Two weeks ago, analysts at Credit Suisse attacked one of the scariest charts of the era, the one showing the record-high mountain that is total US debt to GDP:
The chart, the CS analysts said, combined two different statistics and suffered from double-counting. Specifically, the analysts said, the “financial debt” segment counted whole loans and asset-backed securities as different types of debt when in reality asset-backed securities just repackage whole loans. ($1 billion of mortgage-backed securities and $1 billion of mortgages don’t make $2 billion of debt).
Ned Davis of Ned Davis Research was one of those who produced the original Debt To GDP chart, and he now defends it. He doesn’t dispute the double-counting charge. He simply notes that, even if you exclude financial-sector debt, US debt levels are very high.
Recently someone in our office gave
me an article from analyst Henry Blodget,
who excerpts a study from Credit Suisse
(that a client sent me), that says my chart
E501A is a “false idol” that is “technically
wrong and analytically meaningless.”
They call my chart “arguably the most
dangerous piece of propaganda to come
out of the current crisis.” They argue
our chart uses “spliced data” from two
different sources, and current data is really
not comparable to the 1929 period.
I have had many other criticisms of
my debt chart over the years. The most
frequent is that by including financial
debt, I am double-counting some loans.
This was a key point in the CS report.
After seeing some $1-2 trillion in toxic
assets (loans) that banks have made, I
am now astounded that anyone would
currently want to ignore financial debt.
In any case, the chart [below]
leaves out financial debt, comes directly
from one source, the Fed’s flow of funds
report, and does not compare with the
1920s. While the levels are obviously
lower without financial debt, it shows that
current record debt levels are well above
the mean for the past 60 years. It is not
technically wrong and anyone is free to
derive their own meaning.
By the way, here’s Ned’s original chart, the one the CSFB analysts are objecting to: