The phrase "contempt of court" is a little funny. Normally it just means that, if a judge tells you to do something and you don't do it, the judge can throw you in jail for contempt. If you are Argentina, on the other hand, and you don't do what New York federal Judge Thomas Griesa tells you to do, he can't throw you in jail, because you're way over there in Argentina, and also a country. So it doesn't do him much good to find you in "contempt of court," even if you've expressed your actual contempt for the court in the form of personal attacks on him in advertisements and press releases and speeches.
Anyway, a while back Judge Griesa ordered Argentina:
- not to pay any of its exchange bondholders (the bondholders it likes, the ones who settled with it in 2005 and 2010) without also paying holdout bondholders (the "vultures" it dislikes, who still want full payment from Argentina's previous default), and
- not to do any sort of debt exchange to get around the first order.
And the other day Argentina's government introduced a bill in parliament to do a debt exchange to get around Judge Griesa's orders, which he specifically told it not to do. So yesterday the holdout bondholders went to court to try to get Judge Griesa to hold Argentina in contempt, and Judge Griesa basically said, well, this is pretty contemptuous, but he still didn't hold Argentina in contempt. Because what good will that do? They've ignored him so far.
Still, though, Judge Griesa reiterated -- and then re-reiterated -- that Argentina's proposed exchange would be illegal,1 and that anyone caught helping Argentina to do the exchange would be in Big Trouble. This probably makes the straightforward exchange proposed by Argentina -- investors hand Argentina New York-law (or English, or Japanese) bonds and get back Argentine-law bonds -- impossible, as various U.S.-based intermediaries would need to be involved to make it work.2
But "impossible" just means "not sufficiently engineered yet." Several people have suggested a possible way around Judge Griesa's orders that goes something like this:
- Argentina issues new local-law dollar-denominated bonds for cash to ABC Corp., a private company not subject to U.S. jurisdiction.3
- ABC Corp. buys New York-law bonds from investors for cash, clearing those trades in the usual way using New York institutions like the Depository Trust Co.
- ABC Corp. sells the new local-law bonds to the same investors for cash, clearing those trades in Argentina using local institutions.
- ABC Corp. only does matched trades: Each counterparty that sells it New York-law bonds will buy back the same number of local-law bonds, for the same amount of cash, so that there's no net movement of cash.
- ABC Corp. ends up holding only New York-law bonds.
- Argentina gives ABC Corp. cash to cancel those bonds.4
The problem this aims to solve is:
- everyone who's obviously working for Argentina needs to be out of U.S. jurisdiction, and
- everyone who's in U.S. jurisdiction needs to have plausible deniability that they're working for Argentina.
So bondholders who sell to ABC Corp. need to be able to say, "What, I was just selling my bonds in the market, for cash, I wasn't helping Argentina's evasion." And DTC can record those trades and say, "What, it was just a sale between two private parties, I wasn't helping Argentina." And ABC Corp. can assist Argentina in the exchange and say, "Okay, you got me, I'm in contempt, what are you going to do about it?"5
(Argentina, of course, already says that.)
It's a theory! Perhaps it solves the mechanical problems of doing the exchange, though I'm not sure. Any flavor of evasive exchange, even an exchange so evasive that it evades being an exchange, is a risky business. Anyone who plays any part in structuring or executing this runs some risk of raising Judge Griesa's ire and being held in contempt under his order.6 I might be in contempt just for writing about it. You might be in contempt for reading about it.
But even if it does work, it's still terrible. Bondholders who accept this exchange end up with new bonds that lack the protection (such as it is!) of the U.S. legal system, and that are issued by a sovereign that is hampered by litigation and still has no immediate prospects of returning to full access to the global financial markets. Even worse, by agreeing to the exchange, they encourage Argentina to avoid reaching a settlement with the holdouts, and thus delay its return it to those markets. Bondholders who reject the exchange, on the other hand, end up as ... well, "holdouts."7 And we've seen how Argentina treats holdouts.
For the current exchange bondholders, the optimum response to Argentina's new exchange proposal might well be to ignore it, keep not getting payments, and hope that something eventually changes (a settlement, a new government) and those payments resume. But that response gets less attractive, and swapping into local-law bonds gets more tempting, with each day of nonpayment.
I have heard one other proposal for a quasi-exchange that solves some of these problems. The basic insight here is that, if you can convincingly tell Argentina how much of each New York-law bond you own, Argentina could just mail you a check for the interest.8 Generalizing from that a bit:
- Someone sets up a trust somewhere outside of the U.S., say Switzerland, to work on behalf of current bondholders.
- The trust is set up as an agent of bondholders, not of Argentina: Its purpose is to manage defaulted bonds on behalf of bondholders in a way that maximizes their recovery.
- Bondholders can sell their bonds to the trust in exchange for trust certificates (listed on the Swiss stock exchange, say) corresponding to those bonds.
- The trust, as a bondholder, negotiates with Argentina on behalf of its participants.
