This leads to the fourth characteristic which is major unpredictable swings in volume. In some ways the first three characteristics represent the tinder and this last characteristic represents the match that created the conflagration we have faced for the last several years. If it were possible to reasonably predict the amount of activity that our securites processing systems had to handle our problems would be much less severe-although they would still be present. With proper management it would be possible to predict labor needs and take the necessary steps to recruit and train competent personnel in a timely fashion. Even the danger of demand outstripping the labor supply could be overcome if sufficient leadtime were available to design better salary programs and other inducements.

However, the realities of this business are that volume is unpredictable and when the volume increases rapidly it can quickly outpace the industries' ability to meet its labor needs. In 1968 this resulted in pirating of qualified personnel between firms and, when this did not work, frantic efforts to eliminate some of the labor content through automation.

Conversely when volume drops very rapidly there is a lag in reducing staff because managers don't know how long the drop in volume is going to last or how deep it is going to be and they are reluctant to give up qualified people after they have gone through such difficulty to find and develop them. This lag has a direct and significant impact on profits because all of a sudden these variable costs end up being fixed. The problem has been further complicated during recent years by the investment made in software programs and computer hardware which became fixed costs. Finally when a substantial volume increase is followed by a substantial volume decrease many firms are unable to reduce their labor costs because during a rapid increase certain elements of control tend to be abandoned in favor of "getting the work out the door" and this results in a back-log of investigations. When the volume slackens there is time to handle these investigations but it means that the labor force must be retained until the back-log is cleaned up or a major write-off must occur.

Given this definition of the problem we conclude that there is an overwhelming need to significantly reduce the labor content of processing a security transaction.

The Ideal Solution

In the best of all worlds we would eliminate the securities certificate and all of the associated activities in settling a transaction which require labor input. This would mean that the certificate itself would no longer be part of the transaction. It also would mean that all communication of associated information, all proving, all checking, all verification and all record keeping would be completely automated. Finally, it would mean that there would be no exceptions, no errors, no investigations-nothing that would require manual intervention.

But, this is the ideal and can serve only as a goal. As we have learned in today's computer operations there are exceptions that must be handled manually and the propect for eliminating all such exceptions is nil. Likewise, errors and investigations will undoutedly continue to some extent because no computer system operating today and no foreseeable system can eliminate errors and resulting investigations. This should not deter us, however, in moving toward this ideal of a society in which certificates are not necessary.

Limited examples of such a system are already in existence and offer us the fruits of experience. The Mutual Fund industry with nine million share holders is on a dual system with book entry shares and certificate shares. The dividend re-investment program of First National City Bank is functioning with a dual system of book entry shares and certificate shares. Also the Federal Reserve operates a book entry system for government securities in which most banks participate. So systems and hardware are functioning today whereby shares are being held in book entry form.

But what kind of a system would be required for our ultimate goal of a society in which no certificates are necessary. First it would require a highly complex, automated communications structure. Every participant in the settlement of a transaction, whether he be broker, bank, agent or registrar would have to be tied into the communications network with the capacity to transmit and receive all the required information in a highly standardized manner without manual intervention. It would also require that no one outside of the communications network could be involved in the processing flow. An individual

security holder would have to work through an agent and accept the word of the agent that a transaction has been settled satisfactorily. This would require a change in the attitude of many people including a great number who 5 years ago were amenable to such a system. The credibility of the securities industry has deteriorated in the last several years and many who had left their stocks on account with brokers have now demanded certificates in their own name. Thus, we have to restore the confidence of the investor.

Our first steps for the most part will have to be something short of the utopian solution. We have to recognize the ultimate solution will probably evolve from a series of changes rather than from some quantum jump. For several reasons we believe this evolutionary process is highly desirable. First, the industry cannot afford to lose control of its processing as it did in the last several years. As witnessed by these hearings, customers and the public are unwilling to accept instability in the processing of transactions even though they can survive wide swings in the value of their holdings.

Second, the complexity of this business and the exposure to substantial loss through theft requires that responsible people ensure that any new processes do not increase this exposure. It is a known fact throughout industry that the implementation of change often creates a temporary deterioration in the quality of the product or service. This is commonly referred to as the "debugging phase" and unless properly planned and controlled, such deterioration could have extreme consequences including bankrupting a company.

