We spend a lot of time talking about dark pool transactions. Serving the function of private stock exchanges, dark pools often tell us timely, interesting, and potentially exploitable things about the parallel universe of trading on lit public exchanges.
By now, most people know that dark pools "exist" - whatever that means - though, very few people acknowledge how important they are or that valuable information can be gleaned from them (or indeed, that the information is even available).
Readers will know that bringing this data to light is what SqueezeMetrics is all about, but today, we want to talk about a different parallel market - one that's even darker than dark pools. One that's often called...
The securities lending market
You probably know already that whenever you "short" a share of stock, you're actually selling that share on the market. As in, someone else is actually buying it from you.
Logically, to do that, you need to borrow it from someone else first, because - you know - you never had it in the first place.
Of course, the reverse of borrowing and selling is buying and returning. This is called "covering" a short.
He who sells what isn't his'n, must buy it back or go to pris'n.
Most investors understand this concept well enough, and when they do choose to short shares of stock, it seems that everything just "works." The broker seems to take care of the details.
Since this is all so streamlined (and nobody wants to be the one to ask the stupid questions), nobody ever seems to wonder:
Who am I borrowing that stock from?
Unfortunately, even if you scour your brokerage platform for details, it seems that we are not privy to this information. This might seem odd to you.
Usually, when you draw up a contract to borrow something from someone, your character, creditworthiness, liquidity, and probably your signature, all come into play. Also, a rather strong point is made of who the rightful owner is. Only when something is common, fungible, and easy to collateralize can these contractual agreements be drawn up on the fly.
Which should suggest to you - when you borrow shares to short - that there's a rather large, developed market for lending these shares sitting behind your broker somewhere.
According to Markit, one provider of this data, this large:
Where can fifteen trillion dollars in lendable securities be coming from, you wonder?
Institutions, of course
You know, pension and mutual funds - like Vanguard (read their policy on lending shares). For large institutions, lending shares is a time-tested way to make a little bit of money on the side. Because even mutual funds need a side hustle.
Now, at this point, you may be wondering why this could possibly matter to a normal investor. Try using your imagination.
[S&P quants sought alpha with data from this market - and found it]
It's very common for bondholders to look at the market for Credit Default Swaps (CDS) to get an idea of the default risk of certain debt. The CDS market is not itself the bond market, but it is mutually referential and undeniably important to the bond market.
It's also very common for equities investors to look at levels of implied volatility (IV) derived from prices in the options market to get a sense of risk and fear in the underlying (ever hear of the VIX?).
What's not common is for investors to think the same way about the market for lendable shares - the securities lending market. Why? Only because the data is neither well understood nor readily available.
So, what if we wanted to draw an outline of this shadowy parallel market for ourselves? Short of buying the data outright (it's kind of expensive), we would need to establish some sort of proxy for the "supply" and "demand" sides of the market.
Do you still remember what we were talking about? How institutions are the primary lenders in the securities lending market?
That's right - institutional ownership is our proxy for supply in the securities lending market.
Most readers will already know where to look for an estimate of securities lending demand. It's called "short interest" - the number of shares held short - and it should be part of everyone's investment toolkit already.
Unfortunately (for them), most people use short interest in one of two clumsy ways:
1. Most commonly, investors will look at "days to cover," which divides the number of shares held short by the average volume of shares traded per day. The idea is to guess how many days it would take for short sellers to buy back all of their shares. This metric doesn't even pretend to gauge demand - it's simply a (bad) tool for would-be short squeezers to fantasize about how many days a stock will go up if shorts get scared.
[Reality check: In a market where so much short selling and covering happens in dark pools - where huge transactions don't affect market price at all - short squeezes aren't very dramatic.]
2. The other way investors use short interest is as "short percent of float" and "short percent of shares outstanding." While this can be a useful way to normalize and chart shifts in short interest (if we sort of ignore dilution, buybacks, corporate actions, etc.), it also falls short in that it assumes that supply is essentially unlimited. I.e., it assumes that everyone - not just institutions - is offering shares on the lending market, even when that's far from the truth.
Fortunately, when we assert institutional ownership as the better gauge of supply in this hidden market, we get a whole lot closer to the truth without buying any extra data.
[We aren't crazy, by the way. This is not a new idea.]
Ubiquiti Networks (NASDAQ:UBNT) is an interesting stock with plenty of fans and detractors. It's also an interesting case study because its ownership and shorting statistics are unusual and confusing. Here's all the data we need (available here, or anywhere, really):
Shares outstanding: 82,958,706
Institutional ownership: 28,346,990 (34.17%)
Short interest: 11,096,198
If we were to use normal metrics, we would assess that about 13% of shares are held short, or that there are about 14 days to cover. Both of these figures seem to be on the high end of average, but they're nothing exceptional. We would expect that it would cost less than 2% to borrow a stock like this because supply seems to be a good deal greater than demand.
If, on the other hand, we looked at short interest as a percentage of institutional ownership, we would assess that over 39% of shares are held short. At this level of demand, we would expect borrow rates to be above 3%, in the low "hard-to-borrow" range. By this measure, demand is very high.
I called my broker to see if our method here is a more accurate assessment. I can only get a borrow at 5.5%.
So, if people are participating in the lending market at this borrow rate (and they are), this reflects a rational expectation that UBNT will fall substantially more than 5.5% within the year.
With this information in hand, we can look at daily and bi-monthly short selling data to get an idea of if anyone is currently shorting shares at this cost.
According to NASDAQ, over a million (1.18M) more shares were held short on 29 January than on 15 January.
According to our dark pool data, a total of 1.2M shares were shorted in the same period in dark pools (strange coincidence?).
[sqzme.co - Red volume bars are dark pool short sales.]
That means that a whole lot of investors are willing to pay ~5.5% for the privilege to short UBNT.
That should give UBNT investors pause.
This dynamic is at work in every stock on the market.
That should give you pause.
Learn more about the way the market really works. Follow us for more articles like this.
Comments and discussion always welcome.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.