1.

INTRODUCTION:
1a. Meaning
Over the last few years the Indian financial markets have witnessed wide ranging
changes at fast pace. Intense competition for business involving both the assets and
liabilities, together with increasing volatility in the domestic interest rates as well
as foreign exchange rates, has brought pressure on the management of banks to
maintain a good balance among spreads, profitability and long-term viability.
These pressures call for structured and comprehensive measures and not just ad
hoc action. The Management of banks has to base their business decisions on a
dynamic and integrated risk management system and process, driven by corporate
strategy. Banks are exposed to several major risks in the course of their business credit risk, interest rate risk, foreign exchange risk, equity / commodity price risk,
liquidity risk and operational risks. Asset-liability management refers to the overall
control of the composition of balance sheet accounts to attain key objectives. These
key objectives are to generate optimum levels of quality earnings and to maintain
adequate liquidity to meet both predicted and unexpected cash needs. The
increasing volatility in funding sources and market rates resulting from the removal
of interest rate limitations and rapid fluctuations in the economy have made
effective funds management essential to successful operations.

1b. Why ALM?

In the normal course, banks are exposed to credit and market risks in view of the
asset-liability transformation. With liberalization in Indian financial markets over
the last few years and growing integration of domestic markets and with external
markets, the risks associated with banks' operations have become complex and
large, requiring strategic management. Banks are now operating in a fairly

deregulated environment and are required to determine on their own, interest rates
on deposits and advance in both domestic and foreign currencies on a dynamic
basis. The interest rates on banks' investments in government and other securities
are also now market related. Intense competition for business involving both the
assets and liabilities, together with increasing volatility in the domestic interest
rates as well as foreign exchange rates, has brought pressure on the management of
banks to maintain a good balance among spreads, profitability and long-term
viability. Imprudent liquidity management can put banks' earnings and reputation
at great risk. These pressures call for structured and comprehensive measures and
not just ad hoc action. The Management of banks has to base their business
decisions on a dynamic and integrated risk management system and process,
driven by corporate strategy. Banks are exposed to several major risks in the course
of their business - credit risk, interest rate risk, foreign exchange risk, equity /
commodity price risk, liquidity risk and operational risk. It is, therefore, important
that banks introduce effective risk management systems that address the issues
related to interest rate, currency and liquidity risks.
"Banks need to address these risks in a structured manner by upgrading their
risk management and adopting more comprehensive practice - Asset-Liability
Management (ALM) o. ALM provides a comprehensive and dynamic framework
for measuring, monitoring and managing liquidity, interest rate, foreign exchange
and equity and commodity price risks of a bank that needs to be closely integrated
with the banks' business strategy. It involves assessment of various types of risks
and altering the asset-liability portfolio in a dynamic way in order to manage risks.

2. The ALM process rests on three pillars:

a. ALM information systems
=> Management Information System
=> Information availability, accuracy, adequacy and expediency
b. ALM organization
=> Structure and responsibilities
=> Level of top management involvement
c. ALM process
=> Risk parameters
=> Risk identification
=> Risk measurement
=> Risk management
=> Risk policies and tolerance levels.

2a. ALM Information Systems

Information is the key to the ALM process. Considering the large network of
branches and the lack of an adequate system to collect information required for
ALM which analyses information on the basis of residual maturity and behavioural
pattern it will take time for banks in the present state to get the requisite
information. The problem of ALM needs to be addressed by following an ABC
approach i.e. analysing the behaviour of asset and liability products in the top
branches accounting for significant business and then making rational assumptions
about the way in which assets and liabilities would behave in other branches. In
respect of foreign exchange, investment portfolio and money market operations, in
view of the centralised nature of the functions, it would be much easier to collect
reliable information. The data and assumptions can then be refined over time as the
bank management gain experience of conducting business within an ALM
framework. The spread of computerization will also help banks in accessing data.

2b. ALM Organization

The Board should have overall responsibility for management of risks and should
decide the risk management policy of the bank and set limits for liquidity, interest
rate, foreign exchange and equity price risks.
The Asset - Liability Committee (ALCO) consisting of the bank's senior
management including CEO should be responsible for ensuring adherence to the
limits set by the Board as well as for deciding the business strategy of the bank (on
the assets and liabilities sides) in line with the bank's budget and decided risk
management objectives.
The ALM desk consisting of operating staff should be responsible for analysing,
monitoring and reporting the risk profiles to the ALCO. The staff should also
prepare forecasts (simulations) showing the effects of various possible changes in
market conditions related to the balance sheet and recommend the action needed to
adhere to bank's internal limits.
The ALCO is a decision making unit responsible for balance sheet planning from
risk -return perspective including the strategic management of interest rate and
liquidity risks. Each bank will have to decide on the role of its ALCO, its
responsibility as also the decisions to be taken by it. The business and risk
management strategy of the bank should ensure that the bank operates within the
limits / parameters set by the Board. The business issues that an ALCO would
consider, inter alia, will include product pricing for both deposits and advances,
desired maturity profile of the incremental assets and liabilities, etc. In addition to
monitoring the risk levels of the bank, the ALCO should review the results of and
progress in implementation of the decisions made in the previous meetings. The
ALCO would also articulate the current interest rate view of the bank and base its
decisions for future business strategy on this view. In respect of the funding policy,
for instance, its responsibility would be to decide on source and mix of liabilities or
sale of assets. Towards this end, it will have to develop a view on future direction

