GDP plus Change in Debt—and the US Flow of Funds
on September 7th, 2010 at 4:06 pm
My recent post “What Bernanke doesn’t understand about deflation” has hit a chord, with a number of sites around the world reproducing it—including John Mauldin’s Outside the Box column. But it has raised a couple of queries in people’s minds too:
- Does my definition that “aggregate demand equals GDP plus the change in debt” involve double-counting?
- My figures for the USA are difficult to reconcile with the published US Flow of Funds data.
On the second point first, I produce an aggregate level of private sector debt in the USA from Table L1 of the Flow of Funds (on page 60 of the June 2010 PDF, and in ltab1d.prn in the ltabs.zip data archive) by adding together debt data for the following sectors:
- Household
- Non-financial corporations
- Nonfarm non-corporate
- Farm
- Financial Corporations
This omits some of the debt included in the aggregate debt level in the same table—notably government debt and debt owed by the “rest of the world”. In the interests of making it easier to reconcile my table with the data in the Flow of Funds, here’s the same exercise applied simply to the very first row in Table L1 (and the first column in ltab1d), “Total credit market debt owed by:”
US Flow of Funds Table L1, row 1 (column 1 in the file ltabs1d.prn)
| 200504 |
41267079 |
| 200601 |
42343298 |
| 200602 |
43337326 |
| 200603 |
44258861 |
| 200604 |
45329493 |
| 200701 |
46504304 |
| 200702 |
47528151 |
| 200703 |
48860628 |
| 200704 |
50044489 |
| 200801 |
50812625 |
| 200802 |
51272735 |
| 200803 |
52082473 |
| 200804 |
52524931 |
| 200901 |
52882693 |
| 200902 |
52686684 |
| 200903 |
52549072 |
| 200904 |
52416676 |
| 201001 |
52126900 |
I also transform the data to monthly by interpolation, and the way my data is stored the figure I give for 2006 corresponds to the figure stored for the end of the quarter 200504 by the Fed. I’ve highlighted these numbers in the two tables here to make that more obvious.
Change in debt and aggregate demand
| Variable\Year | 2006 | 2007 | 2008 | 2009 | 2010 |
| GDP | 12,915,600 | 13,611,500 | 14,291,300 | 14,191,200 | 14,277,300 |
| Change in Nominal GDP % | 6.3% | 5.4% | 5.0% | -0.7% | 0.6% |
| Change in Real GDP % | 2.7% | 2.4% | 2.3% | -2.8% | 0.2% |
| Inflation Rate % | 4.0% | 2.1% | 4.3% | 0.0% | 2.6% |
| Total Debt | 41,267,079 | 45,329,493 | 50,044,489 | 52,524,931 | 52,416,676 |
| Debt Growth Rate % | 9.2% | 9.8% | 10.4% | 5.0% | -0.2% |
| Change in Debt | 3,468,111 | 4,062,414 | 4,714,996 | 2,480,442 | -108,255 |
| GDP + Change in Debt | 16,383,711 | 17,673,914 | 19,006,296 | 16,671,642 | 14,169,045 |
| Change in Aggregate Demand % | 0.0% | 7.9% | 7.5% | -12.3% | -15.0% |
On the first point, since I consider that aggregate demand is spent on both goods & services (which are counted in GDP) and the net sum expended purchasing existing assets (which is not counted in GDP), then there is no double counting. A standard textbook aggregate demand figure is the sum spent buying goods and services (for the expenditure definition), which omits of course the sum spent buying existing assets as well. That would be all well and good if we lived in a world without asset sale—which of course we don’t.
Another reason people see a potential error here is that they think that a loan simply represents the transfer of spending power from a saver to a borrower, so that overall there’s no change in spending power because of a loan: money is simply transferred from one group that will therefore spend less (creditors), to another that will therefore spend more (debtors). This is clearly the thinking that Bernanke applied when he, in common with most all neoclassical economists, dismissed Fisher’s “debt deflation” explanation for the Great Depression:
“Absent implausibly large differences in marginal spending propensities among the groups, it was suggested, pure redistributions should have no significant macroeconomic effects. ” (Bernanke 2000, p. 24)
This is not the case in the real world, for two reasons:
- Credit Money is created by banks “out of nothing” by the act of giving a borrower purchasing power (a loan of money) in return for recording a liability by that borrower to the bank (a bank debt). This creates new spending power “ab initio” without removing it from other agents. For the mechanics of this process, see my “Roving Cavaliers of Credit” blog entry (click here for the PDF).
-
As Schumpeter argues cogently, the endogenous creation of money by the banking sector lending to entrepreneurs is an essential reason that capitalism can grow, and it creates spending power that does not originate in the existing “circular flow of commodities”:
“From this it follows, therefore, that in real life total credit must be greater than it could be if there were only fully covered credit. The credit structure projects not only beyond the existing gold basis, but also beyond the existing commodity basis.”
“[T]he entrepreneur needs credit … [T]his purchasing power does not flow towards him automatically, as to the producer in the circular flow, by the sale of what he produced in preceding periods. If he does not happen to possess it … he must borrow it… He can only become an entrepreneur by previously becoming a debtor… his becoming a debtor arises from the necessity of the case and is not something abnormal, an accidental event to be explained by particular circumstances. What he first wants is credit. Before he requires any goods whatever, he requires purchasing power. He is the typical debtor in capitalist society.” (Schumpeter 1934, pp. 101-102)
So there is no double-counting in “aggregate demand equals GDP plus the change in debt”: the rise in debt adds new demand to that generated by the sale of commodities alone (and is a good thing here because it finances a large part of investment); and the increase in debt is spent financing part of investment and consumption (an overlap that could give rise to double-counting) and also on purchases of existing assets (where no overlap is possible).
Bernanke, B. S. (2000). Essays on the Great Depression. Princeton, Princeton University Press.
Schumpeter, J. A. (1934). The theory of economic development : an inquiry into profits, capital, credit, interest and the business cycle. Cambridge, Massachusetts, Harvard University Press.
Just found an interesting perspective on You Tube.
Anybody follow Patrick Dixon? He’s a futurist based in London. He’s also got some interesting ideas on various trends (much better than Celente, Alex Jones and all the rest of those guys).
Within 10 years, stock markets as we know them will disappear. Instead, it will be one global market.
Some problems come with that:
How will you get all of these countries to throw away protectionist tendencies re their economies?
Will this also include a global currency? Lots of people had a go at Robert Fisk when he first broke that story. Realistically that will take what, 5 years to set up?
Considering that the glbal depression started in the States, why would other countries even consider something like this? Publically there would be various nice sounding press releases. But privately a lot of opposition.
I’ve looked at some of his economic clips. Some other problems that he has:
He’s not talking at all about the role of debt in this.
Maybe because if he did his lecture bookings would go way down?
One minute he talks about the paying attention to people in the markets and not just data. But then the next he’s totally ignoring the predatory nature of the banks and other big players in the depression.
If you’re interested, check here:
http://www.youtube.com/pjvdixon#p/u/14/PwPBgubcyI8
Totally off topic. But, if you need a break from the global depression stress (and who doesn’t), check out this link:
http://youtu.be/f7Gbqi2VtgY
someone at beginning of thread was mouthing off about Obama infrastructure bank idea like that was nutty…I found it to be one of the best ideas he has had yet (of course I’m biased as I want public banks and I’m a civil engineer) and the fact the US media has barely mentioned it and has asked no questions about makes me think it is a very good idea…and now that paid trolls masking as grass roots bloggers are ragging on it, I’m becoming even more convinced its a good idea.
Why, in the US, we have let the Fed “print” money in the form of no interest loans to banks only to have banks buy Treasuries with this money we gave them, guarantee them no risk profits and income streams paid for by taxpayers as we pay interest to banks on the money we gave them, is beyond my comprehesion. I know we as a population are captured by Wall Street and thank them daily for taking all our jobs because it clears us of all those inefficient middle class salaries and the spoiled entitlement we felt towards havign a broad middle class, but really, really, pay banks interest on money Fed lent them a no interest?!? That is more stupid than I thougt we could ever be. But this is apparently okay, nothing for Fox News to talk about….but Obama’s bank for infrastructure projects, thats a sink hole.
Why should private banks get all the leverarge, while governments can spend what they take in? If no one else leveraged up, I guess I could be sympathetic to frugal austerity for govt too….but if banks get to create money and leverage and no one is trying to stop that, then any call for govt no do the same is hypocritical.
US infrastructure has been long neglected as politicians would rather give home builders a deal than gix a bridge, maintiaining bridges is not interesting unless the major interstate highway bridge over the Mississippi in a major city about to have a the Republican presidential convention completely collapses and kills 13 people.
Is all debt bad, are public banks funding public works not a good thing, generally? Seems more useful than say printing money to banks so we can pay interest to banks…
by the way gaday…here in the US as profitable companies keep laying people off because their sales are down 2 percent, I keep thinking and saying, depressions are depressing. More than half my co-workers were laid-off, I miss them, as I look at their empty cubes and the piles of work on my desk that they once helped me do. I miss them when people call and get mad they don’t get the service they once did. I’m so lucky to have a job but am constantly on edge that I will be next, I know I can’t service our customers like we used to but I try, because one complaint, one lost job may be the end of my job…I cringe when I hear the owners are paying interest only on loans they took out to build plants that are not fully used…its plain depressing, I can’t imagine how I’d feel if I was unemployed.
My neighbors, who moved into their house in 2001, a young couple both with college educations and seeming success at their jobs and now two kids…one was a health insurance broker that started his own business that has done well, the other an architect, now have their house for sale….it is a short sale, they owe way more than their house value…I think, 40 years ago, only the father would have had to work to pay the mortgage, and if they had to sell ten years after buying, they wouldn’t have been upside down, they’d be set by now, and even more so if wife also worked…but not in the 2000s, instead they are haggling with bank, realtors, lawyers. Architect says 50 percent of architects in MN are unemployed, she is lucky to still be working reduced hours….
