This page has been archived and commenting is disabled.
Guest Post: Short Interest - Does It Mean Anything
From Peter Tchir of TF Market Advisors:
Short Interest - Does It Mean Anything
The WSJ diligently reported short interest today. It increased in March. Unfortunately, I think this is just another example of data that is published, talked about, and even used to make decisions, that has failed to keep up with the times.
Of the 12 stocks listed with the most shares shorted, 4 were ETF's. Is being short a broad ETF the same as being short a single stock? I'm not sure it is. I would definitely want to see a version with and without ETF's.
Which leads to the next question, what ETF's are included? I don't think short interests in TLT or OIL are relevant, and shouldn't be included in any case. A long in SH should be counted as a short interest. A long in an ultra short etf should be counted as double short. Clearly, in order of magnitude, the biggest change would be looking at the short interest with and without ETF's. Properly categorzing the ETF's is a second order effect, but there is no reason it shouldn't be done, its not that difficult.
The next big question, is why do we focus on short interest in terms of number of shares outstanding? Why not in terms of market value shorted? Aren't we more interested in market value? Citi is the biggest short with 372 million shares shorted. SPY has 308 million, but the price of 132 vs 4.5 means the short in SPY dwarfs the short in Citi, in my opinion. Also, once the 10 -> 1 reverse split goes through for Citi, we will see a massive drop in the short interest since its focused on shares. Does that drop tell you anything? Not in my opinion. At the very least I'd like to see a list based on both shares and market value of shorts, but if I could only get one, I think the market value based one is more useful.
There is nothing wrong, per se, with looking at short interest, but it is just one of many pieces of data we get and look at, that have not kept up with the market changes. We need to delineate ETF's from non ETF's and make sure the ETF's are looked at correctly - stock underlying, and ETF's that are shorts, have their long positions count as shorts. Then we should see the data both on a share basis and market value basis. Some of this data is out there already, its just not how the headlines are reported.
Why do I think its important? We are seeing more trading rules of thumb breakdown. When we look to short interest, volumes, etc, as an additional guideline to evaluate the market's overbought/oversold conditions, etc, we should be looking at accurate data. I suspect that the reason so many relationships seem to have broken down, is because the data we look at is now effectively garbage. These changes in reporting should be incredibly easy to implement and I suspect would provide data that is more useful and maintains historical relationships better than what we currently get and discuss.
Volume indicators have broken down in many ways. I think a lot of the issues listed apply to volume indicators as well. (market value vs # shares, ETF's, etc). Volume indicators have some additional issues that also need to be addressed. But I think there too, once we develop a better measurement in this ETF, HFT, driven world, we will find that relationships are still valid, its just that the data we use became useless as it no longer reflected the market.
»
- 5667 reads
- Printer-friendly version
- Send to friend
- advertisements -
The only thing that matters in the markets now are the FRB and other central banks. When they fail in their latest experiment in anti-gravity, it will be epic.
When the entire market/system is corrupt and fraud is the status quo, none of the "reported" numbers are trustworthy, much less useful. Hedge accordingly.
In a mega-beta market, shorting individual names is seeking alpha the hard way.
Good point about ETFs though - when you strip out the SI that's derived from sector/basket/index shorts I wonder how much is left?
What ami I missing? SPY is designed to track almost exactly to a tee the price and movement of the SP 500 which is the cumulation of 500 stocks. So the short interest on SPY has absoultely nothing to do with the price of shit in China. Much like charting AGQ to determine the movement of silver or UCO or DTO with crude. Just more pie in the sky for the sheeple to be confused created by lawyers and regulators to justify their high salaries milking the ignorant.
Why is it in a time of electronic trading that short sales aren't indicated on the ticker. This would be easy to do with a super or sub scripted "s" next to the trade. One should know at the end of the trading day what volume was short as a percent of the total volume. The only reason I can see for not implimenting this is that some one benifits by having the window only half transparent. The short sale data reported today is ancient history and worthless.
All risk capital fleeing to the safety of pork bellies
Now if someone can get Ben to scare the pork belly traders to sell we could free up some cash for bond buying.
TA has become irrelevant, replaced by whim, gut feel, finger in the wind or flip of coin.
