Lapas attēli
PDF
ePub

if deliveries are not made, with the expectation that such Fails will tend to balance out resulting in a need for less frequent deliveries. The system, therefore, accepts the existence of Fails, within controlled limits, for the sake of decreasing delivery requirements, and decreasing the impact which original Fails can have in producing secondary Fails. In this manner, it seeks to eliminate the "domino" effect.

As a result of past work, ADL has recommended that the NASD adopt a continuous netting approach. We consider this procedure highly desirable for a situation where the geographical and size distributions of participants vary widely, where supplies of liquid stock in the securities traded are limited, and where a significant fraction of securities handled is not fungible and thus limits the effectiveness of bookkeeping entry systems. Under these conditions we consider continuous netting as the most desirable technique to employ.

In the case of the NYSE, we do not recommend the adoption of continuous netting because:

(1) We consider the establishment of an effective depository as a superior approach for the NYSE, with its centralized operations, more equal sizes of Clearing Members, large liquid supplies of securities, and a high degree of fungibility in its listed stocks. We believe the stock lending and Fails control procedures we are recommending will be adequate for handling Fails, and we believe participation in these activities will be greater and more effective if the NYSE retains traditional procedures than if it were to adopt a new set of relationships.

(2) Accepting the contra-side position for each trade, as opposed to merely acting as a bookkeeping and monitoring agency, entails a considerable increase in NYSE liability. While the NYSE's liability will increase in any case, since its data resources will be improved and thereby its ability to act on its responsibilities for regulating its membership, this is a different and less direct form of liability.

(3) If the NYSE were on the contra-side of each trade, it would become more difficult for it to equitably police its membership, because it would be an active participant in each operation as well as a "judge" and arbitrator. We believe equitable policing can be done more effectively under present procedures. It should be noted, however, that if effective stock lending and Fails controls are not adopted, continuous netting may become a necessity.

Since we have not recommended continuous netting for the NYSE (although we believe it a necessity for some of the other markets) we have considered the implications of more than one clearing system, which implies a limit on the degree of standardization possible within the industry. As a result we note that complete standardization of input data is not feasible in any case; the NYSE, for example, must make provisions to accept data on $2 brokers, while the OTC market will be entering numbers associated with NASDAQ, etc. Settlement and position reports might be standardized if a single system were adopted. However, we believe the automated participantactivity auditing that we are suggesting is more important in reducing clerical checking, and should reduce clerical loads to a level where the existence of multiple systems is not unduly burdensome. In line with our basic philosophy, we consider the disadvantage of having more than one clearing system more than balanced by the opportunities offered the different markets to develop systems to meet their specific needs, and to express initiative in trying new procedures that can work to the long-term benefit of all markets.

V. SECURITIES MOVEMENT OPERATIONS

In the previous section we described the Broker Module, which employs a bookkeeping- entry system and associated services for controlling the level of broker-to-broker Fails, while simultaneously reducing broker clerical support requirements. The Broker Module, by itself, cannot handle effectively the movement of securities that involves transfer and customer delivery. A requirement of the Securities Handling System is to provide for rapid and efficient conduct of these functions. Here, however, the needs of the end customers the purchasers and sellers of securities - will set constraints on possible procedures. We must start, therefore, by examining the characteristics of these customers.

A. THE MARKET FOR DIFFERENT FORMS OF SECURITIES
OWNERSHIP RECORDS

Table 1 shows certain characteristics for three categories of owners of stock during 1968, and as projected for 1975. The future estimates are based on direct extrapolations of recent trends. Although we consider these estimates to be reasonable for testing the general feasibility of a system design, more study would be required to develop estimates appropriate for detailed marketing or cost planning.

The "Institutions" category in the Table is based on the usual NYSE definition of Institutions and Intermediaries and does not include personal trusts. The "Traders" and "Investors" result from splitting the normal NYSE category of Individual Owners to reflect the fact that a large fraction of all active trading is provided by a small percent of all share owners. Solid data on this subject are not available, so the Table was developed from the assumption that 20% of all share owners (the Traders) are responsible for 85% of all trading.* The implications of this assumption in terms of "activity rates" and "years held" could then be calculated, leading to the values shown in the Table.

*The split is an arbitrary one. However, so long as a relatively large fraction of all trading is accounted for by a small fraction of traders (presumably including the NYSE Members among them) the market segments will have the general characteristics described herein.

[blocks in formation]
[blocks in formation]

Some critical design features for a securities handling system can be derived by examining the needs of the market segments shown in Table 1:

Investors: A large fraction (well over half) of all securities
is owned by individuals whose activity rates are extremely
small. For such persons, a normal certificate is a fairly
cost-effective instrument of ownership. In general, brokers
will not want to hold these securities in their possession,
since their costs for storage and record-keeping will tend to
be greater than the benefits derived from the infrequent
trading commissions. Banks may offer custodian services,
however, especially to individuals with relatively large
holdings.

Traders: The most active component of the customer pop-
ulation is that of individual Traders. Ideally, the system
design should encourage Trader securities to be left with
brokers, where the rapid turnover can be handled effi-
ciently within the Broker Module.

Institutions: The activity of institutional customers is
growing rapidly. These customers will almost always em-
ploy banks as custodians. The current relatively rigid pro-
cedures required to make deliveries for these customers are
a major problem. This component of the market could be
served best, from an overall point of view, through devel-
opment of a means whereby security ownership records
can be transferred nationwide by a wire system, with
central control and record-keeping. The fact that a rela-
tively small number of banks are used by institutions for
the large majority of such securities storage makes this
approach both technically feasible and potentially desir-
able on an economic basis.

The customer market, therefore, appears to be stratified, with a number of transfer, delivery and custodian arrangements desirable to meet the needs of the end customers in a socially desirable fashion. The primary change required is the development of a centrally controlled wire system for transferring ownership records of some components of the institutional market (and some of the larger personal holdings).

B. THE WIRE SYSTEM DECISION

The subject of moving securities by wire has received repeated attention over many years. The legal and institutional barriers to a system are enormous and will take many years to overcome. It seems unlikely that the institutional market could be converted entirely in less than several decades. However, some components (such as bank managed portfolios, or mutual funds, which are subject to fewer and more uniform legal requirements than other institutions) would be far easier to convert than others.

In a separate technical appendix, we present a preliminary design for a securities movement system. We believe this preliminary design furnishes a reasonable basis for starting serious consideration of the problems that will be encountered. The intent of the design is to provide a stimulus for potential participants to begin the open discussions necessary to start action on a serious development program.

« iepriekšējāTurpināt »