TABLE VI-q.-54 stocks consistently in NYSE short interest reports classified by industry (January 1961-June 1962)

NOTE.-Industry groups are based upon U.S. Bureau of the Budget, Office of Statistical Standards, "Standard Industrial Classification Manual."

Table VI-r identifies the 54 stocks by name. The list scarcely needs any additional data to indicate that the most popular stocks for short selling purposes are the so-called "market leaders" or current "trading favorites." 535 Among the leaders are such stocks as American Telephone & Telegraph, Bethlehem Steel, Chrysler, General Electric, and U.S. Steel, while some of the trading favorites are American Motors, Avco, Bell & Howell, Brunswick, Litton Industries, Polaroid, Texas Instruments, and Varian Associates.

TABLE VI-r.—54 stocks consistently in NYSE short interest reports (January 1961-June 1962)

The volume of trading for the 54 stocks during 2 months of the period (December 1961 and June 1962) accounted for 15.5 and 22.6 percent, respectively, of total trading on the NYSE in those 2 months indicating they were among the most actively traded.

4. EFFECT OF SHORT SELLING ON PRICES

An analysis was made of the data provided by the monthly short interest reports to ascertain, if possible, the relationship between short selling, short covering, and stock prices during different types of market movements. As indicated below, the analysis was not productive. For 9 different groups of 30 securities each, data including the following were examined: the short interest, price range, and volume of trading during the period studied, and the industry classification of the issue. The groups included stocks with the largest short interest as of a market low, stocks with the largest decrease in short interest during a period of market advance, and others with the largest increase during a decline; stocks with the largest percentage price rise in a year of rising prices, and those with the largest percentage price change in a period of decline. The changes in short position of each of the 270 stocks were then compared with the stock's price changes in the period for which it was studied.

It was not possible from the data available for these tests to find any relationships, nor did the study of individual price movements provide any corroborative evidence. For example, the rise in the market after the 1960 lows amounted to some 177 points in the DowJones Industrial Average. It might have been assumed, therefore, that the 30 stocks with the largest short interest would not only advance during the subsequent general rise because of short covering, but would perhaps advance even more vigorously than the average. However, the 30 stocks not only failed to advance at a more rapid rate than the general market, but 20 of the group actually declined in price.

The several tests indicate the difficulty of employing the monthly short interest figures to attempt to gage the influence of short selling or covering on the price movements of stock.

5. A CLOSER LOOK AT SHORT SELLING

Accordingly, it is important to examine short selling with respect to the individual stocks where such selling is large, and, as much as possible, to pinpoint this selling to specific instances of time. For this purpose, reference was made primarily to data concerning eight selected stocks obtained from the Special Study's analysis of the May 1962 market break. These data include detailed information on short selling for the 3 days of the market break period, May 28, 29, and 31, and summary statistics for 14 selected days prior and 2 days subsequent to the market break.536

This discussion is divided into three parts. The first provides some highlights on short selling in the eight stocks during the selected 14 days prior to the market break, the second presents an overall view of the market break period, while the third furnishes a detailed analysis of short selling in the eight stocks during May 28, 29, and 31. a. The period prior to the market break

Most of the eight stocks suffered a generally falling trend during the period covered by the 14 selected days and all of them experienced

630 See ch. XIII for a fuller analysis of trading on these days. Short sales by nonmembers were estimated on the basis of reports by 25 leading NYŠE member firms accounting for the highest volume of commission business in 1961.

declines during the latter part of the period. Short selling rose to unusual heights, as indicated in table VI-s, which shows the days on which estimated short selling was 8 percent or more of total sales for each of the stocks. On various days short selling actually constituted over 25 percent of total sales of two of the stocks. As a point of comparison, it may be noted again that from January 1954 through July 1962, short selling in relation to total trading in all stocks varied from 3 percent to over 8 percent.

TABLE VI-S.-Days on which short sales amounted to 8 percent or more of total sales in 8 selected stocks1 (14 selected days, 1961–62)

1 Short sales by members were reported for each of the 8 stocks. Short sales by nonmembers were estimated on the basis of reports by 25 leading NYSE member firms accounting for the highest volume of commission business in 1961.

