I blame spring break both for the lack of news this week and for the fact that what news there is revolves around trading glitches. Apparently spring break has cleared New York not only of responsible adult bankers and traders taking their kids to Disney, but also of responsible adult trading computers who are off doing God knows what, leaving the callow analyst computers alone to man their desks. And no one should be surprised that they got a few things wrong on their first day in charge.
Away from BATS, the glitchy news is in TVIX, a 2x levered short volatility ETN issued by Credit Suisse. And the craziness here seems to be caused not by robots on the fritz or fat fingers. Here is the story:
Credit Suisse Group AG (CSGN), under pressure to restore order in an exchange-traded note tracking U.S. equity volatility, said it will start resupplying the market with shares today after cutting issuance off in February.
Stock will be added to the VelocityShares Daily 2x VIX Short-Term ETN (TVIX), or TVIX. The security, designed to track Chicago Board Options Exchange Volatility Index futures, has whipsawed investors for the past month, climbing 89 percent above its asset value and plunging 29 percent yesterday before Credit Suisse’s announcement. It fell another 19 percent to $8.23 at 9:56 a.m. New York time today, extending its retreat since Oct. 3 to 92 percent. [When I looked it was swinging wildly around in the $7.50ish area, which is a bit under its $7.83 NAV as of yesterday, go figure.]
Creating shares in the ETN will help bring the security back in line with its so-called indicative value, the price implied by futures on the CBOE gauge, said Alec Levine*, an equity derivatives strategist at Newedge Group SA in New York. Credit Suisse’s first round of share issuance is intended to lower the cost of borrowing the note, a step that may aid short sellers who yesterday helped cut the premium by 66 percent even as owners of the security were burned.
“Lending out shares is an attempt to drive down the premium,” Levine said yesterday in a phone interview. “When your product isn’t trading anywhere near NAV, it’s the market telling you that it’s a broken product.”
So that's sort of a hilarious way to put it. "If you offer widgets at a suggested price of $29.95, and people are reselling them at $56, that's the market telling you your widgets are broken." Not ... exactly.
It is, though, the market telling you "hey we will buy widgets at a price much higher than their actual value, would you like to sell us widgets?" Selling widgets for 89 percent above their asset value is a good trade. Selling widgets for only 34%** above their asset value is ... I mean, it's also a good trade, so it's no surprise that CS was open to doing it, but it's about a ... 29% less good trade than it was the previous day. So it's pretty weird that CS watched this run-up for a month, was all "meh, I don't really want the free money," and finally capitulated after a 29% drop.
Now of course Credit Suisse can only sell more widgets if it can make more widgets, and its 2x levered VIX widget factory seems to have shut down last month due to "internal limits" so maybe it had no more supply of VIX? The customary way for an ETN issuer whose ETN is indexed to some futures is to buy those futures, and maybe it hit VIX futures position limits? Izzy Kaminska reads Credit Suisse's press release and comes up with what I guess is a plausible explanation:
As we’ve pointed out before, Credit Suisse has no obligation to hedge this ETN with Vix futures outright. It can (and probably has been) using a multitude of different instruments — including, if not exclusively, its own internal net volatility position.
The fact it is now going to create units that will be aimed solely for market makers – and presumably not resold to the secondary market*** — suggests it’s trying to tie their hands somehow. What’s more, the press release states that Credit Suisse is demanding that the creations are linked to market makers selling the company suitable Vix hedges, including swaps, at the same time.
One might conclude that market makers ramped up the creations leading up to February 21 on purpose so as to specifically breach Credit Suisse “internal limits”, since they had an insight about how much of an internal volatility hedge it carried.
She thinks they basically tried to corner the market to force CS to buy futures from them at inflated prices. I don't get it - like, if I were CS and I was able to sell short-dated index vol at an 89% premium I would be buying tons of correlated vol everywhere I could get it since I'd have to be able to pay more than my competitors, so I could always grow my internal vol book. There is definitely a market breakdown here and I am not smart enough to understand it; is it maybe possible that "internal limits" was meant literally as some combination of actual VIX position limits and a regulator- or whatever-imposed limit on total vol risk and some CS vol trader spent the last month arguing "guys, if I buy vol from clients and sell it to TVIX, I'm flat!" and getting ignored / lectured about correlation?
