Warren's audience seems not to have noticed that this is kind of a weird number to pick. Having more of a particular kind of debt doesn't tell you much; if you take out a home equity loan to pay off your credit card, your financial position has (arguably) slightly improved, even though your housing debt just jumped. What you really want to know is the overall level of indebtedness. Warren covers this by implying that she's just using revolving debt as an example; "I could have used consumer loans or mortgage debt", she says, and "it would have been much the same".
This is desperately, hopelessly wrong. Not an exaggeration or a misreading; just flatly untrue.
I cannot read the little squib at the bottom of Warren's chart to find where she got her data, but the gold standard for this sort of thing is the Federal Reserve Board, which among its many other functions keeps tabs on how much credit is sloshing around various markets. The class will please turn to FRB web page G.19, "Consumer Credit Outstanding (Millions of Dollars; Not Seasonally Adjusted).
You will note that the Federal Reserve didn't even begin measuring revolving debt as a separate item until January 1968; this was the birth of the modern credit card era. You may also notice that revolving debt increased fourfold just between then and the end of 1970. You will probably further note that the growth of non-revolving debt slows down around then, as credit cards began substituting for installment buying.
In December 1970, the American public carried $5 billion worth of revolving debt, and $128 billion worth of non-revolving (secured and fixed-term debt). In December of 2004, the American public had $1.3 trillion worth of non-revolving debt--and $823 billion worth of revolving debt. One number had increased by a factor of 10; the other by a factor of 160. Saying that these two increases are "much the same" is like saying that a newborn kitten is pretty much indistinguishable from a full-grown lion.
Of course, these figures are in nominal dollars, which is to say, not adjusted for the inflation that has taken place. In constant 2004 dollars, according to the excellent inflation calculator provided by the Bureau of Labor Statistics, the outstanding non-revolving debt was $625 billion, while outstanding revolving debt was $24 billion. It doesn't take much math skill to see that the increase in the two figures is nowhere near the same magnitude.
Warren then goes on to do something really, really weird: she puts up a chart showing the savings + revolving debt as a percentage of personal disposable income. This is a useless figure, though her audience appears to think it is deeply significant. For one thing, it doesn't show all forms of debt, which is what we really care about. For another, to get a good financial picture, you don't add those two things; you subtract, which tells you whether you're running a surplus or a deficit. But the biggest problem is that she's comparing a stock to a flow. A stock is everything you have--the inventory in a store, say. A flow is the shipment of canned goods you got in today. You never compare the two directly; it's meaningless. If I tell you that your mortgage is 80 times the size of this month's 401(k) contribution, are you saving too much or too little?
It's especially useless because we don't know what the interest rate on the revolving debt is. In the high-inflation early eighties, the debt would have been smaller relative to people's incomes, but high interest rates would have jacked up their monthly payments; by the late 1990s, the reverse would be true.
The correct comparison would be total assets to total debt, or savings to debt service payments. As it happens, we have some data on that; though the available time period is not quite the same as Warren's, it is close enough for government work. In the Federal Reserve's 1964 survey of consumer finances, the average household net worth was $22,588 ($137,641 in constant 2004 dollars). In 2000, it was about $200,000 in constant terms.
But we're so much more unequal! People will cry. Bill Gates is screwing up the average. Indeed this is true . . . but apparently he was back in 1964 as well. Back then the median net worth was $7,550 ($46,000). In 2000, it was $55,000. To be sure, not an impressive increase. But families hadn't gone backwards--and the 1964 figure counts some things the 2000 figure doesn't, like life insurance policies. And the number of families with negative net worth was about the same in 2004 as it was 40 years earlier: roughly one third.
Having seen that the stock has, in fact, improved for both mean and median households (insofar as we can tell from discontinuous data series), let's look at the flow: debt service payments. Warren's weird chart implies that they must be piling up on households like never before. Well, this is kind of true.
To examine how much it has increased, we turn to a little Fed publication winningly entitled "Household debt service payments and financial obligations as a percentage of disposable personal income; seasonally adjusted". Memorize that: you can recite it at parties.
Now where was I? Ah, yes. Unfortunately, the series only starts in 1980--recession territory. Back then, households were using 11.13% of their personal disposable income to service mortgages and consumer debt. By mid-2004, when Elizabeth Warren was giving that speech, the DSR had soared to . . . 13.5%. All of which seems to have been housing debt and property tax payments, since the Financial Obligations Ratio (the DSR + auto leases, property taxes, and homeowner's insurance) for renters actually fell slightly.
Americans are simply not being bankrupted by their credit cards and mortgages. 2% of my income is a lot--I sure wouldn't want to have to write that kind of check out for no reason. But it wouldn't push me into bankruptcy, or even out of the middle class.
This article available online at: