m                        top

  Making STP work
   Dr Anthony Kirby

In today's business climate of 'show me the money', companies need to continuously improve their bottom line productivity and use information efficiently to increase the quality of customer service that they deliver. The financial services industry in particular recognises that it needs to move from today's sequential processing into a more streamlined process defined by the principles of straight-through processing (STP) -- more automated processing, less human intervention and redundant data entry throughout the life cycle of the trade.
The drivers to achieve STP come in many forms -- operational cost reduction, the pace of globalisation, increased competition, the growing regulatory requirement for shorter settlement cycles and advances in technology, not to mention the growth in trade volumes and the exploitation of the real benefits of e-commerce. This is because in essence, STP is about linking together different parts of an institution and the wider industry to enable businesses to function as part of a greater whole. In other words, STP is achieved when there is no manual intervention at any stage in the transaction lifecycle spanning execution through to eventual settlement.
STP continues to be a top priority for the global securities industry, driven by high costs, exposure to operational risks and the need to mitigate trade failures. Furthermore, achieving STP is about helping firms make the transition towards optimising the handling of information in the internet world. This is because information content is critical to the successful initiation and flow of the trade to its safe completion. So STP has strategic ramifications for the boardroom equal to its relevance in the back office. But rather than being an endgame in itself, STP needs to be considered more than merely an arcane or isolated goal for the financial services industry.
Last year's joint Reuters/Capco/TowerGroup report and the T+1 report on securities processing in the US cited inconsistent, inaccurate and incomplete data as the major cause of failure to achieve STP internally. After all, 40% of a trade record is composed of reference data. Despite the impact of poorly managed data, 37% of respondents stated that their organisation did not have a reference data strategy. In addition, 55% said they did not have, or were unaware of the existence of, defined data reference standards within their organisation. The study concluded that financial institutions would spend upwards of USD19bn over the next 4 years in transforming their businesses to meet the demands of STP and the T+1 deadlines.
These findings highlighted     the fact that, by failing to implement the necessary controls and strategies to manage the quality and consistency of their reference data, banks are exposing themselves to undue risk, inefficiency, costs and failed trades. The research also highlighted the industry's reliance on manual entry and maintenance of reference data, in conflict with the requirements of STP. This is because in today's securities processing world, information relay is often sequential, resulting in poor synchronicity of vital information throughout a transaction lifecycle.
The findings were in accord with an operational survey published by SWIFT last year. The lack of STP was estimated to cost the securities industry USD12bn every year. Failure rates in the back office were reaching new heights, generating significant operational costs owing to the following:
    •    One in five cross-border trades failed to settle on its contractual settlement date.
    •    a single cross- border trade can involve as many as 25 different people, processes and systems before it is finalised; many of these steps add little or no value to the trade process.
    •    60% of settlement instructions required repair at a cost of USD6.00 per repair.
    •    61% of allocation matching was considered to be batch intensive.
So STP is seen as a central component in helping firms deliver scale, by providing the right foundation for financial institutions to respond to critical business issues. STP is important for a raft of other reasons too:
    •    Helping firms boost their productivity by realising greater operational efficiencies
    •    Helping firms improve their bottom line by reducing costs and/or converting more fixed into variable costs
    •    Helping firms deliver scale by opening up capacity to remove volume sensitivity and designing future flexibility via open systems architecture
    •    Helping firms provide great customer service by facilitating information flows and leveraging knowledge management (thus improving performance and differentiating themselves from the competition)
    •    Helping firms link the different areas of their business, e.g. Securities and FX, to deliver better value for their customers
    •    Helping firms manage operational risk and improve their control, particularly at a time when customer expectations are growing yet margins are shrinking
    •    Helping firms meet legal and regulatory requirements in various market centres -- especially under the constraints of shorter settlement cycles, e.g. T+1
    •    Helping firms create greater business resilience through standards, market practices and attention to compliance.