It's all baffling, and it seems to have baffled a lot of people, including apparently people who owned TVIX, which, for the love of God, why? Kid Dynamite, who unlike me actually understands exchange-traded products, has a series of posts trying to puzzle out the TVIX puzzle and he has this to say:
I want to step back, and point out that when you put aside the crazy price action in TVIX over the last month and get back to “basics” it’s still a product that is beyond the understanding of most retail investors. It’s a relatively complicated product, and there’s a lot to understand in terms of the difference between the ETN vs ETF structure, the effect of contango in the VIX futures curves, the long term holding issues based on this contango, etc.
None of that means that the product shouldn’t exist – it serves a specific purpose, and I think that traders should generally be free to seek the exposures that they want.
It seems clear, however, that many people do think that someone needs to protect them from their own – what word do I use here – ineptitude? ignorance? unsuitability? In other words, one can certainly argue that Ma and Pa Kettle probably shouldn’t be trading TVIX from their Etrade/Fidelity account. In my world, my response is “If you don’t understand it, you shouldn’t be trading it.”
True, but. We talked a while back about a paper about ratings agencies that tried to puzzle out why ratings of asset-backed securities performed so much worse than ratings of corporates. The paper authors were resistant to the notion that the greater complexity of CDOs was the main culprit, saying: "Indeed, one could argue that a pool of mortgages or credit card receivables should be less opaque than corporate issuers with synthetic leases and other exotic off-balance-sheet liabilities." This struck me as a useful reminder that a structured product is generally less, not more, complex than, say, a financial institution. Sure the TVIX is a hot mess of leverage and contango, but so is Credit Suisse - which overlays a bunch of other messes and a bunch of humans on top of whatever its undisclosed short-dated equity vol position might be. CS's stock is up like 2% today, after being down like 2% yesterday. TVIX is down 29% and 27% respectively.
Steve Randy Waldman has this theory that finance is complex because hiding risks enhances value; it encourages people to commit capital to finance risky ventures that they might shy away from if the risks were fully transparent. I don't know that that's entirely true but it's fun to think about with this TVIX blooper. Because here's the thing: TVIX is super simple! At a fixed date in the foreseeable future (2030), Credit Suisse will pay you a number calculated by applying a formula in the prospectus to a publicly reported index, which index is a reasonable proxy for expected volatility in the US equity markets over the next month, give or take. It is a fairly transparent, as these things go, bet on a comprehensible sort-of-macroeconomic variable.****
Credit Suisse's stock is super complicated. It will, at an uncertain date far in the future when Credit Suisse winds up its business or is acquired, pay you an amount equal to the value of whatever its bankers and traders have bought between now and then, minus the value of whatever liabilities it has incurred; it will also, if it feels like it, pay you a 5%-ish dividend, or whatever higher or lower number it decides on, between now and then. It is a combination of bets on various macroeconomic variables and also on specific identifiable and unidentifiable humans' reactions to those variables. You can look at its balance sheets but, y'know, good luck with that.
TVIX is so simple that everyone sees the problem when its trading price gets way out of line with its underlying value, which is calculated and disclosed every day. When there's a bubble in TVIX, everyone sort of realizes that more supply is coming so shorts start looking around for borrow. Every bank's trading price is way out of line with its underlying value, but that's fine, because nobody believes those underlying values when they're reported every quarter. New supply normally comes not to pop bubbles but when banks trade below book value. And yet TVIX is what everyone worries about.
* I'll throw out a blanket "no relation" for all Levines mentioned herein and hereinafter, unless specifically noted. Also "Levins" since for some reason that is a source of confusion.
** That's NAV * 189% * (100% - 29%) for those of you working your way through the "are you ready to start studying for CFA Level I?" exercises.
*** I don't read it this way, fwiw, but iw not very much.
**** Here I totally unjustifiably ignore your credit risk to Credit Suisse.