Counterparty Limit

Introduction to Counterparty Risk
Paola Mosconi
Banca IMI
Bocconi University, 18/04/2016
Paola Mosconi Lecture 8 1 / 84

Disclaimer
The opinion expressed here are solely those of the author and do not represent in
any way those of her employers

Outline
1 Introduction
Counterparty Risk
Exposures
2 CVA
General Framework
Unilateral CVA
3 CVA Calculation
General Framework
Wrong Way Risk
Monte Carlo Valuation
4 Case Studies
Case Study 1: Single Interest Rate Swap (IRS)
Case Study 2: Portfolio of IRS
5 Mitigating Counterparty Exposure
Netting
Collateral
6 Selected References
Introduction
Outline
1 Introduction
Counterparty Risk
Exposures
2 CVA
General Framework
Unilateral CVA
3 CVA Calculation
General Framework
Wrong Way Risk
4 Case Studies
Netting
Collateral
Introduction Counterparty Risk
Counterparty Risk: Definition
The counterparty credit risk is defined as the risk that the counterparty to a
transaction could default before the final settlement of the transaction’s cash
flows. An economic loss would occur if the transactions or portfolio of transactions
with the counterparty has a positive economic value at the time of default.
[Basel II, Annex 4, 2/A]
Counterparty risk is affected by:

the OTC contract’s underlying volatility
the correlation between the underlying and default of the counterparty
the counterparty credit spreads volatility

Counterparty Risk: Features
Unlike a firm’s exposure to credit risk through a loan, where the exposure to
credit risk is unilateral and only the lending bank faces the risk of loss, the
counterparty credit risk creates a bilateral risk of loss: the market value of the
transaction can be positive or negative to either counterparty to the transaction.
[Basel II, Annex 4, 2/A]
Loans: exposure at any future date is the outstanding balance, which is certain
(without considering prepayments). Credit risk is unilateral
Derivatives: exposure at any future date is determined by the market value
at that date and is uncertain. Counterparty risk can be:
unilateral: one party (the investor) is considered default-free and only the ex-
posure to the counterparty matters
bilateral: both parties are considered risky and face exposures depending on the
value of the positions they hold against each other

OTC Derivatives
OTC derivatives are efficient and effective tools to transfer financial risks
between market participants.
As a byproduct of such transfer:

they create credit risk between the counterparties
they increase the connectedness of the financial system
The 2008 financial crisis showed that counterparty-related losses (e.g. changes
in the credit spreads of the counterparties and changes in the market prices that
drive the underlying derivative exposures) have been much larger than default
losses.

OTC Market – BIS 2015

The volume of outstanding OTC derivatives has grown exponentially over the past 30
years. According to market surveys conducted by ISDA and BIS:
notional amounts of outstanding interest rate and currency swaps went from $866
billion in 1987 to $17.7 trillion in 1995, $99.8 trillion in 2002 and more than $500
trillion in 2015
the gross market value of outstanding derivatives contracts – i.e. the cost of replacing
all outstanding contracts at market prices prevailing on the reporting date – amounted
to $15.5 trillion at the end June 2015.
the interest rate segment accounts for the majority of OTC derivatives, with a
notional amount of outstanding contracts of $435 trillion (79% of the global OTC
markets)
FX derivatives make up the second largest segment of the global OTC derivatives
market with an OTC market share of 13%, amounting to $75 trillion in terms of
notional outstanding
central clearing has become increasingly important in interest rate derivatives mar-
kets and credit default swap markets

OTC Market – Global Chart

OTC Market – Interest Rate Derivatives Chart

Counterparty Risk: Risk Management vs Pricing
Two approaches to counterparty risk:
counterparty risk management:

for internal purposes and for regulatory capital requirements, following Basel
II
counterparty risk from a pricing point of view:

Credit Valuation Adjustment (CVA), when updating the price of instru-
ments to account for possible default of the counterparty
However, Basel III has made the distinction less clear-cut.

Counterparty Risk Management
Counterparty risk is the risk that one bank faces in order to be able to lend money
or invest towards a counterparty with relevant default risk.
The bank needs to measure that risk and cover for it by setting capital aside.
Credit VaR is calculated through the following steps:

1 the basic financial variables underlying the portfolio, including also defaults of
the counterparties, are simulated under the historical probability measure
P, up to the risk horizon
2 at the risk horizon, in every simulated scenario of the basic financial variables,
the portfolio is priced, eventually obtaining a number of scenarios for the
portfolio value at the risk horizon. “Priced”means that discounted future cash
flows of the portfolio after the risk horizon are averaged, conditional on each
scenario at the risk horizon but under the (pricing) risk neutral measure Q.

Counterparty Risk: Pricing
Pricing concerns updating the value of a specific instrument or portfolio, traded

with a counterparty, by adjusting the price in order to take into account the risk
of default of the counterparty.
The amount charged to the risky counterparty on the top of the default-free cost
of the contract is known as Credit Valuation Adjustment, or CVA.
Since it is a price, it is computed entirely under the (pricing) risk neutral measure
Q.
Under Basel II, the risk of counterparty default and credit migration risk were addressed
but mark-to-market losses due to credit valuation adjustments (CVA) were not. During
the financial crisis, however, roughly two-thirds of losses attributed to counterparty credit
risk were due to CVA losses and only about one-third were due to actual defaults.
[Basel III, Press Release June 2011]

Introduction Exposures
Exposures I
Counterparty exposure at any given future time is the larger between zero and
the market value of the portfolio of derivative positions with a counterparty that
would be lost if the counterparty were to default with zero recovery at that time.
Current exposure (CE) is the current value of the exposure to a counterparty.
Exposure at Default (EAD) is defined in terms of the exposure valued at the

(random future) default time of the counterparty.

Introduction Exposures
Exposures II
Potential future exposure (PFE) for a given date is the maximum of exposure at
that date with a high degree of statistical confidence. For example, the 95% PFE
is the level of potential exposure that is exceeded with only 5% probability. The
curve of PFE in time is the potential exposure profile, up to the final maturity of
the portfolio of trades with the counterparty.
The maximum potential future exposure (MPFE) represents the peak of PFE
over the life of the portfolio. PFE and MPFE are used to determine credit lines.
Expected exposure (EE) is the average exposure on a future date. The curve of
EE in time, as the future date varies, provides the expected exposure profile.
Expected positive exposure (EPE) is the average EE in time up to a given future

date.

CVA
Outline
1 Introduction
Counterparty Risk
Exposures
2 CVA
General Framework
Unilateral CVA
3 CVA Calculation
General Framework
Wrong Way Risk
4 Case Studies
Netting
Collateral
CVA General Framework
Credit Value Adjustment: Introduction
Credit Value Adjustment (CVA) tries to measure the expected loss due
to missing the remaining payments.
CVA has become an integral part of IAS 39 accounting rules and Basel III
regulatory requirements.
CVA is defined as:

1 the difference between the risk-free value and the risky value of one or
more trades or, alternatively,
2 the expected loss arising from a future counterparty default.

CVA Definition 1
The Net Present Value of a derivative at time t is given by:
V (t) = Et [Π(t, T )]
where Π(t, T ) represents (the sum of) all discounted cash flows between times T
and t. In the presence of counterparty risk, the sum of all discounted payoff terms
between t and T is denoted by ΠD (t, T ).
CVA is defined, according to Canabarro and Duffie (2004), as the difference

between the risk-free value and the risky value:
CVA := Et [Π(t, T )] − Et [ΠD (t, T )] (1)
unilateral (asymmetric) CVA, if only the default of the counterparty is

considered
bilateral (symmetric CVA), if also the default of the investor is taken into
account.
CVA Definition 1: Unilateral Case
We will show that, starting from definition (1) and accounting for all the cash-flows,
unilateral CVA is given by:
h i
CVA = LGD E 1{τ ≤T } D(0, τ ) (V(τ ))+ (2)
where:
τ is the default event, as defined in the Bilateral ISDA Master
Agreement
V(τ ) is the (uncertain) close-out amount
LGD is the expected loss given default, expressed as a percentage of the
nominal close-out amount.

CVA Definition 1: Bilateral Case
Analogously, In the bilateral case, the risky value of the derivative takes into account
both the default of the counterparty C and that of the investor I:
Et [ΠD (t, T )] = V (t)

| {z }
risk-free
h i
− LGDC Et 1{τC ≤T ,τC ≤τI } D(t, τC ) (V (τC ))+
| {z } (3)
CVA
h i
+ LGDI Et 1{τI ≤T ,τI ≤τC } D(t, τI ) (−V (τI ))+
| {z }
DVA
The formula is symmetric: the investor’s DVA is equal to the counterparty’s CVA,
but an investor cannot hedge its DVA spread risk by selling CDS protection on
itself.

CVA Definition 2: Expected Loss
We consider here the unilateral case1 .

CVA measures the risk of incurring losses on a portfolio of deals, upon default of
the counterparty. The loss is material when the value of the portfolio at default,
V (τ ), is positive and default occurs before the maturity of the portfolio, i.e.:
Loss(τ ) = 1{τ ≤T } LGD (V (τ ))+
Unilateral CVA is defined as the expected value of this loss, discounted till
evaluation time:
CVA = E[ D(0, τ ) Loss(τ )]

h
+
i (4)
= LGD E 1{τ ≤T } D(0, τ ) (V (τ ))
The resulting expression is in agreement with eq. (2).
1 The extension to the bilateral case is straightforward.

2002 ISDA Master Agreement: Events of Default
Events of Default
Failure to pay or deliver
Breach of agreement; repudiation of agreement
Default under specified transaction
Bankruptcy ...
Termination Events
Illegality
Force majeure event
Deferral of payments and deliveries during waiting period
Tax events ...
Events which cause Early Termination of the Master Agreement are always
bilateral.

Close-out Amount and Netting Set
Close-out Amount
When one of the two counterparties incurs in one of the events that causes
the Early Termination of the contract, the non-defaulting party determines
the amount of the losses or the costs of closing the position and replacing it
with a new one with another counterparty (substitution cost).
Netting Set
The ISDA Master Agreement determines the possibility to net out all the
positions with the defaulted counterparty.
The netting set P is formed by p contracts pi , whose individual value is Vi (t).
Each deal has weight wi = ±1 if the investor is respectively receiver/payer:
p
X
V (t) = wi Vi (t)
i =1

Risk Free vs Substitution Close-out
Unilateral case:
The close-out amount is the net present value of the residual deal, calculated as
a risk-free quantity, since the surviving counterparty is assumed to be default-
free.
Bilateral case:
The close-out amount is the net present value of the residual deal, calculated
by taking into account the risk of default of the survived party. It is called
substitution close-out and can give rise to contagion effects.
See Brigo and Morini (2010).

Positive Part of Close-out
The CVA formulae (2) and (3) depend on the positive part of the close-out
amount: (V(τ ))+ . Indeed, only the positive part of the close-out contributes to
counterparty risk.
Consider the point of view of the non-defaulting party. If:

V(τ ) < 0
the close-out amount is a liability and the non-defaulting party is due to pay
it fully to the defaulted party
V(τ ) ≥ 0
the non-defaulting party is exposed to the risk that the defaulted party does
not pay the close-out amount.

CVA Features
CVA is a credit hybrid.
The positive part of the close-out amount introduces an element of optionality

in the payoff: i.e. a call option with zero strike on V (τ ).
Optionality renders the payoff under counterparty risk model dependent,

even when the original payoff is model independent.
Example: Interest Rate Swap (IRS)
Without counterparty risk, the payoff is linear and model independent, requiring no
dynamical model for the term structure (no volatility and correlation). In the presence
of counterparty risk, the payoff transforms into a stream of swaptions, whose valuation
requires an interest rate model.
Optionality applies to the whole netting set with a given counterparty, making
CVA valuation computationally intensive.

CVA Unilateral CVA
Unilateral CVA – Assumptions
We assume that:
transactions are seen from the point of view of the safe investor, namely the
company facing counterparty risk
such investor is default-free
We denote by ΠD (t, T ) the sum of all discounted payoff terms between t and T ,
subject to counterparty default risk and by Π(t, T ) the analogous quantity when
counterparty risk is not considered.

CVA Unilateral CVA
Unilateral CVA – Cash flows
Cash flows are given by:

1 if default comes after final maturity τ > T , the original payoff:
+ 1{τ >T } Π(t, T )

2 if default occurs before maturity τ < T :
1 the payments due before default:
+ 1{τ ≤T } Π(t, τ )
2 the recovery of the residual net present value at default, if positive:
+ 1{τ ≤T } Rec D(t, τ ) (Eτ [Π(τ, T )])+
3 minus the total residual net present value at default, if negative:
− 1{τ ≤T } D(t, τ ) (−Eτ [Π(τ, T )])+

CVA Unilateral CVA
Unilateral CVA – Formula
By summing up all terms, the total payoff subject to counterparty default risk
becomes:
ΠD (t, T ) = 1{τ >T } Π(t, T ) + 1{τ ≤T } Π(t, τ )
n
+ −
o (5)
+ 1{τ ≤T } D(t, τ ) Rec (Eτ [Π(τ, T )]) + (Eτ [Π(τ, T )])
Recalling the definition on slide 18

CVAt := Et Π(t, T ) − ΠD (t, T )
the CVA at time t turns out to be:

h i
CVAt = LGD Et 1{τ ≤T } D(t, τ ) (Eτ [Π(τ, T )])+
h i (6)
+
= LGD Et 1{τ ≤T } D(t, τ ) (V (τ ))

CVA Unilateral CVA
Unilateral CVA – Proof I
Starting from the definition of CVA and expression (5) of the risky payoff, we derive formula
(6), through the following steps:
1 we express the sum of all discounted payoff terms between t and τ , i.e. Π(t, τ ), as a
function of Π(t, T ) as follows:
Π(t, τ ) = Π(t, T ) − D(t, τ )Π(τ, T )
2 Plugging this result into eq. (5) we get:
ΠD (t, T ) = 1{τ >T } Π(t, T ) + 1{τ ≤T } Π(t, T )

+ 1{τ ≤T } D(t, τ ) −Π(τ, T ) + Rec (Eτ [Π(τ, T )])+ + (Eτ [Π(τ, T )])−

and, using the fact that 1{τ >T } Π(t, T ) + 1{τ ≤T } Π(t, T ) = Π(t, T ),
ΠD (t, T ) = Π(t, T )
+ 1{τ ≤T } D(t, τ ) −Π(τ, T ) + Rec (Eτ [Π(τ, T )])+ + (Eτ [Π(τ, T )])−


CVA Unilateral CVA
Unilateral CVA – Proof II
Recalling that CVAt := Et Π(t, T ) − ΠD (t, T ) , we get:

3
CVAt = −Et 1{τ ≤T } D(t, τ ) −Π(τ, T ) + Rec (Eτ [Π(τ, T )])+ + (Eτ [Π(τ, T )])−

4 We consider the first term in the sum Et −1{τ ≤T } D(t, τ ) Π(τ, T ) .
Using the tower rule of expectations, i.e. Et [·] = Et [Eτ [·]], with τ ≥ t we have:

Et −1{τ ≤T } D(t, τ ) Π(τ, T ) = Et −Eτ 1{τ ≤T } D(t, τ ) Π(τ, T )

= Et −1{τ ≤T } D(t, τ ) Eτ [Π(τ, T )]
5 Plugging this result in CVAt , recalling the definition of close-out amount at default
time, i.e. V (τ ) := Eτ [Π(τ, T )], and using the fact that X = X + + X − , we obtain
the final result, eq. (6):
CVAt = −Et 1{τ ≤T } D(t, τ ) − (V (τ ))+ − (V (τ ))− + Rec (V (τ ))+ + (V (τ ))−

= −Et 1{τ ≤T } D(t, τ ) − (V (τ ))+ + Rec (V (τ ))+

= LGD Et 1{τ ≤T } D(t, τ ) (V (τ ))+


CVA Calculation
Outline
1 Introduction
Counterparty Risk
Exposures
2 CVA
General Framework
Unilateral CVA
3 CVA Calculation
General Framework
Wrong Way Risk
4 Case Studies
Netting
Collateral
CVA Calculation General Framework
General Framework
We consider for simplicity the case of unilateral CVA.

The goal is to calculate CVA expressed as in formula (4), i.e.
h i
CVA = E[ D(0, τ ) Loss(τ )] = LGD E 1{τ ≤T } D(0, τ ) (V (τ ))+
Closed-form formulae are only available:

for certain single deals, i.e. in the absence of a netting set
under the assumption of independence between counterparty risk,
embedded in 1{τ ≤T } and the risks associated to the underlying exposure,
embedded in V (τ )

CVA Calculation Wrong Way Risk
Wrong Way Risk (WWR)
In some cases, the assumption of independence leads to underestimate a significant

source of potential loss. This is due to Wrong Way Risk or Right Way Risk.
ISDA definition of WWR
WWR is defined as the risk that occurs when “exposure to a counterparty is

adversely correlated with the credit quality of that counterparty”.
It arises when default risk and credit exposure increase together.
Specific WWR arises due to counterparty specific factors: a rating downgrade,

poor earnings or litigation.
General WWR occurs when the trade position is affected by macroeconomic
factors: interest rates, inflation, political tension in a particular region, etc...

WWR Examples
1 Monoline insurers (e.g. Ambac and MBIA)
During the sub-prime crisis, the monolines specialized in guaranteeing mortgage-
backed securities, when the mortgage market collapsed, saw their creditworthi-
ness deteriorate and found themselves unable to pay all of the insurance claims.
Almost all exposure mitigation from monoline insurance fell short due to the
guarantors’ increased probability of default under exactly the same conditions
when insurance was most needed.
2 Collateralized loan
Bank A enters into a collateralized loan with Bank B (the counterparty).
The collateral that Bank B provides to A can be of different nature:
bonds issued by Bank B (specific WWR)
bonds issued by a different issuer belonging to a similar industry, or the
same country or geographical region (general WWR). This kind of risk
is both difficult to detect in the trading book, hard to measure and
complex to resolve.

Right Way Risk (RWR)
Right way risk is the opposite of wrong way risk. It is the effect observed when
the exposure decreases as the default probability increases, i.e. when there is
a negative dependency between the two. The size of credit risk decreases as the
counterparty approaches a potential default. RWR occurs when a company enters
into transactions to partially hedge an existing exposure.
Examples:
An airline usually protects itself against a rise in fuel prices by entering into
long oil derivative contracts.
A company would normally issue calls and not puts on its stock.
WWR and RWR are together referred to as DWR (directional way risk).

Overview of Modeling Approaches
Goal: to correlate counterparty exposure with counterparty creditworthiness
1 Intensity (reduced form) models

2 Structural models
3 Cespedes et al (2010)
4 Hull and White (2012)
5 Basel II

Modeling Approaches
1 Intensity (reduced form) models

Default is described in terms of the default time, i.e. the first jump time
of a Poisson/Cox process with (default) intensity λ(t).
The stochastic process for λ(t) is correlated to the stochastic process for
the exposure.
Intensity models work well with exposures in the asset class of interest
rates, FX, credit, commodity.
In the case of equity, not enough correlation. Structural models are
more appropriate.
2 Structural models
The stochastic process affecting the dynamics of the counterparty firm value
is directly correlated to the stochastic process ruling the underlying of the
contract. See Lecture 5 for a list o structural models and their calibration to
the market.

Other Modeling Approaches
3 Cespedes et al (2010)
Cespedes et al (2010) propose an ordered scenario copula model. Default
events and exposures are driven by factor models, while a Gaussian copula is
used to correlate exposure and credit events. The approach builds on existing
exposure scenarios by a non-parametric sampling of exposure via the factor
model.
4 Hull and White (2012)
Hull and White (2012) model the hazard rate as a deterministic monotonic
function of the value of the contract. Wrong-way (right-way) risk is obtained
by making the hazard rate to be an increasing (decreasing) function of the
contract’s value.

Basel II
Basel II deals with wrong-way risk using the so-called “alpha”multiplier to

increase the exposure, in the version of the model in which exposure and
counterparty creditworthiness are assumed to be independent.
The effect is to increase CVA by the alpha multiplier.
The Basel II rules set alpha equal to 1.4 or allows banks to use their own models,
with a floor for alpha of 1.2, i.e.
if a bank uses its own model, at minimum, the CVA has to be 20% higher than
that given by the model where default and exposure are independent
if a bank does not have its own model for wrong way risk it has to be 40%
higher
Estimates of alpha reported by banks range from 1.07 to 1.10.

CVA Calculation Monte Carlo Valuation
Monte Carlo Valuation I
In the presence of a netting set, CVA can be calculated only through multi-
asset Monte Carlo simulation, where financial instruments must be simulated
until maturity. Calibration of the framework and Monte Carlo simulation with
a sufficiently large number of scenarios is very time consuming.
Monte Carlo method allows to estimate the expected value of a variable

as the average of all its realizations across different simulated scenarios ωk .
In the case of unilateral CVA, starting from eq. (4), we get:
N MC
1 X
CVA ≈ D(0, τ ; ωk ) Loss(τ, X (τ, ωk )) (7)
NMC
k=1
where τ ≡ τ (ωk ) and we have made explicit the dependence of the loss vari-
able Loss on the risk drivers affecting the cash flows, denoted compactly by
X (τ, ωk ).

Monte Carlo Valuation II
In order to simplify the calculation of:
N MC
1 X
CVA ≈ D(0, τ ; ωk ) Loss(τ, X (τ, ωk ))
NMC
k=1
two approximations are commonly introduced:

Approximation 1: Default bucketing
Approximation 2: Default bucketing + independence
This is equivalent to ignoring wrong/right way risk

Approximation 1: Default Bucketing
Assuming that default can be observed at discrete times T1 , T2 , . . . , Tb , formula

(7) can be simplified as:
N b
MC X
1 X
CVA ≈ D(0, Tj ; ωk ) Loss(Tj , X (Tj , ωk )) (8)
NMC
k=1 j=1
where defaults have been bucketed but, a joint model for:

the default of the counterparty 1{τ (ωk )∈(Tj −1 ,Tj ]} and
the value of counterparty exposure V (Tj ) = ETj [Π(Tj , T )]
is still needed.

Approximation 1: Models I
In order to simulate all the variables that affect the calculation of CVA, we
need to define their dynamics. We follow two criteria:
simplicity of the models
straightforward calibration of the models to liquid market instruments
Let Z(t) = {X(t), λC , λI } be the set of all processes underlying the calculation
of CVA, where X denotes the risk drivers, λC and λI respectively the default
of the counterparty and the default of the investor2 .
Each process, under the risk neutral measure Q, follows a dynamics
dZi (t) = (. . .)dt + (. . .)dWi
and is correlated to the others through dWi (t) dWj (t) = ρij dt.
2 In the following, we will consider the investor default free λ = 0, such that only the default
I
of the counterparty matters λC ≡ λ.
Approximation 1: Models II
In the following, we will show as an illustrative example of underlying models only

the case of an exposure to interest rate risk.
Analogously, models for other asset classes can be introduced and correlated to the
model for the creditworthiness of the counterparty.

Approximation 1: Credit Model
We consider an intensity model3 for the credit spread

If default of the counterparty were independent of the others sources of risk X , a
deterministic model would be enough to bootstrap the probabilities of default of
the counterparty
When λ and X are dependent, a common choice is the CIR model with jumps:
λ(t)dWλ (t) + dJ β,F (t)

p
dλ(t) = κ(µ − λ(t))dt + ν (9)
β,F
where dJ (t) represents the jump component and the model:
N(t)
X
J β,F (t) = Yi N(t) ∼ Poisson(βt) Yi i.i.d. ∼ F
i =1
F (x) = 1 − e −γx x≥0

The model is exactly calibrated to CDS quotes, assuming deterministic interest rates
3 For exposures to equity products, structural models are more appropriate

Approximation 1: Interest Rates Model
A common choice is the one-factor Hull and White model, which can be equiva-
lently4 described by a shifted short rate model
r φ (t) = r (t) + φ(t) (10)

where φ(t) is a deterministic shift extension and the factor r (t) evolves as a Vasicek:
dr (t) = k(θ − r (t))dt + σdWr (t)
The value of the zero coupon bond is given by:

h RT i RT
P(t, T ) = Φ(t, T ) E e − t r(s)ds Ft where Φ(t, T ) = e − t φ(s)ds

The model is exactly calibrated to swaption prices and the discount curve.
In the multi-curve setting, the deterministic shift captures the differences between
the curves of discounting (OIS) and forwarding (underlying instrument tenor)
4 See Brigo and Mercurio, Chapter 3.3 and 3.8

Approximation 1: Correlation Default/Exposure
Finally, we introduce correlation between credit (9) and interest rates (10), as fol-
lows:
corr(dλ(t), dr φ (t)) = corr(dλ(t), dr (t)) = ρλr dt

Approximation 2: Default Bucketing and Independence
Assuming that default of the counterparty is independent of its exposure (absence

of wrong-way risk) eq. (8) can be further simplified according to:
b
X
Q(τ ∈ (Tj−1 , Tj ]) E D(0, Tj ) (V (Tj ))+

CVA = LGD
j=1
b
X
{Q(τ > (Tj−1 )) − Q(τ > Tj )} E D(0, Tj ) (V (Tj ))+

= LGD
j=1
b NMC
LGD X X
≈ {Q(τ > (Tj−1 )) − Q(τ > Tj )} D(0, Tj ; ωk ) [V (Tj , X (Tj , ωk ))]+
NMC j=1
k=1
(11)
Defaults are bucketed and only survival probabilities are needed (no default model).
An option model for counterparty exposure V (t) is still needed.

Case Studies
Outline
1 Introduction
Counterparty Risk
Exposures
2 CVA
General Framework
Unilateral CVA
3 CVA Calculation
General Framework
Wrong Way Risk
4 Case Studies
Netting
Collateral
Case Studies Case Study 1: Single Interest Rate Swap (IRS)
A (Payer) Interest Rate Swap is a contract that exchanges payments between two legs,
starting from a future time instant.
The fixed leg pays out the amount N αFi K , with a fixed interest rate K , a nominal value N
and a year fraction αFi between TiF−1 and TiF , while the floating leg pays out the amount
N αfi L(Ti −1 , Ti ), where

1 1
L(Ti −1 , Ti ) = f − 1
αi P fwd (Ti −1 , Ti )
is the forward rate resetting at Tif−1 and paying at Tif−1 .
Schematically:
Fixed → fixed rate N αFi K at Ta+1

F
, . . . , TbF → Floating
f f
Leg ← floating rate N αi L(Ti −1 , Ti ) at Ta+1 , . . . , Tbf ← Leg

Case Study 1: Risk Drivers
We consider a EUR denominated IRS, where the fixed leg pays annually and the
floating leg semi-annually.
Risk drivers for a EUR denominated IRS
1 EUR discounting curve (OIS – Overnight Indexed Swap)

2 EUR forwarding curve (6 months Euribor)
3 EUR interest rate volatility
4 CDS spread of counterparty C

Case Study 1: IRS under Independece
IRSD (t, K ) = IRS(t, K )

b−1
X (12)
− LGD Q(τ ∈ (Tj−1 , Tj ]) SWAPTIONj,b (t; K , Sj,b (t), σj,b )
j=a+1
Counterparty model for credit spread Counterparty exposure model

Survival probabilities are bootstrapped One-factor Hull and White (short rate)
from CDS quotes, under the assumption model for r (t), calibrated to swaption
of deterministic interest rates. quotes and zero curve data.

Case Study 1: Exposures Results
10 year payer swap

Notional N = 10 Million
From left to right:
ATM: K = 0.6%
ITM: K = 0.1%
OTM: K = 3.5%

Case Studies Case Study 2: Portfolio of IRS

5 counterparties
30 IRS (payer, receiver), not all deals belonging to given netting sets
Deal ID Cpty ID Netting ID Principal

1 5 5 813450
2 5 NaN 441321
3 1 NaN 629468
... ... ... ...
9 5 5 918177
... ... ... ...
IRS maturities: from 1 to 7 years

CDS market quotes in bps for each counterparty, for different maturities
Maturity Cpty 1 Cpty 2 Cpty 3 Cpty 4 Cpty 5

1y 140 85 115 170 140
2y 185 120 150 205 175
3y 215 170 195 245 210
4y 275 215 240 285 265
5y 340 255 290 320 310

Case Study 2: Mark to Market Swap Prices

MtM swap prices are computed at each future simulation date and for each scenario
×10 4 Swap prices along scenario 32

8
4
Mark-To-Market Price
-2
-4
-6
-8
-10
2007 2010 2012 2015
Figure: MtM of all IRS in the portfolio, for scenario 32.

Case Study 2: Simulated Portfolio Values

The total portfolio value is computed at each simulation date, for each scenario. As the
swaps get closer to maturity, their values begin to approach zero since the aggregate value
of all remaining cash flows decreases after each cash flow date.
×10 5 Total MTM Portfolio Value for All Scenarios

10
4
Portfolio Value ($)
-2
-4
-6
-8
Jul07 Jan10 Jul12 Jan15
Simulation Dates

Case Study 2: Counterparty Exposures I
The exposure of a particular contract i at time t is given by:
Ei (t) = max{Vi (t), 0} = (Vi (t))+
The exposure to a counterparty is the sum of the individual contract exposures:

X X X
Ecpty (t) = Ei (t) = max{Vi (t), 0} = (Vi (t))+
In the presence of netting agreements, contracts are aggregated together and can
offset each other. The total exposure of all contracts in a netting agreement is:
nX o X +
Enetting (t) = max Vi (t), 0 = Vi (t)

Case Study 2: Counterparty Exposures II
Exposure of the entire portfolio, at each simulation date and for each scenario.
×10 5 Portfolio Exposure for All Scenarios

12
10
8
Exposure ($)
0
Simulation Dates

Case Study 2: Exposure Profiles I

(Non-discounted) exposure profiles:
PFE – Potential Future Exposure
A high percentile (95%) of the distribution of exposures at any given future date
MPFE – Maximum Potential Future Exposure
The maximum PFE across all dates
EE – Expected Exposure
The mean (average) of the distribution of exposures at each date
EPE – Expected Positive Exposure
Weighted average over time of the expected exposure
EffEE – Effective Expected Exposure
The maximum expected exposure up to time t
EffEPE – Effective Expected Positive Exposure
Weighted average over time of the effective expected exposure

Case Study 2: Exposure Profiles II
×10 5 Portfolio Exposure Profiles ×10 4 Counterparty 5 Exposure Profiles

6 15
PFE (95%)
PFE (95%)
Max PFE
Max PFE
Exp Exposure (EE)
5 Exp Exposure (EE) Time-Avg EE (EPE)
Time-Avg EE (EPE) Max past EE (EffEE)
Max past EE (EffEE) Time-Avg EffEE (EffEPE)
Time-Avg EffEE (EffEPE) 10
4
Exposure ($)
Exposure ($)
3
2 5
0 0
Jul07 Jan10 Jul12 Jan15 Jan08 Jan10 Jan12 Jan14
Simulation Dates Simulation Dates

Case Study 2: Discounted Exposures
Discounted expected exposures are computed by using the discount factors obtained
from a Hull and White simulation.
×10 4 Discounted Expected Exposure for Each Counterparty

×10 4 Discounted Expected Exposure for Portfolio 6
18
16
5
14
Discounted Exposure ($)
Discounted Exposure ($)

4
12
10
3
6 2
4
1
2
0 0
Jan08 Jan10 Jan12 Jan14 Jan08 Jan10 Jan12 Jan14
Simulation Dates Simulation Dates

Case Study 2: Probabilities of Default
The default probability of a given counterparty is implied from the current market spreads
of the counterparty’s CDS (see Table in slide 55) at each simulation date, through a
bootstrap procedure.
Default Probability Curve for Each Counterparty

0.45
0.4
0.35
Probability of Default
0.3
0.25
0.2
0.15
0.1
0.05
0
Simulation Dates

Case Study 2: CVA Computation
CVA is calculated through formula (11), which we write here in a simplified form:
b
X
CVA ≈ (1 − Rec) [PD(tj ) − PD(tj−1 )] discEE(tj )
j=1
where:
exposures are assumed to be independent of default (no wrong-way risk) and have
been obtained using risk-neutral probabilities
discEE(t) is the discounted expected exposure at time t
PD(t) = 1 − Q(τ > t) is the default probability
Rec = 1 − LGD is the recovery, and for this example it has been assumed equal to
40%

Case Study 2: CVA Results
CVA for each counterparty

6000
5000
Counterparty CVA ($)
1 2228.36 4000
CVA $
2 2487.60 3000
3 920.39 2000
4 5478.50 1000
5 5859.30 0
1 2 3 4 5
Counterparty

Mitigating Counterparty Exposure
Outline
1 Introduction
Counterparty Risk
Exposures
2 CVA
General Framework
Unilateral CVA
3 CVA Calculation
General Framework
Wrong Way Risk
4 Case Studies
Netting
Collateral
Mitigation of counterparty exposure can be achieved through5 :

netting agreements
collateralization
5 See for example Ballotta et al.

Mitigating Counterparty Exposure Netting
Netting
In presence of multiple trades with a counterparty, netting agreements allow,

in the event of default of one of the counterparties, to aggregate the transactions
before settling claims.
In the absence of netting, the exposure is:

n
X n
X
E (τ ) = wi Ei (τ ) = wi (Vi (τ ))+
i =1 i =1
where n is the number of contracts, wi are the asset quantities, and Ei the corre-
sponding exposures.
A netting agreement is a legally binding contract between two counterparties based
on which, in the event of default, the exposure results in:
n
!+
X
Enetting (τ ) = wi Vi (τ )
i =1

Example 1
Assumptions
Two counterparties, a bank B and a counterparty C, such that:
C holds a currency option written by B with a market value of 50
B has an IRS with C, having a marked to market value in favor of B of 80
Exposures
The exposure of the bank B to the counterparty C is 80
The exposure of the counterparty C to the bank B is 50
The exposure of the bank B to the counterparty C, with netting, is 30

Example 2
The following table shows, at different times, the values of five trades as well as
the future exposures to the counterparty, with and without netting.
Time (Months)
Trade ID 1 2 3 4 5
1 10 -7 8 -6 -2
2 9 0 4 -2 2
3 7 7 5 10 -8
4 -7 -6 3 -6 -6
5 -5 -5 3 6 -6
Exposures ($)
No Netting 26 7 23 16 2
Netting 14 0 23 2 0
Table: Source Ballotta et al (forthcoming).

Example 3
Consider a portfolio of 10 homogeneous assets (long forward contracts) with IVol = 38%,
marginal default probability 0.025%, risk free rate r = 3%, dividend yield q = 1% and the
same cross-correlation levels.
The simulated CVA of such portfolio shows that the larger the cross correlation among
assets, the smaller the benefit of the netting clause.
CVA CVA Reduction

No Netting 0.01518
ρ = 0.9 0.01449 4.59%
ρ = 0.5 0.01055 30.48%
ρ = 0.2 0.00649 57.23%
ρ=0 0.00304 80.00%
Figure and Table: Source Ballotta et al (forthcoming).

Mitigating Counterparty Exposure Collateral
Collateral Definition
Collateralization is one of the most important techniques of mitigation of counter-

party risk.
A collateral account is a contractual clause aimed at reducing potential losses

incurred by investors in case of the default of the counterparty, while the
contract is still alive.

Collateral ... in Theory

Consider the bank/investor B and the counterparty C.
Let C (t) be the (cash) collateral amount posted by C to B, at time t.
B has no exposure to the contract up to the collateral amount, while its losses are re-
duced by the collateral amount whenever the exposure exceeds it. The collateralized
exposure EC (t) is defined as:
EC (t) = (E (t) − C (t))+
Equivalently:
EC (t) = E (t) − C (t) − (C (t) − E (t))+

The posting of collateral allows a mitigation of the exposure in favor of the part
receiving it. This mitigation is positive and equal to the amount:
C (t) − (C (t) − E (t))+

Collateral ... in Practice I

The actual amount of the collateral available at time t depends on the contractual
agreement between the parties, specified in terms of:
1 Posting threshold H > 0

i.e. the threshold which triggers the posting of collateral: below the threshold no
collateral is posted.
The underlying commercial reason for a threshold is that often parties are willing
to take a certain amount of credit risk (equal to the threshold) before requiring
collateral to cover any additional risk
2 Margin period δ
i.e. the interval at which margin is monitored and called for:
C (t) = [E (t − δ) − H]+
In case of default at time τ , the last call occurs at (τ − δ).

Most collateral agreements require daily calculations; however, in order to reduce
operational requirements, weekly or monthly calculations can be agreed on, which
result in increased credit risk

Collateral ... in Practice II
3 Minimum transfer amount MTA

i.e. the amount below which no margin transfer is made. The collateral is set to zero
if less than MTA:
C (t) = [E (t − δ) − H]+ 1E (t−δ)−H>MTA
The presence of the MTA avoids the operational costs of small transactions and
contributes to reduce the frequency of collateral exchanges
4 Downgrade triggers
Sometimes, the threshold and the MTA vary during the lifetime of the contract if
the parties agree on the inclusion of downgrade triggers, also known as rating-based
collateral calls. These clauses force a firm to post more collateral to its counterparty,
if it is downgraded below a certain level

Example 1: Downgrade Triggers
AIG (2008)
Soon after the collapse of Lehman Brothers, on September 16, 2008 AIG’s credit
rating was downgraded and it was required to post 15$ billion in collateral with its
trading counterparties, leading to a liquidity crisis that essentially bankrupted all of
AIG. AIG could not collect the required funds on such a short notice.
Citigroup, MS and RBS (2012)

Similarly, in June 2012 Moody’s downgraded three major derivatives dealers (Cit-
igroup, Morgan Stanley and Royal Bank of Scotland) below the crucial single A
threshold, which has led to collateral calls from counterparties.

Example 2
The following table provides the exposure of a bilateral contract at different dates and
under different assumptions on the threshold H and the MTA:
Time (Months) 0 2 4 6 8 10 12
E (t) 0 3 12 19 25 26 0
C (t) (H = 0, MTA = 0) 0 0 3 12 19 25 26
EC (t) 0 3 9 7 6 1 0
C (t) (H = 1, MTA = 0) 0 0 2 11 18 24 25
EC (t) 0 3 10 8 7 2 0
C (t) (H = 1, MTA = 2) 0 0 0 11 18 24 25
EC (t) 0 3 12 8 7 2 0
Table: Source Ballotta et al (forthcoming).
In general, the larger the threshold, the less effective the collateral protection and the
longer the margining period, the higher the risk of upward movements in the value of the
contract, and ultimately in the CVA.

Gap Risk I
Collateralization is not able to fully eliminate counterparty risk.

Sudden movements in the market between two margining dates can increase both
the exposure and the probability of the relevant default event.
This originates gap risk.
Gap Risk is the residual (counterparty) risk which remains because:

the threshold is non-zero
the margin period of risk is the finite time needed to:
1 initiate margin call
2 allow the counterparty to post additional collateral
3 liquidate the position and rehedge
the market can jump (crash) within this time

Gap Risk II
In practice, gap risk is the risk that the corporate defaults, the bank survives and
the contract moves in the money, given that, at the last margining date, the coun-
terparty was solvent and the exposure out-of-the money.
In the presence of downgrade triggers, the counterparty may have downgrade trig-
gers with many other banks thus its downgrade may lead to simultaneous sell-offs
(“crowded market”) leading to falling market (gap risk).

“Unravelling Gap Risk at Deutsche Bank”, FT 26/05/2015

In 2005 Deutsche bought $100bn of insurance from Canadian pension funds against
the possibility of default by some of the safest companies. Since both sides estimated
extremely unlikely the simultaneous default of all companies, they agreed that the
pension funds put up a small amount of collateral, initially 9% of the $100bn.
During the 2008 crisis, the increased risk of the companies going bankrupt made
the trades become more valuable to Deutsche, increasing their value from $2.63bn
to $10.65bn.
However, the crisis did also increase the chance that pension funds would not be able
to live up to their end of the bargain. With such a small percentage of collateral, the
pension funds could ultimately decide it was better to walk away from the trades.
The bank was supposed to account for this risk, known as the gap risk. The bank
used five different methods to calculate gap risk, but instead of increasing the risk,
each of the methods reduced it. In the end, the bank reduced its gap risk from $200m
to zero, though potentially, according to independent estimates by the SEC, Goldman
Sachs and some ex-employees, it would have had to be around to $12bn.
Deutsche was taking the upside of the trades, but not the downside.
Re-hypothecation
Banks can use collateral not only as a way of reducing credit risk, but also as a way
of funding, through re-hypothecation, i.e. the practice of reusing, selling or lending
assets which have been received as collateral.
According to a survey on margin published by ISDA in April 2010, 82% of large deal-
ers reported re-hypothecating collateral received in connection with OTC derivatives
transactions (Risk Magazine, October 2010).
If collateral is segregated and not available for re-hypothecation, banks have to

assume that they need to raise funding to meet the cashflows over the life of the
trade using their own internal funding curves (Risk Magazine, September 2010).

Collateral and Corporates
Very few corporates post collateral because they do not have enough liquid assets
for the purpose. In addition, for a corporate the operational complexity associated
with collateralization (negotiating a legal document, monitoring exposures, making
cash transfers, etc.) may significantly increase the cost and resource requirements.
As a result, hedging with derivative can become so expensive that corporates will
choose to accept higher levels of exposure instead (Risk Magazine, October 2011).
“The airline’s Cologne-based [Lufthansa] head of finance, Roland Kern, expects its
earnings to become more volatile - not because of unpredictable passenger num-
bers, interest rates or jet fuel prices, but because it does not post collateral in its
derivatives transactions.”

Selected References
Outline
1 Introduction
Counterparty Risk
Exposures
2 CVA
General Framework
Unilateral CVA
3 CVA Calculation
General Framework
Wrong Way Risk
4 Case Studies
Netting
Collateral
Selected References
Selected References
Ballotta, L., Fusai, G. and Marena, M. (Forthcoming), Introduction to Default Risk
and Counterparty Credit Modelling
Basel II, Annex 4 (2006). https://www.bis.org/publ/bcbs128d.pdf
Basel III (2010), http://www.bis.org/bcbs/basel3.htm
BIS Semiannual Survey of Over-The-Counter (OTC) Derivatives, November 2015
Brigo, D. and Morini, M. (2010), Dangers of Bilateral Counterparty Risk: the
Fundamental Impact of Close-out Conventions.
Canabarro, E. and Duffie, D. (2004), Measuring and Marking Counterparty Risk,
Asset/Liability Management for Financial Institutions, edited by Tilman
Institutional Investor Books.
http://www.darrellduffie.com/uploads/surveys/duffiecanabarro2004.pdf
Cespedes, J., de Juan Herrero, J., Rosen, D., and Saunders, D. (2010), Effective
modeling of wrong way risk, counterparty credit risk capital, and alpha in Basel II,
The Journal of Risk Model Validation, 4(1):71-98
Hull, J. and A. White,(2012), CVA and wrong way risk, Financial Analysis Journal
68(5):58-69

Counterparty
Risk, CVA, and
Basel III
Harvey Stein
Markets and
risks Counterparty Risk, CVA, and Basel III
Counterparty
risk
Risk
Modifications
Counterparty
Harvey Stein
Valuation
Adjustments
hjstein@bloomberg.net
CVA and CCDS
Hedging CVA
Head, Counterparty and Credit Risk
CDS Proxies Bloomberg LP
Portfolio
counterparty risk
CVA & Basel
CVA VaR
Accounting Columbia University

considerations
Financial Engineering Practitioners Seminar
CVA issues
March 2012
Summary
Appendix —
CDS Spreads
References Id: cva-body.latex.tex 30266 2012-03-05 15:25:40Z hjstein
1 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Outline

1 Markets and risks
Markets and
risks
2 Counterparty risk
Counterparty
risk
3 Risk Modifications
Risk 4 Counterparty Valuation Adjustments
Modifications
Counterparty 5 CVA and CCDS

Valuation
Adjustments 6 Hedging CVA
CVA and CCDS
Hedging CVA
7 CDS Proxies
CDS Proxies 8 Portfolio counterparty risk
Portfolio
counterparty risk 9 CVA & Basel
CVA & Basel 10 CVA VaR
CVA VaR
11 Accounting considerations
Accounting
considerations
12 CVA issues
CVA issues
13 Summary
Summary
Appendix — 14 Appendix — CDS Spreads

CDS Spreads
References 15 References
2 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Need for CVA

Markets and
risks
Counterparty
risk
Risk
Modifications
We live in an increasingly risky world.
Counterparty
Valuation
Adjustments
• Bank failures.
CVA and CCDS
Hedging CVA
• Global recession.
CDS Proxies • Libor rates far from the “risk free” rate.
Portfolio
counterparty risk
Pre-crisis, swaps often traded without counterparty risk being taken
CVA & Basel
into account. Today, the counterparty could be a low quality bank.
CVA VaR
Accounting
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
3 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Size

Markets and
risks
Despite the financial crisis, the OTC derivatives markets continue to
Counterparty be a big part of the market, and interest rate derivatives are the
risk
largest part of the OTC derivatives market.
Risk
Modifications
Counterparty
• OTC derivatives notional outstanding
Valuation
Adjustments
• $547 trillion in December 2008
CVA and CCDS
• 70% in interest rate derivatives
Hedging CVA
• $605 trillion in June 2009
CDS Proxies • OTC derivatives gross market value
Portfolio
counterparty risk
• June 2008 — $20 trillion
CVA & Basel
• December 2008 — $32 trillion — up 60%
CVA VaR • IRD gross market value
Accounting • June 2008 — $9 trillion
considerations
• December 2008 — $18 trillion — doubled in 6 months.
CVA issues
Summary
Appendix —
And despite continued market turmoil, the derivatives market has
CDS Spreads continued to grow.
References
4 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Notional outstanding

Markets and
risks
Counterparty OTC Derivatives Notional outstanding (in Billions)

risk 800000
Risk
Modifications 700000
Counterparty
Valuation
600000
Adjustments
CVA and CCDS

500000
Hedging CVA
CDS Proxies 400000
Portfolio
counterparty risk 300000
CVA & Basel

200000
CVA VaR
Accounting
100000
considerations
CVA issues
0
01/98 01/00 01/02 01/04 01/06 01/08 01/10 01/12
Summary
Appendix —
CDS Spreads
References
5 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Demand

Markets and There is much demand for managing counterparty risk
risks
Counterparty
risk
• Accounting standards — FASB 157 (now “Topic 820, Section
Risk 10”) and IAS 39 — credit risk must be taken into account.
Modifications
• Regulators.
Counterparty
Valuation
Adjustments • Risk managers.
CVA and CCDS
Hedging CVA The IASB even issued a request for comment on counterparty risk
CDS Proxies calculation methodologies.
Portfolio
counterparty risk While the Dodd-Frank Act pushes OTC derivatives into clearing
CVA & Basel houses, thus mitigating counterparty risk, but:
CVA VaR
Accounting
• Risk doesn’t go away — clearing houses will be bearing and
considerations
collateralizing for counterparty risk.
CVA issues
• Corporations may still remain off of clearing houses.
Summary
Appendix — • Collateralization costs must be computed and are similar to

CDS Spreads
References
counterparty valuation adjustments.
6 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Counterparty risk

Markets and
risks
Counterparty risk — the exposure to loss due to a specific
Counterparty
risk counterparty failing to meet contractual obligations, i.e. defaulting.
Risk
Modifications
Often restricted to counterparties on OTC derivatives contracts, but
Counterparty doing so is myopic. If a counterparty defaults, all of their contracts
Valuation
Adjustments are affected:
CVA and CCDS
• OTC derivatives
Hedging CVA
CDS Proxies • bond issues

Portfolio
counterparty risk • stock issues
CVA & Basel • debts, loans, ...
CVA VaR
Accounting Compartmentalization of risks must be based on the type of risk, not

considerations
the type of security.
CVA issues
Summary So, it’s worthwhile to think of counterparty risk as the impact of

Appendix — default risk.
CDS Spreads
References
7 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Bond counterparty risk

Markets and
risks
Counterparty Example — Investor is long a bond.

risk
Risk • Counterparty is issuer.

Modifications
Counterparty • Issuer defaults — investor loses bond cash flows, but gets
Valuation
Adjustments recovery on the bond.
CVA and CCDS
• Recovery is a percentage of the principal of the bond.
Hedging CVA
CDS Proxies
• Default causes loss of interest payments, and early (but partial)
Portfolio return of principal.
counterparty risk
• Could be an improvement if bond is trading at a sufficient
CVA & Basel
CVA VaR
discount.
Accounting
considerations Example — Investor is short a bond.
CVA issues
Summary
• No counterparty risk.
Appendix —
CDS Spreads
References
8 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Swap counterparty risk

Markets and
risks
Counterparty
risk
Risk
Example — Investor enters into a swap.
Modifications
Counterparty • Counterparty is the entity with which the swap was transacted.
Valuation
Adjustments • Investor is simultaneously long one leg of the swap, and short the
CVA and CCDS
other leg.
Hedging CVA
• Counterparty defaults — investor loses swap cash flows, but gets
CDS Proxies
Portfolio
recovery on the swap.
counterparty risk
• Recovery is a percentage of the market value of the swap.
CVA & Basel
CVA VaR • If swap value is positive then there’s a loss.

Accounting
considerations
• If swap value is negative, then there is no loss.
CVA issues
Summary
Appendix —
CDS Spreads
References
9 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Risk modifications — Netting

Markets and
risks Netting agreements:
Counterparty
risk
• Optional part of the ISDA master agreement.
Risk
Modifications • In the event of default, recovery is on the net market value of all
Counterparty
Valuation
contracts covered by the agreement.
Adjustments
CVA and CCDS No netting agreement:

Hedging CVA
• Investor owns two 5 year 5% swaps to counterparty — one is pay
CDS Proxies
Portfolio
fixed, the other is receive fixed.
counterparty risk
• No net market exposure — the two positions cancel out.
CVA & Basel
CVA VaR • Substantial counterparty exposure:

Accounting • Get recovery on the market value of the positive swap
considerations
• Still owe full value on the market value of the negative swap
CVA issues
Summary
With netting agreements, exposure at any given time is on the net
Appendix —
CDS Spreads market value of all securities covered. Risk is reduced.
References
10 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Risk modifications —

Markets and
risks
Collaterialization
Counterparty
risk
Risk
Modifications Collateralization
Counterparty
Valuation • ISDA credit support annex (CSA).
Adjustments
CVA and CCDS • At the end of the period (day, week, etc), if swap value exceeds a
Hedging CVA threshold, collateral must be posted.
CDS Proxies
• Exposure is to the threshold plus the movement of the market
Portfolio
counterparty risk value over the period.
CVA & Basel
CVA VaR • Exposed to threshold plus the market moves between default and
Accounting
considerations
liquidation.
CVA issues • Leads to requirement of initial margin from clearing houses.
Summary
Appendix —
CDS Spreads
References
11 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Unintended consequences

Markets and
risks
Counterparty
risk
Risk Risk modifications change effective seniority structure of debt.

