Sie sind auf Seite 1von 28

The Journal of

Compliance Risk & Opportunity

Banking, Financial Services & Insurance
July 2011 VOL V Issue 9

RBI Publishes its Third Report

HDFC Bank Upgrades to TCS BaNCS Treasury 5.0

Technical Computing in Risk Management

Basel III: Celent Report

2nd Annual Retail Banking Asia Pacic

4th - 5th October 2011, Kuala Lumpur, Malaysia

Discover & Share INNOVATIONS! Sharp minds. Sharp solutions.

What do customers really want with their bank? What motivates customers to stay loyal? How do innovation help?



to enhance protable retail banking strategies

2 Days of Innovative Ideas 5+ Bespoke Networking Hours 12+ Expert Speakers 17+ Resourceful Topics

T: +603 2141 5352 E: Finsights readers benet from 10% off the registration fee

Of i l Travel P Ofcial Travel Partner: l

Research P R Research Partner: ch

Media Sponsor: M di S

Organised By: O i d

Media Partners:

July 2011

Finsight Media
TF-07, Suner Complex Harinagar Crossing, Gotri Road Vadodara 390 021 Tel: +91-265-6533475 Email: Website:

Publisher T S Chandrasekaran

Editors Note
Financial stability had begun receiving global attention at the turn of the millennium, when the Financial Stability Forum was founded in 1999 by the G7. Ten years later, in 2009 it was expanded by G20 and renamed as Financial Stability Board. In 2010, 27 member states of the European Union established The European Financial Stability Facility to ensure financial stability in Europe. In the US, The Office of Financial Stability was created by the Emergency Economic Stabilization Act of 2008, and The Financial Stability Oversight Council was created by the DoddFrank Act 2010. The Reserve Bank of India has been concerned with stability of the financial sector so far. But, the emergence of financial conglomerates on one hand, and products such as unit-linked insurance on the other, gave rise to inter-regulatory disputes over span of controlthe SEBI IRDA spat is still fresh in the memory of Indian financial sector. The post crisis focus on establishing an institutional mechanism for coordination among regulators and the Government has culminated in the establishment of the Financial Stability and Development Council (FSDC) in December 2010 to be chaired by the Union Finance Minister. Still, the role of the RBI as systemic regulator has not diminished, as is exemplified by its release of third Financial Stability Report last month, the gist of which forms the cover story of this issue. A detailed discussion of how to build software to estimate a credit value-at-risk (VaR) measure for a bond portfolio is another highlight of this issue. Launch of a new version of treasury solution by TCS and an automated regulatory reporting system for banks to meet the RBI guidelines, are also covered in this issue.

Editor Hari Misra

Advertising Amita Tiwari Production Sanket Vohra Design Chandrakant Kokje

Copyright Finsight Media. All rights reserved. Neither this publication nor any part of it may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of Finsight Media. Published monthly. MAHENG/2006/19566. Compliance Risk and Opportunity

Published by T S Chandrasekaran Bldg. No. A-4, Flat No. 2-B, Sujata Co-Op. Society, Bund Garden, Pune - 411001 and printed by him at Spectrum Offset, D-2/4, Satyam Estate, Behind CDSS, Erandawane, Pune - 411004

Hari Misra

Price Per Copy: INR 50 Annual subscription: INR 600

HDFC Bank Upgrades to TCS BaNCS Treasury 5.0 iCREATE Launches Automated Regulatory Reporting Solution for Banks FINANCIAL STABILITY
RBI Publishes its Third Report

6 8

10 14 20

Technical Computing in Risk Management Basel III

Navigating Business and Risk Technology Architecture Decisions


Risky Business for Hedge Funds When Selecting a Business-Critical System

23 25

in brief


Stamp out fraud.
With SAS Analytics, you can score millions of transactions a day in real time to detect fraud faster, reduce risk, streamline investigations and prevent losses. Decide with con dence.

Scan the QR code* with your mobile device to view a brief customer video or visit for the complete success story video.

For more information contact

*Requires reader app to be installed on your mobile device

SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute Inc. in the USA and other countries. indicates USA registration. Other brand and product names are trademarks of their respective companies. 2011 SAS Institute Inc. All rights reserved. S75378US.0611

July 2011

HDFC Bank Upgrades to TCS BaNCS Treasury 5.0

TCS Financial Solutions, a strategic business unit of Tata Consultancy Services (TCS) has recently launched TCS BaNCS Treasury 5.0. HDFC Bank, which had implemented TCS treasury solution in 2008, has upgraded to the new version. We had found the features of quick go-to-market with interfaces and facilitation for Sarbanes Oxley compliance, most useful for our bank, recalls Harish Shetty, executive vice president, IT, HDFC Bank. As an existing customer of TCS BaNCS Treasury solution, he lists the personalised single workspace feature along with browser based platform agnostic SOA-ready Java EE architecture features of the latest version, features in the recently released version 5.0, which made him to go for the upgrade. Apart from HDFC Bank, two other banks have opted to upgrade to the new version, informs NG Subramaniam, president, TCS Financial Solutions declining to name them. We are in advanced stage of discussions with other clients to enable them to upgrade, he says. TCS BaNCS Treasury 5.0 is an integrated solution which supports multi-entity, multicurrency, multi-asset class system with process coverage for front-, mid- and back-office operations. One of the key enhancements in Treasury 5.0 is the five-layered hierarchical portfolio structure, which aligns to the desk, book and folder organisation in a treasury, informs Subramaniam. This layered approach helps a treasurer with added transparency in identifying the source of risk generation, advanced features as bootstrapping, various curve fitting techniques and the ability to value a position or deal using multiple curves based on purposes such as from the frontoffice perspective or the backoffice accounting perspective, he says. The personalised single workspace which Shetty likes, provides the user with a capability to configure and personalise screen layouts. This enables users to execute a days operation without navigating through a maze of menus. TCS BaNCS Treasury 5.0 is an SOA ready, browser-based solution with Java EE architecture. The platform agnosticism helps achieve a lower total cost of ownership (TCO) while giving the institution various options for hardware and database platforms, asserts Subramaniam. What were the key drivers for the new version? The Treasury and Capital Markets space is a highly volatile one in terms of new instruments, structures, regulatory frameworks, risk measurement techniques, observes Subramaniam. Our product strategy is tightly aligned with sales and marketing strategies. The teams constantly interact with customers, analysts, industry bodies to provide directions to the roadmap. The personalised single workspace and SOA ready platform agnostic architecture are directions which we took from our customer forums at the backdrop of SIBOS events held every year, while the recent regulatory activism has helped us formulate the deskbook-folder approach and rich

Harish Shetty

and revenue. This is further strengthened by intuitive position transfer capabilities from sales desks to the market desks, who in turn are the real position owners, he elaborates. The mid- and backoffice have always been a strong area for us. With real-time analytics, blotters and comprehensive position keeping in 5.0, we have taken a taken a significant upstream move into the front office, he adds. According to Subramaniam, other enhancements include an open framework for derivative structures and strategies and the ability for market curve management. Through the former, a number of derivative structures and strategies can be created on the fly, or standardised for future use. Traders can quickly cobble up a strategy, price it and then execute the same with the counterpart. The ability to source market information has been expanded to cover synthetic curve