of interest rate movements and decide on a funding mix between fixed Vs floating
rate funds, wholesale Vs retail deposits, money market Vs capital market funding,
domestic Vs foreign currency funding, etc. Individual banks will have to decide the
frequency for holding their ALCO meetings.

Composition of ALCO
The size (number of members) of ALCO would depend on the size of each
institution, business mix and organizational complexity. To ensure commitment of
the Top Management, the CEO/CMD or ED should head the Committee. The
Chiefs of Investment, Credit, Funds Management / Treasury (Forex and domestic),
International Banking and Economic Research can be members of the Committee.
In addition the Head of the Information Technology Division should also be an
invitee for building up of MIS and related computerization. Some banks may even
have sub-committees.

Committee of Directors
Banks should also constitute a professional Managerial and Supervisory
Committee consisting of three to four directors which will oversee the
implementation of the system and review its functioning periodically.

2c. ALM process:

Risk parameters
Risk identification
Risk measurement
Risk management
Risk policies and tolerance levels.

3.

The scope of ALM function can be described as follows:

Liquidity risk management
Management of market risks (including Interest Rate Risk)
Funding and capital planning
Profit planning and growth projection
Trading risk management

Objective:
The primary objective of an Asset Liability Management system is liquidity risk
management and interest rate risk management.
During last couple of years, liquidity risk management did not pose a challenge
due to surplus liquidity in the banking system. The surplus liquidity parked by
banks with Reserve Bank of India touched Rs 554.75 billion in March 2004. The
deposit growth has been phenomenal whereas the credit off-take was negligible.
However, since beginning of fiscal year 2008-09, the liquidity situation has
changed.

Current Scenario

 Bank credit grew by 30.2 per cent as on January 5, 2009 as against 29.7 per
cent a year ago. Nonfood credit grew by 31.2 per cent, at the same rate as the
previous year.
Credit growth has been largely broad-based. About 34 per cent of the
incremental non-food credit was absorbed by industry, 12 per cent by
agriculture, 29 per cent by personal loans and 25 percent by 'other loans'.
Growth in aggregate deposits accelerated to 21.1 per cent as on January 5,
2009, from 16.2 per cent in the previous year. Demand deposits grew by
19.2 per cent Vis-a-Vis 28.7 per cent a year ago. However, accretion to time
deposits was significantly higher at 22.9 per cent from 15.0 percent in the
previous year.
High growth in time deposits, apart from the acceleration in economic
activity, can be attributed to higher deposit rates and tax benefits under
Section 80C for deposits with a maturity of 5 years and above.
Therefore, liquidity risk management has become increasingly a challenging task
in Indian banks. Liquidity risk is often related to banks inability to pay to its
depositors. However, a banks inability to pay to its depositors is the ultimate
manifestation of liquidity risk. Liquidity risk at initial stages may lead to distress
pricing of assets and liabilities. A bank with high degree of liquidity risk may be
forced to borrow funds from inter bank market at exorbitant rates or has to increase
its deposit rates. As the bank may not be able to transfer these increased costs to
borrowers, ultimately its net interest income shall be affected. Further, as banks
cost of funds goes up, increasingly it looks for risky avenues to increase its
earnings. The process may lead to wrong selection of borrowers as well as
venturing into risky areas (such as equity financing, giving un-secured loans etc)

increasing overall risk profile of the bank. Therefore, liquidity risk has strong
correlation with other risks such as interest rate risk and credit risk.

4. LIQUIDITY RISK MANAGEMENT

Liquidity risk is measured through either
Stock approach or
Flow approach.
Stock approach

Under stock approach certain standard ratios are computed. Some of the ratios
widely used in banks are
Liquid assets to short-term liabilities,
Core assets to core liabilities,
Inter bank borrowings to total assets,
Overnight borrowings to total assets etc.
However, management of liquidity through ratios suffers from some drawbacks, as
it does not factor market liquidity aspect of assets and liabilities. For example,
presence of some short term investments may show the improved liquidity risk of
the bank whereas the investment itself may be highly illiquid. Further, the ratio,
though good indicator of liquidity, may be valid good for a point of time only as
balance sheet profile constantly changes. Therefore, the flow approach, the
alternative model for measuring and managing liquidity has been accepted by most
of the banks.

Flow approach

Under Flow approach cash flows are segregated into different maturity ladders and
net funding requirement for a given time horizon is estimated. The net funding
requirement over a given time horizon gives a fair idea of liquidity risk faced by an
institution. Reserve Bank of India has prescribed some statutory returns for
submission of data on liquidity risk and interest rate risk by banks.
The Maturity Profile as given in Appendix I could be used for measuring the future
cash flows of banks in different time buckets. The time buckets given the Statutory
Reserve cycle of 14 days may be distributed as under:
i) 1 to 14 days
ii) 15 to 28 days
iii) 29 days and upto 3 months
iv) Over 3 months and upto 6 months
v) Over 6 months and upto 12 months
vi) Over 1 year and upto 2 years
vvii) Over 2 years and upto 5 years
iviii) Over 5 years
iThere are limits for liquidity mismatches in the first two buckets prescribed by
)RBI. The mismatches as percentage to outflows should not exceed negative
20% in the time buckets of 1-14 days and 15 days to 28 days.

The above hypothetical chart indicates liquidity risk profile of a bank. The bank
has created long-term assets through short-term liabilities as evident from negative
mismatches in the near term buckets and positive mismatches in long term buckets
of 3-5 years and above 5 years. In the process, the bank has exposed its liquidity
profile to a great degree of risk in two ways; first its liquidity position will depend
on the roll over of short term maturing liabilities and second, depending on adverse
money market conditions as well as external liquidity, it may be forced to rollover
maturing liabilities at a higher cost. In case, the long-term assets are in the form of
investments, the bank will be forced to sale assets at loss to meet its liquidity
requirements.
Generally banks perform an in-depth analysis of their liquidity profile analyzing:
1) The profile of liability holders,
2) The extent of purchased fund providers,
3) ABC analysis of depositors etc.
They also prepare scenario analysis with different degrees of rollovers and
different level of asset securitization, asset sale and other realizations to evolve the
contingency funding requirement. Ultimately a plan is prepared to meet liquidity
requirements in the event of crisis and cost thereof to meet the crisis. The Board of
directors and top management periodically review result of the exercise.
Importance of liquidity risk management has also been highlighted by Basel
Committee on Banking Supervision in the document Sound Practices of
managing Liquidity in Banking Organizations. World-wide, banking regulators
are mandating banks to conduct stress test and scenario analysis of liquidity risk
and prepare contingency plans to meet liquidity requirements in unforeseen
circumstances.
The major challenges faced by Indian Banks in liquidity management are:

1. Maturities of bank deposits are shrinking over last couple of years. This is
mainly due to narrowing of liquidity premium paid by banks for long-term
deposits. For example, interest rates on deposits (of over 1 year maturity) were in
the range of 5.25%-6.25% in March 2008 as compared to 9%-11.50% in March
1999. Therefore, a depositor has no incentive to place his money in a banks long
term fixed deposit.
2. With decline in interest rates, a sizeable portion of customers migrated to other
savings avenues like post office deposits, mutual funds etc depriving banks of long
term resources.
3. On the other hand, due to Dis-intermediation in corporate funding and recession
in past years, banks have been aggressively offering asset products with long-term
maturities (e.g. home loans). Falling interest rates and lack of credit growth in past
years prompted banks to put major portion of their deposits in long-term
government securities which may not remain liquid at all times. Further,
government securities are subject to price risk and a bank may not be able to
generate adequate liquidity without selling them at loss.

5. INTEREST RATE RISK MANAGEMENT

Interest rate risk management in banking book is a major objective of any ALM
system. Typically, a bank positions its assets and liabilities into trading book and
banking book. Interest rate movements cause price changes in trading book and
earnings/economic value changes in banking book. While trading book assets and
liabilities are held to take advantage of short term price movements, banking book
assets and liabilities arise out of relationship/regulatory requirements and provide
accrual income to a bank. The top management of banks lay down policies with
regard to volume, maximum maturity, holding period, duration, stop loss,
defeasance period, rating standards, etc. for classifying securities in the trading
book. The assets held in the trading book are marked to market on a daily basis.
The tools and techniques of risk management and the plan horizon also vary
between trading book and banking book. While risk in the trading book are
measured on a real time basis through VAR, PV01 etc, interest rate risk in banking
book is measured on a longer horizon say a quarter or a year.
A banks banking book is exposed to following types of risk:
o Mis-match risk
o Basis risk
o Embedded option risk
o Yield curve risk
o Price risk

5a. Gap or mismatch risk

A gap or mismatch risk arises from holding assets and liabilities and offbalance sheet items with different principal amounts, maturity dates or repricing dates, thereby creating exposure to unexpected changes in the

level of market interest rates. An example of gap risk is exhibited in the following
table:

(Note: the difference of Rs 600 Cr between assets and liabilities, re-present nonsensitive liabilities e.g. Capital, Reserves, Current deposits etc).
As the bank has negative mismatch in the first year (i.e. its maturing liabilities are
more than assets), its net interest margin or earnings will decline in a rising rate
scenario. The reverse will happen in a falling interest rate scenario. Following is
the relationship between a banks net interest income, gap and direction of interest
rate movements:

A bank may strategically position its gaps in anticipation of interest rate

movements to increase its Net Interest Income.

5b.Basis Risk
Basis risk arises due to changes in market rates on different financial instruments
by varying degree. The risk that the interest rate of
a different asset, liabilities and off-balance sheet items may change in different
magnitudes is termed as basis risk. The degree of basis risk is fairly high in respect
of banks that create composite assets out of composite liabilities. For example, a
bank may be funding floating rate loans linked to its Benchmark Prime Lending
Rate (BPLR) through composite liabilities of various maturities. The rates on these
liabilities may change by different degrees whereas the bank may not be able to
change its BPLR by same degree. This may result NII of the bank to shrink in a
rising interest rate scenario.

5c.Embedded Option Risk

Embedded option means possibility of alteration of cash flows to the disadvantage
of a bank. Traditionally banks have offered products with embedded options. The
depositors enjoy freedom to close their deposits at any time by paying penalty.
Similarly, there are embedded options with loan products such as cash credit,
demand loans and term loans. Banks are experiencing embedded option even in
stable interest rate environments due to stiff competition. Pre-payments in home
loans are regular phenomenon. However, in volatile interest rate scenario, the
degree of usage of embedded option goes up. Banks statistically estimate the extent
of embedded options to be exercised by customers and incorporate them in
liquidity and interest rate risk models. The pricing of such risks into deposit and
loan products has also to happen. Though Reserve Bank of India has permitted

banks to deny premature closures in case of large deposits, they are currently not
doing so due to fear of losing business.

5d.Yield Curve Risk

Currently yield curve risk is limited in Indian Banks as mostly the assets and
liabilities in Indian banking system are either at fixed rate or at floating rate linked
to internal benchmarks. For example, through most of the term loans and cash
credit loans in banking system are at floating rate linked to internal PLR of banks,
the degree to which they respond to changes in external benchmark rates is a
matter of debate. Further, the relevance of benchmark PLR is itself questionable
when banks are lending at sub-PLR rates. The concept of yield curve risk may be
explained with the following example:

5e.Price Risk
Banks are required to mark to market their investment portfolio in held for trading and available
for sale category. In the financial markets, prices of instruments and yields are inversely related.
During last three years prior to March 2008, due to slow credit pick up, a large number of banks
had invested in government securities more than regulatory requirements. As per Reserve Bank
of Indias guidelines banks may classify their investments into three categories viz. a) Held for
Trading (HFT) b) Available for Sale (AFS) and c) Held till Maturity (HTM). While securities in
HFT and AFS categories are required to be marked to marked, the securities in HTM are not. As

interest rates declined, banks made huge treasury gains in their investment portfolio. However,
due to rise in yields many banks reported treasury losses in December quarter 2004. The losses
would have been higher, had RBI not permitted banks to shift a portion of their securities to
Held to Maturity class. However, notwithstanding this accounting treatment, the rising interest
rates will result in economic loss of bonds held in held till maturity category.

6. Management of Interest Rate Risk in Banking Book- Earnings

Perspective and Economic Value Perspective.
6a. Earnings perspective
As discussed in previous paragraphs, changes in interest rates will affect net
interest income of a bank in short run and its economic value in long run. The
impacts on earnings are computed with a perspective of a plan horizon, which may
vary from bank to bank. A bank sets a plan horizon for computation of earnings at
risk and groups all assets and liabilities into different time buckets upto the plan
horizon. It assumes a rate shock at the beginning of the plan horizon and applies to
all assets and liabilities re-pricing during the horizon. Depending on the position of
gaps and likely interest rate shock, the analysis will present net gain or loss on NII.
A bank, as a prudent policy may put limits on its earnings at risk (EAR). The limit
may be expressed as EAR as percentage to net-worth, rate sensitive gaps as
percentage to total assets etc. Further, some banks refine this process by applying
different degree of rate changes on assets and liabilities accepting the basis risk
prevalent in their asset and liability products, their customer profile and practicality
of enforcing re-pricing of contracts. The limits on gap positions can actively be
managed either through on or off-balance sheet strategies. For example, a bank
with net negative gap upto 1-year plan horizon and anticipating rates to harden up,
will pay in Overnight Index Swap (OIS) to improve positive mismatches in short
term buckets.

The following tables illustrate gap management strategies:

As illustrated above, both the strategies result in reduction of Earnings at Risk

from Rs (-) 16 Cr to Rs (-) 11 cr. However, the bank may have to begin with
negative earnings on swaps initially (for example while paying in Swap the fixed
leg (pay) will be higher than the floating leg (MIBOR)), which needs to be factored
in analysis. Though gap analysis, as described above, is a very simple tool to
manage interest rate risk in banking book with short-term perspective, it suffers
from some serious limitations which need to be kept in mind:

The gap is computed as the rupee difference between the values of rate
sensitive assets and liabilities in the gapping period regardless of when the
repricing occurs. Thus if all assets re-price at the beginning of the maturity
period and the liabilities at the end of the period the institution will not be
insulated from interest rate risk even though the gap is maintained at zero.

The gap management might be hampered by the objectives of the customers.

In rising interest rate scenarios gap management recommends shifting out of
fixed rate assets to floating rate assets. The customers may however demand
fixed rate assets. Additionally such adjustments in assets/liabilities may have
to be accomplished at the cost of trading off lower interest rate risk for
greater credit and default risk.

 Alternatively, such assets may be swapped at floating rate through OIS.

However, there has to be a definite view on magnitude as well as quantum of
interest rate increase, as there will be opportunity loss through negative carry
at the beginning.
Gap management does not take into account that re-pricing spreads of assets
and liabilities which may not be identical i.e. it is possible that the rise in
interest rates on liabilities is proportionately higher than rise in interest rate
on assets leading to a decline in spreads despite an increase in the general
level of rates.
Gap management concentrates solely on flow of funds and variability of
revenues and does not focus on the effect of interest rate on the market value
of assets and liabilities.

6b.Economic value perspective

One of the objectives of an ALM system is to preserve and enhance long-term
economic value of a bank or its market value of equity.
Market value of equity (MVE) refers to net position of net-worth after marking to
market assets and liabilities. Similar to individual securities, duration analysis can
be used to assess the impact of interest changes on the market value of the assets
and liabilities of a bank. The Duration Gap indicates the effect of interest rate
changes on the net worth of the institution. This serves as a strategic planning tool
for evaluation and controlling the interest rate risk of the institution. Duration gap
analysis can be used to immunize the relevant target account (assets, liabilities or
groups of assets and liabilities) against unexpected changes in interest rates by
setting the duration gap of the target account to zero. In a positive duration gap this
can be accomplished by lengthening the duration of liabilities or shortening the

duration of assets. However since duration drift can occur with passage of time
without any changes in interest rate, the institution will have to regularly
restructure the balance sheet. Similarly, when
interest rates are expected to rise, the composition of balance sheet should be
restructured to obtain a negative duration gap.
Thus the market value of the institution will increase due to a greater reduction in
the value of liabilities than assets. A higher duration gap will result in greater
degree of changes in market value of equity. A banking institution may create a
duration gap to improve market value of equity depending on its interest rate view.

7. Value at risk (VAR)

Risk in the trading book are measured on a real time basis through
VAR
Meaning
Throughout the centuries since man became a commercial animal, investors and
speculators have shared a common intuition that there is a balance between fear
and greed between the risks of investments and their potential rewards. The silk
merchants of ancient China, the crowd that gathered to trade shares under the
buttonwood tree near what would later become the NewYork Stock Exchange, and
the denizens of modern high-tech trading rooms have all shared a common
concern: how to quantify the risk of loss.
For a long time, the risk takers just guessed. After all to know with certainty the
magnitude of potential losses was to know the future! And while methodologies for
quantifying the risk of loss in various circumstances such as "duration" for fixed
income securities were developed in recent decades, foreknowledge of market
conditions was required to answer with certainty.

The problem confronting investors and speculators was not unlike that facing
Modern scientists: predicting future events with certainty is simply not possible.
And it was only when financial market participants adopted methodologies
pioneered in the sciences that a satisfactory gauge for market risk was developed.
In adopting the methods of the empirical sciences, which predict future events
according to the laws of probability and statistics, risk managers were obliged to
accept the highly probable, but nonetheless conditional answers they produce.
Since the question "how much can I lose?" cannot be answered with absolute
certainty, market practitioners learnt to be satisfied with answers to questions of a
significantly different form: "With a degree of confidence chosen by me, what is
the largest loss I am likely to suffer?" And it is questions of this form that is
addressed when we compute the value-at-risk of a position, or its VAR.
Meaning & how it is useful
First, VAR is an amount of money which is what we mean when we ask how
much we can lose. Second VAR is an estimate. It is based upon probabilities, so we
cannot rely upon it with certainty, but rather with a degree of confidence we select.
VAR is defined as the expected maximum loss (expressed in currency units) due to
unfavorable movement of the underlying market factors that an asset or a portfolio
of assets will suffer over a target horizon (T) within a given degree of certainty /
confidence level (a ) selected by the decision maker.
The VAR concept embodies three factors. The first is given time horizon. A risk
manager might be concerned about possible losses over one day, one week, one
month, etc. Second, VAR is associated with a probability. The stated VAR
represents the maximum possible loss over a given period of time with a given

probability. Third, there is the actual rupee amount itself. Consider for example a
dealer with an Rs 20 million position. He might find that his VAR for a one-day
period, with a one percent probability is Rs 500,000. This means that the dealer can
expect to lose at least Rs 500,000 in any given day about one percent of the time,
or in other words, 2.5 times in a year (assuming 250 trading days). Of course the
user can specify any probability or holding period and there are no hard or fast
rules to help one decide on the appropriate probability and holding period.