Modifications
Counterparty • Netting — contracts which are assets are used to pay some
Valuation
Adjustments contracts which are debts in advance of other debts.
CVA and CCDS
• Collateralization — cash and other assets of firm used as
Hedging CVA
collateral are used to pay corresponding contracts in advance of
CDS Proxies
Portfolio
other debts.
counterparty risk
• Reduces value to bond holders and other creditors.
CVA & Basel
CVA VaR So, to a certain extent, using risk mitigants on derivatives debt is
Accounting
considerations
really transfering the risk to the other creditors.
CVA issues
Summary
Appendix —
CDS Spreads
References
12 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Risk mitigation and clearing houses

Markets and
risks
Counterparty
risk
Clearing houses mitigate risk via netting, collateralization, and
Risk
reassignment of contracts.
Modifications
Counterparty • Initial margin plus daily variational margin attempts to make

Valuation
Adjustments needed cash available in the event of a default.
CVA and CCDS • If that proves insufficient, coverage comes from backers of
Hedging CVA
clearing house and equity of the firm itself.
CDS Proxies
Portfolio
• Netting is a two edged sword — with one clearing house, there’s
counterparty risk
definitely increased netting. With multiple clearing houses and
CVA & Basel
some structures not on clearing houses, it’s unclear if netting is
CVA VaR
increased.
Accounting
considerations
• Reassignment of contracts prevents losses due to market impact.
CVA issues
But in a real crisis, reassignment might fail.
Summary
Appendix —
CDS Spreads
References
13 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Risk mitigation clauses

Markets and Another approach to risk mitigation that is becoming popular is to
risks
include risk mitigation clauses (A.K.A. Additional Termination
Counterparty
risk Events):
Risk
Modifications
• Contract can be closed out at replacement value if the
Counterparty
Valuation counterparty’s rating drops.
Adjustments
CVA and CCDS

• Contract can be closed out at market value prior to maturity.
Hedging CVA
CDS Proxies
Issues:
Portfolio
counterparty risk • What exactly is “replacement value”?
CVA & Basel
• How is closing out a swap early any different from entering into
CVA VaR
the reversing swap and novation?
Accounting
considerations
• By forcing closeout at a rating change, are you decreasing
CVA issues
counterparty risk while increasing systemic risk?
Summary
Appendix —
• Will this really work at the next crisis — what if the market is
CDS Spreads
illiquid?
References
14 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Counterparty Valuation

Markets and
risks
Adjustments
Counterparty
risk
Risk
Modifications
Counterparty
Valuation
Adjustments How does the counterparty exposure and the risk of default impact
CVA and CCDS
the value of the security?
Hedging CVA
CDS Proxies • The Credit Valuation Adjustment (CVA) is the cost of the
Portfolio
counterparty risk
potential loss.
CVA & Basel • Risk free price - CVA = price of risky security.
CVA VaR
Accounting
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
15 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Counterparty risk — long only vs

Markets and
risks
long/short
Counterparty
risk
Risk
Modifications Counterparty risk calculations are far more complicated for
Counterparty
Valuation
instruments that are a combination of long/short positions than for
Adjustments long only instruments.
CVA and CCDS
Hedging CVA • In a long only instrument, (like a bond position), counterparty

CDS Proxies risk can be judged by using models that can incorporate a
Portfolio
counterparty risk
discount curve shift.
CVA & Basel • In a long/short instrument, (like a swap position), the instrument
CVA VaR can potentially be an asset or a liability. When it’s an asset,
Accounting
considerations
default results in a loss. When it’s a liability, default results in no
CVA issues
change.
Summary
Appendix —
CDS Spreads
References
16 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Bond counterparty risk

Markets and
risks If a bond with a coupon of C pays f times per year at times ti , with
Counterparty maturity tn , the value of the bond is:
risk
Risk
X
n Z tn
Modifications C
D(ti )S(ti ) + 100D(tn )S(tn ) + 100RD(t)P(t)dt,
Counterparty
Valuation f
Adjustments
CVA and CCDS
Hedging CVA
• P(t) — default probability density function.
CDS Proxies Rt
Portfolio • S(t) = 1 − P(s)ds — survival probability for time t (the
counterparty risk
probability of no default before time t).
CVA & Basel
CVA VaR
• R — bond recovery rate.
Accounting • D(t) — risk free discount factor for time t.
considerations
CVA issues
This is the standard CDS model applied to a fixed coupon bond
Summary
(details in the Appendix). Note that it assumes independence of rates
Appendix —
CDS Spreads and default.
References
17 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Equivalent par curve

Markets and
risks
Counterparty
risk
Risk
Modifications
To get a feel for the impact of credit spreads on bond values, we can
Counterparty compute the par curve for the risky bond. For ti = i/f , and for each
Valuation
Adjustments n, solve for C (tn ) such that
CVA and CCDS
X
n Z tn
Hedging CVA C (tn )
100 = D(ti )S(ti ) + 100D(tn )S(tn ) + 100RD(t)P(t)dt,
CDS Proxies f
Portfolio
counterparty risk
Then C (tn ) is the implied par curve — the coupons that the issuer
CVA & Basel
CVA VaR
with this CDS spread curve would theoretically use to issue debt at
Accounting
par.
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
18 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Equivalent par curve example

Markets and
On the Bloomberg terminal, we do this calculation in YASN - the
risks
structured notes calculation screen.
Counterparty
risk
Risk
Modifications
Counterparty
Valuation
Adjustments
CVA and CCDS
Hedging CVA
CDS Proxies
Portfolio
counterparty risk
CVA & Basel
CVA VaR
Accounting
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
19 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Recovery rate impact

Markets and
Flat curves and equal recovery rates give an equivalent par curve
risks roughly equal to the swap curve shifted by the CDS spreads.
Counterparty
risk Otherwise, there can be significant differences, as we see here for a
Risk flat 100bp CDS curve and a flat 3% swap curve.
Modifications
Counterparty
Valuation
Adjustments
CDS recovery 0% 40% 80% 40%
CVA and CCDS Bond recovery 0% 40% 80% 0%
Hedging CVA Term
CDS Proxies 1 Wk 4.2216 4.2241 4.2366 5.0783
Portfolio
counterparty risk
1 Mo 4.0041 4.0071 4.0223 4.7141
CVA & Basel
6 Mo 3.9861 3.9948 4.0385 4.6741
CVA VaR 1 Yr 4.0206 4.0361 4.1144 4.7196
Accounting 2 Yr 4.0333 4.0420 4.0857 4.7252
considerations
3 Yr 4.0333 4.0419 4.0855 4.7251
CVA issues
Summary
5 Yr 4.0322 4.0408 4.0845 4.7233
Appendix —
7 Yr 4.0321 4.0407 4.0843 4.7231
CDS Spreads
10 Yr 4.0316 4.0403 4.0839 4.7223
References
20 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Par curves vs default probabilities

Markets and
Once the risky par curve is computed, one is tempted to strip it and
risks use it for discounting.
Counterparty
risk Good points:
Risk
Modifications
• This is simple, straight forward, and in line with common
Counterparty practices.
Valuation
Adjustments • This will properly price par bonds back to par.
CVA and CCDS • If the bond recovery rate is zero, this properly prices all bonds!
Hedging CVA
CDS Proxies
In the zero recovery rate case, this makes the risky discount factors
Portfolio
counterparty risk
D(ti )S(ti ),
CVA & Basel
so the risky spot rate curve R̄ is given by
CVA VaR
Accounting
considerations
R̄(ti ) = − log(D(ti )S(ti ))/ti = R(ti ) − log(S(ti ))/ti ,
CVA issues
where R is the risk free rate. So, the survival probabilities add a
Summary
Appendix —
spread of − log(S(ti ))/ti (the average hazard rate) to the risk free
CDS Spreads rate. This spread is roughly the CDS spread, adjusted by the CDS
References
recovery rate.
21 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Par curves vs default probabilities

Markets and
risks
Counterparty
risk Using the risky par curve directly for calculations also has drawbacks.
Risk
Modifications • For nonzero recovery rates, prices produced by this method on
Counterparty
Valuation non-par bonds will differ from the default based method.
Adjustments
CVA and CCDS There is a difference between the two methods when the bond is a
Hedging CVA couple of hundred basis points away from par (roughly 5 to 15 basis
CDS Proxies points for a 100 bp CDS spread), but given other general
Portfolio
counterparty risk
uncertainties (such as the recovery rate, or the spread between bonds
CVA & Basel and CDS), this is a reasonable margin of error, and can be folded into
CVA VaR OAS adjustments.
Accounting
considerations Bottom line — adding a spread to the risk free curve is a reasonable
CVA issues way to price risky bonds.
Summary
Appendix —
CDS Spreads
References
22 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Par curves and OAS

Markets and
risks
Counterparty
risk
Risk
Modifications
Counterparty
Valuation
Adjustments If the spread curve is flat, it roughly amounts to a shift of the par
CVA and CCDS curve, which is roughly adding an OAS.
Hedging CVA
CDS Proxies
Using the risky par curve in such a calculation is an improvement, in
Portfolio
that it factors in the shape of the CDS spread curve.
counterparty risk
CVA & Basel
CVA VaR
Accounting
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
23 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Advanced risky bond calculations

Markets and
Once we have a risky par curve, we can use it to apply risky spreads
risks on top of a risk free interest rate derivatives models to analyze risky
Counterparty
risk
bonds with embedded options:
Risk
Modifications
Counterparty
Valuation
Adjustments
CVA and CCDS
Hedging CVA
CDS Proxies
Portfolio
counterparty risk
CVA & Basel
CVA VaR
Accounting
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
24 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Counterparty risk in swaps —

Markets and
risks
Characteristics
Counterparty
risk Consider a five year swap in a flat interest rate environment.
Risk
Modifications There is volatility dependent risk:
Counterparty
Valuation • Zero volatility:
Adjustments
CVA and CCDS

• Swap is always nearly zero market value.
Hedging CVA
• Minimal default risk.
CDS Proxies • High volatility:
Portfolio • Swap value in future can be substantial.
counterparty risk
CVA & Basel

• Potentially substantial loss upon default.
CVA VaR
There is curve dependent risk:
Accounting
considerations
• In a steep interest rate environment, the swap is expected to be
CVA issues
Summary
heavily off the money for the duration of its life.
Appendix — • Far more counterparty risk.
CDS Spreads
References
25 / 79
Counterparty
Risk, CVA, and
Basel III

Markets and
risks
Valuation
Counterparty
risk
Let V (t) be the value of the risk-free swap at time t, and let R be
Risk the recovery rate on the underlying swap.
Modifications
Counterparty
If the counterparty defaults at time τ , the payoff for holding the swap
Valuation
Adjustments
is:
CVA and CCDS
• If V (τ ) > 0 we get R × V (τ ).
Hedging CVA
CDS Proxies • If V (τ ) < 0 we still owe V (τ ).

Portfolio
counterparty risk The above payoff is
CVA & Basel
CVA VaR R max(V (τ ), 0) + min(V (τ ), 0)

Accounting
considerations = V (τ ) − (1 − R) max(V (τ ), 0)
CVA issues
Summary The quantity max(V (τ ), 0) is the payoff of an option to enter into

Appendix — what’s left of the swap at time τ — i.e. - a swaption maturing at
CDS Spreads
References
time τ .
26 / 79
Counterparty
Risk, CVA, and
Basel III

Markets and
risks
Valuation
Counterparty
risk The loss at default time τ is:
Risk
Modifications
Counterparty
(1 − R) max(V (τ ), 0)
Valuation
Adjustments
CVA and CCDS The cost of this loss is:

Hedging CVA
CDS Proxies CVA = N(0)E [(1 − R) max(V (τ ), 0)/N(τ )1τ <T ]

Portfolio Z T
counterparty risk
= (1 − R)N(0)E [ max(V (t), 0)/N(t)δ(t − τ )dt]
CVA & Basel
0
CVA VaR Z T
Accounting
considerations
= (1 − R)N(0) E [max(V (t), 0)/N(t)δ(t − τ )]dt
0
CVA issues
Summary where the expectation is the equivalent martingale measure with

Appendix —
CDS Spreads
respect to the numeraire N.
References
27 / 79
Counterparty
Risk, CVA, and
Basel III

Markets and
risks
Valuation
Counterparty
risk
If max(V (t), 0)/N(t) (the value of the call) and δ(t − τ ) (the default
Risk
Modifications event) are independent, then the expectation factors and the CVA is:
Counterparty
Valuation Z T
Adjustments
CVA and CCDS

CVA = (1 − R)N(0) E [max(V (t), 0)/N(t)δ(t − τ )]dt
0
Hedging CVA Z T
CDS Proxies
= (1 − R) N(0)E [max(V (t), 0)/N(t)]E [δ(t − τ )]dt
Portfolio 0
counterparty risk
Z T
CVA & Basel
= (1 − R) S(t)P(t)dt
CVA VaR
0
Accounting
considerations
where S(t) is the current value of the swaption to enter into the
CVA issues
Summary
remainder of the swap at time t, and P(t) is the default time
Appendix —
probability density function.
CDS Spreads
References
28 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein CVA subtleties

Markets and
risks CVA valuation can be a little subtle. First of all, it’s impractical to
Counterparty
risk
compute the above integral. One approach is discretization. Divide
Risk the time interval [0, T ] into periods [ti , ti+1 ], and select t̄i ∈ [ti , ti+1 ].
Modifications
Then
Counterparty
Valuation Z T
Adjustments
CVA and CCDS

CVA = (1 − R) S(t)P(t)dt
0
Hedging CVA X
CDS Proxies ≈ (1 − R) S(t̄i )P̄(ti )
Portfolio
counterparty risk R ti+1
CVA & Basel
where P̄(ti ) = ti
P(t)dt is the probability of defaulting in interval
CVA VaR [ti , ti+1 ].
Accounting
considerations The finer one subdivides the time period, the more accurate the
CVA issues calculation. We’ve found that it suffices to divide the time interval
Summary according to the cashflows of the swap as long as one values the call
Appendix —
CDS Spreads
options at the midpoints of the intervals.
References
29 / 79
Counterparty
Risk, CVA, and
Basel III

Markets and
risks
Another issue is accurately valuing S(t̄i ). This is the value of the
Counterparty
risk option to enter into the tail of the swap, which is slightly different
Risk
Modifications
from the swaption maturing at time t̄i .
Counterparty Swap cash flows:
Valuation
Adjustments
CVA and CCDS
Hedging CVA Floating Cashflows

CDS Proxies
Portfolio
counterparty risk
Time
CVA & Basel
CVA VaR
Fixed Cashflows
Accounting
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
30 / 79
Counterparty
Risk, CVA, and
Basel III

Markets and
risks
When exercising a swaption to enter into a swap with an odd first
Counterparty
coupon, the first coupon is adjusted — the floating leg references a
risk
shortened index, and the fixed leg only accrues over the remainder of
Risk
Modifications the period.
Counterparty
Valuation
However, in the event of default at that time, the full cash flows are
Adjustments
lost.
CVA and CCDS
Hedging CVA
CDS Proxies Floating Cashflows

Portfolio
counterparty risk
CVA & Basel
CVA VaR
Exercise Time
Accounting
considerations Fixed Cashflows
CVA issues
Summary
Appendix —
CDS Spreads
References
31 / 79
Counterparty
Risk, CVA, and
Basel III

Markets and
The difference between the value of the forward start swap and the
risks corresponding tail of the underlying swap will typically be substantial,
Counterparty
risk
as is illustrated in the following graph of the two for a 5 year at the
Risk
money payer swap, on a 1 million notional in a rising interest rate
Modifications
environment.
Counterparty
Valuation
Adjustments
CVA and CCDS
Hedging CVA

CDS Proxies
Portfolio
counterparty risk

CVA & Basel

CVA VaR
Accounting
considerations
CVA issues

Summary
Appendix —
CDS Spreads
References
32 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Forward swap rates

Markets and
risks
Consider the pay fixed swap with fixed rate F that the holder of a
Counterparty swaption would receive on exercise (at time t).
risk
Risk
Let the underlying floating (fixed) leg pay at times ti (ti′ ). Then the
Modifications
time t value of the swap is1
Counterparty
Valuation
X X
Adjustments S(t) = L(t, ti , ti+1 )Z (t, ti+1 )αi − F αi′ Z (t, ti′ ),
CVA and CCDS Since t ≤ t1 ,
Hedging CVA
L(t, ti , ti+1 ) = (1/αi′′ )(Z (t, ti )/Z (t, ti+1 ) − 1)
CDS Proxies
Portfolio
(with Libor accrual factor αi′′ ), so if we assume αi = αi′′ , then S(t)
counterparty risk
can be written in terms of Z as:
CVA & Basel X
CVA VaR
S(t) = Z (t, t1 ) − Z (t, tn ) − F αi′ Z (t, ti′ ),
Accounting This gives us the standard expression for the time t value of a forward
considerations
start swap.
CVA issues
Summary 1 L(w , x, y ) is the time w forward Libor rate setting at time x and paying at time y ,
Appendix —
CDS Spreads
Z (x, y ) be the time x price of a zero coupon bond paying at time y , and αi (α′i ) are
the accrual fractions for floating (fixed) payment periods ti to ti+1 (ti′ to ti+1
′ ),
References
respectively.
33 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Tail swap rate

Markets and
The tail of the existing swap is slightly different. Its cash flows and
risks accruals are the same as for the forward swap, except for the first
Counterparty
risk
period, where the reset date (t̄1 ) is earlier than the valuation time t,
Risk and the accrual factors correspond to the full period rather than the
Modifications
partial period. The time t value of the first floating cash flow is
Counterparty
Valuation
Adjustments Z (t, t2 )
CVA and CCDS
− Z (t, t2 ),
Z (t̄1 , t2 )
Hedging CVA
CDS Proxies
so the value of the tail is
Portfolio
counterparty risk
Z (t, t2 ) X
CVA & Basel S̄(t) = − Z (t, t2 ) + L(t, ti , ti+1 )Z (t, ti+1 )αi
CVA VaR
Z (t̄1 , t2 ) 2
Accounting
X
considerations −F ᾱ1′ Z (t, t1 ) − F αi′ Z (t, ti′ )
CVA issues 2
Summary Z (t, t2 ) X
Appendix — = − Z (t, tn ) − F ᾱ1′ Z (t, t1 ) − F αi′ Z (t, ti′ ).
CDS Spreads Z (t̄1 , t2 ) 2
References
34 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Forward swaps and tail swaps

Markets and
risks
compared
Counterparty
risk
Risk
Modifications
Counterparty
Valuation
Adjustments The difference between the values of the two is
CVA and CCDS
Z (t, t2 )
Hedging CVA
S̄(t) − S(t) = − Z (t, t1 ) − F (ᾱ1′ − α1′ )Z (t, t1 ).
CDS Proxies Z (t̄1 , t2 )
Portfolio
counterparty risk
This is the adjustment that needs to be made to convert the swaption
CVA & Basel
payoff to the payoff of the option on the tail swap.
CVA VaR
Accounting
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
35 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Forward swap vs tail

Markets and
Under the equivalent martingale measure with respect to numeraire
risks N, the time zero value of the forward start swaption is
Counterparty
risk N(0)E0 [max(S(t), 0)/N(t)]
Risk
Modifications Instead of this, we need the value of the option on the tail of the
Counterparty
Valuation
swap. Concentrating on the payoff, if
Adjustments
Z (t, t2 )
CVA and CCDS D = S̄(t) − S(t) = − Z (t, t1 ) − F (ᾱ1′ − α1′ )Z (t, t1 )
Hedging CVA Z (t̄1 , t2 )
CDS Proxies we have
Portfolio
counterparty risk
max(S̄(t), 0) = max(S(t) + D, 0) = max(S(t), −D) + D
CVA & Basel
CVA VaR So the option to enter into the tail of the swap is the same as the
Accounting option to enter into the swaption with an associated fee of
considerations
S(t) − S̄(t) with an additional cash component of S̄(t) − S(t).
CVA issues
Summary
One could apply a convexity adjustment technique to work this out,
Appendix —
but it replacing Z (t, t1 ) and Z (t, t̄1 ) with their corresponding time
CDS Spreads zero forward values is a reasonable approximation and it makes these
References
adjustments fixed values and thus easily handled.
36 / 79
Counterparty
Risk, CVA, and
Basel III

Markets and
risks
CVA<Go>
Counterparty
On the Bloomberg, the CVA function does this calculation.
risk
Risk
Modifications
Counterparty
Valuation
Adjustments
CVA and CCDS
Hedging CVA
CDS Proxies
Portfolio
counterparty risk
CVA & Basel
CVA VaR
Accounting
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
37 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Visualizing counterparty risk

Markets and
The calculation and risks can be better visualized by exposure graphs
risks
over time:
Counterparty
risk
Risk
Modifications
Counterparty
Valuation
Adjustments
CVA and CCDS
Hedging CVA
CDS Proxies
Portfolio
counterparty risk
CVA & Basel
CVA VaR
Accounting
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
38 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein CCDS

Markets and
risks Contingent credit default swaps are a variant of credit default swaps.
Counterparty
risk With both CDS and CCDS, the buyer of insurance pays the spread
Risk
Modifications
until default or maturity (whichever comes first), and receives (1 − R)
Counterparty
of something at default time.
Valuation
Adjustments What makes the two contracts substantially different is what exactly
CVA and CCDS is being made whole.
Hedging CVA
CDS Proxies • CDS — Receives (1 − R) on the principal of a reference secrity.

Portfolio
counterparty risk • CCDS — Receives (1 − R) on the value of a reference security.
CVA & Basel
CVA VaR With a CDS contract, the only uncertainty regarding the former’s
Accounting payout given default is the recovery rate, because recovery is on a
considerations
fixed principal amount.
CVA issues
Summary With a CCDS contract, there is uncertainty on the value on which

Appendix —
CDS Spreads
recovery is applied as well.
References
39 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein CCDS and CVA

Markets and
risks
Consider a CCDS contract on an interest rate swap.
Counterparty
risk
• The payoff of the floating leg of the CCDS contract is the loss
Risk given default for the interest rate swap.
Modifications
Counterparty
• The value of the floating leg is then the CVA value.
Valuation
Adjustments • The fixed leg of the CCDS contract must then be the annuitized
CVA and CCDS (to default time) value of the CVA.
Hedging CVA
CDS Proxies The CCDS contract is thus the perfect hedge for the counterparty
Portfolio exposure on the interest rate swap, except that
counterparty risk
CVA & Basel • CCDS contracts are thought to be expensive forms in which to
CVA VaR buy this protection.
Accounting
considerations • CCDS contracts have counterparty risk too!
CVA issues
Summary
Any hedge that would be used for the counterparty risk in a swap can
Appendix — be adjusted to serve as the replicating strategy for the CCDS contract
CDS Spreads
(except for the counterparty exposure of the CCDS).
References
40 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Hedging CVA

Markets and
risks
Counterparty
risk
Risk
Modifications
Hedging counterparty risk can be difficult. Since
Counterparty X
Valuation
Adjustments
CVA ≈ (1 − R) S(t̄i )P̄(ti )
CVA and CCDS
Hedging CVA It’s natural to hedge with a portfolio of swaptions and CDSs:
CDS Proxies
• Take (1-R)P̄(ti ) positions in swaptions maturing at t̄i
Portfolio
counterparty risk
CVA & Basel

and
CVA VaR
• Take positions in CDS that neutralize exposure to credit moves.
Accounting
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
41 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Hedging issues

Markets and
By virtue of the swaption positions (and the low sensitivity of CDS to
risks interest rates:
Counterparty
risk • Changes in CVA due to changes in interest rates and interest rate
Risk
Modifications
volatility are hedged.
Counterparty • Changes in CVA due to changes in credit spreads are hedged.
Valuation
Adjustments
However
CVA and CCDS
Hedging CVA • CDS position needs rebalancing when interest rates and vols
CDS Proxies change — Dynamic hedging with CDS is expensive.
Portfolio
counterparty risk • Swaption position needed changes when CDS spreads change —
CVA & Basel Dynamic hedging with swaptions is expensive.
CVA VaR • Cross gamma risk.
Accounting
considerations • Where can you get a risk free swap?
CVA issues
• What about the CVA of the CDS and the swaptions for that
Summary
matter?
Appendix —
CDS Spreads • How do you neutralize credit exposure and default exposure at
References
the same time?
42 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein No CDS CDS

Markets and
risks
Counterparty
risk If CDS are not available for a given name, there are alternative ways
Risk to estimate default probabilities:
Modifications
Counterparty
Valuation • Back out a CDS spread from a bond par curve.
Adjustments
CVA and CCDS

• Use a generic spread (CDS indices).
Hedging CVA • Use CDS spreads from a similar name.
CDS Proxies
• Use CDS spreads from a similar sector and credit rating.
Portfolio
counterparty risk • Use historically estimated default probabilities for a similar sector
CVA & Basel
and credit rating.
CVA VaR
Accounting
• Estimate credit risks from low strike puts.
considerations
CVA issues Each of these methods has its own problems.

Summary
Appendix —
CDS Spreads
References
43 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Bond spread method

Markets and
risks
Counterparty
risk
Risk
Modifications
Counterparty Backing out spreads from bonds is closest to yielding the price of a
Valuation
Adjustments hedge. Problems include:
CVA and CCDS
Hedging CVA • Liquidity risk is priced into bond yields.

CDS Proxies
• Issued bonds might trade rarely or not trade at all.
Portfolio
counterparty risk • To actualize the hedge, one would have to short the bond.
CVA & Basel
CVA VaR
Accounting
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
44 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Proxy CDS method

Markets and Using other CDS spreads includes:
risks
Counterparty • Use a generic spread (CDS indices).

risk
Risk • Use CDS spreads from a similar name.

Modifications
Counterparty
• Use CDS spreads from a similar sector and credit rating.
Valuation
Adjustments
On the plus side:
CVA and CCDS
Hedging CVA • Can give reasonable estimates of CDS spreads.

CDS Proxies
• Can be highly correlated with a particular name.
Portfolio
counterparty risk
CVA & Basel

On the minus side:
CVA VaR
• They all break down when that name deteriorates.
Accounting
considerations • They fail to capture idiosyncratic changes.
CVA issues
• They fail to capture the actual default event.
Summary
Appendix — • In the latter case, a name will deteriorate well before its credit
CDS Spreads
References
rating changes, leading to an underestimate of credit risk.
45 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Historical default probabilities

Markets and
risks
Counterparty
risk
Risk
Modifications
Counterparty Using historical default probabilities has its own problems.

Valuation
Adjustments
CVA and CCDS

• Same problems as the sector/credit rating approach.
Hedging CVA • Cannot even hedge market moves.
CDS Proxies
• Does not respond to changing market conditions.
Portfolio
counterparty risk
• Not a market price (historical, not implied).
CVA & Basel
CVA VaR
Accounting
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
46 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Option estimation approach

Markets and
risks
Counterparty Using low strike puts to estimate and hedge credit risk is an
risk
interesting approach.
Risk
Modifications
Counterparty
• Options can be purchased to effect the hedge.
Valuation
Adjustments • Options capture the default event.
CVA and CCDS
• Good correlation to credit risk.
Hedging CVA
CDS Proxies On the other hand:

Portfolio
counterparty risk • Can be a disconnect between stock prices and credit (GM and
CVA & Basel
Kerkorian in 2005).
CVA VaR
Accounting
• Adds substantial model risk — model must capture equity moves
considerations
along with credit risk and default events.
CVA issues
Summary
• Too volatile — adds exposure to spot price moves.
Appendix —
CDS Spreads
References
47 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein General proxy issues

Markets and
risks
Counterparty
risk
Risk
Modifications
More generally, if a proxy is used which cannot effect a hedge, then
Counterparty
Valuation
Adjustments • The price is not an arbitrage free price.
CVA and CCDS • The credit risk cannot be hedged.
Hedging CVA
CDS Proxies In this situation, one should consider managing the risk differently:
Portfolio
counterparty risk • Collateralize.
CVA & Basel
• Capitalize based on real world potential future exposures.
CVA VaR
Accounting
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
48 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Portfolio counterparty risk

Markets and
risks
Counterparty
risk
Risk
Modifications
In the presence of netting agreements, the counterparty exposure
Counterparty
Valuation must be computed on the portfolio of securities covered by the
Adjustments
netting agreement.
CVA and CCDS
Hedging CVA
• Loss given default is no longer the sum of the losses in the
CDS Proxies
individual positions.
Portfolio
counterparty risk • Loss given default is (1 − R) of the payoff of a call on the
CVA & Basel
underlying portfolio.
CVA VaR
Accounting
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
49 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Calculating portfolio counterparty

Markets and
risks
risk
Counterparty
risk
Risk
Modifications
Counterparty
Valuation
Calculation options:
Adjustments
CVA and CCDS • Feed call options on the portfolio to your defaultable interest rate
Hedging CVA derivatives valuation system.
CDS Proxies
• Compute value of appropriate call options on the portfolio via
Portfolio
counterparty risk your interest rate derivatives valuation system, and proceed as
CVA & Basel above.
CVA VaR
• Make some assumptions and get formulas.
Accounting
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
50 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Calculating portfolio counterparty

Markets and
risks
risk
Counterparty
risk
Risk
Modifications Brigo and Massimo take the latter approach.
Counterparty
Valuation • Net all of the interest rate swaps covered by a given netting
Adjustments
CVA and CCDS

agreement.
Hedging CVA • The floating “leg” of the portfolio is now some sort of amortizing
CDS Proxies floating leg with time dependent leverage.
Portfolio
counterparty risk • The fixed “leg” is now some sort of amortizing step coupon fixed
CVA & Basel leg.
CVA VaR
Accounting
Difficulties ensue because the leverage could be positive or negative
considerations
(i.e. - at some times the aggregate can be a payer swap while at
CVA issues
other times it can be a receiver swap).
Summary
Appendix —
CDS Spreads
References
51 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Portfolio woes

Markets and
risks
Counterparty
risk
Risk Consider a portfolio consisting of an at the money 5 year payer swap

Modifications
with fixed rate F1 , and an at the money 3 year receiver swap with
Counterparty
Valuation fixed rate F2 .
Adjustments
CVA and CCDS

• Credit exposure 1 year out is to the difference between the value
Hedging CVA
of the two tails being positive.
CDS Proxies
Portfolio
• Similar to the difference between the 4 year swap rate and the 2
counterparty risk
year swap rate.
CVA & Basel
CVA VaR
The portfolio behaves roughly like a spread between two rates, so the
Accounting
considerations credit exposure is like a spread option, and thus, difficult to price.
CVA issues
Summary
Appendix —
CDS Spreads
References
52 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein CVA & Basel

Markets and
risks
Counterparty
risk
Risk
Modifications
Basel II requires capital to be held for credit risk.
Counterparty
Valuation • Standardized approach — Weight by credit rating.
Adjustments
CVA and CCDS • Current exposure method — EAD = CE + PFE - Collateral.

Hedging CVA PFE is estimated as the notional times a multiplier based on
CDS Proxies asset type and maturity.
Portfolio
counterparty risk • Internal Rating Based approach (AKA Internal Model Method)
CVA & Basel — 1.4 × Effective Expected Positive Exposure (EEPE = average
CVA VaR Expected Exposure).
Accounting
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
53 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Basel III

Markets and
risks
Counterparty
risk
Risk Basel III expands on Basel II.

Modifications
Counterparty
Valuation
• IMM approved banks must account for credit spread vol on OTC
Adjustments derivatives — CVA VaR.
CVA and CCDS
• Advanced CVA charge — market risk rules of 10 day 99% VaR
Hedging CVA
• Standardized CVA charge — uses formula which assumes 1 year
CDS Proxies
99.9%.
Portfolio
counterparty risk • Incremental Risk Charge — account for credit default and
CVA & Basel
migration risks.
CVA VaR
Accounting
• Comprehensive Risk Measure.
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
54 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Basel III CVA formula

Markets and
risks
Counterparty
risk
Risk
Modifications Basel III stipulates a formula for the standard CVA charge. In our
Counterparty notation, it’s
Valuation
Adjustments
CVA and CCDS
Hedging CVA X −si−1 ti−1 −si t i S(ti−1 ) + S(ti )

CVA = (1 − R) max(0, e 1−R − e (1−R) )
CDS Proxies
2
Portfolio
i
counterparty risk
CVA & Basel

where si is the credit spread at time ti , and S(t) is the current value
CVA VaR
of the option to enter into the underlying portfolio.
Accounting
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
55 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein CVA VaR

Markets and
CVA:
risks
Counterparty
• Price embedded default risk of all securities under netting
risk
agreement with counterparty, taking collateralization into
Risk
Modifications account.
Counterparty
Valuation
• Requires pricing portfolio of options on the portfolio, which itself
Adjustments
might contain options.
CVA and CCDS
Hedging CVA
• Complicated model.
CDS Proxies • American Monte Carlo.
Portfolio
counterparty risk CVA VaR
CVA & Basel
CVA VaR • Compute distribution of CVAs one year out.

Accounting
considerations
• Compute best of the worst — best case loss of the worst X%
CVA issues losses (99.9% VaR on a 1 year horizon for credit risk, 99% and
Summary 10 day for Basel III).
Appendix —
CDS Spreads
• Simulate market moves, credit spread changes and defaults
References (under the real world measure) to the horizon date.
56 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein CVA VaR Issues

Markets and
risks
Counterparty
risk
Risk
CVA VaR, while conceptually straight forward, raises a number of
Modifications issues.
Counterparty
Valuation
Adjustments
• A one year VaR horizon is much longer than typically computed
CVA and CCDS in VaR — trading impact, hedging impact, and life cycle events
Hedging CVA all become significant.
CDS Proxies
• Data is sparse — CVA can be priced using market data (spreads,
Portfolio
counterparty risk option prices, etc). CVA VaR needs real world default
CVA & Basel probabilities, recovery rates, ...
CVA VaR
• Exploring the variation of a complicated model under
Accounting
considerations perturbations of inputs we cannot estimate.
CVA issues
Summary
Appendix —
CDS Spreads
References
57 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Accounting considerations

Markets and
risks
Counterparty
risk
Risk
Modifications Accounting for counterparty risk is required by accounting standards
Counterparty FASB 157 (US) and IAS 39 (Europe).
Valuation
Adjustments
FASB 157, Appendix B, Paragraph 5:
CVA and CCDS
Hedging CVA • B5. Risk-averse market participants generally seek compensation

CDS Proxies
for bearing the uncertainty inherent in the cash flows of an asset
Portfolio
counterparty risk or liability (risk premium). A fair value measurement should
CVA & Basel include a risk premium reflecting the amount market participants
CVA VaR would demand because of the risk (uncertainty) in the cash flows.
Accounting
considerations
CVA issues
Summary
Appendix —
CDS Spreads
References
58 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Accounting considerations

Markets and
risks
Counterparty
risk
Accounting rules are currently being updated and globalized.
Risk Topic 820, Section 10 replaces FASB 157, but it strengthens the
Modifications
Counterparty
position that credit risk must be accounted for:
Valuation
Adjustments • 55-8. A fair value measurement should include a risk premium
CVA and CCDS
reflecting the amount market participants would demand because
Hedging CVA
of the risk (uncertainty) in the cash flows. Otherwise, the
CDS Proxies
Portfolio
measurement would not faithfully represent fair value. In some
counterparty risk cases, determining the appropriate risk premium might be
CVA & Basel difficult. However, the degree of difficulty alone is not a sufficient
CVA VaR
basis on which to exclude a risk adjustment.
Accounting
considerations
While accurate valuation of the embedded default risk is preferred, a
CVA issues
Summary
number of alternative approaches have traditionally been accepted.
Appendix —
CDS Spreads
References
59 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Shifty calculations

Markets and One alternative is the discount shift method. The CVA is calculated
risks
by shifting the discount curve by the credit spread (like in the bond
Counterparty
risk approach).
Risk
Modifications When all of the swaptions are in the money, this is the zero volatility
Counterparty version of the CVA. For a 5 year 5% receiver swap, ignoring volatility
Valuation
Adjustments can lead to errors of 15% to 20%.
CVA and CCDS
Hedging CVA
Table: 5% 5 year receiver swap, 10 million notional, with market data
CDS Proxies
yielding swap rate of 3.16%
Portfolio
counterparty risk
CVA & Basel

CDS rate CVA Discount Shift
CVA VaR
0 0 0
Accounting
considerations 100 17,610 15,227
CVA issues 200 35,751 29,970
Summary 300 52,950 44,249
Appendix —
CDS Spreads
400 69,181 58,085
References
60 / 79
Counterparty
Risk, CVA, and
Basel III

Markets and
The errors from the shift approximation are illustrated by the results
risks on an at the money swap.
Counterparty
risk
Risk
Table: At the money 5 year receiver swap, 10 million notional, with market
Modifications data
Counterparty
Valuation
Adjustments CDS rate CVA Discount Shift
CVA and CCDS 0 0 0
Hedging CVA 100 11,852 7,348
CDS Proxies 200 22,870 14,316
Portfolio
counterparty risk
300 33,108 20,923
CVA & Basel 400 42,620 27,189
CVA VaR
In general:
Accounting
considerations • Errors grow as relevant swaptions go out of the money.
CVA issues
• Large errors when swaptions are close to the money (so that
Summary
volatility plays a large part).
Appendix —
CDS Spreads • Cannot be used for out of the money swaps — CVA is negative.
References • Same typically holds for pay fixed swaps.
61 / 79
Counterparty
Risk, CVA, and
Basel III

Markets and
risks
To address the problem of swaps that are a liability, the discount
Counterparty
risk method is modified.
Risk
Modifications
• Positions of positive value:
Counterparty
Valuation • Treat as assets.
Adjustments
• Reduce value according to counterparty risk (by discount curve
CVA and CCDS
Hedging CVA
shift).
CDS Proxies • Positions of negative value:
Portfolio • Treat as liabilities.
counterparty risk
• Increase value (reduce liability) according to investor’s credit!
CVA & Basel
CVA VaR
Accounting
Nicely symmetric and based on the theory that the money is owed to
considerations the counterparty and only a fraction of it will be paid if the investor
CVA issues
defaults.
Summary
Appendix —
But adds problems of bilateral CVA.
CDS Spreads
References
62 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Net current exposure

Markets and
risks
Counterparty
risk Another approach — the current net exposure method.
Risk
Modifications • Net current exposure = value at immediate risk of default to
Counterparty
Valuation
investor = max(current market value, 0).
Adjustments
• Cost of the risk of immediate default = cost of insuring against
CVA and CCDS
Hedging CVA
default via CDS on counterparty’s default.
CDS Proxies • CDS notional = net current exposure.
Portfolio
counterparty risk • CDS maturity = some measure of life of deal (maturity or
CVA & Basel duration).
CVA VaR
• CVA = Cost of insurance = cost of fixed leg of CDS contract ≈
Accounting
considerations the cost of the CDS spread applied over the life of the CDS
CVA issues contract.
Summary
Appendix —
CDS Spreads
References
63 / 79
Counterparty
Risk, CVA, and
Basel III

Markets and
risks
Counterparty The appeal of the net current exposure method lies largely in its ease
risk
Risk
of implementation.
Modifications
Counterparty • If you can value the positions and get the CDS spreads, you can
Valuation
Adjustments easily compute the CVA.
CVA and CCDS • Easily extended to portfolios and to take netting agreements into
Hedging CVA
account.
CDS Proxies
Portfolio
• Easily extended to take collateralization into account — reduce
counterparty risk
the current net exposure to the threshold level (the level beyond
CVA & Basel
which the position must be collateralized).
CVA VaR
Accounting
• Easily extended into a type of bilateral CVA — do as above if
considerations
position is positive, and do the reverse (impact on counterparty
CVA issues
of one’s own default) if negative.
Summary
Appendix —
CDS Spreads
References
64 / 79
Counterparty
Risk, CVA, and
Basel III

Markets and
The negative of this approach comes from its inaccuracy. Consider
risks the CDS position needed to insure the portfolio to maturity.
Counterparty
risk • Forward value of the portfolio is not constant.
Risk
Modifications • Improve hedge by using a portfolio of CDS so that the forward
Counterparty market values of the portfolio are matched.
Valuation
Adjustments • This CDS portfolio hedges against default risk assuming zero
CVA and CCDS
interest rate risk.
Hedging CVA
• Current net exposure method is a rough approximation of this
CDS Proxies
Portfolio
with just one CDS contract.
counterparty risk
CVA & Basel

As such, there are a number of disadvantages:
CVA VaR • Zero volatility approach, so similar problems to the discount
Accounting
considerations curve shift approach — neglects value from interest rate
CVA issues volatility, which can be substantial.
Summary • Even less accurate, in that only one CDS contract is used.
Appendix —
CDS Spreads • Method tends to be unstable — value is proportional to market
References value, which has much higher volatility than the true CVA.
65 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Bilateral CVA

Markets and
risks
In accounting circles, one often finds support for bilateral CVA
Counterparty
calculations.
risk
Risk • Unilateral CVA — value of contract taking into account default

Modifications
of counterparty. What we’ve discussed up until now.
Counterparty
Valuation
Adjustments
• Bilateral CVA — value of contract taking into account both
CVA and CCDS default of counterparty and default of investor.
Hedging CVA
CDS Proxies
Bilateral CVA is a complicated calculation.
Portfolio
counterparty risk
• Need to know relationship between default of counterparty and
CVA & Basel default of investor.
CVA VaR • Often approximated as difference in unilateral CVA of investor
Accounting
considerations and counterparty.
CVA issues • Approximation is true bilateral CVA assuming the probability of
Summary
both parties defaulting before contract maturity is zero.
Appendix —
CDS Spreads • Can lead to significant error with high default correlation.
References
66 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Bilateral CVA

Markets and
Bilateral CVA is often looked upon favorably.
risks
• Reduces CVA charge.
Counterparty
risk • Liabilities behave more like bond liabilities - a drop in credit of
Risk the investor will potentially improve the balance sheet.
Modifications
Counterparty
It also has some drawbacks.
Valuation
Adjustments • If investor’s credit is worse than counterparties, bilateral CVA
CVA and CCDS increases value of the derivative above the risk free value:
Hedging CVA • 2009 — Citigroup, $2.5 billion.
CDS Proxies • 2011, Q3 — Goldman, $450 million, J.P. Morgan Chase &
Portfolio Citigroup, $1.9 billion each, and BofA, $1.7 billion.
counterparty risk
• All derivatives (even assets) increase in value when credit rating
CVA & Basel
CVA VaR
drops.
Accounting • Prices derivative without an associated hedge.
considerations
CVA issues
Because of these issues, accounting boards have been lobbied to
Summary
reject bilateral CVA as an acceptable approach. We can hope that
Appendix — these issues will be addressed as IASB, FASB and other accounting
CDS Spreads
standards boards work together on global convergence of accounting
References
standards.
67 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Issues in CVA computations

Markets and
CVA computations are difficult for a number of reasons.
risks
• Options on a portfolio are difficult to price and hedge
Counterparty
risk
• Weak models
Risk • Lack of securities that effectively hedge risks
Modifications
• Data is difficult to come by and manage
Counterparty
Valuation • CDS spreads
Adjustments
• CSA legalese
CVA and CCDS
• Extra structure within and across portfolios
Hedging CVA
• CVA calculations are 6 orders of magnitude harder than option
CDS Proxies
pricing
Portfolio
counterparty risk • Level 1 — Pricing
CVA & Basel • Level 2 — Calibration
CVA VaR • Level 3 — Sensitivities
Accounting • Level 4 — Trading strategy analysis
considerations
• Level 5 — Portfolio calculations of the above
CVA issues
• Level 6 — VaR
Summary
• Level 7 — CVA
Appendix —
CDS Spreads • Level 8 — CVA VaR
References
This stresses the interfaces between quants, risk management and IT.
68 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Summary

Markets and • It’s important to factor credit risk into valuations.
risks
• Netting agreements and collateralization must be accounted for.
Counterparty
risk • The prices of risky bonds can be calculated based on CDS
Risk
Modifications
spreads.
Counterparty • Bond calculations roughly correspond to just shifting the
Valuation
Adjustments discount curve by an appropriate amount.
CVA and CCDS • Swap credit risk is more complicated because swaps can be
Hedging CVA assets or liabilities, depending on rates.
CDS Proxies
• For swaps, the discount curve shift is sometimes used, but it is
Portfolio
counterparty risk not very accurate.
CVA & Basel • The net current exposure method is used as well, but has similar
CVA VaR
shortcomings.
Accounting
considerations • Swap credit risk can be calculated using swaptions and CDS
CVA issues rates.
Summary • The CVA is the value of the default leg in a CCDS on the
Appendix —
CDS Spreads underlying swap.
References • Portfolio CVA calculations are not so easy...
69 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Appendix — CDS Spreads

Markets and
risks
Counterparty
risk
Risk A CDS swap consists of two legs:

Modifications
Counterparty • A fixed leg which pays a constant spread until default or maturity.
Valuation
Adjustments
• A floating leg which receives (1 − R) on the notional of a
CVA and CCDS
reference bond (i.e. — makes whole on the principal).
Hedging CVA
CDS Proxies R is determined by bankruptcy proceedings, and the fixed leg also
Portfolio
counterparty risk
pays accrued interest at default time (the portion of the next coupon
CVA & Basel that is due).
CVA VaR
The par CDS spread is the spread that makes the prices of the two
Accounting
considerations spreads match.
CVA issues
Summary
Appendix —
CDS Spreads
References
70 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Default probabilities

Markets and
risks
Modeling the price of a CDS contract
Counterparty
risk
• Interest rates and default are independent.
Risk • The recovery rate R is currently known.
Modifications
Counterparty
Valuation Then the par swap spread C (tn ) for a CDS with payment times ti and
Adjustments
maturing at tn satisfies the following relationship:
CVA and CCDS
Hedging CVA X
n Z
CDS Proxies C (tn )α(ti )D(ti )S(ti ) + C (tn )α′ (s)D(s)P(s)ds
Portfolio Z
counterparty risk
CVA & Basel

= (1 − R)D(s)P(s)ds
CVA VaR
Accounting where D(t) is the risk free discount factor for time t, P(t) is the
considerations Rt
CVA issues default probability density function for time t, S(t) = 1 − P(s)ds is
Summary the survival probability for time t, α(ti ) is the coverage for the ith
Appendix — coupon (e.g. roughly 1/2 for semiannual payments), and α′ (t) is the
CDS Spreads
References
accrued interest factor at time t.
71 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Interpretation of terms

Markets and
The par swap spread C (tn ) satisfies:
risks
X
n Z
Counterparty
risk
C (tn )α(ti )D(ti )S(ti ) + C (tn )α′ (s)D(s)P(s)ds
Risk
Z
Modifications
= (1 − R)D(s)P(s)ds
Counterparty
Valuation
Adjustments The summation is the value of the fixed payments, not counting the
CVA and CCDS
accrued interest:
Hedging CVA
• C (tn )α(ti ) — the coupon paid at time ti .
CDS Proxies
• C (tn )α(ti )D(ti ) — its discounted value.
Portfolio
counterparty risk • S(ti ) — probability of getting this coupon.
CVA & Basel • C (tn )α(ti )D(ti )S(ti ) — value of this coupon (by virtue of
CVA VaR
independence assumption).
Accounting
considerations The second term is the value of the accrued interest.
CVA issues • C (tn )α′ (s) — accrued interest that would be paid at time s.
Summary • C (tn )α′ (s)D(s) — its present value.
Appendix —
CDS Spreads
• P(s) — the probability of paying this amount.
References Similarly, the third term is the value of the disbursement made at
72 / 79
default time.
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Solving for the coupon

Markets and
risks
Counterparty
risk
Recall, the par swap spread C (tn ) satisfies:
Risk
Modifications
X
n Z
Counterparty
Valuation C (tn )α(ti )D(ti )S(ti ) + C (tn )α′ (s)D(s)P(s)ds
Adjustments
Z
CVA and CCDS
= (1 − R)D(s)P(s)ds
Hedging CVA
CDS Proxies
Portfolio
counterparty risk Solving for C (tn ), we see that the par spread is given by:
CVA & Basel R
(1 − R)D(s)P(s)ds
CVA VaR
C (tn ) = Pn R
Accounting
considerations
α(ti )D(ti )S(ti ) + α′ (s)D(s)P(s)ds
CVA issues
Summary
Appendix —
CDS Spreads
References
73 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Discrete and Continuous

Markets and
risks
Counterparty
risk If we knew the CDS spread for all times t, then we could use the
Risk above relationship to convert it to a default probability curve.
Modifications
Counterparty
Similarly, we can convert probability default curves back into CDS
Valuation
Adjustments
spread curves.
CVA and CCDS However, CDS spreads are only quoted for a handful of discrete times.
Hedging CVA
CDS Proxies
Like in curve stripping, to compute implied default probabilities, we
Portfolio
need to make an assumption about either the default probability curve
counterparty risk or the CDS spreads.
CVA & Basel
CVA VaR
The common assumption:
Accounting
considerations • Default probabilities are derived from piecewise constant hazard
CVA issues rates.
Summary
Appendix —
CDS Spreads
References
74 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Piecewise constant hazard rates

Markets and
risks
Counterparty If par CDS spreads Ci are quoted for maturities Ti , we let λi be

risk
discrete hazard rates, and define λ(t) = λi for Ti−1 < t ≤ Ti .
Risk
Modifications
Then
Counterparty
Valuation Rt
Adjustments S(t) = e − 0 λ(s)ds
CVA and CCDS
P(t) = −dS/dt = S(t)λ(t)
Hedging CVA
CDS Proxies
Portfolio
For 0 < t ≤ T1 , this yields:
counterparty risk
CVA & Basel S(t) = e −λ1 t

CVA VaR P(t) = e −λ1 t λ1
Accounting
considerations
CVA issues Using this we can solve for λ1 given C1 , and repeat for each λ to
Summary construct the default probability curve.
Appendix —
CDS Spreads
References
75 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein Pricing CDS swaps

Markets and
risks
Counterparty
risk Once a default probability curve is constructed as above, it can then
Risk be used to compute the price of a CDS contract with a non-par
Modifications
spread.
Counterparty
Valuation
Adjustments The price of a receive fixed contract with spread C is then:
CVA and CCDS
Hedging CVA
CDS Proxies X
n
C α(ti )D(ti )S(ti )
Portfolio
counterparty risk Z
CVA & Basel + C α′ (s)D(s)P(s)ds
CVA VaR Z
Accounting
considerations
− (1 − R)D(s)P(s)ds
CVA issues
Summary
Appendix —
CDS Spreads
References
76 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein CDS on the CCP

Markets and
To facilitate trading of CDS through a central clearing counterparty
risks (CCCP, or more commonly CCP), CDS conventions have changed.
Counterparty
risk
On the CCPs,
Risk • Quarterly maturities are traded.
Modifications
Counterparty
• Particular spreads are traded, depending on geographic locale:
Valuation • Standard North American Contract (SNAC) — 100 and 500 bp,
Adjustments
CVA and CCDS

with 40% recovery.
Hedging CVA
• Standard European Contract (STEC) — 25, 100, 300, 500, 750
CDS Proxies
and 1000 bp, 40% recovery.
Portfolio
• Standard Emerging Market contract (STEM) — 100 and 500 bp,
counterparty risk with 25% recovery.
CVA & Basel
CVA VaR
The convention is to quote a par spread for investment grade, and
Accounting
pay the difference between the price of the traded leg and the price of
considerations
the par leg, where the prices are computed using the above model and
CVA issues
a flat curve equal to the quoted spread. This difference is the “points
Summary
up-front” payment.
Appendix —
CDS Spreads For distressed credit, instead of quoting the par spread, the points
References
up-front themselves are quoted.
77 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein References

Markets and
Altman, Kishore; Almost Everything You Wanted to Know About
risks Recoveries on Defaulted Bonds; Financial Analysts Journal, Nov/Dec
Counterparty
risk
1996
Risk Alavian, Shahram; Ding, Jie; Whitehead, Peter; Laudicina, Leonardo;
Modifications
Counterparty
Counterparty Valuation Adjustment (CVA); March 2009
Valuation
Adjustments Bank for International Settlements; Detailed Tables on Semiannual
CVA and CCDS OTC Derivatives Statistics at End-June 2009
Hedging CVA
Bank for International Settlements; Basel III: A global regulatory
CDS Proxies
framework for more resilient banks and banking systems - revised
Portfolio
counterparty risk version June 2011; http://www.bis.org/publ/bcbs189.htm
CVA & Basel
Brigo, Damiano; Masetti, Massimo; A Formula For Interest Rate
CVA VaR
Swap Valuation under Counterparty Risk in presence of Netting
Accounting
considerations Agreements; ssrn.com; April 2005
CVA issues
Gregory, Jon; Being two-faced over counterparty credit risk, Risk,
Summary
February 2009
Appendix —
CDS Spreads
Cooper, Ian A.; Mello, Antonio S.; The Default Risk of Swaps;
References
Journal of Finance, Vol. 46, No. 2, June 1991
78 / 79
Counterparty
Risk, CVA, and
Basel III
Harvey Stein References

Markets and Jarrow, Robert A.; Yu, Fan; Counterparty Risk and the Pricing of
risks
Defaultable Securities; Journal of Finance, Vol. LVI, No. 6, Oct. 2001
Counterparty
risk
Pykhtin, Michael; Zhu, Steven; A Guide to Modelling Counterparty
Risk
Modifications Credit Risk; Global Association of Risk Professionals, Issue 37,
Counterparty July/August 2007
Valuation
Adjustments
Statement of Financial Accounting Standards No. 157, Fair Value
CVA and CCDS
Measurements
Hedging CVA
CDS Proxies Stein, Harvey J.; Valuation of Exotic Interest Rate Derivatives -
Portfolio Bermudans and Range Accruals, December 7, 2007,
counterparty risk
CVA & Basel

http://ssrn.com/abstract=1068985
CVA VaR Stein, Harvey J.; Lee, Kin Pong; Counterparty Valuation Adjustments,
Accounting
considerations
Credit Risk Frontiers: Subprime Crisis, Pricing and Hedging, CVA,
CVA issues
MBS, Ratings and Liquidity, Ed. Tomasz Bielecki, Damiano Brigo,
Summary Frédéric Patras, 2011 (to appear)
Appendix —
CDS Spreads
Topic 820, Fair Value Measurements And Disclosures, Financial
References Accounting Standards Board
79 / 79
Counterparty Risk and CVA
Stephen M Schaefer
London Business School
Credit Risk Elective

Summer 2012
Net revenue included a $1.9 billion gain from debit valuation

adjustments (“DVA”) on certain structured and derivative
liabilities, resulting from the widening of the Firm’s credit
spreads. This was partially offset by a $691 million net loss,
including hedges, from credit valuation adjustments (“CVA”)
on derivative assets within Credit Portfolio, due to the
widening of credit spreads for the Firm’s counterparties.
JP Morgan, 3rd Quarter Results 2001
Counterparty Risk and CVA 2

Counterparty Risk and the Growth of OTC Derivatives
• The OTC derivatives market has grown enormously in terms of
notional amounts outstanding
creates substantial exposure to default of counterparties
• Counterparty risk increased
significantly due to practice of 500
Total IR and currency

“offsetting” rather than 450
Credit default swaps
400
“unwinding” derivative trades 350
• Since crisis both regulators 300
$US Trillion
and banks currently paying 250
much more attention to 200
150
assessing counterparty 100
risk 50
valuing counterparty -
1985 1990 1995 2000 2005 2010
exposure (CVA)
The Basic Idea: Exposure as replacement cost

• Suppose bank A has entered an OTC contract (e.g., a fixed-floating
interest rate swap – not necessarily a credit derivative) with a bank B
• Value of swap:
Initiation Zero
In future (before maturity) Positive or negative
At maturity Zero
• If bank B defaults during life of swap:

if PV to bank A negative: bank A pays PV to creditors of
bank B (assuming bank A is solvent)
if PV to bank A positive: bank A has claim on bank B and
loses (1-R) x market replacement cost of contract
Exposure measured in terms of replacement cost

Exposure – Definitions
• Counterparty exposure:
the larger of (i) zero and (ii) the market value of the
portfolio of derivatives that would be lost if the
counterparty were to default = (1-R) x replacement cost
• Current exposure (CE)
current value of the counterparty exposure
• Potential future exposure (PFE)
highest level of exposure at a given future date expected
with a particular (typically high) degree of statistical
confidence (e.g., 95%)
analogous to VaR
Exposure – Definitions, contd.
• Maximum potential future exposure (MPFE)

the maximum (peak) value of PFE over life of portfolio
• Expected exposure (EE)
the average value of the counterparty exposure on a given future date.
• Expected Loss
expected exposure x (1 – R)
• Right-way / wrong-way exposure
relation between level of exposure and credit quality of counterparty
inverse: wrong way
positive: right way

Methods of Mitigating Counterparty Risk
• Netting agreements
• Collateral agreements
“safe harbour” rule for derivatives
• Clearing houses
• Diversification
• Early termination agreements
Counterparty risk exposure

Example: fixed/floating interest rate swap

Example – A Single Fixed Floating Interest rate Swap
• Value of swap contract (e.g., receiving fixed) at:

inception: zero (typically)
future time t (after inception):
( S0 − St ,T ) × A(t , T )
where;
- S0 is the contract swap rate;
- St,T is the market swap rate at time t for a swap that matures
at T; and
- A(t,T) is the market value of an annuity at time t that matures
at T.
• Value of swap exposed to variation in swap rate

Swap example: Simulated Paths for Future Swap

Rates
14%
Current Swap Rate 5%
12%
Maturity 10
Ann Vol of swap rate 1.00% 10%
Recovery Rate 40%
Swap Rate
8%
Nominal amount 100
6%
4%
2%
0%
0 1 2 3 4 5 6 7 8 9 10
Time
• Method for simulating future interest rates comes from term

structure modelling
should be consistent with current term structure

Value of Swap Contract Simulated Paths
for Contract
Swap value
Value and Future
A
25
20
Swap Rates
15
10
Swap value
0 14%
-5 12%
-10 10%
Swap Rates
Swap Rate
-15 8%
-20 6%
0 1 2 3 4 5 6 7 8 9 10
4%
Time
• If future swap rate is lower than contract 2%
0%
rate (e.g., heavy blue line – “A”), value of 0 1 2 3 4 5 6 7 8 9 10
contract to party receiving fixed is positive Time
A
Counterparty Exposure
Simulated Paths
25
for Exposure and
20
A Contract Value
Counterparty exposure
15
10
5
Contract Value
A
25
0
20
0 1 2 3 4 5 6 7 8 9 10
15
Time 10
Swap value
• Counterparty Exposure is -5
-10
maximum of contract value and -15
zero -20
0 1 2 3 4 5 6 7 8 9 10
Time

Expected Exposure and Potential Future Exposure (95%)
20
Expected Exposure
PFE
15
10
0
0 2 4 6 8 10 12
Valuing Counterparty Exposure (CVA)

Valuation of Counterparty Credit Exposure:
Credit (or Counterparty) Value Adjustment (CVA)
• Consider a portfolio of contracts under which Bank B (the

counterparty) makes payments to Bank A
the value to Bank A of these contracts is reduced by the
possibility that Bank B may default
• The CVA is the difference between the value of the portfolio

when there is :
a. no possibility of default by the counterparty (Bank B); and
b. a positive probability of a default by the counterparty
• We compute this (as in our analysis of CDS) by computing

the discounted risk-neutral expected value of the expected
loss.
Calculation of Credit value Adjustment (CVA)

• For each future period Expected Exposure
5.0
expected loss is EEt
(expected exposure) x (1- 4.0
Exepcted exposure
R). 3.0
• Value of counterparty 2.0

losses is then
1.0
T
(1 − R)∑ EEt ( Qt −1 − Qt ) Dt 0.0
0 2 4 6 8 10
t =1 Tim e
• Where Qt is the risk-neutral survival probability of the

counterparty to time t.

Risk-Neutral Default Probability
• Suppose:
CDS curve for counterparty is flat and equal to 98.36 bps
Recovery rate is 40%
• Then one-year RN conditional default probability is

constant and equal to 1.64% (and independent of the zero
curve!)
Expected Exposure D(t) Q(t) CVA Calc

0 0.0000 1.0000 1.0000
0.5 2.0556 0.9759 0.9918 0.0098
1 2.3102 0.9524 0.9837 0.0107
1.5 3.6125 0.9294 0.9757 0.0162
2 3.8694 0.9070 0.9677 0.0168
Example, contd.
2.5 4.8177 0.8852 0.9598 0.0202

3 5.3980 0.8638 0.9520 0.0219
3.5 4.5181 0.8430 0.9442 0.0178
4 4.7059 0.8227 0.9365 0.0179
4.5 4.6863 0.8029 0.9289 0.0173
5 5.1288 0.7835 0.9213 0.0183
5.5 3.9424 0.7646 0.9138 0.0136
6 2.6491 0.7462 0.9063 0.0088
6.5 2.2238 0.7282 0.8989 0.0072
7 2.0535 0.7107 0.8916 0.0064
7.5 2.2375 0.6936 0.8843 0.0068
8 2.1978 0.6768 0.8771 0.0064
8.5 1.4405 0.6605 0.8699 0.0041
9 0.9937 0.6446 0.8628 0.0027
9.5 0.5724 0.6291 0.8558 0.0015
10 0.0000 0.6139 0.8488 0.0000
Total 0.2244

CVA
• CVA in this case is 0.2244 (per 100 nominal) or 22.4 bps
• This assumes no collateralisation and no netting.
• Cost of counterparty default will affect price at which

bank should be prepared to enter deal.
• In fact, counterparty risk in swaps is two sided and so,
while counterparty default risk reduces portfolio value,
own default risk increases it.
Mitigating Counterparty Risk

Counterparty Risk Mitigation I:
Aggregation of Counterparty Risk and Netting
• For counterparty risk, we first have to aggregate at the
level of a given counterparty (e.g., all our exposures to
Bank of America, Deutsche Bank etc.)
• In doing this we aggregate only over positions with
positive exposure and so do NOT benefit from hedging
between different positions with that counterparty unless
we have a netting agreement in place (usually, there will
be)
Contracts with Counterparty X
• Netting can substantially Exposure
reduce counterparty exposure Contract Value No Netting With Netting
a 5 5 5
• Cross-product netting highly b -3 0 -3
desirable c 6 6 6
d -5 0 -5
• Potential problem with legal e 3 3 3
entities 14 6
Aggregation and Comparison with VaR

• Analysis of uncertainty in future contract value is similar
to analysis of value-at-risk (VaR)
• For VaR bank needs to aggregate across all positions in a
business unit at various levels of aggregation, such as:
a trading desk (e.g., a particular corporate bond desk, a
particular equity derivatives desk etc.)
a geographical region/business area (e.g., European Fixed
Income)
and .. the bank as whole (global basis)
• In VaR we aggregate all positions and benefit from
positions with offsetting exposures (hedging); and
diversification

Counterparty Risk Mitigation II: Collateral
• A second contractual mechanism that reduces counterparty
risk is the use of collateral.
• Banks will increasingly use a legally binding margin
agreement that requires one or both parties to post collateral
when the uncollateralised exposure exceeds a given threshold.
• In addition to the threshold level the agreement will also
specify the call period, i.e., the frequency at which collateral is
monitored and called for.
• The use of collateral to mitigate counterparty risk is subject to
operational risk.
• Derivatives benefit from a “safe harbour” provision that
means that, in the event of default by one counterparty, the
other is able to realise collateral without having to go through
the bankruptcy process.
Risk Mitigation III: Diversification

• Having aggregated all exposures to a given counterparty
(not only derivative exposures but, for example, bonds
issued by the counterparty held in inventory, equity in the
counterparty etc.) counterparty risk is reduced by
diversification across counterparties.
• One simple mechanism employed by banks to achieve this
is to impose a maximum exposure to each counterparty
(equivalent to a credit line).

Risk Mitigation IV: Early Termination
• Optional Early Termination (OET) agreements
give each party the right to seek termination (at the current
market price and without giving a reason).
used when, e.g., an interest rate swap has 10-years to
maturity but counterparty has credit line for only 5 years
mitigates risk in the event that counterparty credit quality
deteriorates.
• Mandatory Early Termination (MET) agreements
example: 20-year interest rate swap with mandatory
termination in 5 years (at market prices)
why do banks do this? Because banks want the exposure to
the 20-year rate but do not want 20-year counterparty
exposure
used less than OETs.
Right-way / wrong way exposure

• Calculation of CVA on previous slide assumes no
correlation between counterparty risk exposure and credit
quality of counterparty. .
• But actual correlation between risk-neutral probability of
default (RN-p) and exposure may be non-zero:
right-way exposure: if RN-p low when exposure is high then
this reduces value (cost) of counterparty credit risk
wrong-way exposure: if RN-p high when exposure is high
then this increases value (cost) of counterparty credit risk
• Examples:
right way exposure: holding forward contract to buy oil from
BP at fixed price: low oil price means BP default risk high
when and contract value low (negative)
wrong way exposure: holding a put option on Lehman
written by …….. Lehman!!
Another example of right way / wrong way risk – Variance
Swaps
Cumulative returns -- VIX - Realised Var - Fixed Qauntity of exposure
350 • Variance swap is contract
300
File: Realised Vol & VIX
where one side pays fixed
and the other pays realised
250
variance of returns (e.g.,
200
on S&P)
150 • On average high risk
100
premium (implied vol. >
50
realised vol. on average)
but in crisis realised vol.
-
n-9
0
n-9
2
n-9
4
n-9
6
n-9
8
n-0
0
n-0
2
n-0
4
n-0
6
n-0
8
n-1
0 was huge.
Ja Ja Ja Ja Ja Ja Ja Ja Ja Ja Ja
• SELLER of variance faces right-way risk .. will pay out in

crisis and receive in normal times
Summary
• Size of notional outstanding in derivatives market means that
management of counterparty risk is a major issue for banks and
many other types of financial institution.
• Key concepts:
exposure
expected exposure (EE)
potential future exposure (PFE)
• Methods of managing counterparty risk
netting
collateral
diversification
early termination
• Valuation of counterparty exposure: credit value adjustment (CVA)

Counterparty Risk
Gabor Fath
Morgan Stanley
Counterparty credit risk - definition
Counterparty credit risk (CP risk) is the risk

that the counterparty to a financial contract will
default prior to the expiration of the contract
and will not make all the payments required
by the contract.
Counterparty credit risk
What to investigate?
• Exposure
The potential/expected loss in case when
CP defaults
• CP default probability
From bond or CDS prices
• Pricing counterparty risk:
The value adjusment on top of the fair
(non-defaultable CP) market value of a
transaction reflecting the „price” of CP risk
Mark-to-market (M2M)
Banks’ trading book valuation is based on fair value

accounting. Positions are „marked to market” (by models
calibrated to the market) daily.
• Trades are „fair” at inception, PV(t=0)=0

• As time passes the contract is „marked to market”
and its M2M value become non-zero PV(t)≠0
• At time zero PV(t) is a random variable, which has a
distribution
• At time t the trade can be
in-the-money or out-of-the-money
Loss in case of CP default
Assumptions:
• Counterparty defaults
• Bank closes out its positions with CP
• Bank enters into a similar contract with another CP in
order to maintain its market position
Since the bank’s market position is unchanged

after replacing the contract, the loss is determined by the
contract’s replacement cost at the time of default.
If the contract M2M value V(t) is negative

for the bank at the time t of default, the bank
• closes out the position by paying the defaulting CP the M2M

V(t) of the contract
• enters into a similar contract with another CP and receives
M2M V(t) for contract
• bank has neither loss nor gain.
If the contract M2M value V(t) is positive

for the bank at the time t of default, the bank
• closes out the position, but only receives recovery RCV(t)

from the defaulting CP
• enters into a similar contract with another CP and pays V(t)
for the contract
• bank has a net loss equal to (1-RC )V(t).
In which case does the Bank have counterparty risk?
Position CP risk Why?

long option
short option
forward
swap

long option yes always an asset, PV(t)>0
short option
forward
swap

short option no always a liability, PV(t)<0
forward
swap

forward yes asset or liability, PV(t)>0 or <0
swap

forward yes asset or liability, PV(t)>0 or <0
swap yes asset or liability, PV(t)>0 or <0

Exposure
• Contract level exposure
• Counterparty level exposure

Contract level exposure
Ei (t )  max[Vi (t ),0]
Bank’s exposure in Forward PV

contract i at time t Value of contract i at time t
Only cashflows in (t,T) should be counted
• Current exposure at t = 0 is certain

• Future exposure at t > 0 is a random variable
PV and Forward PV
deterministic value
PV = V(0) =
Value of contract at present (t=0)
V(0)
• Model provides
probability distribution
• V(0) calculated by
averaging cashflows [0,T] over paths
random variable
Forward PV = V(t) =
Value of contract at future time t
V(t)
• Model provides:
probability distribution for [0,t]
prob distribution [t,T] given path [0,t]
• V(t) calculated by
averaging cashflows [t,T] over paths [t,T]
has a different value for each path [0,t]
t T
pdf of V(t)
0
Forward PV distribution
t
Vi (t )
out of in the
the money money
Exposure distribution
Ei (t )  max[Vi (t ),0]
pdf of V(t)
t
Mitigating CP risk (1) - netting
Portfolio level exposure

without netting E (t )   Ei (t )   max[Vi (t ),0]
i i
with netting E (t )  max[ Vi (t ),0]

i
A netting agreement is a contract between two counterparties

that allows aggregation of transactions between the
counterparties – i.e., transactions with negative values can be
used to offset ones with positive values and only the net positive
value represents credit exposure at the time of default.
Netting plays a crutial role in reducing CP exposure.
Mean positive/negative exposure, potential future exposure
Mean positive exposure (discounted to today):
MPE(t )  ΕD(t )E(t )  ΕD(t )max[V (t ),0]
Usually the discount factor

is part of the definition
Mean negative exposure:
MNE(t )  ΕD(t ) min[V (t ),0]
Potential future exposure (at 95%):
PFE(t )  X : Pr(D(t ) E(t )  X )  0.95

Exposure profile
MPE(t)
t
T
diffusive phase amortization phase
• long remaining maturity • short remaining maturity

• many remaining CFs • few remaining CFs
• sensitivity to market • little sensitivity to market
uncertainty uncertainty
Exposure profile - example
IR swap
Mitigating CP risk (2) – margin agreements
A margin agreement is a contract that requires one or both

counterparties to post collateral when the uncollateralized
exposure exceeds a threshold.
Residual CP risk remains („gap risk”) because:

• Threshold is non-zero
• Finite time („margin period of risk” ~2weeks) needed to
1) initiate margin call,
2) CP to pose additional collateral
3) liquidate position and rehedge
• Market can jump (crash) within this time
Mitigating CP risk (3) – downgrade triggers
A downgrade trigger is a close stating that if the credit rating of the

CP falls below a certain level, than the Bank has the option to close
out the derivatives contract at M2M.
Residual CP risk remains because:

• The CP may have downgrade triggers with many other banks thus
its downgrade may trigger simulataneous sell-offs („crowded market”)
leading to falling market (gap risk)
• Downgrading depends on rating agencies
• Client relationship issues
Enron defaulted in Dec 2001. It was not downgraded until the very
last moment, becase downgrade triggers at BBB- were defined in
many of its derivatives contracts. Premature downgrading would have
killed it immediately, even if had had a chance to survive otherwise.
Mitigating CP risk (4) – central counterparties
Central counterparties (CCP) would provide broader and more

effective netting as well as daily marking-to-market and margining.
Banks would face the CCP as opposed to each other
and the CCP would be sufficiently capitalized and would impose
sufficiently strict margin requirements to eliminate counterparty
risks to a large extent. CCPs will also lead to more standardization of
derivatives over time.
E.g., CCPs for CDS:

ICE Trust, LCH.Clearnet, Eurex, …
Valuating counterparty credit risk
Credit valuation adjustment (CVA) is the difference

between the risk-free portfolio value and the true
portfolio value that takes into account the
possibility of the counterparty’s default.
CVA is the market value of counterparty credit risk.

CVA valuation
If we knew the time of default, the discounted loss would be
L  (1  RC ) D (t ) E (t )
loss
random exposure
variable
recovery random
discount factor variable
CVA valuation
Unilateral CVA is the risk-neutral expectation of the discounted loss
T
uCVA  EQ L  EQ  (1  RC ) D(t ) E (t ) pC (t )dt
0
default time pdf

of counterparty C
risk neutral
Can be computed from the
expectation
CDS term structure of C
CDS - Kinetics
Long credit risk
Buyer of credit risk
Premium leg
Protection Protection
buyer A seller B
Default leg
Short credit risk

Seller of credit risk
Reference
obligation C
N – notional [$]
s – spread [bps/year]
R – recovery [1]
T – maturity [year]
CDS - Kinetics
Long credit risk
Buyer of credit risk
Premium leg
buyer A seller B
Default leg
Short credit risk

Seller of credit risk
Path 1: no default of C
Reference
obligation C t=0 T
sN
N – notional [$] (1-R)N

Path 2: C defaults
s – spread [bps/year]
R – recovery [1] t=0 T
T – maturity [year]
sN t
CDS – Constant hazard rate (toy) model
Premium leg
Protection Protection - Default intensity h=const
buyer A seller B - Continuous premium payment
Default leg
- Fixed recovery
Default probability - Const interest rate
P(t  t  t  t )  e  ht  ht
defaults in interval
survives up to t
Default leg
T
DefPV  N  e rt  (1  R)e ht hdt
0
Premium leg expected default payment at t

T
PremPV  N  e rt  seht dt
0
expected premium payment at t
T
PremPV1  N  e rt  e ht dt
0
CDS – a more general model
Premium leg
Protection Protection - Default intensity h=const
buyer A seller B - Continuous premium payment
Default leg
- Fixed recovery
Default probability pdf - Const interest rate
P(t  t  t  t )  p(t )t
defaults in interval
survives up to t
Default leg
T
DefPV  N (1  R)  e rt  p(t )dt
0
Premium leg expected default payment at t

T
PremPV  Ns  e rt  (1  P(t ))dt
0
expected premium payment at t
T
PremPV1  N  e rt  (1  P(t ))dt t
default
0 P(t  t )  P(t)   p(t ' )dt'
probability
0
CDS – Fair spread
sfair
Premium leg
buyer A seller B
Default leg
The fair spread sfair makes the contract worth 0
DefaultPV  s fair  PremPV1
DefaultPV
s fair   (1  R)h
PremiumPV1
CDS - Term structure
CDS fair spread of a given company („name”)

sfair
maturity
CDS - Term structure
sfair
Simple model:
flat term structure
s fair  (1  R)h
Does not depend on maturity T

maturity
Market term structure of

CDS spreads on 26/2/2009
Default intensity is
time dependent
CDS - Calibration
hazard rate
maturity
T1 T2 T3 T4
Piece-wise constant hazard rate model can be calibrated

to market term structure by bootstrapping
{s1,s2 ,..., sT }  h1  h2  ...  hT

CDS - Calibration
Piece-wise constant hazard rate model can be calibrated

to market term structure by bootstrapping
{s1,s2 ,..., sT }  h1  h2  ...  hT
The survival probability follows:
P(t  t )  P(t)  e  h1T1 e  h2 (T2 T1 )  e  hN ( t TN 1 )

CVA valuation
Unilateral CVA is the risk-neutral expectation of the discounted loss
T
0
default time pdf

of counterparty C
risk neutral
Can be computed from the
expectation
CDS term structure of C
CVA valuation
T
0
In general, RC, D, E, and pC can correlate
Especially:
if corr(E, pC )>0 this is called wrong-way risk
if corr(E, pC )<0 this is called right-way risk

Wrong-way / right-way risk
Wrong-way risk
• Typical if V(t)>>0 („out-of-the-money risk”). CP has huge liability
thus high probability to default.
• Bank enters a swap with an oil producer (CP) where Bank
receives fixed and pays floating crude oil price
• Bank buys credit protection on an underlying reference entity
whose credit quality is positively correlated with that of CP
• Bank selling CDS (selling protection) on its own name
Right-way risk
• Bank enters a swap with an oil producer (CP) where Bank
pays fixed and receives the floating crude oil price
• Bank sells credit protection on an underlying reference entity
whose credit quality is positively correlated with that of CP
• Bank selling call options on its own stock
CVA – decoupling case
Calculation simplifies when correlations can be neglected
T
0
T
 (1  RC )  MPE(t ) pC (t )dt
0
N
 (1  RC ) MPE(ti ) PC [ti 1 , ti ]
i 1
Bilateral CVA
By now bilateral CVA (aka. CVA, simply) is the industry standard.

For the uncorrelated case:
T T
CVA  (1  RC )  MPE(t ) pC (t )dt  (1  RB )  MNE(t ) pB (t )dt
0 0
„unilateral CVA” DVA

or or
„asset CVA” „liability CVA”
CVA = Credit Valuation Adjustment

DVA = Debt Valuation Adjustment
Literature
• M. Pykhtin and S. Zhu, A Guide to Modelling Counterparty Credit

Risk, GARP Risk Review,July–August 2007
• E. Canabarro, Pricing and Hedging Counterparty Risk,

Counterparty Risk, Risk Books, 2010
Understanding the Consequences of the Use of Central Clearing on Counterparty Risk London, February 1, 2012
Understanding the Consequences of the

Use of Central Clearing on Counterparty
Risk
Marcus Zickwolff
Head of Trading & Clearing System Design, Eurex Group
London, February 01, 2012
www.eurexgroup.com
Agenda
• Objectives and aspirations for central clearing

• To what extent can central clearing actually meet objectives
• How many central clearers
• Coordinating central clearers – how shall central clearing counterparties cooperate
• How do we avoid ”CCP arbitrage”
• Counterparty credit risk mitigation
• Bilateral and multilateral netting
• Concentration risk considerations
• Systemic risk considerations
www.eurexgroup.com 2
Major regulatory initiatives: Market safety as core objective

Rule Set Scope Assessment
• Strengthen banking sector regulation, supervision and risk • Initiative has been adopted into EU
management regulation
Basel III • Improve shock absorption, risk management, governance and • To avoid regulatory arbitrage, all
Global
banks' transparency and disclosures regions to adopt principles
• Review of existing standards for financial market • Review and response for CCP and
CPSS - infrastructures/FMIs (CCPs, SSSs, CSDs and TRs) CSD arms completed
IOSCO • New and more demanding international standards for FMIs to be • World-wide harmonization & and
more robust & better placed to withstand financial shocks fine-tuning of principles required
• Reporting and clearing obligation for OTC derivatives, and • Regulation currently discussed in
defining measures reducing risks of bilaterally cleared OTC EU institutions; publication
EMIR derivatives expected end of year
• Common rule setting for CCPs and trade repositories
Europe
• The Capital Requirements Directive transposes Basel III • In Europe it is high priority to
CRD IV requirements into European Regulation pursue the Basel III standards
• Creating a robust common regulatory framework for Europe's • MiFID review ended in Q1/2011
securities markets • EU Commission proposals in
MiFID Q4/2011
• Leading to greater market transparency and efficiency, as well as
investor protection
• Promote financial stability of the US by improving accountability • Dodd-Frank Act partly in effect
and transparency in the financial system already
US
Dodd-Frank • Protect American taxpayer by ending bailouts, to protect

consumers from abusive financial services practices
Objective of future market structure
Trading on
Collateralized OTC OTC trading
Status Uncollateralized OTC trading organized
quo trading using a CCP
markets
Collateralized
Future (preferably third OTC trading
Registration Trading on organized markets
Market
Structure of trades party) using a CCP
OTC trading
Box size represents the notional amount outstanding
Agenda

Objectives and aspirations for central clearing
CCPs increase market safety and integrity by
• Mitigating and management of counterparty risk

• Mitigating liquidity & operational risks
• Addressing information asymmetries
• Reducing complexity and increasing efficiency
• Moving OTC traded derivatives from bilateral settlement to clearing via a CCP is the most effective way
of reducing the described systemic risk inherent in the global OTC segment
Most important function of a clearing house is to protect

and stabilize the markets and its participants
• Transfer of client position to stabilize markets, avoid adverse price movements and
Markets / protect exchange integrity
Regulators • Avoid knock-on effects on other market participants to reduce systemic risks
• Stable and transparent markets resulting in increased market efficiency
• Reduction of capital costs through passing through of client collateral

Clearing • Low operational impact – use of existing interfaces and processes (in particular
Members substitution of collateral)
• Minimal impact on the overall Clearing Fund in case of Clearing Member default
• Continued trading in the event of Clearing Member default

NCMs / • Transfer of positions to new Clearing Member to allow continuation of established
Clients hedge positions
• Avoid double funding when transferring positions where possible
• Protection of assets – achieve portability or at least preferential rights in an insolvency
situation
• Choice of protection level to allow risk/cost considerations
Agenda

Financial instruments universe to be cleared

Securities Derivatives
Users On-exchange OTC On-exchange OTC
Retail • Equities • Equity-linked • CFDs

• Bonds derivatives1)
• ETFs/ETCs/ETNs
• Certificates (e.g. index or bonus
certificates)
• Warrants
• Funds/UCITS
Wholesale/ • Equities • Fixed-income derivatives

professional
• Bonds • Equity-linked derivatives
• ETFs/ETCs/ETNs • Commodity derivatives
• Funds/UCITS
• Structured credit- • Foreign

linked securities exchange
(CDOs, MBS etc.) derivatives
• Other ABS • Credit derivatives
1) Only relevant in few regions, e.g. equity options in the US and the Netherlands, and equity index futures and options in Korea; negligible in most of Europe
Capital efficiency: Portfolio margining with Eurex

Clearing benchmark products – Eurex Clearing Prisma
Increased capital efficiency through next generation portfolio risk management
Initial Margin (IM) requirement

• Portfolio risk management
Diversification effect
provides optimal capital efficiency size is a function of
without taking additional risk onto several arguments:
• Risk method
the clearing house • Listed volumes
New - X%
• New products
related
• Risk netting between existing volumes
products
• Portfolio specifics
Benefits open interest in listed benchmark
products and for new OTC
products Listed
30 Bn.
EUR
• Existing risk management
infrastructure is leveraged to allow IM volumes IM
suitable for considering
low operational entry hurdles for diversification diversification
existing and new customers effects
Benefits are derived through the methodical cornerstones

of the next generation risk management model
Key benefits for customers
Capital • More accurate risk netting effects for listed, and between listed
Efficiency and OTC positions.
• Higher capital efficiency for customers through risk calculation
on portfolio basis.
1 Accuracy
• Cross-product scenarios enable consistent way to account for portfolio
2 correlation and diversification effects.
• Risk covered at 99% through margin calculation - consisting of a
− Mark to market component
− Forward looking component
− Adequate liquidation time horizon
Central benefits 3 Robustness
• Adjustments to enable stable margin requirements: Concentration,
uncertainty in correlations, model error adjustment if compression
is used.
• Flooring for stability: Stressed period scenarios and dynamic volatility
4 scaling to at least long run volatility percentiles.
5 Consistency • Aligned default management process between listed and OTC.
• Positions deemed suitable for simultaneous liquidation are
logically assigned to the same Liquidation Group.
• Risk offsets are only granted within the same Liquidation Group
(e.g. Equity, Commodity or Fixed Income).
Flexibility
• Reduced time-to-market for new products.
Agenda

CCPs cooperate via their trade associations to foster the

dialogue with regulators and industry
www.ccp12.org www.EACHORG.com
GCM model interlinks CCPs today and tomorrow
Legal relationship NCM - GCM Legal relationship GCM - CCP
NCM 1 CCP i
NCM 2 CCP ii
GCM
NCM n CCP m
• The GCM is responsible for the risk management of the NCMs

• CCPs provide the risk management for GCMs and ensure the transmission
of data necessary to enable GCMs to risk manage their NCMs
Agenda

CCPs are strongly regulated and supervised
Regulations Supervisors
• CPSS-IOSCO Principles for • International Monetary Fund
FMIs (IMF)
• European Market • European Securities and
Infrastructure Regulation Markets Authority (ESMA)
(EMIR) • National supervisors in EU
• MiFID / MiFIR • European System of Central
• Basel III with explicit rules Banks (ESCB)
about the capitalisation of • National Central Banks in EU
bank exposures to CCPs • 3rd country supervisors (e.g.
• CRD IV as the European SEC, CFTC, FINMA)
transmission of Basel III
• Dodd – Frank Consumer
Protection Act
Agenda

Eurex Clearing – Counterparty risk effectively

covered by collateralization and lines of defense
CCP risk management Lines of defence
8,772 1. Liquidation of open positions

(Defaulting clearing member)
Netting & 2. Liquidation of collateral
margining 3. Clearing fund contribution
4. Eurex reserves
5. Eurex Clearing AG
guarantee fund
6. Eurex Clearing AG
38 42 equity capital
Clearing Margin Collateral1

volume requirement
1 Cash and securities collateral after haircut, cash collateral ~ 9% of total
Source: Eurex Clearing

Agenda

Netting efficiency – Netting efficiency

multilateral (CCP) netting – bilateral netting
Exposure before netting = 100 Exposure before netting = 100
Multilateral (CCP) netting in an Bilateral netting in an OTC
on-exchange market1) market2)
100
– 85%
– 99%
15
<1
before netting after netting before netting after netting
1) Eurex Clearing example

2) Netting efficiency calculated as gross credit exposure/gross market value, including both netting and offsetting of netted positions under legally enforceable bilateral
netting agreements
Source: Eurex, BIS
Complexity reduction through CCP clearing

Bilateral market organization Multilateral market organization via CCP
CCP
10 market participants 10 market participants

90 counterparty relations – 89% 10 counterparty relations
Agenda

Eurex Clearing’s general rules for collateral

Admission and limitation criteria are aligned with regulatory requirements
• Collateral needs to fulfill quality, liquidity and accessibility criteria

Eligibility • Collateral is composed of highly liquid security collateral as well as of central
bank or commercial bank cash collateral (possibly in a mandatory portion)
• Own issues* are not accepted as collateral due to pro-cyclical effects (wrong-way
Exceptions risk). Enhancement to close link** securities prohibition for same reason
• Debt Securities that have a remaining term of 15 calendar days or less are not
accepted as collateral
• At Clearinghouse level:
Concentration – For bonds the admissible proportion of the issued capital is 25%
Limits – For equities the admissible proportion of free float is 5%
• At Clearing Member level:
– A maximum of 30% of the required collateral can be deposited in equities
* Securities issued by the collateral provider

** Securities issued by companies affiliated with the collateral provider
Eurex Clearing offers a variety of services to guarantee

optimal usage of provided collateral
Eurex Clearing -
Admissible Collateral • Eurex Clearing accepts a
broad range of securities as
collateral, approximately
25,000 ISINs
Bonds* Equities* Cash Other
• Overall collateral pool amounts
Government DAX ® EUR to 42bn EUR, 36bn of which is
in securities
CHF Xetra Gold
Corporate Euro STOXX 50®
Certificates • Online processing including
USD
State Agencies SMI® collateral substitution
GBP
Mortgage • Real time update of collateral

balances
Denominated in Denominated in Denominated in
EUR, USD, CHF, EUR and CHF EUR
JPY, GBP, DKK, • The full list of eligible collateral
NOK, SEK, AUD can be found under:
and CAD
http://www.eurexclearing.com/
* Securities are deposited to pledged securities accounts at Clearstream Banking risk/parameters_en.html
Frankfurt AG (CBF) or SegaInterSettle Zurich (SIS)
Agenda

Safety and integrity of the derivatives market need to be

ensured
Imperatives for a well-functioning Derived principles for a safer Overall
derivatives market derivatives market goal
• Capability of withstanding • Collateralization of all open risk positions

Safety shocks and/or potential • Risk taking of market participants limited to
domino effects within the individual risk tolerance
financial market • Continuous risk controlling/ management Reduction of
• Risks mitigation systemic
• Fair trading free from fraud • Transparency for effective supervision risk
Integrity or price manipulation • Manageable market structure complexity
• Secured price discovery
function
• Price efficiency • Information transparency
Efficiency • Cost efficiency • Process optimization and automation
• Product innovation • Flexible regulatory environment

Innovation • Process innovation • Competition
Source: Geithner 2008, Bernanke 2008, ECB 2008, Larosière 2009 etc.
Eurex Clearing: Safer due to reduced systemic risk

Eurex Clearing’s main concern is to provide a safer market, which is needed if
financial products are to deliver their full economic benefits.
• A safer derivatives market is transparent, to inspire trust
• It‘s efficient, so processes are simple and capital costs are lower
• It ensures that investors‘ positions are protected
• And above all it‘s neutral, so that counterparty risk is mitigated
Which is exactly what Eurex Clearing helps to provide.
Because the more people have faith in the markets, the more they‘ll feel clear to trade.
© Eurex 2012
Deutsche Börse AG (DBAG), Clearstream Banking AG (Clearstream), Eurex Frankfurt AG, Eurex Clearing AG (Eurex Clearing) as well as Eurex Bonds GmbH (Eurex Bonds) and
Eurex Repo GmbH (Eurex Repo) are corporate entities and are registered under German law. Eurex Zürich AG is a corporate entity and is registered under Swiss law. Clearstream
Banking S.A. is a corporate entity and is registered under Luxembourg law. U.S. Exchange Holdings, Inc. and International Securities Exchange Holdings, Inc. (ISE) are corporate
entities and are registered under U.S. American law. Eurex Frankfurt AG (Eurex) is the administrating and operating institution of Eurex Deutschland. Eurex Deutschland and Eurex
Zürich AG are in the following referred to as the “Eurex Exchanges”.
All intellectual property, proprietary and other rights and interests in this publication and the subject matter hereof (other than certain trademarks and service marks listed below) are
owned by DBAG and its affiliates and subsidiaries including, without limitation, all patent, registered design, copyright, trademark and service mark rights. While reasonable care has
been taken in the preparation of this publication to provide details that are accurate and not misleading at the time of publication DBAG, Clearstream, Eurex, Eurex Clearing, Eurex
Bonds, Eurex Repo as well as the Eurex Exchanges and their respective servants and agents (a) do not make any representations or warranties regarding the information contained
herein, whether express or implied, including without limitation any implied warranty of merchantability or fitness for a particular purpose or any warranty with respect to the accuracy,
correctness, quality, completeness or timeliness of such information, and (b) shall not be responsible or liable for any third party’s use of any information contained herein under any
circumstances, including, without limitation, in connection with actual trading or otherwise or for any errors or omissions contained in this publication.
This publication is published for information purposes only and shall not constitute investment advice respectively does not constitute an offer, solicitation or recommendation to
acquire or dispose of any investment or to engage in any other transaction. This publication is not intended for solicitation purposes but only for use as general information. All
descriptions, examples and calculations contained in this publication are for illustrative purposes only.
Eurex and Eurex Clearing offer services directly to members of the Eurex exchanges respectively to clearing members of Eurex Clearing. Those who desire to trade any products
available on the Eurex market or who desire to offer and sell any such products to others or who desire to possess a clearing license of Eurex Clearing in order to participate in the
clearing process provided by Eurex Clearing, should consider legal and regulatory requirements of those jurisdictions relevant to them, as well as the risks associated with such
products, before doing so.
Eurex derivatives (other than EURO STOXX 50® Index Futures contracts, EURO STOXX® Select Dividend 30 Index Futures contracts, STOXX Europe 50® Index Futures contracts,
STOXX® Europe 600 Index Futures contracts, STOXX® Europe Large/Mid/Small 200 Index Futures contracts, EURO STOXX® Banks Futures contracts, STOXX® Europe 600
Banks/Industrial Goods & Services/Insurance/Media/Personal & Household Goods/Travel & Leisure/Utilities Futures contracts, Dow Jones Global Titans 50 IndexSM Futures
contracts, DAX® Futures contracts, MDAX® Futures contracts, TecDAX® Futures contracts, SMIM® Futures contracts, SLI Swiss Leader Index® Futures contracts, Eurex
inflation/commodity/weather/property and interest rate derivatives) are currently not available for offer, sale or trading in the United States or by United States persons.
Trademarks and Service Marks
Buxl®, DAX®, DivDAX®, eb.rexx®, Eurex®, Eurex Bonds®, Eurex Repo®, Eurex Strategy WizardSM, Euro GC Pooling®, EXTF®, FDAX®, FWB®, GC Pooling®, GCPI®, MDAX®,
ODAX®, SDAX®, TecDAX®, USD GC Pooling®, VDAX®, VDAX-NEW®, Xetra® and XTF Exchange Traded Funds® are registered trademarks of DBAG.
Phelix Base® and Phelix Peak® are registered trademarks of European Energy Exchange AG (EEX).
The service marks MSCI Russia and MSCI Japan are the exclusive property of MSCI Barra.
iTraxx® is a registered trademark of International Index Company Limited (IIC) and has been licensed for the use by Eurex. IIC does not approve, endorse or recommend Eurex or
iTraxx® Europe 5-year Index Futures, iTraxx® Europe HiVol 5-year Index Futures and iTraxx® Europe Crossover 5-year Index Futures.
Eurex is solely responsible for the creation of the Eurex iTraxx® Credit Futures contracts, their trading and market surveillance. ISDA® neither sponsors nor endorses the product’s
use. ISDA® is a registered trademark of the International Swaps and Derivatives Association, Inc.
IPD UK Annual All Property Index is a registered trademark of Investment Property Databank Ltd. IPD and has been licensed for the use by Eurex for derivatives.
SLI®, SMI®, SMIM®, and VSMI® are registered trademarks of SIX Swiss Exchange AG.
The STOXX® indices, the data included therein and the trademarks used in the index names are the intellectual property of STOXX Limited, Zurich, Switzerland and/or its licensors
which is used by Eurex Frankfurt AG under license. Eurex derivatives based on the STOXX® indices are in no way sponsored, endorsed, sold or promoted by STOXX and its
licensors and neither STOXX nor its licensors shall have any liability with respect thereto.
Dow Jones, Dow Jones Global Titans 50 IndexSM and Dow Jones Sector Titans IndexesSM are service marks of Dow Jones & Company, Inc. Dow Jones-UBS Commodity IndexSM
and any related sub-indexes are service marks of Dow Jones & Company, Inc. and UBS AG.
All derivatives based on these indexes are not sponsored, endorsed, sold or promoted by STOXX Ltd., Dow Jones & Company, Inc. or UBS AG, and neither party makes any
representation regarding the advisability of trading or of investing in such products.
All references to London Gold and Silver Fixing prices are used with the permission of The London Gold Market Fixing Limited as well as The London Silver Market Fixing Limited,
which for the avoidance of doubt has no involvement with and accepts no responsibility whatsoever for the underlying product to which the Fixing prices may be referenced.
PCS® and Property Claim Services® are registered trademarks of ISO Services, Inc.
The names of other companies and third party products may be trademarks or service marks of their respective owners.
Challenges in Market and Counterparty Risk
Management
How to solve the critical issues? Solutions that can help.
Reinhard Keider; Head of Risk Architecture and

Risk Methodologies in Bank Austria
ZADAR, 14 May 2011

Don´t ever try to understand everything,
some things will just never make sense.
2
I guess some of you might agree with this saying.
Despite this I will try to
1) Make my brief presentation understandable

2) And hopefully it make sense to you.
….so let´s start !
3
AGENDA
 Part 1: Market Risk Management
 Part 2: Counterparty Credit Risk
 Part 3: New regulatory framework (Basel 2.5/3)
4
Part 1: Market Risk Management
Market Risk Management must be seen as an integrated approach.

Successful Market Risk Management should cover at the least the following
topics:
 Risk Management Unit to be independent from business unit
 Consistent framework where to operate (RM handbook, clear and
transparent limit structure, Escalation path in case of limit violation)
 Combination of limits: VaR limits, sensitivity limits, options limits
 State of the art model for monitoring/steering market risk (internal model
internal/external)
 Coverage of the relevant instruments/products
 Consistent Market data, risk factors and time series
 Flexible architecture to include new products fast and efficient
 Experienced people within Risk Management Unit
5
Components of MR Management
 Dependent on asset classes within the bank a market risk tool needs to
cover FX products, Interest Rate Instruments (both linear and non linear
ones), Equity and Credit Spread linked instruments. Currently commodity
and inflation linked products need to be covered properly.
 The number of Risk Factors plays important role to reflect a realistic risk
picture (e.g. FX: 75 currencies, 45 IR curves, 5000 EQ, 250 Spread Curves,
Vega risk: 2000 risk factors)
 State of the art modelling in terms of simulation; Historical simulation, Monte
Carlo Simulation for monitoring/steering market risk (internal model
internal/external)
 Stresstesting: importance still increasing; standard scenarios and macro
economic scenarios (e.g. Greece default, inflation, terror attack)
 Export /Transfer module and data source module allow fast integration of
new requirements
 Reporting engine to fullfill various information needs
 Backtesting gives answer about to model qualtiy
6
Comparison of current and new Market Risk Capital
Requirements for Trading Book
Current Approach New Approach
Value-at-Risk based capital Value-at-Risk based capital

charge charge
+ Incremental Risk Charge (IRC)

Specific Risk Surcharge
+
(capture default risk) + Stressed VaR
+ CRM for correlation trading

+ Capital from standardised portfolio
approach + capital from revised
standardised approach
7
Part 2: Counterparty Credit Risk (CCR)
Basic definition, types of CCR
 Counterparty Credit Risk (CCR) is the risk that a counterparty to

financial contracts fails to fullfil the obligations (payments, delivery of
assets) agreed in the contracts.
 Two types of CCR are distinguished

1. Settlement Risk:
The counterpary fails in setteling its due payments or deliveries
2. Presettlement Risk
The counterpary defaults before contract settlement
Following products are subject to CCR:
 OTC-derivatives (no exchange-traded instruments)

 Securities Financing Transactions (SFT)
(e.g. securities borrowing and lending, repurchase and reverse repurchase
agreements)
8
Why to introduce an internal model for counterparty
credit risk? Key benefits?
 Improvement of Internal risk management:

 A state-of-the-art model allows realistic assessment of actual counterparty credit
risk exposure
 Allows efficient use of credit lines by supporting risk mitigating effects:
 Full Netting effect
 Margining
 Portfolio effects
 Capital calculation:
 Same risk mitigation effects as for internal risk management apply
 Usually Significant RWA reduction compared to Current Exposure Method to be
expected
 Risk adequate pricing:

 Unilateral/Bilateral Credit Valuation Adjusments supported, taking full portfolio
view into account
9
Counterparty Credit Risk
Aspects of Managing Counterparty Credit Risk
means of CCR
management Coverage in
and control description Bank Austria
Pre-Settlement
Internal Limits:
Risk Limit
measure and  measure CCR and link exposure limits to CP credit grade
Exposure
limit counterparty  take advantage of risk mitigation techniques - use netting,
exposures margining, break clauses, reset agreements 
RWA (Risk Weighted Assets):
Basel 2 IMM
 measure Exposure at Default as underlying for RWA
Exposure-at-
hold capital for following Basel 2 Internal Model Method (IMM)
Default
counterparty  enable bank to absorb unexpected CCR losses during
exposures downturn periods 
CVA (Credit Valuation Adjustment)
 pricing of OTC contracts subject to CCR should reflect the
Credit Valuation
counterparty risk default risk
Adjustment
sensitive  fair value adjustment of risk-free prices for counterparty risk,
customer pricing  hedge against CVA-induced P&L variations 
10
10
Quantitative measures for CCR
Exposure measure and purpose
Exposure Measure purpose of application

equal to the current replacement cost of a
Current Positive Exposure (CPE) transaction
maximum exposure estimated to occur on a future
date at a high confidence level
Potential Future Exposure (PFE)  used for counterparty credit limit
• credit equivalent input for risk capital calculation

(regulatory and economic)
• input for cost calculation
Expected Positive Exposure • EPE profile input for Credit Value Adjustment (CVA)
(EPE) • closely related to EE (Expected Exposure)
11
Internal Counterparty Risk Model
Model overview and components
Transaction Counterparty
Market data Product valuation models Risk Management
data data
Spot rates Vanilla options (Black Swap rate Legal Limit

Yield curves Scholes) Reference agreements monitoring
Volatilities Exotic (accrual, rate  Netting Capital

extendable) Settlement data requirements
etc.
Barrier (standard, frequency  Collateral Credit
double) Maturity data Valuation
Swaption, cap/floor Country Adjustment
etc.
etc. etc. Reporting
Stresstesting
Validation
Market data pre-processing Aggregation
Scenario engine Instrument price distributions Output
USD interest rates Value of swap (% of notional)

7.0% 20
USD interest rates (%)
6.0% Limit
15
5.0%
10
Exposure
4.0% EE
5
3.0% PFE
0
2.0% Reduction)
0 1 2 3 4 5
1.0% -5
0.0% -10
0 2 4 6 8 10
0 2 4 6 8 10 12
Time Time Time
12
Part 3: Basel 2.5 and Basel 3
Chapter Title - Chapter Section Title
Challenges to the Banking industry in front of

new Basel regulations
13
Current situation (Basel II)
• VaRbased Trading book capital charge calculated using 99%

quantile of 10 day loss
Common assumption:
• Losses from issuer defaults in trading book positions negligible since
– Mainly high rated issuers in trading book
– Positions are sold in case of downgrading
Financial crisis:
• Losses >> trading book capital charge occurred
(e.g. Lehman Brothers)
• Particularly positions subject to credit spread risk (cds,
cdo,bonds,…)
• Significant part of the losses not caused by actual defaults but by
rating migrations
14
Regulatory Response
Basel Committee proposed changes to the capital requirements for

the trading book: (mainly for internal model)
• Incremental risk charge (IRC), specific risk

• Stressed value-at-risk, general risk
• Comprehensive risk measure for correlation trading activities
• For remaining securitization products the capital charges of the
banking book apply
Implementation date: 31 Dez. 2011
15
Timeline for Basel 2.5 and 3
Dec. 2010
Dec. 2009 Jan 2013
Basel III ~finalized
Basel III first draft,
‘A global regulatory Initial rise in capital
‚Strengthening the from Basel III
framework for more resilient
resilience of the banking
banks and banking systems’ Stressed EPE,
sector‘ Market Risk CVA charge,..
Leverage ratio
…July 2009 Dec. 2011 Liquidity ratio ..
Basel 2.5 proposal Basel 2.5 goes live Final implementation
Finalised June 2010 IRC, CRM, Stressed VaR,.. 2019
IRC, CRM, Stressed VaR,..
Basel Reform Programme: Focus on Counterparty Credit
Risk - Overview
A. Capital Base B. Counterparty Risk C. Leverage Ratio D. Procyclicality E. Liquidity Standard

Strengthen risk coverage, Introduce a "leverage ratio"
amending July 2009's as a supplementary Promote measures for Promote measures for
Raise quality, consistency trading book and measure to the Basel II building up capital buffers building up capital buffers
and transparency of bank's securitization reforms by framework with in order to in good times that can be in good times that can be
capital base adding capital build up excessive drawn upon in stress drawn upon in stress
requirements for leverage in the banking periods periods
counterparty credit risk system
EXAMPLES
 Determine capital requirement for counterparty credit risk using effective EPE calculation for a period
of stress (similar to 2009' proposal regarding market risk stressed VaR) and
 Introduce captial charge for mark-to-market losses (ie credit value adjustment losses = CVA risk)
associated with deterioation in the credit worthiness (not necessarily default) of counterparties
 Strengthen standards for collateral management and initial margining. Eg banks with large and illiquid
derivative exposures will have to apply longer margining periods.
 Banks with exposure to central counterparties (of those are meeting some CPSS/IOSCO "strict criteria"
only) will qualify for a zero percent risk weight.
 Include "wrong-way risk" (cases where the exposure rises when the credit quality of the counterparty
deteriorates) into Pillar 1 requirements, enhance stress test requirements, revise model validation
standards and issue supervisory guidance for sound backtesting practices of CCR.
17
Additional Market Risk Capital Charge
Credit Valuation Adjustment: Cover MtM of unexpected
counterparty risk losses
 In addition to the existing capital charge for unexpected losses arising from
counterparty defaults (Counterparty Credit Risk RWA) a stand-alone capital charge has
been proposed to cover the market risk of potential MtM losses due to spread driven
increase of unilateral credit value adjustments (CVA) of OTC portfolios.
 Unilateral CVA: Adjustment of the risk-free mark-to-model prices of OTC derivatives for
the credit risk of the counterparty (= expected loss).
 Calculation of the new CVA capital charge by evaluating the unexpected losses of a
portfolio composed of bond-equivalents each describing the OTC exposure to a
counterparty and associated hedges.
 Applying the applicable regulatory market risk charge (IMOD) to the bond-equivalents (i.e.
99% VaR: general + specific risk including stressed VaR but not IRC).
 Liquidation horizon = 10 days
 recognises hedges: eligible, single-name CDS / CCDS or other hedging instrument directly
referencing the counterparty
 Central Counterparties (CCP) and Securities finance transactions (SFT) are excluded
 CVAs already recognised by the bank as an incurred write-down, can be used to reduce the
Counterparty Default Risk capital charge (no “double counting”)
18
Now it´s up to you to decide:
1) Was the presentation understandable ?

2) Does it make sense ?
….in any case: it is the end! 
19
Many thanks for your attention !
20
Credit Risk of Traded Products under
Basel II
HEC Conference
Montreal
April 13, 2007
Niall Whelan, Director

Research & Model Risk Management Scotiabank
The views expressed here are those of the author and not necessarily those of
Scotiabank.
Introduction and Terms
 Three components of regulatory capital from trading activities:

1. Credit Risk
2. Market Risk
3. Operational Risk
 Credit Risk: Counterparties may default, leading to losses
 Market Risk: Market conditions may be adverse, leading to losses in the trading portfolio
 Operational Risk: Fraud, terrorism, pandemics, weather disasters etc. may occur, leading to losses to physical or
other assets
Layout of Talk
 Focus:
 Credit risk with some discussion of market risk
 Excluding: operational risk
 Agenda/Topics:
 The need for a special treatment
 History of regulatory capital and traded products
 pre-Basel
 Basel I
 Basel II
 Overview of Basel II requirements
 5 chapters of Trading Book document
 Implementation
 “use test” and connections to other risk measures
 Pillar 1, 2 and 3 issues
Credit Risk of Trading Portfolios
 Trading activities are significant businesses for most large banks:

 OTC derivatives
 repo and reverse-repo transactions
 security lending and borrowing
 They can all generate losses when counterparties default and should therefore be capitalised...
 HOWEVER
 ... we do not generally know how much our exposure will be if and when the counterparty defaults:
 values of traded products are intrinsically uncertain
 we do not know what the portfolio of transactions will be when default occurs
Market Risk of Trading Portfolios
 Derivatives portfolios are marked-to-market on a daily basis

 Typically well hedged but not perfectly so
 There is some residual net open position
 This net open position can deteriorate in value under adverse market conditions
 This effect is captured by Value-at-Risk (VaR) capital provisions. Based on extreme movement over 10 days of
exposure
 No banking book equivalent
 Illiquid positions may require special treatment
 Idiosyncratic risk factors may require special treatment
History of Regulatory Treatment for Capital
 Pre-Basel
 There was no international consensus on setting of capital
 Regulators did not adhere to rigid capital ratios
 relied on judgment of risk by the banks and themselves
 echoed in current "principle-based" approach
 Capital ratios fell through time
 1840 ratio was about 50%
 deposit insurance, access to central bank funds acted to buffer bank risk
 1940 ratio was about 6-8%
 Huge bank failures in the US during the Depression lead to much more active regulatory environment
 1973 collapse of Bretton Woods coincided with emergence of a significant Capital Markets industry to create
a more dynamic and volatile FX and interest rate trading environment
 1980's: Latin American debt crisis as well as the S&L crisis in the US led to more concern about the
robustness of the banking industry. The US brought in a series of ever tightening capital requirements
 1988: the Basel I accord...
Traded Products under Basel I
What is the exposure associated with an OTC contract?
OTC credit risk is the mark-to-market plus notional times the following add-on factor:
Residual Interest FX and Equities Precious Other

Maturity Rates Gold Metals (except Commodities
Gold)
< 1 year 0% 1% 6% 7% 10%
1-5 years 0.5% 5% 8% 7% 12%
> 5 years 1.5% 7.5% 10% 8% 15%
Consider:
• is silver really so different from gold?
• what about a cross-currency equity derivative?
• is a 1½ year swap really identical to a 4½ year swap?
• etc...
only reverse repos attract credit risk capital (not repos), due to collateral
only securities lending attracts capital (not securities borrowing), due to collateral
Basel I cont...
Further multiply by the counterparty riskfactor weight
Counterparty Type Riskfactor Weight

OECD 0%
governments
OECD banks and 20%
public service
entities
Corporate and 50%
other
Capital is then 8% of the product


Consider:
• shouldn't the perceived risk vary with economic conditions?
• are all OECD banks of equal credit risk?
• is a CCC corporate really identical to a AA corporate?
• etc...
Netting under Basel I
 Netting is the legal ability to settle on the net value of a portfolio of derivatives upon default by one counterparty (vs.
deal by deal)
 Potential to lessen the exposure at default
 First recognised as an amendment in 1995
A' = 0.4*A + 0.6*NGR*A
A is the add-on ignoring netting
NGR =(net current replacement cost) / (gross replacement cost)
Consider:
• this is very ad-hoc
• what about netting with collateral?
• deeply out-of-the money portfolios still attract significant capital
• not "coherent"/sub-additive
Market Risk Capital under Basel I
 In 1996 Basel outlined capital treatment of market risk (VaR).