July 2011

analytics. The hierarchical deskbook-folder alignment using the layered portfolio definition is a key element in our product differentiation, says Subramaniam. With increasing regulatory oversight, it is imperative for banks to ensure transparency, which TCS BaNCS Treasury provides. Banks are able to review their performance, exposure, and risk profile in real time, monitor pre- and post-trade compliance results, all with the backdrop of a robust audit trail capturing every action in the system. Further, the comprehensive five-layered portfolio structure provides a logical demarcation of proprietary trading and client trading activities. The risks and revenue recognition is also based along with the portfolio structure, explains Subramaniam. BaNCS Treasury 5.0 enjoys the execution capabilities of the supplier with its solution centres dotted round the globe to ensure the nearness to the client for support. In the near-term, the Middle East (inclusive of Turkey) is a target for us. We are taking a focused approach towards this market in terms of additional support for Islamic products and processes, informs Subramaniam. Outlining the roadmap for the product for the next two-three years Subramaniam says that tremendous potential is emerging in the arena of Cloud Computing. Standardised functions such as settlements, confirmations, among others, can be placed on the cloud to allow multiple tenants to utilise standard functions. Secondly, the solution will be iPad ready, which is another important step for us.

HDFC Bank uses point solutions for derivative pricing, liquidity risk management etc in addition to BaNCS Treasury. It also uses a third party system for managing market risk. But for limit monitoring, we use this solution, says Shetty. Commenting on pros and cons of using a set of best-of-breed point solutions against using an integrated solution, especially in case of treasury solutions, Shetty says that while best-of-breed point solutions do provide rich functionality; integration, standardisation, support availability, reduced straightthrough-processing capabilities and vendor management pose a challenge. An integrated solution provides better integration across asset classes, front-to-back integration, and standardised messaging with enterprise systems. There is a single vendor to manage. But, for an integrated solution time to market is a challenge, he says. However, Subramaniam opines that with regulators permitting new asset classes, an integrated solution will be preferred over point-solutions despite pros and cons attached to each. . Commenting on the replacement market for treasury solutions in India, Subramaniam says that with Indian banks expanding overseas there will be a growing need for a multi-entity solution. Regulatory compliance, such as impending IFRS convergence in India will call for an overhaul of the existing treasury systems, especially, on the front of IAS 39, among others. With India taking steps towards full convertibility, the existing dual currency systems of Indian

NG Subramaniam

treasuries have to transform to multi-currency systems and that will call for a major overhaul across systems. We also see a potential in the corporate segment in India. Indiabased MNCs are in need of a specialised treasury system with hedge management capabilities over and above the generic ERP coverage, he emphasises. In the evolving regulatory framework, which activity will become more profitablemanaging banks own treasury or offering treasury products to corporates? The objective of managing bank's own treasury is more to protect core income, while offering treasury products to customers is to generate the profits, explains Shetty. Both are equally important, he says. Among his wish-list from a future treasury solution are: offering on cloud, trading on iPad/mobile devices and better integration with other systems.


July 2011

iCREATE Launches Automated Regulatory Reporting Solution for Banks

Bangalore-headquartered iCreate Software launched its reporting solution Biz$core to enable banks comply with automated data flow guidelines outlined in the approach paper of Reserve Bank of India (RBI) published in November 2010 to enhance data quality, ensure data integrity, accuracy and timely reporting. The paper was prepared by a core group consisting of experts from banks, RBI, Institute for Development and Research in Banking Technology (IDRBT) and Indian Banks Association (IBA). The paper suggests the methodology to be adopted by banks to classify themselves into a cluster based on its technology and process dimensions. Banks are required, in the first phase, to ensure seamless flow of data from their transaction server to their management information system (MIS) server and automatically generate all returns from the MIS server, without any manual intervention. In the second phase, RBI plans to introduce a system for the flow of data from the MIS server of banks in a straight through process. Banks have been given sufficient time for completing the first phase of the project. Vivek Subramanyam, CEO, iCreate Software spells out the details of this solution in an exclusive interview with CRO. Here are the edited excerpts: CRO: Main concern of regulators worldwide, RBI included, is transparency in regulatory reporting. How does iCreates Biz$core solution ensure this transparency? features such as pre-built RBI returns, in-built workflows, automated returns submissions supporting XBRL, XML and XLS, business configurations, reference data management, data lineage and metadata management, adjustments and audit trail. The solution also has the additional strategic advantage of helping a bank jumpstart its business intelligence and analytics initiative. The compliance solution that is built around Biz$core can very easily be scaled and extended to be an enterprise business intelligence solution given Biz$cores modular approach and thus can transform to becoming a strategic decision support ecosystem for the bank in addition to helping them comply with RBI reporting requirements. CRO: How does your solution automatically gather data from multiple systems of the bank? How much customisation does the solution require to get data from all such systems? Vivek: Biz$core comes with a data integration framework where predefined adapters exist that are transaction system aware. We have productised the process of connecting to and accessing data from typical banking transactional systems such as core banking, treasury, GL, credit cards, trade finance, etc. This data integration framework drastically reduces the time it takes to have Biz$core up and running in a bank. Typical transaction systems come in various versions and have some level of bank specific customisations as well. To that

Vivek Subramanyam

Vivek: Regulators worldwide seek transparent and accurate reporting, which translates to automated straight-through reporting without any manual intervention. This requires technology solutions which can automatically collate data from multiple banking transactional systems and transform it into reports that the regulator requires periodically, and automatically transmit these to the regulator. Such a solution would ensure transparency and accuracy. Our Biz$core is a packaged business intelligence and analytics solutions built specifically for banks. One of the Biz$core components is a central bank reporting solution that would help banks achieve exactly that. In the Indian context, this solution is called Biz$core RBI ADF (Automated Data flow) Solution. This solution is an extensible and comprehensive regulatory reporting solution with advanced


July 2011

extent, there is a need to customise the data adapter framework to fit in at each bank. However, Biz$core ensures that the data integration is achieved easily and in around 30 to 40 percent of the time it would take using traditional methods. CRO: How much time would a typical bank require to become fully operational on this solution? Vivek: Though RBI has defined a framework for each Bank to assess their maturity and based on that a timeframe is prescribed by when the solution needs to go live. Independent of this, our solution has different deployment flavours available based on the maturity of each bank. Depending on whether a bank has a data warehouse or not, the bank has data that needs some focus on data quality or not and some other parameters, we have three implementation approaches and the right choice of implementation approach would need to be made for each banks specific context. Keeping the above variations in mind, some banks could have the solution up and running in as little as 8 weeks while for some it could take around 24 weeks. CRO: How does the bank extend this solution to include future RBI reporting requirements? Vivek: Banks have different choices available for them to become RBI ADF compliant. They could build such a solution inhouse or outsource the development of a bespoke application to a system integrator or they could choose a product like Biz$core. Each choice has its own implication in terms of effectiveness, lead time, cost and quality.

With the Biz$core solution, in addition to the advantages stated above, banks would be future proofing their RBI ADF solution. RBIs regulatory compliance requirements would continuously evolve and the solution deployed would need to be changed to ensure that banks keep up with the changes in requirements. With a productised solution, banks need not have to worry about this scenario. We will be tracking the compliance requirements closely and will be releasing upgrades to the product that ensure that this is up-to-date and this makes it extremely hassle-free for a bank when it comes to dealing with and staying on top of ever changing regulatory requirements from the central bank. CRO: What would be needed to be done if a bank replaces any of its back office point solution? Vivek: Irrespective of changes to a banks back office solutions, our solution would continue to be relevant, with minimal changes. This is because the product has a data model defined specific to the solution and which expects specific sets of data from specific systems at a certain periodicity. For example, when System A is being replaced by System B, the effort that will need to be put in would be to start sourcing all the data that was earlier being sourced from A to B now. Integration with B is relatively easy again considering the robust nature of Biz$cores data integration framework. CRO: Has the solution been audited for compliance requirements? Vivek: One way to audit the solution would be to run the system on historical data and

compare the output with the reports that were actually submitted at that time. This would need to be done as part of each implementation at every bank. Other than this, there are no specific audits or certifications needed for such a solution, currently. As and when clarity emerges around the need for such a solution to be audited or certified, we would be ensuring that the solution definitely gets certified. CRO: Did you partner with any bank for developing this solution? Vivek: No, we developed this independently and did not partner with any bank to build this. But we did build a team of bankers and ex-bankers who have significant experience in compliance area to help us define, architect and build the solution. CRO: Which platform does it use? Vivek: The Biz$core technology at a high level comprises of three components. We have OEM-ed industry leading and award winning technology around BI/OLAP, data integration platform, and built our solution embedding these. From a customer standpoint, this is abstracted and they just need to buy our product alone and all the technology needed to power the solution will be part of the Biz$core license. One flavour of our product uses industry leading platforms Microstrategy and Informatica and this flavour is platform and database independent. Another flavour of our product is built around the best-in-class Microsoft platform. Depending on the preference of each customer, we deploy the flavour of the solution.


RBI Publishes its Third Report

This article is a gist of the third Financial Stability Report (FSR) published by the Reserve Bank of India (RBI) last month, presenting its assessment of the health of Indian financial sector. The 70-page report, which contains 9 boxes highlighting a concept each, 125 charts, 11 tables, and an annexure on stress testing methodologies, can be accessed by interested readers at The first FSR was released in March 2010 and the second in December 2010. According to RBI, now onwards FSRs will be released bi-annually in June and December every year. In his foreword to the report, Dr D Subbarao, governor, RBI observes: Ensuring financial stability cannot be a formulaic rule-based task. The endeavour for the policy makers should be to not get trapped in commoditised ideas, reductive categories and prepackaged narratives. The FSR is divided into five chapters on macroeconomic outlook, financial markets, financial institutions, financial sector policies and infrastructure, and macro-financial stress testing.

Macroeconomic developments
The global risk scenario has improved during the last six months, though there are signs of a slowdown in growth during 2011 in most countries, including some of the developing economies in Asia. The main factors affecting the global growth are: high food, commodity and energy prices, steps towards fiscal consolidation, sovereign debt problems in the Euro area and high level of government debt in some advanced economies. Also, the main underlying factors behind global imbalances remain largely unaddressed, increasing the uncertainty in global recovery. The sovereign debt crisis in

countries like Greece, Portugal, and Ireland is posing serious challenges for the stability in the entire Euro area. The increasingly high levels of government debt in other advanced countries are also adding to the uncertainty around the fiscal consolidation and its impact on international financial markets. Although the Emerging Market Economies (EMEs) have more comfortable fiscal space and better growth prospects, there are still significant risks on the fiscal front, given the complex inter-play between growth and inflation. The slackening of global recovery, high oil and commodity prices, deceleration in domestic industrial growth, uncertainty



July 2011

about continuation of strong growth in agricultural sector and impact of monetary policy actions pose downside risks to India's Gross Domestic Product (GDP) growth during 2011-12. The slowdown in growth momentum may affect the quality of the assets of financial sector. The international prices of food, energy and commodities are expected to remain high during 2011-12. Although there has been some decline recently in international oil prices, this may not help in inflation management as complete pass-through of previous escalations is still to be affected. Inflation is likely to face upward pressure from higher subsidy expenditure of the government and rise in wages and raw material prices. Housing prices have undergone some correction but continue to stay firm. Gold prices continue to increase on the back of strong demand. Recent growth in India's exports may off-set, at least partially, the expected increase in the import bill due to elevated oil and commodity prices. There does not seem to be an impending pressure on the financing of CAD. However, going ahead, as the advanced economies exit from the accommodative monetary policy, there could be some slowdown in capital inflows. In the wake of high international commodity and oil prices, the budgetary projections of deficits for 2011-12 are expected to come under pressure. Management of government expenditure, especially subsidies bill, will pose challenges to the process of fiscal consolidation. This could be further accentuated by a tempered growth adversely impacting the revenue collections.

Financial markets
During the last six months, global financial markets have been resilient, overcoming a short phase of heightened volatility caused by the earthquake in Japan and political tensions in Libya and other parts of the Middle East and North Africa (MENA). The forecasted value for Financial Stress Indicator (FSI) for India, a measure to capture the severity of contemporaneous developments as they occur in different market segments and the banking sector, suggests benign conditions in the near term. The sovereign debt crisis is threatening to affect some of the bigger economies even as the high deficit and debt levels in Advanced Economies (AEs) like US, UK and Japan could exert further pressure on their sovereign rating outlook. The low economic growth combined with the high levels of debt in these countries is adversely impacting market sentiments. Continued concerns regarding sovereign risk could raise the funding costs of the financial sector and have a negative impact on its balance sheet. Evolving regulatory changes will require financial institutions to raise fresh capital even as they face a wall of refinancing at a time when sovereigns in AEs also have high borrowing programs. The sustained demand and growth in EMEs are providing strong impetus to commodity prices but the increasing financialisation of commodity markets might be adding to the volatility in commodity prices. It could also result in an increased correlation between financial and commodity markets, thereby facilitating faster transmission of shocks across markets.

In spite of a sharp turnaround in Government cash balances with the Reserve Bank during the current financial year, liquidity in the system remained in a deficit mode reflecting an increase in liquidity requirements of the economy. The increase is mainly attributable to strong credit demand and high level of currency in circulation. However, the overnight call rates have remained range-bound. The collateralised markets continued to remain the predominant money market segment of the money market. The government bond yields hardened across all maturities. The increase was more pronounced in the short end resulting in a flattening of the yield curve. Rupee has remained range-bound, reflecting a relatively balanced external account and the general weakness experienced by the US dollar during the period. Availability of alternative channels of funding has reduced the dependence of firms on domestic bank credit over the years. Rising domestic yields are widening the interest rate differentials vis--vis AEs, resulting in a greater access to External Commercial Borrowings (ECB) by Indian firms. This trend is causing a build-up of currency mismatches in their balance sheets. During the period from 2005 to 2008, large amounts were raised through Foreign Currency Convertible Bonds (FCCBs) by many Indian companies with elevated conversion premia. Most of them are nearing maturity by March 2013. Estimates show that a very large proportion of these FCCBs may not get converted into equity thus requiring their refinancing at the much higher interest rates prevalent today.



