flect the identity of the firms participating, the time of the transaction or whether the sales are long or short. To determine the names of participating firms clearinghouse data must be used; to determine the time of each transaction the Exchange resorts to the odd-lot firms' records of the tape time of each transaction.632 However, the fact that these records show tape time makes it difficult to match transactions on the tape with floor trading or specialist reports, which use the actual time of the transaction. Integrating all the data merely to determine buying and selling firms and time of each transaction is a difficult process because of the disparate sources used.633 The more active the market the more difficult the task of market reconstruction becomes; the task of assigning times from one source to transactions from a second source and names garnered from a third source makes the whole process uncertain. And, as already noted, even the successful conclusion of this process yields only a sequence of transactions without providing such elementary data as the amount of short selling; at this point the firms involved must be contacted for further information. The NYSE's projected system of automation will not materially improve this situation. These plans will still rely on the reporter's notation of a transaction, although the form of the notation will be different. Little more information will be captured than at present and there is no reason to believe that there will be fewer omitted transactions.634 The identification of member firms, short sales and floor members' transactions will still depend on methods presently used. A reporting approach which captured information at the point of execution and from the members involved would permit the introduction of an integrated system of data processing. Under such a system, which is being increasingly used by business firms, an original document becomes the basis for accounting and planning operations. Development of such a program of integrated data processing would probably cut costs and add greatly to the reliability of the resulting reports. Through storing the record of each transaction in a computer, improvements could result not only in surveillance and other market studies, but also in the whole clearing process. One impediment to development in the area of automation is an apparent reluctance on the part of the NYSE to impose different reporting duties on floor members than they now assume. This reluctance is understandable, but with other changes in mechanisms the balance of convenience and inconvenience may be rapidly shifting. Moreover, a program which has as its goals the maximum utilization of available information and its reporting by those who have the greatest incentive to provide accurate data (the Exchange members) may be worth the added inconvenience, if any. In view of the important potential advantages for all concerned, the subject of automation clearly warrants examination now and periodic reexamination in the future. 632 See pt. E of this chapter for a description of the records maintained by the odd-lot firms. 633 Complicating the picture is the fact that an analysis of NYSE floor trader reports, which purport to include time, price, and amount, shows omissions of varying importance. The use of these reports for market reconstruction purposes adds to the uncertainty of the resulting picture. See ch. XII for a discussion of floor trading reports. 634 See the recommendation in pt. D of this chapter that the NYSE require that all transactions be reported on the tape. 4. OTHER EXCHANGE USES OF AUTOMATION Thus far, comment has been directed to the events which occur at the time of an order's execution and thereafter. However, it should be noted that the impact of automation is not restricted to the floor of the exchanges. It has been mentioned that some firms use electronic equipment to transmit an order to the floor and at the same time to commence back office work in connection with it, as well as for internal surveillance and supervision. In chapter VIII, it is noted that the Midwest Stock Exchange utilizes data processing techniques to provide comprehensive back office services for its members. By so doing, that Exchange has managed to cut costs for its members, both large and small, to a level which may be significantly below those of many large and efficient NYSE member firms. The NYSE Clearing Corp. is a further example of how cooperative effort can dramatically reduce costs and improve efficiency. However, its history may also serve as a lesson-it was only after several years and in the face of determined opposition to innovation by some members of the Exchange community that the venture was established. The creation by the NYSE in 1960 of a department of operational planning and development to study ways of improving the Exchange's own operations was a basic step in the direction of keeping Exchange mechanics abreast of technological developments. Until recently, the Commission paid little attention to the subject of automation, either as involving the exchanges and other markets or as it might concern its own operations and responsibilities. It would also seem that the Commission, as the statutory representative of the public interest, must give prompt consideration to immediately available uses and to keeping itself informed of progress in this area generally, so that possibilities for automation that may exist now or in the future are not lost sight of. 5. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS Technological innovations have been utilized increasingly in the exchange markets. Some member firms have used automated techniques to transmit orders from the point of origin to the floor of the NYSE as well as in their own back office operations and in supervision. The clearinghouse of the NYSE uses electronic equipment to great advantage and the Exchange's unique "stock watch" procedures rely on these devices for surveillance purposes. The Midwest Stock Exchange has pioneered in the development of centralized bookkeeping for its members. However, despite the NYSE's interest in automation, as signified by the establishment of its department of operational planning and development in 1960, progress with respect to exchanges' trading mechanisms has not been impressive. The possibilities of modern technology may go beyond mechanical arrangements and involve such functional matters as the execution of some orders from off the floor. This does not mean that possible changes of this kind are necessarily desirable but it suggests that they ought to be objectively explored and evaluated. The public, through the Commission, has an interest in such exploration. 96-746-63-pt. 2- -24 Other possibilities for the fruitful use of automated techniques may lie in the accumulation of reliable, comprehensive and current market data. At present the Exchange and member firms file many periodic reports on a variety of subjects, each of which is designed for a specific purpose. The processing of these reports does not make use in any susbtantial degree of modern data processing techniques. Also, since the reports cover different periods and are differently designed, the information they contain cannot be integrated to present a comprehensive picture of market trends or current transactions. As a particular problem, improvement is needed in the reliability and completeness of volume data. Various studies have shown that such data are unreliable and incomplete under the present reporting systems. Furthermore, much information is lost at the time orders are executed which would be valuable for market surveillance and other purposes. The automated reporting procedures recently announced by the NYSE will not result in more accurate reporting or the preservation of more data than under the present system. The Special Study concludes and recommends: 1. Automation, in the form of electronic communicating and data processing devices, seems certain to have an increasingly important impact on exchanges as trading markets and as selfregulatory agencies in various ways. For example, in connection with the handling of odd lots, reference is made to the discussion and recommendations of part E of this chapter; and in connection with surveillance of specialists' activities, reference is made to the discussion and recommendations of part D of this chapter. The potential impact of automation in these and other respects is affected with a public interest and the Commission has a present and continuing responsibility to be informed of developments and potential developments in this area. 2. The NYSE should promptly undertake to revise its floor reporting procedures so that volume data will be complete and accurate. In this connection the Exchange's recently announced automation procedures should be restudied with the view to obtaining and preserving more market data at the time orders are executed than is presently the case. 3. The Commission and each of the major exchanges should jointly undertake studies of the periodic reports filed by member firms with the respective exchanges and by the latter with the Commission, with a view to simplifying and coordinating the reports and maximizing their usefulness. Also, they should jointly consider possibilities for developing and coordinating their automation programs in such a manner as to fulfill the needs of each with maximum effectiveness and minimum burdens, duplication, and expense. CUMULATIVE PERCENT OF EACH MEMBER CLASS STOCK DAYS FALLING AT OR BELOW EACH PRICE⚫ Chart VI-1 NEW YORK STOCK EXCHANGE MEMBER TRADING BY PRICE: DISTRIBUTION BY PRICE OF STOCK DAYS IN WHICH MEMBERS PARTICIPATED During the Weeks Ended January 27, March 24, and June 16, 1961 80 40.00 PRICE DISTRIBUTION OF ALL STOCK DAYS. NOTE: The price used is the closing price of each stock day. • Stock days are based on common stocks only. During the three weeks there was a total of 16.174 stock days. One or more #Specialist data excludes 644 stock days of competing specialists. This exclusion does not materially affect the specialist DS. 4464 Chart VI-2 NEW YORK STOCK EXCHANGE MEMBER TRADING BY PRICE RANGE: DISTRIBUTION BY PRICE RANGE OF STOCK DAYS IN WHICH MEMBERS PARTICIPATED During the Weeks Ended January 27, March 24, and June 16, 1961 CUMULATIVE PERCENT OF ALL STOCK DAYS FALLING AT OR BELOW EACH PRICE RANGE 100 0 8 CUMULATIVE PERCENT OF EACH MEMBER CLASS STOCK DAYS FALLING AT OR BELOW EACH PRICE RANGE 200 PRICE RANGE DISTRIBUTION OF ALL STOCK DAYS. NOTE: Price range is expressed as a percent of the closing price on each stock day. Stock days are based on common stocks only During the three weeks there was a total of 16,174 stock days. One or more Specialists 12,825 #On this chart Odd-Lot Dealers and Specialists have almost identical distribution curves, and therefore only one line is DS-4465 |