 Based on the 99'th percentile of market loss over a 10 day horizon, multiplied by 3
Distribution of Market Losses
99'th
percentile
-3 -2 -1 0 1 2 3
Loss (Arb. Units)
• Banks have wide latitude in deciding how this is determined

• Regulatory focus on self-consistency, back-testing, internal controls and validation
• The result is a key internal risk measure that is regularly reported in financial statements
• “Prototype” for internal modelling approach now embedded in Basel II
• Separate charge for "specific risk" for derivatives subject to idiosyncratic entity-specific
risk (equity derivatives, credit derivatives etc.)
Basel II
 The main document:

 International Convergence of Capital Measurement and Capital Standards: A Revised Framework, June 2004
A document specific to trading activities was published in April 2005
Two months of industry consultation led to...
 The Application of Basel II to Trading Activities and the Treatment of Double Default Effects, July 2005
1. The treatment of counterparty credit risk and cross-product netting
2. The treatment of double default
3. The short-term maturity adjustment in the IRB approach
4. Improvements to the current trading book regime
5. A capital treatment for failed trades and non-DvP transactions
Counterparty Credit Risk under Basel II
   1   
Capital=EAD*LGD*A*  N  G(PD)  G(0.999)   PD 
 1-  
  1-  
1  e 50PD
  0.12 w  0.24(1  w) w
1  e 50
1  (M  2.5)b
A
1  1.5* b 0.45
b   0.11852  0.05478ln(PD) 
2 0.40
0.35
0.30
Capital
0.25
0.20
0.15
0.10
0.05
0.00
0.00 0.05 0.10 0.15 0.20
PD
Consider:
•1-e-50 = 1-10-22 is a built-in underflow
Counterparty Credit Risk under Basel II cont…
 Risk-weighted assets = 12.5*Capital

 RWA now a derived quantity, not fundamental to the capital calculation
 PD and LGD are common to other asset classes

 Trading book requirement is the determination of EAD and of M
Counterparty Credit Risk under Basel II continued...
 Determination of EAD and M for OTC: Three methods

1. Current exposure method (essentially Basel I augmented for credit derivatives)
2. Standardized method (I am not aware of any bank using this)
3. Internal (EPE) method
 Determination of EAD and M for repo-style: Three methods

1. Standardised haircuts
2. Adapted VaR
3. Internal (EPE) method
 EPE method. Use an internal model to estimate expected exposures over a one year time horizon
0.6
Expected
Exposure
0.4
0.2 Effective
Expected
Exposure
0.0
0.0 0.5 1.0
time (years)
Determine "running maximum" through time (roll-over risk)

Average of the running maximum times 1.4 is EAD
M is determined by a similar manipulation of the exposure profile
Factor of 1.4 is regulatory; it can be as small as 1.2
 Other considerations:
 Can be done at the level of a netting set
 Collateral treated as a negative offsetting position
 Under certain conditions can net OTC and repo-style
 Margin agreements effectively cap the potential exposure
 Based on expected exposure, not 99.9’th percentile
 Explains need for the scaling by 1.4
 Can impose one’s own scaling of:
 Subject to a floor of 1.2

 Industry studies indicate 1.1 to 1.2 is typical
EC(simulated exposures)
 Internal economic capital is an explicit component of the calibration
 We find about 20-30% relief relative to CEM
EC(expected exposures)
The treatment of double default
 Credit derivatives that are hedging loan exposures do not attract counterparty capital
 They are treated the same as a credit guarantee
 Under certain conditions guarantees or credit derivatives can attract "double default
treatment"
 Reflects the probability that both the counterparty and the guarantor must default to experience an
exposure.
C '  C  0.15  160PDg 
Consider:
This can actually increase capital if PDg  0.0053
Equation "should" be C '  capital determined using PD'
PD '  probability that both default
The short-term maturity adjustment in the IRB approach
 Technical point.
 By default M is floored at unity but in some cases it can be less than one
 Repo-style transactions with daily remargining are the most important application
Improvements to the current trading book regime
 Not an extensive overhaul, implying an overall regulatory and industry comfort with market risk under Basel I
 Changes include:
 Greater oversight about what can be held in trading books
 Greater concern by regulators about validity of the 10-day unwind period in VaR for illiquid positions
 More rigorous validation and stress testing standards
 Lessening of specific risk charge multiplier from 4 to 3
 Need to capture “event risk” and “incremental default risk” in specific risk
 Still somewhat contentious
 These are based on 99.9’th percentile of loss over one year
 Difficult to combine with market VaR defined as a multiple of 99’th over 10 days
A capital treatment for failed trades and non-DvP transactions
 Meant to plug a small hole in Basel I: how to treat trades which have failed to settle
 DvP - treat as a forward contract
 non-DvP - treat as a loan
 Typically so immaterial that applying a conservative factor is the best approach

Use Test
 Related credit risk measures Related market risk measures
 Economic Capital Desk and business-level market VaR
 Credit Line Utilisation
 Credit loss provisions (expected losses)
 Principle is that the internal model for these measures should be broadly in line (not necessarily identical) to what is
done for Basel
 Systems/infrastructure
 Data
Pillars 1, 2 & 3
 We have focused mostly on Pillar 1 – the determination of capital

 Significant Pillar 2 requirements – supervisory oversight. This includes disclosure to regulators, back-testing, stress-
testing, use-test and other evidence that the measures are embedded within a robust risk framework
 Important consideration for regulators in light of the "principle-based" approach
 Connection to Economic Capital particularly important
 Pillar 3 requirements – public disclosure. This is a relatively light component for the Trading Book
Bank Experiences
 A major improvement. Principle-based/internal modelling approach makes sense

 Avoid “regulatory arbitrage”
 More granular and dynamic reflection of credit and market risk
 Use test requirements
 Little time for consultation between first and second drafts of Trading Book document
 Regulators and Basel Committee were receptive to industry feedback
 Still some aspects of the accord that are somewhat contentious
 Tendency for Trading Activities to be an "after thought"
 Basel guidelines came later
 Gap analysis from OSFI came later and was not paragraph-by-paragraph
 Internally
 A major piece of work involving multiple departments across the bank
 A valuable tool to improve banks’ internal processes
 Data flows and system information
 Consistency of risk assessment across business lines
 Improvement and rationalisation of internal risk measures
 VaR
 Economic Capital
 Loss provisions
 Credit line utilisation
Singl e counterp arty credit
l imits ( SCCL ) for l arge
banking organizations
M arch 2 0 1 6
Executive summary
On Friday, March 4, the Federal Reserve these exposures against a more limited publication of the standardized approach
Board of G overnors (FRB) approved a Tier 1 capital base. This change is offset by for measuring counterparty credit risk
second notice of proposed rulemaking (“ re- a greater percentage limit — 15% instead exposures (SA-CCR).3
proposal” ) to establish single counterparty of 10% of capital for this tier. The three-
credit limits (SCCL) for large bank holding tier limit framew ork is intended to scale
companies (BHCs).1 The re-proposal the constraints to align w ith the greater
Broader definition of ‘single
applies to U S BHCs and foreign banking systemic risks posed by counterparty counterp arty ’
organizations (FBO) w ith U S$ 50 billion or interconnectedness between larger firms. The re-proposal largely adopts the
more in consolidated U S assets, collectively exposure combination and aggregation
defined as “covered companies,” and sets rules in the BCBS LES, w hich mandate an
stricter boundaries for firms with larger
Additional ex emp tions analysis of “ economic interdependence”
Exposures to qualifying central for large counterparties, and includes
systemic footprints. Comments on the re-
counterparties (QCCP s) and highly rated additional criteria to determine “ control
proposal are due by 3 J une 2016 . The re-
sovereigns (those w ith a 0% U S Basel III relationships.” These rules w ould broaden
proposal did not provide an official effective
risk w eight [ RW] ) are exempt in the new the definition of a single counterparty
date, but compliance w ill be required w ithin
proposal, w hich addresses the concern that to combine exposures w here a highly
one or tw o years from the effective date,
limits on exposures to QCCP s w ould create correlated probability of default exists
depending on the size of the institution.
an impediment to the regulatory goal of between specific counterparties. Firms
The re-proposal is more closely aligned encouraging central clearing. that have invested in enterprise legal entity
w ith the Basel Committee on Banking
identifier (LEI) infrastructure and more agile
Supervision’ s international standards for
More flexibility in derivatives counterparty aggregation hierarchies w ill be
measuring and controlling large exposures
better positioned to meet the operational
(BCBS LES — finalized in 2014).2 How ever, ex p osure measurement challenges of this requirement.
it presents some material changes from meth odol ogy
the original notice of proposed rulemaking If adopted as proposed, the re-proposal
The FRB recognizes the lack of risk-
(NPR) that reflect evolution in supervisory would require a significant capability build
sensitivity of the current exposure method
thinking: for many banking institutions. Firms w ill
(CEM) mandated in the prior proposal.
likely look to integrate SCCL into their
Firms w ould now be permitted to apply the
currently existing counterparty credit
T h ree- tier l imit structure for same exposure calculation methodologies
risk infrastructure and try to leverage
used for their U S risk-based capital
aggregate net ex p osures to a calculations (i.e., CEM or the internal model
related counterparty risk identification and
singl e counterp arty measurement capabilities to the greatest
method [ IMM] ), w hich requires supervisory
A three-tiered limit structure is introduced extent possible.
approval). U sing IMM generally results in
(increased from tw o in the original a low er exposure estimate than under the
proposal), effectively placing tighter relatively blunt CEM. How ever, it should be
constraints on exposures betw een the noted that the FRB is expected to reform
largest and most systemically important the counterparty risk-based capital rules in
counterparties in the top tier by measuring the future follow ing the Basel Committee’ s
Background and context
The single counterparty credit limits re-proposal implements part of The FRB estimates that systemically important financial institutions
the Dodd-Frank Act (DFA) building on earlier proposals released by (SIFIs) w ill have to reduce exposures to each other by approximately
the FRB in 2011 and 2012, and seeks to promote global consistency U S$ 100b in aggregate. This excess credit exposure is substantially
by generally aligning to the Basel Committee’ s large exposures less than the estimate in a 2012 industry study, as a result of changes
standards (BCBS LES) released in 2014. in the proposed rule’ s measurement approaches, in conjunction w ith
the progression in industry efforts to reduce bilateral exposure and
The re-proposal addresses regulatory objectives to:
further movement to central clearing.4 Nonetheless, bringing credit
• Reduce the risk of failure of individual BHCs by limiting their exposures w ithin limits w ill require active exposure management by
exposure to any single counterparty. firms. The SCCL is expected to pose a greater compliance challenge
• Reduce the systemic risk in the banking system by curtailing the for firms with substantial capital markets activities, particularly in
level of overall interconnectedness betw een BHCs, especially derivatives and securities lending.
betw een the largest market participants (i.e., major covered Similar to the original NP R, the re-proposal places a quantitative limit
companies). on exposure to a single counterparty, expressed as a percentage of
The re-proposal is more closely aligned w ith the BCBS LES and has a firm’s capital base. The re-proposal specifies the approaches firms
taken into consideration comments received on the original NP R, as may use in calculating counterparty exposure, in addition to the
w ell as the outcomes from both industry and supervisory quantitative definition of connected entities that firms should apply in determining
impact study (QIS) exercises. There are several material differences and aggregating their exposure to a single counterparty.
from the original 2011 SCCL proposal that could present substantial
new operational, data and technology challenges.
T imeline
The re-proposal did not provide a firm effective date but indicated a one-year implementation period for firms with consolidated assets of
$250b or greater (or $10b+ in foreign exposures) and two years for other firms following finalization of the rule. Given the time needed to
finalize the SCCL rules and this one- to two-year implementation period, the SCCL limits would likely be effective sometime in 2018–19 and
potentially as early as Q4 2017 . Figure 1 show s the timeline for the SCCL rulemaking and the development of the related guidelines of the
Basel LES and SA-CCR:
F igure 1
Mar. 2013
Supervisory framework for
Jan. 2017 Jan. 2019
measuring and controlling large
2008–09 Apr. 2014 BCBS SA-CCR BCBS LES
exposures “BCBS LES”
Global financial BCBS LES — implementation implementation
consultative document
crisis final standard effective date effective date
Basel Committee for Banking Supervisions (BCBS)
2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019
US
Dec. 2011 Dec. 2012 Feb. 2014 Mar. 2016 Jun. 2016
SCCL NPR per SCCL NPR EPS rule Revised SCCL Comments due Potential range for
DFA 165e as part for FBOs as finalized with NPR for BHC for the revised US implementation
of Enhanced part of EPS SCCL deferred and FBO SCCL NPR
Prudential rule
Standards (EPS)
rule
Potential range for US NPR
release of SA-CCR
2 | A ll Rights Reserv ed — © 2 0 1 6 Ernst & Y oung L L P .

The re-proposal: summary of key components
The follow ing is a summary of the major components of the re-proposal, as w ell as commentary on considerations and implications.
Limit constraint framework — The constraint framew ork generally follow s the approach from the original NP R — w ith larger and more
systemically important banks having tighter constraints. FBOs are required to report for any IHC and also for combined U S operations (CU SO)
w ith the potential that these dual reporting levels may need to measure limits off of different capital bases, each of w hich could be in a different
tier. Table A summarizes the new framew ork.
Table A
US covered FBO covered Singl e counterp arty l imit Ch ange from original
T ier comp any criteria comp any criteria1 Applicable capital base (% of capital base) SCCL p rop osal
BHCs w ith consolidated U S intermediate holding Total BHC or IHC regulatory 25% Resembles original
assets of U S$ 50b or companies (IHCs) w ith capital plus allow ance NP R
greater but less than consolidated assets of for loan and lease losses
U S$ 250b, and less U S$ 50b or greater but less (ALLL) not included in
than U S$ 10b in foreign than U S$ 250b, and less Tier 2
exposures than U S$ 10b in foreign
exposures
G eneral limit
N/A Combined U S operations P arent entity regulatory 25%
(CU SO) of FBOs w ith capital plus ALLL not
consolidated assets of included in Tier 2
U S$ 50b or greater but less
than U S$ 250b, and less
than U S$ 10b in foreign
exposures
BHCs w ith consolidated U S IHCs w ith consolidated Tier 1 BHC or IHC capital 25% Denominator has
assets of U S$ 250b or assets of U S$ 250b or been reduced to Tier 1
greater, or U S$ 10b or greater but less than capital
greater in foreign exposures, U S$ 500b, or U S$ 10b
but are not a “ major or greater in foreign
covered company” 2 exposures
T ighter limit
N/A CU SO of FBOs w ith Tier 1 capital of the parent 25%
consolidated assets of entity
U S$ 250b or greater but
less than U S$ 500b, or
U S$ 10b or greater in
foreign exposures
Major covered companies, U S IHCs w ith consolidated Tier 1 BHC or IHC capital 15% to major While the limit has
2 U S BHCs that are globally assets of U S$ 500b or counterparties 2 been relaxed to 15% ,
systemically important greater (i.e., G -SIFIs) the denominator has
(G -SIB) been reduced to Tier 1
25% to all other
capital
counterparties
I nter- S I F I limit
N/A CU SO of FBOs w ith Tier 1 capital of the parent 15% to major
consolidated assets of entity counterparties2
U S$ 500b or greater (i.e., G -SIFIs)
25% to all other
counterparties
1
FBOs would need to file two distinct limit reports: one for the IHC level and another for CUSO.
2
“Major covered companies” are a subset of “major counterparties.” Major counterparties also include any non-bank financial companies
supervised by the FRB.
S ingle counterparty credit limits ( S CCL ) for large b ank ing organiz ations | 3
Considerations for foreign banking organizations — Foreign banks Credit risk mitigation and risk transfer methodology — The re-
operating in the U S that are subject to SCCL w ould have to produce proposal adjusts the amount of exposure to be shifted to credit risk
tw o exposure reports: one at the intermediate holding company (IHC) mitigation (CRM) providers, essentially requiring full notional risk
level w here exposure w ould be measured relative to the IHC’ s capital shifting only when the obligor is a financial institution, whereas the
base and one for its CU SO using its parent’ s capital base. It is likely original NP R required this for all counterparties. This methodology is
that FBOs w ould be required to also comply w ith equivalent BCBS LES directionally consistent w ith the BCBS LES approach. An additional
requirements in the home country of their parent, subjecting them to clarification was made to remove the optionality of risk shifting to
multiple limit regimes. The additional requirements could introduce collateral issuers. Risk shifting to collateral issuers is now required
operational complexities for certain FBOs in maintaining related but in the re-proposal. Furthermore, the definition of eligible collateral
parallel reporting capabilities. has been slightly refined to explicitly include government-sponsored
entity (G SE) asset-backed and mortgage-backed securities exposures
Counterparty definition and aggregation — The re-proposal requires
and exclude private labels.
a counterparty’ s aggregate exposure to include exposure via entities
in w hich the counterparty has ow nership interest equal or greater Securities financing transactions (SFT) methodology — The SFT
than 25% (similar to the BHC definition of control), but also adds exposure calculation methodology has been relaxed to allow for a
the BCBS LES concepts of economic interdependence and control 5-day holding period (as opposed to 10 days in the prior proposal),
relationships to determine w hen entities should be aggregated into a w hich is expected to reduce exposure values relative to the original
single counterparty exposure. The new combination rules are more SCCL proposal.
complex, and compliance w ith these new requirements may require Exposures to funds and special purpose vehicles (SPVs) — The
qualitative and subjective data collection efforts — from sources that re-proposal clarifies the approach for funds and securitization
are typically not readily available — in addition to enhancing how vehicles by adopting the BCBS LES treatment, w hich requires the
proprietary counterparty hierarchy systems are managed. largest banks to “ look through” a vehicle to the underlying assets
E x emp tions — The re-proposal broadened the scope of counterparty once the total exposure to the fund exceeds 0.25% of capital. Failure
exposures that would be exempt from SCCL limits. Specifically: to obtain the necessary data to look through a vehicle w ould require
an allocation of such exposure to an “ unknow n counterparty”
a. Foreign sovereigns attracting a 0% risk w eight (RW) under
and subject it to the single counterparty limit. This “ unknow n
risk-based capital rules w ould now be exempt in addition to U S
counterparty” bucket could accumulate exposure if a firm’s data
G overnment exposures. The re-proposal does not detail w hether
quality or data sourcing efforts yield insufficient look-through results,
exempted sovereign exposures w ould need to be reported (as
w hich could ultimately constrain business activity.
required in certain cases under BCBS LES). The re-proposal
maintains that home country sovereign exposures for FBOs
remain exempt regardless of the risk w eight
b. Exposures to QCCP s are now fully exempt in alignment w ith
BCBS LES.
c. U S G overnment-sponsored entities operating under
conservatorship or receivership of the U S G overnment) and
Federal Home Loan Banks are now explicitly exempt.
d. A key point to note, how ever, is that U S BHCs and FBOs that
reduce gross exposures to exempt counterparties using non-
exempt eligible risk mitigation products are required to shift
these exposures to the non-exempt mitigants providers.
e. Intraday exposures continue to be exempt.
Derivative exposure calculation methodology — The re-proposal
expands the allow able methodologies to calculate credit exposure
to derivatives counterparties from the previously mandated
CEM. Covered companies may now use any of the methodologies
authorized under the FRB’ s risk-based capital rules, including the IMM
if approved, w hich could potentially provide a reduction in exposure
vs. CEM.
Though the U S regulators have not proposed the BCBS revised
standardized approach for counterparty credit risk, or SA-CCR, the
direct reference to SA-CCR corroborates the assumption that the U S
implementation of SA-CCR w ill be in line w ith the proposed Basel rules
and effective on or soon after the Basel effective date of 1 J anuary
2017 .5 The SCCL re-proposal notes that the FRB w ill consider the
benefits of incorporating SA-CCR for risk-based capital purposes. It
should be noted that SA-CCR is the only permissible methodology for
measuring derivatives potential exposure under BCBS LES.

Rul e design and imp l ementation ap p roach
Both BCBS LES and SCCL share common required capabilities for monitoring and controlling large exposures. Financial institutions (FIs) must
be able to aggregate credit exposures across all products (trading book, banking book, off-balance sheet, etc.), across all internal in-scope
entities and to all ultimate counterparties as prescribed.
Institutions should consider the extension of existing infrastructure build-out efforts w here possible and develop capabiilities that can be
levered across multiple regulations. In addition to the business constraints of restricting credit exposures against a regulatory limit, compliance
w ith SCCL requires an integrated data and technology infrastructure across front-, middle- and back-office systems within risk, finance
and lines of business. Complex alignment of multiple limits management business processes w ould be also necessary to manage and monitor
the SCCL limits along with other regulatory and internal limits. Largest impacts can be grouped into: counterparty identification and hierarchy
management, exposure calculation and aggregation, and reporting and limits monitoring, as described in Figure 2.
F igure 2
Counterparty identification and Exposure calculation Reporting

hierarchy management and aggregation and limits monitoring
Issuer/reference
- entity
information management
Counterparty/borrow er
information management Lending and investment
book exposure
OTC Derivative/SFT traded
product exposure
Limits and
Exchange traded threshold
product exposure monitoring
Exposure
Issuer risk netting and
aggregation
Risk transfer
Reporting
Netting Collateral
Other
P rotection credit risk
mitigation
Figure 2 presents a logical division of activities required for an SCCL solution:

i. Counterparty identification and hierarchy management — The re-proposal prescribes grouping counterparties together using qualitative
criteria that may be new or different than how counterparties are currently grouped w ithin a bank’ s systems leading to the need to
maintain multiple hierarches. Additionally, the data required to determine economic interdependence and control relationships likely w ould
require new data sourcing and management efforts.
ii. E x p osure cal cul ation and aggregation — Specific rules regarding shifting risk to credit and equity derivative protection providers and
issuers of collateral make this a challenging exercise across several legal entities within a firm. Universal common identifiers or extensive
mapping tables w ould be required to enable proper calculations.
iii. Rep orting and l imits monitoring — Calculating and monitoring exposure on a daily basis present significant data and technology
challenges, highlighting the need for an integrated end-to-end infrastructure to be able to apply gross and net calculations on a timely
basis. Reporting of exposures for counterparties subject to SCCL also poses challenges given the already existing regulatory mandate of
producing various reports on a daily, monthly or quarterly basis.
Considerations and op en items
Overlap with other regulatory and business-as-usual (BAU) processes
• There are significant overlaps in data and process requirements with other regulatory exposure measures, such as Basel III risk-based capital
and leverage ratio rules, CCAR G lobal Market Shock largest counterparty default loss and legal lending limit (LLL) rules, as w ell as banks’
internal risk limits monitoring. The existing capabilities of these BAU processes can be significantly leveraged for SCCL implementation. See
Figure 3 for an illustrative view of the overlaps w ithin the counterparty credit risk landscape.
• These various measures of credit exposures and potentially disparate w ays in w hich they impact capital can create additional complexities for
capital management efforts. Bank capital and risk management will increasingly require enhanced MIS and analytics to efficiently manage
allocation of capital and balance sheet constraints.
F igure 3
Numerous existing and emerging regulatory reform demands require an integrated counterparty risk infrastructure for financial institutions
operating in the U S. The regulatory set below provides an illustrative view of some of the prevailing CCR issues requiring attention.
Stress testing and CCAR
Capital/RWA FR Y-14 Counterparty

projections reporting default scenario
Qualified
Common infrastructure Reg W – Sec.
financial
23A
contracts Counterparty IDs and hierarchy
management (including LEI) OCC:
SR 14-1
lending limit
Exposure calculations and
BCBS 239: risk
analytics (CEM, SA-CCR1, IMM,
data BCBS LES
CVA, SFT Coll. Haircut, VaR)
aggregation
Collateral sourcing, application
FSB common FRB SCCL
and management
data template
Exposure sourcing and

aggregation Standardized
2052a report
Liquidity RWA2 Supplementary
coverage ratio Leverage Ratio
Reporting and limits management
Net stable Advanced
funding ratio RWA
Document management
1
Formerly titled the “ Non-Internal Models Method” (NIMM)
2
Revisions to Standardized Approach currently in BCBS comment status
F inal rul e uncertainty

• The
 re-proposal contains 58 questions for which the regulators are soliciting industry feedback. The questions cover nearly all aspects of the
revised rule. Approximately half of the questions address definitions included in the re-proposal and ask for industry participants to confirm
whether the definitions are sufficiently clear and/or appropriate, or if additional regulatory guidance is required. Several of the questions
address the calculation of gross exposures, the recognition of collateral and short-positions, as w ell the recognition of eligible guarantees and
credit protection. These questions are in regards to the overall exposure calculation, and any changes made by the regulators reflected in the
final rule might have a significant impact on the limit calculations performed by the covered companies. Lastly, a significant proportion (17 of
the 58 questions) addresses the treatment of FBOs and their related IHCs and combined US operations.
• Notably, one question (No. 52) asks w hether FBOs should be allow ed to use internal models to value derivatives transactions for the sole
purpose of complying w ith the SCCL rules.

Actions to be considered
• Coordinate comment letter and program — • Assess data and technology — Assess the
Coordinate roles and responsibilities across impact on overall counterparty risk management
impacted functions to establish preliminary program infrastructure and processes, including
governance, and determine whether the bank foundational exposure identification, existing
will engage in providing comments individually or counterparty hierarchies and capabilities required
through industry efforts. to 1) identify counterparties in scope of the
• Conduct pro forma analyses — Update exposure economic interdependence analysis, and 2)
concentration impact analyses to reflect the look through to collateral issuers, guarantors,
updated proposed rule changes and determine credit protection providers and SPVs. Firms
whether mitigation actions are necessary. should also conduct a holistic application and
architecture design building on the assessment
• Perform a gap analysis — Determine the extent of existing exposure calculation and reporting
to which existing capabilities can be leveraged capabilities, and firms’ ability to support multiple
versus new capabilities that would need to be methodologies and daily monitoring.
developed as part of the implementation efforts.
This would include assessing the connection points
with other internal and regulatory requirements,
such as other limit frameworks (e.g., legal lending
limits, internal risk limits), BCBS LES, Basel risk-
based capital, CCAR, and existing reporting and
monitoring frameworks.
Single counterparty credit limits (SCCL) for large banking organizations | 7

K ey change summary tab le
Table B summarizes additional key changes across the SCCL NP R from the original proposed rule and notes some potential implications.
Table B
Category I tem Commentary
Considerations for FBOs An FBO w ould likely be required to monitor and report on another layer of concentration
limits if its home country regulator has BCBS LES requirements.1
S coping and
definitions Eligible collateral definition The list of eligible collateral now explicitly includes G SE exposures. The eligible
collateral definition differs slightly from the risk-based capital rules, which could cause
implementation challenges.
Combinations New rules prescribing how counterparties must be aggregated around economic
interdependence and control relationships could have significant data sourcing, IT and due
Counterparty diligence implications.
identification Attribution rule The FRB softens language on w hen the attribution rule must be applied. The FRB now
and hierarchy states the intent to prevent “ undue burden.” The attribution rule must only be monitored in
management “ the ordinary course of business.”
Funds and securitizations A new “ look-through approach” and “ unknow n counterparty” bucket may cause business
and operational challenges.
Derivative exposure calculation The new potential to use IMM as another option instead of only CEM presents certain banks
methodology w ith a more risk-sensitive approach to calculating exposure.
Credit risk mitigation methodology — The adoption of BCBS LES-styled methodology in w hich full credit/derivative notional is
eligible credit/equity derivatives shifted to the protection provider only for financial reference assets should reduce overall
Exposure calculation inter-SIFI exposures.
and aggregation Credit risk mitigation methodology — Removing optionality and requiring risk shifting to collateral issuers and protection
risk shifting providers could create an exposure if collateral w as being used to offset an exempt
exposure.
SFT exposure calculation methodology The re-proposal permits a 5-day liquidation period, dow n from 10 days in the original NP R;
this is expected to reduce exposures.
Exemptions The re-proposal features an expanded set of counterparties exempted from the limit,
including QCCP s and sovereign entities that have a 0% RW under risk-based capital rules,
Reporting and limits w hich could alleviate some business concerns.
monitoring Timing Larger banks still must monitor exposures on a daily basis and produce a monthly report to
the FRB; how ever, the operational burden has been softened on smaller banks < U S$ 250b,
w hich now have quarterly calculation and reporting requirements.

G lossary
A L L L Allow ance for loan and lease losses IH C Intermediate holding company
B CB S Basel Committee on Banking Supervision IM M Internal models method
B H C Bank holding company L EI Legal entity identifier
CCA R Comprehensive Capital Analysis and Review L ES Large exposure standards
CCP Central counterparty L L L Legal lending limit
CCR Counterparty credit risk N P R Notice of proposed rulemaking
CEM Current exposure method O T C Over the counter
CRM Credit risk mitigation Q CCP Qualifying central counterparty
CU S O Combined U S operations Q IS Quantitative impact study
D F A Dodd frank act RW Risk w eight
F B O Foreign banking organizations RW A Risk w eighted assets
F RB Federal Reserve Board of G overnors S A - CCR Standardized approach for counterparty credit risk
G S E G overnment-sponsored entity S CCL Single counterparty credit limits
G -S IB G lobal systemically important bank S F T Securities financing transactions
G -S IF I Global systemically important financial institution V aR Value at risk
F ootnotes
1
Board of Governors of the Federal Reserve website, http://w w w .federalreserve.gov/new sevents/press/bcreg/2016 0304b.htm, 4 March 2016
2
Supervisory framework for measuring and controlling large exposures, Basel Committee on Banking Supervision standards, April 2014
(accessed via http://www.bis.org/publ/bcbs283.pdf, Accessed March 2016 )
3
The standardised approach for measuring counterparty credit risk exposures, Basel Committee on Banking Supervision, March 2014
4
“ TCH Study Finds that Measurement Methodologies P roposed for Estimating SSCL Exposures Overstate Actual Risk,” The Clearing House
Association website, https://w w w .theclearinghouse.org/publications/2012/sccl-study, 19 July 2012
5
The standardised approach for measuring counterparty credit risk exposures, Basel Committee on Banking Supervision, March 2014
Contacts
Adam Girling Vipul Karundia
Principal Executive Director
Tel: +1 212 773 9514 Tel: +1 212 773 3985
Email: adam.girling@ey.com Email: vipul.karundia@ey.com
Marc Saidenberg Markus Buri

Principal Senior Manager
Tel: +1 212 773 9361 Tel: +1 212 773 8473
Email: marc.saidenberg@ey.com Email: markus.buri@ey.com
Judy Modica Chris Jackson

Executive Director Manager
Tel: +1 212 773 0814 Tel: +1 212 773 1360
Email: judy.modica@ey.com Email: chris.jackson@ey.com
EY | Assurance | Tax | Transactions | Advisory
A b out EY
EY is a global leader in assurance, tax, transaction and advisory
services. The insights and quality services w e deliver help build trust and
confidence in the capital markets and in economies the w orld over. We
develop outstanding leaders w ho team to deliver on our promises to all
of our stakeholders. In so doing, w e play a critical role in building a better
w orking w orld for our people, for our clients and for our communities.
EY refers to the global organization, and may refer to one or more, of

the member firms of Ernst & Young G lobal Limited, each of w hich is a
separate legal entity. Ernst & Young G lobal Limited, a U K company limited
by guarantee, does not provide services to clients. For more information
about our organization, please visit ey.com.
Ernst & Young LLP is a client-serving member firm of Ernst & Young
G lobal Limited operating in the U S.
EY is a leader in serv ing the glob al financial serv ices mark etplace
Nearly 43,000 EY financial services professionals around the w orld
provide integrated assurance, tax, transaction and advisory services to
our asset management, banking, capital markets and insurance clients.
In the Americas, EY is the only public accounting organization w ith a
separate business unit dedicated to the financial services marketplace.
Created in 2000, the Americas Financial Services Organization today
includes more than 6,900 professionals at member firms in over 50
locations throughout the U S, the Caribbean and Latin America.
EY professionals in our financial services practices w orldw ide align

w ith key global industry groups, including EY’ s G lobal Wealth & Asset
Management Center, G lobal Banking & Capital Markets Center, G lobal
Insurance Center and G lobal P rivate Equity Center, w hich act as hubs
for sharing industry-focused know ledge on current and emerging
trends and regulations in order to help our clients address key issues.
Our practitioners span many disciplines and provide a w ell-rounded
understanding of business issues and challenges, as w ell as integrated
services to our clients.
With a global presence and industry-focused advice, EY’ s financial

services professionals provide high-quality assurance, tax, transaction
and advisory services, including operations, process improvement, risk
and technology, to financial services companies w orldw ide.
© 2016 Ernst & Young LLP .

All Rights Reserved.
EYG no. 00099-161US

1603-1851784
ED None
This material has been prepared for general informational purposes only and is not intended to
be relied upon as accounting, tax, or other professional advice. P lease refer to your advisors for
specific advice.
ey. com
Counterparty credit risk in portfolio risk management
Prominent financial institution failures reminded market participants that over-the-counter
derivatives bring counterparty credit risk. Even as these markets move towards settlement
through clearing houses, significant volumes of existing and new transactions remain bilaterally
settled, especially as non-standard derivatives may not qualify for central clearing. UBS Delta is
providing tools for clients to measure counterparty exposure alongside other investment risk
Prudent risk management of credit portfolios includes measurement Exposure modelling
and limitation of exposure to individual issuers to manage Here we use a simple example to explain exposure measures:
concentration risk. Investment portfolios will have limits, for example, The only trade with counterparty “MadeUpBankLtd” is a foreign
on percentage of current value invested in securities issued by “Bank exchange forward, buying EUR10m for USD, one-year forward at
XYZ”. Where over-the-counter (OTC) derivative counterparties are 1.3600. We allow (for educational purposes only) currency returns
also issuers of securities held, counterparty risk may be incremental to take seven equally spaced values, with defined probabilities, as
to issuer exposure. If a portfolio includes a swap with Bank XYZ as the in figure 1. We calibrate the distribution ensuring that the mean
counterparty, then exposure to them failing on that swap should be price equals the forward and standard deviation of returns equals
considered alongside exposure to them failing on their debt issues. one-year volatility.
For each rate we calculate the value of the trade (and thus the
Counterparty credit risk measurement netting set) and from that the counterparty exposure: positive
Counterparty exposure is properly measured not just by the netting set value equals exposure; negative and the exposure is
current value of trades with a counterparty, but also by how zero, as money is owed to the counterparty. (Note: In either case,
this value can move as markets move. Where sets of trades with we will lose the transaction at default. If we wish to keep the same
counterparties feature multiple risk drivers/asset classes, modelling position in EUR|USD, the trade will need replacing.)
the potential exposure becomes a complex problem. Many The average of these exposure numbers – expected exposure
investment banks have had to tackle this challenge. At UBS we (EE) – is just over USD0.5m. EE is driven by EUR|USD and by the
have recently built a new counterparty risk trading system (see box, volatility of EUR|USD, due to asymmetry of the exposure profile.
Modelling CVA and counterparty exposure), which we are rolling out Potential future exposure (PFE), a measure of the extreme of the
internally and now making available to clients through UBS Delta, exposure, is just under USD3m, at the 97.5% confidence level. PFE is
our portfolio risk management and performance system for clients. commonly used to limit exposure to individual counterparties.
Table A shows exposure measures calculated for an example large EE numbers have the advantage that, when being aggregated
portfolio of uncollateralised in-the-money (ITM) long-dated swaps – across netting sets (that cannot be netted together), the EEs are
four sets of nettable trades (netting sets) with three counterparties. additive, unlike PFEs. EEs are therefore often used for portfolio level
measures and limits.
1 An illustrative model – EE and PFE
Frequency Value of Sensitivities
exposure
45% We can use this simple model to derive useful measures of sensitivity
EE (expected exposure) – mean of
the exposure numbers of exposure to market drivers and volatility changes. Shifting
PFE (potential future exposure) –
20% 20%
maximum exposure at a certain EUR|USD up by 0.01 pushes the EE up on this trade by ~USD32,000
2.5%
5% 5%
2.5%
confidence level
and the exposure rises by ~USD49,000 per one point rise in volatility.
EUR|USD
0.8000 0.9400 1.0800 1.2200 1.3600 1.5000 1.6400 1.7800 1.9200 Where netting sets have large numbers of market drivers, sensitivity
measures are very useful, especially when volatile markets cause large
Buy EUR10m for USD13.6m, one-year forward (only trade with c/p) exposure moves just as counterparty credit may need extra attention.
Probability 2.5% 5% 20% 45% 20% 5% 2.5%
Exposure profiles
Rate 0.9945 1.1015 1.2200 1.3512 1.4965 1.6574 1.8356
(in one year) Performing the same exercise for different time horizons will give
Trade value -4.2m -2.8m -1.4m 0 +1.365m +2.974m +4.756m
EE and PFE exposure profiles. Taking the second row of table A as
Exposure 0 0 0 0 +$1.37m +$2.97m +$4.6m
an example, we can show the exposure profile (EE and PFE) across
Source: UBS Delta
Prob x exp 0 0 0 0 $0.27m $0.15m $0.12m

the maturity spectrum for this set of receive-fixed ITM swaps
(see figure 2). Many risk limit frameworks use peak PFE/peak EE
EE = $0.54m 97.5% PFE = $2.97m
measures, as shown in table A.
SPONSORED EDUCATIONAL FEATURE
A. E xposure and sensitivities – summary

Value Peak 97.5% PFE Peak EE CVA
Counterparty/netting set Spread delta EUR IR delta
(millions) (millions) (millions) (millions)
Source: UBS Delta Risk Analytics
A Bank Ltd 380.7 893.9 380.7 27.8 247,323 -276,062

DEF Ltd (Set 1) 136.2 450.6 136.2 13.2 98,041 -130,416
DEF Ltd (Set 2) 364.8 930.6 364.8 40.8 208,958 -481,542
GHI, Inc. 299.8 762.8 299.8 29.0 162,441 -298,540
Total 1,181.5 – 1,181.5 110.8 716,763 -1,186,560
2 Exposure profile – DEF Ltd Modelling CVA and counterparty exposure

500 The CVA quant team explains our approach to modelling exposure, now
being integrated into UBS Delta:
400 PFE 97.5% The computational kernel of UBS’s counterparty exposure system is
EE based on models for valuing and hedging CVA. This emphasises accuracy
300 as CVA is used to compute initial credit charges and exposure hedges. To
evaluate CVA, we must evaluate portfolio price distributions. From these
Source: UBS Delta Risk Analytics
200
distributions, risk measures including EE, PFE and sensitivities are derived.
100 UBS’s CVA system addresses the main technical challenges of computing
credit exposure as follows:
0 Scenario generation. Counterparty exposure needs a portfolio view to
20 20 20 20 20 20 20 20 20 20
10 15 20 25 30 35 40 45 50 55 account for netting effects – all constituent trades are analysed using the
same set of scenarios. So that the simulation and pricing methodologies
deal with all types of products in a consistent way, models follow a risk-
Credit valuation adjustment neutral dynamic, calibrated every day using the same market parameters
Credit valuation adjustment (CVA) can be a useful measure for used for mark-to-market valuation. Scenario consistency is ensured by
investment managers understanding and hedging counterparty using the same numeraire dynamics across all products.
risk. It is an accounting adjustment to positive replacement values Accuracy. To deal with exotic types of transactions UBS’s CVA system uses
of derivative instruments for financial institutions. It is the estimated American Monte Carlo (AMC) techniques and a generic mathematical
cost of hedging counterparty exposure, based on the EE profile and framework, which can be applied to all types of transactions in a
the credit default swap (CDS) market, accounting for correlation consistent way.
between exposure and probability of counterparty default (‘wrong- Product representation. UBS has developed a Portfolio Payoff Language,
way risk’). Using UBS Delta, clients can calculate CVA for a netting agnostic to the product type, which allows abstract trade representation.
pool and show sensitivities of CVA to movements in credit spread There is then a direct mapping between this representation and the
and underlying market drivers – table A – in order to hedge CVA. AMC pricing algorithm.
C-CDS approach. CVA is computed as the price of a contingent credit
Exposure measurement for portfolio managers default swap. This allows determining the potential future evolution
If a pension scheme chooses to bridge an asset/liability gap by of CVA and taking into account correlation between the default of the
receiving fixed on long-dated interest rate swaps, that hedge will counterparty and the underlying risk factors.
only be effective as long as the counterparties to the trades survive. Sensitivities. To enable proper CVA hedging, UBS’s CVA system computes
both credit and market deltas, as well as cross gammas.
Even where netting sets are collateralised, counterparty
Collateral and close-out. To correctly price counterparty exposure,
risk should be managed. For example, taking the first netting
we take account of collateralisation and all credit support annex
set (“ABank Ltd”) above and modelling as a fully collateralised
characteristics to calculate close-out risk, which can be considerable.
relationship gives a 15-day 99% close-out value-at-risk of USD131m.
For the buy side in general, exposure to counterparties needs Reference: Cesari, G et al (2009), Modelling, Pricing, and Hedging Counterparty
Credit Exposure: A Technical Guide (Springer Finance)
to be monitored, and understanding the sensitivity of exposure
to movements in underlying drivers is very important for
management of counterparty risk.
This material has no regard to the specific investment objectives, financial situation or particular needs of About UBS Delta
any specific recipient and is published solely for information purposes. No representation or warranty, either
express or implied, is provided in relation to the accuracy, completeness or reliability of the information UBS Delta is UBS’s award-winning portfolio analysis and risk
contained herein, nor is it intended to be a complete statement or summary of the developments referred to management system. Clients use UBS Delta to measure and manage
in this material. This material does not constitute an offer to sell or a solicitation to offer to buy or sell any
securities or investment instruments, to effect any transactions or to conclude any legal act of any kind risk, attribute performance and optimise portfolios across asset
whatsoever. Nothing herein shall limit or restrict the particular terms of any specific offering. No offer of classes. We run regular education sessions helping our clients to
any interest in any product will be made in any jurisdiction in which the offer, solicitation or sale is not
permitted, nor to any person to whom it is unlawful to make such offer, solicitation or sale. Not all products make best use of the system’s functionality, including counterparty
and services are available to citizens or residents of all countries. Any opinions expressed in this material are credit exposure.
subject to change without notice and may differ or be contrary to opinions expressed by other business areas
or divisions of UBS AG or its affiliates (“UBS”) as a result of using different assumptions and criteria. UBS Contact: Lindsey Matthews, CFA
is under no obligation to update or keep current the information contained herein. Neither UBS AG nor
any of its affiliates, directors, employees or agents accepts any liability for any loss or damage arising out of
UBS Delta, Head of Client Education & Marketing
the use of all or any part of this material. E: delta@ubs.com
© UBS 2010. The key symbol and UBS are among the registered and unregistered trademarks of UBS. All www.ubs.com/delta
rights reserved.
WHITE PAPER
RISK MANAGEMENT
COUNTERPARTY CREDIT RISK BEYOND

BASEL - IMPROVING SA-CCR FOR INTERNAL
RISK MANAGEMENT
RISK MANAGEMENT
COUNTERPARTY CREDIT RISK BEYOND

BASEL - IMPROVING SA-CCR FOR INTERNAL
RISK MANAGEMENT
Counterparty Credit Risk Beyond Basel 1
The new SA-CCR

INTRODUCTION – IMPROVING SA-CCR FOR
INTERNAL RISK MANAGEMENT The Basel Committee’s new standardised approach
for measuring counterparty credit risk exposures will
The Basel Committee’s new standardised approach for fundamentally revise the calculation of counterparty
measuring counterparty credit risk exposures (SA-CCR), exposure-at-default (EAD) for the purposes of regulatory
effective from January 2017, places banks at an interesting capital adequacy.
crossroads regarding a possible upgrade to their internal
The regulatory EAD is critical because it represents the
risk measurement and control policies. Many banks that
“loan-equivalent” amount, which in turn is multiplied by a
don’t yet apply advanced internal models may benefit from
risk weight (generally reflecting the creditworthiness of the
moving towards an internal risk management approach
counterparty) to give the amount subject to the relevant
inspired by the SA-CCR, e.g., for managing credit limits,
minimum capital ratios (e.g., eight percent).
in addition to adopting the SA-CCR for regulatory capital
adequacy. However, there are some important caveats The new SA-CCR is widely recognised as being more risk
and flexibilities that need to be taken into account, as we sensitive than the Current Exposure Method. Table 1
explain below. sets out the key differences between the CEM and SA-CCR
methodologies, with the main changes being:
●● An improved recognition of offsets between transactions
within “hedging sets”, i.e., groups of transactions that share
The situation today the same (or correlated) underlying risk factors. This
represents a significant improvement compared to the
Until now banks have had to make a choice between two
much-criticised “net-to-gross ratio” (NGR) methodology
main methods of calculating counterparty exposures for
used in the CEM.
the purpose of making internal decisions about credit risk
management: ●● A more accurate treatment of collateralisation and
margining, effectively recognising:
●● Traditional mark-to-market plus add-on approach
(MtM + Add-on). –– The shorter time horizon of margined exposures, based
on the Basel III “margin period of risk” rules.
●● Simulation of potential future exposure (PFE) using
a Monte Carlo engine. –– The effect of “thresholds” and “independent collateral
amounts” (including initial margins) on potential future
Some banks adopt a hybrid policy such as using PFE
exposure.
simulation only for certain products or only as an end-of-
day calculation, with non-simulated or intra-day exposures ●● A closer alignment with the Internal Model Method (IMM).
(e.g., for pre-deal limit checks) calculated as MtM + Add-on. For example, the SA-CCR weighting factors are calibrated
to a period of stress, the formula incorporates an “Alpha”
Whilst a simulation-based approach is deemed to be best
multiplier of 1.4, and it assumes a risk horizon of one year
practice, smaller financial institutions with low volumes of
for all exposures.
derivatives often cannot justify investing in a Monte Carlo
engine integrated with their limits framework.
The MtM + Add-on approach is similar to the “Current

Exposure Method” (CEM) for calculating regulatory capital,
although banks using a non-simulation based approach for
internal risk measurement generally fall into one of two
camps:
●● Those that base their policy exactly on the Basel I & II CEM. Many banks that don’t yet apply
advanced internal models may
●● Those that have developed an internal MtM + Add-On benefit from moving towards an
policy, which may or may not be inspired by the CEM. Such internal risk management approach
policies may include more refined add-on percentages, inspired by the SA-CCR.
time-banded exposure profiles, add-on offsetting based
on risk factor sensitivities, etc.
This is why some banks are now at a crossroads. With the new
SA-CCR regulations making the CEM capital calculation
obsolete from January 2017, should banks continue to base
their internal risk measurement on the CEM and MtM +
Add-on approaches, or replace these with an approach
more in line with the new SA-CCR?
2 Counterparty Credit Risk Beyond Basel
Table 1: Key differences between the CEM and SA-CCR

methodologies
CEM SA-CCR
Basic Formula EAD = Max (0, Replacement EAD = Alpha x (Replacement Cost + Add-on) Alpha = 1.4
Cost + Add-on – Collateral)
Replacement Cost Max (MtM, 0) Unmargined: Max (MtM - Collateral, 0)

Margined: Max (0, MtM - Collateral, Threshold +
MTA - Net Independent Collateral Amount)
Add-on Time Dimensions <1Y, 1-5Y, 5Y+ 1Y

Scaled by sqrt(t) for transactions less than 1Y
Scaled by “supervisory duration” for IR & Credit Derivatives
Margined Add-on No scaling Scaled by sqrt of Margin Period of Risk x 1.5

(i.e., 0.3 for a typical 10-day MPOR)
Margined Add-ons Up to 60% based on “net-to- Up to 100% within “hedging sets” (long vs short in same
gross” ratio underlying asset)
Partial offsetting (based on “supervisory correlation”)
across equities, credit & commodity types
No offsetting across hedging sets
Add-ons Calibration Based on pre-crisis volatilities Based on period of stress
Offsetting of Negative None “Multiplier” formula allowing reduction

Replacement Cost against Add-ons floored to 5%
Delta-weighting of Add-ons None Simple Black formula using conservative

”supervisory volatilities”
Table 2: Comparison of CEM versus SA-CCR exposure for sample

transactions
CEM SA-CCR Details of SA-CCR

Transactions (% of notional) (% of notional) calculation SA-CCR/CEM
5-year Interest Rate

0.5% 3.1% 0.5% Add-On x 1 Maturity Factor (MF) x 4.4 620%
Swap – Unmargined
Supervisory Duration (SD) x 1.4 Alpha
5-year Interest Rate 0.5% 0.93% 0.5% Add-On x 0.3 MF x 4.4 SD x 1.4 Alpha 186%
Swap – Margined
3-month FX Forward –
1% 2.8% 4% Add-on x 0.5 MF x 1.4 Alpha 280%
Unmargined
Two completely
0.2% 0% Transactions are long/short in same 0%
offsetting 5Y IR Swaps (0.5% x 40% due hedging set hence completely net Add-ons
to zero NGR)
Two 6M FX Forwards 0.4% 3.96% 4% Add-on x 0.71 MF x 1.4 Alpha 990%

(USD/EUR + GBP/USD) (1% x 40% due Transactions are in different hedging sets
with opposite MtMs to zero NGR) (based on currency pairs) and hence
add-ons don’t offset
This additional complexity may lead banks to delegate the

The methodological differences between the CEM and
responsibility for the regulatory EAD calculation from the
SA-CCR regulatory approaches have a significant effect
finance department to the risk department. In particular,
on the exposure amounts calculated for some common
banks may wish to leverage their investment in an SA-CCR
transaction types (Table 2). For example, the regulatory
calculation engine by applying it to their internal
exposure calculated for an unmargined 5-year interest
measurement of exposures against limits. This would have
rate swap will be six times higher under the new SA-CCR
the advantage of addressing the reconciliation challenge
approach.
between the finance and regulatory numbers, on the one
The regulators intend to replace the CEM with the SA-CCR hand, and internal exposure numbers on the other.
on 1 Jan 2017. Banks will have to implement the new measure
because it plays a central role in the Basel capital adequacy
framework. Even banks that are IMM-approved will need Adjusting SA-CCR for internal risk
to implement the SA-CCR for certain purposes, e.g., for management
reporting “large exposures” and for calculating the
regulatory leverage ratio. Banks embarking on the implementation of the SA-CCR will
soon realise that there are some aspects of the measure
that make it ill-suited for internal risk management and limit
utilisation.
Whilst it is likely that many banks will embrace the general

Capital is ever more important to methodology, as it is more risk sensitive, they will need to
bank management. Keeping internal make some adjustments to reflect their internal needs and
approaches to risk measurement and their credit risk policies. Here are some aspects of the
risk control in line with regulatory SA-CCR that may require revision as the approach is
capital can only be considered a adapted to support internal risk management.
positive move.
Time banding
Conceptually, the regulatory “PFE Add-on” percentages
represent the one-year expected exposure point (weighted
by the square root of time for transactions of less than one
year). The EAD is a single number, with no time profile
attached to it. Some banks may therefore wish to refine
Internalising SA-CCR this treatment to allow for:
Banks that currently use the CEM for both capital adequacy ●● A scaling of (single number) exposures to reflect a
and internal risk management purposes need to consider transaction’s residual maturity.1 This could be achieved
upgrading their internal risk measurement approach to by applying a square root of time multiplication to both
keep in step with the latest regulatory mandate. short-term and long-term exposures.
Capital is ever more important to bank management. ●● Multi-date exposure profiles reflecting the true dynamics
Keeping internal approaches to risk measurement and risk of potential future exposures over time. For example,
control in line with regulatory capital can only be considered “root-t” profiles for forward transactions and “humped”
a positive move. Furthermore, having to support different profiles for swap transactions (Figure 1a-b); and of course
logics for the various methodologies will introduce additional portfolio profiles that reflect exposures rolling off as
support and maintenance overheads for the bank, and transactions mature.
mean that more time will be spent explaining and reconciling
numbers. Converting from an expected to potential loss measure
There are two key issues here:
Banks that use a different MtM + Add-on method should,
in our view, also consider upgrading to an SA-CCR type of ●● Weighting factors, i.e., add-on percentages, may need
methodology, because the SA-CCR approach may be more reviewing to represent potential future exposure at a given
risk sensitive and a consolidation of methodologies offers confidence interval, rather than expected exposure.
operational benefits.
●● Banks may wish to ignore the 1.4 Alpha multiplier and
It should be noted that the new SA-CCR calculation requires instead calibrate PFE weighting factors to a 97.5 percent or
significant system enhancements. The “hedging set” concept 99 percent confidence level. We see no point in multiplying
will require access to additional trade data in order to the MtM component of the exposure by 1.4 (as is the case
identify the underlying risk factor that determines the right in the SA-CCR formula) if the PFE component is sufficiently
set for each transaction, the direction of the trade relative conservative.
to that risk factor, the delta of options, and so on.
4 Counterparty Credit Risk Beyond Basel
Figure 1a: Root-t exposure profile, e.g., forward transaction
16
Time-banded profile
14.1 Scalar (SA-CCR)
14 13.2
Root(t) profile
12.2
12 11.6
10
10
Exposure
8.4
8
7
6
5
2
0
0
Today 3m 6m 9m 1Y 15m 18m 21m 2Y
Time
Figure 1b: Humped exposure profile, e.g., interest rate swap Time-banded profile
Scalar (SA-CCR)
IRS Profile
16
14
12
10
10 9
8.6
Exposure
8
8
7 6.8
6 5.6
5
3.6
4
2
0 0
0
Today 6m 1Y 18m 2Y 3Y 4Y 5Y
Time
Completeness and accuracy for risk measurement Conclusion

The SA-CCR weighting factors lack granularity, having
been placed on broad asset classes such as interest rates, We believe that many banks that are not currently applying
foreign exchange, equities, etc. Policy-makers may wish to advanced models could benefit significantly from moving
implement more granular weighting factor matrices, e.g., towards an SA-CCR inspired system for measuring credit
by classifying underlying risk factors such as interest rates, exposures for internal risk management purposes. However,
exchange rates, equities and commodities into volatility they may need to adapt the SA-CCR methodology in the
buckets. ways that we highlighted above.
In addition, the SA-CCR does not apply to spot or cash From a systems perspective, it is therefore important that any
transactions with a market-standard settlement period. implementation of the SA-CCR methodology should retain
Some banks may still wish to recognise that a small amount some flexibility in its structure and parameters. This will allow
of pre-settlement risk attaches to such transactions, the bank to use the same underlying systems infrastructure
especially if they do not have settlement limits and/or for parallel and methodologically consistent calculations of
wish to impose some volume control on such transactions. counterparty exposures, whether for regulatory capital,
financial reporting, limit utilisation, or internal risk
Contentious or overly conservative parameters management purposes.
The rather high “supervisory volatilities” imposed by the
SACCR standards have the perverse effect of privileging
put options, so banks may wish to apply a different set of ABOUT THE AUTHOR
volatilities in the calculation of option deltas.
Some other parameters used in the SA-CCR may also Jean-Marc Schwob is product manager, credit risk, for
prove contentious and hence could be modified or FIS’ Adaptiv. In this role, Jean-Marc is responsible for
removed by an internal risk management policy: the overall functionality of Adaptiv Credit Risk, a leading
credit exposure measurement, management and control
●● The 0.05 floor used in the “multiplier” formula that is
solution. This includes determining industry and regulatory
applied to netting sets with negative MtMs or that feature
trends, future product direction, documenting current and
over-collateralisation may be removed. The effect of
proposed system functionality, ‘product specialist’ support
the floor is to always record positive exposure, even for
for sales, plus generating global awareness and interest
transactions that are significantly out-of-the-money
in the product.
or netting sets that are heavily over-collateralised.
In particular, banks may wish to recognise that a portfolio
of sold options should not result in positive exposure.
●● The 1.5 multiplier applied to the PFE Add-Ons of
margined exposures2 may not be justifiable under
a risk management policy. We believe a conservative
application of the “margin period of risk” rules should
be the correct treatment. We believe that many banks that are
not currently applying advanced
Simplifications models could benefit significantly
Due to the challenges involved in the delta-weighting of from moving towards an SA-CCR
option positions, banks may wish to: inspired system for measuring
●● Simply apply a delta of (1) or (-1) to all options. credit exposures for internal risk
management purposes.
●● Simply apply a delta of (½) or (-½) to all options.
●● Or use delta values calculated by front-office systems.
2
THIS MULTIPLIER HAS THE PERVERSE EFFECT OF CAUSING THE PFE OF A SHORT-TERM
(≤ 10 DAYS) MARGINED TRANSACTION TO BE HIGHER THAN THE SAME PFE IF IT WERE
UNMARGINED. INDEED MARGINED PFE ADD-ONS ARE MULTIPLIED BY 1.5 AND THE
1
THE SA-CCR METHODOLOGY DOES INCLUDE A SCALING BASED ON THE “SUPERVISORY SQUARE ROOT OF THE “MARGIN PERIOD OF RISK”, WHEREAS UNMARGINED
DURATION”, I.E., THE TENOR OF THE UNDERLYING RISK FACTOR, FOR INTEREST RATE TRANSACTIONS ARE MULTIPLIED BY THE SQUARE ROOT OF THE RESIDUAL
AND CREDIT DERIVATIVES. THIS IS PRESUMABLY BASED ON THE INSIGHT THAT LONGER MATURITY. ADMITTEDLY, THIS EFFECT IS NEGATED BY THE FLOORING OF MARGINED
TENORS HAVE A HIGHER VOLATILITY THAN SHORTER TENORS. THIS IS HOWEVER EXPOSURES TO THE OTHERWISE UNMARGINED CALCULATION, BUT IT IS IN OUR VIEW
DIFFERENT TO A (SQRT-T) WEIGHTING BASED ON THE DURATION OF THE EXPOSURE. UNNECESSARY AND INCONSISTENT.
About FIS’ Solutions for Risk Management
FIS’ solutions for risk management cover pre- and posttrade
risk management; integrated, enterprise-wide market,
liquidity, credit and operational risk management; asset
liability management; and trade surveillance. These
solutions can be used across trading and clearing platforms
and around multiple asset classes to help organizations
better understand their exposure, improve the visibility and
understanding of risk across the enterprise, and comply with
regulations globally. FIS’ customers include banks, broker-
dealers, securities firms, clearinghouses, hedge funds,
pension funds, asset managers, insurance companies,
corporations and government entities of varying sizes,
geographical locations and organizational complexities.