The Evolutionary Process

Given this framework of conviction, we believe that the most constructive step that participants in this industry can take at this time is to support the establishment of securities depositories be they central, regional or transfer agent. Our main reasons for this belief are rather pragmatic. First the mechanics of a depository and much of the legislation required to establish a depository are already in being, have been tested and are under daily use in the Central Certificate Service. Second, there appears to be a great deal of agreement in the industry that depositories make sense-although there are disagreements as to control and financial structure. Given the usual diversity of opinion in this business and the lack of any strong central authority covering brokers, banks and customers such agreement must be considered of significant value in gaining momentum toward change. Third, the recent study completed by the Rand Corporation indicates that a full depository system would have a most significant impact in reducing the costs of failures to deliver. And forth, the additional legal changes needed to make a depository acceptable for the majority of transactions in this business involves only a few changes in trust and commercial law.

Although we are convinced that depositories are a logical first step in solving our overall problems, this should not be construed as general support for the existing depository under the New York Stock Exchange. Many problems such as financial responsibility of the parties involved, equitability of fee structures, control over rule making, and selection of key personnel have all been touched on by various elements of the industry.

In addition banks outside of the New York area have voiced their deep concern about their inability to participate in the existing depository except through New York correspondents. We believe all of these problems must be resolved. We see no reason why regional and transfer agent depositories cannot operate along side the Central Certificate Service and along side each other so long as each accepts the paper of the other.

In addition to the development and participation in depositories other steps can and should be taken at the same time. According to various estimates the cost of completing an individual transaction runs between $15 and $25 under today's procedures. The advent of a interconnected depository system would probably reduce these costs from $4 to $8 per transaction. This leaves somewhere between $7 and $21 of clerical costs that would still be incurred if other steps were not taken. Thus, we believe efforts should also be devoted to the standardization and we hope mechanization of the various documents associated with the securities transaction. As I have mentioned, the American Bankers Association has been active for many years in attempting to promulgate standards in this area. These include the standardized securities numbering systems referred to as CUSIP and the machine-readable standardized forms used for delivering

securities, reclaiming them and sending them to the transfer agent for reregistration.

It must be recognized that no matter how effective a depository system is and how soon it can be put into operation, it will be a long, long time before the documents needed to complete a transaction are eliminated. In fact, it is likely that for at least an interim period a depository may add to the number of forms used for settling transactions. As we see it, it will be much easier to continue to use hard copy to process transactions but to the greatest extent possible, this paper work should be standardized and put in machine-readable form. This could significantly reduce transaction labor content.

The third comment I wish to make about this evolutionary process concerns the machine-readable certificate. As I mentioned several times, if the industry is to retain control of their processing they must evolve toward the ideal solution. This will take time and to us, the machine-readable certificate represents a kind of insurance program for the industry. In effect, the value of a machine-readable certificate is directly related to the growth rate of volume over the next several years and inversely related to the success of depositories. For example, if depositories are effective in immobilizing only 50% of the certificates in circulation and volume over the next several years doubles, we will still have to process physically as many certificates as we do today and thus we will have the same problems.

We agree that there will be a natural tendency toward greater use of depositories because of the faster processing of transactions and because of less risk of loss of securities through insurance afforded by the SIPC but we still believe it will be a long while before certificates are immobilized. Even if brokers were to charge special fees for transactions handled outside of depositories there would still be those stockholders who would require their certificates be delivered directly to them. Legislation requiring that all securities be left on deposit would, we believe, be unacceptable to the public. Thus, we would hope the industry would look at machine-readable certificates as a back-up program for a full depository system and support the change if it appears that the depositories cannot absorb certificates faster than volume is creating them.

These steps we believe can be undertaken in the short range-that is, the next two to five years. As I mentioned earlier we view this as an evolutionary process and depositories built around existing systems along with standardization of accompanying documents would be the first step in this evolution. The virtual elimination of certificates we view as the long range goal. We see nothing inconsistent between what we are proposing for the short run and the long range goal. If you look at them carefully, you will find that depositories can be a springboard for the elimination of certificates. Thus, we urge support for these short range steps which can be initiated immediately.

Banking has devoted a great deal of time and energy to solving the paper work problem. It has developed plans and programs and has put them on the table. We have gone as far as we can go unilaterally. We stand ready to join with the exchanges and the securities industry in implementing the steps I have outlined or any other sound program.

Mr. Moss. Our next witness is Mr. Junius W. Peake, general partner, Shields & Co.

STATEMENT OF JUNIUS W. PEAKE, GENERAL PARTNER,

Mr. PEAKE. Thank you.

SHIELDS & CO.

Mr. Chairman and members of the subcommittee, I am a senior partner of Shields & Co., a member firm of the New York Stock Exchange, Inc., the Pacific Coast Exchange, the Midwest Stock Exchange, and NASD.

