Stress Testing for IRRBB

Sanjay Basu, NIBM.

March 2022

1
 Outline
• RBI IRRBB guidelines 2017 - snapshot.
• Modelling Yield Curve Risk.
• Modelling Basis Risk.
• Modelling Options Risk.

2
 IRRBB – Basel 2016/RBI 2017
• Capture three types of scenarios – internal, historical
and subjective stress shocks and six regulatory ones.
• Assess repricing, yield curve, basis and options risks.
• Slot cash flows from all RSAs and RSLs (including
future interest receipts and payments), in 19 repricing
buckets, from 1 day to above 20 years.
• Discount all cash flows, with appropriate yields, to
compute EVE.
• If EVE falls by more than 15% of T1 capital, reduce
exposure or add to capital for IRRBB.
3
 IRRBB – Basel 2016/RBI 2017
• Parallel shock up;
• Parallel shock down;
• Steepener shock (short rates down and long rates up);
• Flattener shock (short rates up and long rates down);
• Short rates shock up; and
• Short rates shock down.

4
 Yield curve versus Basis Risks
• Yield curve risk captures the impact of rate shocks
which are common to all products within a given
segment, but unequal for different maturity segments.
➢ It ignores changes in product-specific spreads
• Basis risk captures the impact of rate shocks which
are equal across various tenors, but dissimilar for
different products within a given segment.
➢ The focus is on capturing changes in product-specific spreads.

5
 Modelling Yield Curve Risk
• Depending on the complexity of the bank, choose an
appropriate model of the yield curve which is also
relevant to the Indian market.
• Using the parameters, generate model yields, for at
least 250 days in the past, corresponding to the
midpoints of the maturity buckets.
• Calculate annual changes in yields, for each maturity
bucket, to create a history of model yield shocks.
• Estimate the NII and EVE impact with either
Historical or Monte Carlo simulation methods.

6
 Historical Simulation
(1) Measure annual changes in historical yields for
various tenors for the past 250 or 500 days.
(2) Rank shocks, in ascending order, from the sharpest
rise to the sharpest decline.
(3) Mark off the 99% or 95% worst shock for each
tenor.
(4) Calculate the NII and EVE impact as before.

7
 Modelling Basis Risk
• Pick up average advance and deposit rates, from at
least one year of historical data from the bank,
ensuring that there are no volatility or regime shifts.
• Calculate the annual changes in advance and deposit
rates, to create a history of rate shocks.
• Use Historical Simulation, as before, to estimate rate
shocks at a particular c.l..
• Estimate NII or EVE impact for such rate shocks.

8
 Options Risk
• Effect of options embedded in bank assets, liabilities
and OBS items on future NII and Net worth. Such
options provide a customer the right, but not the
obligation, to alter the cash flows of these products.
(1)Borrowers have the right to prepay their term-loans.
(2)Depositors have the right to rebook their term
deposits.
(3) Savings and current account holders have the right
to introduce or withdraw money at any point.
 Modelling Options Risk
• Estimate the extent of premature withdrawal or
prepayment, on liabilities and assets, for each tenor.
• Redistribute RSAs and RSLs, across buckets, based
on estimated options exercise.
• Compute MVAs and MVLs after the reallocation.
• Calculate repricing and duration gaps again.
• Calculate the NII and EVE impact after options
exercise.

10
EVE Impact as on 31.3. xx (in Rs. Cr.)

Pillar II
EVE Impact Share Capital

200 bps shock -328.096 13.68% Nil

Yield Curve Risk – S1 -303.234 12.63% Nil

Yield Curve Risk – S2 -612.80 25.53% 252.78

Basis Risk -713.224 29.72% 353.21

 Reasons for EVE Impact
• The EVE loss is the highest for Basis Risk because
of two reasons:
1. The positive duration gap is implicit. A rise in rates
will still push up cost of funds first and yield on
assets later. This will squeeze future NII and EVE.
2. Since rise in asset rates > rise in liability rates, the
opportunity loss on assets exceeds opportunity gain
on liabilities, even if the durations are the same.
➢ The bank may make an NII gain, since future interest income
rises more than interest cost, but EVE will decline.
Liquidity Risk Management

