2 take possession of their shares in certificate form, there is no increase in book entry shares, but in that case the total number of certificate shares held by investors in their own names increases instead.) Consequently it is commonplace, when a significant short interest has developed in the stock of some firm (call it XYZ Corp.), for individual brokerage firms to show on their customers' accounts that the customers own more shares of XYZ Corp. than the brokerage firms own as record owners on behalf of their customers. The difference is exactly matched by the short positions those brokerage firms hold for other customers or for their own accounts, or by shares they have lent to other firms for short sales. In the aggregate, therefore, when there is a significant short interest outstanding in the shares of XYZ Corp., the shares owned directly by investors in their own name, when added to the shares owned beneficially by other investors in their accounts with brokerage firms, will generally add up to more shares than XYZ Corp. has ever issued. Short Sales and Proxy Voting One important implication of this multiplication of shares through short selling is its effect on proxy voting. Whenever the quantity of shares of a company's stock held by investors has been increased through short sales, the investors who are shown in brokerage firms' records as owners of stock in that company cannot all vote their shares at the company annual meeting, either in person or by proxy. In practice, we are told, many investors do not bother to try to vote, so brokerage firms allocate their limited voting rights (determined by their record ownership of shares) among the customers who take the trouble to sign proxy cards. The fact remains however and the SEC acknowledges this that if every investor wanted to vote his or her shares, they could not all do this if short sales have increased the total outstanding shares. This situation creates the potential for serious abuse because brokerage firm customers are generally not aware of this potential limitation on their right to vote. To the best of our knowledge it is not disclosed to them when they sign lending agreements with their brokerage firms, and there appears to be no mechanism to inform them later if their shares have been lent out and their vote may not count. On the contrary, they are routinely sent proxy cards as if their votes would count, thus giving the misleading impression that their vote is assured of being counted if they return a signed proxy. Moreover, based upon the subcommittee's informal inquiries to date, there seem to be no rules or standards that govern which excess proxy votes 3 by customers will be discarded or invalidated, if a brokerage firm receives more customer votes than it can actually turn in. In particular, it would appear to be consistent with existing rules for a brokerage firm to select for discard the particular votes of certain customers whose voting choices it finds objectionable. Finally, there appears to be a possibility, under present regulations, for interested parties to make use of large block short sales to manipulate the outcome of corporate proxy contests. To accomplish this an interested party would sell short the selected number of shares while simultaneously purchasing the same number of shares in a different account name or through a different firm. In this way the interested party would get voting control of a certain block of shares without affecting the stock price and without taking any investment risk. Such a transaction could then be unwound immediately following the record date for determining who was entitled to vote, and any such short sales would typically not show up in the short interest statistics reported monthly. The voting rights would come automatically in such a transaction if shares are borrowed to complete the short sale normally. It also may be possible, however, for an accommodating brokerage firm to obtain voting rights for a particular customer at less cost through overnight "naked" short sales that are unwound the next day. It seems to be routine practice for brokerage firms to forward proxy material to all customers owning a particular stock, even if the brokerage firm does not have record ownership of that number of shares. Consequently, if large numbers of other customers who own that stock do not return marked proxy cards, the brokerage firm would appear to be free, within the constraints of its total record ownership of that stock, to honor the voting preferences of the preferred customer to whan it sold shares through an overnight naked short sale. Market Price Effects of the Multiplication of Shares Expanding the quantity of tradeable book-entry shares of an issuer's stock in this manner through short selling also has a direct supply effect on the market price of those shares. This opportunity for short sellers to alter at will the supply of a corporation's shares that must be absorbed into investors' portfolios raises fundamental issues about the nature and purpose of equity markets. Through this mechanism, corporations lose control of the quantity of their shares outstanding in the public equity markets. This analysis shows the fallacy in the usual argument that an outstanding short interest is bullish for a particular stock. The usual argument has been that the short sellers must eventually cover their positions by buying shares, which should support or drive up the stock price. This argument relies on the false premise that there is an increased potential demand on the buy side without any increase of supply on the supply side of the market for that stock. In fact, since short sales increase the effective supply of shares by the amount of the short interest, there are more potential sellers of shares also, holding more shares in total, than was true before the short sales. This supply increase effectively offsets the effect of any potential demand increase. "Naked" Short Selling An additional set of questions arises in connection with the practice of naked short selling, by which I mean the