I have served personally as a member of the Association of Stock Exchange Firms Operations Committee, and the American Stock Exchange Operations Advisory Committee. I am a former member of the "Locked-In Trade" Committee of the New York Stock Exchange.

67-228-72-pt. 3—5

I was a member of the National Uniform Practice Committee of the NASD. I am presently on the District Committee of the NASD. I am presently on the Advisory Committee of the National Clearing Corp., and I am a governor-elect of the NASD.

I also am a member of the SIP task force which developed the report that was presented to the New York Stock Exchange in June of 1969, and I believe I have submitted that report as part of my direct testimony.

The securities industry has reached a critical time of decision: Does the existing business and regulatory structure have the resources, authority, motivation, and ability to solve rapidly and permanently what has been described in the past several years as the "operational crisis?” Without your assistance, which will make possible certain fundamental advances in specific areas, the answer is probably, "No."

The most rapid, most effective, and least costly solution would be the creation of an agency which I suggest be called the Securities Industry Service Corporation. This entity should have a Federal charter, be organized regionally like the Federal Reserve System, be owned by its users, and be operated as a profitmaking entity.

Its users would all be members or stockholders in the Corporation. Membership would be open to all commercial banks that belong to the Federal Reserve System, and would also be open to all broker-dealers who are registered with the Securities and Exchange Commission.

Foreign agency banks which have signed required agreements with the Federal Reserve would be permitted associate membership.

This new entity should be created using all existing resources within the industry. The cornerstone, or more correctly stated, the foundation of the proposed Service Corporation would, of course, be the Central Certificate Service, the present depsitory developed and run by the New York Stock Exchange.

When Central Certificate Service was conceived, it represented a farsighted, logical step for the industry; but it was created without the benefit of our present knowledge. Experience has demonstrated and the New York Stock Exchange recognizes the absolute necessity for inclusion of the commercial banking system in any future operations organization.

The Corporation should perform the following functions:

(1) Compare, clear, and settle all securities transactions for its broker-dealer members.

(2) Act as a truly nationwide depository for all securities of publicly owned corporations and for all exempt securities, except those securities which are physically in the possession of the individual public investor.

(3) Keep all records and perform all clerical functions necessary for the issuance and cancellation of stock certificates on behalf of the present transfer agents.

(4) Hold as a fiduciary those securities which are held in trust or under other similar conditions.

(5) Transmit money and securities over a nationwide wire system, which would permit the Corporation, as agent for commercial banks. to extend credit and hold collateral to finance members of the securities industry.

(6) Have both rulemaking and enforcement power to regulate the operational and financial aspects of its broker-dealer members, specifi

cally those aspects now governed by recordkeeping, financial responsibility, and net capital rules.

(7) Assume the present duties of the Securities Investor Protective. Corporation, act as trustee for its funds, and liquidate any of its members whose business must be terminated.

If the proposed Securities Industry Service Corporation is created with at least these elements, numerous current problems in the industry will be solved or made less troublesome.

For example, protection of the individual and institutional investor will be increased many fold, broker-dealers will be relieved of overlapping and sometimes conflicting regulation, segregation of customers' free credits and securities will be simplified or made totally unnecessary, regulation of operational and financial responsibility will be more efficient and effective by its concentration in one body, the cost of essential support services to the securities industry will be reduced by at least 50 percent, the cost paid by the public for participating in the securities market will be reduced, and the new entity will be sufficiently flexible to interface with any system of markets developed in the future.

The operational requirements of the securities industry are large enough, are sufficiently specialized, have a great enough effect on the public interest, and have created such pervasive national problems to justify the formation of this type of entity.

This corporation would not regulate the marketplace, which should evolve through its own efforts and which should continue its own. self-regulation.

THE GOVERNING BOARD

In the construction of such a new administrative entity, the composition of the governing board should be given great thought because numerous "interest groups" should be represented on it. The board should be 13 in number. The President of the United States should appoint a chairman and two vice chairmen who would be paid and who would serve full time. Four governors should be nominated by the members of the Corporation who are commercial banks; and four governors should be nominated by broker-dealer members. The Federal Reserve Board and the Securities and Exchange Commission should nominate one governor each. All nominations would be subject to Senate approval.

The Board of Governors should act by an affirmative vote of twothirds of the total Board. This would prevent one interest group from voting as a block and preventing the implementation of a proposal which the remainder of the Board deemed beneficial to the industry and to the public at large.

STRUCTURE

A national organization ought to provide a network of regional facilities which would give local representation and local service. The views and ideas from all sections of the country could be communicated easily and directly in any national entity with regional facilities. Therefore, regional boards should be established which would satisfy the operational needs of particular areas, would be spokesmen