George meets Steve Keen: not a pretty sight

The official view, as articulated by Ben Bernanke, chairman of the Federal Reserve, is that both the first Great Depression and the current crisis were caused by a lack of base money. Base money, or M0, is money that the central bank creates. It forms the reserves held by private banks, on the strength of which they issue loans to their clients. This practice is called fractional reserve banking: by issuing amounts of debt several times greater than their reserves, the private banks create money that didn\’t exist before. Conventional economic theory predicts that when the central bank raises M0, this triggers a \”money multiplier\”: private banks generate more credit money (M1, M2 and M3), boosting economic growth and employment.

Bernanke, echoing claims by Milton Friedman, believed that the first Great Depression in the US was propelled by a fall in the supply of M0, which, he said, \”reinforced … declines in the money multiplier\”. But, Keen shows, there is a weak association between M0 supply and depression. There were six occasions after the second world war when M0 supply fell faster than it did in 1928 and 1929. On five of these occasions there was a recession, but nothing resembling the scale of what happened at the end of the 1920s. In some cases unemployment rose when the rate of M0 growth was high and fell when it was low: results that defy Bernanke\’s explanation. Professor Keen argues that it\’s not changes in M0 that drive unemployment, but unemployment that triggers changes in M0: governments issue more cash when the economy runs into trouble.


President Obama justified the bank bailout on the grounds that \”a dollar of capital in a bank can actually result in eight or 10 dollars of loans to families and businesses. So that\’s a multiplier effect.\” But the money multiplier didn\’t happen. The $1.3 trillion that Bernanke injected scarcely raised the amount of money in circulation: the 110% increase in M0 money led not to the 800% or 1,000% increase in M1 money that Obama predicted, but a rise of just 20%. The bail-outs failed because M0 was not the cause of the crisis.

Jeez. It\’s as if MV = PQ never existed, isn\’t it?

Sure there\’s a multiplier, sure there\’s a V, the velocity of circulation.

But V changes in recession. So, in a recession we need a greater increase in M to stop declines in either P or Q.

Keen hasn\’t discovered anything new, he\’s simply ignoring (or ignorant of) the conventional wisdom.

If V were constant and we pumped $1.3 trillion of new Mo into the economy then we\’d have hyperinflation! Because we would have created $8 trillion or whatever of new wider money supply.

As to the rest of Keen\’s thesis, excess debt leveraging, sure, this is pretty much the Austrian view, you know, Hayek, Mises etc. Hardly new.

Instead, Keen says, the key to averting or curtailing a second Great Depression is to reduce the levels of private debt, through a unilateral write-off, or jubilee.

And no, that ain\’t the solution. We want to liquidate the unproductive debts, the stupid stuff, but not the good stuff. So we want bankruptcy of those who cannot pay, not simply the wiping away of all debts.

19 thoughts on “George meets Steve Keen: not a pretty sight”

  1. I’m not sure V does change much in a recession. Friedman didn’t think it changes much, and on that I’m inclined to agree with him. Remember, “V” is the rate of transfer of dollar bills from one person to the next, it is kind of the “transaction rate”. Now think what a dollar does- it gets paid to somebody, some time in the next month they buy something in a shop, it sits in the shop’s bank account then gets paid either to an ordinary shopgirl, or to a wholesaler. It sits in the wholesaler’s account until they pay a worker or supplier, and so on. The average transaction rate for most dollars is (I think) two weeks on a monthly pay economy (most bills and these days wages in the West are monthly). So, how can V change much?

    If anything is changing it is M, not V. But not M0. What happens during the boom is that everyone starts considering debts to be money; they start paying each other with bills for debts, on the assumption that they will be paid back later. Mx starts to rise and, as the boom intensifies, rises a great deal.

    The bust happens when the realisation comes that most of the debts can’t be paid back (e.g. with the house boom) since ultimately the debts have to be paid in M0. (A mortgagee can’t pay her mortgage provider in Mx, only M0). So the notional M collapes from the debt inflated Mx towards the actual money supply, M0, taking down vasty amounts of the Mx supported finance industry. Hilarity ensues.

    Anyway, if you visualise dollarpounds “clocking” through the economy with bill payments, it becomes pretty clear that V can’t change much; unless you go all Weimar and pay everyone twice a day, in which case the entire liquid money supply is exchanging hands twice a day and inflation goes roaring into the stratosphere.

  2. The other problem that I’ve got with all of this excessive money creation through QE, etc. is that, yes at the moment it is going to banks to prop up their balance sheets and act as cover for the stupid debts that they’ve got on thier books.

    At some point they are going to have to destroy this money, preferably by writing down their foreign bonds to reasonable levels.

    If, heaven help us we do start to see some real recovery and that newly created money starts flooding out of our zombie banks and into the real economy then inflation will skyrocket.

    Why don’t these idiot politicians and bankers see that. It would be better to create a ‘bad bank’ with all the toxic debt in it (as well as all of the newly created money) and then balance that off over time to stop it polluting the real economy.

    Thus far we’ve felt the earthquake from the distant eruption, but the tsunami of bad debt still hasn’t hit us yet. When it does, then the real pain will begin.

    I’m heading for the hills…

  3. Hmm. If lots of new M is created and swapped for assets held by investors (QE), but all those investors do is dump the cash into insured deposit accounts and safe haven currencies, V falls off a cliff. So you have a great increase in M but V declines. Result is that P & Q continue to decrease despite the increase in M. Which I think is Keen’s point. The multiplier does not work because banks and investors are not passive agents. They can choose not to lend and not to spend. Which is what they did, and are continuing to do.

  4. But Frances, the money that is exchanged for deposits in a safe haven currency or nationally-backed savings account or other supposedly rock-solid investment still has to go out and earn its keep. And this for me is the problem. If you are going to have a long-term mismatch of production and consumption in one group (too much consumption relative to production in group A for example), and this (arithmetically of course) has to be matched by the opposite effect in group B, you have constant accumulation of debt from people in group A to those in group B.

    Redeeming that debt requires both group A to start producing more, and group B to start consuming the products of group A. The only alternative is forgivement of the debt – a financial transfer from group B to group A. And if group A simply cannot pay then the transfer becomes enforced by reality, whethey group B agrees to it or not.

  5. The fact that “new”(=bogus) money is not getting out of the banks clutches is all that is keeping inflation from really taking off (and it is bad enough now).

    Obummer or KingDick or any of them can create all the currency they like–they cannot create wealth to back up that currency. If they had Superman or The Flash on their team and could create x million million dollars of goods and services at super-speed to back up their cash creation there would not be an economic crisis at all.
    Arguing about M, V or BDSM for that matter does not get around that reality.

  6. @Mr Ecks:
    Obummer or KingDick or any of them can create all the currency they like–they cannot create wealth to back up that currency.

    True, but as with any form of dilution (and QE = printing money = currency debasement whatever way you look at it), those that have cash in the bank will find the value of goods and services that can be purchased with said cash to be less because there is more money chasing those goods and services.

    The problem with QE is that no-one is quite sure of the effects and if those effects are disasterous (i.e. inflation or god forbid even hyperinflation), then there is no easy way to put the genie back in the bottle.

    If you want to understand the dynamics of hyperinflation (from a real world perspective rather than just pure economic theory), then Adam Fergusson’s book “When Money Dies” is a telling reminder.

    One thing that should be realised is that investing in assets doesn’t guarantee that you will survive a period of hyper-inflation as the assets themselves tend to be severely mispriced.

    After the Weimar hyperinflation, it was found that those who survived best was almost a lottery due to timing payments and personal decisions as to whether or not to honour a contract.

