may desire to stablize the price of the security during the offering or he may want to protect against a possible flooding of the market after the offering. This is most often done by over-alloting the issue at the time of the offering. The resultant short position provides the underwriter buying power to stabilize the price of the security during the offering, to offset any cancellations by selling dealers and to absorb, through covering purchases in the open market, any unusual selling of the securities after the offering is completed. d. A brief review of the regulation of short selling (1) The exchange rule; 1935 The Commission did not at first promulgate a short selling rule of its own, but in 1935 requested the exchanges to regulate the practice. The rule adopted that year by 16 exchanges required that a member should not effect a sale which would demoralize the market. This merely codified requirements which had been in force on at least the New York Exchange since 1931. A portion of the financial community had long considered that a sale at a price lower than that of the last sale exerted a demoralizing effect, particularly when the lower price represented a short sale. Accordingly, the 1935 exchange rule prohibited the short sale of a security at a price below the last sale price. The Commission voiced a hope that the rule would "preserve those features of short selling which are in the public interest." 521 (2) The Commission rules; 1938 The autumn of 1937 brought a sharp drop in the market, and the Commission immediately began a study of the market decline and a reassessment of the exchanges' short selling rule. This study led to the promulgation of a set of Commission short selling rules, which went into effect in February 1938. The Commission's three rules were designed to correct some of the limitations of the 1935 provision. The first rule defined a short sale, something which the exchange rule had failed to do. The definition, now Exchange Act rule 3b-3, describes a short sale as * [Alny sale of a security which the seller does not own or any sale which is consummated by the delivery of a security borrowed by, or for the account of, the seller. But the heart of the new provisions was embodied in paragraph (a) of the second rule, the important part of which is: No person shall *** effect a short sale of any security at or below the price at which the last sale thereof, regular way, was effected on such exchange. This meant that all short sales were prohibited unless made above the last sale price, usually by 1% point or more. In this aspect, the rule met three objectives which the Commission felt a short selling rule should accomplish: (1) Allow relatively unrestricted short sales in an advancing market; (2) Prevent short selling at successively lower prices-thus, eliminate the use of the short sale by the "bear raider" to drive the market down; (3) Prevent short sellers from accelerating a declining market by exhausting all remaining bids at one price level, causing successively lower prices to be established by longer sellers. 61 8.E.C., 1st Annual Rept., p. 16 (1935). Paragraph (b) of the second rule contained provisions which required all sales to be marked either "long" or "short" 522 while paragraph (c) described the conditions for marking the sale "long." Paragraph (d) provided exceptions, including any sale of an odd lot and certain offsetting transactions by the odd-lot dealer. The third rule related to the conditions under which a broker could borrow securities for a "long" account. (3) The adoption of the present rule; 1939 In the months after adoption of the Commission's rules, the total short interest on the New York Stock Exchange declined more than 50 percent, and Exchange officials suggested, and had extensive discussions with the Commision about, modification of the rules. In March 1939 the Commission promulgated what is now the main portion of rule 10a-1(a). Whereas the 1938 rule had prohibited all short selling at the last long sale price, the new rule permitted it at such price if that was higher than the last different price. The rule's operation is illustrated by the following example. (The asterisks denote short sales which would not have been permitted under the 1938 rule.) 1 A sale on a'minus tick" is made at a lower price than that of the preceding transaction; on a “zero-minus tick," at the same price as the previous sale, the last change in price having been downward. Sales on "plus ticks" and "zero-plus ticks" are simply the respective opposites. On several other occasions the New York Stock Exchange has urged the Commission to modify the existing rules, usually suggesting that short sales should be permitted without restriction at any price above the security's closing price on the preceding day. The Commission, however, has indicated that it did not believe such modification to be in the public interest. In 1955 the Commission reiterated this position to the Senate Banking and Currency Committee, which was investigating factors affecting the stock market.523 At various times during this period and in later years, the Commission has continued to support the need for a strong short selling rule. e. The data regularly available on short selling The only data regularly compiled and published concerning short sales are daily aggregate figures for all stocks on the New York and American Stock Exchanges and monthly figures on the short positions in certain stocks on the NYSE and in all securities on the Amex. Analysis of such data permits only broad conclusions about short selling practices. The series of reports are here described, and their limited usefulness is indicated. 522 Such a requirement had been in effect for some years at least on the New York Stock Exchange, but its purpose and use were solely statistical. 59 Hearings before the Senate Committee on Banking and Currency, 84th Cong., 1st sess., p. 949 (1955). (1) Records filed on a continuing basis with the Commission Three types of reports containing short selling data are filed on a continuing basis with the Commission. One is a weekly report of daily round-lot transactions on the New York and American Stock Exchanges, in which daily aggregate short selling in all stocks is reported as well as aggregate short selling effected by members for their own account, classified into the three categories of specialists, floor traders, and members off the floor. Secondly, the Commission receives weekly reports of aggregate daily short sales by odd-lot customers. For the NYSE, these data are reported by the odd-lot dealer firms, and for the American Stock Exchange, by the Exchange itself. The data are similar to round-lot short sales in that only aggregate short selling in all stocks is reported. A third category of reports, containing information on short positions, are released by the NYSE each month to the press and are filed with the Commission. These reports provide an aggregate midmonth short interest figure for all stocks, the number of issues in which short interest was reported, and the actual short interest for identified individual issues in which there is a short position of 5,000 shares or more, or in which there was a change in short position of 2,000 shares or more from the previous month.524 They also report as a separate figure the total short interest of the odd-lot dealer firms. The value of these reports is reduced because they cover only the short positions as of monthly points, rather than reporting as of more frequent intervals or, also, giving total short sales volume in at least the most active "short" issues. Their value is further reduced because they fail to classify the short interest into categories of traders as do the roundlot short sales reports, or to classify the short interest by such types of short selling as sales "against the box" or sales for speculation. While these three types of reports constitute the primary sources of information regularly available to the Commission, two additional reports concerning short sales are regularly filed: daily reports on the floor traders' transactions and month-end stock positions reported by the two largest odd-lot dealer firms on the New York Stock Exchange, Carlisle & Jacquelin and DeCoppet & Doremus. The daily reports of floor traders' transactions represent the most complete record available for any single group, as such members are required to report each transaction effected on the floor of the Exchange, giving the time, tick, price, and number of shares. These reports are quite comprehensive but apart from their limitation to one class of traders their use for a study of short selling is limited because floor traders are not required to report on this form trades initiated off the floor of the NYSE or those effected on other exchanges. The reports of odd-lot dealers' month-end positions in each stock cover periods differing from the total short interest report by approximately 2 weeks, and are limited, like the short interest reports, in what they reveal. Thus, the drawback common to most of the short selling information on file with the Commission is that it does not provide, with respect to either round lots or odd lots, the total volume of short selling occurring in single issues over continuous periods of time. 24 The Amex also makes midmonth short interest reports covering all securities traded on that exchange. (2) Records on file with the exchanges Reports of short-selling data available at the New York and American Exchanges though not regularly filed with the Commission are those of the individual members and firms, which are used to compile the summaries of trading data regularly submitted to the Commission. On file with the exchanges are the individual reports of all clearing members' daily total sales and short sales, and all members' daily total transactions for their own account. These are filed at the end of the week following that in which the transactions occurred. Summaries of these reports provide the data for the weekly round-lot reports to the Commission on aggregate reported short sales for each trading day. Also on file with the exchanges are tabulations of members' midmonth short positions in each stock. These tabulations, as compared with the short-interest report submitted to the Commission, have two useful features: they contain the total short interest in each stock whereas the published short interest reports do not, and they permit a stock-by-stock analysis of the large positions or monthly changes in any reporting member's short interest. The data available in exchange records afford a means of further classification of short sellers, but these records still do not provide the most basic material necessary for an appraisal of short selling-a record of total short sales effected in any particular issue either classified by type of seller or, ideally, in terms of each short sale transaction. 2. GENERAL FEATURES OF SHORT SELLING Assuming that the technical uses of short selling are responsive to operating needs of the market and that short selling in its various forms may contribute liquidity to an exchange market under ordinary conditions, the important question is the extent to which the current rules operate as originally intended to restrict short selling in sharply declining markets. For this purpose, the Special Study examined short selling in a group of selected stocks during a period culminating in the market break at the end of May 1962. As a background, an examination was made of the relative volume of short sales by the various classes of participants in the securities markets and the movements of stock prices.525 The results, presented in chart form, provide some information about the timing of short selling by the different types of sellers during the major swings of the market. a. Total volume Chart VI-d illustrates the pattern of short selling on a monthly basis from January 1954 through July 1962. The chart shows that short sales in the aggregate vary considerably in volume, ranging for the period shown from less than 2 million shares per month to more than 11 million shares during the break in May 1962. In relation to total trading, short selling during this 812-year period varied from 3 percent to over 8 percent. 525 All references to market levels in this part of the report are to Dow Jones Industrial Average (DJI). |