Wednesday, March 28, 2012

Debt, Income and Aggregate Demand: Scoring Krugman vs Keen

Not having learned from my earlier foray into MMT-land, I’m at it again, sucked into a debate between Paul Krugman and Steve Keen over how to think about debt, money and macroeconomics. It is remarkable that these questions could still be unresolved after eons of economic theorizing; this stuff is not very complicated, after all. Someone has to be right, but who?

Keen argues that mainstream economics is wedded to a false conception of what banks are and how they operate. According to conventional wisdom, people who earn more than they spend puts their savings into banks, and the banks turn around and lend this money to others who wish to spend more than they earn. The result is a wash: banks are simply intermediaries who play a role in the distribution of financial resources but not in their quantity. This view recognizes that money is created through the process of credit creation, but (1) the total volume of new money is given by the money multiplier and therefore reflects choices made by the central bank (M is exogenous), and (2) the amount of money in circulation influences aggregate demand via its impact on interest rates—beyond this, more money just raises the price level. For more detail, consult IS-LM.

Keen says this is all wrong. Banks don’t lend because someone has deposited, they lend because they have found a profitable lending opportunity. In the aggregate, credit creation is also deposit creation, since a portion of the new money eventually finds its way back into the banks, and banks borrow and lend between one another to meet their reserve requirements. But it is lending that leads, not saving. This is seen is a refutation of the “loanable funds” doctrine. Meanwhile, financial resources acquired through borrowing support the demand for goods and services just as effectively as income; thus aggregate demand is a function of the sum of the two, and not just income alone. Unfortunately, to add Minsky into the equation, people and businesses do not borrow only to purchase newly produced goods (including investment goods), but also to speculate on existing assets, thus bidding up their price. Borrowing in this fashion is inherently a Ponzi activity in the aggregate, and the perception of rising asset values in good times paves the way for a wave of insolvencies in the bust.

To sort this out, it may help to be a bit schematic. Let us say that there are three time periods of arbitrary length, and let’s tell two stories. The first is the orthodox (O) story: See, a guy walks into a bank and…..deposits some money, say $100, into his account. That’s period 1. In period 2 the bank lends $90 of this new-found cash to individual #2, taking care to hang on to a required 10% to meet its reserve requirements. In period 3 our borrower spends the money on something, generating income for individual 3. We could go on, but you get the point. This story can be found in any introductory textbook you might care to (or be required to) read. The upshot is that, at the end of period 3 there is $90 in new money, created by lending, and $90 in new income as well, accounted for by individual #3. Individual #1 is a creditor to the tune of $90, individual #2 is a $90 debtor, and the bank is simply an intermediary, with both its asset and liability columns up $100. (Assets: $90 in loans, $10 in reserves. Liabilities: $100 in deposits.)

So what might be the Heterodox (H) story? In period 1, a guy walks into a bank and….asks for a $90 loan. The loan officer figures the default risk is covered by the interest rate, so she extends it. In period 2, the borrower buys something from individual #2, which was the purpose for going into debt. In period 3, individual #2 deposits some of this new revenue in the bank. It happens that period 3 is also designated as a point at which the bank must meet its required reserve ratio. It could be that they will have extra reserves to lend out, or that they will have to borrow from some other bank that has reserves to spare. One way or another, they settle up. What is the result? There is $90 in new money, $90 in new income, one individual is a $90 creditor, another is a $90 debtor, and the bank’s books are balanced.

So the H story is beginning to look a lot like the O story. This should not surprise us, since mainstream theory comes close to endorsing Keen’s pronouncement that “aggregate demand is income plus the change in debt” (italics in the original). You can see this directly in the national income accounts: the sum of expenditures is equal to the sum of income plus net saving or dissaving. Obviously there can be net saving or dissaving only in an open economy with a current account imbalance. The reason I use the word “close” is that this identity holds for expenditures, but expenditures equal aggregate (desired) demand only in equilibrium. In textbooks you see this with some variables having an asterisk (equilibrium) and others not.

Is this a tempest in a teapot? I’m inclined to say, partly yes, but not entirely. I think it does matter, perhaps a lot, what sequence these three activities—borrowing, spending, and saving—occur in. A world in which investment and growth are driven by perceived profit opportunities is different from one in which they are the passive result of prior decisions to not spend. The first is a Keynesian, and also a Schumpeterian, world; the second is the world of monetary orthodoxy. Similarly, money really is more endogenous in the H world, since financial commitments are, in general, made first, and then monetary authorities have to deal with how much to accommodate them, especially if the banking sector finds itself spread thin. (I have a similar view of how trade and international finance work: first people decide how much they wish to import or export, and then holders of currencies have to figure out how to manage the accumulations and decumulations of foreign exchange.)  Same dance, same partners, but it matters who leads.

