Arnold Kling  

Thoughts on Short-Selling

PRINT
Two Interesting Things James H... Instructive Commentary...

What is short-selling? How can short-selling destroy a good company?

1. Short-selling explained

The easiest form of short-selling to understand is "naked" short-selling. I think that XYZ stock is going down, and so I tell my broker to sell 100 shares for me. Eventually, I will have to buy 100 shares and either realize a gain or a loss.

If the stock goes down, then it's up to me when I buy those shares. I have a profit on paper, and I take the profit when I buy the shares. I sold at $20 (say) and bought at $10 (say), so I make $1000.

If the stock goes up, my broker starts pestering me with margin calls. She asks me to put cash or short-term Treasury bills in my account. That is because she is afraid I will walk away and leave her to have to buy the 100 shares, and she'll have to take my loss for me. If I don't respond to the margin calls, she buys the shares, closes out my position, and makes me take the loss.

I don't claim to understand non-naked short selling. I imagine it works something like this. My broker gets 100 shares of the stock and gives them to me. I immediately sell them. I promise to buy them back later.

Suppose that I didn't sell the shares. Suppose I just took shares from my broker, and promised to buy them back later. Then that would be an ordinary repo loan from me to my broker, using the stock as collateral. The difference with short-selling is that I dump the collateral!

With non-naked short selling, my obligation to buy shares is with a specific counter-party, my broker. With naked short-selling, I just have a vague obligation to the "market."

In my opinion, there is no practical difference between naked short-selling and non-naked short selling.

2. How can short-selling destroy a good company?

The simple answer is that it can't.

First of all, short-selling can't force down your share price. Short-selling only forces down your share price if buyers don't emerge to defend your share price.

Banning short-selling cannot protect a bad stock. If nobody is willing to buy XYZ at a price higher than $.02 a share, then the price at which XYZ will trade will be $.02 a share (or lower). It doesn't matter whether you have short-sellers or not.

What drives stock prices down is the lack of people willing to buy them at the higher price. If the company has sufficient value, there will be sufficient buyers.


Comments and Sharing





COMMENTS (21 to date)
ErikR writes:

I was under the impression that "normal" short-selling (I guess it would be non-naked in your nomenclature?) involves the broker borrowing the shares from another client of the broker who is long the shares (and, I assume, each share can only be borrowed by one short-seller). Since the shares are borrowed, if a dividend is issued, then the short-seller must pay the dividend to the share lender.

Have I misunderstood something?

ErikR writes:

I wanted to add that I do not understand why short-selling is apparently popular relative to stock options, at least for highly-traded stocks. A short-sale is exactly replicated by buying a put option and selling a call option at a strike of the current stock price.

But the options allow more control, such as just buying a put option, or just selling a call option (depending on whether you believe the volatility is being mis-priced). Or you could trade the long put/short call and also buy a call-option with strike above stock price as insurance against a strong surge in stock price.

Now, this is not attractive if the options are thinly traded, especially if you are looking at long-dated options. But for highly-traded companies (such as AIG, Lehman, etc.) the options are liquid, so why sell short instead of using options?

Brad Hutchings writes:

Bravo Arnold. Tyler hit it on the head. McCain and Palin don't have the language to deal with Wall Street's problems. They should be saying "let it be, it needs to sort itself out". But they think that won't sell right now. Actually, that's exactly what they were saying last week, but then the Bush Administration butted in and bought AIG! So now McCain wants Cox's head for allowing naked short selling, which is just odd. Cox is finally addressing all the problems that festered during the 90s and early part of this decade, like accounting for stock options. And now he's getting smacked down for not being proactive on a non-problem.

I really look forward to Mark Cuban's comments on this. It's like McCain, in a senior moment, started channeling Patrick Byrne. Unreal.

dWj writes:

The obligation to repurchase isn't to "the general market", it's to whomever you borrowed the stock from. For most retail customers, in fact, you can be forced to close your short position if the person who lent the stock wants it back, though typically your broker will seek to borrow it from somewhere else, making you cover only if it can't.

Your description of naked short-selling raises the question, "How do you sell something you don't have?" (If I go to Home Depot, and they don't have hammers, they can't sell me a hammer.) The answer usually is by borrowing it. The answer, naked, is that "sell" means "agree to terms under which to trade paper for money in three business days". You don't have to borrow the stock if you don't mind failing to deliver on your promise.

The mechanics of shorting a treasury security are exactly what you indicate: you repo it in and sell it. The reason securities in the repo market sometimes go "on special" is because people are willing to forego a normal "risk-free" rate of interest in order to be able to short the security.

Dan Krabach writes:

One needs to borrow a stock before short selling so that the short selling does not cause more open positions than there are actual shares of stock. If there is a significant amount of naked short selling going on (which no one has demonstrated to be the case, just blamed), this could cause an artificial downward pressure in stock prices since there would be more sales than actual available shares. Also, short selling requires borrowing real shares for voting and dividend purposes. This is why the SEC requires brokers to make sure their house has shares available to be borrowerd (usually from customers who have margin accounts) before fulfilling short orders. This is one reason of why options markets have been.....invented.

Marcus writes:

"In my opinion, there is no practical difference between naked short-selling and non-naked short selling."

I think you're confusing issues as there's a big difference. Naked-shorts is essentially counterfeiting, increasing the number of shares in the market (thereby reducing their value).

In short-selling, you have to borrow the shares first, making them unavailable to somebody else. Your obligation is that you must cover their position and pay them dividends.

*YOU* probably can't perform a naked short. If you tell your broker to sell a hundred shares of something you don't have, behind the scenes the broker borrows those shares from other clients with margin accounts who own those shares.

Naked shorts occur when your broker can't deliver the shares you shorted to the broker of the person who bought them. This can happen because it takes 3-days to clear a transaction. Lots can happen in 3-days.

cjh writes:

"Naked short" can mean two different things. The one that you refer to is the sense of being unhedged - naked as opposed to covered - i.e., you don't have a long stock or options position that offsets the short position.

The definition that the SEC is worried about has to do with whether you have actually borrowed the shares you're selling prior to the sale. If not, you could try to borrow them between the sale and the settlement date, or have a fail-to-deliver.

All explained here: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=982898. The authors conclude that the latter definition of naked doesn't have economic significance.

cjh writes:

I think I misread the first part of the post. But basically the difference has to do with the fact that some people (as Marcus says, not ordinary investors) can sell stock on an exchange on one day but then don't have to produce the shares until settlement several days later. Nakedness has to do with when the seller acquires the shares that he sold.

Ray writes:

The simplest way to understand why naked short selling depresses the "true" market price of a security is that it effectively increases the actual supply of the stock and (slightly) decreases the demand (by satisfying some of it), hence a lower price. In theory, there's no limit to the amount you could naked short sell, which has the potential to increase the supply of the stock wildly.

Some stocks have short positions larger than the entire outstanding supply of "real" shares. Defend, if you can, the position that this doesn't decrease the market value.

Covered short selling doesn't increase the *actual* supply of the stock, because the shares are taken off one person's books and added to another's. However, it does increase the *effective* supply, because the original owner presumably wasn't willing to sell at the current market price. You've supplied a new willing seller, which can depress the prices, but not as severely as naked shorting.

You're also assuming that people buy stocks rationally, which isn't supported by the evidence.

There isn't 1 true market price in stocks, there's a curve of various people willing to buy at various prices. By satisfying the demand at a particular price, you're effectively lowering the market price. As you say, if someone steps up to protect the price, it won't go down, however, there's an arbitrage period in there where new buyers can't be found, resulting in temporary lower prices. The *psychology* of market forces makes it more difficult to find new buyers when the price is decreasing. Long term that may correct, of course -- if people are acting rationally. But in a panic, great damage can be done.

Walt French writes:

"I don't claim to understand non-naked short selling."

An overlooked feature of the borrow is that the lender often has NO CLUE that his broker has loaned out stock XYZ. Why? Because all economic benefits stay with the lender. She STILL owns it, legally. Her broker doesn't know, either; only a green-eyeshade guy in the back office knows.