- The trust's offer is: If you pay us an amount equal the interest on our bonds (by check, or by deposit in Buenos Aires, or whatever), even though that doesn't technically comply with the terms of the bonds (which require payment through trustees and so forth), then we will agree not to sue for default, and we will waive our claims about the technicalities of the interest payments.9
- The trust then passes the quasi-interest along to its certificate holders.
- The trust and its payment mechanism stay out of Judge Griesa's jurisdiction.
The crucial benefit here is that the trust keeps the New York-law bonds, avoiding some of the problems of the local law exchange. If things go wrong, the trust and its participants are in no worse shape than they are now: They still have New York-law bonds (albeit defaulted ones (albeit defaulted ones that are paying interest)). And if things go right -- if Argentina reaches a settlement and can resume regular-way payments -- then the trust can be unwound, the bonds returned to the participants, and this all looks like a bad dream. It's a way to move the payments on the bonds outside of U.S. jurisdiction, without moving the actual bonds out of the jurisdiction.
It's a neat trick.10 Don't hold your breath for it to happen tomorrow, but as Argentina's default goes on without resolution, you might expect to see more creativity born of necessity.
I want to be very clear, and I want to state it right now. This proposal is a violation of the current orders of this Court and of the Second Circuit. It is illegal, and the Court directs that it cannot be carried out.
2 See this post: Depository Trust Co. (the depository for the bonds) and/or Bank of New York Mellon (the trustee) would need to be involved just to keep track of who's handed in their bonds for exchange.
3 An Argentine bank would be a natural candidate to be ABC Corp., though you might want a newly created entity -- perhaps owned by several Argentine banks -- to make it a bit more difficult to sanction.
4 The same cash ABC Corp. gave Argentina in step 1, so there's no net cash flow. This is probably more of a rolling trade than an all-at-once one. It's not like Argentina issues billions of dollars upfront to ABC Corp. for cash and then waits to see who takes the deal; it's more like, you find an investor willing to swap, you do all of these steps for that investor, and then repeat as necessary.
5 The answer is maybe "seize ABC Corp.'s U.S. assets," but those would presumably just be New York law bonds that were going to be cancelled anyway.
6 Key language is in paragraphs 2(e) and 2(f):
Such Participants shall be bound by the terms of this ORDER as provided by Rule 65(d)(2) and prohibited from aiding and abetting any violation of this ORDER, including any further violation by the Republic of its obligations under Paragraph 1(c) of the FAA, such as any effort to make payments under the terms of the Exchange Bonds without also concurrently or in advance making a Ratable Payment to NML.
“Participants” refer to those persons and entities who act in active concert or participation with the Republic, to assist the Republic in fulfilling its payment obligations under the Exchange Bonds.
7 Sort of tangential, but here is an argument for calling the holdouts in this case "vultures" (a perfectly good, nonderogatory term in debt markets) rather than "holdouts" (since many of them bought after default). I do not especially agree with it, but I also do not especially disagree.
8 See Scheme 3 in this post, which is essentially a less sophisticated version of this proposal. I owe both of them to conversations with an emerging markets hedge fund manager.
9 I think it's more technically like: We will temporarily waive our claims for default as long as you are making these quasi-interest payments, and we will permanently waive our right to receive any interest payment for which you've actually made a quasi-interest payment. This probably requires a bunch of other technicalities too. Like, the trust needs to convince Argentina that it holds the bonds, which probably means showing some sort of broker statement and promising never to sell any bonds as long as the agreement is in effect (to avoid double-counting). Once you put bonds into the trust, your liquidity comes from selling trust certificates, not taking the bonds out.
10 Elsewhere in neat tricks I suppose I should wrap up the other day's discussion of credit default swaps on Argentina's bonds. Here are the terms of the CDS auction. The objection to the Japanese yen bonds that were momentarily misplaced -- revolving around when exactly they were issued -- has been resolved, and those bonds are on the final list.
One version of the problem that I've heard is that the "issue date" of those yen bonds is actually well before the June 1, 2005, cutoff, if by "issue date" you mean "date that they start accruing interest" or "date they were authorized by statute" or some other technical definition. If that's how you count "issuance," then the yen bonds would not be eligible unless they were "New Securities" under the terms of the prospectus blah blah blah. The determinations committee's reasoning is not public, but either it decided they were "New Securities" (unlikely but not crazy) or it decided that they were actually issued when they were settled, on June 2, 2005 (which seems right to me). If this is all gibberish, read Wednesday's post, though that's probably gibberish too.
Anyway this result means that the recovery for CDS buyers will be higher than it otherwise would have been. But there's now just $813 million of net notional CDS on Argentina outstanding (see here for the official explanation of "net notional"), so CDS recoveries are going to be relatively small compared with the amounts at issue here. Thus this theory, that Elliott Management pushed Argentina into a default to make money on giant CDS positions, seems extremely unlikely, though it remains popular in some circles.
To contact the writer of this article: Matt Levine at email@example.com.
To contact the editor responsible for this article: Stacey Shick at firstname.lastname@example.org.