More firms need to adopt a diagnostic response which will help firms pre-empt STP problems occurring later in the transaction cycle by supplying them with the right business information and tools at the outset. After all, surely prevention earlier in the cycle is preferable to more expensive cure days later. By way of example, many STP problems are the result of a lack of data integrity or the lack of information synchronicity. What is clear is that success in raising STP rates isn't just a matter of better standards per se, although standards are indeed key to supporting efforts between enterprises. The problem has to be tackled intra-enterprises first, and that means that the right information needs to be presented before the right user at the right time for effective decision-making.
In my role at Reuters I am responsible for formulating and driving the company's STP strategy from a vision whereby Reuters works with the industry to act as a critical enabler of infrastructure change over the next 3-5 years. The focus needs to be customer-centric, to work on what is happening within each firm's trading, settlement and operations areas, and thereby to improve the quality and integrity of data which passes through these areas. This means that businesses will be able to fulfil their transactions with a greater degree of confidence.
As accurate and well-managed data is the foundation of STP, particularly market reference data as applied to business counterparties, Reuters data management solutions (including our recent joint venture with CapCo) will offer a range of downstream data management tools and diagnostic services. Using a unique set of algorithms, Reuters will be able to assist its customers to configure their own data models, to look for and predict where inconsistencies or discrepancies occur which contribute to failed trades. These approaches to improve exception-based management will be popular because they will help our customers make better use of increasingly scarce IT resources, representing a better return of investment.
In the area of Transactions Management, the acquisition of InterTrade Direct from Liberty as far back as 1995 coupled with last year's acquisition of Bridge and their IOE2 order routing product added the combined liquidity of several hundred money managers. This strategy complements the 'View & Do' capabilities of Smart Trading which allows customers to transact directly across a wide range of liquidity pools, and Smart Data in the Fixed Income space which allows users to view data such as prices, and then allow the user to form a link with critical information such as analytics prior to transacting. Reuters also provides enterprise risk management solutions which enable our customers to manage transactions across their entire lifecycle.
Finally, the acquisition of leading middleware integrator Tibco in 1995 allowed Reuters to pioneer integration systems which share data across an enterprise. The Rendezvous suite of Tibco services is already used by financial institutions across the globe. Reuters won the right in 1998 to distribute a full range of Tibco products and services. In 2001, Tibco additionally acquired a middleware stack with full transactional capabilities called Talarian, further demonstrating Reuters commitment in meeting the STP challenge in the integration space. Reuters will focus on providing solutions in Data Management and Business Process Management, working with solution providers who are market leaders in the Order Management, Portfolio Management or Workflow Management domains. We will continue to evaluate all options for growth, including co-marketing, alliances, joint ventures and acquisitions. Reuters will leverage its open systems technology and drive alliances with third parties, acting in effect as a 'vendor of vendors' providing greater choice and future flexibility for our customers. Reuters will also continue to be supportive and inclusive of industry efforts such as GSTPA, OMGEO and CLS which are taking active steps towards delivering STP. The use of virtual matching utilities (VMUs) such as GSTPA's TFM or Omgeo's CTM as a response to shorter settlement cycles will highlight the importance of good-quality reference data because high matching rates are a function of compatible data,
Despite conventional views, STP and T+1 are not only a back office issue. They affect front office trading applications, risk management and many middle office practices. The need to enhance the settlement process requires significant involvement of front office staff, technology and processes. Data integrity and synchronicity will be key for T+1 and an instrumental part of end-to-end trade processes will be the availability of dynamic/static reference data charging the trading applications. Certain reference data must be accessible by the front office at the point of trade entry and execution for validation purposes. The front office requires RIC codes for identifying asset classes while the middle office depends on ISINs. Access to counterparty identifiers such as BIC codes or SSIs for identifying settlement instructions in order to expedite the trade are equally relevant. Enhanced risk management and credit risk control requires real time access to matching and counterparty information.
To quote the late Hagay Shefi of MINT and GoldTier Technologies, "Upon execution of a trade today, the order sets off a complex, multi-day procedure. Faxes and phone calls are exchanged back and forth between broker/dealers, asset managers, custodians and other players -- even for a simple domestic equity order. What's more, as trading volumes explode, settlement fail rates increase, degrading the quality of customer service, leading to increased challenges of client retention. Most financial institutions do not have a sufficient degree of STP to be able to achieve settlement by T+1. The internal and external manual processes that are in place today require the full three days between trade and settlement to complete all the steps included in the trade process. Moreover, in about 12% of cases, 3 days is not sufficient and affirmation does not occur by the trade settlement date."
