Christopher Culp & J.B.Heaton have written an essay on The Economics of Naked Short Selling. They review the mechanics and economic theory of short selling to conclude:
There is little meaningful economic difference between the two forms of short selling … The only difference is who acts as the effective lender of the security … The buyer, after all, is now in the position of the security lender and has a very solvent counterparty in the NSCC [National Securities Clearing Corporation].
The Depository Trust and Clearing Corporation itself has Question & Answer page … from 2005!
Certainly there have been cases in the past where it has, and those cases have been prosecuted by the SEC and other appropriate enforcement agencies. I suppose there will be cases where someone else will try to break the law in the future. But I also don’t believe that there is the huge, systemic, illegal naked shorting that some have charged is going on. To say that there are trillions of dollars involved in this is ridiculous. The fact is that fails, as a percentage of total trading, hasn’t changed in the last 10 years.
Now, as far as I can see from SEC Form X-17A-5 PART II, shorts on a firm’s books resulting from a fail to receive are merely marked-to-market; there is no requirement that the position be over-collateralized by either the customer or the firm.
And this is the crux of the issue. If I am correct, and naked short-selling is simply a methodology to get around margin requirements, then it is the margin requirements that need to be fixed.
I have posted a question on Jim Hamilton’s blog … we shall see!
Update: The above is rather cryptic, isn’t it?
I’ve been puzzled about naked short selling and why it is considered the epitome of evil; by-and-large taking the view of Culp & Heaton. The problem – as I see it – is counterparty risk. If somebody naked-short-sells you a million shares of Morgan Stanley, that, in and of itself, is no big deal. You don’t have to pay for them and as long as the price doesn’t change, there’s no big risk.
The risk is that the shares will go up a lot and the counterparty will go bust, leaving you high and dry … particularly if you’ve taken other market action based on your purchase.
As I noted yesterday, Accrued Interest thinks a lot of hedgies are going to go bust in the near future and I agree with him. The ones who shorted financials on Thursday go first.
And we have seen a lot of problems lately with undercollateralization of exposure. If CDS exposures had been adequately collateralized, there would not have been nearly so much of a problem caused by MBIA, Ambac and AIG. If the parties at risk on naked shorting of financials turn out to be the financials themselves, we could have a very interesting co-dependency conundrum!
So I want to know, but I don’t know: what are the over-collateralization requirements, if any, on Fails-to-Receive?
Update, 2008-9-21: I have found a paper by Leslie Boni of the UNX and University of New Mexico titled Strategic Delivery Failures in U.S. Equity Markets, abstract:
Sellers of U.S equities who have not provided shares by the third day after the transaction are said to have “failed-to-deliver” shares. Using a unique dataset of the entire cross-section of U.S. equities, we document the pervasiveness of delivery failures and provide evidence consistent with the hypothesis that market makers strategically fail to deliver shares when borrowing costs are high. We also document that many of the firms that allow others to fail to deliver to them are themselves responsible for fails-to-deliver in other stocks. Our findings suggest that many firms allow others to fail strategically simply because they are unwilling to earn a reputation for forcing delivery and hope to receive quid pro quo for their own strategic fails. Finally, we discuss the implications of these findings for short-sale constraints, short interest, liquidity, price volatility, and options listings in the context of the recently adopted Securities and Exchange Commission Regulation SHO.
In the text:
Any clearing member with a failure-to-receive position has the option of notifying the NSCC that it wants to try to force delivery of (“buy in”) some or all of that position. Evans, Geczy, Musto, and Reed (2003) provide evidence that buy-ins may be rarely requested. Using fails and buy-in data from one major options market maker for the period 1998-1999, they find that the market maker failed-to-deliver all or at least a portion of the shares in 69,063 transactions. The market maker was bought-in on only 86 of these positions. An interesting question is why clearing members that fail to receive shares allow the fails to persist.27 The following explanations have been suggested by market participants.
1) Costs of failures to receive are small. Regardless of whether shares are delivered, long and short positions are marked-to-market each day. Although long positions that fail to receive shares forego the opportunity to lend them, short interest levels and lending as a percentage of outstanding shares are low on average.
2) Clearing member may have to recall stock loans that have been made via the National Securities Clearing Corporation (“NSCC”) before requesting buy-ins.
3) Bought-in shares will themselves have a high probability of delivery failure.
4) Firms that fail to receive, by not forcing delivery, hope to bank future goodwill with those that fail to deliver.
I’m not concerned about the price-discovery process, or possible distortions thereto that might be created by naked short selling. I am concerned about counterparty and systemic risk. These risks are best addressed through ensuring that fails are adequately covered by capital; applying a capital charge – with mark-to-market – is the most direct way of addressing these risks.
I’m not that bright but it seems to me that any market transaction, whether going long, shorting, buying on margin, buying puts or calls, requires at least two parties – a buyer and a seller – commonly called a market. Naked short selling seemingly does not require two parties. A naked seller can “sell” shares not borrowed from anyone which, under the existing rules, seems legal, if he can “reasonably expect” to get them before settlement date. What if no broker or no shareholder wanted to lend him these shares? How would he live up to his responsibility to produce them? Well, if allowed to short without taking possession or having an agreement in place with a ‘lender”, it’s no problem under today’s rules. Short till they tank, then simply buy them up, no second party required and, to me, that’s the entire crux of the thing. Would Lehman have been anxious to lend blocks of their shares to shorters? Lehman shareholders? I doubt it. The lender is betting the share price will go up, not down, but if no lender is required and you can short enough shares (no problem if you don’t have to have them to begin with) it becomes self-fulfilling, inevitable and unstoppable. Short selling leaves actual shareholders without any input and totally out of the market equation and, in my naive opinion: it ain’t right
David Patch shares your view and decries the ability of large hedge funds to induce panic selling by intra-day naked short selling. He states:
Well, I don’t consider those who panic to be investors either. An actual investor, by and large, will buy when the price is less than value, selling when the price is more than value; he will say “thank you very much” whenever price and value get out of whack by enough to justify a trade. His ability to determine value will determine his returns; hopefully, his returns will have a great influence on his AUM.
SAC Capital should be thanked for (a) providing liquidity, and (b) helping to drive bozos out of the market.
My only concern is that when I buy a block of shares of $10, I want to get delivery. I am concerned about counterparty risks. If these naked shorts and fails to receive are overcollaterallized and margined properly, my concerns are greatly diminished.