How does one choose appropriate holding period and confidence

level?
The choice of the holding period depends on the liquidity of the market under
consideration. As far as probability is concerned, the choice is purely subjective.
Bankers Trust uses a 99 percent level; Chemical and Chase, a 97.5 percent level;
Citibank a 95.4% level; Bank America and J.P.Morgan a 95 percent level. For
internal models of capital allocation, the Basle Committee recommends a multiple
of three (with an additional factor depending on the failure rate of the VAR limit in
the past) of the VAR, calculated using 10 day holding period and 99 percent level
of confidence.
The concept of VAR is a very appealing one. It can be developed for any kind of
portfolio and can be aggregated across portfolios of different kinds of instruments.
For example, a bank might have a portfolio of interest rate swaps, a portfolio of
currencies, positions in some commodities and a portfolio of common stocks, as
well as its regular loan and bond portfolios. The VAR for each separate portfolio

can be calculated and aggregated across all portfolios. This does not imply that
estimating VAR for a portfolio is a simple process; the correlations across asset
classes must be accounted for. VAR does, however provide a consistent measure
across portfolios. Thus with appropriate considerations for all correlations, VAR
can provide the bank with an overall measure of exposure.
An attractive feature of VAR is that it is stated in terms of rupees. Senior
management and anyone without much technical knowledge of financial products
can still easily understand the concept.

How is VAR computed?

Now let us examine how VAR is computed. First we begin with a simple example
that takes the required inputs as given. Later we shall look at how this information
is obtained. Suppose that a firm has compiled the following information about the
change in the value of a portfolio over a one-week period.
Change in Value

Probability

<= Rs 1,000,000

0.01

-Rs 500,000 to Rs 999,999

0.04

-Rs 250,000 to Rs 499,999

0.15

Rs 0 to Rs -249,999

0.30

Rs 1 to Rs 249,999

0.30

Rs 250,000 to Rs 499,999

0.15

Rs 500,000 to Rs 999,999

0.04

>=Rs 1,000,000

0.01

From this information we can say that VAR at 1% probability is Rs 1,000,000.

This means that the firm would expect to lose atleast Rs 1,000,000 one percent of
the time. A common misconception about VAR is that it is the maximum expected

loss. It should be apparent that this is not what VAR is. We do not say that the most
the firm could lose 1 percent of the time is Rs 1 million. Obviously the firm could
lose more than Rs 1 million.
From the above table we can also easily define VAR at 5 percent, which is Rs
500,000. There is a 4 percent chance of a loss between Rs 500,000 and Rs 999,999
plus a one percent chance of a loss of atleast Rs 1 million. Thus the firm can expect
to lose atleast Rs 500,000 five percent of the time.

The probability distribution given above is a simple discrete distribution.

Sometimes a firm will prefer to use a continuous distribution, the Normal
Distribution being a common one. When that is the case, then the user must know
the expected change in value, E (D V), and the standard deviation, s (D V). From
there, one can compute VAR as
VAR (1%) = E (D V) 2.33 s (D V)
VAR (5%) = E (D V) 1.65 s (D V)
These numbers 2.33 and 1.65, represent the number of standard deviations to the
left of the expected value that leave 1 percent and 5 percent, respectively in the tail
of the distribution.
For example, suppose a firm estimates that a change in its portfolio value is Rs
1,000,000 with a standard deviation of Rs 1.5 million. Thus VAR (1%) = Rs
1,000,000 2.33 (Rs 1,500,000) = Rs 2,495,000 and VAR (5%)= Rs 1,000,000
1.65(Rs 1,500,000)= -Rs 1,475,000). Thus the firm could expect to lose atleast Rs

2,495,000 one percent of the time and lose atleast Rs 1,475,000 five percent of the
time.
A discrete distribution is, in many cases, merely a simplification of a continuous
distribution. In finance virtually all risks are associated with continuous
distributions. The more critical problem however, is not whether to use a discrete
or continuous distribution but how to obtain the inputs.