practice of engaging in short selling without borrowing shares to deliver to the buyer's broker. This practice, according an article in the February 8, 1988, issue of Forbes, is permitted in the over-the-counter markets, although it is against the rules of the major exchanges. Under current NASD rules it is particularly easy for market-makers to engage in naked short selling. The normal requirement that borrowed shares be obtained by the short seller for delivery to the buyer's broker imposes a significant cost on the short seller. Naked short selling thus provides a significantly less costly way to engage in short selling that presumably encourages more widespread short selling. In fact, the argument has been presented to the subcommittee that naked short selling should be generally permissible for all investors in all markets, since this would (it was argued) reduce "trading friction" and increase market efficiency. If naked short selling were generally permitted, a shortage of shares available for borrowing would no longer prevent short sellers from acting on their negative investment judgments about certain stocks, and as a result (it was argued) market prices would more accurately reflect all the available information, both good and bad, about each company. On the other hand, the absence of any transfer of borrowed shares to complete such naked transactions creates fails in the securities clearing mechanism. The seller's broker still owes shares to the buyer's broker in each such case, and the failure to deliver borrowed shares leaves an unresolved or open position in the clearing mechanism. If such fails are allowed to persist and build up in the clearing mechanism, this could cause problems, but in an active market in which many investors are shorting naked through different firms, netting in the clearing mechanism may keep the unresolved fails quite low. In general, the implications of such delivery failures in the securities clearing mechanim are not clear to us. We mailed a questionnaire letter to approximately 200 companies in May and June asking them a series of questions about their experiences as a result of heavy short selling activity in their stock. This letter was accompanied by a "Statement of Concerns". Both of these documents are attached. A followup letter was sent in late July to the companies which had not responded to the first letter. The questions dealt mainly with what practical effect the short selling and the related activities of short sellers had had on their company, whether they had experienced any disruptions or distortions of the proxy voting process, and how they felt about three suggested changes in the regulation of short selling. The three regulation ideas proposed in this letter were (1) mandatory public reporting of their short positions by short sellers if their positions exceed some percentage of a company's outstanding shares; (ii) an uptick rule for short sales on NASDAQ stocks; and (iii) a mandatory buy-in rule to reduce naked short selling. The mailing list for this survey was developed by selecting companies that had shown a ratio of total short interest to their public float of shares of at least 10 percent at some time in the period from December 1986 through April 1989. (In the case of New York Stock Exchange listed companies, I could not obtain float data and therefore compared the short interest of each company to its total shares outstanding.) I developed my own database for tabulating the monthly short interest data released by the NYSE, American Stock Exchange, and NASD. Tables showing the maximum short interest ratios of the companies in the survey are also attached to this memo. We received a total of 68 responses, for a response rate of about 34 percent. 30 companies reported no problems or complaints arising from short selling activity, while 38 reported problems of various sorts. Several of those reporting no problems had very substantial short 2 positions arising from hedging or arbitrage transactions involving convertible securities, but did not feel there was any need to complain. Widespread circulation of false rumors around the time of heavy short sale activity was cited by 21 companies as a serious problem. They generally reported that these rumor problems, at the very least, made it necessary for company officials to devote inordinate amounts of time to reassuring stockholders, regulators, customers, and sources of debt financing, who were often seriously unsettled by the reports being circulated. Thirteen companies characterized the short sellers' activities as involving improper interference with their relationships with customers, major shareholders, suppliers, banks, etc. Generally they complained of numerous phone calls, often anonymous, to these parties from "analysts" attempting to suggest very negative, frequently false, conclusions. Only three companies cited shareholder problems obtaining possession of their shares. Only a small number of companies asserted that naked short selling was a significant problem. Most companies responded that they had no factual basis for determining whether naked short selling was a problem or not because they could not obtain the necessary data from their exchange or the NASD. No companies reported any complaints related to shareholder proxies. Many companies that reported specific complaints also expressed the view that short selling is a legitimate market practice and that the only need is to curb specific abuses. RULE PROPOSALS Out of the 68 substantive replies received, 37 commented in some manner on one or more of the three policy ideas suggested in the survey letter. The views expressed in these 37 letters were as follows: 1. 2. 3. Requiring public reporting of large individual short positions: Imposing an uptick rule on short sales of NASDAQ stocks: Among OTC companies, 13 in favor; 3 opposed; 21 no comment Imposing a mandatory buy-in rule or some other step to prohibit Various rule changes to assure informed consent by investors whose shares are lent to short sellers were suggested by 7 companies. |