About FIS
FIS is a global leader in financial services technology,
with a focus on retail and institutional banking, payments,
asset and wealth management, risk and compliance,
consulting and outsourcing solutions. Through the depth
and breadth of our solutions portfolio, global capabilities
and domain expertise, FIS serves more than 20,000 clients
in over 130 countries. Headquartered in Jacksonville,
Florida, FIS employs more than 55,000 people worldwide
and holds leadership positions in payment processing,
financial software and banking solutions. Providing software,
services and outsourcing of the technology that empowers
the financial world, FIS is a Fortune 500 company and is
a member of Standard & Poor’s 500® Index. For more
information about FIS, visit www.fisglobal.com
www.fisglobal.com twitter.com/fisglobal
getinfo@fisglobal.com linkedin.com/company/fisglobal
©2016 FIS
FIS and the FIS logo are trademarks or registered trademarks of FIS or its subsidiaries in the U.S. and/or other countries.
Other parties’ marks are the property of their respective owners. 1158
Counterparty Risk and CVA Survey
Current market practice around
counterparty risk regulation, CVA
management and funding
February 2013
Contents
Preface 1
Executive summary 2
Glossary 4
Survey methodology 5
Introduction 6
Survey findings 8
1. Regulation 8
1.1. Overview 8
1.2. Exposure modelling approach 8
1.3. Collateral modelling approach 9
1.4. From Basel II to Basel III 10
1.5. Central counterparties 11
1.6. Modelling 12
1.7. Technology 16
1.8. Backtesting and validation 16
1.9. Stress testing 19
1.10. Wrong way risk 21
2. CVA 23
2.1. Overview 23
2.2. Platform description 23
2.3. CVA modelling 24
2.4. Calibration 26
2.5. Implementation 28
2.6. Incorporation of risk mitigants 28
2.7. Hedging 30
2.8. Return on capital 35
3. Funding and valuation 36
3.1 Overview 36
3.2. OIS discounting 36
3.3. Collateral value adjustment 37
3.4. Unsecured funding (FVA) 38
3.5. Organisation 40
Conclusion 41
Contacts 42
Preface
It is with great pleasure that we present this Counterparty Risk and CVA Survey, the result of a collaborative effort
by Deloitte and Solum Financial Partners. Counterparty risk management has been a key area of focus for financial
institutions over the past few years, and the aim of this survey is to take stock of the industry’s response to the
numerous theoretical issues and operational challenges raised as a result of the evolving regulatory, accounting and
risk management environment.
We would like to express our thanks to the institutions and individuals who participated in the survey. The time and
dedication put in by the respondents in articulating their views was a key contributing factor to its success.
We trust you will find this survey topical and insightful, and we hope the contents will help you navigate this rapidly
changing environment.
Tim Thompson Vincent Dahinden

Partner, Risk & Regulation Chief Executive Officer
Deloitte LLP Solum Financial Partners LLP
Counterparty Risk and CVA Survey 1

Executive summary
Counterparty risk is a topic which has been elevated

to the forefront of the front office, risk management
and regulatory agendas following mark-to-market
volatility and defaults over the global financial crisis.
Universal acknowledgement of credit valuation adjustment (CVA) and debt valuation adjustment (DVA) as essential
components within the fair-value of derivatives and securities financing transactions has reinforced the importance
of counterparty risk management across a much broader spectrum of financial services firms. As a result, banks
are facing a much stricter regulatory environment, the impact of which will have far-reaching implications for the
way they manage their counterparty credit risk (CCR) through CVA and how they ensure that they are generating
sufficient return on capital. There are additional requirements on financial reporting under revised international
accounting standards. Finally, the uncertainty in the international financial markets has also resulted in sizeable
increases in the cost and scarcity of funding available to banks.
Since the previous survey conducted by Solum Financial Partners in 2010 there have been significant changes to
the regulatory framework governing financial institutions, and we see such supervisory considerations permeate
almost every area of the survey responses. We have adopted an approach that provides three different analysis
perspectives: a regulation point of view, a CVA standpoint and finally a focus on trading and valuation challenges
related to counterparty risk modelling. The first part of the survey in particular focuses on the implementation
challenges associated with the new regulations, and how respondents are managing the capital cost and the
operational and methodological challenges of transitioning to the new regime.
The forthcoming Basel III revisions to the counterparty risk capital standards represent a meaningful departure from
the existing regime, and the introduction of CVA VaR will materially increase the capital held against bilateral credit
exposure.
The survey found that the perceived capital savings that could come from leveraging the advanced CVA approach
is incentivising a new set of respondents to pursue advanced ‘internal model method (IMM)’ approval from their
respective supervisory bodies and existing IMM banks to expand their product coverage.
The introduction of a low risk weight to central counterparties (CCPs) will force banks to hold capital for exposures
to CCPs which was not required before and would require the modelling of exposures to CCPs as well as default
fund contributions. Emerging securities markets legislation which is designed to mandate the use of CCPs for
standardised derivatives and requires robust margining for bilateral trades, has placed renewed emphasis on banks’
ability to model collateralised exposure.
The ability to model collateral has also come under regulatory scrutiny with Basel III introducing additional
conservatism into the so-called shortcut method, on which a quarter of those IMM banks surveyed were reliant.
The responses revealed that banks have a considerable way to go in this space, with a large majority of the
respondents unable to perform full collateral modelling over the entire duration of the trade, and fewer still
capturing other credit support annex (CSA) specific features, FX mismatches or price variation in non-cash collateral.
There is however, an acute awareness amongst those surveyed that this is fast-becoming an urgent priority in order
not only to allocate capital efficiently, but also to price these instruments correctly.
2
The valuation challenges presented by collateral agreements were explored within the survey, especially as
consensus is emerging amongst practitioners for the need to move away from LIBOR discounting for secured
funding trades – and in fact survey responses indicated the overwhelming majority of participants are moving
towards overnight index swap (OIS) discounting. A smaller, but growing subset of those respondents also
commented that they had the capability to capture the optionality associated with multi-currency CSAs within the
discount rate.
It is however not just collateralised exposures for which participants have recognised the need to integrate more
closely the funding costs and benefits into pricing. Such considerations are encapsulated within what is known as
a funding valuation adjustment (FVA) for their uncollateralised equivalent; a theme explored throughout the survey.
Virtually all participants acknowledged the necessity of such an adjustment, even if the accounting standard setters
appear to be less convinced. Furthermore, the majority of respondents already claim to charge for FVA at the trade
level and charge it to the relevant trading desks, analogous to CVA and DVA. That said, the extent to which all
three components can be simultaneously incorporated within the fair-value and in what proportion, is something
which is still the subject of much debate and academic interest.
The widespread acknowledgement that such considerations materially impact the price, must then necessitate an
integrated framework within which banks can adequately risk manage their exposure to each component. The final
part of the survey explores the operational and organisational challenges faced by banks and looks at how they are
overcoming such difficulties and implementing solutions within the context of their own operations.
What is clear is that the regulatory, accounting, front office and risk-management perception of counterparty risk
has changed dramatically in recent years, bringing to the forefront new technical challenges for banks. In particular,
areas such as OIS discounting, collateral optimisation and funding have become increasingly important. This survey
is designed to capture market practices in these new areas, and in particular to highlight the heterogeneity in how
these risks are measured, managed and mitigated given the unique set of organisational constraints specific to
each participant.
Despite having much more clarity as to the final form and substance of the emerging banking and securities
markets regulations, and the fact that banks are further advanced in developing their CVA risk management
capabilities, future trends remain very hard to predict. Certainly, we expect CVA, DVA and FVA to remain at the
forefront of the risk, regulatory and accounting agenda for some time to come.
What is clear is that the regulatory, accounting, front

office and risk-management perception of counterparty
risk has changed dramatically in recent years, bringing
to the forefront new technical challenges for banks.

Glossary
AMC American Monte Carlo
BIS Bank for International Settlements
CCDS Contingent credit default swap
CCP Central counterparty
CCR Counterparty credit risk
CDS Credit default swap
CEM Current exposure method
CollVA Collateral valuation adjustment
CSA Credit support annex
CVA Credit valuation adjustment
DVA Debt valuation adjustment
EAD Exposure at default
EEPE Effective expected positive exposure
EPE Expected positive exposure
FVA Funding valuation adjustment
HJM Heath Jarrow Morton (model)
IFRS 13 International Financial Reporting Standard 13 ’Fair Value Measurement’
IMM Internal model method
LGD Loss given default
LMM LIBOR market model
MTM Mark-to-market
OIS Overnight index swap (rate)
OTC Over-the-counter
PD Probability of default
PFE Potential future exposure
P&L Profit and loss
RWAs Risk-weighted assets
SCSA Standard credit support annex
VaR Value at risk
WWR Wrong way risk
4
Survey methodology
This survey has been conducted jointly by Deloitte and Solum Financial Partners. The survey examines the
approaches used to manage CCR in light of the financial crisis and increased regulatory focus covering CVA, DVA
and FVA. We surveyed 21 banks in 2012 and their responses were given as a current state of the situation that
existed at that time. Subsequent changes may have occurred.
This survey report is based solely upon the responses received from the participant banks. Not all participants have
provided the same level of detail in relation to all sections and questions. In addition, the participants represent
a wide cross-section of the industry and, as such, the extent and granularity of their responses will be limited by
the extent of their operations.
The approach involved having each of the participating banks complete the survey. In some instances follow up
interviews were conducted for consistency and completeness. The answers were anonymised and analysed for
key trends.
Within the survey the number of banks represented can be broadly described in two ways. The first are those
banks who already have much of their CVA infrastructure in place in terms of models, systems, CVA desks and
regulatory approvals. These banks are focusing more on enhancing their capabilities across FVA, CVA hedging and
capital optimisation. The second group of banks are in the process of developing their CVA infrastructure with
respect to accounting rules, trade pricing, CVA desk setup and obtaining advanced regulatory approval.

Introduction
There continue to be significant shifts in the financial landscape as a result of increased regulatory scrutiny and the
tougher operational environment for banks. The extent of change is evident when comparing results of this survey
to the one carried out by Solum Financial Partners in 2010. The scope is broader primarily as a result of the growing
importance of CVA in light of accounting requirements and Basel III capital rules. The survey questions were
designed to span a broad spectrum of topical issues, including how banks are positioning themselves ahead of the
revised Basel III counterparty risk requirements, CVA pricing and risk-management solutions; and their integration
within the existing architecture, valuation challenges for collateralised counterparties and the incorporation of
funding costs. Before analysing the results, we first consider the key background areas and themes that are the
subject of this survey.
Accounting
International Financial Reporting Standard (IFRS) 13 ‘Fair Value Measurement’ is effective from 1 January 2013.
It is based largely on the accounting standard applied in the U.S. One of the aims of IFRS 13 is to harmonise the
definition of fair value and in doing so harmonise the approaches to determining fair value in accounting.
Fair value is characterised as an exit price, which is described as the price that would be received or paid in an
orderly transaction between market participants. An important but complex component of fair value is the CVA
(and DVA).
There appears to be market consensus that the reference to an exit price in the accounting standards will
necessitate a move from historically-based to risk-neutral (market-implied) parameters in CVA quantification. This
is very significant in terms of default probability estimation. Whilst many large banks have for a number of years
used market implied default probabilities to calculate their CVA, this practice has been less common in smaller
banks that have not been subject to the U.S. accounting standard, FAS 157 (generally those domiciled outside
the U.S. and Canada). A natural consequence of the remaining banks moving to risk neutral CVA is that overall
accounting CVA numbers will be significantly higher and more volatile. This is due to the well-known existence
of a significant risk premium within a credit spread, making the proportion of risk-neutral default probabilities
significantly larger than real world ones, especially for high quality ratings.
The CVA profit and loss resulting from the systemic component in a credit spread can be essentially offset with
the analogous component within a bank’s own credit spread. This latter component is contained within the DVA
component which is also a requirement of IFRS 13. IFRS 13 requires an institution to account for the fair value of
the non-performance risk (also referred to as the entity’s own credit risk) of their liabilities. Some banks question
the use of DVA as it implies they profit from their own declining credit quality and leads to hedges which may
create wrong way and systemic risk. Other banks see DVA as a completely logical component, alongside CVA,
which can be monetised (albeit with some difficulty). Some banks see DVA more as a funding benefit and
therefore the links between DVA and funding must be considered carefully.
Regulatory capital
The first version of the Basel III capital requirements had a large focus on CCR and CVA, and left little doubt that
the associated capital requirement needed to be substantially increased. It explicitly mentioned that essentially
two-thirds of the risk, due to CVA volatility, was not capitalised at all. The Basel Committee introduced the concept
of a new capital requirement for CVA VaR which makes a clear reference to credit spreads as the driver of default
probability in the CVA formula. Under Basel III, this risk-neutral default probability requirement is explicit. It should
also be noted that, although DVA is an accounting requirement under the fair value measure, the benefit arising
from it must be removed from Tier 1 equity and is therefore not allowable in quantifying capital requirements
under Basel III. This represents a double blow as Basel III forces the use of comparatively high risk-neutral default
probabilities without giving the associated benefit of own default risk. Furthermore, Basel III does not consider
market factors other than credit spreads (for example interest rates and FX rates) which limits the scope for
potential capital relief through hedging.
Basel III gives two possible frameworks for the calculation of CVA VaR: the standardised and the advanced. The
framework used depends on whether a bank currently has IMM and specific interest rate risk approval for bonds.
Capital relief is given for hedging with single name and index credit default swaps (CDS) and it seems that Basel III
is intending to push banks to hedge their CVA credit component where possible.
6
This is potentially controversial as the CDS market is not particularly liquid for all counterparties, and it is not
clear to what extent banks hedging their CVA relating to illiquid counterparties with credit indices represents a
reasonable form of risk transfer. Furthermore, the more straightforward CVA related underlying asset hedges may
actually consume, rather than reduce, capital. The unintended consequences of CVA hedging have already created
problems in terms of market instability such as in spiralling sovereign CDS spreads driven by CVA desk hedging.
This, together with the need to reduce CVA VaR charges for sovereign exposures (resulting from interest rate
hedging of large debt issuance), has led to an exemption in Europe for sovereign CVA VaR (under CRD IV covering
the implementation of Basel III capital rules). A further exemption for European non-financial counterparties is also
under consideration. Possible capital relief achieved through other hedging strategies, such as that provided by
synthetic securitisation for example, is another possibility for potentially improving efficiency.
Implementing changes in capital rules will clearly represent a very significant cost for banks (and therefore their
clients). However, the complexity of capital methodologies, together with the uncertainty around specific rules and
possible exemptions, makes the overall magnitude of this hard to gauge.
Alignment of front office, accounting and regulatory practices

Within a given bank, there can exist multiple definitions of CVA. The most obvious examples are accounting CVA
(for books and records), front office CVA (for pricing new transactions) and regulatory CVA (for defining capital
requirements). This is particularly important to consider as misalignment between CVA definitions can lead to
inappropriate trading decisions, incorrect assessment of risk and mismanagement of capital. For example, if
accounting and front office CVA definitions do not match then apparently profitable trades may not appear that
way to shareholders, and profit & loss (P&L) volatility as seen by a CVA desk may not be equivalently represented
in earnings volatility. Another example would be that if front office and regulatory CVA were misaligned then a
reduction in capital may increase CVA volatility and vice versa.
Whilst accounting standards and regulatory capital rules appear likely to create more uniformity over CVA
quantification (for example by use of risk-neutral parameters such as credit spreads), they also create ambiguity (for
example in terms of DVA benefit). It is therefore not clear how rapid and complete the convergence will be, and to
what extent a bank should attempt to align these calculations.
CVA, DVA, funding and risk-free valuation

Since CVA and DVA should adjust the non-credit risk value of a trade or portfolio, it is crucial to determine the
correct way to perform a benchmark risk-free valuation. In recent years, the significant rise in short and long term
funding rates has seen attention placed on both risk-free valuation and funding costs. LIBOR rates, previously
seen as a close proxy for risk-free rates, are now seen as inadequate discount rates due to their credit and funding
component divergences with respect to both tenor and cross currency basis effects. This has driven the need to
use dual curve, or OIS discounting (at least for valuing collateralised derivatives). There has been a trend to switch
to these more sophisticated valuation methods, led by CCPs and banks. Related to this discounting issue there is
a need to account for currency and type of collateral posted under the CSA (or other) agreement and ideally the
optionality inherent in collateral posting requirements and substitution rights.
The financial crisis has driven short-term rates such as LIBOR away from benchmark risk-free rates. Additionally,
banks are being required to rely less on short-term funding and more on longer-term, more costly borrowing.
These aspects have led to the notion of FVA due to the need to assess funding costs and benefits in the valuation
alongside other elements such as CVA and DVA. There is controversy over whether or not FVA should form a
component of pricing and also to what extent it overlaps with the existing notion of DVA. Coupled with the fact
that there are no specific accounting and regulatory requirements governing the use of FVA, this leads to very
different treatments of funding benefits and costs.

Survey findings
It is evident that the investment in IMM programmes is

paying off as an increasing number of banks are heading
towards IMM compliance.
1. Regulation
1.1. Overview
Over the course of 2012, banks’ CCR programmes were mainly focused around obtaining IMM approval prior
to the Basel III ‘deadline’ imposed by the Bank for International Settlements (BIS), previously set to January 2013
and recently extended to later in 2013. Failure to calculate CCR exposure under IMM would have had a significant
double blow on banks from both the regulatory capital charge as well as the regulatory CVA charge: banks would
have had to calculate the CVA VaR charge under the standardised approach whilst pursuing the much-needed IMM
approval which would permit the use of the advanced CVA approach.
It is evident that the investment in IMM programmes is paying off as an increasing number of banks are heading
towards IMM compliance. Banks that are compliant with IMM do not have full coverage across their portfolios as
some exotic trades are calculated using a semi-analytical approach, and for which regulatory capital requirements
are determined based on the current exposure method (CEM).
1.2. Exposure modelling approach

About 70% of the banks interviewed are currently calculating regulatory capital associated with their CCR
exposures for at least a part of their portfolio using the CEM. However, there is continuing effort towards gaining
full IMM approval by means of organisation-wide large-scale projects, using the prescribed alpha factor in the first
instance followed by the assessment for use of their own alphas.
Figure 1. Regulatory capital calculation approach
Internal Model Method with

internally calculated alpha
Internal Model Method with

regulatory prescribed alpha
Current Exposure Method
Standardised Method
0% 20% 40% 60% 80%
Future plan Current
8
The alpha factor applied to effective expected positive exposure (EEPE) in order to capture portfolio diversification
and general wrong way risk (WWR) effects, is currently prescribed at 1.4, unless the regulator deems it necessary
to increase this factor (on a case by case basis), in which instance the regulator will provide the particular bank with
an alpha factor which it deems appropriate. Whilst not many banks have done internal analysis to assess the ‘true’
alpha associated with their own portfolios, 1.4 is deemed to be conservative and, using subjective judgement,
alpha is generally expected to be between 1.2 and 1.4.
For banks which use a combination of IMM and CEM, the proportion of trades for which exposure is calculated
using IMM is either small (less than 80%) or large (more than 95%), indicating bimodal behaviour amongst the
participants, and, potentially, the market. Interestingly, it is not necessarily the larger banks that have a greater
proportion of trades under IMM.
Figure 2. Exposures measured under IMM
100%
95% to 100%
90% to 95%
80% to 90%
< 80%
0% 10% 20% 30% 40%
1.3. Collateral modelling approach

The majority of banks are using, or planning to use, the full collateral modelling approach. About 70% of
banks apply haircuts to non-cash collateral, and just under 60% of banks consider the FX risk associated with
nondomestic currency collateral. The challenge lies in the modelling of the collateral portfolio composition, with
only 43% of banks ensuring future margin calls and postings are anticipated and incorporated in the future. Given
the increased focus on collateral management that will flow as a result of increased interaction with CCP clearing
houses, there is likely to be an increased effort to improve collateral modelling.
Figure 3. Collateral modelling approach
Full collateral modelling Shortcut method

Figure 4. Collateral model characteristics
Valuation of
non-cash collateral
Haircuts on
non-cash collateral
FX risk on cash in
different currencies
Changes in collateral
portfolio composition
0% 20% 40% 60% 80%
We questioned respondents on the common issue of the allocation of collateral between netting sets which
contain trades which are modelled using a combination of IMM and CEM approaches. The two approaches mostly
observed to deal with this issue are:
• Ensure all IMM trades are fully collateralised, using the collateral for IMM trades first and then allocating any
remaining collateral to the CEM trades.
• Allocate collateral proportionally between IMM and CEM trades, based on the absolute mark-to-market (MTM)
at day 0.
Both approaches only allocate collateral at the current time, and re-allocation of collateral across time does not
seem to occur over the life of trades belonging to that particular netting set. This is mainly attributed to system
restrictions since most banks calculate CEM and IMM exposures in different systems (or sub-systems).
1.4. From Basel II to Basel III

The implementation of projects that will ensure compliance with Basel III requirements are generally well underway,
although there is a sense of relief that the Basel III/CRD IV effective timelines for CCR have been postponed.
In order to use the advanced CVA approach under Basel III, the bank is required to hold regulatory approval for
the Specific Interest Rate Risk VaR model for bonds. As there will be a significant difference in the amount of
capital required, hedging permissions and intuitive representation between standardised CVA and advanced CVA,
internal debates as to whether a bank should use standardised CVA or advanced CVA continue. Almost 70% of the
participating banks already have Specific Interest Rate Risk VaR model for bonds approval (either partially or fully),
with those who currently do not have this approval planning to do so in early 2013.
However, the challenge lies in determining the CDS for names that do not have actively traded CDSs. The proxy
methodology to be used should be based on general industry, region and rating, which poses a question on the
derivation of this proxy CDS level. The question remains as to whether the CDS should be based on the specific
intersecting dimensions only, or whether a proxy should be considered based on an average of the industry, region
and rating, whilst ensuring appropriate representativeness when incorporating hedging. Sourcing and mapping
names to the appropriate proxy is a practicality which is proving to be unnecessarily challenging. Half of the banks
surveyed have indicated that they will be adding a specific credit risk spread to the general proxy based estimate to
account for the fact that counterparty is not traded.
10
Once the cost of the new capital charge has been determined, the costs will be passed from the bank to the client
by capturing it in the return on capital charge, with the treatment of the cost of capital being mixed between full
lifetime of the trade versus the first year of the trade. The majority of banks consider this cost at trade level, and on
a case by case basis, especially for larger trades. Additionally, trades are reviewed against hurdle rates to ensure the
target revenues are achieved, with target revenues reviewed as part of management planning to account for the
larger CVA and FVA charges. Typically, only trades that meet the hurdle rate are approved.
1.5. Central counterparties

As a direct result of the financial crisis, regulatory bodies are placing increased pressure on banks to move the
industry towards centralised clearing. Whilst such a regime has advantages and disadvantages, the integration of
the new requirements into banks will require significant effort. Since CCPs were not previously deemed to be risky,
and to optimise the portfolio exposure and regulatory capital calculations, some banks did not previously include
the trade exposure to CCPs in the overall exposure and regulatory capital calculations. Also, the differentiation
between qualifying CCPs and non-qualifying CCPs is still being embedded within some banks, both from a business
user perspective and from a systems perspective.
Following significant investments in technology projects, banks are now heading towards calculating bilateral
exposure for all CCPs and performing credit monitoring of these exposures. Margins posted are tracked and
sensitivity to CCPs is monitored.
Figure 5. CCP risk measure calculations
Regulatory Capital
Exposure
None
0% 10% 20% 30% 40% 50% 60% 70% 80% 90%
Infrastructure programmes to calculate the capital requirements for the exposures and default fund charges have
been initiated at almost 65% of participants, with a further 30% of banks planning to initiate these programmes
once the CCP regulations and requirements have been finalised.

Figure 6. Risk weight calculation programme
Yes
No, but plans are in place
No
1.6. Modelling
All banks calculating regulatory capital using IMM do so for vanilla interest rates and FX products, with just
under 80% calculating regulatory capital using IMM for credit derivatives. A third of the participating banks are
calculating regulatory capital using IMM for all vanilla over-the-counter (OTC) derivatives. Some banks model first
generation exotics under IMM for regulatory capital purposes but more complex exotics are generally capitalised
based on the exposure generated under the CEM approach.
Internal models are being used considerably more for exposure monitoring than regulatory capital calculations, in
particular for the calculation of exposure associated with more exotic derivatives.
Figure 7. Internal model for regulatory capital
Repos/ﬁnancing transactions
Equities
Commodities
Credit
Foreign Exchange
Interest rates
0% 20% 40% 60% 80% 100%
Exotic Vanilla
12
Figure 8. Internal model for exposure monitoring
Equities
Commodities
Credit
Foreign Exchange
Interest rates
0% 20% 40% 60% 80% 100%
Exotic Vanilla
All banks calculate exposure for credit risk monitoring purposes at least daily, and almost all banks use the previous
day’s trade and market data to perform this calculation. Just over 10% of banks have the capability to update
exposure in real time (as soon as the trades have been traded). Regulatory capital calculations generally occur daily
or monthly, with the calculation mostly based on the same or the previous day’s data, but some banks are going as
far back as the previous month’s data.
Figure 9. Calculation frequency
Yearly
Quarterly
Mo1thly
Weekly
Daily
I1tra-day
0% 20% 40% 60% 80% 100%
Regulatory capital Exposure mo1itori1g

Monte Carlo simulation is mostly used for the internal models, with the number of scenarios ranging between
1,000 and 10,000. The majority of banks consider a single set of scenarios across all portfolios and asset classes.
However, there are some banks that vary the number of scenarios depending on the complexity or size of the asset
class and convergence capabilities of the underlying stochastic processes. Potential future exposure (PFE) is mostly
calculated at the 95th, 97th or 99th percentiles, but some banks consider loan-equivalent exposure measures as
their PFE. Time steps are usually tighter in the near future, in particular to capture the potential effect of margining,
but become further apart over long time horizons, with time horizons varying between 30 and 50 years.
The modelling of exposure for exotic trades is done by means of a variety of different methods, ranging across:
• off-line calculations with manual upload into the risk systems;
• MTM + add-on approach;
• semi-analytic approach using approximations;
• decomposition of the trades into replicating structures of simpler, more vanilla products; and
• valuations using front office models.
When banks use the MTM + add-on approach, the add-on is either taken to be the regulatory add-on, or it is
calibrated internally using a proxy simulation and inferring the MTM from the simulated exposure.
The majority of banks generally incorporate the more traditional risk mitigants such as netting, cash, bonds and
Figure 10. Internal model calculation approach
Exotics
Commodities
Equity
Credit
FX
Interest rates
0% 20% 40% 60% 80% 100%
MtM plus add-on Semi-analytical approach Monte Carlo simulation
equities for credit risk monitoring and regulatory capital reporting.
There remains debate around the inclusion, treatment and modelling of optional and mandatory break clauses,
downgrade triggers, letters of credit and guarantees. Some banks only incorporate mandatory break clauses, and
monitor downgrade triggers and optional termination events as part of the credit risk monitoring process. Letters
of credit and guarantees may be considered on an ad-hoc basis.
14
Collateral modelling seems to focus on the modelling of margin calls and break events rather than the collateral
deterioration or improvement itself.
The parameters used in the stochastic process models underlying the Monte Carlo simulation consist of implied
and historical parameters, where banks choose to use implied parameters if these are available and can be sourced
appropriately from up-stream systems. 45% of banks re-calibrate the model parameters on a daily basis, and the majority
of banks comply with the regulatory requirement of at least quarterly calibration when using historical parameters.
Once parameters have been estimated, an impact review is performed and the results are assessed at various
methodology committee meetings, where a decision is made on whether or not to implement the recalibrated
parameters. Discussing the impact of new parameters with heads of business illustrates a strong example of
satisfying the ‘use test’ requirement as exposure calculation outputs are used more widely across the bank.
Figure 11. Calibration frequency
Ad-hoc
Annually
Semi-annually
Quarterly
Monthly
Weekly
Daily
0% 10% 20% 30% 40% 50%
About 90% of banks using historical data for the calibration of Monte Carlo simulation parameters consider at least
three years of data, with 42% of banks considering more than three years of data in order to represent an entire
business cycle.
Figure 12. Historical data series length
More than three years
Three years
One to three years
Less than one year
0% 10% 20% 30% 40% 50%

Under the new Basel III/CRD IV regulations, banks will be required to calculate an additional EEPE based on stressed
parameters, therefore requiring the inclusion of a stressed period in the calibration dataset. The definition of this
‘stressed period’ is subjective, and banks are currently defining the approaches to be taken to identify them. More
than half of the banks take the stance that, given the recent economic downturn, a stressed period is automatically
included in the last three or four years and they are therefore compliant with new regulatory requirements.
In addition to calibrating a set of stressed parameters, the bank needs to calculate two sets of exposures,
stressed and ‘normal’, and use the most conservative exposure figures to calculate the capital charge. This will
affect the run-time of the exposure calculation, and is also expected to increase exposures (and therefore capital
requirements) significantly.
1.7. Technology
More than 75% of banks surveyed use internal systems and are investing significant time and resources to migrate
multiple legacy or asset class systems into a single, all-encompassing system.
There is also a shift towards the use of integrated systems between CCR and the front office CVA systems, allowing
for increased efficiency, leveraging off a single golden source of data and enabling scenario consistency.
1.8. Backtesting and validation

Backtesting of the CCR models has proven to be challenging, with various factors making portfolio backtesting in
particular much more complicated than market risk backtesting:
• difficulties in obtaining historical data to test models over ‘sufficiently long time horizons’;
• changes in portfolio composition over long periods; and
• changes in simulation models and associated parameter calibration over long periods.
The majority of banks are in the process of implementing risk factor as well as portfolio backtesting programmes as
part of model assessment and performance monitoring.
Risk factor backtesting is either performed for the most important risk factors to which the bank is exposed or, if
banks have sufficient infrastructure in place, all risk factors are backtested.
For portfolio backtesting, a combination of complete, sample and hypothetical portfolios is generally used.
The hypothetical portfolios are selected to be representative of the book, considering asset class concentrations
in notionals, trade numbers or uncollateralised exposure, or building hypothetical portfolios representing the
complete actual portfolio based on key risk drivers, and considering collateralised and non-collateralised portfolios.
When samples of the portfolio are selected, they are chosen subjectively as key counterparties, countries or
asset classes.
16
Figure 13. Portfolio backtesting
Hypothetical portfolio
Sample of actual portfolio
Complete actual portfolio
0% 10% 20% 30% 40% 50% 60% 70% 80%
All banks consider MTM distributions when performing portfolio backtesting, although only 40% consider current
exposure and only 25% EEPE. The majority of banks backtest over multiple time horizons up to one year, with a few
banks also considering time horizons beyond two years.
Figure 14. Portfolio backtesting: risk measures considered
Effective EPE

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Figure 15. Backtesting: time horizons considered
1 year
6 months
3 months

2 weeks
1 week
0% 20% 40% 60% 80% 100%
In addition to performing the ‘quantitative’ exercise of backtesting, banks are improving governance frameworks
and processes in order to obtain more business involvement in model performance assessment.
Regulatory requirements around validation of CCR models have increased significantly following the crisis. In the
US, The Office of the Comptroller of the Currency (OCC) published the Model Risk Guidance in April 2011 which
highlights the importance of clear and solid validation guidelines, and US banks feel that they are ‘expected’ to
tailor their validation exercises to these higher standards and ensure effective challenge is in place. These guidelines
are also used at non-US banks to ensure all aspects of model risk are covered. Furthermore, the proposed Basel III
and CRD IV requirements list increased objective requirements with regards to validation of CCR models.
However, whilst banks have independent review units in place who approve models before implementation, at
initial development and on an on-going basis (usually annually), the banks interviewed feel that a good balance
still needs to be found between qualitative judgement and pre-defined criteria, and validation standards still
need to be increased.
Figure 16. Validation approach
Independent model review
Code review
Conceptual review
0% 20% 40% 60% 80% 100%
Qualitative judgement Pre-defined criteria
18
1.9. Stress testing
Another area continuously under regulatory scrutiny is stress testing, not just within the CCR trading book but
across the bank. The ‘Regulator’s expectation’ is that stress testing should not be considered a one-off, quarterly
or annual ’ticking the box’ exercise, but rather a bank-wide integrated effort illustrating that the firm continuously
considers the impact that market or macroeconomic stresses could have on its business and CCR exposure.
The majority of banks perform a combination of daily, weekly, monthly and quarterly stress tests. These range from
sensitivity tests of the key risk factors to macro-economic stress tests. Ad hoc stresses are also considered where
necessary in order to test the firm’s capability to withstand potential immediate stresses as well as the ability of the
firm’s technology to calculate the impact of ’stress on demand’.
Figure 17. CCR stress testing frequency
Ad hoc
Annually
Quarterly
Monthly
Weekly
0% 10% 20% 30% 40% 50% 60%
The majority of banks interviewed consider macroeconomic scenarios relating to the Eurozone crisis and the credit
crisis. Reverse stress testing is also under development, generally as part of larger-scale stress testing programmes.
Whilst banks aim to obtain complete coverage of the portfolio when considering stresses, there is sometimes a
small, immaterial proportion that is not covered as part of the wider stress testing programme. Monitoring of these
exposures is performed on a continuous basis in order to ensure that these segments remain immaterial and that
the exposures would not increase significantly were they subject to stresses.
The stress test performed by the CCR areas mainly focuses on the calculation of stressed MTMs, although at least
40% of banks also focus on risk-weighted assets (RWAs), current exposure and potential exposure.

Figure 18. CCR stress testing measures
RWA
PE
EPE
EE
CE
MtM
0% 20% 40% 60% 80% 100%
About 70% of banks perform stress testing of credit worthiness, through probability of default (PD) and loss given
default (LGD) when performing CCR stress testing, generally stressing point in time PDs and downturn LGDs when
performing these joint stress tests. Having said this, 64% of banks do not explicitly model the correlation between
credit worthiness and market movements when performing the stress tests; rather, the correlation is assumed to be
implied by the macroeconomic scenarios.
Reporting of stress testing results takes a varying number of degrees, with almost all banks providing at least high
level reporting to the senior board and detailed reporting and integration into the day-to-day CCR management.
Figure 19. CCR stress testing reporting
Regulator
Senior Board
Capital management
CCR management
0% 20% 40% 60% 80% 100%
Detailed High level
20
Stress testing programmes are currently integral in the development and enhancement of banks’ CCR frameworks.
Future enhancements to banks’ stress testing framework focus on an improvement in the technology infrastructure
to enable more frequent and faster stress testing, combined with additional flexibility in the specification of
scenarios. In addition, firms are considering the use of stress testing limits against certain counterparties, industries
or sectors to identify vulnerabilities and manage risk appetite to these vulnerabilities accordingly.
1.10. Wrong way risk

The interaction of WWR between the front office and the risk perspective remains an interesting debate. Whilst
WWR should be identified, monitored and controlled, it should also be accounted for in front office pricing. During
our interviews we have seen that almost all banks have processes in place to identify WWR, in particular specific
WWR as banks move towards meeting Basel III regulatory requirements.
Identification
Banks are investing in the integration of risk practices into the front office environment, by rolling out training
programmes and enforcing procedures whereby new trades should be checked for potential WWR (specific or
general).
Specific WWR trades are identified at origination by performing systematic checks between the various
counterparty and collateral entity hierarchies. General WWR is usually identified by comparing the trade type,
direction and counterparty to pre-defined general WWR scenarios, with the scenarios reviewed on an on-going, at
least annual, basis.
In addition to identification of WWR at trade origination, automated triggers are in place to report and identify any
WWR trades.
The identification of specific WWR is required prior to trade approval, and approvals are assessed on an individual
basis with sufficiently senior Risk Manager sign-off required in some instances.
Measurement
The degree to which PFE is adjusted to incorporate effects of specific WWR on exposure has varied over time, with
some banks initially making an expert-based adjustment to capture the specific WWR in the trades, as a percentage
of MTM or notional, and other banks working towards the more stringent Basel III framework whereby these Specific
WWR trades are segregated into another netting set and exposure at default (EAD) is assumed to be full notional.
A few of the banks interviewed have the capabilities to capture and measure general WWR, but are only using this
for trades strongly affected by general WWR and not as a blanket approach for all trades. Banks that do not currently
have the capability to measure general WWR are investing in the enhancement of their own or vendor systems to
enable the measurement of WWR, mainly through simpler, expert based or deterministic correlation measures but
some with more advanced modelling.
One potential driver of the focus on simpler general WWR measurement approaches rather than more advanced
modelling techniques is the introduction of the EEPE using stressed parameters under Basel III. In the interim, stress
testing has also been used to identify and measure general WWR, by jointly simulating credit spreads and underlying
risk factors and therefore linking credit worthiness and exposure.
Of banks that measure WWR, 85% measure specific WWR at trade level, whereas general WWR is measured across
various dimensions. Where banks measure general WWR at portfolio level, regional and sector dimensions are
incorporated and concentrations within these dimensions are monitored.

Figure 20. WWR measurement dimensions
Region/country
Portfolio level
Product level
Counterparty level
Trade level
0% 20% 40% 60% 80% 100%
General WWR Speciﬁc WWR
Monitoring and control

The majority of banks that consider WWR have processes in place to monitor the WWR exposure. These processes
encompass the identification of WWR at inception as discussed previously, the measurement where possible
against WWR limits, and continuous reporting of specific and general WWR trades.
Approvals and policies are in place to limit specific WWR, and positions are closed-out if specific WWR limits are
breached. During the pre-approval process, trades are reviewed against the specific WWR limit and if there is no
longer appetite the trade will not be executed. For approved specific WWR it is expected that a limited appetite
and structural mitigations such as reduced tenor, enhanced collateral requirements and minimum credit risk rating
requirements are in place.
WWR risk limit management is supported by regular reporting, ranging from high-level regular management
reporting to detailed daily reports listing the trades leading to specific WWR exposure.
22
2. CVA
2.1. Overview
As expected, all banks surveyed incorporate CVA into Fair Value for IFRS purposes. In terms of portfolio
stratification, the vast majority of the banks’ CVA is driven by interest rate trades with interest rate swaps obviously
being the most significant product type. This is followed by foreign exchange (dominated by FX forwards and cross
currency products) and credit derivatives (single name and index CDS). Equity derivatives, commodities and exotics
tend to have less significance in driving the total CVA. The contribution to overall CVA is driven by several factors,
with high notional amounts (e.g. interest rate swaps), long-dated trades (e.g. cross currency swaps) and overall
complexity (e.g. credit derivatives) being most significant. Exotic products, even in the large banks, do not tend to
make up a large part of the overall CVA but this is balanced by the inherent problems involved with dealing with
exotics in a reasonably efficient manner.
Figure 21. Total CVA by asset class
70%
64%
60%
50%
40%
30%
19%
20%
10% 6% 5% 4%
2%
0%
Interest Foreign Credit Exotics Commodities Equity
rates exchange derivatives derivatives
Collateralised trades, often ignored or modelled with very favourable assumptions made in relation to collateral
receipt, are increasingly appreciated as contributing significantly to the overall CVA bottom line. This is largely
driven by an appreciation that the margin period of risk can be material and much longer than the contractual
collateral call frequency under a CSA (often daily). Other important components accounted for are the
imperfections of collateral agreements (thresholds etc.) and the quality of the collateral itself. Banks are tending to
accept that whilst collateral reduces CVA by a significant amount, even well collateralised portfolios have a CVA
reduced by a low single digit multiplier.
There is a growing understanding of the future impact of central clearing. Whilst CCPs apparently remove CVA as
an issue as they are unlikely ever to (or to be allowed to) fail, banks are seeing their exposure to the default fund of
a CCP as representing a complex CVA with respect to the other clearing members, and are seeking to quantify such
exposure. Furthermore, the funding cost of clearing trades, due to initial margin for example, both a bank’s own
and that of clients, is being assessed.
2.2. Platform description

As the move to reduce risk and manage banks’ balance sheets and profitability intensifies, it is not surprising
that the survey showed that 80% of participants who have a CVA desk have this set up as a hedge centre (risk
mitigation), with the remaining CVA desks set up as a profit centre (risk taking). However, the distinction between
these two types of setup is not completely clear. For example, even hedge centre CVA desks with a zero P&L target
will have reasonable discretion with respect to hedging choice, which amounts to taking proprietary positions.

For those banks that charge for CVA/DVA, the majority charge at inception with the remaining performing
some form of on-going reallocation process. Reallocation is obviously difficult to manage as most trades have a
profitability which is very dependent on the CVA and so not knowing this value at inception can lead to incorrect
pricing and the potential to experience some form of winner’s curse. Subjectivity does enter into trade pricing to
some degree. One example is giving a reduction for the first trade with a client and under-pricing certain trades
(e.g. long-dated trades) on the assumption that the associated CVA loss will be compensated for via other trades
with the same client. In addition, banks often incorporate various assumptions regarding trade lifetime in terms of
aspects such as break clauses, restructurings and unwinds, especially in terms of defining the cost of capital.
Traditional counterparty risk mitigation methods such as credit lines are not made obsolete by the existence of
a CVA desk. CVA desks generally have a front office alignment and a CVA charging mechanism will naturally
incentivise more concentrated positions so as to extract maximum benefit from netting agreements. Credit lines
have a risk management focus and encourage a maximisation of portfolio diversification rather than netting.
Despite the apparently complimentary roles of credit lines and CVA, some banks (especially the more sophisticated
ones) rely less on credit lines as a result of active CVA management. Indeed, 40% of participants confirmed that the
existence of an active CVA trading desk affects credit risk monitoring. One example of this is the concept of a liquid
single name book where there is credit line relief or benefit for hedging the credit risk.
2.3. CVA modelling

It is clear that the sharing of models and systems between front office and risk management is not particularly
common with a significant proportion of banks not even planning such a convergence. This may at first glance
appear unusual due to the potential for duplication of effort. However, it is important to note that front office
CVA and risk management counterparty risk models have very different key requirements. Front office CVA models
need to be accurate and extremely fast (to support real time pricing and a high volume of sensitivities and scenario
analysis) and often have more complex underlying calibrations. Balancing this, front office CVA normally focuses on
a relatively small sub-population of the total portfolio with short-dated and collateralised trades, and trades with
high quality counterparties, often ignored.
Conversely, risk management and regulatory models have to support an extremely large trade population
irrespective of the perceived risk of those trades (upwards of 95% of the trade population). However, such
approaches do not need to have the same level of model sophistication, calibrations may be more straightforward
and they do not give rise to the same intensity of computation in terms of both time and volume of calculation.
The figure below shows the stratification amongst participants with respect to the sharing of the same exposure
models for CVA and CCR.
Figure 22. Exposure models sharing for CVA and CCR
Shared
Not shared but plans to

share in future
Not shared and no plans to share
24
Almost 80% of participants confirmed that they utilise Monte Carlo techniques for simulating exposures. The range
of Monte Carlo paths varies from 1,000 to 100,000 with sometimes, by necessity, fewer paths used for calculating
the many required greeks. The number of time-steps also varied widely amongst participants, but the maximum we
observed was 200. The number of parameters chosen generally fulfils a need to run all calculations in an overnight
batch although we noted that two banks calculated greeks only on a weekly basis. In certain institutions, the
number of paths and time-steps were contingent on certain factors, most notably the nature of the counterparty –
for example, the more complex portfolios with liquid counterparties would attract a higher investment in time-
steps and paths compared with illiquid counterparties with few vanilla trades. Collateralised counterparties will
also typically require a greater number of time-steps to account for the relevant margin period of risk. All banks
utilising Monte Carlo techniques for simulation believed that satisfactory convergence for CVA exposure and greeks
purposes is achieved.
We have found that CVA modelling varies substantially in terms of sophistication. This level of sophistication, not
surprisingly, is driven by the size and complexity of the bank’s OTC derivatives portfolio. Whilst some banks believe
that the simplicity of their underlying portfolio does not warrant very sophisticated modelling, others believe that it
is important to have complex models capturing curve dynamics and volatility behaviour.
In terms of interest rate models, both short-rate, Heath Jarrow Morton (HJM) and LIBOR market model (LMM)
approaches are used where the greater complexity of a non-Markovian approach such as LMM may be rationalised
by the benefit of sophistication in terms of calibrating to volatility, and pricing exotics.
Other asset classes follow along the same lines, with some banks favouring simplistic Black Scholes approaches and
others making more attempts at including effects such as mean reversion and calibrating more fully to volatility
surfaces. Front office focused implementations tend to be more sophisticated compared to those with a risk and
regulatory aim.
Whilst CVA systems are becoming more advanced, implementations are still required to make a number of
shortcuts so as to not require excessive computational resources. One example of this is that less than 40% of
participants use the same revaluation model for the CVA calculation and the main trading system. Additionally,
effects such as stochastic volatility, that have for many years been a part of exotic derivatives valuation, are still
seemingly too complex to incorporate in CVA modelling approaches.
There is also a divide within the overall simulation approach. The majority of banks use their own pricing models for
revaluation within their counterparty risk simulation. Whilst this approach is fundamentally inconsistent in terms of
approach, it does provide time zero pricing consistency, is probably the simplest approach and can be implemented
in a piece meal fashion. Some banks rely instead on a generic optimised American Monte Carlo (AMC) (for example
Longstaff-Schwartz approach) which requires quite a significant up-front implementation cost and can produce
divergent time zero pricing. However, the internal consistency of this approach, the fact that exotic and path
dependent products are better represented, together with the ability to produce faster valuations and sensitivities,
may be viewed as an overall benefit, especially for banks with more complex portfolios.
Correlation between and within asset classes is generally handled within a historically based correlation approach
with differences existing in length of time series used. Only 31% of banks model general WWR within their CVA
calculation although several more note the future intention to do this.

2.4. Calibration
The following charts show how participants calibrate model dynamics. This is broadly split into historical (real
world) and market implied (risk-neutral) measures. In market risk terms, it is volatility, correlation and other model
parameters that are important. On the credit risk side, this relates to the calculation of default probability and
recovery rates.
In relation to counterparties whose CDS trade in the market, most of the banks surveyed imply the PD from
the applicable observable CDS level. With the exception of some smaller banks that use an internally derived
credit spread which is generally based on a historical rating-based default probability added to a risk premium
component. Most of the banks (especially the larger banks) mark recoveries for liquid counterparties in a consistent
way to the CDS and bond recoveries on the relevant credit trading desks. These standard recoveries are frequently
adjusted for those counterparties where the bank is ranked senior in the waterfall (for example, where they hold
additional security) compared to the senior unsecured level of the comparable CDS. Again, the exception is some of
the smaller banks that mark to an internally derived recovery.
As noted above, the use of historical default probabilities for illiquid names seems to be declining driven by future
IFRS 13 accounting rules and Basel III capital requirements. It is therefore interesting to look at the ways in which
banks calculate a spread-based (risk-neutral) PD for counterparties which do not trade with sufficient liquidity in
the market and cannot therefore be derived directly from a CDS price or suitable alternative. The results indicate
that the majority of banks map to tradable CDS primarily by way of credit rating and then may take into account
geography, followed by industry. This is not surprising as mapping by rating, industry and geography is quoted as
the way to define spreads in line with the Basel III Advanced Method CVA VaR. Nevertheless, mapping via indices is
also used by a proportion of banks, all of which are engaged in active CVA credit hedging.
Figure 23. Probability of default mapping for illiquid counterparties
Mapping to tradable CDS –

categorised by rating

categorised by geography

categorised by industry
Internal spread
Index
0% 10% 20% 30% 40% 50% 60% 70%
The marking of recovery rates is another key issue. The way that banks mark recoveries on illiquid names is
typically in line with the way they mark recoveries on the liquid name population. It should be noted that,
together with the marking of the PD under the advanced CVA method under Basel III, the new capital
requirements for CVA VaR refer to the market assessment of recoveries, rather than an internal estimate.
We asked the participants how often they remarked curves for CVA, DVA and LGDs. As expected, the majority of banks
remark their CVA curves on a daily basis, DVA on a daily, weekly, or monthly basis, and LGDs on an ad-hoc basis. Many
banks have a regular review system in place to facilitate timely reviews of LGDs.
26
Figure 24. Frequency of curve remark
DVA
LGD
CVA
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Intra-day Daily Weekly Monthly
Quarterly Yearly Ad-hoc Not marked
Regarding market risk components, it is interesting to note that the majority of participants mark volatility to
market (risk neutral) for front office purposes.
This is driven by the exit price concept under the fair value measure for accounting requirements under IFRS 13
and the wish to hedge movements in the exposures driven by volatility. Not all volatilities can be easily calibrated
as some asset classes have less developed volatility markets and long-dated volatilities are often unavailable.
Finally, very off-market trades require either in or out-of-the-money volatilities which may not be observable. Often
assumptions for extrapolating volatility skew across strike and maturity are important considerations.
Whilst volatilities are reasonably well accessible, the same cannot be said of correlation parameters. These are only
sparsely available via a limited selection of basket, quanto and spread option products and correlations, including
those representing general WWR. As a result, it is common to mark correlation parameters to historical data. This
would imply a need to identify key correlation sensitivities and potentially seek hedges for these risks on a portfolio
basis. Specific WWR approaches are calibrated to market parameters if they exist (e.g. quanto CDS) and otherwise
are estimated empirically and with a degree of judgement.
The need to mark to both risk-neutral and historical parameters was also found in relation to model parameters.
For example, participants commented that, with certain models, some parameters are marked as risk-neutral (for
example, mean reversion levels) and some as historical (for example, mean reversion speed).
Figure 25. Exposure model dynamics marking
C

V

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
H
I

2.5. Implementation
More than 80% of participants (including all Tier 1 banks) reported having an internal system for front office
pricing, with the remaining having vendor systems. There are clear advantages and disadvantages when comparing
an internally built system and a vendor product. One of the primary advantages of an internal system is the control
over the specification and build, and the ability to respond in a timely manner to the ever changing landscape of
counterparty risk pricing and hedging.
With respect to trade pricing, 82% of participants confirmed that new trades are priced in a real-time incremental
framework, accounting for netting at inception. The remaining banks do not have the operational and technology
capacity to calculate such real time CVA. The deployment of CVA pricing tools is, as expected, concentrated in the
front office sales and trading areas.
The trend over the last few years, certainly for the more sophisticated banks, has been to devolve incremental CVA
and DVA pricing to the relevant marketer (and subsequently priced into the trade by the trader) for small vanilla
deals within certain agreed limits. This then allows the CVA trading desk to concentrate on the more structured,
more risky deals, and incremental deals against a large portfolio.
Figure 26. Access to pricing tools
Sales
Trading
Risk
Other users
0% 10% 20% 30% 40% 50% 60% 70% 80%
2.6. Incorporation of risk mitigants

CVA is naturally reduced by a wide range of risk mitigants, most of which are traditional in CCR management
and not specific to CVA. Whilst some mitigants such as netting, recouponing and mandatory break clauses are
relatively straightforward to model, other components such as DVA, collateral and non-mandatory breaks are
more subjective.
When pricing CVA into trades, it is generally accepted that the presence of CVA charges, both to clients and other
banks, can be prohibitive to certain types of trading activity. The most common ways in which CVA charges are
reduced are by including a DVA component, or using an historical or blended default probability. We emphasise
that these aspects are mutually exclusive and, as mentioned previously are not consistent with Basel III capital rules.
Another common method used to reduce CVA charges is to assume a higher recovery on the claim than is assumed
in the default probability estimation, either due to structural subordination or based upon the assumption that the
claim process will be managed to achieve a superior recovery than that which would have been achieved at the
time of default (CDS auction).
28
Figure 27. Risk mitigants applied to CVA calculations
Recovery
PFE (Credit line threshold)
PD calculation
DVA
0% 10% 20% 30% 40% 50% 60% 70%
Due to the debate around DVA, it is particularly interesting to explore to what extent DVA is incorporated into the
pricing of new trades. Many banks include full DVA into pricing whilst a lesser number give only partial benefit.
A significant proportion give no benefit at all, although these banks tend to be those using historical (or blended)
default probabilities. Whilst the survey results indicate a strong trend of giving full DVA benefit, anecdotal evidence
suggests that even for the most aggressive pricing, the full DVA benefit may not be given and would also be
capped at the CVA (so as to not ’pay through mid’ in a situation where the DVA benefit exceeds the CVA).
We further note that DVA benefit given depends on the type of trade and counterparty. For example,
collateralised trades with counterparties of similar credit quality may, implicitly1 or explicitly, be given full DVA
relief whereas uncollateralised trades with end-users and/or weaker credit quality counterparties may give a
small or no DVA benefit.
Figure 28. Inclusion of DVA in pricing
Fully included
Partially included
Inclusion is planned
Not included
There is growing appreciation of the importance of closeout assumptions in relation to CVA and DVA calculations.
Although only a few banks incorporate closeout assumptions in their methodology, a significant number plan to do
this in the future.
1 Meaning that the trade is executed at mid and CVA and DVA are not even quantified. This could be rationalised by the similar
credit spreads and that the use of a two-way CSA will symmetrise even a relatively asymmetric exposure profile. In such
situations the CVA and DVA would be expected to be approximately equal and opposite.

In terms of other risk mitigants, collateral is, not surprisingly, always included in pricing with the only issues
being having the correct legal information and the computational burden associated with providing real
time calculations including such mitigants (discussed below). Other strong risk mitigants such as contractual
recouponing and mandatory break clauses are also generally included when present. Softer risk mitigants are less
likely to be included. These include rating based triggers, either in relation to a break or collateral receipt, which is
not surprising given the difficultly in modelling rating transitions in relation to a potential default and the potential
cliff edge effects that such triggers introduce. Optional break clauses are not often included: whilst these breaks
can be more freely exercised, there are clear issues in defining this exercise boundary. With respect to the inherent
asymmetry between the CVA desk always wanting to exercise such breaks2 (to reduce risk) and the originator
of the trade never wanting to break (to preserve the client relationship), the former component is becoming
increasingly dominant.
One important aspect of including DVA and FVA in valuation is that apparently risk mitigating actions do not
always result in P&L gains. Examples of this are consolidation of netting agreements and bilateral reduction of
collateral thresholds. A CVA desk should always price in the potential losses driven from DVA/FVA before such
agreements are renegotiated.
Figure 29. Risk mitigants accounted for in the payoff
Collateral
Break clause – Mandatory
Break clause – Optional
Break clause – Rating Based
Re-couponing
Triggers/rating downgrades
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
2.7 Hedging
In terms of the calculation of greeks, we found the majority of banks calculated both exposure and credit risk
related sensitivities at least daily. The range of greeks calculated is also quite sophisticated with components such
as jump to default and cross gamma seen as being important alongside more traditional measures such as delta,
gamma and vega.
2 At least if the MTM is positive, as otherwise the DVA or funding benefit may be significant.
30
Figure 30. Calculation of greeks
Correlation
Vega
Jump to default
Cross gamma
Gamma
DV01
(Expsure)
SPV01
(Credit risk)
0% 10% 20% 30% 40% 50% 60% 70% 80%
Figure 31. Frequency of greek calculation*
Correlation
Vega
Jump to default
Cross gamma
Gamma
DV01
(Exposure)
S 1
(Credit Risk)
0% 20% 40% 60% 80% 100%
Intraday Daily Weekly Other
*Expressed as a percentage of the total number of participants that calculate the relevant greek.
Whilst a CVA desk has many greeks to monitor, not all are actively hedged and rebalancing may be infrequent.
Participants indicated that the majority of hedging is discretionary in nature, which is understandable given the
complex non-linear nature of CVA/DVA risk and significant transaction costs, especially in relation to credit risk.
The following figures present a breakdown of which greeks banks hedge and the frequency of such hedge
rebalancing.

Figure 32. Greeks hedged
Correlation
Vega
Cross gamma
Gamma
DV01
(Exposure)
SPV01
(Credit Risk)
0% 20% 40% 60% 80% 100%

Full Discretionary No
Figure 33. Frequency of greek hedging*
Correlation
Vega
Cross gamma
G&&
DV01
(Exposure)
SPV01
(Credit Risk)
0% 20% 40% 60% 80% 100%

! D"#! O$%
*Expressed as a percentage of the total number of participants that hedge the relevant greek.
Given that most CVA desks are set up as hedge centres rather than profit centres and that in either case CVA
hedging is to some degree discretionary, it is not surprising that the majority of desks are subject to various risk
limits, the most common of which are credit risk and exposure risk limits (typically assessed against a VaR measure).
The risk that is most predominantly hedged is exposure DV01s, which is not surprising since the underlying hedges
are generally liquid and often exchange traded. On the other hand, credit risk, volatility and correlation hedges may
be illiquid, subject to their own CVA and in many cases simply unavailable. By focusing on the hedging of material
and liquid market risk components, a CVA desk can reduce its MTM volatility even if it is not hedged against its
idiosyncratic jump to default credit risk on illiquid names. It should be noted that generally hedging practice may
change under future Basel III capital requirements as capital relief is achieved for the hedges which are illiquid and
have WWR (single name and index CDS) whilst more liquid hedges (for example, interest rate hedges)
may actually consume capital.
32
Figure 34. Limits applied to the CVA desk
Vega limits
Concentration limits
Jump to default limits
DV01 limits (Exposure)
SPV01 limits (Credit risk)
0% 10% 20% 30% 40% 50% 60%
Participants hedge their credit SPV01s with the products that are most liquidly available in the market and best match
their hedge methodology and mandate. Given that the population of counterparties with actively traded CDSs is
limited, it is not surprising that the most traded product is CDS indices. The number of respondents that trade CDS and
proxy CDS compared to CDS indices also indicates that banks would use CDS for those counterparties which actively
trade and indices for the more illiquid counterparties. iTraxx and CDX indices offer the ability to hedge the systemic risk
of the illiquid counterparties in a commoditised way by region, counterparty type and tenor. Whilst some of the more
sophisticated banks indicated that they have traded contingent credit default swap (CCDS) in the past, this market has
never lived up to the expectations that were created when the technology was first developed.
Figure 35. Instruments used for credit SPV01 hedging
CDS Indices
Contingent credit default swaps
Proxy CDS (single-name/basket)
CDS where available in the market
0% 10% 20% 30% 40% 50% 60%
During financially stressed periods, exposures tend to increase at the same time as credit spreads widen,
a phenomenon generally described as WWR. Not surprisingly, Basel III regulation has placed a stronger emphasis
on both general and specific WWR, the former arising from macroeconomic relationships and the latter from
badly structured trades. Some more attention is being given to WWR although, as noted above, only 31% of
banks model general WWR. Specific WWR is receiving more attention although most approaches are relatively
ad-hoc. The most sophisticated banks are implementing programmes to identify and attempt to work out WWR
positions and often to avoid such trades entirely. As can be seen from the following chart, the main WWR under
consideration are FX, followed by interest rate, CDS and commodities. For those banks that do identify WWR and
attempt to calculate it, the instruments used to hedge are mainly out-of-the-money options and CDS in different
currencies.

Figure 36. Treatment of wrong way risk
CDS
Commodities
FX
Interest Rates
0% 20% 40% 60% 80% 100%
Quantify and hedge Quantify and don’t hedge Ignore
Given the current emphasis on controlling the linkage between CVA/DVA, as measured under accounting and front
office measures, and CVA capital, as measured under regulatory requirements, it is helpful to understand how
banks are looking at strategies to manage their approach to obtaining capital relief by hedging and other structural
approaches. Such approaches are outlined in the figure below. CSA renegotiation and tactical unwinds and
novations are used by most banks whilst hedging with single name or index CDS is slightly less common since
a number of banks do not activity hedge their CVA credit risk. CVA securitisation has been tried by a few banks
that participated in the survey, but the uncertainties over future capital relief (none is permitted according to
Basel III) presumably has led the majority of banks to not yet pursue this option.
Figure 37. Strategies looked at to obtain capital relief
CSA (renegotiation)
Tactical Novations/Unwinds
Securitisation
Index CDS
Single name CDS
0% 10% 20% 30% 40% 50% 60% 70% 80% 90%
The move from using historical PDs and market parameters to using risk-neutral parameters, coupled with the
adoption of DVA and FVA into front office pricing, has resulted in significant increases in P&L volatility. This has
necessitated the increased control and visibility of P&L movements. Banks are generally focusing on building P&L
explains, which attempt to cover as many parameters as possible and hence reduce unexplained P&L. Of the banks
surveyed, 73% had a P&L explain (which constitutes all the banks following an active hedging approach) which
for the more sophisticated banks has a residual unexplained amount of less than 5% of the gross daily movement
in reserves.
34
2.8. Return on capital
Counterparty risk capital increases under Basel III are anticipated to be substantial due to both CVA VaR and the
more conservative modelling assumptions such as stressed calibration and increase in the margin period of risk.
Given the increased capital usage of derivative trades under Basel III, not surprisingly, 65% of participants
confirmed that their institutions price trades with clients in the context of a return on capital (equity) hurdle. For the
banks that include capital costs in pricing, the trades considered are shown below.
Figure 38. Trades priced in the context of a return on capital hurdle
All trades
Trades without “strong CSAs” only
Trades with no CSA only
Additionally for those participants who price derivatives in terms of a target/hurdle:
• 30% adjust the return to reflect costs and a tax efficiency ratio.
• 70% reflect capital usage over the life of the trade in equity. Of these banks:
– 85% discount capital at the risk free rate; and
– 15% discount capital at the cost of capital rate.
At the time of writing, future capital requirements are mired in uncertainty with respect to the actions of local
regulators. For example, in Europe the sovereign exemption under CRD IV has provided relief and banks (and
their clients) are hoping that this will be followed by a similar non-financial exemption for which they have been
lobbying. Clearly such an exemption would be extremely beneficial in terms of relieving the capital charges
associated with, for example, corporate counterparties. Other potential methods of achieving capital relief such as
CVA securitisations, and gaining recognition for market risk hedges, do not yet have a clear impact as they depend
on the views of local regulators.
Given the regulatory uncertainty, defining a return on capital hurdle rate is challenging, a problem that is
particularly acute for long-dated trades. Banks are either dealing with this by making their best estimate of future
regulatory rules or by being conservative and viewing any future concessions as producing gains (that may be
partially passed back to clients).

3. Funding and valuation
3.1. Overview
The issues surrounding CVA have spread out to cover three new but related areas that are proving just as difficult
and controversial in their own ways. Firstly, CVA is defined as an adjustment to the risk-free value of a trade and
therefore defining risk-free valuation is paramount. Secondly, the same intuition and mechanisms behind CVA
also appear, analogously, to give rise to an FVA adjustment that defines the costs and benefits derived from the
funding of a derivatives book. Finally, the concept of valuing the optionality derived from collateral agreements is
increasingly being viewed as a significant valuation component, collateral valuation adjustment (CollVA).
3.2. OIS discounting

Under stylised assumptions that can be loosely defined as representing a perfect collateral agreement, it can be
shown that OIS discounting is the correct valuation mechanism and no further adjustments are required. Whilst such
a theoretical ideal never exists in practice, it is a useful starting point. Furthermore, certain trades such as interbank
and centrally cleared ones (from the point of view of the CCP at least) are reasonably close to this limiting case.
In light of the above comments, it is not surprising that 90% of participants confirmed that all desks are already, or
are planning in the near future, to use OIS discounting for the valuation of collateralised (secured funding) trades.
Whilst such dual-curve discounting is much more complex due to the need to calibrate both OIS and LIBOR curves
where the instruments defining the former are often illiquid except for the major currencies, OIS discounting is
becoming the market standard for risk-free valuation.
Figure 39. Desks using or planning to use OIS discounting Figure 40. Stochastic models considered for OIS discounting
All desks Yes – implemented
Certain desks only Yes – not implemented
No
In addition to the switch to OIS discounting approaches, some banks have looked into stochastic models for
modelling the behaviour between OIS and LIBOR rates, although the majority of these are still in development
since banks consider them too complex to implement at the current time.
36
3.3. Collateral value adjustment
It has become widely appreciated that typical CSAs may give rise to a large degree of optionality due to the
flexibility over the collateral that can be posted, both in terms of currency and asset type.
A bank can attempt to post (and substitute, if relevant) the collateral that is most beneficial in terms of return and
balance sheet opportunity. In the case of non-cash collateral, the relevant haircuts and repo considerations must be
factored into these decisions. Obviously a bank must consider the optionality that their counterparty has and the
fact that this will be exercised optimally.
The first way in which CollVA is seen is via the choice of (OIS) discount curve used for valuation which differs
between the trade currency, posted collateral and theoretical cheapest-to-delivery collateral.
Figure 41. Discount curve for secured funding trades
Cheapest to deliver currency
Posted collateral currency
Currency of trade leg
0% 10% 20% 30% 40% 50%
Whilst banks are attempting to monetise CollVA where possible, it is generally recognised as a component which is
highly subject to price. Going forward, it appears likely that it will instead be structured out of trades via changes to
collateral agreements. For collateralised counterparties a significant number of participants envisaged simplification
of CSAs (either bilaterally or through the standard credit support annex (SCSA) in order to mitigate the complexity
of modelling the components of current CSAs.
Figure 42. Changes to CSAs envisaged
N'()
Simplifying CSAs bilaterally
Move to SCSAs

3.4. Unsecured funding (FVA)
For uncollateralised trades in particular, it has become increasingly common for banks to consider funding costs and
benefits via FVA in pricing. 52% of participants charge for FVA at the trade level with most charging it to the relevant
trading desk at inception. The remainder recover FVA on an accrual basis or not at all. This reflects the growing
consensus that FVA is not only an important component but ideally must also be charged on an upfront basis to
prevent funding intensive trades appearing profitable when they are not.
Figure 43. Charging the trading desk for funding*
Accrual based upon the current

mark-to-market
FVA charge at inception
*Exp*+ss+d as a percentage of the total number of participants that charge for funding.
Almost 80% of banks use a rate based on the bank’s internal funding policy for marking unsecured funding, with
the remaining basing their calculation of FVA on a bond spread.
Interestingly, we did not generally observe a thorough treatment of partially collateralised trades (e.g. the case of a
relatively high threshold in a CSA) that tend to fall in between the extreme cases of OIS discounting and unsecured
funding.
In the debate as to whether DVA should be included alongside CVA, it has often been rationalised as a funding
benefit. This has caused further debate as to how to price CVA, DVA and funding into trades at inception (and on
an on-going basis, for example on assignments and novations).
The situation has not been clarified, rather the opposite, by a proliferation of mathematical and theoretical
literature on the topic of CVA, DVA and funding, and potential overlaps between the numbers and the variables
that go into producing the numbers. Many authors on the topic have expressed a variety of divergent views. For
participants that consider DVA and FVA, the figure below presents how they currently view CVA, DVA and FVA
from a front office perspective.
38
Figure 44. Inclusion of CVA, DVA and funding
CVA + DVA + symmetric funding
CVA + symmetric funding
CVA + DVA + partially asymmetric

funding
Interestingly, it seems that banks are increasingly seeing DVA as a funding benefit and not as a benefit in the event
they default. We note that whilst a bank may consider CVA + symmetric funding to be relevant, they may still refer
to the funding benefit as DVA. An obvious reason for this is to remain consistent with accounting requirements.
However, the role DVA plays (whether it is viewed as a funding benefit or not) in terms of damping the overall
volatility of the P&L of a CVA desk is also important.
Despite the obvious disagreements over the treatment of funding, there seems little disagreement from practitioners
generally that funding should be a component of valuation. This is at odds with some eminent academics3 who
have published work suggesting that funding costs should not be a component of valuation. There is potentially
some middle ground to this debate relating to the fact that the theoretical and practical views of funding may
differ substantially due to the fact that the market for funding does not operate in an idealised Capital Asset Pricing
Model manner. In addition, devolving group treasury funding requirements across multiple businesses in a bank is
highly complex. Calculating a funding cost per trade, which is tied to how the bank would fund the trade on
a group-wide netted basis, is not a trivial undertaking.
3 See J. Hull and A. White, Risk July 2012, pp. 83-85 and Risk September 2012, pp. 18-22.

3.5. Organisation
The way that the majority of the larger banks have dealt with the issues resulting from the financial crisis has been
to set up desks to manage the resources of the bank, from funding to credit and capital. As these resources have
become scarcer, banks have been forced to focus far more on management of capital, and the return thereon. One
practical application has been a trend in the larger institutions to implement a central funding desk which has been
tasked with optimising the way the front office trading desks fund their derivative trades on a mark to market basis,
a role which has been traditionally performed by the banks’ treasury department on a global accrual basis. As can
be seen from the following graph, 60% of participants confirmed they had a central funding desk in place currently,
with the remaining 40% stating that they have future plans to implement a central funding desk.
Figure 45. Central funding desk setup
Central funding desk in place
Proposed for future implementation
Not planned
One mechanism the central treasury function uses to fund the bank is the issuing or redeeming of bonds.
As the trend over the last few years has been to more actively hedge the risks resulting from CVA and DVA, so has
the trend more recently been to hedge the funding risks. Half of those participants who have a central funding desk
in place hedge the market risk on the funding position.
40
Conclusion
The landscape around CVA has changed dramatically in the last two to three years and related areas such as
OIS discounting, collateral optimisation and funding have become increasingly significant. Market practices are
evolving rapidly, catalysed by changes in accounting requirements and regulatory capital guidelines. In keeping
with the findings of Solum Financial Partners’ 2010 survey, there is still an evident divergence in approaches and
the nonuniformity of methodologies across the market.
Given the changes brought about by future Basel III capital rules, it is not surprising that there is a large focus on
the regulatory side of counterparty risk. Banks are investing more resource into building models for advanced
capital treatments. This usually includes the smaller banks that are looking to gain IMM and/or specific risk
approvals to allow them to use the advanced capital methodologies charges for both default risk and CVA capital.
There is a growing trend to model collateral rather than rely on simpler routes such as the shortcut method and
increased emphasis on quantifying WWR. Effort is also being placed on building backtesting frameworks and
establishing effective model validation procedures. Finally, the move towards central clearing is focusing efforts on
quantifying CCP trade and default fund exposures, and calculating the associated capital charges.
The use of risk-neutral default probabilities via credit spreads is becoming a standard practice in the quantification
of CVA, driven by accounting and capital rules. The associated problem of mapping illiquid credit spreads is
receiving significant thought. Divergence still exists over DVA and the extent to which it should be used to reduce
CVA charges. Return on capital considerations are being incorporated into pricing decisions and are considered
especially important in light of the Basel III CVA capital charge. The potential impact of capital charges is also
leading to increased focus on capital reducing CVA hedging strategies, despite the potential misalignments
between capital relief and hedging in relation to DVA and non-credit related hedges.
A number of areas have developed around CVA which have recently received substantial consideration. There is a
general switch to OIS discounting as the best standard valuation method (at least for collateralised trades), although
some divergence exists over the correct choice of OIS currency. The optionality around collateral terms has also
led to debate around the value inherent in CSAs and how best to optimise this. Finally, the consideration of FVA as
a material component of valuation has become common, although this remains probably the most controversial
aspect, with debate on the validity of FVA and the potential overlap between FVA and DVA creating variation in
approaches across the market.
Despite substantial effort around CVA practices and related areas over the last few years, future trends remain very
hard to predict. This is largely due to ambiguity over the implementation timescale and potential exemptions in
Basel III. Uncertainty over aspects such as DVA and FVA, which are outside the Basel III mandate but are the subject
of increased focus under accounting rules, adds to the confusion. The one thing that is certain is that CCR, CVA and
FVA will remain hot topics for regulators, practitioners and academics for some time to come.

Contacts
About Deloitte LLP

Deloitte LLP offers professional services to the UK and European market. With over 14,500 exceptional people in
28 offices in the UK and Switzerland, Deloitte has the broadest and deepest range of skills of any business advisory
organisation. We provide professional services and advice to many leading businesses, government departments
and public sector bodies, and publish many influential studies and thought leadership pieces. Our annual revenues
reached £2,329m for the financial year ended 31 May 2012.
Deloitte LLP is the UK member firm of Deloitte Touche Tohmatsu Limited, a UK private company limited by
guarantee, and its network of member firms, each of which is a legally separate and independent entity.
About Solum Financial Partners LLP

Solum Financial Partners (www.solum-financial.com) is a unique consultancy firm which offers advisory services
provided by front office experts across all areas of capital markets and risk management. The Solum Financial
Partners team consists of detail-oriented front office experts who have held senior managerial positions and are
highly experienced in the areas of trading, structuring, principal investing, risk management, quantitative analysis,
modelling and control and support functions. The Solum Financial Partners team has established industry leadership
in counterparty credit risk, CVA and associated regulatory matters.
Contacts
Deloitte
Tim Thompson Vishal Vedi

Partner, Risk & Regulation Partner, Risk & Regulation
+44 (0)20 7007 7241 +44 (0)20 7303 6737
tthompson@deloitte.co.uk vvedi@deloitte.co.uk
Zeshan Choudhry Liesbeth Bodvin

Director, Risk & Regulation Senior Manager, Risk & Regulation
+44 (0)20 7303 8572 +44 (0)20 7303 6597
zchoudhry@deloitte.co.uk lbodvin@deloitte.co.uk
Solum Financial Partners
Vincent Dahinden Jon Gregory

Chief Executive Officer Partner
+44 (0)20 7786 9235 +44 (0)20 7786 9233
vincent@solum-financial.com jon@solum-financial.com
Thu-Uyen Nguyen Su Green

Partner Senior Consultant
+44 (0)20 7786 9231 +44 (0)20 7786 9232
tu@solum-financial.com su@solum-financial.com
Rowan Alston Ilya German

Senior Consultant Senior Consultant
+44 (0)20 7786 9238 +44 (0)20 7786 9239
rowan@solum-financial.com ilya@solum-financial.com
42
Notes

Notes
44
Deloitte refers to one or more of Deloitte Touche Tohmatsu Limited (“DTTL”), a UK private company limited by guarantee, and its
network of member firms, each of which is a legally separate and independent entity. Please see www.deloitte.co.uk/about for a
detailed description of the legal structure of DTTL and its member firms.
Deloitte LLP is the United Kingdom member firm of DTTL.
This publication has been written in general terms and therefore cannot be relied on to cover specific situations; application of the
principles set out will depend upon the particular circumstances involved and we recommend that you obtain professional advice
before acting or refraining from acting on any of the contents of this publication. Deloitte LLP would be pleased to advise readers
on how to apply the principles set out in this publication to their specific circumstances. Deloitte LLP accepts no duty of care or
liability for any loss occasioned to any person acting or refraining from action as a result of any material in this publication.
Deloitte LLP is a limited liability partnership registered in England and Wales with registered number OC303675 and its registered
office at 2 New Street Square, London EC4A 3BZ, United Kingdom. Tel: +44 (0) 20 7936 3000 Fax: +44 (0) 20 7583 1198
Solum Financial Partners refers to Solum Financial Partners LLP.
This report is provided for your information only and does not constitute legal, tax, accountancy or regulatory advice.
Although all opinions and recommendations expressed in this document were formed in good faith based on the information
provided by the survey participants, no representation or warranty, express or implied, is made as to the accuracy of those opinions
or as the appropriateness of those recommendations. You must make an independent assessment of this report, including any
representations made herein, with your professional advisors in the context of your particular circumstances before taking any
course of action. No responsibility or liability is accepted by us or any of our principals, officers, consultants or agents in relation
to the accuracy, appropriateness or completeness of this report, any recommendations made herein, or any other written or oral
information provided to you in relation to the subject matter of this report.
This report is based on the documents, communications, reports and other written information provided by the survey participants.
We have no responsibility for, and have not verified, the accuracy or completeness of such information.
All information and opinions contained in this report are subject to change without notice, and we have no responsibility to update
this report after the date hereof.
© 2013 Deloitte LLP and Solum Financial Partners LLP. All rights reserved.
Co-authored by Deloitte LLP and Solum Financial Partners LLP. Designed and produced by The Creative Studio at Deloitte, London.
24671A
WHITE PAPER
April 2016
Federal Reserve Board Proposes New

Single-Counterparty Credit Exposure Limits for
Large Banking Organizations
On March 4, 2016, the Federal Reserve Board (“the Board”) proposed a rule that would
limit the credit exposures of large banking organizations to a single counterparty. The
proposal implements part of the Dodd-Frank Act and is designed to mitigate risks to the
financial stability of the United States that can result from interconnectivity among major
financial institutions. The proposal would apply increasingly stringent single-counterparty
credit limits to U.S. bank holding companies (“BHC”), intermediate holding companies
(“IHC”), and foreign banking organizations (“FBO”) with total consolidated assets of $50
billion or more as systemic importance increases.
The new proposal builds on two Board proposals issued in 2011 for domestic BHCs and
in 2012 for FBOs while reflecting some revisions based upon public comments and incor-
porating many features of the Basel Committee on Banking Supervision’s large exposures
framework issued in 2014. In conjunction with the new proposal, the Board released
a White Paper, Calibrating the Single-Counterparty Credit Limit between Systemically
Important Financial Institutions, that provides support for the Board’s more stringent single
counterparty credit exposure limit between the largest financial institutions.
TABLE OF CONTENTS
KEY FEATURES OF THE NEW PROPOSAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Statutory and Regulatory Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Increasingly Stringent Limits Based upon Systemic Importance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Aggregate Net Credit Exposure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Covered Counterparties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Applies to Covered Companies on a Consolidated Basis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Separate and Independent from Bank Investment and Lending Limits . . . . . . . . . . . . . . . . . . . . . . . . . 6
Compliance Requirements and Effective Date . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
IMPORTANT CHANGES FROM THE ORIGINAL PROPOSALS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Measure of the Eligible Capital Base Against which Exposure Limits Would Be Determined . . . . . . 7
Credit Exposure Limits Between SIFIs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Broader Scope of Covered Subsidiaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Broader Scope of Covered Counterparties with More Complexity to Making Determinations . . . . 8
Statutory Attribution Rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Exposure Methodology for Derivatives Transactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Exposure Methodology for Securities Financing Transactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
“Look-Through Approach” to Exposures to Special Purpose Vehicles . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Exemptions for Exposures to the Government and Government-Sponsored Enterprises . . . . . . . . 11
Compliance Requirements and Reporting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
KEY DIFFERENCES BETWEEN THE LARGE EXPOSURES FRAMEWORK

ADOPTED BY THE BASEL COMMITTEE ON BANKING SUPERVISION AND
THE NEW PROPOSAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
SUBJECTS FOR POSSIBLE COMMENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
LAWYER CONTACTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
ENDNOTES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Jones Day White Paper

ii
A principal tenet of the Dodd-Frank Wall Street Reform and • Second Category: A domestic BHC, FBO, and U.S.
1
Consumer Protection Act of 2010 (“Dodd-Frank Act”) was IHC with total consolidated assets of $250 billion
creation of a comprehensive approach for mitigating threats or more, or $10 billion or more in on-balance-sheet
to the financial stability of the United States (“U.S.”) posed by foreign exposures, would be prohibited from having
systemically important financial institutions (“SIFI”). aggregate net credit exposure to a single counterparty
in excess of 25 percent of the company’s Tier 1 capital.
The Dodd-Frank Act mandates a comprehensive set of
regulatory reforms designed to address threats to U.S. • Third Category: A domestic BHC that is a global systemically
financial stability. These reforms cover enhanced prudential important banking organization (“G-SIB”), any U.S. IHC with
regulation of large bank holding companies (“BHC”) and total consolidated assets of $500 billion or more, and any
nonbank companies that are designated by the Financial FBO with total worldwide consolidated assets of $500
Stability Oversight Council (“FSOC”) for additional oversight by billion or more—defined collectively as a “major covered
the Board; enhanced regulation of over-the-counter derivatives company”—would be prohibited from having aggregate
and other core financial markets and financial market utilities; net credit exposure to a “major counterparty” in excess of
and orderly liquidation authority for financial companies, 15 percent of the major covered company’s Tier 1 capital.
among other important reforms. A “major counterparty” would include a major covered
company, any other FBO that has the characteristics of a
While the federal financial regulators have adopted key G-SIB, and any nonbank financial company designated by
macroprudential rules to fulfill the requirements of the Dodd- the FSOC for additional oversight by the Board.
Frank Act, several significant rules are still under development.
Chief among these is the adoption of a rule setting credit The New Proposal is designed to “enhance the resiliency and
exposure limits for large domestic and foreign banking stability” of the U.S. financial system by setting a “bright line on
organizations. The Board originally proposed rules on single- total credit exposures between one large [BHC] and another
counterparty credit exposure limits in 2011 for domestic BHCs large bank or major counterparty.”6 Overall, the Board intends
and in 2012 for FBO and U.S. IHC as part of a proposal to the New Proposal to mitigate risks to financial stability that
establish a set of enhanced prudential standards, but the Board can arise from the interconnectivity among major financial
did not adopt these proposed rules (collectively, the “Original institutions because “trouble at one big bank will [often] bring
Proposals”) in the final rule on enhanced prudential standards.2 down other big banks.”7 In her opening statement on the Board’s
consideration of the New Proposal, Board Chair Yellen explained:
On March 4, 2016, the Board invited public comments on a
new proposal (the “New Proposal”),3 pursuant to Section 165(e) In the financial crisis, we learned that the largest and
of the Dodd-Frank Act (“Section 165(e)”),4 that would apply most complex banks and financial institutions lent or
increasingly stringent single-counterparty credit exposure promised to pay large amounts to other institutions
limits to large domestic and foreign banking organizations as that were also very large and complex. These credit
their systemic significance increases, in accordance with the extensions and promises did not eliminate risk, and in
following framework: many cases they magnified it.8
• First Category: A domestic BHC, FBO, and U.S. IHC The limits on single-counterparty credit exposures in the New
5
with total consolidated assets of $50 billion or more Proposal are thus intended to be “tailored to the systemic
would be prohibited from having aggregate net credit footprint of covered companies”9 by imposing increasingly
exposure to a single counterparty in excess of 25 stringent limits as the systemic significance of a covered
percent of the company’s total regulatory capital plus company increases. The New Proposal would not apply
allowance for loan and lease losses (“ALLL”) not included to credit exposures of any nonbank SIFI that the FSOC has
in Tier 2 capital. This is the limit set by Section 165(e).

1
designated for additional oversight by the Board; however, enhanced prudential standards as BHCs.16 The Board’s notices
these nonbank SIFIs are considered counterparties for of proposed rulemaking17 that preceded adoption of the final
purposes of measuring credit exposure limits for major covered rule on enhanced prudential standards18 included single-
companies, the largest and most complex financial firms.10 counterparty credit limits that ultimately were not adopted
while the Board considered public comments, considered
The Board has invited public comments on all aspects of the New the revised lending limit rules applicable to nationals banks19,
Proposal through June 3, 2016, and has raised specific questions conducted empirical analyses of the quantitative impacts of
and alternatives for comment throughout the New Proposal. credit exposures, and consulted with the Basel Committee,
which adopted the Large Exposures Framework in April 2014.20
This White Paper describes key features of the New Proposal,
highlights important changes made to the New Proposal, and The Original Proposals established a two-tier structure for
explains differences between the New Proposal and the Basel setting limits on single-counterparty exposures. Under those
Committee on Banking Supervision’s (“Basel Committee”) large proposals, a domestic BHC, FBO and U.S. IHC (“covered
exposures framework for internationally active banks (“Large company”) with $50 billion or more in total consolidated assets
Exposures Framework”) issued in 2014.11 generally would have been prohibited from having aggregate
net credit exposure to a single counterparty in excess of 25
percent of the covered company’s total regulatory capital
KEY FEATURES OF THE NEW PROPOSAL plus ALLL not counted in Tier 2 capital.21 This definition of
capital stock and surplus is consistent with the definition
Statutory and Regulatory Background of the same term that appears in the Board’s Regulation
Section 165 of the Dodd-Frank Act directs the Board to O on loans to executive officers, directors, and principal
establish enhanced prudential standards for BHCs with total shareholders of member banks; the Board’s Regulation W on
consolidated assets of $50 billion or more.12 These enhanced affiliate transactions; and the Office of the Comptroller of the
prudential standards must include requirements for risk- Currency’s (“OCC”) national bank lending limit regulation.22
based capital, leverage capital, stress testing, liquidity, risk In addition, a covered company with $500 billion or more in
management, and single-counterparty credit limits.13 total consolidated assets would have been prohibited from
having aggregate net credit exposure to another banking
Section 165(e) authorizes the Board to establish single- organization with $500 billion or more in total consolidated
counterparty credit limits for domestic BHCs and FBOs with assets, or to a nonbank financial company designated by the
total consolidated assets of $50 billion or more in order to FSOC for additional Board oversight, in excess of 10 percent of
limit the risks posed to a covered company by the failure of the covered company’s total regulatory capital plus ALLL not
14
any individual company. This section prohibits domestic counted in Tier 2 capital under the capital adequacy guidelines
BHCs and FBOs with total consolidated assets of $50 billion or applicable to that BHC under the Board’s Regulation Y on
more from having credit exposure to any unaffiliated company BHCs and change in bank control.23
that exceeds 25 percent of the company’s capital stock and
surplus, or such lower amount as the Board may determine by The Board received 48 comments, representing approximately
regulation to be necessary to mitigate risks to U.S. financial 60 parties, on the Original Proposal related to BHCs, and 35
stability.15 In 2014, the Board adopted a final rule to implement comments, representing more than 45 organizations, on the
enhanced prudential standards for risk-based and leverage Original Proposal related to FBOs. Staff of the Board also met
capital, capital planning and stress testing, risk management, with industry representatives and government representatives
and liquidity. As part of that final rule implementing the to discuss issues relating to the Original Proposals. While
enhanced prudential standards contained in section 165 of some commenters expressed support for the broader goals
the Dodd-Frank Act, the Board required that FBOs with total of the Original Proposals, most commenters were critical of the
consolidated assets of $50 billion or more and total non-branch approach to almost every aspect of the Original Proposals and
U.S. assets of $50 billion or more consolidate U.S. subsidiary to the absence of a stated foundation to support many parts
activities under a U.S. IHC that would be subject to the same of the Original Proposals.

2
Generally, commenters on both Original Proposals were covered companies is significant because measuring single-
critical of the same features. Industry commenters criticized counterparty credit exposures against Tier 1 capital imposes a
the Original Proposals for taking an overly broad approach stricter limit than using total regulatory capital plus ALLL not
to credit exposures, which, they argued, would constrain included in Tier 2 capital. Therefore, while the percentage of
market liquidity, decrease lending capacity, and push banking single-counterparty exposures for covered companies with $250
activities to the so-called shadow banking companies. One billion or more in consolidated assets remains unchanged from
industry study estimated that counterparty exposures the Original Proposals, these second category companies could
exceeding the proposed limits of the Original Proposals on face stricter single-counterparty exposure limits overall because
domestic BHCs would have totaled nearly $1.3 trillion.24 Some of the more limited measure of Tier 1 capital. The preamble to
industry commenters characterized the foundational focus on the New Proposal indicates that Tier 1 capital represents, on
interconnectivity losses as conceptually flawed and asserted average, about 82 percent of total regulatory capital plus ALLL
that losses during the financial crisis were not due to the not included in Tier 2 for these covered companies.25
actual interconnectedness of financial institutions but were
attributable to investor anxiety that financial institutions had The third category of credit limits in the New Proposal would
similar shared investments and risk issues. apply to U.S. G-SIBs26 and U.S. IHCs with total consolidated
assets of $500 billion or more, as well as FBOs with total
consolidated worldwide assets of $500 billion or more.
INCREASINGLY STRINGENT LIMITS BASED UPON Whereas the Original Proposals prohibited these major covered
SYSTEMIC IMPORTANCE companies from having aggregate net credit exposure to any
entity that is a major counterparty, in excess of 10 percent of
The New Proposal would establish three increasingly stringent total regulatory capital, plus ALLL not included in Tier 2 capital,
single-counterparty credit limits, whereas the Original Proposals the New Proposal would set the limit at 15 percent of Tier 1
would have established two limits. As with the Original Proposals, capital, even though many commenters recommended aligning
the baseline standard would prohibit a covered company with the exposure limits with the statutory limits and criticized the
$50 billion or more in total consolidated assets from having Original Proposals for failing to provide adequate reasons for
aggregate net credit exposure to a single counterparty in excess departing from the 25 percent limit set forth in Section 165(e) for
of 25 percent of the covered company’s total regulatory capital major covered companies. The aggregate net credit exposure
plus ALLL not included in Tier 2 capital. The New Proposal, limit that would apply to major covered companies’ exposures to
however, would establish a new second category of credit limits other counterparties would be set at 25 percent of Tier 1 capital.
that would apply to covered companies that have $250 billion
or more in total consolidated assets or $10 billion or more in In conjunction with the issuance of the New Proposal, the Board
cross-border exposures. Covered companies that fall within released a White Paper that explains the rationale for a more
this second category of credit limits would be prohibited from stringent single-counterparty credit limit and the calibration of
having aggregate net credit exposure to a single counterparty the proposed limit of 15 percent of Tier 1 capital for the largest
in excess of 25 percent of their Tier 1 capital. This change to and most systemically important institutions.27 The White Paper
the eligible capital base against which the single-counterparty responds to commenters who were critical of the Board’s
credit exposures are measured for these particular covered Original Proposals for failing to expressly provide a foundation
companies is a change from the Original Proposals, which set for deviating from the 25 percent single-counterparty credit
the eligible capital base as all regulatory capital plus ALLL not exposure limits set forth in Section 165(e) for covered companies
included in Tier 2 for all covered companies. with total consolidated assets above the $500 billion threshold.
Section 165(e) does not define how capital and surplus are to be According to the White Paper, separate SIFIs often share
measured. In the Original Proposals, the Board defined “capital common business lines and funding sources, and as a result,
and surplus” in the same way for all covered companies— they often exhibit similar economic performance. Thus, factors
total regulatory capital plus ALLL not counted in Tier 2 capital. that adversely affect one SIFI would also likely adversely affect
The Board’s movement away from this definition for larger another SIFI, and default by a SIFI borrower and a SIFI lender

3
would cause greater adverse consequences to the stability largest institutions under the Original Proposals. While smaller
of the financial markets than would the default of a non-SIFI covered companies in the first and second categories generally
borrower to a single SIFI lender. face fewer restrictions than GSIBs in the third category, the
New Proposal would place stricter limits on the covered
The White Paper analyzes data on the default correlation companies in the second and third categories than the Original
between SIFIs as well as data on the default correlation between Proposals would have. Under the Original Proposals, all covered
SIFIs and a sample of non-SIFI companies. The analysis companies with less than $500 billion in assets were treated the
supports a finding that the correlation between SIFIs—and same and were prohibited from lending more than 25 percent
hence, the correlation between major covered companies and of their total regulatory capital, plus ALLL not included in Tier 2
major counterparties—is measurably higher than the correlation capital, to a single counterparty. The New Proposal adds a third
between SIFIs and other counterparties. According to the category for companies with assets between $250 billion to
White Paper, this finding further supports the view that credit $500 billion and applies a more stringent capital base against
extensions of major covered companies to major counterparties which to measure these companies’ single counterparty credit
present a higher degree of risk than credit extensions between exposure—Tier 1 capital as opposed to total regulatory capital
28
a major covered company and other counterparties. and surplus, plus ALLL not included in Tier 2 capital.
The three-category approach in the New Proposal reflects both The following chart summarizes the single-counterparty credit
a tightening and a loosening of the credit limits imposed on the limits for covered companies in the New Proposal:29
Category of Covered Company Applicable Credit Exposure Limit

Covered companies—U.S. BHCs, FBOs, and U.S. For U.S. BHCs, aggregate net credit exposure to a counterparty
IHCs—that have: cannot exceed 25 percent of a covered company’s total regulatory
capital plus ALLL not included in Tier 2 capital.
between $50 billion and $250 billion in total consoli-
dated assets, and For U.S. IHCs, aggregate net credit exposure cannot exceed 25
percenwwt of a covered company’s total regulatory capital plus the
less than $10 billion in on-balance-sheet foreign balance of its ALLL not included in Tier 2 capital under the capital
exposures adequacy guidelines in 12 C.F.R. part 252.
For FBOs with respect to U.S. combined operations, aggregate net

credit exposure cannot exceed 25 percent of the FBOs’ total regu-
latory capital on a consolidated basis.
Covered companies—U.S. BHCs, FBOs, and U.S. For U.S. BHCs and U.S. IHCs, aggregate net credit exposure to a
IHCs—that have: counterparty cannot exceed 25 percent of a covered company’s
Tier 1 capital.
more than $250 billion in total consolidated assets, or
For FBOs with respect to U.S. combined operations, aggregate net
more than $10 billion in on-balance-sheet foreign credit exposure to a counterparty cannot exceed 25 percent of the
exposures, FBOs’ worldwide Tier 1 capital.
but are not major covered companies

Major covered companies— For U.S. BHCs, U.S. IHCs, and FBOs with respect to combined U.S.
operations, aggregate net credit exposure to a major counterparty
U.S. G-SIBs, U.S. IHCs that have total consolidated cannot exceed 15 percent of a major covered company’s Tier 1
assets of $500 billion or more, capital.
FBOs with total worldwide consolidated assets of For U.S. BHCs, U.S. IHCs, and FBOs with respect to its combined
$500 billion or more U.S. operations, aggregate net credit exposure to other counterpar-
ties cannot exceed 25 percent of a major covered company’s Tier 1
capital.

4
Aggregate Net Credit Exposure The lending limits rule adopted by the OCC for national banks
The New Proposal would set limits on a company’s “aggregate applies to similar types of extensions of credit following
net credit exposure” to a single counterparty. Aggregate net amendments made to cover derivatives transactions,
credit exposure would be defined as “the sum of all net credit repurchase and reverse repurchase agreements, and
exposures of a covered company to a single counterparty.”30 securities lending or securities borrowing transactions
Under the New Proposal, a covered company would be pursuant to Section 610 of the Dodd-Frank Act.32
required to first calculate its “gross credit exposure” resulting
from credit transactions with that counterparty. “Gross Covered Counterparties
credit exposure” would be defined to mean, with respect to Under the New Proposal, a counterparty would include:
any credit transaction, the credit exposure of the covered
company to the counterparty before adjusting for the effect of • A natural person and the person’s immediate family;
any qualifying master netting agreements, eligible collateral, • An unaffiliated company and all persons that the company
eligible guarantees, eligible credit derivatives and eligible (i) owns, controls, or holds with power to vote 25 percent
equity derivatives, and other eligible hedges (i.e., a short or more of a class of voting securities; (ii) owns or controls
position in the counterparty’s debt or equity securities). 25 percent or more of the total equity of the person; or (iii)
consolidates for financial reporting purposes, collectively;
The New Proposal sets forth the method for calculating gross • A U.S. State and any of its agencies and instrumentalities,
credit exposure for each type of covered credit transaction.31 and political subdivisions;
In general, the methodologies contained in the New Proposal • Any foreign sovereign entity that is not assigned a risk
are similar to those used to calculate credit exposure under weight greater than zero under the Board’s capital rules33
the standardized risk-based capital rules for BHCs. (and all of its agencies and instrumentalities, but not
political subdivisions, collectively); and
Second, a covered company would next reduce its gross • A political subdivision of a foreign sovereign entity such as
credit exposure amount based on eligible credit risk mitigants, states, provinces, and municipalities; any political subdivision
such as collateral, guarantees, credit or equity derivatives, of a foreign sovereign entity; and all of such political
and other hedges, to determine its net credit exposure for subdivision’s agencies and instrumentalities, collectively.34
each credit transaction with a counterparty. Finally, a covered
company would then sum all of its net credit exposures to the A credit exposure to a counterparty would also include a credit
counterparty to calculate the covered company’s aggregate exposure to any person the counterparty owns or controls and
net credit exposure to a counterparty. any credit exposure to counterparties that are “economically
interdependent.” This aspect of the New Proposal is
The credit exposure limits in the New Proposal are identical significant because identifying economically interdependent
to the credit exposures set forth in Section 165(e) and would counterparties and counterparties where a control relationship
apply to: may exist by way of a “controlling influence” is likely to present
significant operational challenges to covered companies
• Extensions of credit; that may not have access to that type of information. In this
• Repurchase or reverse repurchase agreements; regard, the New Proposal expressly raises questions for
• Securities lending or securities borrowing transactions; commenters concerning the operational and other challenges
• Guarantees, acceptances, and letters of credit; that covered companies may face in identifying economically
• The purchase of, or investment in, securities issued by the interdependent counterparties and asks whether companies
counterparty; have access to the information needed to complete the
• Credit exposures in connection with certain derivative analysis of economic interdependence.35
transactions; and
• Any transaction that is the functional equivalent of the Despite numerous comments criticizing the Original Proposals
above as well as any similar transaction that the Board for creating substantial compliance burdens by treating
determines to be a credit transaction. individuals as covered counterparties, the New Proposal

5
continues to treat “a natural person and the person’s immediate • The Board determines, after notice and opportunity for
family” as a counterparty. In the preamble to the New Proposal, hearing, that the company directly or indirectly exercises
the Board indicates its belief that large credit exposures to a controlling influence over the management or policies of
individuals can create risks similar to those created by large the bank or company.40
credit exposures to companies.
In the preamble to the New Proposal, the Board reasons that
Applies to Covered Companies on a Consolidated Basis “[a] bank holding company should be able to monitor and
Like the Original Proposals, the New Proposal would apply control … the credit exposures of its subsidiaries” and that
to the credit exposures of a covered BHC and U.S. IHC single-counterparty credit limits must apply at the consolidated
to any unaffiliated counterparty on a consolidated basis, level in order to “avoid evasion of the rule’s purposes.”41 This
including any subsidiaries. As part of the enhanced prudential expansion of the meaning of subsidiary could pose operational
standards adopted by the Board in 2014, an FBO that has total challenges by capturing a broad scope of companies for which
consolidated assets of $50 billion or more and total non-branch credit exposures may not otherwise be consolidated with the
U.S. assets of $50 billion or more must establish a U.S. IHC to covered company and for which the covered company and the
hold its interests in U.S. bank and nonbank subsidiaries.36 subsidiary may not have common, integrated systems. Under
Credit exposure limits as applied to an FBO as opposed to an the Original Proposals, the scope of subsidiaries would have
IHC or BHC would apply only with respect to credit exposures been narrower under what the Board had viewed as a “simpler,
of that FBO’s combined U.S. operations (i.e., any branch or more objective” definition of “control” than that in the Bank
agency of the FBO; exposures of the U.S. subsidiaries of the Holding Company Act.42 In the Original Proposals, a subsidiary
FBO, including any U.S. IHC; and any subsidiaries of such of a covered company would have captured only those where
subsidiaries (other than any companies held under Section 2(h) the company (i) owns, controls, or holds with power to vote
(2) of the Bank Holding Company Act of 1956 (“Bank Holding 25 percent or more of a class of a company’s voting stock;
Company Act”)),37 although the FBO’s total consolidated assets (ii) owns or controls 25 percent or more of a company’s total
on a worldwide basis would determine whether the credit equity; or (iii) is consolidated for financial reporting purposes.43
exposure limits apply in the first instance.38 In determining
whether a U.S. IHC complies with the single-counterparty limits, Separate and Independent from Bank Investment and
exposures of the U.S. IHC itself and its subsidiaries would need Lending Limits
to be taken into account.39 While the New Proposal is silent Section 165(e) is a separate and independent limit from the
about whether a U.S. IHC must consider the exposures of investment securities and lending limits that apply to insured
its branches and agencies, the fact that the Board explicitly depository institutions under the National Bank Act and the
requires FBOs to do so suggests that U.S. IHCs need not. Federal Reserve Act44 and through rules for federal- and state-
chartered banks. The total amount of investment securities
Under the New Proposal, a subsidiary of a covered company of any one obligor that a national bank may purchase for its
would mean a company that is directly or indirectly controlled own account is generally limited to no more than 10 percent
by the covered company for purposes of the Bank Holding of the bank’s capital stock and surplus.45 The total amount of
Company Act. Under the Bank Holding Company Act, a outstanding loans and extensions of credit to a single borrower
company has control of a bank or another company if: may not exceed 15 percent of national bank’s capital stock and
surplus, plus an additional 10 percent of the bank’s capital and
• The company directly, indirectly, or acting through one or surplus, if that amount is fully secured by readily marketable
more other persons owns, controls, or has power to vote collateral. Similar limits generally apply to state-chartered banks.
25 percent or more of any class of voting securities of the
bank or company; The New Proposal would require covered companies to apply
• The company controls in any manner the election of single-counterparty credit exposure limits on a consolidated
a majority of the directors or trustees of the bank or basis, and for this reason, the proposed exposure limits could
company; or diminish a subsidiary bank’s lending and other extensions of

6
credit that would otherwise be permitted under applicable IMPORTANT CHANGES FROM THE
lending limit rules. 46
ORIGINAL PROPOSALS
Compliance Requirements and Effective Date Covered companies should see the New Proposal as an
The New Proposal would phase-in compliance and reporting improvement over the Original Proposals due to the reduction
with less time offered for larger covered companies to come into in the overall number and aggregate amount of counterparty
compliance. Covered companies with $250 billion or less in total exposures that would exceed the limit. Based upon the estimate
consolidated assets or $10 billion or less of total on-balance- of the impact of the New Proposal provided by staff of the Board,
sheet foreign exposures would have two years from the effective the total amount of excess credit exposure of U.S. BHCs would
date of a final rule to comply on a quarterly basis and submit a be less than $100 billion, with most of this exposure between
report demonstrating compliance on a quarterly basis. the largest BHCs and the largest counterparties.48 Nonetheless,
some troublesome features of the New Proposal remain intact,
Covered companies with $250 billion or more in total consolidated and some parts of the New Proposal are more stringent than
assets, or $10 billion or more of total on-balance-sheet foreign what was originally proposed. Covered companies will incur
exposures, would have one year from the effective date of a final compliance costs and burdens in order to attain full compliance
rule to comply on a daily basis and submit a report demonstrating if the New Proposal is adopted without change.
compliance on a monthly basis, unless the Board determines that
more frequent compliance and reporting is necessary. Measure of the Eligible Capital Base Against which
Exposure Limits Would Be Determined
Under the New Proposal, covered companies that fail to comply Section 165(e) directs the Board to set single-counterparty
with the final rule for a very limited set of reasons may be granted credit limits based on a company’s “capital stock and surplus”
a “temporary exception” from enforcement actions for a period and allows the Board to set “such lower amount as the Board
of 90 days (or a different period as determined by the Board to may determine by regulation to be necessary to mitigate risks
preserve safety and soundness or U.S. financial stability) as long to the financial stability of the United States.” Section 165(e)
as the company uses reasonable efforts to return to compliance does not define the measure of “capital stock and surplus”
during the period of time that the temporary exception is against which the single-counterparty credit exposure applies.
in place. A covered company may be granted a temporary The Original Proposals defined “capital stock and surplus”
exception from enforcement actions based upon a decrease in broadly to mean, for BHCs, total regulatory capital and any
the company’s capital stock and surplus, in the case of certain ALLL that does not count as Tier 2 capital, and for IHCs as the
mergers of two companies or unaffiliated counterparties, and in “sum of the [U.S. IHC’s] total regulatory capital as calculated
other appropriate circumstances as determined by the Board. under the risk-based capital adequacy guidelines applicable
to the [U.S. IHC], plus the balance of the ALLL of the U.S. IHC
A covered company that is subject to the 90-day temporary not included in Tier 2 capital under the capital adequacy
exception period would be prohibited from engaging in guidelines.”49 For an FBO, “capital stock and surplus” was
any additional credit transactions with a counterparty in defined as “the total regulatory capital of the [FBO] on
contravention of the rule during that period, except where the a consolidated basis, as determined in accordance with
Board determines that “such credit transactions are necessary [enhanced risk based capital and leverage requirements].”50
or appropriate to preserve the safety and soundness of the
covered company or financial stability.”47 The Basel Committee’s Large Exposures Framework employs
the more stringent Tier 1 capital measure for all companies. The
Despite commenters having suggested the addition of a Board highlighted that at least one commenter to the Original
compliance transition period for any company that becomes Proposals noted that a central finding of the financial crisis
a major covered company or major counterparty, the New was that only common equity was reliably loss absorbing, and
Proposal does not add such a feature, and there is no further observed that the Basel III capital standards reflect this
articulated consideration of these comments in the preamble through redefinition of capital instruments. This commenter also
to the New Proposal. argued that there are advantages to coordinating regulatory

7
capital definitions around a limited number of capital definitions Broader Scope of Covered Subsidiaries
51
that include only instruments that are reliably loss absorbing. As with the Original Proposals, the New Proposal applies to the
credit exposures of a covered company on a consolidated basis,
The New Proposal reflects consideration of both the Basel including any subsidiaries, to any unaffiliated counterparty.53
Committee’s Large Exposures Framework and public However, where the Original Proposal adopted what amounts to
comments filed in response to the Original Proposals. The a more a modest approach to identifying subsidiaries to those
relevant capital base in the New Proposal distinguishes where the company (i) owns, controls, or holds with power to vote
between BHCs that are internationally active and those that are 25 percent or more of a class of a company’s voting stock; (ii)
not. For BHCs that are internationally active, the New Proposal owns or controls 25 percent or more of a company’s total equity;
would apply the stricter Tier 1 capital measure adopted in the or (iii) is consolidated for accounting purposes, the New Proposal
Basel Committee’s Large Exposures Framework. BHCs with expands the definition of subsidiary to include entities to which a
$50 billion or more in total consolidated assets that are not covered company has “a controlling influence” over the entity’s
internationally active would use total regulatory capital plus management or policies. Commenters had recommended a
ALLL not included in Tier 2 capital as the eligible capital base. more simplified approach where the aggregate exposure of a
company would be based on accounting consolidation only.
The New Proposal would use Tier 1 capital as the eligible
capital base against which single-counterparty credit limits The New Proposal does not include as subsidiaries any
would be measured for the covered companies in the second investment funds or vehicles advised or sponsored by the
and third categories. This is based upon the Board’s stated company, but the Board has requested comment on whether
concern that the failure of a large, complex institution is more those types of companies should be included.
likely to have an adverse impact on the financial stability of
other financial institutions. Tier 1 capital is a higher-quality, Broader Scope of Covered Counterparties with More
more reliable form of capital that can absorb losses on a Complexity to Making Determinations
going-concern basis. Total regulatory capital plus the ALLL Economic Interdependence of Counterparties. Under the
not included in Tier 2 capital includes capital elements that New Proposal, a covered company would be required to
do not absorb losses on a going-concern basis, such as aggregate exposures to counterparties that are considered
subordinated debt, which is senior in the creditor hierarchy to “economically interdependent.”54 Economically interdependent
equity and takes losses only after a company’s equity is gone. counterparties are those where the failure or distress of one
counterparty would cause the failure or distress of the other.55
Credit Exposure Limits Between SIFIs All covered companies would be required to assess the
The New Proposal applies a tighter credit exposure limit between economic interdependence of counterparties, in accordance
SIFIs—15 percent as opposed to 25 percent for other exposures. with a proposed list of factors, when a covered company’s
The 15 percent limit is an expansion from the 10 percent limit exposure to one of the counterparties exceeds 5 percent
in the Original Proposals. Commenters criticized the Original of the covered company’s eligible capital (Tier 1 capital for
Proposals for failing to provide adequate reasons for selecting covered companies with $250 billion or more in assets and
a $500 billion asset threshold as the cutoff for the higher 25 total regulatory capital plus ALLL not included in Tier 2 capital
percent limit set by Section 165(e) of the Dodd-Frank Act. for companies with $50 billion or more in assets). The concept
of economic interdependence and the 5 percent threshold
With the New Proposal, the Board included a White Paper is derived from the Basel Committee’s Large Exposures
describing the reasons for imposing stricter single- Framework, but the factors are slightly different under the New
counterparty credit limits on larger institutions.52 Proposal. A comparison of the factors is set forth in the chart
below:56

8
New Proposal Basel Committee’s Large Exposures Framework
Whether 50 percent or more of one counterparty’s gross Same
revenue or gross expenditures are derived from transac-
tions with the other counterparty
Whether one counterparty (counterparty A) has fully or Same
partly guaranteed the credit exposure of the other coun-
terparty (counterparty B), or is liable by other means, and
the credit exposure is significant enough that counterparty
B is likely to default if presented with a claim relating to
the guarantee or liability
Whether 25 percent or more of one counterparty’s pro- Where a significant part of one counterparty’s production/
duction or output is sold to the other counterparty, which output is sold to another counterparty, which cannot easily be
cannot easily be replaced by other customers replaced by other customers
Whether one counterparty (counterparty A) has made When the expected source of funds to repay each loan one
a loan to the other counterparty (counterparty B) and is counterparty makes to another is the same and the counter-
relying on repayment of that loan in order to satisfy its party does not have another source of income from which the
obligations to the covered company, and counterparty A loan may be fully repaid
does not have another source of income that it can use to
satisfy its obligations to the covered company
Whether the expected source of funds to repay any credit No mirror provision; Basel refers only to “loans” and funding
exposure between the counterparties is the same and
at least one of the counterparties does not have another
source of income from which the extension of credit may
be fully repaid
Whether the financial distress of one counterparty Where it is likely that the financial problems of one counter-
(counterparty A) is likely to impair the ability of the other party would cause difficulties for the other counterparties in
counterparty (counterparty B) to fully and timely repay terms of full and timely repayment of liabilities
counterparty B’s liabilities
No mirror provision; “financial distress” provision likely Where the insolvency or default of one counterparty is likely to
includes insolvency or default be associated with the insolvency or default of the other(s)
When both counterparties rely on the same source of the When two or more counterparties rely on the same source for
majority of their funding and, in the event of the common the majority of their funding and, in the event of the common
provider’s default, an alternative provider cannot be found provider’s default, an alternative provider cannot be found,
the funding problems of one counterparty are likely to spread
to another due to a one way or two-way dependence on the
same main funding source
Any other indicia of interdependence that the covered No similar provision
company determines to be relevant to this analysis
Critics of the Large Exposures Framework have argued that This may be a significant undertaking for covered companies,
the factors are subjective and will likely require fact-intensive particularly smaller covered companies, which may not have
reviews of counterparty interconnectedness. This would access to this information. Moreover, even though investment
require extensive due diligence, which can be especially funds and vehicles advised or sponsored by a counterparty
burdensome for smaller-sized covered companies. would not need to be aggregated under the New Proposal,
the analysis that covered institutions must perform is likely to
Control Relationships of Counterparties. The New Proposal capture these entities anyway.
would require companies to add exposures to counterparties
that are connected by certain control relationships, including Statutory Attribution Rule
the presence of voting agreements, a counterparty’s influence Like the Original Proposals, the New Proposal includes the
over another counterparty’s management or policies (applying statutory attribution rule, which provides that a covered
the Bank Holding Company Act’s “controlling influence” company must treat a transaction with any person as a credit
test),57 and the ability of a counterparty to appoint or dismiss exposure to a counterparty to the extent the proceeds of the
members of another counterparty’s management or board.58 transaction are used for the benefit of, or transferred to, that

9
counterparty.59 In the Original Proposals, the Board recognized FBOs would be placed at a significant disadvantage under the
that “an overly broad interpretation of the attribution rule … New Proposal because they would not be permitted to use
would lead to inappropriate results and would create a the internal models methodology for measuring derivatives
60
daunting tracking exercise for covered companies” and exposure to a counterparty. FBOs would be required to use
proposed to minimize the scope of the application of the CEM or the calculation set forth at 12 CFR § 217.132(c), which
attribution rule but did not set forth how it would do that. establishes the methodology for determining exposure at
default for an over-the-counter (“OTC”) derivatives contract
As part of the New Proposal, the Board states its “intention to that is not subject to a qualifying master netting agreement.
avoid interpreting the attribution rule in a manner that would
impose undue burden on covered companies by requiring Risk-Shifting for Credit and Equity Derivatives
firms to monitor and trace the proceeds of transactions made
in the ordinary course of business” and, therefore, “credit The Original Proposals would have given a covered company the
exposures from transactions made in the ordinary course of option to reduce exposures to a counterparty based on eligible
61
business will not be subject to the attribution rule.” Again, collateral or an eligible guarantee and would have required a
however, the Board fails to provide any guidance on how covered company that purchased credit default swap (“CDS”)
covered companies should determine which transactions are protection to hedge the credit risk of making a loan to another
“made in the ordinary course of business.”62 firm or a sovereign, to recognize the dollar-for-dollar increase
in exposure to the CDS protection provider. Many commenters
Exposure Methodology for Derivatives Transactions believed that this approach was too conservative since a CDS
The Original Proposals required the use of a “current exposure purchaser would realize a loss only upon default of the original
method” (“CEM”) of measuring derivatives counterparty credit borrower and the CDS protection provider. These commenters
risk. A substantial number of commenters objected to the suggested measuring exposures from derivatives hedges using
use of the CEM measure on the ground that it significantly the same methodology used for derivatives.
overstates derivatives counterparty exposures due to the
limited recognition of netting benefits. After issuance of The New Proposal is more stringent than the Original Proposals
the Original Proposals, the Basel Committee developed as covered companies would be required to reduce exposure
a Standardized Approach to Counterparty Credit Risk of to a counterparty based on eligible collateral and would be
63
Derivatives (“SA-CCR”), and the Basel Committee’s Large required to recognize the exposure to the CDS protection
Exposures Framework employs this approach. provider. However, in cases where a company hedges its
exposure to an entity that is a non-financial counterparty, the
Under the New Proposal, instead of a requiring use of a New Proposal would permit covered companies to calculate
CEM measure, covered companies would be permitted exposure to protection providers using the counterparty credit
to calculate their potential future exposure to derivatives risk methodology for derivatives under the risk-based capital
counterparties (other than credit and equity derivatives) using rules, consistent with commenters’ suggestions.
any methodology that they are allowed to use under the risk-
based capital rules. These methodologies would include Exposure Methodology for Securities
CEM for all covered companies and the internal models Financing Transactions
methodology for covered companies subject to the Board’s The exposure methodology for securities financing
advanced approaches risk-based capital rules. Notably, these transactions (repos, reverse repos, and securities lending and
other methodologies would permit netting and the recognition borrowing transactions) in the New Proposal remains largely
of collateral that will mitigate counterparty credit risk. The New unchanged from the Original Proposals. The Original Proposals
Proposal does not adopt the Basel Committee’s SA-CCR for assumed a 10-day collateral liquidation period and employed
measuring credit exposure to a derivatives counterparty, but several conservative assumptions about correlations among
the Board notes that it may consider incorporating SA-CCR securities that are loaned and securities that are received
into single counterparty credit limit requirements at a later time. as collateral. The proposed methodology in the Original and