July 2011

During 2010, Indian capital markets received a significant amount of net portfolio capital flows. These flows tend to be more volatile, though their impact on the domestic macroeconomic situation so far has been limited. While equity markets in India have undergone some downward correction with Foreign Institutional Investors (FIIs) pulling money out, the bond markets have seen incremental flows on account of attractive yields and the recent enhancement of limits for FII investment in corporate and government bonds. An internal study points to tendency of the portfolio capital flows to be 'autocorrelated' thus implying 'herd behaviour', both in good times as well as during times of stress. Program trading systems in Indian stock markets Encouraging the use of Algorithmic trading and High Frequency Trading (HFT) adds to the efficiency and liquidity of markets but carries some risks too. Indian securities markets have withstood systemic events in the past, without any major disruption. Even as facilities like Smart Order Routing (SOR) are introduced in Indian stock exchanges, events like 'flash crashes' witnessed in US equity markets in May 2010, need to be guarded against.

22.6 percent in credit off take. The growth in deposit mobilisation, at around 18 percent did not keep pace with the growth in credit, the gap being funded through an increasing share of market borrowings. This increased reliance on borrowed funds raised concerns about the liquidity position of banks arising from growing maturity mismatches, in conjunction with a reduction in the share of liquid assets in total assets. Asset quality improved mainly on the back of the credit growth which outpaced the growth in NPAs. The write-offs of NPAs by banks to cleanse their balance sheets also helped in achieving a lower gross NPA ratio. The contribution to the credit growth was disproportionately high for three sectors retail, commercial real estate and infrastructure. As each of these sectors have a peculiar set of asset quality propositions, the brisk growth in exposure seen during 2010-11 poses some concerns. The asset quality under the priority sector lending, especially agriculture, deteriorated at a faster rate as compared to the overall asset quality. The system level CRAR under Basel-II norms stood at 14.3 percent as at end March 2011 which was well above the regulatory minimum of 9 percent. There was, however, a slight decline over the CRAR of 14.5 percent as at end March 2010, largely due to robust credit off take. All the bank groups had CRAR above 12 percent as at end March 2011 under Basel-II norms. An increase in NII facilitated growth of around 20 percent in aggregate net profit of the banking system, even with an

almost stagnant non-interest income and increase in risk provisions. The public sector banks registered a lower growth in profits mainly due to reduction in trading profits, increase in provisions towards staff expenses (including those for pension liabilities) and towards impaired assets. Going ahead, with hardening interest rates and the imminent increase in cost of funds, the credit growth is expected to slow down, which could adversely affect the profitability. The hike in savings account interest rate, amortisations of pension liabilities and potentially enhanced provisioning requirements for NPAs may also impact profitability. Basel II and III Indian banks, at the aggregate level, remain adequately capitalised at present. The progress towards the advanced approaches under Basel II remains on a firm footing, amidst some challenges. The main implementation issues for the migration relate to constraints of data, tools, methodologies and necessary skills for quantification and modelling of risks. As the phase-in period for Basel III measures commences in 2013, the banks will need to gear themselves for the demanding data and analytical requirements for the revised liquidity framework. The position in respect of capital remains comfortable though some individual banks may need capital infusions which could pose some difficulties if the sluggish performance of the equity markets persists. The capital needs of banks will also be impacted due to the unamortised portion of

The recovery in economic growth during 2010-11 has been accompanied by a strong credit growth and slight decline in Non Performing Assets (NPAs). The banking sector balance sheet increased by 19 percent during the year ended March 31, 2011, spurred by a robust growth of



July 2011

pension liabilities to be absorbed by April 01, 2013 on migration to International Financial Reporting Standards (IFRS). The calibration of the countercyclical buffers proposed under Basel III will require accurate assessment as to whether the credit growth is excessive and/or is leading to the build-up of systemic risks. The commonly used indicators, including the ratio of credit to GDP, may not be suitable for India and a combination of qualitative judgment and quantitative indicators may be the way forward for assessing the requirement for, extent of and timing of imposition and removal of the buffer. An analytical framework to assess the network of the Indian banking system reveals that the system is substantially connected and clustered. This intertwined nature of the banking system in any system could leave it vulnerable to domino effects in case of idiosyncratic failure of one or more banks. While the contagion impact is relatively contained due to regulatory limits on interbank exposures, there remains need for continuous monitoring of the interconnectivities in the financial system to identify build up of risks /excesses in the system and to guide policy action to address the same.

synergies arising out of such interdependencies comes bundled with risks as stress on credit/liquidity aspects in one segment/institution/process may affect the other parts of the settlement system due to the cross-linkages. The management of liquidity risks faced by the CCPs entails addressing vulnerabilities arising from the quality and range of collateral, quantum of margins and model risks. There are vulnerabilities in the Indian context arising from dependence on committed backup liquidity for funds and securities from financial institutions for completion of the settlement process (in the case of Clearing Corporation of India Limited, ie CCIL) and exposures to the banking sector as collateral is accepted the form of bank deposits, bank guarantees, etc. The risks of the failure of a CCP, however unlikely, need to be addressed given the potential collateral damage from such an event. There are, however, no easy solutions given the moral hazard concerns which the provision of central bank liquidity for CCPs entails.The OTC markets in India with their skewed participation structures need greater attention towards standardisation and introduction of central clearing even as some segments face low volumes making it difficult to mandate guaranteed clearing for these markets. The existing reporting arrangements for OTC markets encompass foreign exchange, interest rate, government securities, corporate bonds and money market instruments. There is a need for consolidation and building on the existing reporting arrangements of CCIL while

ensuring that the governance issues emanating from CCIL acting as both, a trade repository as well as a CCP, are addressed.

Stress testing
Two new stress testing tools were added to the set of techniques used in the previous FSR. Banking Stability Measures in the form of Banking System's Portfolio Multivariate Density (BSMD) approach for analysing financial stability from different combinations of distress dependencies infers that during the periods of crisis, the systemic risks rise faster than individual risks. Vector Autoregression (VAR) approach for judging the resilience of banking sector on various macroeconomic shocks by capturing the interaction among macroeconomic variables and banks' stability variables, shows that interest rate had the most significant (negative) impact on slippage ratio of the banks. The resilience of the projected balance sheets of the commercial banks was studied through stress testing, in respect of credit risk, interest rate risk and liquidity risk. Under stress conditions based on NPA shocks, the profitability of the banks was seen to be affected significantly though the capital adequacy position appeared to be reasonably resilient. The study indicates that some banks may face extreme liquidity constraints, under severe stress scenario. Overall, the results of the macrostress tests using different scenarios, suggested that the banking sector would be able to withstand macroeconomic shocks though the prevailing inflation and interest rate situation is expected to have an adverse effect on asset quality of banks.

Financial market infrastructure

The operational performance of the payment and settlement infrastructure in India continued to be robust though vulnerabilities could emerge from the high degree of integration and interrelationships among systems, processes and institutions involved in various segments of the payment and settlement systems. The benefits from



July 2011

Technical Computing in Risk Management

Steve Wilcockson and Michael Weidman
As the importance of enterprisewide risk management systems increases, it is useful to review the value of integrated technical computing platforms that offer development of custom analytics, flexible deployment and distributed computing capabilities. This article builds on a credit risk case study (an abridged version is available as a web case study [2]), to highlight themes relating to integration, deployment and distributed computing for risk platforms. Automate a workflow to reduce waste whenever data, models or assumptions change; and Have web- or spreadsheetbased front-ends for final reports. In the case study, we demonstrate how the team can use a particular functional programming environment such as MATLAB, as an integrated technical computing platform to perform these tasks. An integrated platform here means a unified platform where different teams can share analyses, data and models; but also can integrate with existing data warehouses, with other computational engines, and many front-end interfaces (spreadsheets, web browsers, etc.). We use the term technical computing to emphasise a key philosophy of using numerical and scientific coding to build and customise analytics and models. To help implement the in-house credit rating system, the team uses a database with a long list of observations of financial ratios and corresponding credit ratings. The database contains the same financial ratios as in Altmans zscore [1], with the ratings assigned by a consultant. They use the information in the database to train a classifier, where given financial ratios of companies are used as predictors, and the credit rating is the response. The team is not interested in a credit score in the case study (as in [1]). They want to link the financial ratios directly to credit ratings, both because their final goal is to use the ratings for the credit VaR estimation, but also because they have no credit score information in their historical database, just credit ratings. In the case study, the team fits a simple regression model first, but then tries a sophisticated statistical learning tool, bagged decision trees, a decision-tree-based classifier (more on this later). The classifier performs considerably better than the regression model in the web case study. The classifier can then assign ratings to new issuers or update ratings of issuers already present in the portfolio. The team might have considered other tools, say discriminant analysis (as in Altmans original work [1]). A credit committee may also review and approve some or all credit ratings. Though reduced in scope, the workflow in the case study illustrates the convenience of working on a platform that lets users read data from multiple data sources, and ideally offers ready-to-use validated (and openly viewable) algorithms to perform certain tasks. Regarding the estimation of risk parameters, the case study specifically details the estimation of historical transition probabilities. This emphasises the collaboration between team members that determine the credit ratings, and those who use them later to estimate historical transition probabilities. Indeed, the time series of credit ratings assigned using, for example, the credit rating models described above, is the main input for the estimation of historical transition probabilities. When using the same platform, analysts across

A credit VaR example on an integrated platform

Throughout this section, we review the example specified in the web case-study, highlighting specific workflow and integration aspects in it. We assume the persona of a fictional risk management team that needs to estimate a credit value-at-risk (VaR) measure for a bond portfolio. From a computational perspective, the team needs to assign internal credit ratings for all issuers, they need to estimate transition probabilities and other risk parameters, and they need to estimate a loss distribution to determine the credit VaR. Operationally, the team needs to Read in data from databases, flat files, and spreadsheets; Clean and preprocess data; Share results and discuss inputs and outputs with other members of the team; Test different models and assumptions;



July 2011

Figure 1. Schematic diagram of the case study.

benefits of a framework for sharing results with others in the company; see Figure 1. The final step leads us to consider deployment.

The need for flexible deployment

Beyond the workflow elements illustrated in the case study, there are advantages offered by the integrated technical computing platform. These engage other areas of the institution accessing data and calculation engines developed by the risk management team. Take the credit ratings, for example. Credit ratings assigned by an automated classifier may require review or approval from a credit committee, especially for large transactions. To make things efficient, the automated classification tool could be deployed using a web server. Credit committee members can open a web browser, enter information of a new customer, or extract information about an existing customer whose rating is under review, and get a comprehensive report on the automated rating and other relevant information. The classifier in the case study for example could report more than one possible rating for an issuer, complemented with a classification score, a measure of the certainty of the classifier about each possible rating. Once the committee determines the credit rating, they can use the same web application to enter the new credit rating into the system. Both the committee and the credit rating team get consistent financial ratio information from the same database, work with the

teams can more easily share information. Upstream updates are picked up downstream with greater ease and efficiency (as compared to having to export datasets, transfer files, etc.). The primary output is a single transition matrix, an input for a third team simulating credit-rating migrations. We will revisit other outputs of interest for other parts of the institution later. For the estimation of the credit VaR, the fictional team uses a standard simulation-based approach. The methodology consists of generating credit rating migrations over the time horizon, one year in the case study, and valuing the portfolio under each simulated scenario. Credit-rating downgrades cause a bond to lose value; upgrades have the opposite effect. Thus, there is a portfolio value for each scenario. When simulating a large number of scenarios, say, 10,000, one gets a simulated empirical distribution of the possible values of the portfolio, helping estimate the expected loss and the credit VaR of the portfolio. In the case study, we use readily available tools and functions to implement the simulation, and determine how to estimate the credit VaR. Rather than

elaborating on the technical methodology, let us emphasise some workflow issues. First, team members need information on the actual bond portfolio, loaded from a spreadsheet, but it could be loaded from any data source. They also use as inputs parameter estimates obtained by other team members, such as transition probabilities. The results of the simulation, namely expected loss and credit VaR of the portfolio, need to be shared with other areas of the institution. In the case study, a front-end spreadsheet not only contains the results, but also allows a user to run a simulation. The link between the spreadsheet and the simulation engine is created with a few simple steps. We will discuss additional use cases of reporting and deployment in the following section. Despite the somewhat linear workflow (first determine credit ratings, then estimate transition probabilities, then estimate credit VaR), the case study highlights certain advantages of integrated platforms, specifically the comfort of reading in data from different data sources, the convenience of sharing intermediate information between team members, and the



July 2011

same, up-to-date automated classifier, and as soon as the committee enters a new rating into the system, this new piece of information is available to the credit rating team. For smaller transactions, credit committee review may not be required, so the automated credit rating may suffice. In this case, account executives may have access to the same web form, or a limited version of it, and assign a credit rating automatically to new or existing customers. Regarding estimated risk parameters, such as transition probabilities, the situation is similar. Departments elsewhere, other than the team supporting the credit VaR engine, can benefit from access to risk parameters. Regulatory reporting, for example, will benefit from periodic updates on key risk parameters. Their workflow can be more efficient and less error-prone if they can run a report using a custom application interface providing onthe-fly analytics, instead of receiving periodic e-mails with file

attachments. Better still, report generation could be exploited to automate the generation of regulatory reports a necessary update upstream can propagate downstream with a few clicks. There is no need for painful cut and paste. New regulatory requirements can be added, as needed, in the report generation tool, thus reducing human intervention leaving more time for analysis and communication. Others may benefit from the credit VaR engine, perhaps the aforementioned regulatory reporting team, senior management performing what-if analyses, or a portfolio manager considering a new, large transaction wanting to understand its impact on the portfolios credit VaR. A deployed application where the portfolio manager can load the portfolio information and scenario test with the proposed transaction at the click of a button using up-to-date risk parameters is a valuable capability. Figure 2 extends Figure 1 to include other areas that might

utilise the same integrated platform by means of flexible deployment.Not only are all of these solutions possible, many customers are already implementing them (see, for example 7). It is certainly a long term investment, but is worthwhile as such infrastructures facilitate traceability, easy customisation and allow institutions to evolve their risk architectures.

The value of distributed computing

A valuable addition to the integrated platform is a distributed computing infrastructure. It can help speed up computation, add robustness to models and estimates, and provide greater accuracy to model output. Let us return to the development of an automated credit rating tool. The team in the case study chose a sophisticated statistical-learning tool, bagged decision trees, as the automated credit rating tool. The team tries this technique because it is readily available and easy to use, but it also outperforms their alternative simple regression model. Bagged decision trees can be a very robust predictor. Bagging (and we will get a little technical here) is an acronym for bootstrap aggregation, and bootstrap is used in the statistical sense (randomly sampling with replacement from a dataset). To fit the bagged decision trees, many bootstrap replicas of the dataset are generated, and one decision tree, a classifying tool of its own, is grown on each replica. A bootstrap replica is created by randomly selecting N observations

Figure 2. Extended diagram with deployed applications and multiple users.



July 2011

with replacement out of the N original observations in the dataset. To find the predicted rating for a new observation, one prediction is made with each individual tree, and the rating assigned to the new observation is the rating predicted by the largest proportion of individual trees. Beyond the technicalities, the underlying principle is relatively simple: This is a substantial whatif analysis. How would a different dataset influence my model or my estimates? Repeat this many times, perhaps focusing on the variabilities of estimated parameters or the predictions of fitted models. Alternatively, you may want to forecast, as with bagged decision trees, taking a

An important characteristic is that the what-if analysis can be performed in parallel. By design, each what-if scenario is independent, and the individual models can be distributed. Many off-the-shelf models (such as the bagged decision trees in the case study) have built-in support for distributed computing. However, many statistical algorithms applied in risk management can benefit from distributed computing, adding model robustness and speeding up computations. Two common, obvious uses of distributed computing include speeding up computations over a large portfolio, and running Monte-Carlo simulations. In the former, different segments of a

magnitude in some cases. After the serial code is optimised, distributed computing considerations arise, for example, the communication overhead of sending a job to a different node and retrieving the outputs. In a simulation, for instance, it may be worth running a large group of scenarios per node, and the communication cost may be reduced by generating the scenarios inside the same node instead of creating scenarios in one lab and sending to another. Ideally, your platform should handle simpler parallelisation instances for you, but give you tools to handle jobs and schedules when necessary. Another, less obvious benefit of distributed computing is in evaluating the accuracy of the results. Monte-Carlo simulations are random experiments. If one gets a credit VaR of, say, 8.1472 percent of the portfolio value, how confident are we in its accuracy? If we repeated the simulation, might the new estimate be closer to 9 percent? The larger the number of scenarios, the lower the variability of the results; but the variability never disappears. So, how many scenarios are necessary to achieve a desired accuracy level? One way to measure the variability of results is by repeating the simulation many times (say 100 times), using different numbers of scenarios (say, 10,000, then 100,000, then 1 million, etc.). You can then measure the range of variation for each number of scenarios (eg, with a standard deviation, or a percentile interval) and determine when the range is within the desired accuracy level. This is an example of a two-level simulation. In the outer level, we

When using programming languages, code optimisation usually through profiling should be a first port-of-call. Code optimisation can improve execution time, by orders of magnitude in some cases.
range of fitted models into consideration, and make a final forecast by averaging predicted values, or by taking the value most often predicted by individual models, or any other reasonable aggregation procedure. You might also want to measure the prediction error, the main goal of a related procedure called cross validation. This can make your predictions more robust, less sensitive to a particular dataset, and to better understand their potential variability when the data changes (4) and (6). portfolio can be evaluated on separate processors. Different scenarios of a Monte-Carlo simulation can be generated and used on different computational nodes. However, parallelism and distributed computing does not always speed up code that can conceptually be executed in parallel. When using programming languages, code optimisation usually through profiling should be a first port-of-call. Code optimisation can improve execution time, by orders of



July 2011

repeatedly run simulations to get many observations of the credit VaR and estimate its variability. In the inner level, a simulation is run to get one particular estimate of the credit VaR. This is relatively simple case of a two-level or nested simulation, solved efficiently in a distributed environment. Implementation variations can help further. The literature on two-level or nested simulations can suggest other approaches and some more complex cases (see, eg, 3 or 5). When discussing deployment earlier, we presented isolated examples of users requesting information, generating reports, or running ad-hoc analyses ondemand. In reality, of course, many users across the institution perform these tasks at the same time, as illustrated in Figure 2. Distributed computing too can help. Different portfolio managers, may be in different geographies, may want to run VaR analyses several times a day. If the platform is integrated with a cluster or cloud, jobs can run simultaneously. It is possible from a single platform to support multiple custom uses. Where this works well, it can streamline an institutions operations.

the importance of distributed computing capabilities in an integrated enterprise-wide risk management system, and how it could be utilised by analysts, managers and developers. References 1. Altman, E., Financial Ratios, Discriminant Analysis and the Prediction of Corporate Bankruptcy, Journal of Finance, Vol. 23, No. 4, (Sep., 1968), pp. 589-609. Biography 2 Credit Risk Modeling with MATLAB, available on demand at company/events/webinars/ wbnr49601.html. 3 Gordy, M., and S. Juneja, Nested Simulation in Portfolio Risk Measurement, Finance and Economics Discussion Series 200821, Federal Reserve Board, Washington, DC, 2008. 4 Hastie, T., R. Tibshirani, and J. Friedman, The Elements of Statistical Learning, second edition, Springer, 2009. 5 Lan, H., B. Nelson, and J. Staum, Two-Level Simulation for Risk Management, Proceedings of the 2007 INFORMS Simulation Society Research Workshop, 2007. 6 Martinez, W., and A. Martinez, Computational Statistics Handbook with MATLAB, second edition, Chapman & Hall / CRC, 2008. 7 UniCredit Bank Austria Develops and Rapidly Deploys a Consistent, Enterprise-Wide Market Data Engine, MathWorks, User Story, 2009. Web: computational-finance/ userstories.html?file=45641.
Steve Wilcockson has worked for MathWorks for 14 years. He is Industry Manager for Financial Services with global accountability, ensuring industry trends in risk, trading, insurance, portfolio management, econometrics and valation are effectively incorporated into our development process. Steve holds degrees from the University of Cambridge and University of British Columbia.

Final remarks
Referencing a credit risk case study, we have reviewed some aspects of enterprise-wide risk management systems that benefit from an integrated technical computing platform. We suggested that flexibility of deployment is a key component of such a platform, as it facilitates easy incorporation of custom analytics components into a risk framework. We also highlighted

Michael Weidman joined MathWorks in 2007, working in the Application Engineering Team on computational finance applications. He is B A in Physics from Harvard University and has completed Part III of the Mathematical Tripos from DAMTP at the University of Cambridge.