Recently, the "short interest ratio" -- how many shares are loaned out, divided by the number of shares in circulation -- exceeded 100% on some (financial) ETFs. Owner A loaned the shares to short Z, who sold them to another would-be owner, B. B's broker then loaned them to Y, who sold them to C, etc. In total, adding the number of shares held long by A, B and C, minus those sold short by X and Y leaves exactly as many shares outstanding as before. A price increase or decrease still increased the total wealth of A, B, C, Y and Z by exactly as much if only A had held onto the shares.

Except that the brokers get a pretty penny for arranging these loans. Shorts have to be more right than the average investor to make any money. They'd better also have at least $1 of LONG positions on similar names so that they don't get beaten up when the market goes up overall, irrespective of how right they are on their anti-picks.

The price of the shares will reflect the consensus of all parties, in the honest, vote-with-your-wallet system that economists generally revere for finding a fair price.

Except that whereas stockholders -- especially, you might think, as I do, money managers and corporate insiders -- have an incentive to spread happy horsemanure about the stock so they can sell it higher, before the truth comes out, while short sellers have an incentive to spread vile rumors and drive the price down before the truth comes out. Insiders often believe it's part of their job to talk up the stock price, as do the parade of managers each day on the financial TV shows. They're only are prosecuted in the most blatant cases when they knowingly lead investors to buy into utterly Potemkin companies, with provably deceptive intent.

Shorts are generally individuals without so much as a soapbox, merely the same analytical skills as the rest of us, and the courage to take some really risky bets in the name of making a few bucks. Yes, some can spread false rumors that the market will knee-jerk respond to, and make a fast buck, but they poison their own nest because the "tips" exploit the people they spread their lies to. You have to keep moving.

There are times, such as today, when investors are extremely jittery about whether there are skeletons in firms' closets. Rumor-mongering works especially well in those times. But the recent victims of short-selling? Virtually every one of them were actually victims of their own bad behavior: excessive credit, toxic waste hidden off the corporate balance sheet, and occasionally some old-fashioned malfeasance. My statistical studies find that short-sellers, who enjoy reputations as pariahs, stick with it ***because*** they're so often right that they can make a bunch of money, perfectly honestly, by pointing out that the emperor has no clothes.

If we wanted to make dishonest short-selling uneconomical, we would go back to the SEC's original charter and strengthen the accounting standards and give investors more confidence in the stocks. Recognize that American stock markets are the choice of investors because dealing is open and honest. That the SEC prohibits Enron-type off-books cesspools. That an investor doesn't want to find out after the fact that the company he just bought part of has promised half its gains to employees' options. Hey, Mr. McCain, a reason to call for Cox's head!

Then, and only then, will unethical short-sellers have no hope of moving markets against good firms.

Methinks writes:

I think you're confusing issues as there's a big difference. Naked-shorts is essentially counterfeiting, increasing the number of shares in the market (thereby reducing their value).

Well, then selling a futures contract on oil is counterfeiting barrels of oil. There is no difference between a naked short sale and a future on the stock.

Moreover, naked short selling is not and has never been illegal. Market makers (both options and stock) are permitted to sell naked to fulfill their duty of providing liquidity by maintaining a two sided market. Also, pre-borrowing is not the rule. Everyone but market makers and specialists must "locate" the stock before shorting, not borrow it. Whether your locate and eventually borrow or not, the seller is required to deliver the stock at some point in the future.

Most of the persistent failures to deliver stem from a loophole used by options market makers. When time is up and they get a buy-in notice, they simply sell the stock to themselves (remember: no locate requirement) and "cover" their short. Of course, that covers nothing, but it restarts the clock for delivery.

I wanted to add that I do not understand why short-selling is apparently popular relative to stock options, at least for highly-traded stocks.

I agree in general. When there's heavy demand for the put options, the put premium increases. The higher the option premium, the lower the price at which you're effectively selling the stock. At some point, it's cheaper to sell the stock short than to buy the put.

Methinks writes:

The *psychology* of market forces makes it more difficult to find new buyers when the price is decreasing. Long term that may correct, of course -- if people are acting rationally. But in a panic, great damage can be done.

I think you put too much weight on psychology and I'm not sure you can prove that people don't buy and sell shares rationally. It's very difficult to judge rationality because everyone's trading is dictated by their individual goals. What looks like a dumb sale to you may make perfect sense in the context of the three legs that make up my trade. So, how do you objectively measure rationality to be able to make such an assertion?