Sanjay Basu, NIBM,

March 2022
 Lecture Outline
• Liquidity Risk – definition, objectives and
implications.
• Liability Side Liquidity Risk: (i) Purchased Liquidity
Management (ii) Stored Liquidity Management .
• Asset-Side Liquidity Risk.
• Structural Liquidity Statements and liquidity risk
limits.
 Goals of Liquidity Risk Management
• Fix liquidity risk tolerance in line with the bank’s
business strategy – policies, limits and procedures.
• Include liquidity costs and benefits in FTP policies -
penalize long-term assets and reward stable funds.
• Monitor weaknesses in cross-border funding and
collateral valuation, on a regular basis.
• Develop a variety of short-term and long-drawn
bank-specific and market–wide stress scenarios and
associated Contingency Funding Plans (CFPs).
 ILAAP in Basel II
• The Internal Liquidity Adequacy Assessment Process
(ILAAP) is part of Pillar 2 in the Basel II Accord,
dedicated to the supervisory review process.
• The ILAAP comprises a clear assessment of the risks
to liquidity, supported by well-structured risk
governance and risk escalation systems.
➢ The objective of the ILAAP is to identify, analyse and
wherever possible, quantify the material risks faced by a Bank.
➢ The ILAAP exercise enables the Bank to reduce stakeholder
uncertainty about its continuity with a combination of high-
quality liquidity buffers, stable funds, liquidity risk limits and
control systems.
 Steps in ILAAP
• Describe the roles and responsibilities of the bank
hierarchy, in liquidity risk management.
• Create a liquidity adequacy statement, which contains
the drivers and threats to liquidity buffers, ILAAP
parameters, inputs, outputs and business strategies.
• Assess material risks, project internal liquidity needs
and compare with available buffers.
• Forecast business growth, conduct stress tests,
formulate liquidity plans and strategies to raise it.
• Revise limits, pricing decisions and liquidity
allocation to optimize risk-adjusted performance.
 Liability-Side Liquidity Risk
• Focus on the volume of stable, long-term, core
deposits and net deposit drains (deposit outflows
less new deposit and other inflows each day).
• Need to predict the distribution of net deposit drains.
1. Behavioural analysis of NMDs.
2. Behavioural analysis of premature withdrawal, especially for
high-value deposits.
3. Analysis of renewal patterns of maturing term deposits.
 Purchased Liquidity Management
1. Borrowing from the call money or repo market.
2. Issuing additional deposits or selling bonds.
• Managing the liability side without touching the
asset side of the balance sheet.
➢ Purchased liquidity can insulate the asset side from deposit
drains.
• Borrowing funds at higher rates to offset drains in
low-interest deposits.
• Uninsured funds may not be available for large
deposit drains.
 Stored Liquidity Management
• Liquidation of assets to meet deposit drains.
➢ Compute the MV of all assets by assigning conservative (or
worst-case) discount rates.
➢ Adequacy of cash capital – the difference between the MVs
of tradable assets and short-term interbank funding plus
volatile deposits.
➢ Threat of depreciation or interest revenue losses.
➢ Focus on liquid, short-term T-bills – conflict between
liquidity and returns.
• Maintenance of short-term interbank deposits and
excess CRR.
➢ Opportunity cost of noninterest-bearing cash reserves to be
compared with losses on sudden deposit runs.
 Asset-side Liquidity Risk
• Risk from loan commitments and other credit lines:
➢ (i) borrowing funds or (ii) running down reserves.
• Maintaining high levels of savings deposits
➢ As loan commitments (e.g. unutilized portion of CC/OD
balances) are utilized during tight liquidity conditions, banks
with high levels of S/B deposits are in a better position to pay
their borrowers on demand.
➢ Correlation between liquidity demand from depositors and
borrowers have been found to turn negative during crises.
• Cash-flow uncertainty from credit risky assets.
 Measuring Liquidity Exposure
• Structural Liquidity Statements (SLS): Based on
Maturity Ladder approach followed by BIS.
➢ Based on residual maturity of assets and liabilities.
➢ Cash inflows are ranked by future dates or intervals when
assets mature or credit lines can be drawn down.
➢ Cash outflows are ranked by future dates on which liabilities
and OBS commitments fall due.
➢ Interest should be included.
➢ The difference between expected cash inflows and outflows
(gap) in each bucket captures the liquidity surplus or deficit of
within a specific time interval in future.
 Measuring Liquidity Exposure
• The difference between expected cash inflows and
outflows (gap) in each bucket captures the excess or
deficit of liquidity within a specific future interval.
• Depending on balance sheet structures and strategies,
one bank can have negative gaps while another has
positive gaps, unlike trading liquidity risks.
➢ Banks reliant on short-term funding will actively monitor the
first few buckets.
➢ Monitoring longer-term gaps allows the bank to frame
strategies which alter product maturities.
RBI Liquidity Buckets – 31.3.2016
Next day.
2 days - 7 days.
8 days – 14 days.
15 days – 30 days.
31 day - 2 months.
2 months – 3 months.
3 months - 6 months.
6 month - 12 months.
1yr. - 3 yrs.
3yrs. - 5yrs.
Over 5 years.
 NBFC SLS: Share below 1 year
A B C D E F

