RLMdridEco-RegCapital

8/8/2019 RLMdridEco-RegCapital

1/31

1

Company Confidential 2001 Algorithmics Inc.

Enterprise Credit Risk

Management Framework

for Economic &

Regulatory Capital

RiskLab Madrid

Madrid, November 14 2002

Dr. Dan Rosen

VP Marketing, Algorithmics Inc.

[email protected]


Enterprise PortfolioCredit RiskModelling

RiskLab InternationalConference,

Madrid, October 18 2001

Enterprise credit risk framework

integrate credit risk

integrate market and credit

valuation and MtM

portfolio credit risk management

Modelling exposures/LGD accurately

is key for accurate PCR measurement

ECR Framework solid basis for

managing and reconcilingregulatory and economic capital

pillar I I

providing transparency (Pillar III)

Last Years Talk..


2/31

2


Last Year - Case Study

Standard portfolio c redit risk models assume deterministic exposures (LGDs)

If stochastic, they are generally assumed independent

Case study: demonstrates impact of stochastic & correlated exposures on

credit capital (results generally extend to LGDs & collateral)

Substantial effect of vols., credit & market-credit correlations, granularity

Results have important implications for both economic and regulatory c apital

Substantial benefits from

A flexible, integrated market and credit-simulation model analytical

approximations generally show strong limitations

Accurate models for exposures, LGDs, collater al and other mitigation techniques


Summary todays talk

Business Requirement:

Enterprise Risk Framework evolving BIS II requirements

Risk architecture - scalable

Measure and manage market, credit, operational risk (ALM, & collateral)

Enterprise coverage: consistent treatment of banking and trading books

Risk engine

Compute and reconcile economic and regulatory credit capital

Analyze risk along multiple dimensions: consolidate disparate exposure acrossmultiple business lines, portfolios and produc ts

Reporting infrastructure: comprehensive enterprise risk reporting Flexible drill-down analysis and reporting in many dimensions

Accommodate users at every level of the firm, and multiple regulators

Integration of global market & credit risk information into business processes


3/31

3


Outline

Introduction


Management

BIS II and Enterprise Credit Risk

Enterprise Framework for Regulatory

& Economic Capital

Data Architecture

Risk Engine

Reconciling Regulatory &

Economic Capital

Reporting Infrastructure


Financial Institution

Trading BookBanking Book

RetailRetail Commercialmedium/small

Commercialmedium/small

CommercialLarge

CommercialLarge

mortgagesmortgages Credit

cards

Credit

cardsLines ofcredit

Lines ofcredit

Corporates(Public and

Private)

Corporates(Public andPrivate)

SectorsSectors SectorsSectors

SectorsSectors

PrivateFirms

Private

Firms

SectorsSectors

DerivativesCounterparties

DerivativesCounterparties

SovereignBond Issuers

SovereignBond Issuers

CorporateBond Issuers

CorporateBond Issuers

CreditDerivatives

CreditDerivatives



4/31


Obligor Creditworthiness Analysis

Instrument Valuation

Transaction Management

CounterpartyExposures

Measurement & Control

Portfolio

Management

Enterprise Credit Risk Functions



Instrument ValuationTransaction Management



Portfolio

Management


Sovereign

Public firms

Private: large & medium

Small businesses

Retail consumers


5/31

5



Instrument Valuation

Transaction Management



Portfolio

Management


Derivatives

Credit Derivatives

Bonds

Syndicated loans

Large corporate loans

Mi ddle & small market

Retail

- collateral

management






Portfolio

Management


M easurement and limits

Aggregation of positions by

- obligor/counterparty

- sector

- country, etc.

Derivatives:

- actual & potential exposures

Mitigation

- netting, collateral, etc.


6/31

6


SAPPHIRE - AA

0.0

30.0

60.0

90.0

6/4/97 6/4/01 6/4/05 6/4/09 6/4/13 6/4/17

Time

CreditExposure

(Millions)

SAPPHIRE - AA

0.0

30.0

60.0

90.0

6/4/97 6/4/01 6/4/05 6/4/09 6/4/13 6/4/17

Time

CreditExposure

(Millions)

TURQUOISE - AA

0.0

20.0

40.0

60.0

80.0

6/4/97 6/4/01 6/4/05 6/4/09 6/4/13 6/4/17Time

CreditExposure

(Millions)

TURQUOISE - AA

0.0

20.0

40.0

60.0

80.0

6/4/97 6/4/01 6/4/05 6/4/09 6/4/13 6/4/17Time

CreditExposure

(Millions)

Exposure Profiles & Limits

Counter Party Exposure Limits






Portfolio

Management


Portfolio credit risk capital

- economic & regulatory

Portfolio Management tools

- risk contributions

-marginal risk

-capital allocation

-performance

-optimization & efficient frontiers


7/31


Portfolio Credit Risk Reports

Unexpected losses (99.5%)

Expected losses


BIS II Proposal for new capitaladequacy framework

Three pillars:

M inimum capital requirements

gives the explicit rules that define the minimum ratio of capital to risk

weighted assets

Supervisory review process

requires supervisors to undertake a qualitative assessment of capital

allocation techniques and compliance w ith standards actually in place in

an institution

Market discipline high disclosure standards & adequate capital which facilitate market

discipline


8/31

8


BIS II

BIS II: proposal for new Capital Accord

Foster a strong emphasis on risk management practic es

Encourage ongoing improvements in banks risk assessment capabili ties

Regulatory framework c overs

Credit Risk, Mar ket Risk of trading act ivities, and Operational Risk (Pillar I -

minimum capital requirements)

Interest rate management (ALM) and liquidity risk, Collateral M anagement

(supervisory reviews)

Implementation currently scheduled for end of 2006

Requires substantial resource commitments on the part of banks andsupervisors


Minimum Capital Under BIS II

Summary of minimum capital requirements

Three approaches to calculation of risk-weighted assets:

(Revised) standardized approach

Foundation internal ratings-based (IRB) approach

Advanced Internal ratings-based (IRB) approach

Explicit capital charge for operational risk

Market risk capital as defined in the 1996 Amendment to remain

largely unchanged


9/31

9


Portfolio Credit Risk & BIS II

Credit r isk models for the banking book

- Although portfolio credit risk models are not allowed for the

calculation of minimum capital requirements,

- The functional form and coefficients of the BRW and GA already embed

portfolio credit risk model

- Satisfying Pillar II w ill likely require that institution on the advanced IRB

approach have implemented in practic e a portfolio c redit risk

management system


GA = Granularity AdjustmentRWj = Risk Weight for asset/obligor j

Ej = Exposure at default for asset/obligor j

RW x E x 8% represents the capital for a perfectly diversified

portfolio (asymptotically fine grained; with only systemic risk)

The granularity adjustment adjusts the capital for the level ofdiversification of the actual portfolio

now likely to be out of Pillar I

BIS II Advanced IRB Approach

GARWEn

j

jj +

= %8CapitalRegulatory


10/31

10


For example, the RW for an exposure to a corporate obligor :

wherePD= obligors probability of default

LGD= loss in event of defaultM= maturity of transaction

b(PD) = sensitivity of the maturity adjustment to M (from a calibration)N(x) = c df for standard normal random variableG(x) = inverse cdf for standard normal random variable


G(PD) + R0.5x G(0.999)RW(PD,LGD,M) =[1 - 1.5 x b(PD)]

[1 + (M - 2.5) x b(PD)]12.5 x LGD x N[ (1 - R)0.5 ]x




[1 + (M - 2.5) x b(PD)]12.5 x LGD x N[ (1 - R)0.5 ]x

Adjustment for LGD

Expected & unexpected default losses of ahypothetical, asymptotically fine-grained, portfolioof one-year loans (from one-factor credit portfoliomodel)

Maturity adjustment to reflect aportfol io of maturity 2.5 years

Used to offset 8% capital charge


11/31

11


Enterprise Credit Solution:Architecture

Obligor Creditworthiness Data

RatingsPD/TM

LGD Creditcorrelations

Internal Systems External Systems

Obligorrelationships

MarketData

CreditDrivers(factors)

IRs. FX,EQ., etc.

Bonds:Prices/spreads

Loans:Prices/spreads

CreditDerivs.

Inter

nalSystems

ExternalSystems

CollateralGuaranteesMitigation

Terms &Conditions

Exposures

Positions

InternalSystems

T ra

ns

ac

tionD

ata

Mapping Interface (extract , map, load)

M

appingInterface M

appingInterface

Data Staging, Results Management Database

Obligor

Transaction

Collateral

Market

Input DB

Standard

Regulatory

Capital

Report DB

Exposure/M tMBIS PCR Limits

Financial Engines

CollateralManagement


Credit Data Architecture

BIS II advanced data infrastructure: collect, aggregate, validate and

reconcile enterprise-wide credit data - one common data architecture.