July 2011



July 2011

Basel III
Navigating Business and Risk Technology Architecture Decisions
Cubillas Ding
Celent, a member of Oliver Wyman Group, recently added Basel III: Navigating Business and Risk Technology Architecture Decisions report to their series of Basel III and risk offering. Released last month, the report reveals that recent Basel III regulations on bank capital requirements will represent one of the largest drivers of industry economics for the banking world in this coming decade. Celent examines the key elements required by banks to ensure a sufficient level of preparedness by comparing the current and emerging Basel regimes. The report also provides recommendations for firms to navigate this transition from a strategic business and technology perspective. This is a heavily abridged version of the 34 page report which contains 11 figures and 4 tables. Interested readers may find more details at Against a backdrop of growing uncertainty in global credit conditions and sustainable growth, the financial services landscape continues to evolve rapidly. Post-financial crisis responses continue to be characterised by issuance and implementation of numerous complex and formidable array of banking reforms, with Basel III being one piece of the jigsaw, albeit one of the most significant pieces. The report reveals that Basel III solvency and liquidity regulations will represent the largest driver of changing industry economics for the banking world for this decade. With bank capital requirements becoming increasingly onerous and additive as more types of risks and capital buffers have to be taken into account, it is expected that this will eventually have a negative impact on overall investment banking return on equity (RoE) by 3 to 5 percent. As firms navigate this regulatory chessboard of options and rules, they need to bear in mind business and risk management imperatives. Increasingly, risk resources (capital, liquidity, and talent) are expected to remain key items. The industry is likely to move towards a more integrated operating model involving the risk, finance, and treasury functions. It is imperative for firms to chart out a risk technology roadmap by sound guiding principles and lessons learned from previous Basel II and risk technology implementation to navigate this transition from a strategic business and technology perspective. Basel Committee to safeguard financial stability and address the gaps in Basel IIgaps that were exposed by the credit crisis. Figure 1 highlights the evolution of the Basel framework and a summary of what each version entails. First-wave firms that have executed their Basel II programmes and implemented their IT systems in a sustainable manner are likely to find an easier path to Basel III. Conversely, firms that have taken inappropriate shortcuts or selected an incomplete solution for Basel II may face an uphill battle to pull

Basel banking regulations Looking back, surging forward

Basel III essentially stems from the resolve by the Bank of International Settlements (BIS)

Figure 1: The Basel Capital Chessboard

Source: Celent



July 2011

all that is required together in a coherent manner for Basel III. Integrated risk management solution suites are emerging to support not only regulatory aspects but also advanced capabilities. Table 1 presents a list of Basel II and Basel III risk solution vendors based on the market research conducted by Celent.

Table 1: Basel II / Basel III Risk Solution Vendors (Representative)

Business Imperatives
From a business perspective, to ensure a sufficient level of preparedness at this early juncture, financial institutions can look to address a number of issues. First, banks need to execute a comprehensive health check of their risk infrastructure making significant changes to their risk data, models, and analytics frameworks. The issues that must be addressed include: The definition and treatment of additional risk factors to be captured in the valuation and risk engines; Risk model adjustments to enhance back-testing capabilities; and Risk analytics, potentially upgrading the internal dashboards and setting more forward-looking metrics to improve capital forecasting. Once the comprehensive risk framework is designed, they should draw out a clear mapping of trading and banking risk capital to understand capital hot spots and utilisation by business line, desk, and product. Thirdly, banks must ensure that boardlevel risk reporting and oversight are connected across
Source: Celent Analysis

the firm. Accurate, timely, and comprehensive reports will help the risk committee to evaluate the current and projected risks that institution may be exposed to, including stressed conditions. This will allow the management to collectively manage risks, as well as provide assurance of a comprehensive risk framework across the lending, trading, and investment value chain. Lastly, banks need to redefine the strategic portfolio of businesses in the light of changing industry economics as defined by Basel IIIs new

provisions. The changes to trading risk capital are likely to have knock-on effects throughout the mechanisms used to run and steer the bank. This would require banks to discipline all businesses, especially corporate derivatives, based on which client segments they target.

Technology imperatives
From a technology perspective, considerations for architectural capabilities should bear in mind established guiding principles, emerging requirements, and best practice capabilities from a firm



July 2011

Table 2: Selection ConsiderationsBasel Risk Management Applications

wide perspective rather than a departmental standpoint. Celent observes number of instances based on enterprise risk technology initiatives in line with an end-state architecture. We discuss these briefly below. Design-in robust reconciliations applications in order to support comprehensive data quality processes and measures. Aim for a regulatory capital calculator that facilitates seamless migration between various Basel regimes.

Strengthen the firms model management framework by building a streamlined process that can deliver robust quality control around the entire model lifecycle. Ensure that integrated stress and scenario analysis capabilities effectively undergird a scenario-based planning framework. Design and implement a management scorecard of health indicators to dynamically link complete risk and finance data. A robust general risk framework and in time

reporting will result in effective firm wide governance. Adopting right technology for liquidity management. Databases, aggregation, and middleware need to increasingly support a dynamic or near real time flow of data that can give a live snapshot of the state of affairs. In the long run, the emerging best practice capability is to achieve a unified/aligned risk and finance data model. This includes consolidating fragmented risk, finance, and treasury applications along with



July 2011

designing a single risk and financial performance view of the customer. Financial institutions which have yet to make investments need to exercise caution to ensure that their Basel II / Basel III and broader risk software purchases and projects benefit from the hindsight provided by early adopters. Table 2 points out the key considerations that firms need to assess before structuring their Basel risk management framework, which will also help them in selecting the right technology and solution vendors.

be charted and planned at an early stage. Banks, particularly, will need to assess various architecture options considering how the risk management function will operate in the next five years and also to what extent each option will facilitate increasing collaboration between strategic planning, treasury, capital management, finance, and risk groups. Overall, firms need to have a clear view of what ambitions it wants to achieve before building, selecting, or blending Basel toolsets. This will involve designing a risk IT architecture which delineates policies, rules, guidelines, and standards for the technical layers to support the evolutionary path towards end game risk management capabilities.

Road ahead
In order to avoid costly mistakes and dead ends, an institutions risk technology roadmap should

in brief
Capgemini, a consulting, technology and outsourcing services provider, through its Financial Services global business unit, and Murex, a front, middle, back-office and capital markets risk management specialist, have signed a global partnership. With revenues of more than 1.6 billion in the financial services sector, and over 17,000 experts worldwide, Capgemini is pursuing its development strategy in capital markets activities through partnerships. This responds to a sharply increasing demand from financial institutions for the industrialisation and package development of their IT systems. This global agreement positions Capgemini as a reference integrator in financial package services. It provides a significant enhancement of the technical and functional skills for Capgemini consultants, in the Murex package suite. Building on over 25 years of successful presence in capital markets, Murex has developed competence in the design and implementation of integrated trading, risk management and processing solutions for top financial institutions, clearing houses, corporations and utilities located across the globe. Its 200 clients range from leading market makers to large-sized or medium-sized buy-side and sell-side institutions. Over 36,000 users rely on MX.3, the latest Murex platform. Implementations powered by the MXpress approach leverage the wealth of business content accumulated by Murex over the two decades through pre-packaged components of the platform while offering an accelerated process of delivery. The implementation strategy of Murex products is centered around certified partnerships. Biography
Cubillas Ding is a research director in Celent's securities and investments practice and is based in the firm's London office. His expertise lies in global financial markets, securities IT strategy, and ERM. Before joining Celent, Ding was a senior analyst at Datamonitor. Prior to this role he held positions at Euro RSCG Circle as a business consultant, at Hewlett Packard European Labs & Direct Marketing Association as a lead research analyst, and at Accenture's Financial Services Group as a consultant. Ding received a master's degree in international business from the University of Bristol and a B Sc in computer science from Monash University in Australia.