Cash & Inv 52.41% 58.05% 69.28% 7.56% 74.45% 78.99%

Advances 48.65% 15.75% 21.07% 87.31% 21.69% 6.97%

Deposits 28.78% 30.10% 45.29% 59.40%

Bank Borr. 24.62% 27.93% 19.75% 22.34% 42.57%

Mkt. Borr. 33.82% 18.57% 37.11% 24.16% 25.58% 27.40%

Cgap/Cout 57.23% 13.74% -17.99% -37.24% 65.84% -71.54%

Percentage of TDs maturing < 1y

Bank A Bank B
Q1 57.37% 55.46%
Q2 77.59% 49.10%
Q3 45.15% 45.63%
Q4 53.73% 42.21%
Q5 63.38% 39.10%
Q6 60.97% 44.47%
Q7 70.69% 43.73%
Q8 76.96% 42.98%
 Cost of Funds (in %)

Y1 Y2 Y3

Bank A 5.74 6.83 7.27

Bank B 5.85 5.78 5.92

 Uses of SLS
• Allows a bank or FI to
1. Monitor liquidity gap management in shorter-term buckets
over time.
2. Monitor changes in the maturity patterns of key assets and
liabilities over time.
3. Set liquidity gap limits beyond the first four buckets, in line
with its available or planned buffer of liquid assets.
4. Design a variety of stress tests on assets and liabilities and
estimate their impact on liquidity gaps in short-term buckets.
5. Capture the impact of LCR and NSFR on assets, liabilities and
liquidity gaps.
 Liquidity Risk Limits
• Liquidity Risk Limits can be based on:
➢ Cumulative gap/ cumulative outflows: Measures the
cumulative deficits across short-term buckets as share of total
outflows across these bands.
➢ Allows surplus in earlier buckets to offset deficits in later ones
– bigger negative gaps and higher margins permitted.
➢ Allows maintenance of bigger deficits because of comparison
with cumulative outflows.
➢ Liquid (temporary) assets to short-term (volatile) liabilities:
Whether the sum of cash and near-cash assets can cover short-
term outflows (up to 1 year) – industry average: 60%.
Matching (around 100%) preferred.
 Liquidity Risk Limits
➢ Illiquid Assets/Core Deposits: The extent to which assets like
Loans & Mandatory SLR + CRR & Fixed Assets are funded
by stable deposits – industry average: 150%. Moderate value
preferred.
➢ (Volatile Liabilities – Temporary Assets)/(Earning Assets –
Temporary Assets): The extent to which earning assets are
funded by short-term, rate sensitive, liabilities – industry
average: 40%. Low value preferred
➢ Temporary Assets/Total Assets: The stock of low-yielding,
liquid assets in total assets – industry average: 40%. High
value indicates financial stability but low returns.
Behavioural Analysis for Non-
Maturity Deposits
Sanjay Basu, NIBM,
March 2022
 Outline
• Bucketing Savings Deposits.
• Common Flaws in bucketing and consequences.
 Problems
• Theory does not indicate the possible determinants
of NMD balances.
➢ Modelling NMDs is an attempt to answer the following
questions:
1. How are balances affected by passage of time and other
underlying risk factors?
2. How to allocate these balances in different buckets, given
that they do not have any contractual maturity dates?
3. How to compare and choose between different models?
 Solution
1. The current balance is decomposed into core or
stable portions and unstable or volatile portions.
➢ Assign volatile portions to short-term buckets.
2. Choose an appropriate statistical model, or rely or
expert judgment, to (i) determine the maturity and
(ii) volume of the core portions.
➢ The core portion represents the long-term outflow (for
liabilities) component of current balances.
 Trend Analysis for CASA
• Daily/weekly/fortnightly returns capture changes in
balances in the respective buckets.
• Changes in 8d -14d balances = (Fortnightly balance –
weekly balance)/Today’s balance.
• The same approach applies to all other tenors.
