TABLE VI-V.-Trading data for 8 selected stocks combined (May 28, 29, and 31, 1 Short sales by nonmembers were estimated on the basis of reports by 25 leading New York Stock Exchange member firms accounting for the highest volume of commission business in 1961. NOTE. Figures do not necessarily add to totals due to rounding. Moreover, in spite of the Commission rule limiting short selling to up-ticks,538 and without evasion of that rule, in six of the eight stocks most or much of the short selling occurred at times when these stocks were under the greatest pressure. United States Steel Corp. This stock is the only one of the eight studied that had substantial short selling by the stock's specialists on May 28. Of a total of 14,600 shares that were sold short, the two competing specialist units in this stock accounted for 12,700, or 87 percent. The remaining 1,900 shares were accounted for by nonmembers. Each specialist unit had an opening position on May 28 of 1,100 shares long. During the first 2 hours the stock traded in a range from its opening price of 52 down to 512. At 11:42 a.m., both units together sold 4,500 shares at 5134, of which 3,300 shares were short sales, apparently made to supply part of an accumulation of buy orders. As a result, by 1:00 p.m. one specialist unit was short 1,500 shares and the other was short 400 shares; the stock had not gone below 514. At the end of the next 2 hours, however, it had declined to 50%. During that period the specialists did not use their short position to absorb selling pressure, which is the usual reason offered by specialists for their building up a short position. Instead, the two units purchased 5,300 shares and sold 5,600, each unit accounting for about onehalf of the purchases and sales. Of the 5,600 shares sold, 4,500 were 538 "Up-ticks" include both "plus ticks" and "zero-plus ticks." short sales, which were made in small lots just as were their covering purchases. By 3 p. m., one unit was short 1,600 shares and the other, 600 shares. They did no more trading until the closing 7 minutes, when they began covering, the first unit purchasing 2,500 shares (and selling 300 shares in this time) and the second purchasing 800 shares. The purchases were at 504 and 50%, made just after the stock traded at the day's low of 50%. The specialist units ended the day with a combined long position of 800 shares, 1,400 shares less than they had at the opening. The stock proved to be one of the least affected by the avalanche of selling on May 28, but when the stock started its decline after 2 p.m., the specialists did not use their short position to absorb the selling pressure. E. J. Korvette, Inc. Korvette appears to offer a relatively clear-cut example of stock in which short selling may have been a factor in the decline on May 28. All but 100 shares were sold by nonmembers and most of this was concentrated in late trading when the stock suffered its worst decline. The stock opened at 40%, unchanged from the previous day's close; rallied to 41% toward noon; remained at or above the opening until after 1:00 p.m. and then fell on increasing volume to a low of 34 between 3:00 p.m. and 3:25 p.m. It closed at 371⁄2 on a sharp recovery of 312 points in the final few minutes of trading. Based on the substantial sample of nonmembers' transactions, most of their short selling occurred at prices which placed such selling during the rapid decline from around 40 to 34. American Telephone & Telegraph Co. Nonmembers were the principal short sellers of AT&T. While this selling took place steadily during the day as the market declined from about 109 to almost 100, a large part of it occurred as the market neared its low in the latter part of the day. Augmenting this short selling, floor traders sold 3,000 shares. About one-third of this was done between 1 and 2 p.m., during the initial decline of the stock, and most of the rest was concentrated during the final half hour of trading. On balance, floor traders increased their long position slightly during the day. Members off the floor (though ending the day with a net purchase balance) sold 1,500 shares short, most of it as the market began its downward slide. Avco Corp. All the short selling in Avco was done by members off the floor. Although they made fairly substantial net purchases for the day, they sold 3,100 shares short, the bulk of it toward the end of the first hour and prior to the inception of the decline. Standard Oil Co. (N.J.) Short selling was modest relative to total sales in Standard Oil but most of it occurred during the weakest part of the decline near the close. General Motors Corp. Short selling represented 3.8 percent of total selling in the stock, most of it by nonmembers. About one-third of this selling occurred during the final half hour of trading, near the lowest price of the day. Brunswick Corp. Short selling represented 3.2 percent of total selling in the stock, and all of it was by nonmembers. Most of this selling occurred soon after the opening, near the high of the day. International Business Machines Corp. Nonmembers and floor traders accounted for about equal amounts of all the short selling in IBM. The nonmembers' selling was distributed throughout the day. On the other hand, floor traders concentrated more than a third of their short sales between 1 and 2 p.m., when the market was falling rapidly. In addition, they were heavy sellers on balance during the day. 6. AN EVALUATION OF THE SHORT SELLING RULES The preceding analysis highlights two important facts: First, the volume of short selling tends to expand during declining phases of the market and thereby, contrary to the classic argument, augments already existing downward movements. Second, the current rules have been unable to prevent even the concentration of short selling in times of critical market decline, or the concentration of substantial short selling in individual stocks, frequently at moments of great selling pressure in those stocks. Not only did the current rules fail to eliminate the aggravating influence of short sales during the market break of May 1962, but as the table below indicates, a much larger amount of short selling might have been effected than actually took place. TABLE VI-W.-Maximum opportunities to sell short in 8 stocks Despite the weak condition of the market on May 28, there were over 1,500 up-ticks on which short sales could have been effected in the eight stocks selected for study. During the market break particularly, the supply of stock which short selling introduces adds further selling pressure to an already unbalanced market. One specialist commented as follows: Q. During the break in May did you feel any pressure from short selling at all? A. There was a great deal of short selling; there can't be too much pressure because we can only sell on plus ticks. Q. Did you A. They certainly lengthen the time that it took a stock to go up, probably. Q. In other words, when you had an up-tick A. There had to be substantially more buyers to move the stock up because of the heaviness of the sell orders-the short orders, excuse me. Q. So that when there was an up-tick, there would be the short sellingA. Right. Q. That would take place? A. Right. Not in every stock, just certain ones. Another specialist testified along substantially the same lines: Q. What were the circumstances in General Foods on the morning of May 28? A. On May 28 there were in the marketplace, not only in General Foods but in many other stocks, many orders to sell short. * * We had a downtrend of the market with news and you might say at that time that bids were being satisfied for over a period of time in the marketplace on top of which were these orders to sell short. Q. You are talking about May 28? But on May 28 there were bids and, of course, the specialists have had to temper his [sic] markets with his bids on the way down. It was a most difficult job because no matter what he did he had short orders coming in right on top of him. Q. The short selling rule would prevent A. Artificial depressing of the stock but nevertheless they came down with the market. An important reason for the continued downward pressure that short selling may exert even during severely shrinking markets is the weakness of the tool designed to prevent this pressure. The tool takes the form of the tick test which establishes a "trade-to-trade" basis for determining the permissibility of short selling, but that does not serve to prevent short selling in a declining market.539 Ample illustrations have been brought to light of the fact that plus or zero-plus ticks may be commonplace during sharply declining markets. Thus, when measured against the preceding different price, short selling may appear to have no important influence on the movement of the market; measured against a basic trend, however, which could have been in motion for some time, short selling may exert a heavily deleterious effect on the movement of the market. 7. SHORT SELLING EXEMPTIONS Since 1939 the only change made with respect to the short selling rules has been the adoption of certain exemptions. Of these, the most important probably are two arbitrage rules; the first refers to arbitrage transactions between equivalent securities and the second to international arbitrage. Another exemption was provided for so-called equalizing transactions, i.e., those effected on an exchange for the purpose of equalizing the price of a security traded on that exchange with the current price on another exchange where the security has its principal market. All of the exempt transactions are exempt only from paragraph (a) of rule 10a-1, that is the provisions which determine when a short sale may be made, and not from paragraph (b), which requires all sales to be marked either "long" or "short." The Commission did not specifically prescribe special markings for exempt short sales. However, after meetings with the staff of the Commission, the Exchange was advised that marking such orders "short exempt" would satisfy the marking requirements of rule 10a-1. In the statistics that are regularly compiled and published, these "short-exempt" transactions For the correspondence between ticks and price trends, see pt. D.6.e(2) of this chapter. are not counted as short sales but are included with regular sales. Accordingly, the only record available on the volume of "short exempts" is located in the accounts kept by the individual member firms. To throw some light on the extent of such sales, a limited survey was made of certain member firms by the Special Study. Although the survey was not broad enough to yield useful data on the volume of such transactions, since it focused upon firms known or believed to be engaged in arbitrage, some information was gained in regard to their practices. The investigation covered arbitrage transactions at nine member firms, in American Telephone & Telegraph Co., Avco Corp., and Brunswick Corp., principally for the 2 days, May 28 and 29, 1962, but in some cases for other periods: a. International arbitrage As one firm pointed out, the amount of international arbitrage is now much less significant than in the past and this very specialized operation is kept alive only among a few firms. Perhaps for this reason, the investigation found so few transactions effected during the periods covered that only a small amount of data was produced. In regard to short selling by the foreign participants in arbitrage accounts, the same firm stated that in executing trades for their foreign customers, their Amsterdam office is instructed to inquire and make note if a sale is "long" or "short." The firm indicated, however, that while this is the rule, it has been left largely to the client (usually large and reputable) to designate the trade. Another firm's comment adds still further doubt concerning the degree of compliance with the short selling rule and the efficacy of available records or other means of which compliance can be checked. The firm was quoted as saying: Short exempt sales must be reported to the NYSE only if they originate on the Floor. The comment may be ambiguous but it does emphasize the need for more complete records on short exempt sales in international arbitrage as well as on other sales based on the technical exemptions under the rule. b. Arbitrage based on convertible issues The majority of firms visited appeared to be conducting arbitrage in convertible issues properly under the rule, by conversion of the security to cover share positions acquired through short exempt sales. However, in the case of at least one firm there appears to be some question of full compliance. For example, during the period from May 8 through June 29, 1962, the firm sold at least 23,350 shares of American Telephone & Telegraph Co. "short exempt" without converting bonds, apparently covering their short position in the stock by purchases of the stock instead. A similar situation appeared in the firm's arbitrage account in the securities of Avco Corp. These investigations raise a question of possible apathy toward, or misuse of, the exemption provisions of the rule. 8. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS Short selling not only is used by exchange members and members of the public for speculative purposes, but is used also by arbitragers, specialists, and odd-lot dealers to facilitate market operations, and is |