In accordance with the general pattern, the predominant_amount of short selling on the 14 days was done by the members. In some instances, this selling came from the specialists; one important specialist testified that he tried during this period to end each day with a short position in three of his stocks.587 In other instances, the short selling came from floor traders and members off the floor, probably prompted by speculative motives. For example, the large amount of short selling of U.S. Steel on April 27 and 30 may be ascribed, at least in part, to the price controversy between the Government and the steel companies that occurred during this period and to the announcement on April 27 of a Federal grand jury indictment against several companies, including U.S. Steel, on charges of price fixing.

The consistently large amount of short selling in Korvette both from members and from the public very likely was inspired by speculative motivations. During this entire period the stock was in the public eye and its sharp price changes probably influenced the substantial short selling activity that occurred. Such short sales, carried into the period of the market break itself, undoubtedly contributed to the severe weakness of the stock that developed around the spring of 1962. Avco presents an interesting case of an entirely different kind. Over half of the unusually high volume of short selling on May 11 and 14 represented hedged transactions against the company's convertible bonds by two firms which do a great deal of arbitraging.

537 See pt, D.6.e (4) of this chapter.

Thus, during the period preceding the market break, short selling was created by varying motivations. All three of the stocks just mentioned were very much in the public eye during this period and therefore the burden of absorbing a large volume of short selling on certain critical days may have contributed more to the general weakness of the period than is implied even by the amount of short selling in the particular stocks.

b. An overall view of the market break period

Before presenting the data in regard to the eight stocks during the market break period, it may be helpful to show the extent of short selling in general on those three days. These data are summarized in table VI-t. The table shows that short sales increased from 368,000 shares (or 3.7 percent of total sales) on May 28, to 774,000 shares (5 percent) on May 29, and to 1,417,000 shares (12.9 percent) on May 31.

TABLE VI-t.-Summary of markets and short selling by members and nonmembers (May 28, 29, and 31, 1962)

1 The volume of sales shown here represents total volume as contrasted with reported volume used in other tables in this section. Reported volume is obtained from the ticker tape and is generally 3 to 5 percent less than total volume.

Members accounted for 58.2 percent of all short selling on the first day, 79 percent on the second, and 84.8 percent on the third day. The major part of such selling was by specialists, who increased their proportion from 50.3 percent on the day of decline to 68.5 percent and 72.9 percent respectively on the following 2 days of rapid recovery in the market. Floor traders accounted for about 2 to 3 percent on each of the first 2 days but increased their proportion on May 31 to 4.4 percent. Members off the floor, whose ratio of short sales to total sales

rose from 5.2 percent to 8.6 percent, accounted for less than their usual average. On the second day of recovery, May 31, about a third of all selling by members on the floor, i.e., specialists and floor traders, was for short accounts.

Nonmembers increased the volume of their short selling from 154,000 shares on May 28 to 215,000 shares on May 31. However, because of the much greater rise in short selling by members during these days, the nonmembers effected their greatest proportion of total short sales on May 28. On that day they did 41.8 percent of total short selling, compared with 15.2 percent on May 31. On the other hand, the nonmembers' short selling relative to their own total sales rose from 2 percent on May 28 to 2.8 percent on May 31.

c. Short selling in the eight stocks during the break (May 28)

Tables VI-82 through VI-89 (which appear at the end of the chapter) contain comparative data on total selling and short selling by class of seller for each of the eight stocks on each of the 3 days during the market break period. Table VI-v summarizes this information for all eight stocks. The relatively large amount of short selling in these stocks on May 28 is indicated by the fact that they accounted for 15.5 percent of total short selling, whereas their total volume of trading represented less than 10 percent of all trading. Also, on that day of drastic decline, short sales in the eight stocks amounted to about 6 percent of total sales in these stocks; the ratio increased substantially on the following 2 days of rapid price recovery. By the third day, short sales in the eight stocks exceeded those on the 28th by some 171,000 shares, and about 65 percent of this increase was by members, principally specialists.

It is clear that in these stocks the market had to absorb a large amount of short selling, which reached unusually high proportions in the cases of U.S. Steel and Korvette on the day of sharp decline, as shown in table VI-u below: