Wednesday, October 26, 2011

Steve Keen on “Capital Account”

A nice interview with Steve Keen on Russia Today.

10 comments:

  1. LK, do you regularly watch RT in real life? Forgive any potential lack of awareness, but I've yet to see Steve Keen appear on other channels besides this one...

    ReplyDelete
  2. So, he did not explain how the growth of debt in itself is a drag on demand, but he briefly touched on why I think it is - the fact that propensities to consume of creditors and debtors are vastly different! Isn't this the key point? If it weren't so, then Krugmans of the world would have a point in dismissing the "balance sheet recession" thesis by saying that for each creditor there is a debtor. But the fact that creditors are not as likely to spend their income (or are not in the business of spending - like banks) is what causes the level of demand to drop. Or am I missing some other argument that Keen has in mind?

    ReplyDelete
  3. "So, he did not explain how the growth of debt in itself is a drag on demand"

    (1) The growth of private debt is a boost to aggregate demand (AD) during the bubble, but the fall in net annual new private debt (a flow) after the bubble pops is what causes a contraction in AD.

    (2) Paying down the principal of a debt destroys money supply and subtracts from AD. The shift from money spent on producible commodities to debt principal repayment is a contraction of AD.

    (3) of course, many rich creditors/investors are unlikely to spend much of their money on producible commodities. Mostly their money will be recycled through secondary financial asset markets.

    ReplyDelete
  4. Peter,

    Keen shows mathematically that to increase the level of aggregate demand in our economy, private debt has to accelerate:

    http://www.debtdeflation.com/blogs/2010/09/20/deleveraging-with-a-twist/

    He defines AD as the sum of GDP plus the change in debt. Flick through his posts and you will find info.

    ReplyDelete
  5. LK,

    >>(1) The growth of private debt is a boost to aggregate demand (AD) during the bubble, but the fall in net annual new private debt (a flow) after the bubble pops is what causes a contraction in AD.

    Yes, the question is why. We know that banks create loans out of thin air, and those loans create deposits. When the loan is repaid to the bank, the bank now has "funds" but the bank is not in the business of spending, so, this particular creditor (the bank) getting the funds is not going to boost AD, which was my point.
    Now, I know that repayment of loans actually destroys "money", but we can still treat it as "funds" coming back to the bank. Why? Because the banking system in aggregate creates loans as a multiple of its capital, so, when the loan is repaid a portion of the capital is "freed" - but the banks are not in the business of spending their capital!
    Am I correct here?

    >>of course, many rich creditors/investors are unlikely to spend much of their money on producible commodities.

    Exactly! For those rich persons, their MPC is low compared to debtors, while banks' MPC is zero. Which was my point.

    Cahal:
    >>Keen shows mathematically that to increase the level of aggregate demand in our economy, private debt has to accelerate

    I guess that's because his definition of AD includes the derivative term (debt velocity and/or acceleration). But what is the underlying mechanism on actual demand of consumption/investment items? I still think it is the MPC of creditors/debtors, but would love to have a short explanation/link as to why there is something else I might be missing.

    Now, this is another question I have about something from Cahal's link. Keen says: "With the debt to GDP levels for all non-government sectors of the American economy at unprecedented levels, the prospect that any sector can be enticed to take on yet more debt is remote. Deleveraging is America’s future."

    That's interesting. I assumed that some sectors might have been in cumulative surplus and thus net lenders. Is it the foreign sector that is the net lender then? Otherwise we'd have some sectors as net lenders, some as net borrowers and then the differential of (cum. surplus - debt) would be what should matter after netting the debt repayment between the sectors, not the absolute level of debt: suppose that two sectors A and B each have debt of $100 but they owe it to each other. In this case their debt payments cancel each other and do not present any drag on their income.

    ReplyDelete
  6. "Now, I know that repayment of loans actually destroys "money", but we can still treat it as "funds" coming back to the bank. Why? Because the banking system in aggregate creates loans as a multiple of its capital, so, when the loan is repaid a portion of the capital is "freed" - but the banks are not in the business of spending their capital!"

    It's fairly easy to get your head around this if you think of banks literally making credit in a room somewhere. That then gives you the concept of a 'credit stock' - essentially an inventory of potential loans.

    Credit stock is the difference between the maximum amount of loans the bank could issue under the extant regulations and the loans currently in issue.

    So when a bank charges interest it takes some money out of the credit stock and allocates it to a loan. However since the bank equity has also increased by the same amount (and loans allowed are usually a multiple of capital) the credit stock goes back up again.

    Which explains how bank earnings increase loans available and bank losses (the reverse of the process above) reduce the capacity of banks to make loans.

    Once you have credit stock, then repaying loans doesn't 'destroy' money. It just moves it from the loan to the bank's credit stock. Similarly making loans just moves the money from 'credit stock' to 'loans outstanding'

    A bank with a large remaining credit stock is going to be more hungry for loan making business than one with a small remaining credit stock. So it explains the incentive of the banks to lend as well.

    I think the 'credit stock' idea reconciles the 'repaying destroys money' idea (which is more accurately 'repaying destroys money in circulation') with Steve Keen's notion of some sort of relending process.

    I'm still fleshing the idea out, so apologies if the above isn't that clear.

    ReplyDelete
  7. Neil, thanks.

    >>So when a bank charges interest it takes some money out of the credit stock and allocates it to a loan. However since the bank equity has also increased by the same amount (and loans allowed are usually a multiple of capital) the credit stock goes back up again.

    I don’t entirely follow. Do you mean at the moment of making the loan? Because equity does not go up with the act of making the loan. I guess you mean when the bank gets interest payment, its equity increases, but then I fail to see how this is supposed to be “taking some money out of the credit stock and allocating it to a loan”? Isn’t it the other way around – it actually increases the credit stock?
    (BTW, a beautiful tool for visualing the balance sheet operations involved, for anybody interested: http://econviz.com/balance-sheet-visualizer.html# )

    >>Which explains how bank earnings increase loans available and bank losses (the reverse of the process above) reduce the capacity of banks to make loans.

    I think I follow that: the credit stock is a multiple of banks capital, where the multiple goes down as more and more loans are made.

    >>Once you have credit stock, then repaying loans doesn't 'destroy' money. It just moves it from the loan to the bank's credit stock.

    Yes, because banks’ capital is unaffected.

    >>I think the 'credit stock' idea reconciles the 'repaying destroys money' idea (which is more accurately 'repaying destroys money in circulation') with Steve Keen's notion of some sort of relending process.

    Cool, thanks!

    ReplyDelete
  8. Also, any advise on the net debtor/net creditor position of the sectors in the US economy? And am I right that it is not the total amount of debt but how it is distributed that matters? In that if there are sectors that are in debt to each other then what matters is the net position?

    ReplyDelete
  9. Peter,

    It needs a picture really, but the charging of interest by a bank on an existing loan just increases the size of the loan. So there is a balance sheet transfer from credit stock to loans outstanding and an equivalent transfer from credit reserves to equity.

    You can see this as banks paying themselves with their own money. Which in timing terms means the banks are able to earn the money (and deploy it in extra loans) before the borrowing firm has earned the money to pay the interest.

    That has the potential to have fascinating dynamics and it isn't in Steve Keen's model.

    I have plans to write this up, so again apologies if it is unclear.

    ReplyDelete