• For longer-tenors, due to balance sheet growth,
percentage changes might be positive for all c.l.s.
• We take the absolute values of all fluctuations and
ignore their signs – as in SD.
 Trend Analysis for CASA
(1) Measure current outstanding balance in rupees
(current weights).
(2) For each bucket, measure appropriate returns to
capture all the possible shocks, based on historical
scenarios.
(3) Multiply all past returns by current outstanding
balance, i.e. (1)  (2).
(4) Rank from worst outflows to highest inflows.
(5) Mark off the 1% or 5% worst sample outflow as the
amount to be bucketed.
 Trend Analysis: Replication
• CASA deposits are invested at their FTP rates across
various buckets.
• The aim is to maximize the average spread.
• The SD is bounded, to ensure allocation across all
buckets.
➢ This is similar to tenor-wise exposure limits.
• Often simple or Moving average FTP rates are used.
➢ Past rate cycles affect CASA allocation across buckets.
 Trend Analysis: Common Flaws
• Many banks use a mixture of behavioural and
regulatory bucketing rules for estimating outflows of
NMDs. The volatile portion is slotted (pro-rata) in the
first three buckets, the core portion above 5 years and
the balance in the 1y-3y bucket.
• They use fortnightly average balances, per month, to
find out core and volatile portions.
• They use one or two years of data.
• They use regression models with high R2, to choose
core and volatile portions.
 Trend Analysis: Common Flaws
• Many banks do not conduct bucket-by-bucket
behavioural studies on premature withdrawal of TDs.
• Many banks do not adjust the amounts bucketed in
SLS after trend analyses of premature withdrawal.
• Many banks do not conduct behavioural studies on
drawdown of the unutilized portion of CC/OD.
➢ They assume that the unutilized portion would be drawn down
in equal amounts, in all buckets, up to 1 year.
• Many banks assume that the amount of term deposits
rolled over would stay with them beyond 5 years.
 Common Flaws: Consequences
• RBI guidelines distinguish between two mutually
exclusive approaches: (i) regulatory allocation, in
which the buckets (up to 14 days and 1y-3y) and
core/volatile (15:85 or 10:90) ratios are specified by
RBI and (ii) behavioural studies, which are left to
each bank.
• The process of averaging fortnightly balances
smoothens out sharp fluctuations within a month.
• Fluctuations in 1y-3y buckets and above cannot be
captured with two years of data. .
• Large shocks may have occurred more than two years
ago.
 Common Flaws: Consequences
• Use of standard deviations underestimates large
negative shocks.
• SDs may also capture impact of business growth.
• The usage of regression models with high R2, by
including many variables, might fail to capture large
future shocks which are not present in historical data.
• The upshot is that these flaws might over-esimate
stability of some segments of NMDs.
• A bank may be unable to detect that, e.g. its CA
balances are highly volatile or stagnant while its SA
balances are stable.
 Common Flaws: Consequences
• It could be hit by large outflows in the volatile
segments without prior warning.
• It will not know which business to focus on and
which ones to downplay.
• A similar logic applies to premature withdrawal of
TDs or drawdown of unutilized limits.
• Renewed TDs need not always be kept in the ‘above
5 years’ segment. They are more likely to be bunched
in the bucket that pays the highest rates.
Stress Testing for Liquidity Risk

Sanjay Basu, NIBM,

March 2022
 Outline
• Scenario Analysis.
• Multifactor Stress Tests
• Risk Interactions.
• Contingency Funding Plans.
 Need for Liquidity Stress Tests
• Recent financial crises have shown that many banks
have failed not due to shortage of capital, but under
liquidity pressure.
➢ Capital adequacy concerns arose later when they sold off
assets or borrowed at high cost, to solve liquidity problems.
• There is a need to develop bank-specific liquidity
stress tests over short horizons, to estimate the
additional liquidity and capital buffers in tight
markets.
Liquidity Stress Test - Components
• Focus on both severity and probability of shocks.
• Develop forward looking scenarios - to capture a
wide range of possibilities.
• Identify the factors driving short-term liquidity stress
and establish channels to obtain contingency funding.
• Capture liquidity shocks to both asset and liability
sides of the portfolio.
• Capture interactions with market and credit risks.
• Estimate the impact on NII and net worth.
 Liquidity Stress Test – Strategies
• Set limits on funding composition (counterparty,
maturity, location, currency and markets) under
normal and stressed conditions.
• Develop conservative estimates of the liquidity values
of traded assets, in addition to regulatory guidelines.
• Capture the effect of downgrade on the MTM value
of traded assets, which serve as liquidity buffers.
• Analyze the impact of mark-to-market (MTM) losses
in collateral value on margin requirements for SFTs
and derivatives.
 Liquidity Stress Test – Strategies
• Assess the impact of default on cash inflows from
non-traded (e.g. Loans and advances) assets.
• Predict the severity of outflows due to off-balance
sheet commitments (credit lines, LC and BG) over
various horizons.
• Estimate the adequacy (in terms of volume and
composition) of liquidity buffers over different
tenors.
 What-if Scenarios – Liability Side
• A primary liquidity risk is deposit run-offs in a bank-
specific event. Assumed shocks are based on a
combination of firm-specific historical data, industry
data from prior stress events and/or guesstimates.
➢ Run-offs during stable historical periods may not adequately
proxy depositor panic during a future stress event.
➢ Generally, most retail depositors are covered by deposit
insurance and are less likely to withdraw their funds.
➢ For corporate and bank (i.e. bulk) deposits, which are
uninsured, assumed shocks should be stronger.
➢ The duration of the crisis should be captured in the stressed
scenarios.
 What-if Scenarios – Asset Side
• Unexpected drawdown on committed lines of credit.
➢ As markets tighten and rates shoot up, firms may quickly draw
down unutilized credit limits.
➢ This will force banks to borrow short-term at high cost or sell
traded assets at large MTM losses.
• Non-contractual obligation to fund B/S growth and
mitigate reputational risk – LC and LG.
• Delay in scheduled inflows.
➢ Firms may have their own liquidity issues and not be able to
pay back on time. They may also default during stress events.
➢ Sharp decline in MTM value of traded assets.
 Scenarios with Key Risk Factors
1. Establish an empirical relationship between
fluctuation in NMD balances and possible causes.
➢ On the basis of sample data, reject those factors which do
not appear to explain fluctuations in losses.
➢ Construct ‘what if’ scenarios with the relevant risk factors.
• Choose a model which does not contain too many
causal factors.
➢ The estimated NMD balances should be close to observed
balances not only when the values of causal factors (say
time) are near sample data range, but also far away.
 Simple Scenario Analysis
1. Estimate a large decline in SA balances, under
normal conditions, from the regression equation.
2. Create risk factor scenarios to capture correlation
breakdown and large outflows during stress events.
➢ The number of scenarios may be much larger than historical
sample size.
3. For assumed loss in SA balances, back out the
combination of underlying shocks to risk factors, from
the regression equation – reverse stress tests.
Contingency Funding Plans (CFPs)
• A CFP is a projection of funding sources and needs
under various scenarios with liquidity disruption.
➢ Information flows are timely, uninterrupted and precise during
a stress event.
➢ Division of responsibility is clear, so that concerned personnel
know in advance what is expected of them in a stress event.
➢ Strategies for altering asset and liability behaviour during
stress, including identification of assets which can be marketed
more aggressively during a liquidity strain.
➢ Relationships with liability holders, borrowers and
counterparties, to source funds better during an adverse
liquidity or credit event.
 Triggers for CFP
• A good CFP should specify a number of triggers
which indicate a situation of abnormal liquidity in the
near future, though there might not be any current
funding problems. For instance, the CFP can be
activated when six out of ten triggers are hit. The
triggers can be both quantitative and qualitative.
➢ The CFP should be proactive, i.e. it should try to pick up early
warning signals and pre-empt a liquidity crisis. The current
trend, in India, is to start implementing the plan only after a
bank is deep into a crisis.
 Bank-specific Triggers for CFP
• Increase in spread on uninsured deposits, borrowed
funds or asset securitization.
• Downgrade by a rating agency.
• Reduction in tenors lenders are willing to accept.
• A decline in earnings.
• An increase in the level of NPAs or loan losses.
• A decline in the bank’s stock price relative to the
stock prices of similar banks.
• Significant asset growth or acquisitions.
 Systemic Triggers for CFP
• Significant volatility in exchange rates.
• An unexpected shift to a tight monetary policy.
• Changes in macroeconomic variables like growth or
unemployment rates, indicating prolonged recessions.
• Indications that an asset bubble might burst.
• Difficulty in selling securities that are normally
liquid, reduced availability of interbank funds and
system-wide increase in loan demand.
• Political instability and large-scale interference in
lending and bank supervision.
 CFPs for Crises
• Bank-Specific
➢ Use of stored liquidity in traded securities.
➢ Use of unutilized limits with NABARD, RBI and other FIs.
➢ Freezing of unutilized limits to borrowers.
➢ Borrowing against collateral eligible in the repo market..
➢ Arrangement of lines of credit with banks and other FIs.
➢ Asset Securitization.
• Systemic
➢ Full use of refinance facility from RBI.
➢ Deposit acceptance at differential rates of interest (DRI).
➢ Incentives for prepayment of advances.
The Liquidity Coverage Ratio

Sanjay Basu, NIBM,

March 2022
 Outline
• Stress Scenarios in Basel III.
• Liquidity Coverage Ratio and Net Stable Funding
Ratio – Features and Implications.
 Stress Scenarios
• 3-notch downgrade in the bank’s public credit rating.
• Run-off on a fraction of retail deposits.
• Loss of unsecured wholesale funding and partial
erosion in the ability to source secured funds.
• Secure short-term funding only for very liquid assets.
• Sharp rise in market volatility that erodes collateral
values.
• Unexpected drawdown on committed lines of credit.
• Non-contractual obligation to fund B/S growth and
mitigate reputational risk.
 Liquidity Coverage Ratio
• A stock of very liquid assets to manage net cash
outflows, under acute stress, for at least 30 days.
• Asset Features
1. Low credit & market risk: High credit quality, low duration,
low volatility, low inflation risk, issued in hard currency.
2. Easy to value: Widely accepted valuation models and inputs;
no exotic products.
3. Low correlation with risk: Remains liquid under severe
stress.
4. Listed in developed exchanges: Transparent and popular.
 Level 1 Assets
• These assets can be included in the stock of liquid
assets without any limit and haircut:
➢ Cash including cash reserves in excess of required CRR.
➢ Government securities in excess of the SLR requirement.
➢ SLR securities within the mandatory requirement to the extent
allowed by RBI (2% of NDTL at present).
➢ Special/Committed Liquidity Facility (9% of NDTL at
present).
➢ Marketable securities issued or guaranteed by foreign
sovereigns satisfying all the following conditions:
 Level 1 Assets
1. Assigned a 0% risk weight under the Basel II standardized
approach;
2. Traded in large, deep and active repo or cash markets
characterized by a low level of concentration;
3. Proven record as a reliable source of liquidity in the markets
(repo or sale) even during stressed market conditions; and
4. Not issued by a bank/financial institution/NBFC or any of its
affiliated entities.
5. Adjusted Level 1 assets are arrived at by adding back the
amount of cash lent (reverse repo) and by subtracting the
amount borrowed (repo) upto 30 days against corporate
bonds.
 Level 2 Assets
• Level 2 assets can be included in the stock of HQLA,
subject to the requirement that they comprise no more
than 40% of the overall stock after haircuts.
➢ The portfolio of Level 2 assets held by the bank should be well
diversified in terms of type of assets, type of issuer and
specific counterparty or issuer.
➢ All L2 assets must also be traded in large and deep cash or
repo markets.
➢ A minimum 15% haircut should be applied to current market
value of each L2 asset held in the stock.
• January 2013 – Level 2B assets, consisting of less
liquid securities, with higher haircuts, up to 15% of
HQLA (within the 40% L2 cap).
 Level 2 Assets
1. L2A: Marketable securities that are claims on or claims
guaranteed by sovereigns, PSEs or MDBs with a 20% risk
weight under the Basel II Standardised Approach for credit
risk and not issued by a bank/FI /NBFC or affiliated entities.
2. L2A: Corporate bonds and CP (not issued by bank/FI/NBFC
or affiliated entities), rated AA- or above .
3. Level 2B Assets: RMBS with rating ≥ AA (haircut 25%)
Corporate bonds and CP with rating ≥ BBB- (haircut 50%)
and common shares (haircut 50%)
4. Adjusted Level 2 assets are arrived at by adding the amount
of Level 2 securities placed as collateral (after applying the
haircut) and by subtracting the amount of Level 2 securities
acquired (after applying the haircut).
 Rationale for Adjustment
• It is assumed that the entire cash lent, under a reverse
repo, will return in 30 days’ time. Similarly, any cash
borrowed must be paid back by 30 days. Such cash
will be adjusted against Level 1 assets.
• Once the repo transactions are unwound, all corporate
bonds placed as collateral will be available as Level 2
liquid assets (with a 15% haircut). Similarly, any
corporate bonds accepted as collateral will be treated
as level 2 assets of another bank and must be
deducted (with a 15% haircut).
 High Quality Liquid Assets - Formula
• The maximum amount of adjusted Level 2 assets is
equal to two-thirds of the adjusted amount of Level 1
assets after haircuts have been applied. Any excess of
adjusted Level 2 assets over 2/3rd of the adjusted
Level 1 assets needs to be deducted from the stock of
liquid assets.
➢ Level 1 assets must be at least 60%. Level 2 assets can at most
be 40%. So, Adjusted L2≤(2/3)×Adjusted L1
• HQLA= L1 + L2A + L2B – Max ((Adjusted
L2A+Adjusted L2B) – 2/3 ×Adjusted L1, Adjusted
L2B – (15/85)×(Adjusted L1 + Adjusted L2A), 0)-
Adjusted L2B – (15/60)×(Adjusted L1), 0)
 Cash Inflows and Outflows
• The total net cash outflows is defined as the total
expected cash outflows minus total expected cash
inflows for the subsequent 30 calendar days.
➢ Total expected cash outflows are calculated by multiplying the
outstanding balances of various categories or types of
liabilities and off-balance sheet commitments by the rates at
which they are expected to run off.
➢ Total expected cash inflows are calculated by multiplying the
outstanding balances of various categories of contractual
receivables by the rates at which they are expected to flow in
up to an aggregate cap of 75% of total expected cash outflows
 Cash Inflows and Outflows
• Total net cash outflows over the next 30 days =
Outflows – Min (inflows; 75% of outflows).
• This comprises the denominator of the LCR.
➢ Compliance with LCR ensures that HQLA is at least 25% of
net cash outflows.
➢ HQLA should not be for trading purposes or hedging or credit
guarantees/enhancements.
➢ HQLA exclusively as contingent source of funds.
➢ Need for CFPs to identify the triggers and the sources of
funds.
 Implications
• Make banks hold more liquid assets, to prepare for
short-term liquidity stress. Accumulation of short-
term debt was the trigger for the global crisis.
➢ Make them rank traded assets in descending order of liquidity.
➢ Discourage investments in bonds issued by FIs.
➢ Spur growth of non-financial corporate bonds.
• Demand for liquid bonds will go up further, while
demand for illiquid bonds may fall more.
➢ Aggregate market liquidity may worsen.
➢ Demand for CDs may decline since they are non-HQLA.
➢ Demand for CPs may improve since they are under HQLA.
 Implications
• The run-off factors and horizons are standardized.
➢ Some banks could have more unstable liabilities than others.
➢ Banks borrowing in the repo and call markets can collapse
well before 30 days, while banks reliant on retail deposits
might not face any run-off within a month.
• Make banks aware of the differences in stability
among various sources of funds.
➢ Push them towards branch-based, retail, insured deposits.
➢ The run-off factors (extent of premature withdrawal) for
wholesale deposits are assumed to be much higher.
➢ Interbank lending may decline, since run-off factors for
interbank deposits are the highest.
 Implications
• Banks with higher (lower) HQLA are also likely to
have lower (higher) cost of funds.
➢ Banks with lower HQLA can impose much higher withdrawal
penalties to reduce the outflow of TDs, within next 30 days.
May encourage 31-day noncallable term deposits.
➢ Negative carry on callable deposits to be compared with
premium on noncallable deposits.
➢ Large clients may borrow against such deposits as well.
• Demand for short-term liabilities just beyond the 30-
day window, like 3-month deposits, may increase.
 Margin Comparison

LCR India

L1 assets 0% 10%

L2A assets 15% 22%

L2B assets 50% 59%

 Volumes (Rs. Bn) & Growth Rates

Dec-12 Nov-17 Growth (%)

CP 1,817.7 4,736.8 95.78%

CD 3,327.7 1,218.9 -100.43%

 Recent Developments
• LCR standards made less stringent in January 2013
1. Pool of HQLA allowed to be drawn down below 100% of
NCO (i.e. LCR allowed to fall below 100%), both under
specific and systemic stress, with the knowledge and consent
of the national supervisors.
2. More securities added under HQLA.
3. Outflow rates on some items (e.g. stable deposits) reduced.
4. Phase-in period from 1.1.2015, when minimum LCR is
expected to be 60%.
5. The possibility of using central bank funds, to supplement
LCR, has been recognized.
 FALLCR
• RBI has allowed banks to include G-secs up to 9% of
NDTL, as L1 HQLA, for meeting LCR requirements.
• This is known as Facility to Avail Liquidity for
Liquidity Coverage Ratio (FALLCR).
➢ Can be used to obtain liquidity from RBI only under stress, at
2% above LAF rate for at most 90 days, after exhausting all
other sources.
➢ BIS assessment team feels that this facility makes RBI the
lender of first resort.
➢ Since this is a CLF and can be used after tapping all other
options, it should be treated as an L2B asset.
 RBI Revisions - March 2016
• Corporate Bonds and CP, rated between BBB- and
A+, allowed as L2B assets.
• Run-off factor on guarantees, LCs and Trade Finance
reduced to 3%.
• HUFs, partnerships and trusts with balances up to Rs.
5 Cr. treated as retail deposits with relevant run-off
factors, while other legal entities attract 100%.
• Deposits against loans excluded from outflows, iff: (i)
loans do not mature within 30 days (ii) deposits
cannot be withdrawn before loan repayment and (iii)
amount of deposits < value of outstanding loans.