoid inflation – e.g. by raising taxes. In short, private bank credit / money creation is a trick that enables private banks and those they lend to, to rob taxpayers.

geerussell said...

Ralph,

Going back to this discussion I think the case could be made for horizontal money creation as the signaling mechanism from the private sector that it wants/needs more NFA.

The private sector proposes an increase in NFA by creating horizontal money and the government can then "ratify" this by supplying new NFA or it can "veto" and let the debt deflate back down.

This has the benefit of gearing down the very slow, blunt instrument of deficit spending to fine transmission into the specific and timely needs of individual private sector actors.

In the absence of private credit money it's unclear what the signal would be from the private sector to say "Hey, I'm growing here and I need an injection!"

Ralph Musgrave said...

Geerussell, Good question. My answer is that even where there is adequate NFA, private banks will still try to create credit / money, so money creation by private banks is not a good signal that more NFA is needed.

To illustrate, imagine an economy where private banks didn’t exist, but there WAS a supply of money supplied by the central bank, and NFA was adequate. In that scenario, borrowing by one person would only be possible where someone else forgoes consumption. However, in that scenario, a private bank can set up, and supply “fake savings” at a lower rate of interest than “genuine savers” charge.

Coincidentally, I set out the above argument in much more detail in my latest (27th Jan) blog post:

http://ralphanomics.blogspot.com/

As to what the signal should be that more NFA is needed, we already have and employ such a signal: a combination of high unemployment and low inflation is such a signal, and indeed governments / central banks react to that signal by implementing stimulus. They don’t do in the way Abba Lerner advocated, nor in the way most MMTers would like. But never mind – some time in the future they may see the light and just have government net spend new money into the economy as advocated by Abba Lerner when unemployment is hi and inflation is lo.

Matt Franko said...

Clonal:

"So the bank has the choice of going to a depositor or th the Fed for making sure that it is meeting its reserve requirement."

They are not really "borrowing" imo when they take in deposits (just as I view UST issuance as NOT that the US govt is "borrowing"). They are establishing liabilites on their BS. Deposits are not reserves. Reserves are assets, deposits are liabilities, ie on different sides of the bank BS.

Reserve reqs are a function of these deposits. ie Deposits drive req reserve levels. So just for discussions sake, if a bank had no deposits, it would have no reserve reqs. It would not have to borrow FFs to meet reserve reqs. because it has no deposits.

But a bank would still have to figure out what liabilities to establish to build its balance sheet. I guess it could issue corporate bonds.

Even under zirp, it would seem to me that a bank would still want to have deposits as a a low cost way of establishing liabilites on its BS.

Resp,

Clonal said...

Matt,

As I said, the banks borrow to increase their reserves at the central bank. They borrow from the depositors, or they borrow from other banks, including the Fed. When one does double entry accounting, for every asset there is a liability (I am counting shareholder equity as a liability - as it is something that a corporation owes to its shareholders - in other words, a liability is something that the corporation owes - either to shareholders or to non shareholders,) and for every liability there is an asset. The books always have to balance.

When I deposit some money in the bank, The bank owes me what I have deposited when I want it. In the mean time, the bank can do what it wants with my money. It can use it as a reserve (deposit it at the central bank) or do anything else with it. Since the only other thing that a traditional bank does is to loan money, it makes no sense to use deposits for anything other than vault cash, or to meet reserve requirements.

In any corporation, I want to maximize shareholder equity - so why would I take on new liabilities if I don't need to, and if it makes no business sense?

Joe said...

Thanks for the replies! Things are clearer but still a bit confused. Does anyone have any links to a more in depth explanation?

Tom Hickey said...

The quickest way is to keep asking questions.

wh10 said...

Matt - we're in agreement. My first post was poorly worded.

If investors invest 10M into the bank, then there's 10M as cash on the LHS (as an asset) and 10M in capital on the RHS (as equity). However, I do not believe that cash will be reserve balances; RBs come from deposits as far as I understand. The cash however can be used for hiring, capital investment, marketing, etc.

Now have the bank make 100M in loans:

Assets = 10M (cash) + 100M (deposits)
Liabilities = 100M (loans)
Equity = 10M (owner equity)

LHS = Assets
RHS = L + E
110 = 110.

Equity or capital is the residual of assets over liabilities (E = A - L).

And it all balances.

wh10 said...

Gah I keep screwing up. "Loans" should be on the asset side, and "deposits" on the liabilities side, like you said.

Matt Franko said...

wh10,

Hopefully Joe got his question answered! ;)

It's all good wh10... Resp,