Ask HN: Please explain short selling?

That's it. In it's entirety.

Depending on the value of your portfolio/margin account, if the value of the stock goes up too much, they may ask for more collateral, etc. But other then that, there isn't anything to it.

Shorting a stock is a bet against the price of the stock. Honestly, it doesn't make sense as much as buying a put option in many cases, but sometimes it does. http://en.wikipedia.org/wiki/Put_option

As an accountant, all I see is that someone sold 1200 XYZ shares at $23.00 each, receiving a grand total of $27,600 cash. To simplify the example, I am not subtracting any commission there.

Now the funny thing is, this account didn't have any XYZ shares before the trade. Nevertheless the trade happens, and the account has -1200 XYZ shares, and $27,600 more cash than it had before.

At that moment, the account has incurred a liability to buy back 1200 XYZ shares -- eventually. The account can hold this liability indefinitely, so long as it has enough total capital to reassure the brokerage that it can easily buy back the 1200 XYZ shares at any time.

I've read the stories about "naked short selling," and how that's a giant scam because the seller doesn't even have to locate any shares to borrow. In this telling of the story, the seller is issuing brand new shares and selling them into the market -- in effect, counterfeiting. I spoke with a very experienced money manager about this, and that's his take.

I also spoke with a friend who is very wise about markets, and he had an entirely different take. He made the point that no matter how the short sale occurs, either "naked" or "covered", in both cases the seller ends up with the liability of -1200 XYZ shares, so what's the difference? The only difference is that if the short sale is naked, the seller owes the shares to the buyer. But if the short sale is covered, the seller owes the shares to the lender. Either way it's a liability of -1200 XYZ shares, and the only difference is to whom the shares are promised.

Now I'm a rational man, so it bothers the hell out of me to agree with both sides of a contradiction. So what gives? I don't know. The only other shred of evidence I have, and it's a very small shred, is the phenomenon of "Payment in Lieu of Dividends." If I have negative quantity of shares, and those shares pay a dividend, then that dividend is charged to my account as an expense. Normally, as an accountant, I see that as a simple negative dividend, which we call a "Dividend Expense." But sometimes I see those negative dividends labeled "Payment in Lieu of Dividends." I read an article once saying that the only difference is that "payment in lieu" occurs when the shares were sold short in a "naked" fashion.

So maybe, just maybe, there is a real difference between the two forms of short sales, as my money manager friend asserts. The money manager asserts that naked short selling actually increases the total supply of shares, even past the official "float" of shares issued by the company!

The people at the Gold Anti-Trust Action Committee (GATA) even claim that short sellers are creating this artificial counterfeit supply in the gold market. I would suggest calling their bluff and redeeming the physical gold into allocated Swiss storage -- but lo and behold, these financial instruments have no redemption contract. They're cash-settled only. How conveeeeeenient.

You asked if the people lending the shares actually hope for the stock's success. Of course they do. If someone borrows 1200 XYZ shares and sells them, the lender still has an asset of 1200 XYZ shares, and clearly still wants their price to rise. The seller has a liability of -1200 XYZ shares, and clearly wants their price to fall. The buyer has an asset of 1200 XYZ shares, and clearly wants their price to rise.

Note that in my example there, the total net quantity of shares in the three accounts involved is precisely 1200, both before and after the short-sale. So where is this alleged increase in supply from short-selling? Perhaps there's no increase in this case because it wasn't a naked short sale.

OK fine, let's make it a naked short sale. A trader simply sells 1200 XYZ shares without either having them or bothering to borrow them. Another trader buys those 1200 XYZ shares. Now the seller has a liability of -1200 XYZ shares, and the buyer has an assert of 1200 XYZ shares. So after the short sale, the total net quantity of XYZ shares in those two accounts is precisely 0. But the net quantity was precisely 0 before the short-sale as well! So once again, where oh where is this alleged increase in supply caused by short-selling?

When you trade, you must trade through a broker-dealer. A broker-dealer is authorized to trade on behalf of it's customers. A broker-dealer must uphold certain regulatory requirements put in place by the SEC, and policed by a variety of government and non-government entities, including organizations like FINRA.

Originally, shorting was managed by the broker-dealers. It was the responsibility of the broker-dealer to manage finding an entity to "borrow" the stock from (usually a large institutional client). These institutional clients own very large positions in the stock, and the broker-dealer normally provides a guarantee that it will be returned. If for some reason this process was mismanaged, when the trades cleared and settled, there would be a "fail to deliver" (meaning they weren't able to come up with the stock) This is a severe problem for the broker-dealer, and can result in them losing their license. Abusing this system became known as "naked shorting" where you never made an attempt to "locate" (borrow) the stock. Naked shorting was severely curtailed in 2005 under Regulation SHO.

You may also be curious how a customer knows what stocks its broker-dealer can borrow. Originally there was a black-list of sorts called the Hard-To-Borrow list. This list was stocks that were relatively illiquid and that couldnt be shorted freely. If you wanted to short these, you had to request a "locate" from your broker for a specified number of shares, and they would go looking for someone to provide it, and confirm how many shares they could find. Under Regulation SHO, this changed to a white-list "Easy-To-Borrow" model where you were only allowed to short stocks on the list freely and had to request locates for the remainder of the symbol universe.

Finally - another interesting note on shorting called the "uptick rule". The uptick rule was originally put into place in the 1980's under Rule 390. This rule was built to prevent a stock from being run into the ground by repeatedly shorting it while the price was already falling. It required that in order to short, the previous print (quote) must have been higher than the one before (the stock was heading up). This really didn't help the problem too much, and was repealed in 2007.

I don't play the market ... I buy mutual funds for my Roth IRA, and I sit on them. And sit on them. And sit on them. But if you were to reduce my actions down to the bare basics, it would sound just like the initial comment: "you buy some stock, believing it will go up, so you can sell it at a profit at a later date". Just because my window is 30 years, and not 30 days, it doesn't change the basic mechanics or principles.

If you expect a stock value to go down lower than X dollars, you commit to sell them stock at >X$ (without owning) and, when the time comes and your expectations (about the stock going down) become true, you buy for X$.

-- MV