So in the end I think there is a real and important difference, but I don’t think it shows up at the level of accounting. For what it’s worth, I will be at the Berlin conference that Steve is speaking at, and I look forward to meeting him. I doubt I will get involved in any brouhaha over this, since I’m basically a micro guy, and I just want to figure this out to reduce my general level of confusion.

12 comments:

Viscid Konrad said...

NET debt. NET debt. It would be nice if Keen came around to recognizing that.

Of course, I'm still confused by this idea of AD being a number rather than a schedule, but that's for another day.

For now, expenditures equal income plus NET dissaving-- not Keen's GROSS.

Unknown said...

You should not label Keen as an MMT guy; he has differences with that school. Orthodox economists --whether they be fresh water or salt water makes no difference as they argue over minutiae such as informational asymmetries-- should be invited to see the workings of money center banks, specifically a dealing room. No global bank in the world can state their liabilities (client's savings) at the end of each day such is the complexity of flows. Loans are made based on their risk appetite; reserves are a secondary concern. The money multiplier argument is a fallacy as is the idea of one man's debt is another's savings. I have no problem in siding with Keen on this one.

Viscid Konrad said...

The money multiplier argument is a fallacy as is the idea of one man's debt is another's savings. I have no problem in siding with Keen on this one.

I've never heard "one man's debt is another's savings." I've heard "one man's debt is another's asset," which is in fact true.

I don't think there's a lot of disputing that one may create financial assets out of nothing. The creation of net financial assets is another question.

Letsgetitdone said...

I think if you want to get into MMT land, then you ought to go to MMT blogs. The top 3 MMT blogs are: http://moslereconomics.com/

http://neweconomicperspectives.org/

http://bilbo.economicoutlook.net/blog/

kevin quinn said...

Well,damn: I skimmed the post and thought I read "a man goes into a bar...." and was all ready for some jokes! Maybe after he goes into the bank, he goes into a bar and the bartender says,"What is this? Some kind of a joke?"

Peter Dorman said...

Well, in one version of the joke a man goes into a bar to sell the bartender a keg, and then the bartender goes out onto the street to find some poor sucker to drink it. In the other version, a man goes into the bar, orders a beer, the bartender pours it and orders another keg. Funny enough for you? ;)

Lord said...

The emphasis has to be the lag between when these debt and lending decisions are made and when the central bank ratifies or repudiates them. When the debt increases asset values it can continue for years before inflation pressures rise and the central bank responds to them or never really responds to them but the bubble bursts. When created by lower lending standards and destroyed by higher ones a boom and bust is established, the bezel expanding and contracting over a period of years.

Unknown said...

I did not phrase clearly: from my understanding the standard orthodox analysis is that private debt reflects a trade between an impatient borrower and a patient saver hence my allusion to debt and savings. Certainly, there are capital constraints in banking -- as a practitioner rather then a researcher/academic the idea of "capital consumption" is on the top of my concerns but fundamentally financial institutions make loans all the time based on risk limits in place. If the deal appears profitable on paper it will get done. And Europe's zombie banks do so without leverage rules; I don't see new capital adequacy rules making things safer BTW. Derivative contracts are amongst the many financial innovations that are not adequately accounted for in traditional macro models and add to the argument of endogenous money creation and against Krugman's stance (in my opinion). Can NET DEBT properly capture such contracts when they remain opaque and MTM is open to interpretation (e.g. if an auditor is dissatisfied with a valuation she may ask for 'market' levels).
Regards,

annasalvator said...
This comment has been removed by the author.
annasalvator said...

Some of points from this article are very helpful for me as I haven’t considered them yet. I would like to say thank you for sharing this cool article. Bookmarked and sharing for friends. More about savings calculator.

Unknown said...

ARIJIT BANIK said...
"private debt reflects a trade between an impatient borrower and a patient saver hence my allusion to debt and savings"
This is not true. Banks do NOT lend savings, they have a legal privilege to create credit. It is savings that create deposits. But they cannot create net financial assets as loans = deposits.
This is why while the majority of the money supply is bank credit, the only net financial assets held by the private sector are government debt (money, bonds).

Unknown said...



Blogger David said...
"NET debt. NET debt. It would be nice if Keen came around to recognizing that."

Why?