    During the hyperinflation, by the time a debtor took someone to court over a debt, the full settlement of it wouldn’t pay for the bus fare home.

    In our modern world, where people are mostly paid on a monthly basis, it might actually prevent some of the worst aspects of hyperinflation caused by the velocity of money effect.

    However, this is difficult to judge as we haven’t had any recent experience of it in a modern westernised economy.

  7. Frances-

    Hmm. If lots of new M is created and swapped for assets held by investors (QE), but all those investors do is dump the cash into insured deposit accounts and safe haven currencies, V falls off a cliff.

    In that case, the new money doesn’t count in the variable “M”. This is an equation that describes money circulating for produced goods (PQ). In the same way that a million pounds hidden in a mattress isn’t in the “M” variable either.

    The equation should be interpreted as saying something like “In some cycle of transactions, in which each currency unit is used once and only once [1] the sum of money transacted (M) will equal the total cost of the goods purchased (PQ) [2]”. So, “dead” money doesn’t come into it.

    It’s another problem with aggregate statistical thinking, the problem of what the variables actually represent.

    [1] The variable “V” is a kludge to make the thing fit multiple transaction cycles in an economic measurement period. The thing ought to be a partial differential equation anyway.

    [2] P, the hypothetical “price level” is another kludge. It would be more helfpul to talk in terms of delta Q.

  8. Ian B, it depends who owns the deposits (banking system liabilities) created via QE. If it is the private sector, the deposits will count within M4 (broad money) and do matter for the velocity calculation.

    Velocity increased after QE1 started, FWIW:

    This implies the “inflation expectations” channel worked for QE – what Scott Sumner calls the “hot potato effect”, that people expect higher inflation so start moving money around faster.

  9. Gareth, I think the problem is with that MVPQ equation is people are interpreting it all kinds of different ways. Is M broad money? Which definition of broad money?

    One thing though, is that if you look at the official calculated value for V for e.g. Britain or the US, you find it has a value around 2. Two transactions per year per dollarpound? That is a very low, suspiciously low, value. It suggests strongly to me that much of the “M” in the definition of M shouldn’t be there.

  10. JamesV

    Yes, if you look beyond the domestic economy you do indeed find where that money is invested….as you say, it has to go somewhere, but it doesn’t go into domestic spending if demand is flat and banks aren’t lending (for whatever reason).

    Ian B

    If QE money is mostly hoarded and therefore mostly omitted from M, it doesn’t make much difference, does it? If V is falling and QE money is hoarded, P & Q will also fall. The effectiveness of QE in maintaining P & Q when V is falling still depends on what is DONE with the money.

  11. Frances, what I was trying to say is that (as Friedman argued, IMV correctly), “V” doesn’t change much at all. If you increase M- which is liquidity, effectively, V will not alter much and PQ will rise (P will be >1).

    I don’t know of any evidence at all that V in the real economy *is* falling, has fallen, or will fall.

    People didn’t start spending money less frequently. They had less to spend. M fell (because of the collapse of the imaginary debt part of “wealth”), not V.

    It follows that naively QE makes sense, by reinflating M. The problem is, it is trying to raise M0 towards the old boom value of Mx (broad money measure, take thy pick). The problems are-

    a) You can’t actually print enough to do that anyway.

    b) It’s not going into the consumer economy, it’s being handed to banksters.

    c) Once the economy booms again, it’ll cause an even greater bubble of Mx. Rinse, repeat.

  12. I mean, what we’re down to is an economic model in which (a) money is nationalised and (b) handling it is contracted out to a cartel of banks. The result is that rather than the banks simply being participants in a market economy, they literally own the money, so you can’t let them fail. Hence, the obvious recipe for total chaos that we currently have. The least free part of the market is the money supply itself. This is preposterously wrongheaded.

  13. Ian B

    I don’t know what if any evidence there is for changes in V. But your second point – about QE not getting out into the consumer economy – is the nub of the matter.

