Trills, warrants and kickers

  This post is not about organized crime. This post is about a proposal from Robert Shiller and Mark Kamstra featured in today’s New York Times. (Sanchez got the first down! At the opposing 46.)  The idea is for the government to issue bonds whose payout is linked to GDP. (Sanchez sacked!  Commentator yells, “Are they trying to kill Sanchez?”) In the case of Shillers proposed “trills,” the payout would be directly linked:  each one would yield one-trillionth of quarterly GDP every three months.

It’s a fine idea, although I suspect that the benefits would be, well, limited. They protect against inflation, but so do inflation-linked bonds, and investors would presumably demand some sort of equity-ish premium. That said, they’d offer an easy way to diversify ... you get to buy a security linked to the income of every single income-earning entity in the country! On the other hand, the income stream won’t come directly from those entities, but via the tax revenue that they generate for the federal government. 

(Holy s--t! Did that just happen? Challenge. Why yes, it did. Two-point conversion, 18-15.)

But that is not what this post is about. This post is just to point out that Argentina issued something not unlike the trill back in 2004. The “GDP kicker,” pronounced “keek-air,” aka a warrant, offered creditors payments equal to one-twentieth of the dollar value of all GDP growth above a threshold of 4.2% per year. Bosnia, Bulgaria, Costa Rica, and Singapore have issued similar securities. None of them are as simple as the trill, of course, but still. When have you heard of Argentina being a financial pioneer, in a good way?  (Field goal! 21-15.)

Of course, there are some people who think that Argentina gave away the store with the kicker. And they may be right! Which is what worries me about the trill. It might turn out to be a rather expensive way to finance the borrowing needs of the government ... and that, ultimately, is what I tend to think government debt is for. Although Alexander Hamilton would not agree.

Comments

[David Allen]

[From the linked article] "Each trill would represent one-trillionth of the country’s G.D.P. And each would pay in perpetuity, and in domestic currency, a quarterly dividend equal to a trillionth of the nation’s quarterly nominal G.D.P."

The "in perpetuity" part really makes no sense. You can only sell so many eternal trills before they start to be a drag on the economy. On the other hand, a five year or ten year trill might be more practical.

One possibility would be for times like now, when anyone buying a long bond is looking at an eventual loss when interest rates go up. A ten year trill, OTOH, would be interest-rate protected, inflation protected, and likely to grow as the economy grows. The Treasury could sell a bunch of them now,and later, when the economy starts to grow and interest rates go up, shift to selling fixed-rate securities. Could be useful.

[Bernard Guerrero]

re: Perpetuity, google "Consol"

Also, "C/r"

[Doug M.]

I assume you've already seen the Felix Salmon column?

http://blogs.reuters.com/felix-salmon/2009/12/28/the-return-of-gdp-bonds/

Doug M.

[David Allen]

're: Perpetuity, google "Consol"'

'CONSOL Energy Inc. (NYSE: CNX) is the largest producer of high-Btu bituminous coal in the United States.'?

Seriously, I knew about consols. There are two big differences between them and trills, if I'm understanding and remembering correctly (not guaranteed). For one thing, British governments back then tended to be much more religious about balancing the budget than American governments are now. More importantly, the payment on a consol is a set amount, not a set fraction of GDP.

Either a set (or limited) payment in perpetutity or a fraction of GDP for a set number of years would work. The combination doesn't.

[Bernard Guerrero]

David, I think the two situations you describe are pretty close in terms of ultimate effect.

a) A consol or other standard perpetuity in a low-inflation, low budget deficit world

b) A trill (nominal GDP indexed) or other inflation adjusted perpetuity in a world where inflation is not insignificant

The trick with consols is that the British government retains the right, to this day, to redeem them at par. Return the capital and the perpetual debt service goes away. There is no reason I can think of why you couldn't make trills callable. There remains the practical problem of large deficits as far as the eye can see, but that's why the capital is looking for inflation hedges in the first place, eh?

Noel's right that it might prove to be a very expensive way to float national debt, but that's partly because the government is giving up the option of inflating the currency and nominal GDP in order to lower its real debt-service burden. The question is, will the current situation (where the market still doesn't ask for a lot of protection on that front) continue to obtain going forward? I think Doug just put up a post on an unrelated debt topic that falls into a similar trap.

[Noel Maurer]

Hi, Bernard,

That wasn't quite what I was thinking when I wrote that trills might prove to be an expensive way to raise money. Rather, I was thinking that in general equity is more expensive than debt, and trills are sort-of equity-like.

In addition, they have the weird effect of causing the governments interest-rate burden to increase at the rate of economic growth, unlike fixed-rate issues. Growth only causes the interest-rate burden on fixed-rate issue to rise as debt is rolled over, assuming that rates are higher when nominal GDP is growing quickly.

The removal of the inflation option runs in the other direction, I think, reducing borrowing rather than increasing them. No?

[Bernard Guerrero]

Would it necessarily reduce the expense of borrowing? I'm picturing it this way:

1) The government has a relatively fixed need to borrow

2) The market is willing to finance said borrowing at rate X

3) However, given the need to guard against a natural tendency on the part of government to try and inflate away the debt, it requires an additional Y as protection

4) Government might be able to commit to not using the inflation tool, via some instrument like TIPS or trills, which should bring the pricing back down to X from X+Y

5) A long-sighted rational government, having given away one very powerful tool for controlling its debt burden, might try harder to balance the books and reduce the need to borrow, as per your last sentence

5B) But when was the last time anybody saw one of those? The consistent political pressure on any government is to spend more and tax less. Everybody wants a piece (or two.) The only recent example of a balanced budget that we have was partly a result of historical accident, with the late-90s bubble inflating. (This is not to denigrate Clinton's relative discipline; Bush also had a bubble to work with.)

So I guess that giving up the inflation option would tend to lower the coupon cost of borrowing even as it gives government an additional incentive to borrow less. I just don't see the latter dominating the former.

"they have the weird effect of causing the governments interest-rate burden to increase at the rate of economic growth, unlike fixed-rate issues"

Fair point, though it's increase at the nominal rate of growth. Revenue extraction should be capable of growing at about the same rate, no?

I'm not really sure that I disagree that it's a bad idea, BTW. It seems to me like a solution looking for a problem. If the markets eventually become so spooked by the idea of inflation that the government can't borrow, then it can be revived. In the meantime, the TIPS aren't showing overwhelming demand or fear.