Achieving STP in Reconciliations and Exception Management
Reconciliation and exception management have long been a challenge for buy-side institutions due to the manual nature of the processes involved. These processes not only translate into higher operational costs but reconciliation mistakes and untimely resolutions can also create financial risks due to market and regulatory dynamics. An erroneous reconciliation can lead to a short sale of securities, which invites regulatory penalties and leaves an institution open to intra-day market volatility and saddles it with huge borrowing costs. Failing to pick up the problem on time exacerbates the situation, which means performing batch reconciliation on a daily or weekly basis only increases the risk.
But despite this reality, the funds management industry has been slow to automate the process of exception management and spends a fortune each year employing people to do the job. Industry reports show that about 60 per cent of the total non-interest cost of an asset manager is attributed to personnel costs, which indicates that there is still a high level of manual processing.
The reconciliation needs of each asset manager are different, but in general, managers reconcile their asset balances, securities transactions and corporate actions entitlements with their custodians. They also reconcile their allocation instructions with confirmations given by their broker/dealer.
A generic reconciliation process involves receiving data from different sources, mapping the data by adding reference data, setting matching criteria, matching records, identifying unmatched transactions/holdings, raising exceptions, receiving resolved exception information and re-matching records and applying adjustments.
The level of automation used to achieve each of these stages differs. Recently, there has been a lot of progress in the area of data mapping between different sources of information/data. Most custodians now have asystem or are using the new ISO15022 Swift standard to map data before the reconciliation process begins at the asset manager’s end. And there is evidence to suggest that this process is working. Using mapping tools to converge statements onto a common platform helps reduce both cost and reconciliation errors.
The most challenging stage to apply automation to is the area of exception investigation and resolution. User intervention in this area is still high. To successfully automate the process would involve the following achieving the following steps: preliminary investigation of “break cases”; allocation of break cases to correct business segments: resolution of break cases; receiving break case resolution data; and applying adjustments to the incorrect data. It also involves maintaining ageing data of exceptions with a proper audit trail. It is the investigation part of this process that is never likely to be fully automated. There always needs to be some sort of manual intervention in the investigation of breaks. Though other areas lend themselves well to automation, but before exploring this further it is worth looking at the benefits of implementing straight through processing (STP).
Reconciliation and exception management is one of the most easily justifiable STP projects an asset manager can implement, in terms of return on investment. Significant saving in operational cost cans be achieved by reducing headcount alone.
Analysis conducted by Infosys Technologies shows that an asset manager processing 600 trades per day, having 1,000 securities in its portfolios and experiencing a break case rate of 5 per cent, can save about $600,000 per year by automating the exceptions process. This is a return on investment of around 40 per cent over a five-year horizon. If the financial cost of operational risk due to incorrect reconciliations is factored into the analysis, savings of over $1m a year can be achieved. In this risk-factored scenario, the returns can reach 65 per cent with a payback period on the investment being less than two years.
Other non-financial benefits from implementing STP in reconciliations and exception management are improved quality of electronic reporting, reduced operational risks and costs, and improved productivity. Automation can change the timing from being end-of month process to an end-of-day process and gradually to a near real-time intraday process. This helps asset managers to prepare for the shorter settlement cycles. Many other factors such as risk reporting requirements as per Basel II, disaster recovery and business continuity planning initiatives further help an asset manager’s cause for implementing STP in reconciliations and exception management.
The first step in achieving STP in exceptions management is to address the issue of poor documentation. Having a system that documents the processes and regulations of different markets – covering instruments, taxes and valuations – is essential. This helps to avoid any potential exceptions before they occur. To put things in perspective, an asset manager might not have correct information about tax rates on government securities issued by the Italian government before 1980 and therefore the debits appearing in the asset manager’s cash statement have not matched with the bank’s cash statement. By developing a comprehensive repository of such market information, the break cases can be resolved very efficiently and immediately. The next step involves setting up a generic framework that covers all of the reconciliation processes that an asset manager undertakes. This framework should address the pain points in the existing process with an underlying focus of achieving STP in all areas. The solution should be web-based and should integrate with the interfaces used by custodians, brokers and automated data transmission providers. This allows for speed and more efficient control.
Flexible rules engine: The flexible rules engine should allow user defined rules for each stage of reconciliation processing; with rules defined and stored for each kind of reconciliation process. For example, in the data receiving process, the system should recognize that all messages relating to one statement have arrived, and should import all the messages as one statement by looking at the reference number.
Data mapping and enrichment engine: This engine is expected to map incoming data with existing internal data to format the data to be reconciled. Formats should be automatically mapped and enriched with reference data using predefined business rules. If one custodian is using CUSIP as a security identifier and an asset manager is using ISIN as a security identifier, the data mapping tools should be able to correlate the linkage and map the data between the two records for reconciliation.
Generic matching engine: The generic matching engine should automatically reconcile the statements of asset balances/ transactions on a pre-determined basis – weekly, daily or intra-day. The criteria defined to compare each reconcilable item should be user defined and reconciliation process specific. The matching items should match on any data element associated with the item and the matching should be first configured to pair items on portfolio, statement date and security identifier, and then to match on other identifiers for specific reconciliations and specific portfolios. The engine should allow data to be compared exactly, or by set tolerance limits and default rates can be set for each matching field, with possible exceptions to this default by individual portfolio or reconciliation process. The matching engine should segregate reports with “complete match”, “partial match” and “no match” status.
Multiple interface integration and support: The proposed solution should support multiple source data in multiple formats like proprietary formats and SWIFT 15022. The solution should also be able to interface with internal systems on a real-time basis in order to receive updated data from portfolio accounting and management systems and other internal databases for reconciling with the records of custodians/brokers.
Interactive web interface: The proposed solution should have an interactive web interface for internal business users as well as external interfaces like custodians and brokers in order to receive statements. The web interface should also allow business users and external parties to send exception resolution data to an asset manager for an automatic upload onto the reconciliation system for authorization (by asset manager), rematching and sign off.
Strong workflow-based processing engine: The part of the solution allows you to assign work cases to business users for exception resolution. The work case allocation should be rule based and user-defined. For example, in the case of non-receipt of securities as per asset manager records, the workflow module should automatically send a correspondence to the custodian to credit the portfolio account with the non-received securities.
Intelligent exception investigation and resolution tools: The proposed solution should have intelligent exception investigation and resolution tools and should be able to prioritize exceptions based on severity. The exception management tools should automatically send exception reports to business users, investigate databases for any preliminary investigations, track and escalate the exceptions on priority basis, and automatically calculate adjustments for and against the asset manager. The prioritization rules need to be defined on the basis of the possible financial and risk impacts of an open break case and the investigation of internal databases should be based on strong rules to identify possible reasons for break cases, which can be highlighted to business users, thus saving the manual effort of investigation to every extent possible.
Straight-through processing of reconciliation and exception management helps save time and risk of manual matching, especially as global transaction volumes increase and more complex products are introduced. Fund managers should treat their approach to managing these processes as differentiators because it makes them more cost efficient and more accurate at what they do. For fund managers that invest in multiple markets and use multiple custodians, some sort of exception management framework is essential in order to keep on top of things. Asset managers are aware that they need to be spending more time on managing portfolios and less on reconciling accounts and resolving exceptions, which is why automation in this area makes sense.
