cro-july-11

8/2/2019 CRO-July-11

1/28

Banking, Financial Services & Insurance

The Journal of

Compliance

Risk &

Opportunity

VOL V Issue 9July 2011

Technical Computing

in Risk Management

Basel III:

Celent Report

HDFC Bank Upgrades to

TCS BaNCS Treasury 5.0

FINANCIAL STABILITYRBI Publishes its Third Report

8/2/2019 CRO-July-11

2/28

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

2nd AnnualRetail Banking Asia Pacific

T: +603 2141 5352 E: [email protected]

Finsights readers benefit from 10% off the registration fee

www.flemiggulf.com

REGISTER & PARTICIPATE

to enhance profitable retail bankingstrategies

Discover & ShareINNOVATIONS! Sharp minds. Sharp solutions.

EYE ON CUSTOMERS

1 ESSENTIAL EVENT

2 Days of Innovative Ideas

5+ Bespoke Networking Hours

12+ Expert Speakers

17+ Resourceful Topics

What do customers really want with their bank?

What motivates customers to stay loyal?

How do innovation help?

Research Partner:Official Travel Partner: Organised By:

Media Partners:

Media Sponsor:

ravel Partner:Official T

Media Partners:

h Partner:Resear Sponsor:Media Organised By:

8/2/2019 CRO-July-11

3/28

J u l y 2 0 1 1

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 StabilityFacility 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.

Finsight MediaTF-07, Suner Complex

Harinagar Crossing, Gotri Road

Vadodara 390 021

Tel: +91-265-6533475

Email: [email protected]

Website: www.finsight-media.com

Publisher

T S Chandrasekaran

[email protected]

Editor

Hari Misra

[email protected]

Advertising

Amita [email protected]

Production

Sanket Vohra

[email protected]

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

Price Per Copy: INR 50Annual subscription: INR 600

Hari Misra

8/2/2019 CRO-July-11

4/28

HDFC Bank Upgrades to 6


iCREATE Launches Automated 8

Regulatory Reporting Solution

for Banks

FINANCIAL STABILITY 10RBI Publishes its Third Report

Technical Computing 14

in Risk Management

Basel III 20Navigating Business and Risk Technology

Architecture Decisions

in brief 23

ONE SIZE DOESNT FIT ALL 25Risky Business for Hedge Funds When Selectinga Business-Critical System

in brief 26

CONTENTS

8/2/2019 CRO-July-11

5/28

Scan the QR code* with your mobile device to view abrief customer video or visit sas.com/india/bankfraud for the complete success story video.

Stamp out fraud.

With SAS Analytics, you can score millions of transactions a day in real time to detect fraud faster, reducerisk, streamline investigations and prevent losses. Decide with condence.

ANALYTICS

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

*Requires reader app to be installed on your mobile device

For more information [email protected].

8/2/2019 CRO-July-11

6/28

COMPLIANCE, RISK & OPPORTUNITY6

J u l y 2 0 1 1

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, mostuseful 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 whichsupports multi-entity, multi-

currency, 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 transparencyin identifying the source of risk

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 back-

office 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 openframework 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 beenexpanded to cover synthetic curve

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 front-

office perspective or the back-

office accounting perspective, he

says.

The personalised single

workspace which Shetty likes,

provides the user with a capabilityto 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 marketingstrategies. 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 desk-book-folder approach and rich

HDFC Bank Upgrades to


Harish Shetty

8/2/2019 CRO-July-11

7/28

COMPLIANCE, RISK & OPPORTUNITY 7

analytics. The hierarchical desk-

book-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 arobust 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 isanother 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-breedpoint solutions do provide rich

functionality; integration,

standardisation, support

availability, reduced straight-

through-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 replacementmarket 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 existingdual currency systems of Indian

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. India-

based 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 profitable-

managing banks own treasury or

offering treasury products tocorporates? 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 integrationwith other systems.

J u l y 2 0 1 1

NG Subramaniam

8/2/2019 CRO-July-11

8/28

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 coregroup 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 thistransparency?

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 regulatoryreporting solution with advanced

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 ofhelping 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. Wehave 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 specificcustomisations as well. To that


J u l y 2 0 1 1

iCREATE Launches Automated

Regulatory Reporting Solution for

Banks

Vivek Subramanyam

8/2/2019 CRO-July-11

9/28

J u l y 2 0 1 1


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 toassess 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 in-

house 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 andquality.

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 tothe 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 thedata 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 thesystem 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 thetechnology 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 theflavour of the solution.

8/2/2019 CRO-July-11

10/28


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 financialstability 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 developingeconomies 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 itsimpact 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 domesticindustrial growth, uncertainty

FINANCIAL STABILITYRBI 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

http://www.rbi.org.in/scripts/PublicationReportDetails.aspx?UrlPage=&ID=635--Ed.

8/2/2019 CRO-July-11

11/28


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 maynot 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 incapital 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 therevenue 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 marketsegments 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 havehigh 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 acrossmarkets.

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 marketsegment 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 periodfrom 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 ratesprevalent today.

J u l y 2 0 1 1

8/2/2019 CRO-July-11

12/28


J u l y 2 0 1 1

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 limitsfor FII investment in corporate and

government bonds. An internal

study points to tendency of the

portfolio capital flows to be 'auto-

correlated' 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 beguarded against.

Banks

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

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 theregulatory 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 thebanking 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 theprofitability. 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 gearthemselves 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 dueto the unamortised portion of

8/2/2019 CRO-July-11

13/28


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 quantitativeindicators 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.

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 inter-

relationships among systems,

processes and institutions

involved in various segments of

the payment and settlementsystems. The benefits from

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 arevulnerabilities 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 introductionof 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 reportingarrangements 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 differentcombinations 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 bereasonably resilient. The study

indicates that some banks may

face extreme liquidity constraints,

under severe stress scenario.

