Wednesday, August 18, 2010

Tracking Progress on Deleveraging

The Federal Reserve Bank of New York has started a great new quarterly report with a ton of information on household credit status in the U.S.  This will help us to assess the process of deleveraging as it continues.  The data is based on a random sample of credit reports from Equifax, one of the three big credit reporting agencies.

The most interesting graph is this one for overall household debt and it's composition:

The only problem with this is that these are nominal figures, and so tend to overstate the extent of the debt run-up, and understate the progress in paying it down (since even if nominal debt stays flat, inflation will slowly erode it in real terms).  One possibility would be to inflation adjust the data, but instead what I decided to do was compute the ratio to disposable personal income (ie after taxes) from the BEA as this  is what we really care about - how big is the debt relative to the income potentially available to service it.

The blue line is the data, and the red line is a rough extrapolation of the pace of deleveraging of the last eighteen months.  At its peak, total household debt was 115% of DPI, and it's down to 103% as of Q2 of 2010.  At this pace, it will be paid down to a reasonable level in less than a decade (obviously many things could happen to change the pace of deleveraging, so this should be seen as a scenario, not a forecast).

Under the circumstances, that seems like a decent pace of progress.


basil said...

What program do you use to make your graphs?

Stuart Staniford said...


Excel usually.

basil said...

Hm, the graph looks excessively shiny for Excel... Maybe I have an older version.

Anyway, interesting analysis as always.

Stuart Staniford said...

The "Total Debt Balance and It's Composition" one is a screenshot from the fed report.

Alexander Ac said...

Here is longer-term graph of the debt:

so we have a long way to go down. I do not think the process will be linear, rather I think it is reinforcing itself in positive feedback loops... anyway, the recession is far from over, I think


Nick G said...

I'm confused by leverage.

How do we know what is a "correct" level?

For instance, a societ-wide move away from renting will raise leverage, but has society's financial risk really increased?

rjs said...

stuart: is this also being tracked?

Fidelity Sees Record Number Raid Their 401(k)s

In the wake of news about a spike in new applications for unemployment benefits comes another potentially troubling sign: A record number of workers made hardship withdrawals from their retirement accounts in the second quarter.
What's more, the number of workers borrowing from their accounts reached a 10-year high, according to a report issued Friday by Fidelity Investments.

Stuart Staniford said...


Yes, I agree, it's very hard to be precise about it. I had a few thoughts about it here. I think we can say with some confidence that it was quite a bit too much in 2007, and in prior similar episodes it fell a long way (to a half or a third of the peak level) before reaching a final bottom. But I'm not aware of any way of saying with high confidence that it must be so in this episode, or that there can be some way of avoiding it.

George said...

Stuart, the debt to income ratio has much further to fall. See a 100 year chart here instead of the 10 year chart that you put up: