RISK MANAGEMENT

RISK MANAGEMENT
for Financial Institutions
D.A.Georgoutsos
5. Credit Risk
 5a. Defining the Credit Risk
 The loss from any claim the bank might have, due to the default of the
counterparty to the contract,
E ( L)  e  ( pd )  LGD
 where: (pd) the probability for the default event, E(b)=pd,
b=1(default), b=0( no default)
e : the exposure at default
LGD: Loss Given Default (its value depends on the
value of the guarantees in the case they are liquidated)
 The expected loss is related to the reserves that the bank generates
in its income statement accounts
 The unexpected loss can be estimated from an extreme value of the

probability of default (e.g. if a recession hits the economy). The
difference between the unexpected and the expected loss is known
as the Economic Capital or Capital at Risk ). The Economic Capital
should be accounted for in the capital adequacy of the bank.
 5b. An alternative, more general definition
of the credit risk states that the credit risk
is any loss due to the deterioration of the
creditworthiness of the counterparty.
 Question: How do we estimate the PD

and the LGD ?
 Qualitative models based on subjective

estimate of the probability of default
 The 5 C’s (Character, capacity, capital
collateral, conditions).
 Quantitative Models (Credit Scoring Models)
Linear Discriminant models
 Ζ = α1 Χ1 + α2 Χ2 + α3 Χ3 + α4 Χ4 + α5 Χ5.
Historical Data
Historical data provided by rating agencies

can be used to estimate the probability of
default
Risk Management and Financial Institutions 4e, Chapter 19, Copyright © John C. Hull 2015 9
Cumulative Average Default Rates %
(1970-2013, Moody’s) Table 19.1, page 402
Time (years)
1 2 3 4 5 7 10
Aaa 0.000 0.013 0.013 0.037 0.104 0.241 0.489
Aa 0.022 0.068 0.136 0.260 0.410 0.682 1.017
A 0.062 0.199 0.434 0.679 0.958 1.615 2.759
Baa 0.174 0.504 0.906 1.373 1.862 2.872 4.623
Ba 1.110 3.071 5.371 7.839 10.065 13.911 19.323
B 3.904 9.274 14.723 19.509 23.869 31.774 40.560
Caa 15.894 27.003 35.800 42.796 48.828 56.878 66.212
Interpretation
 The table shows the probability of

default for companies starting with a
particular credit rating
 A company with an initial credit rating of
Baa has a probability of 0.174% of
defaulting by the end of the first year,
0.504% by the end of the second year,
and so on
Do Default Probabilities Increase
with Time?
 For a company that starts with a good
credit rating default probabilities tend to
increase with time
 For a company that starts with a poor
credit rating default probabilities tend to
decrease with time
Hazard Rate vs. Unconditional
Default Probability
 The hazard rate or default intensity is the
probability of default over a short period of
time conditional on no earlier default
 The unconditional default probability is the
probability of default as seen at time zero
 Quantitative Models (Credit Scoring Models)
 Capital Market Models

 Estimating the pd from bond yields
p (1  ri )  (1  ri )
c, j
i
c, j g
Estimating the (pd) from bond yields (cont.)
e.g. p1c , BBB (1  0.065)  (1  0.05)  (1  p1c , BBB )  1.4%
 case where the recovery rate is not zero (k=0.50)
p1c , BBB (1  0.065)  (1  p1c , BBB )k (1  0.065)  (1  0.05)  (1  p1c , BBB )  2.81%
 Calculating the forward probability of default
p2c , BBB (1  f1c, 2 )  (1  f1,g2 )  (1  p2c , BBB )  2.28%
 calculating the cumulative probability of default
p1c , BBB p2c , BBB  0.986  0.972  0.958

Risk-free Rate
 The risk-free rate used by bond traders when
quoting credit spreads is the Treasury rate
 The risk-free rate traditionally assumed in
derivatives markets is the LIBOR/swap rate
 By comparing CDS spreads and bond yields it
appears that in normal market conditions traders
are assuming a risk-free rate 10 bp less than the
LIBOR/swap rates
 In stressed market conditions the gap between
the LIBOR/swap rate and the “true” risk-free rate
is liable to be much higher
Real World vs Risk-Neutral
Default Probabilities
 The default probabilities backed out of
bond prices or credit default swap spreads
are risk-neutral default probabilities
 The default probabilities backed out of
historical data are real-world default
probabilities
Real World vs Risk Neutral Default
Probabilities , 7 year averages (Table 19.5,
page 415)
Possible Reasons for These Results
(The third reason is the most important)
 Corporate bonds are relatively illiquid

 The subjective default probabilities of bond
traders may be much higher than the
estimates from Moody’s historical data
 Bonds do not default independently of
each other. This leads to systematic risk
that cannot be diversified away.
 Bond returns are highly skewed with limited
upside. The non-systematic risk is difficult to
diversify away and may be priced by the
market
Models estimating the unexpected loss (cont.)
 The Credit Risk + model.
 focuses on the event of default
 The probabilities to default of different positions are not related
 the number of defaults, n, in a portfolio follows the Poisson
distribution (m is the average number of defaults)
p (n)  (2.7182  m m n ) / n!
 the distribution of losses depends on the number of defaults and

the Loss Given Default, LGD, or Loss severity.
 Distribution of Losses (source: Linda Allen, Credit Risk Modeling for Middle
markets, Baruch College, 2004)
 The expected loss is equal to:
E ( L)  m  e  LGD
 the average number of defaults is equal to the
variance of defaults.
 The economic capital shows the contribution
the shareholders should make in order to cover
unexpected losses.
 the expected losses are related to the reserves
in the incomes statement.
Determinants of the recovery Rates
Recovery Rate
The recovery rate for a bond is usually
defined as the price of the bond 30 days
after default as a percent of its face value
Recovery Rates; Moody’s: 1982 to 2013
(Table 19.2, page 404)
Class Ave Rec Rate (%)

Senior secured bond 52.2
Senior unsecured bond 37.2
Senior subordinated bond 31.0
Subordinated bond 31.4
Junior subordinated bond 24.7
Recovery Rates Depend on
Default Rates
 Moody’s best fit estimate for the 1982 to
2007 period is
Ave Recovery Rate =
59.33 − 3.06 × Spec Grade Default Rate
 R2 of regression is about 0.5
Credit Default Swaps (pages 404-409)
 Buyer of the instrument acquires protection from the

seller against a default by a particular company or
country (the reference entity)
 Example: Buyer pays a premium of 90 bps per year
for $100 million of 5-year protection against company
X
 Premium is known as the credit default spread. It is
paid for life of contract or until default
 If there is a default, the buyer has the right to sell
bonds with a face value of $100 million issued by
company X for $100 million (Several bonds may be
deliverable)
CDS Structure (Figure 19.1, page 406)
90 bps per year
Default Default
Protection Protection
Buyer, A Seller, B
Payoff if there is a default by
reference entity=100(1-R)
Recovery rate, R, is the ratio of the value of the bond issued

by reference entity immediately after default to the face value
of the bond
Other Details
 Payments are usually made quarterly in arrears
 In the event of default there is a final accrual
payment by the buyer
 Increasingly settlement is in cash and an auction
process determines cash amount
 Suppose payments are made quarterly in the
example just considered. What are the cash
flows if there is a default after 3 years and 1
month and recovery rate is 40%?
Attractions of the CDS Market
 Allows credit risks to be traded in the

same way as market risks
 Can be used to transfer credit risks to a
third party
 Can be used to diversify credit risks
Credit Indices (page 408)
 CDX IG: equally weighted portfolio of 125
investment grade North American companies
 iTraxx: equally weighted portfolio of 125
investment grade European companies
 If the five-year CDS index is bid 165 offer 166 it
means that a portfolio of 125 CDSs on the CDX
companies can be bought for 166bps per
company, e.g., $800,000 of 5-year protection on
each name could be purchased for $1,660,000
per year. When a company defaults the annual
payment is reduced by 1/125.
Use of Fixed Coupons
 Increasingly CDSs and CDS indices trade
like bonds
 A coupon and a recovery rate is specified
 There is an initial payments from the buyer
to the seller or vice versa reflecting the
difference between the currently quoted
spread and the coupon
Credit Default Swaps and Bond
Yields (page 409-410)
 Portfolio consisting of a 5-year par yield

corporate bond that provides a yield of 6% and a
long position in a 5-year CDS costing 100 basis
points per year is (approximately) a long position
in a riskless instrument paying 5% per year
 What are arbitrage opportunities in this situation
is risk-free rate is 4.5%? What if it is 5.5%?
Return on a loan
 Example
 Basic interest rate = 12%

 Risk Premium (credit risk) = 2%
 Cost of raising funds = 9%
 E(pd) = 3%
 LGD = 30%
 ROA = 12% + 2% - 9% -[3% (30%)] = 4.1%
Risk Adjusted Return on Capital (RAROC)
Expected net income (loan)
RAROC 
Capital at Risk
Capital at Risk (CaR)99% = L99% - E(L)
Expected interest= 0.12 x $1,000,000= $120,000

Servicing fees= 0.0050 x $1,000,000= $5,000
Provisions = 0.0100 x $1,000,000 = -$10,000
Less cost of funds= 0.10 x $1,000,000=-$100,000
Net interest and fee income = $15,000
RAROC = $15,000/(0.08x 50% x 1,000,000) = 37.5 percent. Since

RAROC is higher than the required rate of return, 10%, the bank
should make the loan.
Credit Risk on Off-balance sheet, OBS
 The credit equivalent of an OBS item

Current replacement cost – Future potential loss
Example: A F.I. holds a long position in USD and a short position in € in
a FX Forward contract that will be settled on 31/8/18. The strike price of the
contract is 1.25 USD per 1 €. Assume that today, 31/5/18, a corresponding
contract, that it is settled on 31/8/18, is priced at 1.20 USD. It is also known
that the 3 months Libor rate on USD is 5% and the spot exchange rate is
1.24 USD per 1 €. What is the current replacement cost ?
Answer: In case the counterparty defaults today the replacement cost as of
31/8/07 will be equal to:
(1.25-1.20)χ 1 mil. €= 50,000 USD.
The current replacement is equal to:
(50,000 USD) / (1+0.05(3/12))= 49,382 USD.
In euros the current replacement is equal to :
49,382 USD / 1.24 = 39,824 €.
Estimation of Credit Risk in portfolios of
claims
 Expected Loss, due to default of the counterparty, and its
variance E ( L p )  e A ( pd A ) LGD A  eB ( pd B ) LGDB
and E (bA )  pd A , bA 10((default )

nodefault )
 2 ( LP )  e A2 2 (bA ) LGD A2  eB2 2 (bB ) LGDB2 

2e AeB (bA ) (bB ) LGD A LGDB   ,
 2 ( LP )  cov[(e A pd A LGD A ), LP ]  cov[(eB pd B LGDB ), LP ] 
 cov( LA , LP ) cov( LB , LP ) 
 2 ( LP )   .
  2
( LP )  2
( LP ) 
 Management of the credit Risk through credit

derivatives (e.g. credit default swaps) and / or
securitization
Estimation of Credit Risk in portfolios of
claims (cont.)
 The problems with the procedure above are: a) the accurate use of
date e.g. the probabilities of default, the correlation of defaults, b)
the back-testing of the model

Differentiation index: 1/ 2
 (0.2  0.3) 2  (0.2  0.3) 2  (0.5  0.3) 2  (0.1  0.1) 2 
     0.1224
 4 
 Concentration index:
 claims to sec tor 
Equity  0.10  (claims to sec tor )  0.30     0.10  0.333.
 equity  0.30
 
Capital Adequacy Rules (Basle Rules)
 Supervision of the Banking Industry

 Regulation
 Monitoring
 Supervision
 What is the best degree of supervision ?
 Why is the Banking Sector supervised;
 Asymmetry of information between depositors and bank
managers
 The reduction of the intervention cost in case of bank failures
 The social cost in case of a “big” bank failure exceeds the
private cost (shareholders – creditors)
Capital Adequacy (cont.)
 Basle I (1998)
 The regulatory capital should be at least equal to 8% of the risk weighted assets
CRC≥8%x (risk weighted assets)
Table 1: Risk weights

0% Cash, claims against OECD governments et. cet.
20% Claims against banks in OECD countries et. cet.
50% Mortgage loans et. cet.

100% Remaining claims
 Risk weights for Off- Balance sheet items – conversion to credit
equivalent amounts
Table 2: Conversion coefficients
100% Letters of Credit et. cet.
50% Letters of credit connected with a specific transaction (e.g. imports) et. cet.
20%
0% Not used credit facilities et. cet.
 Capital Adequacy for Off-balance sheet derivatives
CE=NRC+ Potential exposure {Nominal value x coefficient

x (0.4+0.6 x NGR)
NRC=Net Replacement Cost, NGR = Ratio of net to gross
positions
 The Capital Adequacy will equal:
CRC (OBS )  8%  {Weighting coefficient  CE}
 Regulatory Capital
 Tier I Capital (at least 4% of risk weighted assets)
 Upper Tier I consists of : Paid-up share (ordinary) – Disclosed reserves (e.g. share
premium reserve, retained earnings et. cet.)
 Lower Tier I : Preference shares (no more that 15% of Tier I)
 Tier II Capital (no more that 100% of Tier I)

 Upper Tier II: undisclosed reserves, Revaluation Reserves, General Loan Loss
provisions, Hybrid Capital Instruments (subordinated, not redeemable at the holder’s
initiative, can be deferred if the bank’s profits do not allow payment et. cet. )
 Lower Tier II: Subordinated debt (original term to maturity of at least 5 years) – no more
than 50% of Tier I
 Amendment to the Basle Accord, 1996) (Π.Δ./Τ.Ε. 2397/1996) –
positions in the Trading portfolio
 Standardized Approach
 Internal Approaches ( Var Methodology)
60
VaR t 1
MRC  max{VaR t 1 , k i 1
}  SR
60
 VaR is estimated at the 99% level of significance and has been converted to
cover a 10-days period
 k≥3
 There is provision for a Tier III regulatory capital which consists of
subordinated debt with a maturity of at least 2 years. Also Tier I (for the market
risk) + Tier III > MRC
 Capital Adequacy for the Market risk (standardized Approach)

 Capital Adequacy according to Basle I
 Problems with Basle I rules

 Dis-incentives to lend companies with high
creditworthiness
 Incentives to lend governments
 Regulatory Capital Arbitrage, RCA (i.e. securitization)
 The building blocks approach does not accommodate

for the non-perfect correlation of claims
 It doesn’t provide for the existence of methods that
reduce credit risk
 The structure of Basle II

Pillar I
 Standardized Approach
Table 3: Risk weights

Classification based on ΑΑΑ ως ΑΑ- Α+ ΒΒΒ+ ΒΒ+ ως Β- Below Β- Unrated
Standard & Poor’s ως ως
Α- ΒΒΒ-
Central Governments 0% 20% 50% 100% 150% 100%

Banks 1 20% 50% 100 % 150% 100%
2 20% 50% 50% 100% 150% 50%

Corporations 2 20% 50% 100% 100% 1 150% 1 100%
Securitized assets 20% 50% 100% 350% Deduction from regulatory
capital
Residential Real Estate 35%
Mortgages
Retail Banking 75%
( 1 ) 100% down to BB- , 150% starting from B+ , ( 2 ) past due loans (payments overdue by over 90 days) carry a 150% weight
 Credit risk mitigation methods

 guarantees offered on the loans (e.g. government bonds, listed shares
et. cet.)
 credit derivatives
 real estate
 Reduction of needed regulatory capital

 Simple method – for the capital adequacy use the guarantor weights
 Comprehensive approach - haircuts
C
CA 
1 H
 Problems with the standardized aproach
 Companies in Europe are not rated
 Companies below ΒΒ- have an incentive not to be

rated
 rating shopping
 The Credit Rating Agencies (CRAs) provide ratings

which are not accurate (not forward looking)
 There isn’t a close relation between the weights of the
rating classes and the realized default rates.
 Capital Adequacy (cont.)
 Internal Models
 The F.I. are given continuous functions which allow the
calculation of the risk weights on the basis of inputs supplied
by the banks
 classification of claims in risk categories
 Estimation of the probability of default, for one year, for each risk
category (the probability that assets’ value will be lower within a year
than a default point (Merton’s approach). Returns are determined by a
single- factor model:
Z i  wZ  ( 1  w2 ) i
 other parameters: e (exposure at default), LGD (Loss Given Default),
M (Maturity – calculated as Duration with a zero interest rate – the risk of
downgrading increases with maturity), Granularity (expresses the degree
of concentration in the portfolio) , correlation (calculated for a few
standard categories of borrowers – not individually).
 Foundation Internal Ratings Based, (IRB)
method
 The F.I. estimates the probability of default for each
risk group but the supervisory authority provides
estimates for the e (exposure), LGD (Loss Given
Default), M (Maturity)
 Advanced Internal Ratings Based method
 F.I. calculate, themselves, the inputs e, LGD, M
and can use any form of collateral.
 Those groups are: “corporate, banks, sovereigns”, “Residential real estate mortgage
loans”, “Retail exposures”, et. cet.
 The Risk weights are given by the formula:
  N 1 ( pd )  w  N 1 (0.1%  

RW  b  LGD  N  
  pd  LGD 
  1  w 2
 
Where: w2 = ρ (correlation), b is a formula that expresses the effect of maturity, N stands for
the standardized normal distribution and the losses are calculated under the assumption
that they will be exceeded only in 0.1% of all cases.

Comparison of capital requirements between Basle I and II (source: P.Jackson, Bank of England, Quarterly Bulletin, spring 2003)
Table 4:Capital required as a percentage of exposure
PD Standardized Standardized Internal Models –

(Basle I) (Basle II) Foundation
Approach
ΑΑΑ 0.03 8 1.6 1.13

ΑΑ 0.03 8 1.6 1.13
Α 0.03 8 4.0 1.13
ΒΒΒ 0.20 8 8.0 3.61
ΒΒ 1.40 8 8.0 12.35
Β 6.60 8 12.0 30.96
CCC 15.0 8 12.0 47.04
 Problems with the use of internal models

 The measurement of the risk should apply on a horizon over
which the assets ate not expected to change – one year is a very
short horizon
 The pro-cyclicality factor : in recession the default risk is higher
and the internal models lead to higher capital adequacy ratios.
This retains liquidity from the market and in doing so it intensifies
the recession problem.
 Capital Adequacy on Operational Risk

 Basic Indicator Approach
n
 (a  GI i )
K BIA  i 1
 Standardized Approach
n 8
 max[{ ( β  GI i )},0]
K SA  i 1 i 1
n
Coefficient β takes values from 12% - 18%.

 Advanced Measurement Approaches, ΑΜΑ
 scorecard approach
 Internal Measurement Approach
ELij  EI ij  PEij  LGEij

ORC ij  γij  ELij
 The Loss Distribution Approach

VaR ij  ELij  ULij
CAPITAL ADEQUACY
 Pillar II : A New Role for supervisory authorities
 Supervisors should check for the following:
 Compliance with the requirements
 control of risks which are not quantified in a proper way (e.g.
concentration risk)
 risks which are not included in pillar I (e.g. interest rate risk in the
banking book)
 the role played by some bank-external factors
 Pillar III: Disclosure and Market Discipline

 size and composition of capital and risky assets
 risk measurement and control systems
 Credit exposures among different pd groups and the default rate
recorded on each rating bucket
CAPITAL ADEQUACY
 Basle ΙΙΙ
o Micro-prudential side
 coverage of additional risks (trading book exposures, counterparty credit risk,
securitization activities)
 stressed Value at Risk – one-year observation period relating to significant losses
 over the counter derivatives will be subject to increased capital needs
 Re-securitization exposures (e.g. CDOS) will have a different treatment from positions on simple
securitizations
 Increased Capital Adequacy (full implementation from 2019)

 Tier I από 4% σε 6%
 Common equity από 2% σε 4.5%
 Conservation Buffer (2.5% το 2019 – αρχή από 0.6% το 2016)
 Συνολικό Tier I κεφάλαιο σε 8.5% το 2019
 Tier II capital from 4% falls to 2%
 Countercyclical capital 0%-2.5% (2019)
 Total regulatory capital 13%
60
CAPITAL ADEQUACY
 Basle ΙΙΙ
 Quality of regulatory capital
o Restriction of DTAs (Deferred Tax Assets up to 15% of the common equity
capital)
o Limited use of hybrid products
o Minimum Liquidity Standards
o Liquidity Coverage Ratio (stock of high quality liquid assets to Net cash
outflows over a 30 day stressed period should be higher than 100%).
o Net Stable Funding Ratio (available amount of stable funding to required
amount of stable funding should be higher than 100%).
o Macro-prudential regulation
 Capital Buffers
 Leverage Ratio ( Tier I capital to Total exposures higher than
3%)
61

RISK MANAGEMENT - ERASMUS - 2019 - Bplus

Uploaded by

Copyright:

Available Formats

RISK MANAGEMENT - ERASMUS - 2019 - Bplus

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

RISK MANAGEMENT - ERASMUS - 2019 - Bplus

Uploaded by

Copyright:

Available Formats

for Financial Institutions

 The unexpected loss can be estimated from an extreme value of the

 Question: How do we estimate the PD

 Qualitative models based on subjective

Linear Discriminant models

Historical data provided by rating agencies

Aa 0.022 0.068 0.136 0.260 0.410 0.682 1.017

A 0.062 0.199 0.434 0.679 0.958 1.615 2.759

Baa 0.174 0.504 0.906 1.373 1.862 2.872 4.623

Ba 1.110 3.071 5.371 7.839 10.065 13.911 19.323

B 3.904 9.274 14.723 19.509 23.869 31.774 40.560

Caa 15.894 27.003 35.800 42.796 48.828 56.878 66.212

 The table shows the probability of

 Capital Market Models

e.g. p1c , BBB (1  0.065)  (1  0.05)  (1  p1c , BBB )  1.4%

 case where the recovery rate is not zero (k=0.50)

 Calculating the forward probability of default

p2c , BBB (1  f1c, 2 )  (1  f1,g2 )  (1  p2c , BBB )  2.28%

 calculating the cumulative probability of default

p1c , BBB p2c , BBB  0.986  0.972  0.958

 Corporate bonds are relatively illiquid

 the distribution of losses depends on the number of defaults and

Class Ave Rec Rate (%)

 R2 of regression is about 0.5

 Buyer of the instrument acquires protection from the

90 bps per year

Recovery rate, R, is the ratio of the value of the bond issued

 Allows credit risks to be traded in the

 Portfolio consisting of a 5-year par yield

 Basic interest rate = 12%

Capital at Risk (CaR)99% = L99% - E(L)

Expected interest= 0.12 x $1,000,000= $120,000

RAROC = $15,000/(0.08x 50% x 1,000,000) = 37.5 percent. Since

and E (bA )  pd A , bA 10((default )

 2 ( LP )  e A2 2 (bA ) LGD A2  eB2 2 (bB ) LGDB2 

 Management of the credit Risk through credit

 Supervision of the Banking Industry

Table 1: Risk weights

50% Mortgage loans et. cet.

Table 2: Conversion coefficients

100% Letters of Credit et. cet.

 Capital Adequacy for Off-balance sheet derivatives

CE=NRC+ Potential exposure {Nominal value x coefficient

 Tier II Capital (no more that 100% of Tier I)

 Capital Adequacy for the Market risk (standardized Approach)

 Problems with Basle I rules

 Regulatory Capital Arbitrage, RCA (i.e. securitization)

 The building blocks approach does not accommodate

 The structure of Basle II

Table 3: Risk weights

Central Governments 0% 20% 50% 100% 150% 100%

2 20% 50% 50% 100% 150% 50%

 Credit risk mitigation methods

 Reduction of needed regulatory capital

 Companies below ΒΒ- have an incentive not to be

 The Credit Rating Agencies (CRAs) provide ratings

Table 4:Capital required as a percentage of exposure

PD Standardized Standardized Internal Models –