Sunday, May 22, 2011

A new look at debt

It is 1:30 in the morning, and I have no post written for 4:00 AM. But I woke up with a thought.

Graph #1
That's the number one most popular debt graph at FRED today:
Federal Government Debt: Total Public Debt (GFDEBTN)... 1966/3/31-2010/12/31... The units are Mar Jun Sep Dec, End of Month, which I will call MJSD going forward. (These units are incompatible with "quarterly" units. That affects the combinations of data I can look at.) The graph goes back to 1966, the year I graduated high school.

Note that the number tops out at 14 trillion.

Graph #2
My graph #2 is number six on FRED's list of debt graphs:
Total Credit Market Debt Owed (TCMDO)... 1949:Q4 to 2010:Q4... Quarterly data. The graph goes back to 1949, the year I was born.

Graph #3
Graph #3 shows the first two, together. TCMDO is given in billions. Total Public Debt is given in millions, so I divided by 1000 to convert it to billions.

Using the data above, I cannot divide the one debt number by the other because the "frequency" differs: The one data set is quarterly and the other is MJSD. FRED simply reports an error.

Graph #4
This is number four from the FRED list of debt graphs:
Gross Federal Debt (FYGFD)... 1939 to 2010 ... Annual. The value of the latest observation here is 13528.8 billion, as compared to 14025.215 billion for the GFDEBTN graph above. Close, but not a match.

Graph #5
That's the two graphs of Federal debt, combined. Indistinguishable, in my book. The red line starts earlier, and the blue line finishes later, that's about the only difference. Oh, and FYGFD is annual numbers.

Q: Why does the one graph show 14 trillion federal debt, and the other show 13.5 trillion?

A: Apparently because the 13.5 number is from an earlier date. That's what the graph suggests to me.

You with me so far, Hayward?

It's 2:27 AM... How'm I doing?

Graph #6
Graph #6 is total debt as a multiple of total Federal debt.

Graph #7
Graph #7 is like Graph #6, but inverted.

2:35 AM. The clock is running.

Graph #8
Graph #8 shows non-Federal debt as a multiple of the Federal debt. It has been pretty high for a long time.

This graph is my carefully-done version of a comparison between public and private debt. The old graphs are not trustworthy, so I'm doing some new ones.

Graph #9
Graph #9 is like Graph #8, but inverted. Federal debt (relative to non-Federal debt) starts high but drops quickly, then slows. But the decline comes to an end with the 1974 recession.

The Federal number is not quite as low as it looks, because this graph does not start at zero. But the Federal debt has been stuck in the neighborhood of 30% of non-Federal debt since the late 1960s. A long time.

2:53 AM.

Graph #10
This is the graph I woke up wanting to see. Year-to-year change in total debt, compared to year-to-year change in the Federal debt.

The big red spike on the left is Federal spending during World War II.

In the 1950s and '60s the growth of Federal debt was substantially lower...

Graph #11
Graph #11 is the same as Graph #10, but it eliminates the big red spike by starting in 1949. So we get a close-up view of what remains.

In the 1950s and '60s the growth of Federal debt was substantially lower than that of non-Federal debt. There is a red spike in 1968, associated with the near-recession of 1967.

There are red spikes associated with the 1970 recession and the 1974 recession and the 1980 and 1981 recessions. That is what pushed Federal debt growth upward: recessions. The Federal government was trying to promote economic growth. They were trying to do right by us, I think. Things just got away from them.

I've been meaning to re-do my debt graphs for quite some time. Gene Hayward found an error in my ways, back in April.

I was like a deer in the headlights. Didn't know what to do first. And I ended up doing nothing. Dumbass.

But it was getting harder and harder for me to write. And I went to bed last night with nothing scheduled to post this morning. But I woke up early and I'm on the case now.

Gene -- Thanks for keeping an eye on me. Everybody, really. Thank you.


// 3:25 AM. Gonna take a nap now.


Jazzbumpa said...

Didn't know what to do first. And I ended up doing nothing. Dumbass.

If you don't know what to do, doing nothing is a reasonable option - as long as there aren't . wolves howling at your door. Sitting on this for a month gave you time for unseen things to happen in the nether regions of your fevered brain.

Then, the incubation period was over, and THIS popped out.

What strikes me is the inflection point around (roughly) 1980, easily seen in graphs 6 - 9. Another cause (or effect - you decide) of the great stagnation.