10
New Proposals includes the use of standardized supervisory would be required to recognize an exposure to each issuer of the
haircuts and accounting for any market fluctuations in eligible assets held by the SPV if the company could not demonstrate
collateral. Companies would be prohibited from applying that its exposure to the issuer of each underlying asset held by
internal estimates for haircuts and would be required to an SPV is less than 0.25 percent of the company’s Tier 1 capital.
disregard any collateral that does not meet the definition of If the covered company can demonstrate that its exposure to
64
“eligible collateral” in the New Proposal. each underlying asset in an SPV is less than 0.25 percent of
Tier 1 capital, the company would be permitted to recognize
Many commenters to the Original Proposals argued that an exposure solely to the SPV and not to the underlying assets.
the Board’s methodology for netting securities financing
transactions was too conservative. These commenters pointed If a covered company in applying the “look-through” approach
out that under the Board’s risk-based capital rules, collateral is unable to identify an issuer of assets underlying an SPV,
volatility haircuts for securities lending and repurchase the company would be required to attribute the exposure to a
transactions may be multiplied by the square root of ½ to single “unknown counterparty.” The covered company would
reflect a five-day liquidation period, rather than the ten-day then be required to aggregate all exposures to an unknown
period for other transaction types.65 counterparty as if they related to a single counterparty. In
addition, covered companies with more than $250 billion in
The preamble to the New Proposal explains that the Board assets would be required to recognize an exposure to third
considered several alternatives such as (i) applying valuation parties whose failure would likely result in a loss in the value
adjustments on one side of the transaction, (ii) the formula of the company’s investment in the SVP that is equal to the
recently proposed by the Basel Committee66 where an entity’s value of the investment in the SVP. The look-through approach
exposure for repo-style transactions would be equal to 40 is consistent with the Basel Committee’s Large Exposures
percent of its “net exposure” from the transaction plus 60 Framework and is designed to strengthen the oversight of
percent of its “gross exposure” divided by the square root so-called “shadow banking” institutions.
of the number of security issues in the netting set, and (iii)
using standardized correlation matrices. Ultimately, the Board There are numerous critics of the Basel Committee’s look-
rejected these alternatives on the ground that they sometimes through approach that are likely to be similarly unhappy
make improper assumptions about the correlation of securities with the New Proposal. In particular, the requirement that
(as is the case with valuation adjustments) and, with respect a company treat an exposure to an SPV as an exposure to
to the Basel Committee’s formula and the standardized interconnected third parties as well creates a risk of counting
correlation matrices, on the grounds that they may be overly these exposures twice. Complying with the look-through
complex and subject the framework to arbitrage. However, the approach would also require that funds provide frequent
Board did move from a 10-day liquidation period to a five-day information to banks about the funds’ holdings, which may
liquidation period in the New Proposal. prove to be an expensive endeavor.
“Look-Through Approach” to Exposures to Special Exemptions for Exposures to the Government and
Purpose Vehicles Government-Sponsored Enterprises
The Original Proposals reserved the Board’s authority to require The Original Proposals would have exempted exposures to the
banks to “look-through” their securitization funds, investment U.S. government, including Fannie Mae and Freddie Mac while
funds, and other special purpose vehicles (altogether “SPVs”) operating under conservatorship or receivership of the Federal
either to the issuers of the underlying assets in the vehicle Housing Finance Agency.67 The Basel Committee’s Large
or to the sponsor. The New Proposal would adopt the “look- Exposures Framework exempts exposures to all sovereigns
through” approach. and central banks and to many government-sponsored entities.
Under the “look-through” approach, covered companies that The Original Proposals also treated transactions with central
have $250 billion or more in total consolidated assets or $10 counterparties the same as any other derivatives transaction.
billion or more in total on-balance-sheet foreign exposures Many commenters supported expanding this exemption to

11
include creditworthy non-U.S. sovereigns, U.S. States and their foreign exposures, would face the same compliance
agencies, political subdivisions, and instrumentalities, and requirements as the Original Proposals. These companies
central counterparties. would be required to comply on a daily basis and submit a
compliance report on a monthly basis.
The New Proposal is more aligned with the Basel Committee’s
Large Exposures Framework and the comments to the Original Despite commenters having suggested the addition of a
Proposals. The New Proposal would provide exemptions or compliance transition period for any company that becomes
exclusions for credit exposures to: a major covered company or major counterparty, the New
Proposal does not add such a feature and there is no
• The U.S. government, including U.S. government agencies, articulated consideration of these comments in the preamble
and Fannie Mae and Freddie Mac while operating under to the New Proposal.
conservatorship or receivership of the Federal Housing
Finance Agency;
• Foreign sovereign entities that are assigned a zero percent KEY DIFFERENCES BETWEEN THE LARGE
risk weight under the Board’s capital rules; EXPOSURES FRAMEWORK ADOPTED BY THE
• Trade exposures to qualifying central counterparties; BASEL COMMITTEE ON BANKING SUPERVISION
• Intraday credit exposure to a counterparty; and AND THE NEW PROPOSAL
• The Federal Home Loan Banks.68
Following the financial crisis, the Basel Committee began
In addition, the New Proposal provides an exemption for the revising its existing capital adequacy guidelines and developed
exposures of an FBO to its home country sovereign. However, new capital and liquidity requirements (“Basel III”) designed
the Board did not exempt U.S. States and their agencies, to strengthen the regulatory capital regime for internationally
instrumentalities, and political subdivisions from the credit active banks.71 In 2011, as part of the Original Proposals, the
exposure limits although commenters expressed support for Board announced that it would implement substantially all of the
these exemptions. Basel III capital rules.72 In April 2014, after the Board published
the Original Proposals, the Basel Committee finalized its Large
Compliance Requirements and Reporting Exposures Framework, which establishes credit exposure limits
The Original Proposals treated all covered companies the for internationally active banks.73 One of the goals of the Large
same with respect to their compliance obligations—all Exposures Framework is to “help ensure a common minimum
covered companies would have been required to comply on a standard for measuring, aggregating and controlling single
daily basis and submit a monthly report.69 Covered companies name concentration risk across jurisdictions.”74
with $250 billion or less in total consolidated assets and less
than $10 billion in on-balance-sheet foreign exposure face less The Large Exposures Framework is generally less stringent than
stringent compliance requirements under the New Proposal. the Board’s Original Proposals. A 25 percent credit exposure
These covered companies would be required to comply on limit to a single counterparty is imposed on all internationally
a quarterly basis and report compliance on a quarterly basis. active banks under the Large Exposures Framework, except
However, the New Proposal states that these institutions would for G-SIBs, which are subject to a 15 percent credit exposure
need to have systems in place that allow them to calculate limit for exposures to other G-SIBs (as identified by the Basel
compliance on a daily basis, and would need to calculate Committee). Jurisdictions may consider applying stricter
compliance more often if directed to do so by the Board.70 limits to domestic SIFIs and for exposures to G-SIBs of smaller
banks. Unlike the Original Proposals or the New Proposal,
A covered company with total consolidated assets of $250 the Large Exposures Framework uses Tier 1 capital as the
billion or more, or $10 billion or more of on-balance-sheet eligible capital base against which the single-counterparty

12
credit limits apply for all covered institutions. Like the New • All “internationally active banking organizations” are
Proposal, the credit limits imposed under the Large Exposures subject to the Large Exposures Framework with Basel
Framework apply to a bank’s exposure to identified groups of Committee member jurisdictions having the option to set
connected counterparties. That is, the credit limits apply to more stringent standards and to extend the application to
counterparties that are interdependent and may be likely to a wider range of banks. Only domestic BHCs, FBOs, and
fail simultaneously. U.S. IHCs with total consolidated assets of $50 billion or
more are subject to the New Proposal.
The Large Exposures Framework is scheduled to take effect on • The Large Exposures Framework uses Tier 1 capital as the
January 1, 2019. The European Union (“EU”) has implemented denominator for all banks; the New Proposal uses Tier 1
much of Basel III through the Capital Requirements capital as the denominator only for banks with $250 billion
75
Regulation (“CRR”), a EU-wide “single rulebook” for capital or more in total consolidated assets.
requirements, which took effect starting January 1, 2015. The • The Large Exposures Framework defines an affiliate as a
CRR has comprehensive provisions on large exposures and company that is owned 50 percent or more by the bank,
counterparty concentration risk76 which build on those of whereas the New Proposal sets the ownership threshold
Basel II (“Limits to Large Exposures”). at 25 percent or more.
• Under the Large Exposure Framework, a banking
In brief, CRR’s Limits to Large Exposures provides that an organization must report to its supervisor when its
institution shall not incur an exposure, after taking into exposure to a single counterparty reaches 10 percent of
77
account the effect of credit risk mitigation to a “client” or eligible capital. There is no similar provision in the New
group of connected “clients” the value of which exceeds Proposal, which instead requires periodic compliance
25 percent of its eligible capital. Eligible capital in the EU reporting based upon asset size.
under the CRR is defined as the sum of Tier 1 capital and • The Large Exposures Framework exempts exposures to all
Tier 2 capital that is equal to or less than one-third of Tier 1 sovereigns; for BHCs, the New Proposal exempts exposures
capital.78 Where that client is an institution or where a group of to the U.S. government (not states), foreign sovereigns with
connected clients includes one or more institutions, that value a 0 percent risk weight under the Board’s Basel III capital
cannot exceed 25 percent of the institution’s eligible capital or rules, and, for FBOs and U.S. IHCs, the New Proposal
EUR 150 million, whichever is higher, provided that the sum of exempts exposures to the home country sovereign.
exposure values, after taking into account the effect of credit • The Large Exposure Framework adopts SA-CCR for
risk mitigation in accordance with Articles 399 to 403,79 to all measuring credit exposures to a derivatives counterparty,
connected clients that are not institutions does not exceed 25 which is not a component of the New Proposal.83
percent of the institution’s eligible capital.80 Large exposures • The Large Exposures Framework is expected to be fully
are defined in the CRR as exposures to a client or group of implemented by January 1, 2019; the earliest compliance
connected clients which, in the aggregate, equal or exceed date for certain covered companies subject to the New
10 percent of the institution’s eligible capital.81 There is a list Proposal is one year after the final rule’s effective date.
of exemptions from the definition of large exposures which
allows some flexibility from the strict application of the rules.82 Covered companies subject to the New Proposal are most
However, for thinly capitalized banks with a relatively small likely to be impacted by the difference between the Large
client base, the rules prove perennially difficult from the point Exposures Framework and the New Proposal with respect
of view of capital constraints. to the aggregation of connected or affiliated counterparties.
Covered companies subject to the limits of the New Proposal
While the New Proposal incorporates much of the Basel are likely to see far more single-counterparty credit exposures
Committee’s Large Exposures Framework, there are several than under the Large Exposures Framework due to the 25
important differences between the New Proposal and the percent ownership or control threshold that applies under the
Large Exposures Framework: New Proposal (versus the 50 percent threshold under the Large

13
Exposures Framework). This ownership information is also likely preventing evasion of the single-counterparty credit limits
to be more difficult to obtain as 25 percent ownership may not without imposing undue burden on covered companies?
be reflected in an organization’s financial statements. • Is additional regulatory clarity around the attribution
rule necessary?
• What is the potential cost or burden of applying the
SUBJECTS FOR POSSIBLE COMMENT attribution rule as proposed?
Several features of the New Proposal are ripe for comment, 6. Exemptions
including specific questions posed by the Board. These topics • Should all trade exposures to QCCPs be exempt
and questions include: from the proposed rules? Is the definition of “QCCP”
sufficiently clear? Should the Board consider exempting
1. Capital Base any different or additional exposures to QCCPs? Would
• Are the definitions relating to capital stock and surplus additional clarification on these issues be appropriate?
and Tier 1 capital clear? • Should the Board exempt any additional credit
• Should the single-counterparty credit limits applicable exposures, such as exposures to U.S. states, from the
to covered companies with $250 billion or more in total limitations of the proposed rule? Why?
consolidated assets be based on a different capital
base than that used for other covered companies? 7. Aggregation of Company and Subsidiary Exposures
• Is it appropriate to apply the limits of the New Proposal
2. Asset Thresholds on a consolidated basis?
• Should more stringent credit exposure limits apply to • Should the definition of subsidiary be based on the
credit exposures of a major covered company to a major definition in the Bank Holding Company Act or should it
counterparty than would apply to other exposures? be limited to an entity that a covered company (i) owns,
• Are the definitions of major covered company and control, or holds with power to vote 25 percent or more
major counterparty appropriate? of a class of voting securities; (ii) owns or controls 25
• Should more stringent credit exposure limits apply to percent or more of the total equity; or (iii) consolidates
exposures of major covered companies to a nonbank for financial reporting purposes?
financial company subject to Board supervision? • Should funds or vehicles that a company sponsors
• Should the Board consider other limits or modifications or advises be included as subsidiaries of a covered
to the proposed limits? company?
3. Differences from Basel Committee’s Large Exposures 8. Economic Interdependence and Control Relationship of
Framework Counterparties
• Will the differences from the Basel Committee’s Large • Should covered companies be required to aggregate
Exposures Framework cause difficulty for internationally exposures to entities that are economically
active banks? interdependent? Should they be required to aggregate
• Should the Board adopt SA-CCR? exposures to entities that are connected by certain
control relationships?
4. Exposures to Individuals • Are the factors for determining economic
• Does including exposures to individuals raise interdependence sufficiently clear? For determining
compliance burdens for covered companies? What are whether a control relationship exists?
the burdens? • Are the thresholds for recognizing economic
interdependence appropriate?
5. Attribution Rule • Will companies have access to all of the information
• What ways can the Board apply the statutory attribution required to complete an analysis of economic
rule in a manner that would be consistent with the goal of interdependence? Is this type of information collected

14
in the ordinary course of business as part of an • Should the rule provide a cure period for covered
underwriting or similar process? companies that fall out of compliance? Under what
circumstances should such a cure period be provided,
9. Eligible Collateral for Securities Financing Transactions and how long should such a period be?
• Should the list of eligible collateral be broadened or • If a cure period is provided, would it be appropriate to
narrowed? What should be added? Deleted? generally prohibit additional credit transactions with the
• Should covered companies be given the option to reduce affected counterparty during the cure period?
their gross credit exposures by recognizing eligible • Are there additional situations in which additional
collateral in some or all cases? Are there situations in credit transactions with the affected counterparty
which full shifting of exposures would not be appropriate? would be appropriate? What additional modifications
• Are the market volatility haircuts appropriate for the or clarifications should the Board consider with respect
valuation of eligible collateral? to any cure period?
10. Look-Through Approach for SPVs

• Is the proposed treatment of a covered company that LAWYER CONTACTS
has less than $250 billion or more in total consolidated
assets and less than $10 billion or more in total For further information, please contact your principal Firm rep-
on-balance sheet foreign exposures with respect resentative or one of the lawyers listed below. General email
to its exposures related to SPVs appropriate? What messages may be sent using our “Contact Us” form, which can
alternatives should the Board consider? be found at www.jonesday.com/contactus/.
• Is the proposed treatment of a covered company with
$250 billion or more in total consolidated assets or Lisa M. Ledbetter Robert J. Graves
$10 billion or more in total on-balance-sheet foreign Washington Chicago
exposures with respect to its exposures related to SPVs +1.202.879.3933 +1.312.269.4356
appropriate? lledbetter@jonesday.com rjgraves@jonesday.com
• Are there situations in which the proposed treatment
would result in recognition of inappropriate amounts of Courtney L. Snyder Alban Caillemer du Ferrage
credit exposure concerning an SPV? Pittsburgh Paris
• What alternative approaches should the Board +1.412.394.7910 +33.1.56.59.38.18
consider? clsnyder@jonesday.com acf@jonesday.com
• Is the proposed treatment of exposures related to SPVs
sufficiently clear? John C. Ahern Philippe Goutay
• Would further clarification or simplification be London Paris
appropriate? What modifications should the Board +44.20.7039.5176 +33.1.56.59.46.58
consider? jahern@jonesday.com pgoutay@jonesday.com
11. Compliance Burdens Brett P. Barragate Edward J. Nalbantian

• Should the Board consider a longer or shorter phase-in New York London
period for all or a subset of covered companies? +1.212.326.3446 +44.20.7039.5145
• Is a shorter phase-in period for covered companies with bpbarragate@jonesday.com Paris
$250 billion or more in total consolidated exposures, +33.1.56.59.39.23
or $10 billion or more in total on balance-sheet foreign enalbantian@jonesday.com
exposures, appropriate compared to firms below these
thresholds?

15
ENDNOTES
1 Public Law 111-203, 124 Stat. 1376 (2010). 24 The Clearing House Association, Single Counterparty Credit Limits:
The Clearing House Industry Study (July 2012).
2 77 Fed. Reg. 594 (Jan. 5, 2012); 77 Fed. Reg. 76628 (Dec. 28, 2012)
(collectively “Original Proposals”); 79 Fed. Reg. 17240 (March 27, 25 New Proposal, 81 Fed. Reg. 14333.
2014) (codified at 12 C.F.R. part 252).
26 The G-SIBs are Bank of America, Bank of New York Mellon,
3 Single-Counterparty Credit Limits for Large Banking Organizations, Citigroup, Goldman Sachs, J.P. Morgan, Morgan Stanley, State
81 Fed. Reg. 14327 (proposed March 16, 2016) (to be codified at 12 Street, and Wells Fargo, as of March 16, 2016. Financial Stability
C.F.R. part 252) (“New Proposal”). Board, 2015 update of list of global systemically important banks
(Nov. 3, 2015).
4 12 U.S.C. § 5365(e).
27 Board of Governors of the Federal Reserve System, Calibrating
5 The proposed limits would apply to FBOs with $50 billion or more Single-Counterparty Credit Limit Between Systemically Important
in total worldwide consolidated assets. New Proposal, 81 Fed. Reg. Financial Institutions (March 4, 2016) (“White Paper”).
14345.
28 Id. at 6.
6 Opening Statement on the Proposed Rule Establishing Single-
Counterparty Credit Limits for Large Banking Organizations by 29 Comparable charts appear in the Staff Memo, at 4-5, 11, see supra
Board Chair Janet Yellen (March 4, 2016). note 9.
7 Id. 30 New Proposal, 81 Fed. Reg. 14351, proposed § 252.71(b).
8 Id. 31 Id., proposed § 252.73.
9 Memorandum to the Board of Governors of the Federal Reserve 32 12 C.F.R. part 32; 78 Fed. Reg. 37930 (June 25, 2013).
System from Governor Tarullo on Proposed rules to implement
single-counterparty credit limits in section 165(e) of the Dodd-Frank 33 12 C.F.R. part 217, subpart D.
Act (February 26, 2016) (“Staff Memo”).
34 New Proposal, 81 Fed. Reg. 14350, proposed § 252.71(e).
10 Specifically, nonbank SIFIs are included as counterparties for pur-
poses of the 15 percent limit that applies to major covered compa- 35 New Proposal, 81 Fed. Reg. 14332 (Question 6).
nies. See New Proposal, 81 Fed. Reg. 14330.
36 Enhanced Prudential Standards for Bank Holding Companies and
11 Basel Committee on Banking Supervision, Bank for International Foreign Banking Organizations, 79 Fed. Reg. 17240 (March 27, 2014)
Settlements, Standards: supervisory framework for measuring and (codified at Regulation YY, 12 C.F.R part 252).
controlling large exposures (April 2014).
37 Section 2(h)(2) of the Bank Holding Company Act allows qualifying
12 12 U.S.C. § 5365. Section 165 of the Dodd-Frank Act defines a “bank FBOs to retain certain interests in foreign commercial firms that
holding company” consistent with section 2 of the Bank Holding conduct business in the U.S.
Company Act of 1956 (12 U.S.C. § 1841), and includes “a foreign
bank or company that is treated as a bank holding company for 38 New Proposal, 81 Fed. Reg. 14346; Staff Memo, see supra note 9 at
purposes of the Bank Holding Company Act of 1956, pursuant to 9. Some commenters to the Original Proposals as applied to FBOs
section 8(a) of the International Banking Act of 1978 (12 U.S.C. § argued that, in light of the Basel Committee’s Large Exposures
3106(a)).” Id. Framework that would apply to an FBO on a consolidated basis,
it was unnecessary for the Board to develop single-counterparty
13 12 U.S.C. § 5365(b)(1). credit limits for a FBO’s combined U.S. operations. New Proposal, 81
Fed. Reg. 14330.
14 Id. § 5365(a) and (e)(1).
39 81 Fed. Reg. 14346.
15 Id. § 5365(e)(3).
40 12 U.S.C. § 1841(a).
16 Id.
41 New Proposal, 81 Fed. Reg. 14331.
17 See supra note 2.
42 Original Proposals, 77 Fed. Reg. 614.
18 Id.
43 Id.
19 Lending Limits, 78 Fed. Reg. 37930 (June 25, 2013) (codified at 12
C.F.R part 32). 44 See, e.g., 12 U.S.C. 24(7); 12 U.S.C. 84; 12 C.F.R. parts 1 and 32; see
also 12 U.S.C. 335 (applying the provisions of 12 U.S.C. 24(7) to state
20 See supra note 11. member banks).
21 12 C.F.R. part 225 45 See 12 U.S.C. § 24(7); 12 C.F.R. part 1.
22 See 12 C.F.R. § 215.3(i), 12 C.F.R. § 223.3(d); see also 12 C.F.R. 46 New Proposal, 81 Fed. Reg. 14329.
§32.2(b).
23 See 12 C.F.R. Part 225.
48 Staff Memo at 3, see supra note 9.

16
49 Original Proposals, 77 Fed. Reg. 76655. 66 Basel Committee, Second Consultative Document, Standards:
Revisions to the Standardised Approach for credit risk, issued for
50 Id. comment by March 11, 2016 (December 2015).
51 Staff Memo at 15, see supra note 9. 67 Original Proposals, 77 Fed. Reg. 622.
52 See supra note 27; see also discussion supra pp. 3-4. 68 New Proposal, 81 Fed. Reg. 14343-44.
53 See discussion, supra p. 6 69 See discussion, supra p. 7.
54 See discussion, supra pp. 5-6. 70 New Proposal, 81 Fed. Reg. 14344.
55 New Proposal, 81 Fed. Reg. 14354, proposed § 252.76(a)(1)(ii). 71 Basel Committee on Banking Supervision, International Regulatory
Framework for Banks.
56 Id., proposed § 252.76(a)(2).
72 Original Proposals, 77 Fed. Reg. 599.
57 12 U.S.C. § 1841(a)(2).
73 See supra note 11.
74 Basel Committee on Banking Supervision, Press Release:
59 Id. at 14337, proposed § 252.73(c); see also 12 U.S.C § 5365(e)(4). Supervisory framework for measuring and controlling large expo-
sures—final standard (April 2014).
60 Original Proposals, 77 Fed Reg. 618.
75 Regulation (EU) No 575/2013 on prudential requirements for credit
61 New Proposal, 81 Fed. Reg. 14337. institutions and investment firms (CRR).
62 Id. 76 Id. at Articles 387-403.
63 Basel Committee on Banking Supervision, The Standardized Approach 77 Id. at Articles 399-403.
For Measuring Counterparty Credit Risk Exposures (April 2014).
78 Id. at Article 4(71).
64 Despite commenters request that the Board expand the defini-
tion of eligible collateral, the New Proposal retains the definition 79 Id. at Articles 399-403.
of eligible collateral from the Original Proposals, which limits debt
securities to bank-eligible investments that are investment grade. 80 Id. at Article 395.
New Proposal, 81 Fed. Reg. 14337. The definition of “eligible collat-
eral” for FBOs and IHCs excludes debt or equity securities issued 81 Id. at Article 392, 394.
by an affiliate of the FBO or IHC. Id. at 14347. This definition is more
restrictive than eligible collateral under the Basel Committee’s 82 Id. at Article 400.
capital rules, which would permit recognition of mutual funds and
money market shares. 83 The Board is considering the benefits of incorporating the SA-CCR
into the Single-Counterparty Credit Rule. See New Proposal, 81 Fed.
65 New Proposal, 81 Fed. Reg. 14336. Reg. 14337.
Jones Day publications should not be construed as legal advice on any specific facts or circumstances. The contents are intended for general
information purposes only and may not be quoted or referred to in any other publication or proceeding without the prior written consent of the
Firm, to be given or withheld at our discretion. To request reprint permission for any of our publications, please use our “Contact Us” form, which
can be found on our website at www.jonesday.com. The mailing of this publication is not intended to create, and receipt of it does not constitute,
an attorney-client relationship. The views set forth herein are the personal views of the authors and do not necessarily reflect those of the Firm.
© 2016 Jones Day. All rights reserved. Printed in the U.S.A.

QUANTITATIVE RESEARCH GROUP
JUNE 2015
MODELING Quantifying Risk Appetite in Limit Setting

METHODOLOGY
Authors Abstract
Andrew Kaplin
Amnon Levy An institution’s Risk Appetite Statement (RAS) specifies the aggregate level and types of risks the
Qiang Meng firm is willing to take (or avoid) in order to achieve its business goals. The translation of the RAS
Libor Pospisil
into limits allows an organization to achieve its strategic objectives and business plan while
Acknowledgements adhering to its risk capacity.
We would like to thank Ankit Rambhia, In this paper, we explore leveraging an organization’s economic capital framework to quantify the
Christopher Crossen, and Julie Sykes for their
comments.
RAS via risk- and macro scenario-based limits. Risk-based limits create a level playing field, using
risk-based metrics to align with the organization’s risk appetite. Macro scenario-based limits
Contact Us control exposures to adverse macroeconomic scenarios and are frequently viewed as more
Americas intuitive and more tangible than risk-based limits.
+1.212.553.1653
clientservices@moodys.com We also describe a number of approaches for setting risk- and macro scenario-based limits: top-
Europe of-the-house (TOTH) risk limits, standalone sub-portfolio (SASP) risk limits, portfolio referent sub-
+44.20.7772.5454 portfolio (PRSP) risk limits, as well as Stressed Expect Loss (SEL) Limits, and macro risk-based
clientservices.emea@moodys.com limits. To demonstrate the approaches for limit setting, we use a sample portfolio that consists of
Asia-Pacific (Excluding Japan) C&I, CRE, and retail exposures. We illustrate the relationship between varying sub-portfolio
+85 2 3551 3077 characteristics and the various approaches to setting limits.
clientservices.asia@moodys.com
Japan
+81 3 5408 4100
clientservices.japan@moodys.com
Table of Contents
1. Introduction 3
2. Risk-Based Limits 4
2.1 Top-of-the House Risk Limits 4
2.2 Sub-portfolio Risk Limits 5
2.3 Portfolio-referent Sub-portfolio Risk Limit 6
3. Macro Scenario (MS) Limits 10

3.1 Top-of-the House Macro Scenario Limits 11
3.2 Sub-portfolio Macro Scenario Limits 13
4. Practical Applications 16

4.1 Business Strategy and Operating Plan 16
4.2 Portfolio Dynamics 17
4.3 Sub-portfolio Limits and TOTH Limits 17
5. Conclusion 17
References 18
2 JUNE 2015 QUANTIFYING RISK APPETITE IN LIMIT SETTING

1. Introduction
An organization’s risk appetite statement (RAS) is the aggregate risk level and types of risk it is willing to accept (or avoid) in order
to achieve its business objectives. The RAS includes qualitative statements as well as quantitative measures, expressed relative to
earnings, capital, risk measures, liquidity, and other relevant measures.1 The RAS translation feeds into quantitative risk-based
limits and allows an organization to achieve its strategic objectives and business plan, while adhering to its risk capacity. Risk-
based limits allocate the financial institution’s aggregate risk appetite statement (e.g. measure of loss or negative events) to
business lines, legal entities as relevant, specific risk categories, concentrations, and other levels.
There is no question that articulating stakeholder preference and appetite for risk is challenging. The challenges are intimately
linked with the implications RAS has on metrics related to its monitoring and enforcement. Risk limits are a key mechanism of
enforcement and part of the overall risk appetite framework.
The limit setting process is multifaceted, with risk appetite playing a central role. And while the quantitative measures in the risk
appetite statement provide limit setting guidance, the qualitative portions must be incorporated as well.The challenge is in
quantifying an RAS that may not have obvious quantitative translations. A specific statement such as “maintain an Aa rating or
better” (perhaps with a certain likelihood), was part of the Royal Bank of Canada’s (RBC) RAS, profiled in the 2011 Institute of
International Finance (IIF) study on risk appetite. 2 This statement can be linked to quantitative performance measures by
quantifying the requirements needed to maintain the Aa rating. Some judgment is required, however, given agency ratings are not
entirely based on quantitative metrics.
An example of a more open-ended statement, “maintain low exposure to ‘stress events,’” is another item in RBC’s RAS. The
process of quantifying this statement requires defining “stress events” as well as “low exposure.” In the U.S., natural interpretations
of stressed events are the CCAR Adverse or Severely Adverse scenarios. A low exposure can be related to the sensitivity of
expected loss to the associated macro-economic variables.
In some cases, one can relate multiple elements in the RAS to allow for a more holistic quantitative view of risk appetite. For
example, the two statements above can describe RBC’s aversion to losing their Aa rating under the Adverse or Severely Adverse
scenarios. This can be made more precise by interpreting the risk appetite to target a portfolio composition and capital buffer so
that, say, the likelihood of losses exceeding those that would result in a downgrade are no higher than one-in-25 years, conditional
on the Adverse scenario.
Quantifying risk appetite should rely on sound economic principles, along with strategic overlay. It is natural to leverage an
economic capital framework in this context. For example, an economic capital framework can help us understand the likelihood of
a portfolio incurring losses that drive the institution to a downgrade. An EC framework can also allow further drill-down by
providing estimates for the likelihood of, for example, losses related to a sub-portfolio resulting in the downgrade. Alternatively,
the framework can be used to help understand portfolio loss dynamics, conditional on macroeconomic scenarios, to define limits
that will help avoid losses associated with stress events.
A subtle point worth mentioning is the role that limits play and how their role relates to other tools used to align incentives across
the organization. Limits are generally broad-brush mechanisms that ensure risk exposure conforms to the level and likelihood of
loss. Other tools, such as RORAC and EVA-style measures used in deal pricing and incentive compensation, allow an organization
to optimize their risk-return profile. The various tools complement each other, as the complex nature of the organization’s desire
to align various incentives and achieve its goal of maximizing return, while adhering to a desired risk profile, is too complex to
achieve using a single mechanism. We analyze a number of quantitative approaches in translating an RAS to limits, and we use
case studies to better understand the ensuing dynamics.
The remainder of this paper is organized as follows:
» Section 2 considers risk-based limits.
» Section 3 considers macro scenario-based limits.
» Section 4 considers practical issues when applying an economic capital framework when setting limits.
» Section 5 concludes.
1
Financial Stability Board, “Principles for an Effective Risk Appetite Framework.” November 18, 2013.
2
Institute of International Finance, “Implementing Robust Risk Appetite Frameworks to Strengthen Financial Institutions.” June 2011.

2. Risk-Based Limits
As defined by the IIF, risk limits represent quantitative measures based on forward-looking assumptions that allocate the financial
institution’s aggregate risk appetite statement (e.g., measure of loss or negative events) to business lines, legal entities as relevant,
specific risk categories, concentrations, and, as appropriate, other levels. This approach to limit setting is a departure from
traditional, notional-based limits. Risk-based limits create a level playing field, utlizing risk-based metrics that align with the
organization’s risk appetite. In this section, we use a quantification case study to demonstrate how to design risk-based limits that
address a statement such as, maintain an Aa rating (referenced in the Introduction). There are three sub-sections: 2.1 describes
how top-of-the house (TOTH) risk limits can be defined, 2.2 describes how stand-alone sub-portfolio (SASP) risk limits can be
defined, and 2.3 describes how portfolio-referent sub-portfolio (PRSP) risk limits can be defined.
2.1 Top-of-the House Risk Limits

Maintaining a high credit rating is not an easy task. There were 32 banks rated Aa or higher in 2014 (2004 – 2006 there were 58
such banks). When quantifying downgrade risk, one must understand the likelihood of various loss levels under the portfolio’s
existing composition and the required buffer that allows retention of an entity’s Aa rating. At the top-of-the house, a rating
transition matrix can provide a rough sense of the likelihood of a downgrade. An economic capital framework can refine this
process by more accurately mapping out the portfolio loss distribution, specifying the existing buffer, and quantifying the
likelihood of losses exceeding the threshold associated with a downgrade. In other words, the currency loss would be the
difference between capital thresholds associated with the current capital buffer and an A rating, as in Figure 1. The figure depicts
the loss distribution for the U.S. corporate sub-portfolio of the IACPM portfolio. Table 2 describes details of the portfolio. Under
the assumption3 that an Aa rating is associated with a 3bps – 10bps range of target probabilities, the portfolio maintains its Aa
rating as long as the existing capital falls into that range. The notional TOTH limit can be calculated using the ratio that reflects
how much growth the current capital can support without additional contributions, assuming that the portfolio composition
scales up proportionally. For example, if the current capital is at 10.5%, then the portfolio notional can grow by (10.5% – 8.3%) /
8.3% = 26.5% and still maintain that rating without injecting additional capital. This can be expanded by assuming a certain rate
of growth for capital that leads to higher limits, keeping everything else constant, as well as using a more conservative lower limit
(higher than 8.3%), which results in lower notional limits.
Two aspects to this approach are worth recognizing. First, the portfolio composition does not change as it grows. And, the buffer is
driven entirely by loss/gain on assets (i.e., liabilities are not changing). Section 4 further discusses these assumptions and also
considers practical applications.
Figure 1 Loss Distribution and Capital Thresholds Associated with Aa and A Ratings
3
We assume that the Aa rating will be maintained up to the 10bp threshold. Alternative thresholds are excluded for exposition.

2.2 Sub-portfolio Risk Limits

There are a number of ways to translate RAS into risk-based sub-portfolio limits. In this sub-section, we explore two approaches
that define sub-portfolio limits: stand-alone sub-portfolio and portfolio-referent sub-portfolio limits. We conduct the analysis
within the context of the case study above, where there is an appetite to maintain the Aa rating to construct industry-based risk
limits.
Proceeding with the case study, we analyze the portfolio in question for 14 industry segments. Figure 3 provides the notional
amount as well as the Risk Contribution. Notice the difference in the rank-ordering of the notional and the risk as it relates to limit
setting.
A practical consideration in setting limits is the wide cross-sectional variation in commitment amount that can proxy for the
organization’s strategic investment opportunity set. When setting limits, the idea of constructing a level playing field in the
notional or risk space makes sense from a theoretical perspective, but it does not consider the practicalities of the organization’s
investment strategies. The annual operating plan (AOP) must be overlaid in a way that allows the limits to provide realistic
guidelines that can relate back the RAS, which provides guidelines on the risk the financial institution is willing to accept (or avoid),
in order to achieve its business objectives. Section 4 discusses these considerations in greater detail.
SASP RISK LIMITS
We can construct stand-alone industry risk limits using each industry’s loss distribution and mapping the likelihood that loss
exceeds a threshold associated with a downgrade for the overall organization. Conceptually, the limit is not portfolio-referent, in
that, it does not account for correlation across industries, and the likely losses associated with correlated industries, when the
industry in question deteriorates. This property has drawbacks and benefits. By not accounting for portfolio concentration in
correlated industries, the approach does not recognize that more highly-correlated industries are riskier from a portfolio-referent
perspective. This said, that very characteristic can make it easier to manage, as various stakeholders can manage proximity to their
relevant limits without concern over how investments are managed elsewhere in the organization. This point is discussed further in
Section 2.3, which describes portfolio-referent sub-portfolio limits.
Table 1 provides some basic characteristics of the portfolio, segmented based on industry. Included is the commitment, stand-
alone Unexpected Loss (UL), risk contribution, and other relevant parameters of each sub-portfolio.
Table 1
Characteristics: U.S. Corporates Sub-portfolio
INDUSTRY NOTIONAL NOTIONAL NOTIONAL NOTIONAL NOTIONAL STANDALONE UL
WEIGHTED PD WEIGHTED WEIGHTED WEIGHTED RC
LGD RSQ
Agriculture 103,000,000 0.61% 38.66% 17.45% 0.71% 1,480,307
Banks and S&Ls 1,743,000,000 0.72% 45.80% 46.27% 1.72% 49,526,730
Business Products 176,000,000 3.49% 38.56% 11.26% 1.33% 5,612,098
Consumer Products 5,922,000,000 1.12% 39.27% 34.86% 0.92% 67,191,014
Equipment 1,147,000,000 3.36% 43.56% 32.23% 4.13% 81,834,392
Financials (Other) 2,525,000,000 0.33% 40.39% 51.54% 0.50% 19,801,516
High Tech 1,703,000,000 0.11% 44.38% 45.94% 0.38% 13,298,554
Materials/Extraction 9,057,000,000 1.81% 40.53% 33.02% 1.62% 169,944,117
Medical 212,000,000 1.04% 38.23% 11.45% 0.59% 2,610,645
Real Estate 2,166,000,000 0.82% 37.98% 34.64% 1.05% 53,584,208
Services 1,455,000,000 0.80% 41.59% 20.34% 0.62% 16,106,039
Tel/Cable/Printing/Publishing 6,604,000,000 1.74% 37.93% 35.26% 1.24% 114,211,073
Transportation 2,207,000,000 1.14% 40.43% 31.06% 1.21% 47,408,142
Utilities 1,212,000,000 3.07% 41.11% 38.50% 2.30% 36,953,750

SASP risk limits can be quantified using a risk space (e.g., UL or a tail-based) as the level playing field, whereby risk at the industry
segment is considered. An approach to defining sector limits would be to require that even a catastrophic case in that sector
would not downgrade the overall portfolio. For this process, we calculate capital for each sector at a very small target probability
and then calculate what sub-portfolio increase would lead to a currency-based loss leading to a downgrade
In the TOTH risk limit calculation in Section 2.1, the threshold for a currency-based loss is 2.2% (difference between 10.5% and
8.3%) of portfolio notional. Suppose Banks and S&L sector capital is 69.6%, and we know that it represents approximately 4.8%
of the overall portfolio (by commitment). Then we can increase the holdings in Banks and S&Ls (assuming no injection of new
capital and current capital is at 10.5%) by approximately 2.2%/69.6% or 3.2%4 of the overall portfolio. So the limit is
(3.2%+4.8%)*Current Overall Portfolio Commitment.
2.3 Portfolio-referent Sub-portfolio Risk Limit

Portfolio-referent sub-portfolio risk limits consider the risk of a segment in the context of the overall portfolio. The limits recognize
segments likely to incur substantial loss when the rest of the portfolio performs poorly as more risky. In this sense, segments that
contribute to a larger portion of loss when the overall portfolio deteriorates to the point of a downgrade would face a tighter
notional limit.
The choice of risk allocation measure, tail-based or Risk Contribution (RC)-based, is an important one, and a choice specific to an
institution’s risk preferences. For the purpose of this discussion, we focus on RC-based risk allocation, which measures an
instrument or portfolio segment‘s contribution to the portfolio UL. RC is formally defined as the change in portfolio standard
deviation (in currency unit) resulting from one additional currency unit of this particular instrument or portfolio segment. It can be
shown that RC equals UL of the instrument or segment multiplied by the correlation between the value of the instrument or
ULP
segment and value of the entire portfolio. Mathematically represented as follows: RCi    i ,P  ULi
wi
Visually, Figure 2 presents a traditional decomposition of an instrument or segment’s UL in the context of an overall portfolio. The
market. UL represents the stand-alone standard deviation, and it can be decomposed as a diversified portion, a diversifiable (but
not diversified) portion, and a systematic or non-diversifiable portion. The Risk Contribution is the sum of the systematic portion
and the diversifiable portion.
Figure 2 Decomposition of UL
From a diversification perspective, the worst situation is that the segment is perfectly correlated with the entire portfolio. This
issue occurs if the entire portfolio consists of only one segment, or if all segments are perfectly correlated. For such portfolios,
none of the segment’s stand-alone risk can be diversified away, and the segment RC is equal to segment UL. On the other hand, in
the ideal case where all diversifiable portions in segment UL are diversified away, RC will consist of only the systematic portion of
UL. In the real world, it is usually not possible to diversify away all diversifiable portions in segment UL, and the lower bound for RC
4
Capital rate associated with a catastrophic case — a 1bps event in this example — for the Banks and S&L sector.

is typically higher than the systematic portion in segment UL. This lower bound is portfolio-specific and we refer to it as portfolio-
referent systematic portion in this paper.
Continuing with the case of downgrade avoidance, portfolio-referent sub-portfolio risk limits can be constructed by measuring the
proportion of risk associated with each segment when a downgrade occurs. As a starting point, Figure 3 presents Commitment and
normalized Risk Contribution for each industry segment as a proportion of the overall portfolio Commitment and Unexpected
Loss, respectively. While many industries have a similar relationship between the proportion of risk and commitment, there are a
number of striking exceptions, including Financials and High Tech, whose portfolio-referent risk statistics (measured as proportion
of the portfolio’s UL) are lower than half the sector’s commitment as a proportion of portfolio’s commitment. On the flip side, a
proportion of the portfolio’s UL is more than three times higher than the sector’s commitment, as a proportion of the portfolio’s
commitment to the Equipment segment.
Figure 3 Commitment and Risk Contribution (as a proportion of the portfolio Commitment and Unexpected Loss,
respectively)
35.00%
30.00%
25.00%
20.00%
15.00%
10.00%
5.00%
0.00%
% of commitment Risk Contribution (% of portfolio UL)
When defining risk based limits, it is necessary to understand the relationship between the risk and amount invested in a segment.
Figure 4 illustrates how the RC of a segment changes when the weight of that segment increases.and presents the relationships for
the 14 sectors. Take Equipment as an example. The current RC is 4.13% and comprises 10% of the portfolio UL. As discussed
earlier, if the Equipment were 100% of the portfolio, the RC would represent the UL of Equipment. On the other hand, the
normalized RC when Equipment is set at 0% represents a measure of portfolio-referent systematic risk related to Equipment.
We can now apply the analysis more broadly and define Risk Contribution-to-Portfolio (RCP) as the segment RC measured as a
proportion of the entire portfolio UL: the share of portfolio UL attributed to a particular segment. The RCP is a function of segment
weight and increases from 0, when the weight is 0, to 100%, when the weight is 100%. Now we are ready to translate the RC-
based limits to notional-based limits.
Suppose no more than 35% of portfolio risk should be allocated to any segment in order for a bank to maintain an Aa rating.5 For
RC-based capital allocation, this mandate is equivalent to requiring that the RCP should not exceed 35% for any segment. Since
RCP is an increasing function of segment weight and takes a value between 0 and 100%, there is a unique value for segment
weight, such that, RCP equals 35%. Figure 4 illustrates the translation from RC-based limits to notional-based limits.
5
In practice, sub-portfolio limits should not be determined purely based on the requirement of maintaining a high credit rating. For instance, an organization
can maintain an Aa rating by investing in only high credit quality assets. But this strategy does not allow the organization to recognize high fees. In this paper,
we assume that the upper bound for RCP not only satisfies the requirement of a high rating but also allow business development in areas where the
organization has strategic advantage.

Figure 4 Relationship Between RC and the Amount Invested in a Segment
It is interesting to notice that the notional-based limits for the segments take a wide range of values, from around 9.9% (for
Equipment sector) to 52.2% (for High Tech segment), even though the RC-based limits are identical for all segments. This trait is
due to the different characteristics of each segment. Generally speaking, for a fixed RC-based limit, the notional-based limit is
higher if the segment has lower stand-alone credit risk (as measured by PDs) and lower concentration risk (reflected by low
correlation with other segments in the portfolio). Figures 5 and 6 demonstrate the notional-based limits as a function of sector’s
average PD and R-squared values, respectively. It is evident that the notional-based limit is lower for segments with higher PD
values (higher credit risk). This finding is intuitive and corresponds to the traditional rating-based limits. But the rating-based limits
miss the other, equally important side of the story, namely the concentration risk. As Figure 6 shows, the RC-based approach to
limits allows us to capture these dynamics naturally.
Generally, the more concentration risk added by a sector, the more limiting the result will be. Thus, RC-based limits take into
account both credit and concentration risks and actually allow risk managers to reflect the effects of both in a single limits
measure. For example, just looking at the credit risk, it might seem counterintuitive that the limit for the Business Products sector
is twice as large as the limit for the Equipment sector (20.9% vs. 9.9%) even though Business Products is riskier than Equipment
(average PD values are 3.49% and 3.36%, respectively). But the concentration risk graph explains why this is the case. We see the
Equipment sector sub-portfolio introduces considerably more concentration when compared to Business Products (average R-
squareds are 32.23% and 11.26%, respectively). Similar dynamics can be observed when the limits are higher, even if the
concentration risk brought by the sector’s sub-portfolio is higher (for example, compare High Tech and Equipment limits, average
PDs and R-squareds).

Figure 5 Notional Limit as a Function of Sector’s Average PD
Figure 6 Notional Limit as a Function of Sector’s Average RSQ6
For strategic planning, it is also helpful to compare the notional-based limits with the weights of each segment in the current
portfolio, illustrated in Figure 7. All segments in the graph fall below the 45-degree line, meaning that the current weights are
below the notional-based limits. For the segments close to the 45-degree line, such as Materials/Extraction, the current weight is
just slightly below the notional-based limit. The implication is that even a relatively small increase in the investment in this
industry could jeopardize the banks Aa rating. On the contrary, the High Tech industry lies far below the 45-degree line, implying
that the bank can significantly increase the weight of the High Tech industry in the entire portfolio and still maintain the Aa rating.
6
The trendline and the corresponding regression equation are calculated after omitting the four outliers.

Figure 7 Limits and Current Weights
3. Macro Scenario (MS) Limits

MS limits control exposure to adverse macroeconomic scenarios. Macro scenario limits have some appealing qualities. In the U.S.
and other jurisdictions, regulators require organizations to maintain capital levels that can withstand adverse economic
environments. Defining limits that ensure adherence to these requirements has appeal from a regulatory compliance perspective.
Separately, groups outside of risk functions can find risk-based limits abstract and difficult to intuit at times. Meanwhile, macro-
based analysis is frequently viewed as more tangible and thus easier to relate with.
There are at least two broad classes of macro-based limits: Stressed Expect Loss (SEL) Limits and Macro-Risk-based limits. Similar
to notional or EL-based limits, SEL grows linearly with notional and does not capture concentration or diversification effects. It
speaks to traditional loss stress testing and is typically viewed as transparent. SEL is additive, and it does not require as complex a
modeling infrastructure. Meanwhile, Macro-Risk-based limits consider portfolio effects. A limit ensuring that an organization
maintains an IG rating under an adverse economic scenario with a certain probability is an example of a Macro-risk-based limit.
Such an analysis requires expansion of an economic capital framework linking portfolio loss with macroeconomic scenarios.7
To demonstrate the risk appetite quantification process in this context, we create a sample portfolio containing exposures to large
U.S. and global corporate entities, U.S. retail and U.S. CRE. The sub-portfolios have characteristics typical of those asset classes
(e.g., corporate entities have lower PDs and higher RSQs than retail). The portfolio is designed to illustrate variation in sensitivity
to macroeconomic variables across asset classes.
Table 2 provides summary statistics for each of the sub-portfolios and the combined portfolio.
7
Libor Pospisil, Andrew Kaplin, Amnon Levy, and Nihil Patel, “Applications of GCorr™ Macro: Risk Integration, Stress Testing, and Reverse Stress Testing.”
Moody’s Analytics, September 2014.

Table 2
Summary Statistics for Sample Sub-portfolios and Combined Portfolios
U.S. CORPORATES GLOBAL CORPORATES U.S. CRE U.S. RETAIL AGGREGATE PORTFOLIO
Total Commitment 36 billion USD 26 billion USD 13 billion USD 14 billion USD 89 billion USD
# exposures 2,266 3,734 130 498 (homogenous pools) 6,130 individual exposures plus
498 homogeneous pools
# counterparties 1,133 1,867 130 498(homogenous pools) 3,130 counterparties plus 498
homogeneous pools
Weighted Avg. PD 1.40% 1.58% 1.44% 2.41% 1.62%
Weighted Avg. LGD 40.25% 40.73% 25.00% 30.00% 36.55%
Weighted Avg. RSQ 35.61% 39.19% 35.05% 9.38% 32.46%
Locations and Types of Exposures Diversified across Exposures from Japan, U.S. Commercial Diversified across U.S. Exposures from Japan, Europe,
U.S. industries Europe, Australia. industries (Agriculture, Australia. Diversified across
(Agriculture, Banks Diversified across Banks and S&Ls, Business industries, similarly to the U.S.
and S&Ls, Business industries, similarly to Products, High Tech, …). portfolio.
Products, High the U.S. portfolio.
Tech, …).
Concentration The 10 largest The 10 largest exposures The 10 largest The 10 largest pools The 10 largest pools account for
exposures account account for 8% of the exposures account for 23% of the 10% of the total commitment
for 16% of the total total commitment. account for 66% total commitment
commitment. of the total
commitment.
EL 0.71% 0.86% 0.49% 0.64% 0.71%
Capital wrt EL, 10bps 8.35% 7.84% 13.99% 5.44% 6.97%
3.1 Top-of-the House Macro Scenario Limits
MACRO SEL LIMITS
As referenced above, SEL-based limits are appealing for a number of reasons. In jurisdictions that require financial institutions to
adhere to regulatory stress tests, SEL-based limits are a natural mechanism to help ensure compliance. In the context of the
sample portfolio, one can define a limit requiring SEL to fall below the regulatory thresholds, perhaps with an additional buffer. The
process of setting macro scenario limits requires estimating SELs, which can be involved and is sub-portfolio specific. For the
purposes of this document, we provide only a rough set of the steps needed, but for those interested, please see Pospisil, et. al.,
“Using GCorr™ Macro for Multi-Period Stress Testing of Credit Portfolios.”
To begin, each sub-portfolio differs in its sensitivity to macro-economic variables.8 Thus, modeling SEL frequently entails relating
sub-portfolios with a different set of macroeconomic variables. For example, HPI for retail portfolios and the DJ Total Stock Market
Index for U.S. corporate exposures. Notice, one can mechanically aggregate the SEL across the sub-portfolios to arrive at the
TOTH SEL, despite conditioning on a different set of variables for each sub-portfolio. This will be a more complex issue when
considering macro scenario risk, where conditioning is conducted at the portfolio level, and we require a consistent set of macro
variables across sub-portfolios.
In analyzing the sample portfolio, Figure 10 below provides SEL for each sub-portfolio, assuming a scenario based on
Unemplyment Rate, BBB Corporate Spread, Dow Jones Total Stock Market Index, and the VIX, and their changes from 2014Q3 –
2015Q3 under the CCAR 2015 Severely Adverse scenario.9 One can imagine defining a limit based on SEL limit that recognizes the
organization’s capital structure, along with the organization’s appetite to the sensitivity of loss to macroeconomic variables. For
8
A variable selection procedure leads to the following sets of macroeconomic variables that best describe each sub-portfolio from both a statistical
perspective (a parsimonious model containing only significant variables, high adjusted R-squared value when regressing the portfolio losses on the
macroeconomic variables), as well as an economic perspective:
» U.S. Corporates: Unemployment Rate, BBB Corporate Spread, Dow Jones Total Stock Market Index, and VIX; with adjusted R-squared of 49%.
» Global Corporates: Japanese Equity Market Index, UK Equity Market Index, UK Unemployment Rate, Eurozone BBB Corporate Spread; with
adjusted R-squared of 44%.
» U.S. CRE: Real GDP Growth, Dow Jones Total Stock Market Index, CRE Index; with adjusted R-squared of 39%.
» U.S. Retail: Unemployment Rate, Dow Jones Total Stock Market Index, House Price Index (HPI); with adjusted R-squared of 60%.
The variable selection employed here is described in Pospisil, et al., “Applications of GCorr™ Macro: Risk Integration, Stress Testing, and Reverse Stress
Testing.”
9
Note, these are the four macroeconomic variables constituting a model best describing the U.S. Corporates portfolio. Thus, when we use these variables for
stress testing another sub-portfolio, the explanatory power of this model is lower than the explanatory power of the best model selected for that sub-
portfolio. As an example, these variables lead to an adjusted R-squared of 49% for the U.S. Retail portfolio, while the best model, which contains HPI,
explains 60% of variation in losses.

example, CCAR requires at least a 5% capital buffer remaining under the Severely Adverse scenario. In the interest of exposition,
we assume that: (1) portfolio composition does not change under stress, (2) the capital buffer before stress is 11.5%, (3) EL is 0.7%
and SEL is 5.0%, and (4) the buffer absorbs the SEL. Under these assumptions, the amount of capital buffer consumed to absorb
SEL is 4.3% (i.e., 5.0%-0.7%) of notional. The remaining capital buffer is 11.5%-4.3% = 7.2%. The TOTH SEL limit can now be set
as 144% of current notional amount, since the available capital buffer 7.2% is 144% of the CCAR required capital buffer 5%. If we
further assume that the capital buffer under the Severely Adverse scenario should be higher than the CCAR required capital rate to
account for variation in modeling across the institution and the Fed, the TOTH SEL limit will be lower than 144%. For instance, if
the desired capital rate under the Severely Adverse scenario is 7%, the limit should be set as 103% (i.e., 7.2%/7%) of current
notional amount.
MACRO SCENARIO RISK LIMITS
The preceding section focused on SEL as a portfolio statistic under a scenario. Despite its value for understanding portfolio
behavior under a scenario, SEL itself does not measure risk at the portfolio level, in the sense that it does not depend on portfolio
loss distribution, nor does it account for interactions among instruments in the portfolio (instrument level SELs are agnostic to the
composition of the portfolio).
We therefore consider other analyses that take into account the entire distribution of a portfolio under a macroeconomic scenario.
Figure 8 shows the unconditional and the conditional distribution of the U.S. Corporate portfolio from Table 2. We base the
scenario utilized for that conditional simulation on shocks to Unemplyment Rate, BBB Corporate Spread, Dow Jones Total Stock
Market Index, and the VIX and their changes from 2014Q3 to – 2015Q3 under CCAR 2015 Severely Adverse scenario.
Figure 8 Conditional and Unconditional Dstributions of U.S. Corporates Sub-portfolio
As Figure 8 shows, the conditional loss distribution centers around higher losses, because it assumes the Adverse scenario. But
another important point: there is still a dispersion in the conditional distribution, due to the fact that the risk in the portfolio is
unrelated to the macroeconomic environment. Such a risk can arise from name concentration (idiosyncratis risks of dominating
borrowers affect loss distribution even after conditioning on macroeconomic variables) and country- or industry-specific risks
uncorrelated with the macroeconomic variables considered in the scenario (scenario macroeconomic variables are typically broad
economic indicators that cannot capture shocks to individual industries not neccesarily associated with economy-wide
downturns). The U.S. Corporate portfolio is sufficiently diversified across names and contains only U.S. names. Therefore, the
dispersion in its conditional distribution comes primarily from industry-specific effects.