How does one obtain the inputs to be entered into a generic VAR
model?
Determination of Value at Risk necessitates modeling the movement in market
factors. The modeling can be done either by use of simulation methods or analytic
methods. Simulation methods are heavily dependent on either the values assumed
by the market factor in the past or what we expect it to assume in future.
The Historical Simulation method requires that the user obtain historical
information on the movement of market factors (share prices, interest rate yield
curve, exchange rates, commodity prices etc) which determine the performance of
the financial instruments that are in the current portfolio. This means a time series
of the prices or returns on these assets. From the historical data, the user can
calculate the observed changes in the market factors over the time horizon to be
used in the VAR calculation. Obviously this method relies heavily on the
assumption that the portfolio distribution from the past continues to hold for the
future.
Alternatively the user might do a Monte Carlo simulation. Using the parameters
estimated from the past data, the user can then generate random numbers to
represent market prices, interest rates etc. and for each simulation run, the value of

the portfolio is determined. The simulation enables one to construct a frequency

distribution from which VAR can be picked off from the left tail. These
descriptions provide a general idea behind the historical method but the
implementation is not as quite straightforward as it might seem. To produce a
portfolio Variance it is necessary to estimate correlations among the different
classes of instruments. For a Monte Carlo simulation, one must generate random
numbers representing prices, interest rates, and exchange rates according to
relative frequency with which these numbers would appear in practice.
Analytic methods, on the other hand, solve for the VAR limit, i.e. the (1-A ) the
percentile, based on distributional assumptions about the movement of market
factors and the estimated parameters. For example, under the assumption of a
normal distribution with zero mean and standard deviation s , the 95% VAR limit
(the absolute value of the fifth percentile of the distribution) is given by 1.65*s.

What are the key steps for computation of Value at Risk by analytic
method?
In general, the analytic method for calculation of Value at Risk follows the
following steps:
Identifying exposures and cash flows
Cash flow is the building block for describing any position. A cash flow is defined
by an amount of currency, a payment date and the credit standing of the payer once
determined these cash flows are marked to market. Marking to market a position's
cash flows means determining the present value of the cash flows given current

market rates and prices. This procedure requires current market rates, including a
zero coupon yield curve to value fixed income securities.
Mapping cash flows on certain vertices
Financial instruments , in general, can generate numerous cashflows , each one
occurring at a unique time. This gives rise to an unwieldy number of combinations
of cash flows dates when many instruments are considered. As a result, we are
faced with the impractical task of having to compute an intractable number of
volatilities and correlations for the VAR calculation. To more easily estimate the
risks associated with instruments' cashflows, we need to simplify the time structure
of these cashflows.
One of the methods of simplifying time structure involves cash flow mapping, i.e.,
redistributing (mapping) the observed cash flows onto certain vertices.
Computing Value at Risk
Two analytical approaches can be used to measure Value-at-Risk : Simple VAR for
linear instruments, and Delta-Gamma VAR for nonlinear instruments, where the
terms "linear" and "nonlinear" describe the relationship of a position's underlying
returns to the position's relative change in value.
In the simple VAR approach, we assume that returns on securities follow a
conditionally MultiVariate normal distribution and that relative change in a
position's value is a linear function of the underlying return. Defining VAR as the

5th percentile of the distribution of a portfolio's relative changes, we compute VAR

as 1.65 times the portfolio's standarddeviation. Where the multiple 1.65 is derived
from the normal distribution. This standard deviation depends on the volatilities
and correlations of the underlying returns and on the present values of the cash
flows.
In the Delta-Gamma approach, we continue to assume that returns on securities are
normally distributed, but allow for a nonlinear relationship between the position's
value and the underlying returns. Specifically we allow for a second-order or
gamma effect , which implies that the distribution of the portfolio's relative change
is no longer normal

Is VAR a panacea to prevent all risks?

VAR is one of the essential tools of the risk manager but it is not the whole story.
Its purpose is to give an estimate of losses over a short period under "normal"
market conditions. It is not going to tell you what might happen during a market
crash for that stress testing and scenario analysis are necessary.
Also as the Barings debacle showed, any risk measurement system is only as good
as the numbers that go into it. A rogue trader entering false positions will
undermine the usefulness of VAR reports. So the method, while extremely
powerful for day to day risk management , is no substitute for wider risk
management process of analysing crash scenarios and keeping tabs on operational
and legal risks.

8. Implementation of the Guidelines - Instructions from RBI to

Banks
In this context RBI gave the following instructions with reference to
implementation of the guidelines.
1. Banks should give adequate attention to putting in place an effective ALM
System. Banks should set up an internal Asset-Liability Committee (ALCO),
headed by the CEO/CMD or the ED. The Management Committee or any
specific Committee of the Board should oversee the implementation of the
system and review its functioning periodically.
2. Keeping in view the level of computerization and the current MIS in banks,
adoption of a uniform ALM System for all banks may not be feasible. The
final guidelines have been formulated to serve as a benchmark for those
banks which lack a formal ALM System. Banks which have already adopted
more sophisticated systems may continue their existing systems but they
should ensure to fine-tune their current information and reporting system so
as to be in line with the ALM System suggested in the Guidelines. Other
banks should examine their existing MIS and arrange to have an information
system to meet the prescriptions of the new ALM System. To begin with,
banks should ensure coverage of at least 60% of their liabilities and assets.
As for the remaining 40% of their assets and liabilities, banks may include
the position based on their estimates. It is necessary that banks set targets in
the interim, for covering 100 per cent of their business by 1ST April , 2000.
The MIS would need to ensure that such minimum information/data
consistent in quality and coverage is captured and once the ALM System
stabilizes and banks gain experience, they must be in a position to switch

