Bank of New York Mellon is back in the news for offering a special promotion to its valued FX customers: if you act now, instead of screwing you with the worst possible price for your FX trades, they will not do that. OWS is working!
The thing about that is ... well, wait, let's start with something more important: I don't really think that Gretchen Morgenson understands anything about derivatives. That would be ridiculous. Good to have that off my chest.
What I meant to say yesterday was not that she did, or that anything she's said about derivatives was technically correct. It was that getting all excited about how she mislabeled a repo a swap misses the point. If a repo and a swap have substantially the same cash flows and achieve substantially the same economic effect – here, letting MF Global leverage a position by separating funding from credit risk – then there’s nothing substantive about calling one thing a "repo" and another a "swap."
BoNY Mellon, though, shows that what you call a thing actually can matter. Thinking that everything is a derivative may lead to confusion and anger if you’re, say, Gretchen. Because Derivatives Are Bad. But, if you’re me, thinking that everything is a derivative might make you a little bit more sympathetic to BoNY. Because I don’t think that what they were doing was – or was only - screwing their customers by secretly giving them the worst price of the day. I think that they were “long a floating-strike, intra-day option on their FX transaction." They bought a derivative from their clients and paid for it in the form of lower (zero) FX commissions. Now the new, we-won’t-screw-you pricing model offers no optionality, just a specific London fixing – at a 4-to-10-basis point commission. Is that better? It depends – probably it is; certainly it is in periods of high volatility – but it’s a different risk profile. For a client who figures they will do lots of trades at various prices and have no ability to predict intraday volatility, saving the commission might be attractive.
But seeing everything in terms of optionality bought and sold doesn't mean you can completely let BoNY off the hook. It depends on how exactly they were buying that optionality.
There are, let's say, three reasons to enter into a derivative transaction. One is to consciously parcel out risks between parties to a trade. I am a bank who needs to make loans, but I don't like having too much concentrated credit risk. You are a hedge fund that wants that credit risk, but you don't have the balance sheet to buy the loans. We do a trade. Everyone wins. I mean, one of us wins and one of us loses, but ex ante we conveyed the risk that each of us wanted to have.
The second, and far more important, reason is to achieve some regulatory end. This is the best way to understand the "repo" versus "CDS" distinction. They are approximately the same, economically (if you gloss over interest rate risk). But the name and superficial structure you apply to a transaction can affect how much margin you post, whether you account for it on balance sheet or off, how it's taxed, how it's treated for regulatory capital, etc.
The third reason to do a derivative trade is, though, to obtain hidden profits by capitalizing on counterparty ignorance. The option that BoNY Mellon was long was a real option: they had a floating-strike put on exchange rates. I have no doubt that BoNY had a model that could value that option based on the intraday volatility of exchange rates. I have even less doubt that BoNY's clients - not all of them, maybe, but certainly the ones who were doing "standing instruction" FX trades - did not have a model that could do that.
The pitch to those people is seductive: "do this trade, we'll charge you zero commissions, you'll just get the rate at some point during the day!" If the client - the assistant to the assistant assistant treasurer at a pension fund, it seems - noticed that he was getting the *worst* rate of the day, the bank has a response that is almost as good. "Yes, true, you're getting the worst rate of the day. But you don't know how much worse that will be than the 4pm London fixing. If it's, say, 2bps worse, then you'd rather get that worst rate with no commission than the 4pm rate with a 4bps commission. More importantly, even if it's 10bps worse, you can tell your boss/auditors/whatever that you transacted at a market rate that day - which is true - with no commission - which is also true."
The important thing to understand about any of these trades is why they looked the way they looked, and why they took the form that they did. Could MF Global have gotten the same exposure to European sovereign debt by writing CDS on it as by doing its repo-to-maturity transactions? Maybe - but repo transactions offer the possibilities of both more opaque accounting (regulatory arbitrage) and a more receptive/less sophisticated investor base to stuff with the trades (counterparty ignorance).
Could BoNY's clients have freely chosen to sell time-of-trade optionality in order to avoid paying a fixed commission on their FX trades? Sure, whatever. Did they? It seems really, really unlikely. How can you know? Well, one quick test is actually what they called it. If BoNY told clients "here's our Intraday Option FX Pricing Product," then, you can at least sort of presume that clients understood the tradeoffs - if not in detail, at least in the brute fact that they were selling a derivative. Then it's at least possible that there was a genuine shifting of risks, intended by both parties.
If BoNY told clients "here's our Standing Instruction product, where we give you the best possible rate!" - well, that seems like a pretty good tipoff that they were purely banking on ignorance.