Jarda Borovicka writes to me

In the whole discussion about whether the U.S. should borrow more now when the interest rates are low, I miss one crucial thing – what happens when the debt comes due?

Risk-free debt is really risk-free only when the maturity also coincides with terminal repayment. But assume that the U.S. borrows an extra trillion of dollars now, due in 10 years (the average debt duration of the U.S. debt is something like 4 years?). Sure, the interest rate is low, but the borrowing is cheap only as long as we assume that during the 10 years the U.S. repays this whole extra debt, compared to what would have happened in the baseline world.

If not, the U.S. will need to refinance this debt in 10 years, and potentially at much higher interest rates. There is a potentially very large risk looming. Greece has experienced this the hard way – it collapsed not because it necessarily needed new borrowing but because it had to roll over the old borrowing at impossibly high rates.

And I have not heard the advocates of more borrowing to suggest any credible offsetting mechanism that would lead to repayment (not replacement with other borrowing) of this extra debt at or before maturity. But then the idea "let's borrow now because it's cheap" becomes seriously flawed.

I would compare those who advocate such outright borrowing without committing to credibly repay at maturity to people who fall for teaser mortgage rates and are rather negatively surprised to see the rate adjust later. It is interesting though that there seems to be a large positive correlation between people who advocate government borrowing because rates are low NOW, and those who call for protecting consumers against reckless lenders who tease them with a low temporary rate. To quote a famous Canadian singer, Isn't it ironic?


Indicates that 30 yr yields are only 0.5pc higher than 10yr yields. 30 years sounds like a much better way to get deeply in debt.

Isn't the obvious argument that cheap credit right now could finance more stimulus, and that stimulus would increase revenues to pay for the additional debt incurred?

Enough inflation to bring the debt to a manageable fraction (half??) of nominal GDP is the only way I can imagine.
Paying off the debt requires taxes higher than spending for several years. We rarely elect politicians who will do that.
I would love to have an economist to refute this since inflation,e.g. 100%, could lead to extreme social unrest among babyboomers who have tried to save for their retirement and would be hurt badly. Any takers?

Isn't it ironic?

Only if you dont understand the difference between a budget constrained household and a fiat currency issuing government.


I think the "difference between a budget constrained household and a fiat currency issuing government" may be the second biggest red herring.

I guess this post shows why I am not a macroeconomist. It has always seemed obvious and critical to me that the carrying costs of the debt we are amassing could (and would) skyrocket when interest rates rise. And, this seems far more serious to me than many of the obtuse discussions of macro modeling. But I guess I just don't understand macro after all.

Jarda Borovicka articulates well a key point in the current debate on fiscal stimulus. Another point worth emphasizing is that the classical Keynesian fiscal policy prescription calls for the deficit to act as a stabilizer, with surpluses in good times coupled with deficits during recessions, so that the budget is balanced over the business cycle. The idea that the government would both run large deficits in normal and boom times, and run much larger deficits during recessions, would surely have appalled both Keynes himself and those who created "Keynesian economics" in his name.

If the downturn is cyclical, and the stimulus programs funded by the debt will in fact spur growth, then yes, the extra debt can be paid back.

Now if the downturn is secular, for example due to natural resource contraints, and the period of constant growth interrupted by short (less than ten years) recessions, then the extra debt will be a problem, since the stimulus wouldn't work to rejuvenate the economy. But if the stimulus doesn't rejuvenate the economy, its not really worth trying anyway. But then who knows what happens to interest rates in this scenario.

So its critical to figure out if this is a "normal" or cyclical downturn, like all the others we have had since the industrial revolution, or if we are at the end of the technology-spurred long period of growth started by that event. But know one except for a few fringe environmentalists is considering this.

30 year bonds? Now is the time to issue perpetuals!

Isn't it awesome how people don't seem to grasp the difference between a economic and a military argument?
Jarda is obviously quite right that it's impossible for a government to repay a debt that was spend on any kind of so called "stimulus". Be it booze and hookers, education for retards or whatever.
But as JSK points out, there is a difference between a household and a government. Academic economists call that difference: "A technology called a nuke...so fcuk off and go to work if you need money".
The U.S. government could inflate, default or just tell Canada to pay its debt. What are they gonna do, throw maple syrup at U.S. forces?
The only reason everything has to be packed in fluffy sounding econo-speak is that these are the rules for polite conversation between grown ups. Or do you guys really think these people at the entrance of your supermarket honestly want to know how you are?

Comments for this post are closed