cro-july-11
TRANSCRIPT
-
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
Editor
Hari Misra
Advertising
Amita [email protected]
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
Price Per Copy: INR 50Annual subscription: INR 600
Hari Misra
-
8/2/2019 CRO-July-11
4/28
HDFC Bank Upgrades to 6
TCS BaNCS Treasury 5.0
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
TCS BaNCS Treasury 5.0
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$core up
and running in a bank. Typical
transaction systems come in
various versions and have some
level of bank specificcustomisations as well. To that
COMPLIANCE, RISK & OPPORTUNITY8
J u l y 2 0 1 1
iCREATE Launches Automated
Regulatory Reporting Solution for
Banks
Vivek Subramanyam
-
8/2/2019 CRO-July-11
9/28
J u l y 2 0 1 1
COMPLIANCE, RISK & OPPORTUNITY 9
extent, there is a need to
customise the data adapter
framework to fit in at each bank.
However, Biz$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
COMPLIANCE, RISK & OPPORTUNITY10
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
COMPLIANCE, RISK & OPPORTUNITY 11
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
COMPLIANCE, RISK & OPPORTUNITY12
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
COMPLIANCE, RISK & OPPORTUNITY 13
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
COMPLIANCE, RISK & OPPORTUNITY14
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
COMPLIANCE, RISK & OPPORTUNITY 15
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
COMPLIANCE, RISK & OPPORTUNITY16
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
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 byrandomly selecting N observations
Figure 2. Extended diagram with deployed applications
and multiple users.
-
8/2/2019 CRO-July-11
17/28
COMPLIANCE, RISK & OPPORTUNITY 17
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,
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 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
COMPLIANCE, RISK & OPPORTUNITY18
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
COMPLIANCE, RISK & OPPORTUNITY 19
J u l y 2 0 1 1
-
8/2/2019 CRO-July-11
20/28
COMPLIANCE, RISK & OPPORTUNITY20
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
COMPLIANCE, RISK & OPPORTUNITY 21
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
COMPLIANCE, RISK & OPPORTUNITY22
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.
COMPLIANCE, RISK & OPPORTUNITY 23
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
COMPLIANCE, RISK & OPPORTUNITY24
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
COMPLIANCE, RISK & OPPORTUNITY 25
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.
COMPLIANCE, RISK & OPPORTUNITY26
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-