Comprehensive product coverage of trading and banking book exposures

Default/migration and LGD data

Credit mitigation including netting, collateral and guarantees

Advanced counterparty and own bank data structures

Credit information must fundamentally be integrated into business processes

Limits management, capital allocation, pricing & origination, performance

Execution: development of a comprehensive solution in a timely manner Must start data collecting efforts before architecture is completed

Leverage: substantial cost savings through the development of common

infrastructure for global limits management and economic c apital.


12/31

12


Enterprise Credit Solution: Economicand Regulatory (BIS-II) Capital


Internal Bank Systems

RatingsPD/TM

Estimation/Calibration

LGDs Creditcorrelations

RatingsPD/TM

LGD Creditcorrelations

External Bank Systems

Retail Corporates SovereignsFinancial

Mapping Interface (extract, map, load)





CreditDrivers

(e.g. Macrofac tors)

Da ta

Se

lectio

nan

dCali

bratio

n

IRs. FX,

EQ., etc.

BondPrices/spreads

MarketData

LoanPrices/spreads

CDPrices

(e.g. Default Swaps)

M

appingInterface(extract,map,

load)

InternalBankS

ystems

ExternalSystems


13/31

13



T ra

ns

ac

tionD

ata

M

appingInterface(extract,map,load)

Data

consolidationCollateral

GuaranteesMitigation

Terms &Conditions

Exposures

PositionsInternalBankSystems


BIS II IRB & Portfolio CreditBuilding blocks

The minimum capital calculation requires

Probabilities of default for each obligor (PD)

Maturity of each transaction (M)

Exposure at default for each transaction (EAD)

Loss given default for each transaction (LGD)

Credit Mitigant information

Full portfolio c redit risk modeling further requires

Obligor correlation model Full MtM for each transaction (for MtM models)


14/31

1


Regulatory & Economic CapitalEngine

Requirement: comprehensive capital engine

Flexible (evolving BIS II requirements & multiple regulatory environments)

BIS I, BIS II (standard, foundation IRB and advanced IRB)

smooth transition to advanced IRB consolidated reporting of BIS I and

multiple BIS II approaches for tw o years

Integrated - regulatory, economic capital and Credit MtM

Coverage - enterprise coverage of trading, corporate and retail exposures

Comprehensive treatment of CP structures, credit mitigation & collateral

Trading CP exposures: advanced internal models using full simula tion Risk management tools:

scenario analysis and stress testing

robust risk decomposition, Hot Spots and optimization analytics


Example: syndicated deal (14/04/00): $115 M to fund acquisition of PlayCore

Holdings Inc. (unrated holding company: interests in sporting and games)

$30million revolver, $25 million term loan A, $60 million term loan B.

Secured credit: 85% of eligible accounts receivable, 60% of elig ible

inventories, plus $3,000 monthly from November through M arch

Covenants require hedging of IR risk, minimum fixed-charge coverage

ratios, limitat ions on dividends, etc.

Pricing tied to: Funded debt / EBITDA In default, pricing increases by 200 bps

Prepayment w ithout penalty at any repricing date.

Credit Risk Pricing: banking loans


15/31

15


Term-loan B c omponent (marketed to loan funds):

Maturi ty July 1, 2006 (87 months term)

20 quarterly payments of $150,000, starting on October 1, 2000

Followed by eight quarter ly payments of $7,125

Loan amortization over several quarters

Initially, facility priced at

PRIME + 225bps (LIBOR+ 400bps)

Pricing grid determines pricing

Credit Risk Pricing: banking loans


Modeling a Bank Credit Facility

Choice of credit from among a set

instrument types:

a term loan

a funded revolving line

a letter of credit

bankers acceptance.

Vital to model cash flows accurately


16/31

16


Modeling a Bank Credit Facility


Modeling Embedded Options

Default option: in default, borrower may not pay an obligation in full

affects CFs explicitly through the probability of default

Prepayment option: right to prepay or cancel the contract before maturity

affects CFs explicitly through the probability of prepayment

function of obligor credit state, risk-free interest rates and spreads

contingent on credit events other than default (e.g. credit down grades)

Credit line utilization option: right to choose the usage level of a commitment

affects implicitly several CFs and outstanding amounts - as obligorscreditworthiness diminishes, draw on credit line increases

embedded option on credit events other than default (e.g. downgrades)


17/31

1


Credit Valuation Framework

In summary:

The cash flows from credit facilities are a function of: borrowe r

creditworthiness (e.g., risk rating), interest rates and credit spreads.

e.g. a decrease in interest rates or cr edit spreads or an improvement in

borrow er risk rating may trigger prepayment

credit facilities include pricing grids, graduated utilization fees and

amortization schedules

Underlying credit risk model must describe each state of the world by

obl igor creditworthiness (e.g. a ratings and default probabilities) the term structure of default-free interest rates

the term structures of credit spreads for non-defaulted securities.


Large Corporate Example: $10 MillionPrimary Participation in Playcore

Playcore 7-Year Term Loan B Tranche: B- Counterparty

NPV Duration*

Base Case Valuation -$267k 2.31 years

No Prepayment: -$126k 4.87 years

Prepayment Option $141k

No Pricing Grid -$270k

No Amortization (NPV) -$286k

Key point: substantial impact on value of loan structure components

(NOTE* Duration is risk and opt ion-adjusted)


18/31

18


Evolution of Credit ExposureMeasurement

First level (Common practice): Exposure = Not ional

simple, easy

traditional loan products

handles derivatives as loans

does not consider potential exposures

Second level: Exposure = MtM + Add On (potential exposure) (BIS)

easy to implement

better for derivatives

may not capture properly offsets, netting, mitigation

Third level: Exposure profile over time - Simulation

accurate for derivatives... but c omputationally intensive

multiple time limits


Mark-to-Future

Values

Counterparty Exposures throughMark-to-Future

Scen

arios

Scen

arios

Mark-to-Future

Instruments

aggregation, netting, collateral,aggregation, netting, collateral,

credit mitigation, etc.credit mitigation, etc.

Mark-to-Future

CounterpartyPortfolios

CounterpartiesCounterparties

Sce

narios

Sce

narios

SecuritiesSecurities


19/31

19


Exposure: Importance of Netting


A Simple NettingExample: Daily collateral calls

Instantaneous collection(payment) of collater al

Static CSA variables

Result:

Exposure to CP is capped at

$35M Threshold level

Daily Margin Calls flowdirectly through to CollateralBalance

Exposure: Importance of Collateral


20/31

20


Reality is not soSimple:

3 day lag for receivingcollateral but must postcollateral immediately (CPExposure drift s beyond Thresholdas you wait for collateral)

23%

Thresholds are only checkedweekly (CP Exposure drifts aroundThreshold between call dates)

30%

CP downgraded during horizon(CP Exposure reduc ed due to loweracceptable threshold)

32%

Exposure: Importance of Collateral


Mark-to-Future Framework forPortfolio Credit Risk

1. Scenarios:

market factors

credit drivers

. . . . .

. . . . .

2. Conditional obligor

default probabilities

3. Obligor scenariolosses

(exposures X LGD)

4. Conditional

portfolio losses

. . . . . . . . . .

. . . . .

. . . . .

5. Unconditional Portfoli o

loss distribution

+

+

_________


21/31

21


1. Scenarios 2. Conditionalprobabilitiesp(X)

Systemic and idiosyncraticPortfolio Losses

X=x1

X=x3

X=x2

pj(X=x1)

j=1,,n

pj(X=x2)

j=1,,n

pj(X=x3)

j=1,,n

3. Obligorlosses l(X)

lj (X=x1)

j=1,,n

lj (X=x2)

j=1,,n

lj (X=x3)

j=1,,n

4. Conditional(total) portfoliolosses P(L= l|X)

5. Unconditional Portf olio

loss distribution

4a. ConditionalSystemic losses

E{L|X}

+

+

}{

}|{

3

1

3

xpl

xXLE

n

j

jj

=

==

}{

}|{

2

1

2

xpl

xXLE

n

j

jj=

==

}{

}|{

1

1

1

xpl

xXLE

n

j

jj=

==


Portfolio Credit Risk:Regulatory Capital Breakdown

Regulatory Capital

Default

MtM/Migration

GARWEn

jjj +

= %8CapitalReg

Systemic

Idiosyncratic (GA)

BIS II formulae provide a transparent w ay to break dow n regulatory capital calc ulations.


[1 + (M - 2.5) x b(PD)]12.5 x LGD x N[ (1 - R)0.5 ]x

GA = (TE / n* ) x (0.6 + 1.8 x LGDAG) x (9.5 + 13.75 x PDAG/ FAG)

where


22/31

22


Portfolio Credit Risk: Economic& Regulatory Capital

Economic Capital

Regulatory M odel

single factor

Integrated market-credit

Multi-Factor, X

(best) Single-Factor, X

StandardMulti-Factor, X

Systemic Idiosyncratic

Default MtM/Migration

Regulatory Capital

Systemic Idiosyncratic (GA)

Default MtM/Migration


Comprehensive enterpriseReporting Infrastructure

Integrated -supports all enterpr ise aspects of an organizations credi t risk

Global limits and exposures; economic and regulatory capital; MtM

Consolidation model and data management tools (support definition of complex CPdata and hierarchies; management of current and future exposures)

Historical analysis and auditing

Enterprise Business support

Desk level (limits & Analytic s), Middle-off ice risk management (Enterprise analysis &

drilldown), senior Management (exec dashboards, capital allocation, Raroc)

Costumizable: analyze r isk along multiple dimensions

Flexible slice & dice funct ionality (capital, CP information, exposures &, limits)user-specified criteria (industry, country, rating, product type)

Extensive drill-down capabilities

Flexible and extensible framework

Web-based reporting tools

Support multiple reporting technologies and user reporting requirements


23/31

23


Reporting Infrastructure


Ratings

PD/TMLGD Credit

correlations

Internal Systems External SystemsObligor

relationships

MarketData

CreditDrivers(factors)

IRs. FX,EQ., etc.

Bonds:Prices/spreads

Loans:Prices/spreads

CreditDerivs.

Interna

lSystems

ExternalSystems

Collateral

Guarantees

Mitigation

Terms &Conditions

Exposures

Positions

InternalSystems

T ra

nsa

ct io

nD

ata

Mapping Interface (extract , map, load)

M

appingInterface M

appingInterface

Data Staging, Results Management Database

Obligor

Transaction

Collateral

Market

Input DB

Standard

Regulatory

Capital

Report DB

Exposure/M tMBIS PCR Limits

Financial Engines

CollateralManagement

Enterprise Reporting Infrastructure


Enterprise Capital Report


24/31

2


Credit Capital Report


Credit Capital Report


25/31

25


20K MC scenarios

Credit Reports: Loss Distribution

Expected LossLoss (99%)

Credit VaR

Unexpected Loss (99%)


Marginal Contributions:HotSpots Report

Risk Contribut ions: Obligor contributions to portfolio credit risk

By Industry

By Credit

State

By Industry

By Credit

State


26/31

26




Concentrations in Marine, RealEstate and Retail

Concentrations in B2and Ba3




Removing Caa ratedCounterparties


27/31

2


Marginal Portfolio Credit Risk

Risk contribution ~ Marginal risk (additional risk/unit) X size of exposure

CP 28

CP 11

CP 14

Outliers


Marginal Portfolio Credit RiskRisk contribution ~ Marginal risk (additional risk/unit) X size of exposure

Mar ginal Risk influencedby: Rating, Credit Driversand Risk Sensitivity

CP12 & CP16: Exposure

CP63 & CP62: Credit Driver

CP69 & CP72: Rating

CP55 & CP48: Duration


28/31

28


Marginal Portfolio Credit Risk

Taking into account Credit Derivatives

Before CD

CP10

After CD

CP10


Global Exposures & Limits Reports


29/31

29


Real Time Limits Reports


Market Risk Report


30/31

30


OpRisk Reports


OpRisk Reports


31/31


Summary todays talk

Business Requirement:Enterprise Risk Framework evolving BIS II requirements

Risk architecture - scalable

Measure and manage market, credit, operational risk (ALM, & collateral)

Enterprise coverage: consistent treatment of banking and trading books

Risk engine

Compute and reconcile economic and regulatory credit capital

Analyze risk along multiple dimensions: consolidate disparate exposure acrossmultiple business lines, portfolios and produc ts

Reporting infrastructure: comprehensive enterprise risk reporting

Flexible drill-down analysis and reporting in many dimensions

Accommodate users at every level of the firm, and multiple regulators

Integration of global market & credit risk information into business processes


www.algorithmics.com

See

Enterprise Credit Risk with

Mark-to-Future

Algo Research Quarterly

[email protected]

RLMdridEco-RegCapital

Documents