Overall, the results of the macro-

stress 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 adverseeffect on asset quality of banks.

J u l y 2 0 1 1

8/2/2019 CRO-July-11

14/28


J u l y 2 0 1 1

As the importance of enterprise-

wide 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 aweb case study [2]), to highlight

themes relating to integration,

deployment and distributed

computing for risk platforms.

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 andassumptions;

Automate a workflow to reduce

waste whenever data, models or

assumptions change; and

Have web- or spreadsheet-

based front-ends for final

reports.

In the case study, we demonstrate

how the team can use a particular

functional programmingenvironment 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 z-score [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 becausetheir 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 classifierperforms 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 transitionprobabilities. 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 thesame platform, analysts across

Technical Computing

in Risk Management

Steve Wilcockson and Michael Weidman

8/2/2019 CRO-July-11

15/28

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 enginesdeveloped 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 fromthe same database, work with the


J u l y 2 0 1 1

teams can more easily shareinformation. 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 eachscenario. 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 thecredit VaR. Rather than

elaborating on the technicalmethodology, 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 followingsection.

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 informationbetween team members, and the

Figure 1. Schematic diagram of the case study.

8/2/2019 CRO-July-11

16/28


J u l y 2 0 1 1

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 newor 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 on-

the-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 thecredit 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 distributedcomputing

A valuable addition to the

integrated platform is a


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 byrandomly selecting N observations

Figure 2. Extended diagram with deployed applications

and multiple users.

8/2/2019 CRO-July-11

17/28


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 what-

if analysis. How would a differentdataset 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

range of fitted models into

consideration, and make a final

forecast by averaging predicted

values, or by taking the valuemost 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 datachanges (4) and (6).

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 upcomputations.

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

portfolio can be evaluated on

separate processors. Different

scenarios of a Monte-Carlo

simulation can be generated andused 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 improveexecution time, by orders of

magnitude in some cases. After

the serial code is optimised,


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 thesimulation 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-levelsimulation. In the outer level, we

J u l y 2 0 1 1

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.

8/2/2019 CRO-July-11

18/28


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 on-

demand. 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.

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 riskframework. We also highlighted

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.

2 Credit Risk Modeling with

MATLAB, available on demand at

http://www.mathworks.com/

company/events/webinars/

wbnr49601.html.

3 Gordy, M., and S. Juneja,

Nested Simulation in Portfolio

Risk Measurement, Finance and

Economics Discussion Series 2008-

21, 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:

http://www.mathworks.com/

computational-finance/userstories.html?file=45641.

J u l y 2 0 1 1

Biography

Steve Wilcockson has worked forMathWorks 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.

Biography

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.

8/2/2019 CRO-July-11

19/28


J u l y 2 0 1 1

8/2/2019 CRO-July-11

20/28


J u l y 2 0 1 1

Against a backdrop of growing

uncertainty in global creditconditions 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 overallinvestment 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 integratedoperating 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 banking regulations

Looking back, surging forward

Basel III essentially stems from the

resolve by the Bank of

International Settlements (BIS)

Basel Committee to safeguard

financial stability and address thegaps 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 IIINavigating 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

http://celent.com/node/28834.

Figure 1: The Basel Capital Chessboard

Source: Celent

8/2/2019 CRO-July-11

21/28


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.

Business Imperatives

From a business perspective, toensure 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

infrastructuremaking 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 moreforward-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

capitalto understand capital hot

spots and utilisation by business

line, desk, and product. Thirdly,

banks must ensure that board-

level risk reporting andoversight are connectedacross

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

businessesin the light of

changing industry economics asdefined 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 requirebanks 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 bestpractice capabilities from a firm

J u l y 2 0 1 1

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

Source: Celent Analysis

8/2/2019 CRO-July-11

22/28


wide perspective rather than a

departmental standpoint.

Celent observes number of

instances based on enterprise risk

technology initiatives in line withan 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 betweenvarious 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 riskframework and in time

reporting will result in effective

firm wide governance.

Adopting right technology for

liquidity management.

Databases, aggregation, andmiddleware 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, andtreasury applications along with

J u l y 2 0 1 1

Table 2: Selection ConsiderationsBasel Risk Management Applications

8/2/2019 CRO-July-11

23/28

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 toachieve 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.


J u l y 2 0 1 1

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 managementframework, which will also help

them in selecting the right

technology and solution vendors.

Road ahead

In order to avoid costly mistakes

and dead ends, an institutions

risk technology roadmap should

Biography

Cubillas Ding is a research director inCelent'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.

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 implementationstrategy of Murex products is centered around certified partnerships.

in brief

8/2/2019 CRO-July-11

24/28


J u l y 2 0 1 1

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: www.sagepublications.com

Published in 2011

Pages: 403

Dare to LeadThe Transformation of Bank of Baroda

BOOK REVIEW

8/2/2019 CRO-July-11

25/28


J u l y 2 0 1 1

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 thesystem 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 thoroughevaluation 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 bestto 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 aresufficiently addressed,

maintained, and adaptive is

fundamentally important, and

provides the necessary flexibility

as business needs change. Follow-

through 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 topredict 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 increasedoperational 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-outversus 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 aholistic, 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 andthe ongoing cost of maintenance,

ONE SIZE DOESNT FIT ALLRisky Business for Hedge Funds When Selecting

a Business-Critical SystemBrian Roberti

8/2/2019 CRO-July-11

26/28

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 asystem, 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.


J u l y 2 0 1 1

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 ever-

evolving 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 ofrisk. 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

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-

cro-july-11

Documents