over to more sophisticated techniques like Duration Gap Analysis,

Simulation and Value at Risk for interest rate risk management.
3. In order to capture the maturity structure of the cash inflows and outflows,
the Statement of Structural Liquidity Annexure-I should be prepared, to
start

with,

as

on

the

last

reporting

Friday

of

March/June/

September/December and put up to ALCO/Top Management within a month

from the close of the last reporting Friday. It is the intention to make the
reporting system on a fortnightly basis by 1 ST April, 2000. The Statement of
Structural Liquidity should be placed before the bank's Board in the next
meeting. It would also be necessary to take into account the rupee inflows
and outflows on account of previously contracted forex transactions (swaps,
forwards, etc). Tolerance levels for various maturities may be fixed by the
bank's Top Management depending on the bank's asset - liability profile,
extent of stable deposit base, the nature of cash flows, etc. In respect of
mismatches in cash flows for the 1-14 days bucket and 15-28 days bucket, it
should be the endeavour of the bank's management to keep the cash flow
mismatches at the minimum levels. To start with, the mismatches (negative
gap) during 1-14 days and 15-28 days in normal course may not exceed 20%
each of the cash outflows during these time buckets. If a bank in view of its
structural mismatches needs higher limit, it could operate with higher limit
with the approval of its Board/Management Committee, giving specific
reasons on the need for such higher limit. The objective of RBI is to enforce
the tolerance levels strictly by 1ST April , 2000.
4. In order to enable the banks to monitor their liquidity on a dynamic basis
over a time horizon spanning from 1-90 days, an indicative format
Annexure III is enclosed. The statement of Short-term Dynamic Liquidity

should be prepared as on each reporting Friday and put up to the ALCO/Top

Management within 2/3 days from the close of the reporting Friday.
5. We advise that in the Statement of Interest Rate Sensitivity Annexure - II
only rupee assets, liabilities and off-balance sheet positions should be
reported. The statement should be prepared as on the last reporting Friday of
March/June/September/December and submitted to the ALCO / Top
Management within a month from the last reporting Friday. It should also be
placed before the bank's Board in the next meeting. The banks are expected
to move over to monthly reporting system by 1st April, 2000. The
information collected in the statement would provide useful feedback on the
interest rate risk faced by the bank and the Top Management/Board would
have to formulate corrective measures and devise suitable strategies
wherever needed.
The guidelines for ALM cover the banks' operations in domestic currency. In
regard to foreign currency risk, banks should follow the instructions contained in
Circular AD (MA Series) No.52 dated 27th December, 1997 issued by the Exchange
Control Department."

9. CONCLUSION
Interest rate risk and liquidity risks are significant risks in a banks balance sheet,
which should be regularly monitored and managed. These two aspects should be a
key input in business planning process of a bank. Banks should make sure that
increased balance sheet size should not result in excessive asset liability mismatch
resulting in volatility in earnings. There should be proper limit structures, which
should be monitored by Asset Liability Management Committee (ALCO) on a

regular basis. The effectiveness of ALM system should be improved with a good
Fund Transfer Pricing system.

10. Bibliography
www.google.com
Indian Institute of Banking and Finance
www.rbi.org
www.economicstimes.com

EXECUTIVE SUMMARY

Asset-liability management refers to the overall control of the composition of

balance sheet accounts to attain key objectives. These key objectives are to
generate optimum levels of quality earnings and to maintain adequate liquidity
to meet both predicted and unexpected cash needs. The increasing volatility in
funding sources and market rates resulting from the removal of interest rate
limitations and rapid fluctuations in the economy have made effective funds
management essential to successful operations.
The main objective of ALM is Liquidity Risk Management and Interest Rate
Risk Management. These two aspects is key input in business planning process
of a bank. Liquidity risk and Interest Rate risk can be measured through various
tools. To measure liquidity risk stock approach and flow approach is used while
Gap analysis and Value at Risk are important tools to measure interest risk.
To achieve the above objective of ALM Asset and liability management
Committee (ALCO) is delegated the responsibility from the risk management
committee. ALCO manages the market risk faced by the bank while ensuring
adequacy of capital for the various market risk related activities/products of the
bank. ALCO also frames the liquidity policies and manage interest rate risk in
the Banking book and in the trading book. ALCO frames the policies according
to the guidelines given by RBI.