I'll have to spend some time digesting all this.


P.S - BTW standard deviation is a built-in function in Excel, and also the FREE Open Office Suite spreadsheet that I'm using these days. The formula is rally ugly. You'd never want to calculate it by hand

Jazzbumpa said...

Oh -

To manipulate the qrtly and MJSD data, you can download the numbers and work with them in a spreadsheet.


The Arthurian said...

My fevered brain.:)

"Sitting on this for a month gave you time for unseen things to happen in the nether regions of your fevered brain. Then, the incubation period was over, and THIS popped out."

Yeah, that part of it fascinates me.

"What strikes me is the inflection point around (roughly) 1980, easily seen in graphs 6 - 9. Another cause (or effect - you decide) of the great stagnation."

Remarkable coincidence with economic events. (I'm looking at Graph #8.) A massive run-up of private debt all during the golden age. Troubles arose then after the recessions of 1974 and 1981 with private debt falling and/or federal debt rising. New policies in place by 1982 -- the rapid growth of federal debt -- produce a different pattern until the late 1980s. Then possible the elimination of the tax deduction for interest costs causes the 1989-1992 drop. Then your "bronze" age appears on the graph as an uptrend in the late 1990s....

I think one can see both the end of growth (1973) and the change due to Wanniski-nomics (1980+) in these and other graphs. Important to distinguish between the two. Most of the economic complaints people have these days are direct results of the policies of Reaganomics.

And yet, there were problems existing before the supply-side approach was put in place. Those are the problems that must be solved. Other than that it is simply a matter of undoing all the bad that was done in the guise of fixing things.

Everybody on TV talks about the federal debt. Nobody looks at it in a context like these graphs.

I did graphs of this nature back in mid-november 2010 but I made the mistake of thinking "financial debt" was all private-sector debt. Gene Hayward wondered why my federal debt number was well below 14 trillion on those graphs. Ouch!

Thanks for the tips... I am also using Open Office these days.

Gotta go mow something... I don't have time to organize these thoughts.

Gene Hayward said...

Thanks, Art. However, you give me too much credit! You went way beyond the scope of what I was inquiring about BUT I like the result. This will be a graph I show my students next school year. I have a couple of ambitious students who visit your site and are interested in your perspective.

Can I pose another question for you to ponder at your leisure. It is just and extension of what you have worked so hard on with this latest post.

The national debt is divided into two parts---debt held by the public and "Intergovernmental transfers"--borrowing from the various trust funds. I read alot about how the latter does not matter much to economists and policy makers because it does not affect the financial markets. It is the Public Debt that makes the most impact on the financial markets because of the potential for the "crowding out effect on private investment" (which I have to teach as part of the College Board Curriculum). Do you agree with this? I fear I over-emphasize the raiding of the trust funds to students and I dont want to fall into a partisan trap. I try to stay on an even keel and let the students make up their own minds...tThanks!!! :)

LiminalHack said...

How do real and nominal interest rates figure in all this, in your view?

Jazzbumpa said...

Gene -

I suggest you have a look at Krugman's blog relative to crowding out. IIRC, at or near the 0-interest bound, where we've been living for a while, there isn't any.


The Arthurian said...

Liminal -- Certainly interest rates, both real and nominal, have become increasingly important in the decades since WWII, as the reliance on credit increased.

Apart from that, I don't have much to say about real rates. I have spent little time on that particular aspect of the problem. Perhaps you can see something I cannot?

LH said...

All I would say is that you seem focus on the stock of debt (and the implied cost/drag) of that stock, without much recognition of the fact that the cost, if such exists, cannot be independent of the real and/or nominal interest rates.

So its not obvious to me that the total 'cost' of the debt load has changed materially since the war.

The Arthurian said...

LH -- You deny that debt has a cost. That is a little funny. By way of contrast, Krugman recognizes the cost of debt, but denies it has any significance. Now, that's hilarious!

LiminalHack said...

Arthurian, it may have some costs or may not but my issue is that:

* you have not quantified in detail what the cost is or how that cost actually arises.

* you don't factor interest rates into what the cost is.

I'm willing to entertain the notion for the sake of argument that debt has a net cost of some kind, but some more details need to back that up. And also, I think we can agree the cost is not independent of interest rates.