Debt burden

Tuesday, May 13th, 2008

Megan McArdle argues that increasing levels of debt do not necessarily imply a greater debt burden:

I would argue that this [chart of increasing debt] doesn’t tell us what we really want to know, which is: how much is this debt costing us?

Look at the first chart. What’s striking is how much the top line [debt as share of disposable income] differs from the bottom line [debt as share of assets]. If what we were seeing was America plunging itself into debt to finance consumption, they should all be rising roughly in line with each other. In fact, the top line breaks away from the others.

What’s going on here? Ezra misstates just slightly: it’s not all debt as a share of income, it’s all debt as a share of disposable income. That is to say, income after taxes. In 1949, personal disposable income was 95% of GDP; by 2007, it was 89%. So part of that increase is simply that the share of income dedicated to taxes has risen substantially.

That is not, of course, the entire story. There’s also the fact that in 1940, homeownership rates were around 45%; it’s now almost 68%. All of that transition was financed by debt, and not only by debt, but by a dramatic shift in the type of debt: the emergence in the 1950s of the thirty year fixed-rate amortizing mortgage as the dominant form of home financing. Prior to that, mortgages had been much shorter, and usually featured balloon payments.

Meanwhile, the dramatic rise in effective income taxes on the middle class at federal and state levels made the mortgage interest tax deduction much more valuable, encouraging people to take on more debt. As you can see, most of the increase is actually housing debt.

On the revolving debt side, you’ll notice that the largest increases take place in the 1950s, 1960s, and 1970s, is basically flat in the 1980s and 1990s, and then ticks up again in 2000.

This represents a number of different trends: first there was the auto loan revolution in the 1950s and 1960s; then came the introduction of Diner’s Club, shortly followed by Amex, in the 1960′s. Almost all of this debt was either secured by automobiles, or paid off each month; the latter represents float, not real revolving debt.

In the 1970s, the effective repeal of bank usury laws made Mastercard and Visa ubiquitous, causing debt to march upwards again. This is actual credit expansion. But it’s hard to be sure how much, because this era is the death of another kind of debt: installment buying.

If people massively expand their asset base by a house and a couple of cars, not to mention labor-saving appliances like dishwashers and washing machines, 40% of it all debt financed, this is not an obviously worrisome trend. Especially since a lot of that represents a shift from expenses like rent to debt payments, which is not actually a net deterioration in people’s finances. Nor is it clear that we should mourn for the days when the repo man could take away your furniture, television, and appliances.

Leave a Reply