Do I hear $202 trillion?

By Buttonwood

LARRY Kotlikoff of Boston University has another Bloomberg column on the state of the US national debt in which he declares that

Our country is bankrupt. It's not bankrupt in 30 years or five years. It's bankrupt today.

His calculation is that the Federal debt is not $9 trillion as the (net) figure officially states but $202 trillion. How does he arrive at that number?

In a sense, this is a process rather like the one used to derive the theoretical value of a company, involving the discounting of cashflows to a present value. Professor Kotlikoff takes the process rather further than most, tracking the revenues and expenditures all the way out to 2085 and then calculating a "terminal value" for the post-2085 numbers. The revenues and expenditures are increased at a 2% real rate, to allow for GDP growth and then the debt numbers are discounted at a 3% real rate.

He takes the underlying numbers from the alternative fiscal scenario calculated by the Congressional Budget Office. These figures are based on assumptions in the long-term budget outlook, which presuppose that tax reform does not take place (see page 3 of Chapter 1 for the assumptions, which include the permament extension of the Bush tax cuts).

Now $202 trillion looks a very scary number but how does it compare with GDP? As commenters on my original draft pointed out, there is not much point in calculating future GDP and discounting it back (my misinterpretation of the professor's complex spreadsheet). So given current GDP is somewhere north of $14 trillion, the US debt-to-GDP ratio is almost 14 times, worse even than Greece's position when calculated on the same basis.

The real problem is the issue of labelling. As Professor Kotlikoff points out, the government makes promises to pay citizens benefits in the future. If it borrows the money to fund those promises, this counts as debt and goes into the official debt-to-GDP ratio; if it simply assumes that the promises will be paid out of future taxes, the debt doesn't count.

But this an accounting trick of the sort practiced by many a public company in the past. To boost current profits, compnaies have tended to recognize revenues early and costs late. (A good example is counting the assumed rate of return on the pension fund as profit, even when it isn't achieved.)

As for the headline, it all depends on what one means by bankrupt. Clearly the US has no problem financing itself on the markets at pretty low rates. What the calculations show is that the US will eventually default on its promises to somebody - bondholders, taxpayers, future pension or healthcare recipients. The last three groups are the most likely sufferers. But it doesn't tell us when.