Short Scream

Minyanville’s Mr. Practical had this yesterday on the *new* naked shorting rules.

The rule changes proposed by the SEC for selling stocks short are meant to curtail “naked” short selling, selling stock short without locating a borrow. They tighten up procedure between lender and borrower through broker dealers. For example, in the past a short seller would sell stock and then call the broker for a “locate,” a stock available at the dealer to borrow. And even if he called first and the dealer had no “available locate” at the current time, the dealer probably told the seller to go ahead knowing that the locate would come at some point.

It’s still not clear how to interpret the new procedure, but it looks like now “the locate” must be currently in the box. This is going to raise the cost of borrowing. In other words, the rebate rate credit to the short sellers account (the interest they earn on the cash they generate by selling the stock) will be less.

In general this will cause those making markets in options, convertible bonds, CDS, and other derivatives where short stock is used to hedge to need better prices. In general, liquidity will go down.

I can’t speak to the CDS and convertibles and all, but I can confirm this to be the case on the options floors. And then some. This is massive.

As a market maker, you get positions that would seem insanely large to the “retail” trader. But you don’t need a giant account as you have way reduced “market maker margin” requirements. They also have an exemption on the locate requirement. Which may go out the window now as well. The obvious goal is to provide liquidity, as their job is ostensibly to take the other side of order flow.

There’s some actual thought behind that. Taking the other side of order flow often results in positions that are physically short tons of stock. Remember, the most common public order is a Call Write, so market makers tend to own tons of calls, and ergo short stock against it. It’s not actually a directional bet though per se, he may also be short puts, or he may be long too many calls, or whatever.

Any move that makes it more difficult for a market maker to borrow shares will indeed produce all sorts of unintended consequences to cure something that’s not a problem to begin with. These are not Cramer’s mythical Short Cowboys. They’re probably not even short delta.

The consequences? Like Mr. Practical says, reduced liquidity as a starter. And how about increased put premiums? Shorting stock gives MM’s ammo to short puts as well, especially when they already own calls against the short stock.

Higher puts of course translate to higher volatility. Which on the margins will creep into the volatility of the stock itself. And that will be the ultimate outcome.

As to positions MM’s have now, it’s possibly a disaster. Put/call parity is based on the current interest and short stock rebate rates. Let’s say short stock rebate instantly evaporates, or goes negative. A trader sitting with a pre-existing short stock, long call, short put posture will get ripped.

I realize I am biased to favor the interests of Market Makers. So take what I say on this from that perspective. But imho, changing this rule as it regards to MMs is truly a ridiculous response to some misinformed bleating about shorts. Again, these are stock shorts in name only, it’s just a natural byproduct of market making, It’s clearly the same deal with anyone taking down a convertible offering that needs short stock as a hedge. None are going around trashing stocks left and right. If anything, these sorts of positions set them up as stock buyers into weakness.

So while naturally no one will hold a pity party for MM’s caught with a one time hit on bad positions for this, keep in mind it will have ripples to all of us in the form of higher volatility and shallower markets.

So if SKF is the new Fear Index, how about volatility on that Fear? And better still, longer term anticipation of the volatility of Fear?

Here’s a chart of 90 day normalized options volatility on SKF (the lower one), basically an October option now. And yes Houston, we have record numbers here.

Again, this pup essentially doubles XLF volatility, so keep the absolute numbers in perpective.

Here’s something bizarre though. If you watch pretty much every tick here, as I do now, you would expect shorter term stock volatility is at a record too.

Oddly, no. The higher chart shows 10 Day historical volatility, i.e. the volatility of the stock itself over the past 10 days. And that’s rising, but well off the March highs. In fact, the Oct. options are assuming the extraordinary stock volatility we see right here, right now will persist for 3 more month’s.

Now no reading is perfect. I suspect the 10 Day calculation underestimates the volatility you see actually trading this thing now. It only considers the day’s range, not the way it got there. So in other words a 20 point range in one direction goes into the calculation just the same as a 20 point range that saw SKF round trip up and back 10 times from high to low. But the pattern is closer to the latter than the former. Yesterday for example SKF went up 20, then all the way to down 10 before finally closing up 10. The range was 30ish points, but the round trips felt like 70 points.

Still, this all won’t last forever. I have no position after Friday. But If I want to buy paper, I will look at the nearer month’s, and if I want to sell it, I will look out in time.

OK, this is normalized 180 day volatility in FNM, but only going back the past 2 month’s in order to best visually depict what has gone on here in the past week.