Preventing short selling makes the markets less liquid and more prone to bubbles. It also removes the ability to hedge. Even if you're hedging by buying puts, the options market maker is hedging his short put position by shorting the stock. If the options market maker can't hedge his position by selling the stock, the option premium will be pretty close to the difference between the price of the stock and strike price.

If you think that dumping the stock has a negative psychological effect on people, just wait and see what happens when an inability to hedge levered long positions combines with an illiquid market and an asset bubble.

Oh wait! We don't have to wait because that exactly describes the market conditions for those "toxic" assets held by all these fine, now bailed out institutions. I'm a little dramatic about it, but it's not hyperbole.

Methinks writes:

An overlooked feature of the borrow is that the lender often has NO CLUE that his broker has loaned out stock XYZ.

So what? Do you know that when you "own the stock" you don't actually own the stock? Moreover, when you sign your brokerage agreement, one of the things that you give your broker permission to do is to lend out your stock - provided it's in a margin account.

Shorts have to be more right than the average investor to make any money.

SO true. The tail risk of a long position is the difference between the price paid and zero. The risk of a short position is infinity...because there's no limit on how high a stock price may rise. Your point about hedging is also correct. The majority of short positions are to hedge long positions - either via options or straight shorting. A short position without a long position

Recently, the "short interest ratio" -- how many shares are loaned out, divided by the number of shares in circulation -- exceeded 100% on some (financial) ETFs

If true, it's not a big deal. Unlike a regular stock, shares of ETF shares are constantly created, thus the float is always changing.

Then, and only then, will unethical short-sellers have no hope of moving markets against good firms.

Walt, I agree with much of what you said - particularly in the last few paragraphs. However, the shorts in heavily shorted stocks - like the financials - were outnumbered 5 to 1. It's practically impossible to manipulate a heavily traded stock with a giant float. Although, I agree that manipulation does happen, it's rare and should be prosecuted as all fraud should be prosecuted. The problem is that in response to the ignorance of the general population and to make its job of ferreting out the very fraudsters easier, it slaps blanket Orders on all short sellers. This market is jittery enough. Adding regulatory risk and sucking liquidity out of it now is not making things better.

ErikR writes:

When there's heavy demand for the put options, the put premium increases. The higher the option premium, the lower the price at which you're effectively selling the stock. At some point, it's cheaper to sell the stock short than to buy the put.

Methinks you are a bit confused about using options to replicate a short sale. The price of an option is made up of intrinsic value and time value. The intrinsic value only depends on the strike price and the stock price. The time value depends on several things, most notably the volatility of the stock.

If the put option is "mispriced" for some reason, then a way to think about it is that the stock price volatility is not being properly accounted for by "the market". If the volatility is overvalued (i.e., the put is overpriced as you mention), then you may overpay to buy a put, but when you write the call option, the buyer of the call option will overpay YOU for the volatility. So the long put short call is still a viable alternative to a short sale. Or you could just sell a call (without buying a put) to take advantage of the overpriced volatility. But then you would not replicate a short sale exactly.

The high cost of replicating short sales with options is when the options are thinly traded and the bid/ask spreads are high. But this usually is not a serious problem with high-volume stocks and at-the-money options.

Methinks writes:

Erik,

You're right. However, there's a market in volatility - even in very liquid stocks. The more uncertainty there is, the wider that market. I'll explain more about what I meant later as it's very busy here this morning - what with the short sale ban.

Not so incidentally, I agree that it's more efficient to hedge in the options market. It can also be much more efficient to create synthetic long positions using options. I think the better answer to your puzzlement is that many people (even money managers) don't understand derivatives very well. Also, if you have a long/short strategy, the market in volatility in the options on the stocks you're trading may be wider than than the theoretical edge in the trade. This is often the case in very liquid stocks.

Marcus writes:

"Well, then selling a futures contract on oil is counterfeiting barrels of oil. There is no difference between a naked short sale and a future on the stock."

Hi Methinks. I very much value your opinion and when you disagree with something I've written, I try to rethink my position.

