Actually he neglected to mention this fact in his column this morning. (It's less than 1.0 percent of GDP and only about 0.5 percent of GDP if we net out the interest rebated by the Fed.) Samuelson tells us:

"The true national debt could be triple the conventional estimate, anywhere from $11 trillion to $31 trillion by my reckoning. The differences mostly reflect explicit and implicit “off-budget” federal loan guarantees. In another economic downturn, these could result in large losses that would be brought “on budget” and worsen already huge deficits. That’s the danger.

"My purpose is not to scare or sensationalize. It’s simply to illuminate the problem."

Actually, Samuelson may have inadvertently done the latter.

If you want to make the jump from the $11 trillion commonly used number, or the $16 trillion debt subject to the legal debt ceiling, to get to Samuelson's $31 trillion, you have to add $2.9 trillion in loan guarantees (largely student loans and small businesses), $5.1 trillion in mortgages guaranteed through Fannie and Freddie, and $7.3 trillion in federal deposit insurance. What's neat about these additional debts is that they are tied to assets.

In the case of small businesses, the assets are the businesses. In the case of mortgages, the assets are the houses. In the case of deposit insurance, the assets are the deposits and the banks' assets. (I left out student loans -- we can't force people to work, but it is pretty hard to imagine a situation where all of our doctors and lawyers can't pay any of the debt they owe.)

Should we be worried if the government has more debt, even if it is tied to assets that have considerably larger values? It is difficult to see why. Is a business that has debt of $2 billion and has $10 billion in assets in worse shape than a business with no debt but only $5 billion in assets? In Samuelson's scare story it is, because he only looks at liabilities, not assets. 

We certainly should have been worried about Fannie and Freddie's debt when they made massive commitments in a hugely over-valued housing market. Some of us, who don't have access to the Post's opinion pages, did raise such concerns. But now that the housing market has fallen by 30 percent in real terms, what is the probability of large-scale losses on loans that generally do not exceed 80 percent of market value?

Similarly, if we don't get the Robert Rubins of the world again playing crazy games at government insured banks, is it necessary to spend a lot of time worrying about the $7.3 trillion in insured deposits? (Wait, I forgot Robert Rubin is still a highly respected person in DC policy circles, maybe we should be worried.)

Seriously, in a normally functioning economy these liabilities should be no concern whatsoever. If Samuelson wants to suggest (he does) that the economy could go into a further downturn of Great Depression order of magnitude, then these debts would pose a problem for the government, but who gives a damn? If we are facing a decade of double digit unemployment, then worrying about the government's fiscal obligation is a trivial after-thought that is not worth the time of serious people. It's like asking how we pay for Social Security after a nuclear holocaust. That concern might employ people at the Washington Post, but it need not detain the rest of us.

The moral of the story is that we should focus on getting the economy growing again. If we did, we would have more people employed, we would be building up our capital stock (both public and private) and children would be doing better in school because they are being raised in homes with parents who have stable employment, and these government guarantees would be of no concern whatsoever.

Give Samuelson a lollipop for finding yet another way to scare people about the government's debt and then get back to thinking about real economic problems.