Sunday, July 3, 2011

Minimizing the problem?


In my links, clicking The Economy and then Data turns up Flow of Funds, Debt which brings up the current Federal Reserve Statistical Release Z.1 for Flow of Funds Accounts of the United States.

Well that was a mouthful.

Anyway, things one can download include

  • The Complete file
  • Summary statistics and table of contents
  • Debt growth, borrowing and debt outstanding tables

and more. I like that third item there, for the "debt outstanding" tables.

I download the CSV file, extract the three files from the ZIP, and verify that

  • D1.CSV shows "Credit Market Debt Growth by Sector"
  • D2.CSV shows "Credit Market Borrowing by Sector" and
  • D3.CSV shows "Credit Market Debt Outstanding by Sector"

That last is the one I want. Right-click, open with OpenOffice Calc. The Text Import window opens; under "Separated by" I make sure TAB and COMMA are checked, and under "Other options" that "Detect special numbers" is checked. Then I click OK, and the thing opens.
Then you have to remember to SAVE AS a spreadsheet, or all the formatting and figuring you do will be lost. Word to the wise.
The first column is DATE.
Second column is Total Domestic Nonfinancial Debt Outstanding.

Last column is FOREIGN.
Before that: Domestic Financial debt.
And before that: the domestic nonfinancial component of Federal government debt.

Total debt obviously includes both nonfinancial and financial. So to get a number for total debt, I want to add those two together.

I don't know whether "foreign" means we borrowed from foreigners or loaned to them. I omitted it last time around. This time I will include it, because I have an ulterior motive. I am trying to duplicate the "historic low" shown in Sackerson's third graph.

Then I want to look at Federal government debt in proportion to total debt. But this is how I got in trouble before, because the government debt in this file is listed under "Domestic nonfinancial". So apparently the government number does not include the financial part of the federal debt, whatever that may be.

Yes. The most recent (annual) Z.1 number for Federal debt, for 2010, is 9385.6 billion. Nine trillion and some. But the number people throw around these days is 14 trillion. So we're missing quite a bit. Some of that is 2010 versus 2011. But much of it must be the missing financial portion of the federal debt.

I have to assume that the missing financial portion of the federal debt is lumped in with everybody else's financial debt, so that it is in the total. But that just makes the comparison worse, if you do what I did a while back, and compare the nonfinancial portion of the federal debt to the total debt. Gotta be careful with these things.

Anyway when I add up nonfinancial plus financial plus foreign debt, and look at the nonfinancial portion of the federal debt as a percentage of that, this is what I get:


And that is what I thought.

See how the federal debt falls to about 10 percent of total debt, there around 2007?

Hard to believe, isn't it? I think that's because we're counting the federal government's financial debt in the total, but we're not counting it as part of the federal debt.

If we go with the popular number, $14 trillion for the federal debt, and if that's ten percent of total debt, then total debt is about $140 trillion. I don't think that number is right. Grandfather Hodges lists America's total debt as $57 trillion. And the Z.1 numbers add up to $52.4 trillion.

Fourteen trillion is 26 or 27 percent of the 52.4 number, or just under a quarter of Michael Hodges' 57T.

3 comments:

Sackerson said...

ZI Flow of Funds for end 2010:

Line 1 (total): 52,399.5 bn
Line 7 (State & Local govt): 2,464.7 bn
Line 8 (Fed govt): 9,385.6 bn

Lines 7 & 8 / Line 1 = 22.62%

Checks.

Of course, line 7 excludes pensions, so the argument could easily continue!

Sackerson said...

Oh, and a further thing that it took me a while to realize: line 11 is a sub-total of lines 12, 13 and 14 so it's easy to double-count accidentally.

The Arthurian said...

Yes, good as the source data often is, it's not good enough to foil misinterpretation.

I was devastated when Gene pointed out the discrepancy between $14.3 trillion and the 9,385.6 billion number I was using. I hadn't even looked at that.

Worst of all, one is left largely to one's own devices when it comes to figuring out the error.