Steve,
In the statement “and the increase in debt is spent financing part of investment and consumption (an overlap that could give rise to double-counting) and also on purchases of existing assets (where no overlap is possible)”, what did you mean by investment? The debt can be used to purchase commodities and labor that should probably be termed as consumption. Debt that is not used for the above and put in secondary financial markets or RE would come under asset purchase. What are investments that cause double counting?
Part of investment in plant and machinery is debt-financed csr, so adding the change in debt to GDP plus investment there would involve double counting. The dilemma disappears when you include asset purchases as part of aggregate demand, though I’m still treating part of change in debt as a contribution to demand when in fact it isn’t–compounding of unpaid interest. There are nuances in this, but the overall concept–that aggregate demand over all markets is the sum of income plus the change in debt–falls into the “roughly right” rather than “precisely wrong” category.
Thanks Steve, I think I’m able to understand the problem now. Assume if we knew the aggregate domestic investment of a year and aggregate profits of the previous year of all industrial sector, would it be possible to determine the proportion of investment coming from firms’ profit/equity rather than debt? But I suppose such data are not available from flow of funds report. Or do I still ignore some aspect that is to be understood?
That’s reasonable csr.
Steve,
The more I think about adding “Change in Debt” to GDP the more confused I become (although that is unlikely to indicate a problem with your mode). Somehow I came to the conclusion that GDP already contains a “Change in Debt” component. Let me explain:
1. If someone borrows to spend on goods then this borrowing is already accounted for as part of GDP as Consumption.
2. If someone borrows to invest in a new asset then this borrowing is already accounted for as part of GDP as Investment.
3. If someone borrows to invest in an existing asset then the vendor of that asset has the following options:
3.1. The vendor can retire some debt. But since we observe an aggregate increase of the debt levels we can conclude that the debt retirement is more than offset by new debt and we can concentrate on the difference only.
3.2. The vendor can spend the proceeds on consumption. Just like (1) above this is already accounted for in GDP.
3.3. The vendor can invest in a new asset. Just like (2) above this is already accounted for in GDP.
3.4. The vendor can decide to save the proceeds (or even take them as cash and stuff them in the mattress). In this case there will be no increase of the aggregate demand.
3.5. The vendor can decide to buy another existing asset, in which case the next vendor faces again a choice between the 3.X options
It is possible that some of the funds that came from rising debt have become “unstuck” in option 3.5 and are being reused for high frequency buying and selling of existing assets. Still, between the transactions these funds would sit in a cash management account (or whatever account the brokers use for settlement) and perhaps should be classified as “savings” and therefore not affecting the aggregate demand.
As you can see all possible “branches” of the decision tree lead either to “leaves” that are accounted for by the GDP or to “leaves” that do not contribute to the aggregate demand.
In view of all that are you sure that adding “Change in Debt” to GDP is a valid thing to do?
Certainly an analysis of the “debt-fueled” (or rather debt-affected) portion of GDP could prove to be interesting.
I like the concept of considering debt growth, but I also worry about the double-counting questions. On the question of counting debt, is it possible that the right answer is to use only bank loans, which consist of newly-created money, while other components of private debt simply pass money from one entity to another? Growth of bank debt would thus have the same inflationary effect as government deficits, which also inject new money into the economy.
I think you’re technically correct BlackBox.
However there is also the effect of what is accepted as money morphing during times of extreme speculation–and we’re living in the most extreme of times. I have seen plenty of reports of mutual fund transfers being accepted as final payment in the 2000s in the USA for example.
I also expect that when this era is dissected in a half century or so, we’ll find plenty of fraud in the operations of the financial sector that enabled non-banks to create money.
So I err on the side of measuring all debt; when I have time and resources (both unlikely events!) I’ll try to drill deeper.
Steve,
Thanks for the reply. And many more thanks for the blog and book. I have learned a lot.
The way I look at it is that is that spending=income is an identity, but it is important to look at net inflow into the system from budget deficit + new net bank lending + trade surplus. All 3 terms are significant in recent years for the US.
The trade deficit has been very deflationary in recent years, and continues to be so now. This used to be more than offset by the deficit and especially by bank lending during the housing bubble. Now bank lending is contracting too and even the enormous Federal deficit is not enough to offset those effects. Indeed it should not be if we want to deleverage rather than simply moving debt from one place to another.
However, in a Minsky sense, the gross debt is also a problem in its own right, since debt service problems and defaults can mess up the economy, even when the debt is purely between private actors.
You’re welcome BlackBox.
On the identity, Michael Hudson observed during discussion of my paper yesterday that the identity should be “Spending + Capital Gain = Income + Change in Debt”. That ties it together nicely I think.
Steve, yours is the only ‘economics’ I can identify with, being an old systems and accounting executive. Your ‘change in aggregate demand + debt’ table in this post makes intuitive sense to me.
I imagine you keep this table up to date and have constructed it going back decades. Would you consider making that table available for download? I’d love to play with it:)
Will do Debtwatchj1–I usually just put graphs up but I’ll link to a spreadsheet next time.