Questions of interest related to Figure 8 can be of two types:

» What are the quantiles of the conditional loss distributions?
» What is the probability that the loss will exceed a certain threshold under a macroeconomic scenario?
Both of these types of metrics can be used to define limits: if a portfolio’s metric exceeds the limit, an action is triggered.
Let us give an example of the second metric — the threshold could be set to the level consistent with an institution retaining a
certain rating (such as investment grade rating). The metric of interest is the probability that the portfolio loss will be high enough
to breach this threshold under a macroeconomic scenario. If the probability exceeds a limit, then an action must be taken — such
as moving exposures to industries less sensitive to the scenario. We explore this example in more detail in Section 3.2, within the
context of multiple portfolios, including a discussion of how to determine a loss threshold associated with a given rating.
Figure 9 plots two series of rating-implied-EDF measures (for ratings Aa3 and Baa 3). As Figure 9 shows, the meaning of rating, in
terms of default probability, substantially varies over time. For example, an entity rated Aa3 had a one-year default probability of
2bps in 2008Q1, but this probability reached 9bps in 2009Q1, more than a four-fold increase.
Figure 9 EDF Value Change Over Time
Recession associated
with dot-com bust Financial Crisis
3.2 Sub-portfolio Macro Scenario Limits
MACRO SEL LIMITS
Sub-portfolio SEL limits are similar to the sub-portfolio risk-based limits, except that SEL is being used as the level playing field.
Figure 10 illustrates one approach of setting sub-portfolio SEL limits. In the example, the SEL threshold for each sub-portfolio is set
at 2.5 billion, which is very different when translated to notional, when looking across the various portfolios. The relationship
between SEL and notional is linear, and the slope represents the SEL as a percentage of the notional amount.
It is worth highlighting some of the drives that result in different limits. Table 2 provides summary statistics that demonstrates,
while CRE has the lowest EL, global corporates has the highest EL. This said, CRE has the tightest limit when translated to notional,
and global corporates has the highest. This finding is driven by the senativity of loss to stressed macro factors. In this case, global
corporates are much less sensative to the scenarios than CRE.

Figure 10 Stressed EL-based Limits
MACRO SCENARIO RISK LIMITS
Similar to sub-portfolio risk-based limits, one can define limits based on the extent to which a sub-portfolio might contribute to
loses resulting in an organization falling below investment grade. Following the discussion above on risk-based limits, macro-
scenario risk limits can be stand-alone or portfolio-referent. For exposition, we focus the discussion on the stand-alone-macro-
scenario risk limits . Figure 11 depicts how each of the four sub-portfolios behaves under the stressed scenario . The shape of the
distributions reflect how the sensitivity of each portfolio is to the stressed macrovariables. One approach to defining the limit is to
consider the likelihood that each sub-portfolio will incur losses associated with the organization falling below investment grade
under the stressed scenario.
Figure 11 Conditional and Unconditional Distributions of Sample Sub-portfolios and the Aggregate Portfolio

In order to incorporate ratings into the analysis, we must translate rating downgrade/upgrade events in terms of loss levels
experienced by a portfolio. This process can be achieved by mapping ratings to default probabilities, which, however, must reflect
the macroeconomic environment, because the limits are set in the context of a macroeconomic scenario. To this end, we leverage
rating-implied EDF (Expected Default Frequency) values, point-in-time default probability measures estimated for each rating
class.
An institution often holds such an amount of equity capital against a portfolio that the probability of that capital being wiped out
by losses over a given period is low enough to ensure that the instituion retains a certain minimum rating. Let us consider an
example, where the desired rating is Aa3 in 2014Q3 (analysis date for CCAR 2015 scenario), which means that the institution must
hold equity capital that could be exceeded by portfolio losses with a probability of, at most, 2bps. This is the default probability
associated with Aa3 rating in that quarter. Note, this is exactly the same default probability for Aa3 as in 2008Q1. The amount of
equity capital now can be determined using the (unconditional) portfolio loss distribution. Considering the aggregate portfolio
from Table 2, the equity capital should be 8.85%.10
Now let us assume that the institution is interested in the case when it can potentially lose its investment grade rating if the CCAR
2015 Severly Adverse scenario occurs between 2014Q3 and 2015Q3. Losing the investment grade rating means that the
institution’s equity capital over that one-year period will deplete to such an extent that at the end of the period, the institution’s
default probability will be at least 64bps. This is the default probability associated with the Baa3 rating — the poorest investment
grade rating — as of 2009Q1, and we use it as proxy for Baa3 default probability as of 2015Q3. Having a higher default
probability, therefore, leads to a downgrade to a speculative grade.
In our example, we assume that this downgrade to a speculative grade occurs if losses on the aggregate portfolio over the one-
year period exceed threshold 5.39%.11
Let us shift the discussion to setting limits on individual sub-portfolios under the macroeconomic scenario. Within our example,
we define the limit on a sub-portfolio in terms of the probability (under the scenario) that this sub-portfolio will drive the losses on
the aggregate portfolio beyond the threshold and, thus, trigger a downgrade to a speculative rating. To determine this
probability, we must make an assumptiona about the the losses on the other sub-portfolios under the scenario. We assume they
are equal to stressed expected losses in excess of the uncondtional losses. We can, therefore, write the probability in the following
way:
, , , , ∙
Where , , is the loss on a sub-portfolio over the one year period, , , are the stressed expected losses on the
remaining sub-portfolios, and is the value of the aggregate portfolio.
We present these probabilities in Table 3.
10
The value 8.85% is the 99.98th percentile of the distribution of aggregate portfolio losses with respect to expected loss. Expected loss should not be included
in the capital, as it is typically accounted for in allowance for losses on loans and leases (ALLL).
11
This calculation can be carried out with certain assumptions, which are institution-specific. Let us provide a highly stylized description of this calculation.
Denoting equity capital at the beginning of the period and at the end of the one year period , suppose that the is set so that
2 , where is the portfolio loss with respect to unconditional . depends on several factors — the losses experienced over the period
, new volume, amortization, and maturities of exposures, and importantly on the equity capital raised over the period. Many of these factors depend on
investment strategy and capital planning by the institution. However, for our purposes, the most important is dependence of on : .
Downgrade to a speculative grade occurs if drops below a certain threshold , defined by equation 64 , where 64
is the default probability associated with Baa3 rating as of 2015Q3, and is the loss on the institution’s portfolio over the period 2015Q3 – 2016Q3. This
means that the threshold depends on the characteristics of the portfolio as of 2015Q3 — including its composition and credit risk parameters, such as
PDs, LGDs, and correlations. The last step is to translate the threshold on capital to threshold on losses over the first period –— that is, condition
implies a condition , which, in our example, derives as 5.39%.

Table 3
Probability of Causing Downgrade for Sample Sub-portfolios
RANKING OF U.S CORPORATES GLOBAL CORPORATES U.S. CRE U.S. RETAIL
STRATEGY RETURNS
Total Commitment 36 billion USD 26 billion USD 13 billion USD 14 billion USD
Probability of causing downgrade 19.1% 6.1% 5.5% <1bps
to a speculative grade
It is worth noting that patterns in Table 3 depend on several sub-porttfolio characteristics — its overall weight in the aggregate
portfolio, as well as its sensitivity to the scenario and residual volatility after conditioning on a scenario. As a result, the U.S.
Corporates portfolio, which constitutes the largest sub-portfolio of the aggregate portfolio, also has the highest probability of
driving the aggregate portfolio to the downgrade. However, this probabability for the U.S. CRE portfolio is not that lower than for
the Global Corporates portfolio, even though the U.S. CRE portfolio has a substantially lower commitment, which can be
attributed to high stressed EL and high conditional volatility on the U.S. CRE portfolio. On the other hand, the dispersion of the
U.S. Retail portoflio losses is not high enough to cause a downgrade on the aggregate portoflio.
A limit can, for example, prescribe that the probability of causing the downgrade should not exceed 25%. If that limit were
breached, a possible action could be to shift the portfolio balance to exposures less sensitive to the macroeconomic variables
considered in the scenario.
4. Practical Applications
There are a number of practical considerations to implementing the methodologies outlined above. Most notable is ensuring the
limit setting process aligns with an organization’s operating plan and core competency. As an example, a regional bank focused on
domestic lending in the U.S. will find that its country risk limit outside the States will be beyond anything the organization would
consider. Meanwhile, the limit for the U.S. will be too conservative, given that the entire portfolio is domestic.
To address this issue, the limit setting process must consider the geographic composition of the organization’s investment market.
In this example, appropriate geographic-based limits would likely be regional classifications within the United States and account
for the organization’s market concentration. As we formalize below, a more practical risk limit can be defined as a risk-based
growth rate above each of the existing portfolio segments. Where segmentation can be defined based on business process, to
more easily manage incentives and, with roughly similar level of overall risk, to more easily classify the limit hierarchy.
The remainder of this sections formalizes a few approaches that address cases with practical challenges associated with using
quantitative methods when setting limits. The approaches can be used in broader contexts. And while challenges are very specific
to an institution’s specific business process, the spirit of how to deal with these challenges typically has some commonality.
4.1 Business Strategy and Operating Plan

In the introduction to this section, we presented an example of a bank entirely focused on domestic investing. We provided
intuition for why setting country limits using the methodology outlined in the Introduction makes sense. As is obvious, a naive
application of the risk limits at the country-level does not serve a useful business purpose.
A more useful approach defines limits based on geographic segmentation, limits that align with the organization’s structure. In the
example, one can imagine lending business segmented by U.S. region, perhaps Northwest, Southwest, Central, Northeast, and
Southeast. This structure allows the respective business owners to have clear ownership over their respective portfolio risks (not to
say that limits should only be defined by business lines). Even within this structure, one can imagine that the organization has an
unusually high investment concentration in, for example, the Northeast, and an application of the risk limits methodology to the
proposed geographic segmentation still not properly aligned with business practice; limits would likely be too tight for the
Northeast and too lax for the other geographic classifications.
To define risk-based limits that are in-line with business operations, one can begin with the existing portfolio, confirm sub-
portfolio risk-profiles are in-line with the organization’s risk appetite, and then define risk limits that allow for each sub-portfolio to
grow. The degree of growth can be defined in the risk space (e.g., each sub-portfolio can increase risk allocation by no more than
10%). The proportion of growth must be defined so that the portfolio conforms with the organization’s risk appetite when a limit
is hit.

4.2 Portfolio Dynamics

In some instances portfolios are unusually dynamic, with sub-portfolios exhibiting extreme growth rates. For example, when an
organization enters new markets or existing markets. Using the current portfolio as a starting point to define sub-portfolio growth
in risk allocation may be distortive, as risk concentrations change drastically, resulting in portfolio-referent risk characteristics of
other parts of the portfolio being affected. Instead, an organization can consider using scenario-based reference portfolios as the
starting point for defining the growth limits. The portfolio for each scenario can be defined either through expert judgment as the
organization considers its desired market share for a business line. Alternatively, portfolios can be defined using quantitative
methods with dynamics related to various macroeconomic scenarios.12 What matters in the end is that relevant reference
portfolios are used in defining portfolio referent risk measures, as well as the degree to which risk growth is considered appropriate
in the limit setting process.
4.3 Sub-portfolio Limits and TOTH Limits

In some cases, it is important to define limits so that at least one sub-portfolio limit is breached when the TOTH limit is breached.
This process can be done by defining the sub-portfolio limit dynamically as a function of, for example, how far the TOTH risk
measure is to its limit. We use the following algorithm as an example:
Define to be the relative distance of the TOTH risk measure is to its limit, and sub-portfolio j’s risk measure is to its limit at time
of the limit definition, 0:
sub portfolio risk measure sub portfolio limit TOTHrisk measure TOTHlimit
We can now define a dynamic sublimit for j that will be violated whenever the TOTH limit is violated:
sub portfolio risk measure sub portfolio limit TOTHrisk measure TOTHlimit
Alternative mechanisms can produce the same outcome, this is simply an example of one.
5. Conclusion
Quantifying risk appetite is central to the limit setting process. While an organization’s RAS can be qualitative and difficult to
translate into quantitative metrics, this paper defines a number of intuitive approaches. We define both risk and macro-scenario-
based limits that can be directly linked to common components of an organization’s RAS. Aside from advances related to the
quantification of limits, the methodologies this paper outlines align with an organization’s economic capital framework, which
provides a foundation for a granular description of credit portfolio risk.
We illustrate this alignment using a number of case studies that demonstrate the relationships between varying sub-portfolio
characteristics and the various approaches to setting limits. In practice, of course, applying these approaches requires not just
quantitative methods, but practical considerations for strategic business needs.
12
Levy, Amnon, “Quantifying PPNR Modeling,” Moody’s Analytics White Paper, 2014.

References
Levy, Amnon, “Quantifying PPNR Modeling.” Moody’s Analytics White Paper, 2014.
Financial Stability Board, Principles for an Effective Risk Appetite Framework. November 18, 2013.
Institute of International Finance, Implementing Robust Risk Appetite Frameworks to Strengthen Financial Institutions. June 2011.
Pospisil, Libor, Andrew Kaplin, Amnon Levy, and Nihil Patel, “Applications of GCorr™ Macro: Risk Integration, Stress Testing, and
Reverse Stress Testing.” Moody’s Analytics White Paper, 2014.
Pospisil, Libor, Jimmy Huang, Mariano Lanfranconi, Albert Lee, Amnon Levy, Marc Mitrovic, Olcay Ozkanoglu, Nihil Patel, and
Kevin Yang, “Using GCorr™ Macro for Multi-Period Stress Testing of Credit Portfolios.” Moody’s Analytics White Paper, 2015.

© Copyright 2015 Moody’s Corporation, Moody’s Investors Service, Inc., Moody’s Analytics, Inc. and/or their licensors and affiliates (collectively, “MOODY’S”). All rights reserved.
CREDIT RATINGS ISSUED BY MOODY'S INVESTORS SERVICE, INC. (“MIS”) AND ITS AFFILIATES ARE MOODY’S CURRENT OPINIONS OF THE RELATIVE FUTURE CREDIT RISK OF
ENTITIES, CREDIT COMMITMENTS, OR DEBT OR DEBT-LIKE SECURITIES, AND CREDIT RATINGS AND RESEARCH PUBLICATIONS PUBLISHED BY MOODY’S (“MOODY’S
PUBLICATIONS”) MAY INCLUDE MOODY’S CURRENT OPINIONS OF THE RELATIVE FUTURE CREDIT RISK OF ENTITIES, CREDIT COMMITMENTS, OR DEBT OR DEBT-LIKE
SECURITIES. MOODY’S DEFINES CREDIT RISK AS THE RISK THAT AN ENTITY MAY NOT MEET ITS CONTRACTUAL, FINANCIAL OBLIGATIONS AS THEY COME DUE AND ANY
ESTIMATED FINANCIAL LOSS IN THE EVENT OF DEFAULT. CREDIT RATINGS DO NOT ADDRESS ANY OTHER RISK, INCLUDING BUT NOT LIMITED TO: LIQUIDITY RISK, MARKET
VALUE RISK, OR PRICE VOLATILITY. CREDIT RATINGS AND MOODY’S OPINIONS INCLUDED IN MOODY’S PUBLICATIONS ARE NOT STATEMENTS OF CURRENT OR HISTORICAL
FACT. MOODY’S PUBLICATIONS MAY ALSO INCLUDE QUANTITATIVE MODEL-BASED ESTIMATES OF CREDIT RISK AND RELATED OPINIONS OR COMMENTARY PUBLISHED BY
MOODY’S ANALYTICS, INC. CREDIT RATINGS AND MOODY’S PUBLICATIONS DO NOT CONSTITUTE OR PROVIDE INVESTMENT OR FINANCIAL ADVICE, AND CREDIT RATINGS
AND MOODY’S PUBLICATIONS ARE NOT AND DO NOT PROVIDE RECOMMENDATIONS TO PURCHASE, SELL, OR HOLD PARTICULAR SECURITIES. NEITHER CREDIT RATINGS
NOR MOODY’S PUBLICATIONS COMMENT ON THE SUITABILITY OF AN INVESTMENT FOR ANY PARTICULAR INVESTOR. MOODY’S ISSUES ITS CREDIT RATINGS AND
PUBLISHES MOODY’S PUBLICATIONS WITH THE EXPECTATION AND UNDERSTANDING THAT EACH INVESTOR WILL, WITH DUE CARE, MAKE ITS OWN STUDY AND
EVALUATION OF EACH SECURITY THAT IS UNDER CONSIDERATION FOR PURCHASE, HOLDING, OR SALE.
MOODY’S CREDIT RATINGS AND MOODY’S PUBLICATIONS ARE NOT INTENDED FOR USE BY RETAIL INVESTORS AND IT WOULD BE RECKLESS FOR RETAIL INVESTORS TO
CONSIDER MOODY’S CREDIT RATINGS OR MOODY’S PUBLICATIONS IN MAKING ANY INVESTMENT DECISION. IF IN DOUBT YOU SHOULD CONTACT YOUR FINANCIAL OR
OTHER PROFESSIONAL ADVISER.
ALL INFORMATION CONTAINED HEREIN IS PROTECTED BY LAW, INCLUDING BUT NOT LIMITED TO, COPYRIGHT LAW, AND NONE OF SUCH INFORMATION MAY BE COPIED
OR OTHERWISE REPRODUCED, REPACKAGED, FURTHER TRANSMITTED, TRANSFERRED, DISSEMINATED, REDISTRIBUTED OR RESOLD, OR STORED FOR SUBSEQUENT USE FOR
ANY SUCH PURPOSE, IN WHOLE OR IN PART, IN ANY FORM OR MANNER OR BY ANY MEANS WHATSOEVER, BY ANY PERSON WITHOUT MOODY’S PRIOR WRITTEN
CONSENT.
All information contained herein is obtained by MOODY’S from sources believed by it to be accurate and reliable. Because of the possibility of human or mechanical error as well as
other factors, however, all information contained herein is provided “AS IS” without warranty of any kind. MOODY'S adopts all necessary measures so that the information it uses in
assigning a credit rating is of sufficient quality and from sources MOODY'S considers to be reliable including, when appropriate, independent third-party sources. However,
MOODY’S is not an auditor and cannot in every instance independently verify or validate information received in the rating process or in preparing the Moody’s Publications.
To the extent permitted by law, MOODY’S and its directors, officers, employees, agents, representatives, licensors and suppliers disclaim liability to any person or entity for any
indirect, special, consequential, or incidental losses or damages whatsoever arising from or in connection with the information contained herein or the use of or inability to use any
such information, even if MOODY’S or any of its directors, officers, employees, agents, representatives, licensors or suppliers is advised in advance of the possibility of such losses or
damages, including but not limited to: (a) any loss of present or prospective profits or (b) any loss or damage arising where the relevant financial instrument is not the subject of a
particular credit rating assigned by MOODY’S.
To the extent permitted by law, MOODY’S and its directors, officers, employees, agents, representatives, licensors and suppliers disclaim liability for any direct or compensatory
losses or damages caused to any person or entity, including but not limited to by any negligence (but excluding fraud, willful misconduct or any other type of liability that, for the
avoidance of doubt, by law cannot be excluded) on the part of, or any contingency within or beyond the control of, MOODY’S or any of its directors, officers, employees, agents,
representatives, licensors or suppliers, arising from or in connection with the information contained herein or the use of or inability to use any such information.
NO WARRANTY, EXPRESS OR IMPLIED, AS TO THE ACCURACY, TIMELINESS, COMPLETENESS, MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OF ANY SUCH
RATING OR OTHER OPINION OR INFORMATION IS GIVEN OR MADE BY MOODY’S IN ANY FORM OR MANNER WHATSOEVER.
MIS, a wholly-owned credit rating agency subsidiary of Moody’s Corporation (“MCO”), hereby discloses that most issuers of debt securities (including corporate and municipal
bonds, debentures, notes and commercial paper) and preferred stock rated by MIS have, prior to assignment of any rating, agreed to pay to MIS for appraisal and rating services
rendered by it fees ranging from $1,500 to approximately $2,500,000. MCO and MIS also maintain policies and procedures to address the independence of MIS’s ratings and rating
processes. Information regarding certain affiliations that may exist between directors of MCO and rated entities, and between entities who hold ratings from MIS and have also
publicly reported to the SEC an ownership interest in MCO of more than 5%, is posted annually at www.moodys.com under the heading “Shareholder Relations — Corporate
Governance — Director and Shareholder Affiliation Policy.”
For Australia only: Any publication into Australia of this document is pursuant to the Australian Financial Services License of MOODY’S affiliate, Moody’s Investors Service Pty
Limited ABN 61 003 399 657AFSL 336969 and/or Moody’s Analytics Australia Pty Ltd ABN 94 105 136 972 AFSL 383569 (as applicable). This document is intended to be provided
only to “wholesale clients” within the meaning of section 761G of the Corporations Act 2001. By continuing to access this document from within Australia, you represent to
MOODY’S that you are, or are accessing the document as a representative of, a “wholesale client” and that neither you nor the entity you represent will directly or indirectly
disseminate this document or its contents to “retail clients” within the meaning of section 761G of the Corporations Act 2001. MOODY’S credit rating is an opinion as to the
creditworthiness of a debt obligation of the issuer, not on the equity securities of the issuer or any form of security that is available to retail clients. It would be dangerous for “retail
clients” to make any investment decision based on MOODY’S credit rating. If in doubt you should contact your financial or other professional adviser

Counterparty Trading Limits Revisited:
CSAs, IM, SwapAgent®
From PFE to PFL
Chris Kenyon, Mourad Berrahoui and Benjamin Poncet∗
arXiv:1710.03161v1 [q-fin.RM] 9 Oct 2017
09 October 2017
Version 1.00
Abstract
The utility of Potential Future Exposure (PFE) for counterparty trad-
ing limits is being challenged by new market developments, notably widespread
regulatory Initial Margin (using 99% 10-day exposure), and netting of
trade and collateral flows. However PFE has pre-existing challenges w.r.t.
portfolios/distributions, collateralization, netting set seniority, and over-
laps with CVA. We introduce Potential Future Loss (PFL) which combines
expected shortfall (ES) and loss given default (LGD) as a replacement for
PFE. With two additional variants Adjusted PFL (aPFL) and Protected
Adjusted PFL (paPFL) these deal with both new and pre-existing chal-
lenges. We provide a theoretical background and numerical examples.
1 Introduction
The utility of Potential Future Exposure (PFE) for counterparty trading limits
is being challenged by new market developments, notably widespread regulatory
Initial Margin (BCBS-317, 2015), and netting of trade and collateral flows (e.g.
via SwapAgent® , LCHSA (2017)). However counterparty trading limits were
already difficult to compare from portfolio/distribution effects, and as they were
typically changed when a collateral agreement (Collateral Support Annex, or
CSA, of the ISDA®1 ) was put in place. That is, the effects of the change in loss
distribution, and any potential change in recovery are included ad hoc. Fur-
thermore, trading limits with the same counterparty but at different seniorities
are not fungible. In addition overlaps with credit mitigation and CVA are not
typically included in PFE. Thus the typical counterparty limit metric, PFE, as
a high quantile (95%, 97.5%, and rarely 99%) of future exposures, needs up-
dating. Therefore we introduce Potential Future Loss (PFL) which combines
∗ Contacts: chris.kenyon@lloydsbanking.com, mourad.berrahoui@lloydsbanking.com, ben-
jamin.poncet@lloydsbanking.com. The views expressed in this presentation are the personal

views of the authors and do not necessarily reflect the views or policies of current or previous
employers. Not guaranteed fit for any purpose. Use at your own risk.
1 International Swaps and Derivatives Association master agreement.
1
expected shortfall (ES) and loss given default (LGD). With two additional vari-
ants Adjusted PFL (aPFL) and Protected Adjusted PFL (paPFL) these deal
with the new and pre-existing challenges. We provide a theoretical background
and numerical examples.
PFE is generally defined as:
Definition 1 (PFE(t,q)). Potential Future Exposure at time t in the future for
quantile q is
−1
PFE(t, q) := CDF (q) max(ΠM (t), 0)
where CDF−1 (q)(...) is the inverse Cumulative Distribution Function of (...)

for the quantile q, and ΠM (t) is the mark to market of the portfolio in the
netting set of interest under measure M. The measure M is often chosen as the
inverse-T -Forward measure which is defined as the risk-neutral value (which is
measure-independent) inverse-discounted by an observed discount curve (which
implicitly selects the T -Forward measure). Inverse-discount means divide by
the discount factor.
The choice of M is out of scope of this paper, but discussed elsewhere
(Kenyon et al., 2015).
2 Challenges to PFE
Uncollateralized, 0.95 With CTM/STM, 0.95

4 × 106
2.0 × 107
3 × 106
1.5 × 107
PFE
PFE
2 × 106
1.0 × 107
5.0 × 106 1 × 106
0 0
0 2 4 6 8 10 0 2 4 6 8 10
Years Years
CTM/STM, IM, 0.95 Trade and CTM/STM Netting, IM, 0.95
4 × 106 4 × 106
3 × 106 3 × 106
PFE
PFE
2 × 106 2 × 106
1 × 106 1 × 106
0 0
0 2 4 6 8 10 0 2 4 6 8 10
Years Years
Figure 1: PFE(95%) profiles for 10Y USD IRS with using SV LMM exposure
model for: TOP LEFT uncollateralized; TOP RIGHT daily collateral; BOT-
TOM RIGHT daily collateral with (yellow downwards-staircase) superimposed
Schedule-based IM (i.e. 4%, 2%, 1% of notional); BOTTOM RIGHT collateral,
IM superimposed; and netting of collateral and trade flows.
2
Figure 1 illustrates PFE(95%) for a 10Y USD interest rate swap (IRS) as
uncollateralized, daily-collateralized, with Schedule-based IM, and then with
netting of collateral and trade flows. This is a typical profile resulting from
stopping dates every 2W. The spikes in the collateralized profile are a well-
known phenomenon from return of collateral following a trade flow (Andersen
et al., 2016).
Potential Future Exposure (PFE) for a netting set is typically defined as the
profile of a high quantile of future exposure for a given margin period of risk
(MPOR). The PFE profile runs from now (t = 0) up to the point where there
is no further exposure. The PFE profile endpoint may be beyond the maturity
of the netting set because of (non-)return of collateral risks. Usual quantiles
are 95%, 97.5%, and rarely 99%. In addition the future exposure is generally
calculated as loss-on-the-day and not discounted back to t = 0. This makes the
profile measure dependent but we will not go into this aspect here.
PFE is used to limit and manage counterparty trading limits with non-
collateralized and collateralized counterparties.
We define a challenge to PFE as something that makes its interpretation
unclear, incorrect, or meaningless in terms of its objective of being a control for
counterparty trading. We first list the challenges and then go into detail.
New challenges to PFE are:
• Widespread regulatory Initial Margin (BCBS-317, 2015). This is being

phased in from 2016-2020.
• Netting of collateral mark-to-market flows and trade termsheet cash flows.
A recent example is (LCHSA, 2017) which is going live in 2017.
Existing challenges include:

• Portfolio/distribution effects
• Overlap with credit mitigation and Credit Valuation Adjustment (CVA)
• Netting sets with different seniorities
• Collateral
2.1 IM
One regulatory IM definition is as a 10-day, one-sided 99% exposure, calibrated
to a period of stress. Alternatively, a schedule-based method which uses a
lookup table based on notional and maturity can be applied. The schedule-
based method makes no allowance for netting so most large traders will use the
exposure method. Note that the MPOR is defined as nine business days plus
the frequency of collateral calling, so daily calls will give a 10B MPOR.
Figure 2 shows the IM challenge to PFE: with IM, PFE is identically zero.
This is true even before we consider that IM is defined as the 99th percentile
calibrated to a period of stress. In the figure, given a 20% stress, the quantile
for PFE would have to be defined as something above 99.86%. Even if this
re-definition of PFE was done, using such a high percentile for non-IM or non-
collateralized counterparties would be problematic because it would be so high.
Thus the numbers that Credit Officers would be required to sanction would be
3
0.5
0.99 0.5
0.99, 10% stress
0.4
0.975 0.4
0.975
0.3
0.95 0.3
0.95
0.2 0.2
0.1 0.1
-1 1 2 3 4 5 2 4 6
0.5
0.99, 20% stress
0.4
0.975
0.3
0.95
0.2
0.1
2 4 6 8
Figure 2: The IM challenge to PFE. Shifted LogNormal exposure with quantiles,

including the 99% level defined in regulations for Initial Margin (TOP LEFT).
Comparison with 99% quantile given a 10% volatility increase (stress), TOP
RIGHT; and a 20% volatility increase (BOTTOM). Given IM defined as 99%
10-day one-sided exposure, PFE is identically zero in all three cases.
completely outside previous experience. The percentiles would also refer to tail
eventualities equally outside experience.
It may be argued that the collateral eligibility for IM is sufficiently wide to
make the IM worthless. However, regulations are written specifically to avoid
this situation despite wide eligibility. Collateral value is safeguarded by several
mechanisms:
• collateral must be marked to market
• haircuts are required that reference the credit quality, and tenor, of the
collateral
• collateral with insufficient quality is not eligible. Quality is defined in
terms of 1Y default probabilities, and in terms of the collateral tenor.
• collateral with significant correlation with the posting entity is not per-
mitted
In addition we expect both wide and narrow IM collateral agreements to be
signed. That is, some will be cash and (good) government securities only,
whereas others will be as wide as regulatorily permissible. At the extreme
even if we consider wide-eligible IM to be worthless, the narrow-IM agreements
will be valuable and challenge PFE.
Despite collateral and regulatory IM there can still be significant, if brief,
exposure from spikes in exposure profiles due to return of collateral. This is
addressed by another recent market development, covered in the next section.
4
2.2 Netting of Collateral/Settlement and Termsheet flows
With collateralized counterparties spikes in exposure are observed on coupon
and principle payment dates when collateral and termsheet flows are not net-
ted. These spikes are from return of collateral following a termsheet payment.
Andersen et al. (2016) have pointed out that these spikes may mean that regu-
latory IM does not reduce exposure by 99%, but perhaps only by 90% in some
cases (if for brief periods).
Market services are now appearing that net collateral and termsheet flows,
such as SwapAgent® . With SwapAgent® spikes in collateralized (or settled)
exposure profiles may not be present, but this depends on the exact timing of
the default. The timing matters for SwapAgent® because if a counterparty fails
to pay a SwapAgent® call for two days then it is removed from the service and
goes back to the purely bilateral agreements between the original counterparties.
On this basis it is likely, but not certain, that there will be no spikes assuming
that no flows to SwapAgent® act as a signal of default of the counterparty.
Given collateral and termsheet flow netting, the addition of IM will produce
zero exposure below the 99th percentile. This renders PFE(95%), PFE(97.5%),
and PFE(99%) of questionable utility for these counterparties.
2.3 Portfolio/distribution effects

Here we focus on uncollateralized portfolios and later consider changes of expo-
sure distribution from collateralization.
Since PFE(q) is an exposure quantile it is insensitive to any changes of
exposure distribution above q. Thus there can be arbitrary changes in exposure
— provided they are 1-in-20 at any time, for q=95%, say. This means that
Credit Officers have to factor in these possibilities by hand when setting PFE
limits.
Recall that the value distribution of a portfolio is the convolution of the
positions with the value distributions of each position. Thus a heavy-tailed
value distribution of a single position can be duplicated by a set of light-tailed
value distributions. In practice, this means that portfolios will have more un-
usual heavy-tail distributions than any single instrument (whether or not the
underlying dynamics or pricing of individual instruments have heavy tails).
The distribution-insensitivity of PFE is worse than it appears because the
tail of the portfolio-dependence of the exposure distribution. That means that a
change in the trading pattern of a counterparty can change the exposure above q
and this will not show up. This risk-insensitivity of PFE for relatively common
(1-in-20 for 95% PFE) events is undesirable. Suppose now that Credit Officers
change their q from 95% to 99%. This has two effects: firstly Credit Officers
and Relationship manager have to re-calibrate their risk understanding; and
secondly the PFE limits have to be increased for all counterparties. Even if this
is done, there is now an in-sensitivity to 1-in-100 events: and two or three can
be expected each year, per counterparty2 .
2 Negative correlations between large numbers, n, of counterparties are mathematically
√
impossible: the average limit goes as ρmin = −1/ n − 1.
5
2.4 Overlaps with credit mitigation and CVA
If the counterparty is fully credit hedged then there should be no PFE as there
is no possibility of loss. A fully credit hedged status is not generally possible
with traded instruments from the point of view of capital regulations in Basel III
(BCBS-189, 2011). In Basel III if there is credit protection from a single-name
credit default swap (CDS) in place then, effectively, the default probability of
the counterparty changes to that of the CDS (protection) provider. Given that
PFE assumes default, changes in default probability are not relevant, thus the
capital approach may not be relevant. However, the credit protection is relevant
but PFE does not typically take it into account because it is not in the netting
set that is being protected.
A second overlap between CVA and PFE is w.r.t. incurred CVA which Basel
III deducts from exposure at default in capital calculations on the grounds that
this loss has already gone through PnL (BCBS-237 (2012), Section 2d). Incurred
CVA reduces PFE but is not part of the PFE definition. Of course the PFE
limit should also be reduced by the incurred CVA because the money has been
lost. Thus as a counterparty approaches default the PFE limits will get tighter
and tighter as the loss appetite is used up.
2.5 Multiple Seniorities

If there are multiple netting sets with the same counterparty at different seniori-
ties then this is a challenge to PFE. Typically there will be separate counterparty
trading limits against each netting set. However, the risk is to the counterparty
not the netting sets so this is an issue. In addition it is generally not possi-
ble, nor desired, to move limit capacity from one netting set to another with
a different seniority 1-for-1. This is another practical issue for efficient limit
management.
Claim seniority has a major effect on recovery rates. Jankowitsch et al.
(2014) have investigated the US market and observe:
Seniority Median Recovery
Unsecured 42%
Subordinated 5%
Different recovery rates is the main reason that Credit Officers have different
appetites for PFE for netting sets at different seniorities. PFE does not take
this into account, but the Credit Officers do — hence the limit management
inefficiency.
2.6 Collateralization
Collateralization has two effects w.r.t. uncollateralized exposure: change in loss
distribution; and change in recovery rate. PFE cannot capture either of these
effects, and we examine the numerical significance later.
• The exposure distribution changes from a strip of European Call options

(uncollateralized) to a strip of Calendar Spread Call options. Figure 3
illustrates the change. In addition, the effect of the distribution changes
will be portfolio dependent.
6
10
0.8
8
0.6
6
PDF
PDF
0.4
4
0.2 2
0.0 0
0 1 2 3 4 5 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3
Value(T)/Fwd(T) (Value(T-dt)-Value(T))/Fwd(T)
Figure 3: Considering one time-point on the exposure profile where there is

a LogNormal probability distribution function (PDF) of exposure LEFT, the
collateralized (Calendar Spread) exposure changes to the PDF at RIGHT (very
close to a Student-t with 2 degrees of freedom). Setup details in the text.
• When a collateralized counterparty defaults this is typically because it has

debts. Some of these will be via collateralized counterparties. The default
mechanism is often that it cannot raise liquidity to pay collateral calls. In
short, assets (or financialized assets) pledged as collateral are not avail-
able to creditors. Thus we can expect lower recoveries from collateralized
counterparties than uncollateralized, all other things being equal.
We defer a detailed impact analysis to the next section where we introduced
Potential Future Loss
3 Potential Future Loss

Given the recent and pre-existing challenges to PFE for counterparty trading
limits described above, we now introduce Potential Future Loss (PFL), Adjusted
PFL, and Protected Adjusted PFL
Potential Future Loss(t,q), PFL(t,q) is the future profile of Expected

Shortfall(q) times Loss Given Default, where q is the quantile of interest.
Definition 2 (PFL(t,q)). Potential Future Loss at time t in the future for
quantile q is
PFL(t, q) :=EM max(LGD(t) × ΠM (t), b) /q

(1)
−1

b :=CDF (q) max(LGD(t) × ΠM (t), 0) (2)
Notation as for PFE, EM [.] stands for expectation under the measure M, and
LGD(t) is the loss given default at t. The LGD is inside the expectation to
take into account wrong/right way risk (WWR). We expect that with the em-
phasis in FRTB-CVA (BCBS-325, 2015) on WWR modelling this will be widely
implemented in that timescale.
If we were to assume that portfolio value and LGD were independent then
we have :
PFLindependent (t, q) :=EM [LGD(t)] × EM max(ΠM (t), bindependent ) /q (3)

bindependent :=CDF−1 (q) max(ΠM (t), 0)

(4)
7
Potential Legacy Response
Challenge Effect PFE PFL aPFL paPFL
Widespread Initial Margin PFE∼0 Fail OK OK OK
+ Netting of Collateral and PFE≡0 Fail OK OK OK
Trade Flows
Portfolio composition‡ distribution Poor OK OK OK
change
Overlap with: incurred CVA double Absent Absent OK OK
count
Overlap with: credit mitiga- double Unclear Absent Absent OK
tion (e.g. CDS) count
Different seniorities see text Poor OK OK OK
Collateralization distribution Poor OK OK OK
change
+ Portfolio composition‡ distribution Poor OK OK OK
change
Table 1: Comparison of PFE and new PFL-based responses to recent challenges

and pre-existing issues for counterparty trading limits. Unclear means that
some aspect may be present in practice but is not in typical definitions of PFE.
‡Change of distribution affects uncollateralized and collateralized portfolios.
Adjusted Potential Future Loss(Q), aPFL(Q) is the future profile of

PFL(Q) with incurred CVA removed, and where the associated limit has had
incurred CVA removed.
Protected Adjusted Potential Future Loss(Q), paPFL(Q) is the future

profile of aPFL(Q), with associated limit adjustment, where the profile of bought
credit protection has been removed from the underlying exposures.
Initial margin, when present, acts as an exposure reduction according to its

(with haircut) value. That is, there is no special treatment of initial margins,
independent amounts, etc, they are simply part of the calculations.
Given that one credit mitigant is a tradable instrument (single-name CDS),
some coordination between Credit Officers and the CVA trading desk is re-
quired for effective use of aPFL and paPFL. However, credit protection will
typically be purchased exactly where it is needed — and paid for by Sales for
the counterparty — so coordination is important.
Table 1 provides a comparison of PFE and new PFL-based responses to
recent challenges and pre-existing issues for counterparty trading limits. We now
look at some numerical examples to illustrate the magnitude of issues affecting
PFE (apart from IM which renders PFE void) which PFL addresses.
3.1 Recovery Rates

There are no liquid instruments providing market implied recovery rates. CDS
provide expected loss, within this recovery and default probability are almost
indistinguishable. Even if liquid instrument providing market-implied recovery
rates existed, their use would be questionable as exposure, by definition, is
unhedged.
8
Many industry studies exist on sector-wide recovery rates and their variation
with market stress (Chen et al., 2014; Jankowitsch et al., 2014). Seniority-
dependent recovery rate observations are also available (Jankowitsch et al.,
2014). Beyond this bank Know Your Customer (KYC), Relationship Mangers,
and Credit Officers together with internal (real-world) risk models, and market
data service providers give inputs to internally computed recovery rates for use
in PFL.
3.2 Volume Risk Limitation

One concern both with collateralization and with IM is that these may permit
such high levels of trading so as to create a volume or concentration risk. Now
how big must a position be to create a concentration risk? That is, a position
that has a market impact when the surviving party has to close out their posi-
tions after the counterparty defaults. The definitions in (ISDA-SIMM-2, 2017)
provide one benchmark for concentration risk, and we reproduce the interest
rate (IR) concentration thresholds in Table 2.
Currency risk group Delta Threshold Vega Threshold

(M USD/bp) (M USD)
High vol (default) 8 110
Regular, well-traded (USD, EUR, GBP) 230 2700
Regular, less-traded (short list) 28 150
Low vol (JPY) 82 960
Table 2: ISDA® SIMM® IR concentration thresholds. Vega can be LogNormal

or Normal for IR.
In practice IM levels were sufficient in the Lehman default and regulatory

IM is based on CCP IM. Thus we do not expect increases in volume to create
a concentration risk.
Nonetheless, suppose that increases in trading volumes did create a concen-
tration risk. Regulatory IM takes this into account and increases the required
IM. Thus there are two feedback mechanisms to limit volume risk. Firstly the
existence of IM itself which is costly to provide. Secondly, there is the non-linear
increase of IM when concentration thresholds are reached.
3.3 Numerical Examples

Exposure distribution effects can be read off Tables 3 and 4. For a simple Log-
Normal exposure setup (see below) the ratio of expected shortfall (ES) to the
corresponding quantile (i.e. PFE) can be read off the α and β lines for uncollat-
eralized and collateralized portfolios respectively. Just considering PFE(95%)
we see that the ES starts at 9% more than PFE at 1Y and goes to 34% more at
10Y for uncollateralized. The collateralized picture is similar. The point is that
the ES-to-PFE ratio is not constant, or near-constant, even in a very simple
setup whether uncollateralized or collateralized.
To get a feel for the effectiveness of PFL compared with PFE we now look
at distribution effects with collateralization. We briefly consider portfolio com-
position as well. PFE misses all changes in loss distribution above its percentile
9
whereas PFL includes them. Although IM renders PFE largely meaningless
there will be a period where IM is not widespread. Even during this period the
fact that PFE cannot deal with loss distribution changes above its quantile is
significant, and we give numerical indications here.
3.3.1 Normal Distribution

We first consider exposure with a Normal distribution as a limiting case. Sup-
pose exposure follows a Brownian motion with drift, the autocorrelation is
min(t, s) t−m
ρBM (s, t) = √ , so ρBM (t − m, t) = p
ts t(t − m)
with a margin period of risk m. The volatility of the difference is thus
√
σdifference (t − m, t) = σ m
Taking some example parameters: t = 5, drift=0.01, σ = 0.20, m = 1/26:

quantile = 99% Uncollateralized Collateralized Ratio
PFE 1.09 0.09 11.9
PFL 1.24 0.10 11.8
For Normally-distributed exposures there is little distribution effect with collat-
eralization (because it remains Normal). Note that this only applies to Normal
distributions, a mixture distribution composed of many Normals would not show
this behaviour (see also comments on Portfolio composition versus smile effects).
3.3.2 LogNormal Distribution

Many approximations exist to price the spread options produced by collateral-
ization, e.g. (Bjerksund and Stensland, 2014) as well as semi-analytic techniques
(Caldana et al., 2016). For simplicity and clarity we use a Monte Carlo approach.
For accuracy we take sufficient samples, and repeats, that the half-width of the
95% confidence interval of the results is smaller than a one unit change of the
smallest digit reported. Thus the reported results are exact to the precision
displayed with 95% confidence.
The numerical setup is as follows
• Exposure is considered to follow Geometric Brownian Motion: drift, r =
1%; annualized volatility {10%, 20%, 40%}; profile forward times {1Y,
2Y, 5Y, 10Y}; margin period of risk {2W, 1M, 3M, 6M}. Not all results
are shown.
• Autocorrelation from Geometric Brownian Motion.
• Calculations are done with 220 samples repeated 16 times to get the con-
fidence interval.
Tables 3 and 4 show quantiles and expected shortfall for Q = 95%, 99% with
σ = 20%. PFE misses the change in distribution of exposure and this effect is
roughly 20% overlooked for Q = 95% and roughly 10% overlooked for Q = 99%.
Relative effects are almost independent of time as they are driven mostly by the
margin period of risk.
10
Years
Q=95%, σ=20%, MPOR=2W 1Y 2Y 5Y 10Y
Uncollateralized: Quantile/fwd (%) 125 142 181 233
Uncollateralized: ES/fwd (%) 136 161 221 313
α= ratio -1 (%) 9 13 22 34
Collateralized: Quantile/fwd (%) 5.9 6. 6.4 7.
Collateralized: ES/fwd (%) 7.5 8. 9.2 11.2
β= ratio -1 (%) 29 33 44 59
ratio increase, ((β + 1)/(α + 1) − 1, %) 18 17 17 19
Table 3: Effect of collateralization on ES/quantile ratio for Q = 95%. Using

only the quantile (i.e. PFE) misses the 17% to 19% increase in expected shortfall
above the quantile which PFL picks up. For setup see text.
Years
Q=99%, σ=20%, MPOR=2W 1Y 2Y 5Y 10Y
Uncollateralized: Quantile/fwd (%) 143 172 245 358
Uncollateralized: ES/fwd (%) 154 190 288 453
α= ratio -1 (%) 7 10 18 26
Collateralized: Quantile/fwd (%) 8.6 9.2 10.8 13.5
Collateralized: ES/fwd (%) 10.1 11.1 14. 18.8
β= ratio -1 (%) 18 21 29 39
ratio increase, ((β + 1)/(α + 1) − 1, %) 10 9 10 10
Table 4: Effect of collateralization on ES/quantile ratio for Q = 99%. Using

only the quantile (i.e. PFE) misses the 9% to 10% increase in expected shortfall
Table 5 shows the effects of volatility and time for Q = 95%, whereas Table
7 shows the effect of MPOR and time. Increasing MPOR decreases the dis-
tribution effect. Increasing volatility may increase or decrease the distribution
effect depending on time. This is because there is a countervailing effect from
the increase of correlation with time as the MPOR is relatively smaller with
longer times.
3.3.3 Collateralization: portfolio aka smile effects

Theoretically it is difficult to distinguish the effect of smiles versus the effect of
portfolio composition because any complex PDF can be created from a set of
scaled and shifted LogNormal distributions with different volatilities to within
a given tolerance. This is the kernel decomposition of the PDF, which can be
rephrased as, for example, creating smiles using mixture distributions (Brigo
and Mercurio, 2006).
The setup is to have two exposure distributions both driven by the same
Brownian motion but with different means (one increased by 25%) and volatil-
ities (one decreased by roughly 25%).
Table 6 shows the potential effect of portfolio composition w.r.t. collateral-
ization, an increase of roughly half compared to Table 3. This is a major effect
11
Q=95%, MPOR=2W Years
σ 1Y 2Y 5Y 10Y
10% 21 20 19 18
20% 18 17 17 19
40% 16 17 20 25
Table 5: Effect of volatility and time. Using only the quantile (i.e. PFE) misses
the 16% to 25% increase in expected shortfall above the quantile which PFL
picks up. For setup see text.
Q=95%, MPOR=2W Years

σ 1Y 2Y 5Y 10Y
10% 23 23 23 24
20% 22 22 24 27
40% 22 25 30 37
Table 6: Portfolio effect, aka effect of mixture distribution or smiles. Using only
the quantile (i.e. PFE) misses the 22% to 37% increase in expected shortfall
Q=95%, T=1Y MPOR

σ 2W 1M 3M 6M
10% 21 21 20 19
20% 18 17 16 13
40% 16 15 11 6
Table 7: Effect of MPOR versus time. Using only the quantile (i.e. PFE) misses
the 6% to 21% increase in expected shortfall above the quantile which PFL picks
up. For setup see text.
12
that is missed by PFE.
3.3.4 Example Instruments

10Y USD IRS. The 10Y USD IRS from Figure 1 is displayed in Figure 4
showing the relative expected shortfall(95%) of collateralized (10B) with uncol-
lateralized. The interest rate dynamics use a CIR stochastic volatility Libor
Market Model and calibrated to the 5x5 swaption smile as (Green and Kenyon,
2017). The magnitude of the ratio is consistent with Table 3 and the portfolio
effects in Table 6 (an interest rate swap acts as a portfolio of forward contracts).
The ratio with IM is not shown as it is infinity for the majority of the time.
PFL gives the residual risk (ES) output but PFE is mostly zero. That is, PFE
is simply unusable. Recall that with IM trading volume is largely controlled by
the quantity of IM required whilst PFL measures the residual counterparty risk.
10Y USD IRS, q=0.95
2.0
1.5
Relative ES
(coll/uncoll)
1.0
0.5
0.0
0 2 4 6 8 10
Years
Figure 4: Relative expected shortfall(95%) for 10Y IRS from Figure 1 comparing
collateralized with uncollateralized. This is a measure of the change in exposure
because of distribution changes that PFE misses, but PFL captures. Spikes
are from return of collateral as before. (Output from 212 paths shown, hence
the noise level as compared to table results which used 220 samples and 16
repetitions.))
4 Conclusions
Developing challenges to PFE in terms of widespread IM and netting of collateral
and trade flows mean that PFE will become of questionable value (identically
zero) as a counterparty trading limit. Outside of widespread IM and netting of
collateral and trade flows, pre-existing challenges to PFE (exposure distribution
shape, collateralization, multiple seniorities, overlap with CVA) mean that it is
already a poor fit for purpose. We propose using expected shortfall and loss
given default to arrive at Potential Future Loss (PFL). PFL, together with its
adjusted and protected versions (including incurred CVA and credit protection,
with effects on limits) are robust against both pre-existing challenges (such as
13
distribution shape, overlap with CVA, collateralization, and multiple seniorities)
and developing challenges to PFE.
It is not solely because PFE becomes zero with IM that we need an alterna-
tive. PFL provides visibility of what risk remains after IM, but if there was a
situation where PFL was zero we would not propose another alternative. This
is because PFL is a much more comprehensive measure that PFE, because it
looks out to all tail events (above the quantile) and includes them. PFL pro-
vides a monitor on what risk remains after IM but it is IM that is doing the
essential risk control (by removing risk). We propose PFL also because of the
pre-existing challenges to PFE (exposure distribution shape, overlap with CVA
and credit mitigation, collateralization, multiple seniorities) that PFL answers.
Acknowledgements
The authors would gratefully like to acknowledge feedback from participants at
the MVA Round-table (September 2017, Canary Wharf) and discussions with
Sebastian Steinfeld and Helmut Glemser.
References
Andersen, L. B., M. Pykhtin, and A. Sokol (2016). Rethinking margin period
of risk. Available at SSRN 2719964 .
BCBS-189 (2011). Basel III: A global regulatory framework for more resilient
banks and banking systems. Basel Committee for Bank Supervision.
BCBS-237 (2012). Basel III counterparty credit risk and exposures to central
counterparties - Frequently asked questions (Dec 2012). Basel Committee for
Bank Supervision.
BCBS-317 (2015). Margin requirements for non-centrally cleared derivatives.
Basel Committee for Bank Supervision.
BCBS-325 (2015). Review of the Credit Valuation Adjustment Risk Framework.
Consultative Document. Basel Committee for Bank Supervision.
Bjerksund, P. and G. Stensland (2014). Closed form spread option valuation.
Quantitative Finance 14 (10), 1785–1794.
Brigo, D. and F. Mercurio (2006). Interest Rate Models: Theory and Practice,
2nd Edition. Springer.
Caldana, R., G. Fusai, A. Gnoatto, and M. Grasselli (2016). General closed-form
basket option pricing bounds. Quantitative Finance 16 (4), 535–554.
Chen, H., R. Cui, Z. He, and K. Milbradt (2014). Quantifying liquidity and
default risks of corporate bonds over the business cycle. Technical report,
National Bureau of Economic Research.
Green, A. and C. Kenyon (2017). XVA at the Exercise Boundary. Risk.
14
ISDA-SIMM-2 (2017). ISDA SIMM(tm) Methodology Version 2.0. http://
www2.isda.org/functional-areas/wgmr-implementation/.
Jankowitsch, R., F. Nagler, and M. Subrahmanyam (2014). The determinants

of recovery rates in the US corporate bond market. Journal of Financial
Economics 114 (1), 155–177.
Kenyon, C., A. Green, and M. Berrahoui (2015). Which Measure for PFE? The
Risk Appetite Measure, A. Available at SSRN .
LCHSA (2017). LCH SwapAgent, Clearing the Way for Non-Cleared. Technical
report, LCH.
15

Counterparty Limit

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Counterparty Limit

Загружено:

Авторское право:

Доступные форматы

Introduction to Counterparty Risk

Bocconi University, 18/04/2016

Paola Mosconi Lecture 8 1 / 84

Paola Mosconi Lecture 8 2 / 84

Counterparty Risk: Definition

Counterparty risk is affected by:

Paola Mosconi Lecture 8 5 / 84

Counterparty Risk: Features

Paola Mosconi Lecture 8 6 / 84

As a byproduct of such transfer:

Paola Mosconi Lecture 8 7 / 84

OTC Market – BIS 2015

Paola Mosconi Lecture 8 8 / 84

OTC Market – Global Chart

Paola Mosconi Lecture 8 9 / 84

OTC Market – Interest Rate Derivatives Chart

Paola Mosconi Lecture 8 10 / 84

Counterparty Risk: Risk Management vs Pricing

Two approaches to counterparty risk:

counterparty risk management:

counterparty risk from a pricing point of view:

However, Basel III has made the distinction less clear-cut.

Paola Mosconi Lecture 8 11 / 84

Counterparty Risk Management

Credit VaR is calculated through the following steps:

Paola Mosconi Lecture 8 12 / 84

Counterparty Risk: Pricing

Pricing concerns updating the value of a specific instrument or portfolio, traded

Paola Mosconi Lecture 8 13 / 84

Current exposure (CE) is the current value of the exposure to a counterparty.

Exposure at Default (EAD) is defined in terms of the exposure valued at the

Paola Mosconi Lecture 8 14 / 84

Expected positive exposure (EPE) is the average EE in time up to a given future

Paola Mosconi Lecture 8 15 / 84

Credit Value Adjustment: Introduction

CVA is defined as:

Paola Mosconi Lecture 8 17 / 84

CVA is defined, according to Canabarro and Duffie (2004), as the difference

CVA := Et [Π(t, T )] − Et [ΠD (t, T )] (1)

unilateral (asymmetric) CVA, if only the default of the counterparty is

CVA Definition 1: Unilateral Case

Paola Mosconi Lecture 8 19 / 84

CVA Definition 1: Bilateral Case

Et [ΠD (t, T )] = V (t)

Paola Mosconi Lecture 8 20 / 84

CVA Definition 2: Expected Loss

We consider here the unilateral case1 .

Loss(τ ) = 1{τ ≤T } LGD (V (τ ))+

CVA = E[ D(0, τ ) Loss(τ )]

The resulting expression is in agreement with eq. (2).

1 The extension to the bilateral case is straightforward.

2002 ISDA Master Agreement: Events of Default

Paola Mosconi Lecture 8 22 / 84

Close-out Amount and Netting Set

Paola Mosconi Lecture 8 23 / 84

Risk Free vs Substitution Close-out

Paola Mosconi Lecture 8 24 / 84

Positive Part of Close-out

Consider the point of view of the non-defaulting party. If:

Paola Mosconi Lecture 8 25 / 84

CVA is a credit hybrid.

The positive part of the close-out amount introduces an element of optionality

Optionality renders the payoff under counterparty risk model dependent,

Paola Mosconi Lecture 8 26 / 84