Yet, I think that is a very bad example you've put forward there. Naked-shorts aren't anything like oil futures. The person buying the shorted stock does not in any way believe he's purchasing a future claim on a stock.

Perhaps counterfeiting isn't a good analogy either. Perhaps it's more like kiting a check.

gb writes:

Noting these comments from Ray which sum up short selling very well:

The simplest way to understand why naked short selling depresses the "true" market price of a security is that it effectively increases the actual supply of the stock and (slightly) decreases the demand (by satisfying some of it), hence a lower price. In theory, there's no limit to the amount you could naked short sell, which has the potential to increase the supply of the stock wildly>
There isn't 1 true market price in stocks, there's a curve of various people willing to buy at various prices. By satisfying the demand at a particular price, you're effectively lowering the market price. As you say, if someone steps up to protect the price, it won't go down, however, there's an arbitrage period in there where new buyers can't be found, resulting in temporary lower prices.


My question is this:
"Is it sensible for big index-tracking funds to lend stock to short-sellers?"

- and more worryingly and concerning TRACKING funds:
I understand that many tracker funds are not actively managed. They are simply set up to buy, sell and lend shares according to simple financial algorithms, so:
"If a financial company were to lend from a TRACKING fund to a short-seller would it be possible that given enough financial clout and knowledge of the financial mechanisms by which the fund operated that a short seller could externally manipulate that fund by operating within very short time spans and handing back the shares at the point below the threshold where the fund would be triggered to sell or even worse lend again?"

Methinks writes:

Yet, I think that is a very bad example you've put forward there. Naked-shorts aren't anything like oil futures. The person buying the shorted stock does not in any way believe he's purchasing a future claim on a stock.

Please let me try again. A future is an agreement to deliver a specified amount of something at a specified price on a specified day. The thing you will deliver may or may not exist at all at the time the futures contract is executed.

In selling a stock short without a locate or a borrow, you are still agreeing to deliver the stock at some point in the future at a specified price.

There are obvious problems with this. The most obvious and one I don't know how to even think about solving is the fact that when there are persistent failures to deliver on a large enough percentage of the float, there can be problems with proxy votes.

What do you think?

BTW, here's a great paper on naked short selling by the Cato institute that I thought you would enjoy.

http://www.cato.org/pubs/regulation/regv31n1/v31n1-6.pdf

knzn writes:

The argument that short-selling can't "destroy a good company" or "force down your share price" depends on the presumption that everyone except the short-seller agrees about the value of the company (or, alternatively, that the few who do agree have unlimited capitalization, so that they will be willing to buy any number of shares that are short-sold). In the more realistic case where market participants have a variety of opinions and limited capitalization, as I argue in more detail on my blog, short sellers certainly can force down the share price and, under some conditions, destroy a good company.

methinks writes:

knzn,

in the more realistic case where market participants have a variety of opinions and limited capitalization, as I argue in more detail on my blog, short sellers certainly can force down the share price and, under some conditions, destroy a good company.

To do that, the shorts would require not only unlimited capital (which you concede, they don't have) but also an unlimited appetite for risk. Let's also not forget that the maximum risk of a long position is the difference between the price paid and zero, while the risk of a short position is infinity. Shorts are heavily outnumbered in the stock market for this reason and you're going to have to explain to how a minority activity with a much higher risk profile can suddenly drive companies out of business.

Further, why would shorts take such risks and commit so much capital to drive down the shares of a good company when their are so many bad companies which are so much easier and less risky to short?

Also, why not punish the longs who drive up the price of crappy company - or any company, really - to above fair market value? As we all know, bubbles are pretty painful when they burst and irrational exuberance on the part of the buyers causes them.

Martin writes:

I think most of us agree that it is logical to allow short selling as a useful mechanism to maintain market liquidity and allow the market to find the right price. Only we do know that markets behave very irrationally at times; the dot com bubble is just one example. Fair enough when markets go wild and speculators get burned but put that into reverse and, particulalry in the case of banks, the negative sentiment can be enough to start a "run" and ultimately bring the whole business down. Hence the need to ban short selling is as much about removing the fear of short selling as it is about the effect on supply and demand itself.

Comments for this entry have been closed
Return to top