This contributes to Argentine GDP though. Photographer: Diego Giudice/Bloomberg News
This contributes to Argentine GDP though. Photographer: Diego Giudice/Bloomberg News

A fun category of securities is "ones where the issuer gets to decide if or when or how much it feels like paying you." There are the obvious ones: Common stock never needs to be paid back, and companies only pay dividends if they feel like it.1 There are the almost as obvious yet still somehow jarring ones: People get really mad when their preferred stock stops paying dividends.2 There are the devious and highly structured ones, where deep in the back of the document there's some legalese saying "but we don't have to pay you if we don't feel like it."3

And then there's the weird category of indexes that the issuer can mess with. Take oh say Libor. Banks wrote swap contracts requiring that they pay customers Libor.4 Libor was generally defined as "the rate at which banks can borrow from each other," but Libor was specifically defined as the result of a process in which banks submitted numbers to the British Bankers' Association and those numbers were averaged and so forth. So if those numbers were not, in fact, the rate at which banks borrowed from each other, that was irrelevant to the swap contract: You didn't get paid "true Libor"; you got paid Libor. That's all the contract said. If banks wanted to make up Libors that benefited them, then their swap counterparties had no right to complain.

Though, I mean, they've found plenty of ways to complain.5

One guess you might make is: If you get to decide how much you want to pay on your liabilities, you'll probably err on the side of paying a pretty low number. But I'm not sure that's empirically true; companies pay preferred stock dividends for lots of reasons,6 and I sort of assume that banks lost money manipulating Libor.7 And many companies affirmatively like paying dividends on their common stock, and are sad when they can't.

Anyway, here is an amusing story about Argentina. Argentina, you may have heard, ran into some problems a while back, and swapped some of its old debt for new stuff. Among the new stuff was a thing usually referred to as a "GDP warrant," under which Argentina agreed to pay holders 5 percent of the amount that its gross domestic product exceeds some threshold growth rate (this year, 3.22 percent). Except that, as a reader puts it in an e-mail, "the warrants technically aren't tied to Argentina's GDP growth. Rather, they're tied to what Argentina reports as its GDP growth, which often is an entirely different animal."

Yep! Here is the prospectus; the terms of the GDP warrants are on pages S-64 to S-66. The key definition is: "'Actual Real GDP'' is the gross domestic product of Argentina in constant pesos for each calendar year as published by the Instituto Nacional de Estadística y Censos ('Indec')." Indec is an Argentine government agency that has been known to get its GDP numbers wrong; last year the International Monetary Fund went so far as to censure Argentina for mis-reporting its GDP.

Too high, though. Its reported GDP was too high. Which was good for the GDP warrant holders, who got paid based on GDP numbers that even they thought were fictional.8 But yesterday's news is that Indec has revised its process, and its numbers, to be more in line with reality:

GDP warrants tumbled 30 percent today to 6.19 cents on the dollar at 1:04 p.m. in Buenos Aires, according to data compiled by Bloomberg, after the national statistics agency said yesterday the economy grew 3 percent in 2013, below the 3.22 percent trigger. Government dollar bonds rallied as the missed payment will preserve reserves used to pay debt that are at a 7-year low.

That will save about a $3 billion payment, and it is sort of hard to criticize:

“This clearly surprised the market,” Siobhan Morden, the head of Latin America fixed income at Jefferies Group LLC, said in a phone interview from New York. Warrant holders “were so used to receiving this perverse subsidy” ...

“This is good news for the rest of the market because it reflects the reality” of the economy, Jorge Piedrahita, chief executive officer at brokerage Torino Capital LLC, said in an e-mailed reply to questions. “To spend a couple of billion dollars on a fiction would have been Kafkaesque.”

But, remember: They were doing it! They were spending billions on a fiction. The news is that they stopped.

I don't know what to tell you, this made me smile. Argentine GDP warrants were not a bet on Argentine GDP. They were a bet on reported Argentine GDP, and economically the reporting errors swamped the actual growth rate. Your payoff came from the perverse subsidy, and your calculation was not "Argentina's GDP will grow at more than 3.22 percent a year" but rather "Argentina will have its own reasons to report that its GDP grew at more than 3.22 percent a year."

Those reasons are sort of obvious -- high GDP growth is good for political popularity, attracting investment, etc.9 -- though there are equally obvious reasons to cut it out, particularly that fake GDP reporting is bad for attracting investments, etc. But the trick in investing in things where the other side can just decide what to pay you, is predicting what they're going to pay you. It's predicting motives, not finances. And since there are a lot of those things, that's a lesson that applies beyond Argentina.

1 And sometimes not even then.

2 Examples are numerous but, you know, check in on the Fannie/Freddie preferred holders sometime and see how mad they are about their vanished dividends.

3 Hahaha these totally exist.

4 Also loans where they received Libor, which are less relevant for our purposes. "Contracts where the owner gets to decide how much the issuer pays it" are another interesting category.

5 Including, most notably, that the contract didn't say that, which is not a theory that I completely buy. But also extra-contractual stuff like fraud and market manipulation and antitrust.

6 Market access, reputation, not making creditors nervous, wanting to pay common dividends, plus if you don't pay dividends, many preferreds make you give the preferred holders board seats.

7 I have no strong basis for that (here is my weak basis for it), but the whole understating-Libor-to-not-look-like-you-can't-borrow thing was at least non-economic.

8 On the other hand, Argentina's reported inflation was too low, which was bad for inflation-linked bondholders: "The gap between the official and private inflation rates has enabled the country to save about $6.8 billion since 2007, according to Buenos Aires-based research firm ACM Consultores."

9 There's a story I want to tell where GDP warrants are like "equity" in a country, and issuers have fiduciary duties to, and are fond of, their equity holders, and that biases them towards rewarding the equity holders even at the cost of debt holders. I don't think this story especially fits here -- nobody really thinks of these things as equity holders, and certainly no one thinks of them as being owed fiduciary duties -- but there are maybe some useful parallels; in particular, they're a bet on the "earnings" of the "enterprise," and everyone wants to see those earnings go up. So, you know, aligned incentives.

To contact the writer of this article: Matt Levine at mlevine51@bloomberg.net.

To contact the editor responsible for this article: Tobin Harshaw at tharshaw@bloomberg.net.