Sunday, September 18, 2011

"See you at 8 o'clock"


It is 4:47 AM as I start this post. I've been looking at some numbers Clonal put together here:

Clonal said...

1945-1965 debt 5.72% GDP 6.28% Affordable
1966-1976 debt 10.3% GDP 8.68% getting unaffordable
1977-1992 debt 11.3% GDP 8.31% more so
1993-2007 debt 8.25% GDP 5.47%

This is not what I think, but you could look at those numbers and say, "Gee, debt growth was three percentage points lower in the most recent period, that should have been GOOD for GDP, right?"

What I think is, debt is a stock and GDP is a flow. Debt accumulates, GDP does not. Every year we start counting GDP from zero again. But every year, accumulated debt starts out bigger than it was before.
Every year debt is greater than the year before, until the crisis, of course.
Now, Clonal isn't comparing a stock to a flow. He's looking at the growth of debt, which is the flow that adds to the stock, or something like that. ("Stock" and "flow" are not my terms.) (I don't know what the growth of a flow is. Froth, maybe?)

But my point is, every year the growth of debt adds to the accumulation of debt. And it adds to the cost of accumulated debt, a cost that interferes with GDP growth. So it seems there is an important column missing from Clonal's table, the column that would show the accumulation of debt. The column that would show the problem.


Clonal's numbers:

Graph #1
Blue bars, debt growth. Red bars, GDP growth. Debt lagged behind GDP in the first period. Hard to believe, today, that such a thing could happen.

Everything increased into the second period. Into the third period, debt growth increased but GDP growth fell. And into the last period, everything fell.

Interesting, the way the bar-pairs show an increasing height difference: Close in the first period. Farther apart in the second. Twice as far apart in the third. And in the last period, about the same as the third. Graph #2 shows these differences.

Graph #2
Interesting, too, that the decline of debt and GDP growth in the last period does not reflect what has happened since the crisis. Clonal's last period ends with 2007.


What I think is, I think about the column of numbers missing from Clonal's table. The accumulation of debt. The accumulation was low in Clonal's first period and as a result, the economy was able to grow faster than debt.

In the second period, debt was already accumulating to the point that it interfered with growth. GDP growth became slower than debt growth.

In the third period, the debt added during period two made it even harder for the economy to grow. So GDP growth was a lot slower than debt growth.

And in the fourth period, GDP gave up. It didn't even try. Even debt growth fell.

The problem? Excessive accumulation of debt.

5 comments:

The Arthurian said...

Not what I had in mind, if I were to follow-up on this post, but...

It is interesting that the difference between debt growth and GDP growth in the last period (on Graph #2) is *LESS* than it was in the third period. (And this is *before* deleverage began.)

This suggests that the debt growth of the third period led to the problems we have today.

But again, the difference between debt growth and GDP growth in the third period was *twice* what it was in the second. This suggests that second-period debt-growth led to the problems-that-led-to the problems we have today.

In other words, debt growth in the 1966-1976 period was already a problem. Reminds me of Graph #7 here.

Liminal Hack said...

Hi all. The variations in growth rate are very definitely demographically driven. I took the fred TCMDO data and plotted it beside the 5 year change in the dependency ratio data for the US from the UN population division (http://data.un.org/Data.aspx?d=PopDiv&f=variableID%3A44#PopDiv).

I've used the year on year growth rate of TCDMO versus the total dependency ratio, which thus far is dominated by the young dependency ratio.

http://liminalhack.files.wordpress.com/2011/09/dependency_vs_debtgrowth.jpg

The graph clearly shows heightened credit growth rate when the dependency ratio is low, with reversion to the 'standard) credit growth rates either side of that big dip which is where the boomers enter the workforce (which of course causes an oil shock and stagflation).

So I strongly suggest that we stop worrying about the little differences in the post WWII credit trend in this data, it's a red herring.

What matters, is these two things:

1) what sets the credit expansion rate at about 2% even in normal times? I presume the answer is mainly underlying population growth, but credit is still expanding faster than population so as art observes there is still a discrepancy here.

2) does the TCMDO data include any credit created in the shadow banking sector. E.g. is the fred data really the full picture?

Liminal Hack said...

depednecy ratio versus debt growth

The Arthurian said...

Hi, Liminal. I see what your graph shows...

"2) does the TCMDO data include any credit created in the shadow banking sector. E.g. is the fred data really the full picture?"

Not the full picture, I would say. TCMDO is "credit market" debt. It includes only about $10 trillion of the $14 trillion Federal debt, the rest being held by Federal agencies, as I have it.

I looked at TCMDO components in A Shopping List and the follow-up Bagging the Groceries.

The Arthurian said...

"1) what sets the credit expansion rate at about 2% even in normal times?"

Maybe something like this:

"The policy record for [the March 10 FOMC] meeting indicates that the Committee was setting as its objectives a growth rate of 3 percent for the money supply (currency outside banks and private demand deposits) and 5 percent for the adjusted bank credit proxy over the second quarter."

Or maybe I don't get what you are getting at.