It's been a while since we checked in with the infinity thrillion dollars of Libor lawsuits, but the Journal has a good roundup today and, yeah, eep, this is sort of interesting:
Firms facing the biggest potential payouts, according to Morgan Stanley, based on the financial business they do rather than their assumed culpability, include Deutsche Bank AG, Royal Bank of Scotland PLC, Barclays, Bank of America and J.P. Morgan.
It seems almost unfair: you can very easily put a whole lot of leverage on your employees' lame criminality; if you're really really good at selling rate product even a tiny wee bit of criminality can be a disaster.* Shades of this chart - shouldn't you get more points for being more criminal, not just for being bigger?
But this was the most jarring part:
Fund manager Charles Schwab has alleged it deserves damages related to billions of dollars in fixed-rate investments held by its funds, as well as investments with returns pegged directly to Libor. Schwab alleges in lawsuits it filed last year that the fixed rates were set in relation to Libor.
This is actually true; here is the Schwab complaint, which I've seen before but somehow didn't register this:
86. Fixed-rate notes and other LIBOR-based instruments and securities sold or issued to Plaintiffs by Defendants. Throughout the Relevant Period, Plaintiffs bought fixed-rate notes from and issued by Defendants. These notes paid a fixed rate of return. However, the price of these notes and the fixed rate or [sic] return were determined based on LIBOR. Defendants' suppression of LIBOR caused Plaintiffs to receive lower returns on these notes and/or pay more for them than they would have if LIBOR had been properly set. Plaintiffs relied on the accuracy of LIBOR in undertaking these transactions.
87. Fixed-rate notes and other LIBOR-based instruments and securities sold or issued to Plaintiffs by entities other than Defendants. Throughout the Relevant Period, Plaintiffs bought fixed-rate notes from and issued by entities other than Defendants. As is well-known to sophisticated market participants like Defendants, those notes are priced off of, and pay returns based on, LIBOR. Defendants' suppression of LIBOR caused Plaintiffs to receive lower returns on these notes than they would have if LIBOR had been properly set.
You can see why the plaintiffs' lawyers thought it was a good idea to throw this in there. In lots of applications, people think about fixed-rate notes based on a spread to swap rates: if 5-year swaps are 3%, and a bond pays 6.5%, then you can think of that bond as paying 350bps over swaps and you often do. But of course swaps mean I-pay-3%-and-you-pay-me-Libor for 5 years, to the point that you might somewhat loosely describe that bond's pricing as "L+350." But if you can manipulate Libor, then it stands to reason that in expectation you can manipulate swap rates: if banks are gonna push down Libor by 5bps every day for five years, then swap rates should be 5bps lower than they otherwise would be. And if bond prices are based on swap rates and swap rates are based on Libor and Libor was fucked, then bond prices were fucked. By the Libor banks. So every fixed-rate bond is, in theory, fair game for Libor lawsuits. Whee, more money.
But that can't be right, right? Here is a rephrasing of paragraph 86:
We loaned money to banks at market rates. But the banks manipulated the market rates, so we want more money.
But you can't manipulate a market rate by just writing down a different rate: if someone lends you money at that rate, that's the market rate. If investors bought fixed-rate notes from the Libor panel banks, the rates on those notes are self-validating. It's like saying "we loaned money unsecured overnight to the banks at Libor, and we want damages because Libor was manipulated": Libor is just a measure of where banks could borrow overnight; if they could borrow overnight at those rates then there's no manipulation and you don't have a case for damages.
Paragraph 87 is not much better: it says "we gave non-bank companies long-term loans at rates that we thought were fair, but we would have thought otherwise had we known that short-term bank borrowing was more expensive." And they might have? But so what? Felix Salmon has pointed out**:
The Libor understatements actually brought it closer to prevailing interest rates; the fudging basically just served to minimize the degree to which the unsecured credit risk of international banks was embedded in the rate.
And if you were buying notes from non-banks, that unsecured credit risk wasn't relevant to you. "We loaned money to Nuance Communications at 5.375%, but it would have been 5.5% had we known the full extent of Barclays' financial troubles" is ... I mean, is true, probably, but not especially compelling.
[Update: a reader points me to the amended complaint, which limits the claim to notes with 5 to 365 days of remaining maturity (see paragraphs 3 and 193) - so priced off actual "Libor," not swaps. Still, same argument: sure, if you are buying a Company X note, you could decompose the prevailing fixed rate for Company X into an interest rate plus Company X's credit spread, but there's no reason to think that the "right" way to do so is to add (i) a rate incorporating bank credit plus (ii) Company X's credit spread in excess of average bank credit.]
Anyway, I guess it's no surprise that lots of dumb Libor lawsuits have been and will be filed, or that when faced with the prospect of a zillion dollars of potential damages lawyers are being creative about expanding their target lists. The environment is just too tempting: if Libor is a benchmark for everything, then everything that happens can be manipulated by manipulating Libor. Sometimes - swaps! - that's true, or mostly true. Sometimes - fixed-rate bonds of non-financial issuers! - it's surely false. But with the pot of "everything" so large, there's a lot of potential upside in trying to confuse the two.
* And don't say "well the more rate product you sold the more you benefitted from that criminality previously." If you had a basically balanced book of swaps then manipulating Libor down by 40bps for a year makes you roughly zero dollars whether that book is $1bn or $100bn in notional, but the former costs you $4mm and the latter $400mm in damages. The trading is symmetric, the damages are all one-sided.
Also per the Journal Macquarie seems to think Libor was manipulated down by 40bps throughout 2008 and 2009. That seems astonishingly high, no?
** Before vanishing from the internet; fortunately this is one of the few traces preserved from his once-proud online existence.