July 2011

Dare to Lead
The Transformation of Bank of Baroda Author: Dr Anil K Khandelwal

Publisher: Sage Publications India Pvt Ltd, B1/1-1 Mohan Cooperative industrial Area, Mathura Road, New Delhi-110044, India. Website: Published in 2011 Pages: 403



July 2011


Risky Business for Hedge Funds When Selecting a Business-Critical System
Brian Roberti
The most effective way for Hedge Funds to mitigate selection risk when deciding on business-critical systems would be a miraculous ability to time-travel. If only a Fund could have the foresight to anticipate the future needs of the company and resultant gaps in tomorrows software solutions. How about being able to look forward to a time when the system will be truly entrenched within your business, having that aha moment and realising the impact and pain of having to switch to a different platform or to prematurely upgrade the existing one to support business growth and new strategies. Conversely, the ability to travel back in time to when you made a particular decision, to recognise exactly why you made that choice, and be able to defend it on the strength of a well-documented procurement process with buy-in from key stakeholders would prove invaluable for offensive and defensive reasons. Not a bad super-power to have as far as supernatural abilities go. When Hedge Funds shop for business-critical systems, the first step in mitigating selection risk should be to conduct a thorough evaluation of requirements. This analysis typically breaks down into four key categories namely functionality, operational efficiency, ability to integrate with third-party applications and counterparties, and ability to meet local reporting and regulatory obligations. To help properly prepare for the system evaluation step, it is best to assume vendors all too often over-promise and under-deliver. In order to debunk their we do everything myth, requirements are best framed within an evaluation matrix, specifying desired systems components weighted by priority, and then considered in respect to what is supplied out-of-the-box. It sounds simple, but ensuring that these prioritised requirements are sufficiently addressed, maintained, and adaptive is fundamentally important, and provides the necessary flexibility as business needs change. Followthrough is called for here -theres no point in framing this all out if theres no intention to stick with a disciplined due diligence process from beginning to decision to implementation to production. Naturally, this process must involve key stakeholders from across the business functions representing technology, operations, and accounting all of whom will have their own set of priorities. By considering their different views to identify key requirements as well as potential issues, the Hedge Fund will create a more complete picture of present needs, making it easier to predict future firm-wide concerns while also helping to identify and mind the gaps. Employing the services of a consultative third party can also add value, providing an arbitrational and objective view that takes much of the emotion out of the process. The benefit of this type of approach is the ability to address risk concerns across functional areas while promoting increased operational efficiencies, better workflow and systems integration, and enhanced reporting delivery. Once the requirements have been framed and documented and the potential best-fit solutions identified, the next step is to engage and interact with prospective systems providers. Any gaps in vendor solutions must be identified, and build-out versus buy options should be considered to address these limitations. There are typically two delivery options on-premise software or hosted (SaaS)/managed services (aka BPO). The latter can provide a pragmatic alternative in scenarios where it is preferable for staff expertise to remain with the outsource provider. Discussions with the system providers must also cover implementation considerations including any required data conversions, the capacity and capabilities for customising the system, and service levels especially for production support. After all is said and done, the total cost of the system will be much more than the acquisition and initial implementation costs, especially if you do not have a holistic, forward-thinking approach that considers what is needed as the business scales. Indeed, the final piece to the selection risk puzzle is TCO (total cost of ownership). Certainly, it is necessary to be wary of systems pitched as the low-cost alternative, recognising that TCO entails both the acquisition of the platform and the ongoing cost of maintenance,



July 2011

support and future upgrades. The former can literally be crushed by the latter, particularly if the system proves inflexible and unable to adapt to the everevolving business and regulatory environment. As internal and external pressures exert the need for change in systems over time, the solution employed will need to be revisited and customised to the firms unique business objectives and exposure to various aspects of risk. There are clear benefits to having a system that can be tailored, at least to some extent, by the power users versus having to deploy IT resources that a firm may or may not have dependent on their size and make-up. Hedge Funds need to look to the future when undertaking the initial analyses of and decisions based upon system risk. There is a lot to be said for the use of simple tools to get a job done. On the flip side, Hedge Funds that have launched using Excel as their cornerstone system are all too

quickly confronted with Excels limitations in handling the complexities of financial securities and the requirements of an enterprise class solution and too often wait to make a change until the Fund reaches a critical point where the system falls over, becomes too unwieldy or simply does not have what is required to adequately support business operations, not to mention compliance and regulatory requirements. Typically the deeper and farther one goes with a system, the greater and more painful are the switching costs. If managed poorly, it can even put business-critical processes in jeopardy. Therefore, it is advisable to take a consultative, thoughtful approach when making the initial system decision. When it comes to the risky but necessary business of system selection, your starting point has a huge impact on the path you may ultimately travel and consequently where your business has the potential to go. There is no one size fits all. Biography
Brian Roberti is director, sales and marketing at G2 Systems.He has held sales leadership positions at several financial technology firms offering automated trading platforms and services. He successfully grew the sales team at EdgeTrade an agency-only broker and provider of electronic trading and execution services and helped the firm achieve its aggressive revenue growth targets and exit strategy. Prior to that, he worked for 10 years at Advent Software, a leading provider of investment management software, where he served in a number of market-facing management positions spanning Sales Director and VP Strategic Partnerships. Earlier in his career, Brian served in IBM's Wall Street practice, where he managed global account teams servicing large sellside institutions. He holds a BS in Mathematics from the State University of New York in Albany and NASD/FINRA Series 7, 63 and 24 licenses.

in brief
Aite Group's Global Anti-Money Laundering Vendor Evaluation: A Reinvigorated Market ranks SAS Anti-Money Laundering, as the top anti-money laundering (AML) software provider. Aite Group interviewed 36 financial institutions and 18 leading vendors in the global AML space between January and April 2011. Financial organisations spanned five continents and ranged from $800 million to more than $1 trillion in assets. SAS Anti-Money Laundering is part of the SAS Enterprise Financial Crimes Framework for Banking, a technology infrastructure for preventing, detecting and managing financial crimes across lines of business within today's banks. More than 114 institutions use SAS for anti-money laundering. SAS AntiMoney Laundering harnesses analytics to provide more effective alert processing and offers massive cost savings to banks by reducing false positives and freeing investigative resources, observes the report. The report specifies that AML solutions should include customer due diligence, suspicious activity monitoring, case management and watch-list filtering, all of which are found in SAS Anti-Money Laundering.



LOOK FOR US wherever financial decisions are made. We provide software and technology services that account for and manage more than $25 trillion in investment assets and process more than 5 million trades every day, and we never stop working to improve the software that underpins the worlds financial markets. For further infor mation please contact us at or call us on +91 22 3098 1000.


2010 SunGard, all rights reserved. Trademark information: SunGard, the SunGard logo and the products and services of SunGard SunGard, reserved. Trademark information: SunGard, SunGard products SunGar named herein are trademarks or registered trademarks of SunGard Data Systems Inc. or its subsidiaries in the U.S. and other countries. herein are register registered egistered SunGard