For the U.S. securities industry, the implementation in June of a three-day settlement cycle will begin a series of changes as profound as the move toward certificate-less processing that started in the 1960s.
The accelerated settlement process, called T+3, is not just a simple change to the back office. It is the beginning of a fundamental restructuring of the securities processing industry that might also involve a further contraction of the settlement cycle.
The idea that the U.S. settlement cycle can be shortened further is not fantasy.
During most of the history of the national stock exchanges, the time between trade execution and payment was only one and two days. It was extended to five only as the manually processed volume grew. When the settlement process was first automated, the five-day manual cycle was replicated. It has been institutionalized for almost 30 years.
Several other countries already settle in less than three days. Technology makes it possible, and the reduction in risk makes it desirable.
The move to a settlement period of less than three days will require a distinct change in infrastructure and the use of technology throughout the investment industry. But regulators are open to considering the idea.
Whether or not the cycle is shortened further, three-day settlement and the major trends affecting the securities industry will result in a greater use of standards and technology. This will finally set the stage for end- to-end, paperless processing.
The new compressed settlement cycle has four major implications for the banking industry that increase the need to consider process reengineering:
Increased usage of computers and telecommunications.
In order to meet the increasingly demanding requirements of the securities-processing infrastructure, banks, brokers, and investment managers will have to improve their use of computers and telecommunications.
Depository Trust Co., with Interactive ID, is evolving away from the traditional overnight batch process that has historically characterized most of the back-office functions.
Communication between investment managers, banks, and brokers is becoming increasingly automated as message standards are adopted and the cost of telecommunications drops to acceptable levels.
End-to-end, paperless computing is the industry's ultimate goal. It reduces manual processing costs and improves processing quality for all of market participants.
Achieving end-to-end processing requires a different effort by each of the various segments in the industry. banks, brokers, and investment managers will all have to improve their use of computers and telecommunications.
Brokers are increasingly divided into two technology categories: large brokers with their own sophisticated computing and telecommunications processing capabilities; and users of correspondent clearers, which provide a full range of processing services for many of the medium and small brokers in the industry.
Investment managers generally fall into either of two categories of their own.
*Most large, well-financed firms that are "professionally managed" have made the necessary investments in technology to support their increasingly important role in the institutional securities marketplace. They will continue to do so.
*Most small, owner-managed firms target their investment spending to areas that will contribute immediately to profitability. Investment managers in this category have tended to rely on phone calls and faxes as their primary means of communication with banks and brokers. These advisers are beginning to use technology in the back office as the cost and complexity of the technology drops to levels they can accept.
Banks are seeing a leveling of the playing field as the use of standards and technology increases throughout the securities industry.
In the 1980's only the largest banks could afford to develop sophisticated and usually proprietary customer workstations and networks to differentiate themselves.
Today the industry is demanding standardization of messages and protocols. Individual banks' proprietary terminals and communication products are becoming less attractive; they are useful to the banks only if they provide a cost advantage.
Swift, the Society for Worldwide Interbank Telecommunication, has been the basis for much of this evolution, as its message standards have been adapted widely.
Realizing that its standards failed to meet some needs of securities users, Swift began expanding its role. Industry groups such as ISITC (Industry Standards for Institutional Trade Communication) were formed to provide additional guidance and momentum.
Computers were initially installed so businesses could survive, Then they were used for competitive advantage. Now they are again becoming necessary for survival.
Support the move toward interactive processing.
In the 1960s the volume of business forced banks to automate in order to survive.
They used the available technology to replicate the existing five-day settlement cycle, by extensively depending upon the overnight batch process.
The processes were designed around the manual activities; in effect the manual activities of each department were automated as separate components, which became the de facto applications architecture.
Since then, the securities back office has generally been built upon this base of code. The result is the extremely complex "legacy systems" that form the bedrock of most firms' systems.
In the 1980s, an explosion of activity in the front office made extensive use of local area networking and computer workstations.
During this period, the back office added on-line access and increased the use of telecommunications but continued to rely primarily on mainframe systems and batch processing.
A few pioneering banks are implementing newer technologies, but they are finding it hard. Lack of knowledge about the current system works has combined with lack of experience with advanced technologies to slow down progress.
These banks are finding that their programs have been repeatedly "enhanced." The documentation is weak to nonexistent; few people still working in operations or systems remember or understand the details of how to process a transaction.
Nevertheless, as industry moves toward more and more interactive processing, all participants will have to reconsider their back-office activities.
To meet the requirements of a three-day cycle, operations will no longer have the luxury of overnight processing. Each business will have to move toward multibatch processing - or completely message-based processing.
This interactive environment will require most firms to reconsider their entire processing flow, from trade through settlement, and can fundamentally change the way work is done.
Clerks will no longer be able to spend their days working on homogeneous batches of transactions. They will increasingly have to handle multiple types of transactions, in whatever order they arrive.
This may require a different set of competencies in the few remaining clerks. It will certainly require a different level of training and computer support.
Batch processing is inherently redundant, and therefore, inefficient. However, many firms have avoided consideration of reengineering because they didn't feel they could afford to change their batch systems.
Now they can't afford not to change them.
Move settlement closer to the trade.
The reengineering opportunity is not confined to the back office. It should include every aspect of a firm's process.
To function effectively in a three-day settlement cycle, traders, portfolio managers, other participants should become more responsible for the settlement.
If the settlement cycle is further shortened and automated, the trader's role may change significantly.
As technology becomes more pervasive and the connectivity between the front and the back office improves, some firms are beginning to create a nearly completely automated process.
Robinson Humphrey may be a model of the future for banks and brokers. As the firm has automated its trading function, the roles of the trader, the middle office, and the back office have blurred so much that little difference remains.
Kevin Crean, vice president of electronic trading operations at Robinson Humphrey said:
"Technology and the shorter settlement cycle create the need for people to increasingly understand trading, back-office operations, and trading. The process is organizationally, physically, and logically integrated."
One immediate effect of the move to T+3 will be the loss of 40% of the processing time that has been available to settle a transaction.
This time will come off of the front end of the cycle. Its loss will make it impossible for a firm to recover from errors and still settle routinely in three days.
In a five-day environment, confirming or affirming could always be put off until tomorrow. Now it will have to be done immediately. And since exception processing is increasingly expensive, firms will have even more incentive to "get it right the first time."
In most situations, quality assurance is best accomplished through of a firm-wide program with a clear methodology that improves the process and empowers people.
As firms come to understand the full implications of the shorter settlement cycle, many will decide to establish a reengineering project.
The typical reengineering process involves four steps: project management, information gathering, analysis, and implementation. Each step is associated with a particular set of techniques.
In the project management phase, the manager must establish and train teams.
In the information gathering phase, the team will need to use structured techniques to define current and ideal flows.
During analysis, the team members will be involved in design and in consensus building. They will use divergent and convergent thinking to expand their options.
Finally, numerous tools can be used to aid in the testing and implementation of the new process flow.
Mr. McIntyre is a managing partner of Summit Group in Murray Hill, N.J., which provides management consulting, market research, and systems integration services to banks, brokers, and investment advisers.