In other words, this is the implied volatility of what is now pretty much a Jan. option and it has basically doubled in about a week.

We have seen plenty of extraordinary volatility spikes over the course of time. But not sure I have ever seen one where longer-dated options exploded this much and this quickly.

Essentially, the options pretty much have priced in worthless equity. Or at least hedged against worthless equity.

So what did we learn last week?

How about yet another example of why shorting incredibly pumped volatility rarely works. Here’s FNM’s chart for the last week, and that was the less volatile GSE.

Here’s what I had to say the other day.

Free Money Idea

Sell the FNM July 15 straddle for $4.

Yes, that’s $4 for a straddle with a strike price of 15. With 2 weeks to go. That translates into something like a 180 volatility. Don’t they have government guarantees or something? Free cheese!

OK, hoping the sarcasm above is obvious.

That’s some monster price of course, wow. The best thing you can do when something goes on tilt like this is just avoid it. The next best thing is to use spreads. Either calendars or a directional bet via a vertical. The worst thing idea is just sell gamma and cross your fingers and hope for the best. It may work out, but volatility explosions do not happen in a vacuum.

I can’t emphasize that point strongly enough. Conventional wisdom says you actually buy gamma in the near month when this happens. And obviously hindsight is 20/20 and it would have worked here, and in Freddie. But that’s a game better left to the floor traders. I wouldn’t open a position like that off-floor and bank on my ability to handle 180 volatility. And I wouldn’t advise anyone to try it unless they were very comfortable with the VERY active trading it would take to manage such a position.

What about calendars? Obviously there’s a huge volatility premium in the nearer month’s. Well, conventional wisdom says you actually want to sell calendars when this happens. The theory being you can trade stock against the long gamma of that position, meanwhile the longer dated options are overpriced in volatility terms as this sort of trading tends to dissipate within a couple weeks.

Again, I wouldn’t try that at home either, but also would resist the temptation to buy calendars at what look like great prices. I should have noted that the other day when I threw the calendar idea out there.

Vertical spreads probably make the most sense. It involves making a directional bet, but that’s got defined risk, and avoids making what could be a disastrously wrong guess on what will happen to volatility.

Don Fishback‘s on fire this week looking into the the lack of VIX pop.

I didn’t want to take too much time, and go overboard on the analysis. Which means I took some liberties with respect to the mathematics in trying to make things relatively simple. But the result is pretty much in line with what quite a few of you expected.

A major reason why VIX is as low as it is with respect to individual stock option volatility is due in large part to the relatively low correlation in the market. One stock tanks while another stock skyrockets. Both stocks have huge volatility. But the index remains unchanged, so its volatility remains low.

I realize we are getting a bit of a self-confirming circle here (I’ll throw Bill in too in this). But all roads do seem to lead to correlation. Plus we’re all beating this to death. And I would argue that end of the day, we assign too much import to one indicator that clearly doesn’t fully capture precisely what it is designed to measure.

But here’s something with interesting implications on the Dispersion Trade, the one where you go long individual stock gamma and short index gamma.

…..correlation tends to lead the the volatility differential by about three months. That is, when you start to see correlation rise, a few weeks later you’re likely to see the gap between the implied volatility of individual stock options and the implied volatility of index options narrow.

Basically when you put the trade on, you are betting on correlation. Let’s say you go long options gamma in basket of stocks and short gamma in SPY and/or QQQQ. You are rooting for correlation to decline; your best case has every stock going nuts, but in different directions such that the SPY or QQQQ does not budge. I would have expected options moves to predict correlation, not the other way around.

Since most have neither the time, capital, or inclination to put on and manage the trade, consider this more an interesting allocation tool. In other words, if you see correlation rise, you should switch from individual stock options to index options. Which of course makes perfect sense, I’m just surprised the order it happens.

OK, top to bottom here we have volatility charts of BTU, POT, and MA.

Even though they are in disparate industries, the stocks behaved kind of similarly this year. Basically up, but with a few violent shakeouts along the way. Including one’s they are in right now.

And volatility pretty strong in all three. Most pronounced in BTU, which also had the quickest plowage in the stock itself, down 12% in under a week.

POT probably most interesting though in that as you can see, the stock itself never got all that volatile. The decline off the highs was relatively orderly, although net-net it dropped an impressive 20% or so top to bottom in about a 3 week time span.

So we have modest volatility spikes in momo names, and big spikes in everything financial. Where don’t we see much? How about tech. That’s a relatively peaceful spot in 2008, and options act accordingly.