A while back Greece had a problem. The problem was that it wanted to borrow more money, but didn't want to increase its debt. (Because the European Union would frown on it having more debt.) So it went to Goldman Sachs, and Goldman told Greece, well, what you can do is borrow some money from us, but we won't call it debt, because something something something something swaps.
This worked pretty well for everyone, for a while: Greece got the money, but nobody outside the deal understood that it had borrowed the money, because it was part of a derivative trade that was not accounted for as debt. Then it stopped working, and everyone got mad at Greece for disguising its debt, and at Goldman for helping Greece disguise its debt and charging rather richly for the service.
Anyway last week Goldman said it was sorry and wouldn't do that any more:
Goldman Sachs Group Inc. (GS) Vice Chairman Michael S. Sherwood said his firm would refuse another deal like the derivative it sold Greece that masked the nation’s growing debt to help it meet European Union standards.
“We absolutely wouldn’t do a transaction like that today,” Sherwood, 48, said in an interview posted on Channel 4’s website yesterday in London. “There have been transactions of that ilk that have been presented to us by other European sovereigns that we’ve turned down because we felt there wasn’t the appropriate transparency surrounding them.”
I cannot entirely agree -- I'm sort of of the school of "if it's legal and it pays you a lot, you should do it" -- but there's probably a reason Sherwood is giving interviews about reputational risk and I'm a blogger.
But a hedge fund reader points me to an intriguing modern parallel, or at least the rumor of one. Venezuela, it seems, has a problem. In its foreign currency reserves, where most countries keep foreign currency, Venezuela has a lot of gold, because its late President Hugo Chavez really liked gold and wanted to "move away from the 'dictatorship of the dollar.'" It has rather fewer dollars, and it turns out that if you want to buy goods and services in international trade, dollars are more useful than gold. Thus the problem.
The normal way to solve that problem is to sell some gold to someone who has dollars. Then you'll have less gold, but more dollars, and the dollars can be used to buy goods and services. But if you're Venezuela, there are problems with this too. For one thing, it seems like sort of a repudiation of your late president's policies. For another thing, spending down your gold reserves might suggest that you're in a bit of a tight spot financially, which though true is awkward.
Also, gold was worth $1,800 an ounce a year ago and is now worth like $1,240 an ounce, so it feels sort of crummy to sell a bunch now.
In comes, apparently, Goldman Sachs? Here is a thing that seems to be happening:
Venezuela newspaper El Nacional reported Tuesday that Venezuela’s Central Bank and Goldman Sachs are ready to sign an agreement to swap or exchange international gold reserves with a start date in October 2013 until October 2020.
The negotiated amount is equivalent to 1.45 million ounces of gold, valued at US$1.8 billion at today’s prices, which is to be deposited in the Bank of England with the transfers made directly to Goldman Sachs once delivery times are stipulated. Goldman Sachs will then pay U.S. dollars for the gold.
An adjustment of 10% will be made to the asset value as a hedge in case the international gold market price falls. The annual interest rate will be a combination of dollars with the call BBA Libor equivalent to 8%.
Do you know what this means? I do not (here's El Nacional's version, in Spanish but not significantly clearer), but it would seem to be a margin loan against the gold; i.e. something like:
- Venezuela is borrowing about $1.6 billion from Goldman for seven years.
- Venezuela is collateralizing that borrowing with gold worth $1.8 billion at today's prices (i.e. it's 90 percent of the value of the gold; that's the 10 percent haircut), and it's posting that collateral somewhere Goldman can get it (the BoE).
- The collateral will be subject to margin calls as the price of gold increases or decreases.
- Venezuela is paying about 8 percent a year for this loan.
So is that a good deal for Venezuela? It depends how you count but it's hard to imagine the answer is yes. I mean: Why would Goldman do it if the answer was yes? There are some other arguments below but that is surely the main one.
Because, if true, this is not the most pristine deal you ever will see! (Goldman declined to comment.) I mean, one, Venezuela -- it inspires people to feel feelings, plus you might have some weird dynamics around actually getting them to post margin. Two, complex derivatives etc. -- more feelings, though despite the word "swap" here, my best guess about what is going on here is that it is really just a secured loan and so barely a derivative at all.
Three, ask yourself, what is the purpose of this trade? I won't tell you the answer, because I don't know, but it sure seems to be for Venezuela to get some money for its gold without "selling" it. Which is the sort of sleight of hand that, as a bank, in 2013, you might want to avoid. Unless, again, it pays well.
Sherwood goes on, as well he might: "I’m not talking about legality here. Everything we did we felt was legal, but here we’re talking about what’s appropriate and what’s reputationally sensitive."
Multiple problems, apparently, but let's focus on this one.
Not in the bit I quoted, but in El Nacional:
Durante la vigencia del instrumento se constituye una cuenta llamada “de margen”, en la que el BCV se compromete a depositar una mayor cantidad de oro en el caso de que el precio del metal caiga o en la que Goldman Sachs depositará divisas cuando la cotización del oro aumente. “Al vencimiento de la transacción los aportes realizados son devueltos a sus propietarios”, señala el documento.
I see 5-year credit default swaps on Venezuela at around 1200bps. There's a 2022 USD bond (ISIN USP17625AC16) yielding around 14 percent. So figure Venezuela's unsecured 7-year rate is around 13 percent. It's paying 8 percent. It's saving 5 percent a year in exchange for collateralizing 111 percent of the money it's getting with physical gold and posting margin if that collateral's value drops. Doesn't seem great.
One way to look at it is that this Venezuela could just sell 1.45 million ounces of gold for $1.8 billion and spend some of the money to buy call options on the gold price. Plugging in a 7-year term to XAUUSD OV in Bloomberg, and leaving all other fields at their default settings, I see an at-the-money call being worth about 25.5 percent of spot. So you can buy at-the-money call options on the full 1.45 million ounces, giving you all of the gold upside that you have now, for about $460 million. That's just over three years of interest on this deal -- and you've got no downside risk and no margin calls. Math:
Obvious caveats are (1) you probably can't actually buy $1.8 billion of long-dated at-the-money call options at the Bloomberg-default price, or possibly at all, (2) there's something reassuring about holding an actual pile of gold that you can't quite replicate with a call option, and (3) I've probably messed up the math. Still this deal seems pretty rich.
Because socialism. Also because the risk is wrong-way: They have to post margin when gold prices fall, and a lot of their foreign reserves are in gold, so you're gonna be asking them for margin at particularly bad times.
To contact the author on this story:
Matt Levine at email@example.com