Although operational costs can be reduced with greater involvement of front office and trading applications, this does not necessarily justify the business case for STP or for migrating to a shorter settlement cycle as such. We plan to demonstrate how e-commerce approaches to supply chain synchronicity within e-commerce would support the business case for STP by supporting effective CRM. STP in e-commerce is accepted but merely as a consequence of automating information relating to transaction processing rather than as an endgame in itself. Corporate firms have reasons for implementing STP; improving the bottom line through cost reduction and operational efficiency (a common problem), or differentiating from other competitors on the basis of knowledge or workflow management with a view to fulfilling their customer relationships and expectations.
So information such as global reference data concerning counterparties, accounts, assets or transaction workflows will be fundamental to realising the goals of STP over the next five years. These could be fertile times to recover from the mass automation compliance programmes such as the euro and Y2K, and begin to design the architectural features to build future flexibility into transaction processes. But ripping everything out and starting from scratch assuming we have a green field scenario is no longer an option, if it ever was. After all, there are simply too many moving legacy pieces for that.
The bad news for some is that the cost of rolling out a complete STP programme versus the time taken to realise the ROI could well be prohibitive for many players. The Tower Group's projections of STP spending in preparation for T+1 in the US over the 2001-2004 cycle showed that the broker/dealers were likely to pick up the majority of the burden (USD7.01bn or 64%) as compared with the asset managers (USD2.78bn or 26%) or custodians (USD0.65bn or 6%). Furthermore, the SIA's July 2000 study indicated that the investment cost for an STP solution versus the time to achieve ROI varied considerably; the asset managers showed an ROI on the basis of 4.22 years (compared with 2.58 years for a broker/dealer or 1.49 years for a custodian).
A further study by Celent showed that the lion-share of STP expense (74.6% for a typical large US broker/dealer) distributed between altering internal processes to benefit from STP or optimising market or workflow information in order to improve business decisionmaking. The stark asymmetry in the business case justification suggests that broker/dealers and custodians have a better value proposition in offering outsourcing services to their clients, who might not be able to make the systems leap necessary to achieve STP.
Many GSTPA members that I spoke to during 1999-2001 on the buy-side consistently stated that they wanted operational convenience and future flexibility/migratability. This was a natural consequence of their need to manage costs, service clients through enhanced performance and improve their controls, particularly at a time when customer expectations were growing yet margins shrinking. In fact the desire for 'one-stop' shopping was mentioned on several occasions. However, given the marginal business case for STP for the buy-side, it is perhaps unsurprising that asset managers are no longer looking at central matching in isolation. Many instead are finding the means to
justify the business case on the basis of a solution which sees a convergence behind the worlds of pre- and post-trade rather than that of central matching as a process in isolation.
So, STP must represent the means, and not the end, if industry efforts are to be successful. STP has to be achieved intra-enterprise before it can be realised inter-enterprise. This means that firms have to come to grips with the necessary automation, re-engineering of business processes and integration spanning investment decision, execution, trade processing, settlement and finally, reconciliation of the transaction. At the same time, as many enterprises evolve common inter-enterprise solutions to shared problems, community connectivity with rules-based management will help achieve the automated processing of a transaction across the value chain of participants.
Information enablers such as Reuters have not been slow to recognise that the scope of STP must include the entire life of a trade, including business support process, end-to-end execution through to settlement and subsequent reconciliation. These vendors are only too aware that their timely mix of utility and value-added services (such as benchmarking and knowledge that comes from understanding the workflows) could well become the hallmark of an effective end-to-end STP model. The more that STP is seen as the integration and synchronisation of vital transaction information, the more apparent it will become that the information enablers will have a vital role in helping the industry raise and shape its game.

Dr Anthony Kirby is Head of STP at Reuters.