Wedgies and.flags are all bullshit, but extremes can be traded at times.
Nice, thoughtful piece!
I share Mr. Tchir's concern about data quality, but none more so than BLS employment and unemployment numbers. You need a ouija board to figure out what they really mean.
"we should be looking at accurate data....the data we use became useless as it no longer reflected the market."
When it's garbage in it's garbage out. Revisions are made to the released data month after month. This is not going to end well.
Granny demands a %yield, Granny is begging for a %yield, Granny is screaming at the top of her lungs for a %yield!
No more free money for Primary Dealer billionaires! The American people are broke! especially fixed income elders and those that saved dearly to build a nest egg for their retirement.
With Entitlements and the US dollar near total collapse...
Soon granny will be permanently in the workforce contributing sweat shop labor currently being done in Bangladesh.
Anyone ever seen Happy Gilmore? This will soon be reality!
http://www.youtube.com/watch?v=R7Q2XGf4TCY
Granny needs to get a job and pay more.for her insurance and not expect her kids and grandkids pay it.
there are only so many greeter positions available at Walmart. Unless we have a Deflationary Depression along with a restructuring of the system, Health Care costs are simply unaffordable for the Walmart salary.
interesting perspective by Paul Craig Roberts
http://www.counterpunch.org/roberts09142009.html
For most people, trying to trade stocks today on a short term basis is like trying to find a unique solution to a linear system of M equations with N unknowns, where N >> M. A fool's errand.
I find rubbing my crystal balz helps. At least it feels good.
Now that's a gross violation of triboelectrification rulez!
Regression to the mean is about the only certainty in probability theory, but unfortunately means are dynamic.not static.
"""TA has become irrelevant""" as did FA, and its been a good 3 years. As for Short Interest, like most any other Number we get to see, you can't trust it as factual. Just last week I had reason to do some research around this topic. I'll paste a piece of what I came across. The last sentence shows that you can't trust given numbers, no surprise, paste below:
In 2005, complaints from investors about naked short-selling finally prompted thee SEC to try to curb the scam. A new rule called Regulation SHO, known as "Reg SHO" for short, established a series of guidelines designed, in theory, to prevent traders from selling stock and then failing to deliver it to the buyer. "Intentionally failing to deliver stock," then-SEC chief Christopher Cox noted, "is market manipulation that is clearly violative of the federal securities laws."
A nice, if timid idea — except that it's completely meaningless. Not only has there been virtually no enforcement of the rule, but the SEC doesn't even bother to track who is targeting companies with failed trades.
(3) Phantom shares create systemic risk. According to the Depository Trust and Clearing Corp. (DTCC), on any given day "fails to deliver and receive amount to about $6 billion daily ... or about 1½ percent of the dollar volume." Bradley Abelow, a former DTCC director, says FTDs within the settlement system "occur as a matter of course with great regularity," and calls them "endemic." The stock market has turned into a game of "musical chairs" where claims of ownership exceed shares issued. What happens when the music stops?
The DTCC asserts that CNS has “eliminated the need to settle 96 percent of total obligations.”
If the DTCC processes $400 billion in trades daily as claimed, then $384 billion are
netted out and only $16 billion require delivery.
In a 2005 letter to the dtcc, Robert Shapiro, former undersecretary of commerce for economics, observed that the $6 billion in FTD's [Fails/Phantoms] that exist on any given day is 37.5 percent of the $16 billion in trades that require delivery— “25 times the 1.5 percent reported by the DTCC.”
SUPER BULL: Faster than a HFT, Stronger than Timmah's Breath, more powerful than a Scandel, Able to fight off Congressional Hearing, Fend off Bad Earnings, Laughs at Hurricanes and Earthquakes....IT'S SUPER BULL....
We should be trading 20% off the last high before Japan's Earthquake...but nooooooo
You need to understand that ETF open interest is very difficult to analyze. In much the same way that basis mispricing in Spoos can swing the open interest in futures by tens of thousands the same dynamics are at play in ETFs. Desks carry billions in hedged portfolio's to earn carry with an implied dividend call (if long stock short futures). This can be unwound on any open/close or through EFP.