RSS
Home
Could America go bust?
Print
Email
RSS
Subscribe
Could America go bust? asks Robert Samuelson this morning. The answer is of course yes, for many countries have in fact done so by repudiating parts or the whole of either their foreign or domestic debt.
For those interested in the gruesome details "This Time is Different" is an excellent listing of all those places that have done so, when, how, why and what the results were.
At the heart of Samuelson's column is this true and obvious point:
The American situation is similar. Despite huge deficits, interest rates on 10-year Treasury bonds have hovered around 3.5 percent. In time of financial crisis, investors have sought the apparent sanctuary of government bonds. But the correct conclusion to draw is not that major governments (such as Japan and the United States) can easily borrow as much as they want. It is that they can easily borrow as much as they want until confidence that they can do so evaporates -- and we don't know when, how or whether that may happen.
One thing that we need to understand about government debt: it very rarely actually gets paid off. For that to happen would mean that the Government was running a budget surplus: this isn't something that has been common in Washington for a very long time. In my own native England it's happened in 7 years of the last 64: not a really great record in trying to reduce the absolute amount of debt.
This means that when debt comes up to be repaid the investor does of course get back his money on the bond that has just matured: but the government simply issues a new bond to someone else to get the money to pay him. This is called "rolling over" debt. And this is just fine most of the time. However, it is possible for things to go awry: think of what might happen if you keep your credit card debt reasonably level, even at a high level. As long as the card company allows you to roll over that debt each month you can keep going: but if they either start to ask you to reduce the balance, or as an alternative to pay a higher interest rate, you can quite quickly get into trouble.
So it is with those government bonds being rolled over. As Samuelson points out, the Treasury can borrow long term at 3.5% at the moment. But what happens when that debt becomes due? It will be rolled over, new bonds will be issued: but what interest rate will they pay? If it's significantly higher then the whole system is in a whole heap of trouble: far from borrowing (to pay for spending from 5, 15, 30 years ago perhaps) being affordable, it quickly becomes not so. The rough rule of thumb is that when such government debt interest payments go over 12.5% or so of total GDP then there's going to be a default. If interest rates of US debt rise substantially, we could be looking at that sort of scenario in the not too distant future.
There's one other point to consider as well, this from "This Time is Different". One of the signs that debt is becoming difficult to sell is not so much that interest rates rise on the debt being sold. It's that the maturity of the debt being sold starts falling. That is, instead of selling a lot of 20 or 30 year bonds, the Treasury sells many more 1 and 2 year bonds. These of course have to be rolled over much earlier (and many more times of course) and this increases the sensitivity of the national debt to a change in interest rates. Meaning that, other things being equal, we're at more risk of a hike in interest rates substantially raising the interest that has to be paid and thus the likelihood of default.
Indeed, the authors point to a shortening of the maturity of the government debt as a sign, a portent, that a default is more likely. So what is happening to the maturity profile of the US debt?
That's because in the past two years, the percentage of the growing debt pie that is being financed short-term (maturities of less than one year) has risen from 30 percent to more than 40 percent -- the highest level since 1980.
Ah: absolutely no one at all is suggesting that this debt is going to get paid off: it will need to be rolled over.
As David Rosenberg, of Gluskin Sheff, noted this week, all this short-term borrowing by the US government ultimately creates a massive rollover risk, as trillions of obligations will need to be refinanced in the next year. What's more, it leaves the nation at the mercy of the credit markets, due to the very real potential that short-term rates could shoot up, a move that would add billions per year to the cost of debt service.
To the 12.5% of GDP level that makes default a near inevitability? No, not yet, a few more things would need to happen. But not all that much: Current national debt is north of $11 trillion, projected deficits for the next decade appear to be about $9 trillion or so. Current GDP is $14 trillion or so: it's increasingly likely that gross debt will rise over 100% of GDP (the Obama Administration projects as much). At 3.5% interest rates that's not really a problem. At 12.5% interest rates that's the sort of number that will likely trigger a default.
Who really wants to bet that interest rates won't rise that much? Well, obviously, everyone is indeed betting that they won't but is that all that wise a bet to be taking?
Especially since the Treasury's actions in borrwing short make the national debt more vulnerable to a rise in interest rates: and the very fact that they are borrowing short has historically been a sign that a treasury is having problems borrowing long for fear of raising interest rates.