    The money multiplier assumes that banks and consumers are effectively passive agents: give banks money, they will always lend multiples of it out. The reality is that lending is a commercial decision: banks have to want to lend and people/businesses have to want to borrow. If banks don’t want to lend and/or people/businesses don’t want to borrow the money doesn’t get out into the domestic economy.

    The first round of UK QE was aimed at investors rather than banks. But much of it still ended up in the banks because risk-averse investors preferred to put it on deposit rather than invest in riskier assets. Banks then didn’t lend it out, for whatever reason. I suspect the same will happen this time. In which case QE will once again be a damp squib.

    The question of ownership of the money supply is a much wider question, isn’t it? There are indeed strong arguments for doing something about the unlimited license to create money that commercial banks have. But trying to fix that in a crisis is rather like telling a lost traveller “if I were you I wouldn’t start from here”.

  14. Well Frances, arguably the point of crisis is the best time to fix things. Patching up the system for another run around the merry-go-round means that during the next Boom, nobody will want to reform anything, and then, the next Bust, “this isn’t the time for radical reform”, and so on.

    Anyway, all this highlights the specific problem with economic analysis based on aggregates. People think there is some aggregate “growth” or aggregate “spending” and so on and ignore the specifics in the market. The whole Keynesian mess of saying I investment==C consumption, therefore if the government “invests” so much we’ll get the same return in consumption, and it never happens. And so on.

    Maybe one way to look at it is the problem that however much QE is available to base loans on, the fundamental problem is hardly anybody is in a fit state to borrow anything, at least out here in the production/consumption economy. And I think that might be the cause of quite a lot of rather natural anger. Banks get given money. The rest of us have to borrow it. Doesn’t seem quite fair, does it? 🙂

  15. Ian B

    Totally agree. If the cause of Merv’s supposed underlying deflationary trend is the private sector paying down debt, QE as a strategy to offset the resulting reduction in broad money is totally nuts. Why on earth would people who have been paying down debt want to take on more?

    My own view – unpopular in certain quarters – is that if the private sector wants to deleverage, it will, and no power on earth can stop it. Nor should we try to. The quicker people pay off their excessive debts, the quicker the economy can return to real growth. Expecting already-indebted people and businesses to take on more debt to stop the economy sliding into recession is simply unrealistic.

    In the normal run of things, banks have to borrow money too, of course…..But bailouts (and QE) are exceptional. They shouldn’t happen. Let’s not have any more of them.

  16. My own view – unpopular in certain quarters – is that if the private sector wants to deleverage, it will, and no power on earth can stop it. Nor should we try to. The quicker people pay off their excessive debts, the quicker the economy can return to real growth. Expecting already-indebted people and businesses to take on more debt to stop the economy sliding into recession is simply unrealistic.

    I couldn’t agree more. It seems to me that everything that is being done at the moment by the powers that be is working directly against people and businesses in the economy who are attempting to act rationally.

  17. “is that if you look at the official calculated value for V for e.g. Britain or the US, you find it has a value around 2”

    That is not correct. M4ex and annual nominal GDP are both at around £1.5 trillion.

    Again: an increase in household/business debt is *not* necessary for monetary policy to be effective in boosting GDP. The BOE goblins are used to the banking system being the only transmission mechanism for MP. It’s not. Other ways:

    a) currency devaluation -> boosts exports, domestic investment/spending to substitute for expensive imports
    b) equity prices (high equity prices mean less bogus M&A, more real capital investment)
    c) wealth effects (if my pension fund doubles in value I will spend my income not save more)

  18. Monbiot’s and Keen’s dissing of the role of M0 is not wrong, but nor is it new or radical. Here’s Tim Congdon: “Narrow money – in either its M0 or M1 versions – does not determine important economic variables, such as prices or output, but is determined by them.” That was written in 1986. He was right then, and is right now.

Leave a Reply

Your email address will not be published. Required fields are marked *