Articles by kenli

LO 32.2: Explain the treatment o f counterparty credit risk (CCR) both as a credit risk and as a market risk and describe its implications for trading activities and risk management for a financial institution.
The treatment of CCR as a market risk was historically done through pricing in a credit valuation adjustment (CVA). CVA represents the market value of the CCR. Before the 20072009 financial crisis, institutions saw stable credit spreads and CVAs that made up only a small component of a derivatives portfolio. When the financial crisis resulted in unusual losses and gains, institutions began to pay closer attention to risk managing the CVA.
Financial institutions may view CCR as either a credit risk or a market risk and may manage the credit portfolio accordingly, but looking at it as only one type of risk (in a silo) exposes the institution to the risk from the other side.
Treating CCR as a credit risk exposes the institution to changes in CVA; therefore, CVA must be included when valuing a derivatives portfolio. Not including the CVA could lead to large swings in market value. Credit risk is managed at inception or typically through collateral arrangements, but it is not actively managed once the trades are set up. Since at default all trades need to be replaced in the market, emphasis is on risk mitigation and credit evaluation.
Treating CCR as a market risk allows an institution to hedge market risk losses but leaves it exposed to declines in counterparty creditworthiness and default. However, CCR can be hedged through replacing contracts with a counterparty instead of waiting for default to occur. This can be achieved by buying positions in proportion to the counterpartys probability of default (PD). A counterparty with a low PD will only have a small component of its trades replaced this way, while counterparties with deteriorating credit quality will see their trades replaced faster and moved to other counterparties.
The treatment of CCR as both a credit risk and a market risk creates a large variety of measurements that can be complex to interpret. For example, credit risk uses current exposure, peak exposure, and expected exposure, while market risk uses CVA and variability in CVA (measured by VaR of CVA). When stress testing the portfolio, the number of stress
2018 Kaplan, Inc.
Page 243
Topic 32 Cross Reference to GARP Assigned Reading – Siddique and Hasan, Chapter 4
results can be very large. By classifying CCR as both a credit risk and a market risk, the number of stress results would equal at least twice the number of counterparties plus one (stresses are run for each counterparty as well as the aggregate portfolio), and would be at least double that amount again if instantaneous shocks were considered in addition to stressed risk measures.
S t r e s s T e s t i n g C u r r e n t E x p o s u r e

LO 32.1: Differentiate am ong current exposure, peak exposure, expected exposure, and expected positive exposure.
The concept of counterparty credit risk (CCR) and its measurement and management gained prominence in the 1990s, and it now forms a critical part of most organizations risk governance. Financial institutions incorporated CCR through analyzing their derivatives exposures and by tracking the current exposure to their counterparties. Institutions measured regulatory capital for CCR as add-ons to current exposures, calculated as a percentage of gross notional derivatives values.
With the rise in importance of measuring CCR, modeling CCR also evolved. Initially, potential exposure models were used to measure and limit CCR. This approach evolved into expected positive exposure models, which allowed derivatives to be incorporated into portfolio risk models along with loans. The measurement of CCR also formed the basis for regulatory capital under Basel II and allowed for the incorporation of credit mitigants into risk modeling, including netting agreements.
There are four important definitions of exposure measures:
Current exposure. Also called replacement cost, current exposure is the greater of (1) zero or (2) the market value of a transaction (or a portfolio of transactions) that would be lost if the counterparty defaulted and no value was recovered during bankruptcy. Peak exposure. Peak exposure measures the distribution of exposures at a high percentile (93% or 99%) at a given future date prior to the maturity of the longest maturity exposure in the netting group. Peak exposure is usually generated for many future dates.

Page 242
2018 Kaplan, Inc.
Topic 32 Cross Reference to GARP Assigned Reading – Siddique and Hasan, Chapter 4

Expected exposure. Expected exposure measures the mean (average) distribution of exposures at a given future date prior to the maturity of the longest maturity exposure in the netting group. Expected exposure is also typically generated for many future dates. Expected positive exposure (EPE). EPE is the weighted average of expected exposures over time. The weights represent the proportion of individual expected exposures of the entire time interval. For the purposes of calculating the minimum capital requirement, the average is measured over the first year or over the length of the longest maturing contract.
One of the issues with CCR is wrong-way risk. Wrong-way risk is the risk that the exposure from a counterparty grows at the same time that the risk of default by the counterparty increases. Note that wrong-way risk does not arise with fixed-rate loans.
C C R T r e a t m e n t

LO 31.3: Discuss the im pact o f wrong-way risk on collateral and central counterparties.
Collateral can be viewed as a way to reduce exposure. Therefore, when exposure is increasing significantly, its important to evaluate the overall impact of collateral on WWR. In cases where exposure is gradually increasing (before default), collateral is typically taken to minimize the impact of WWR. In this scenario, the benefit from collateral will increase as WWR increases, because additional collateral is relatively easy to request and receive. However, in cases where exposure jumps at a certain point in time, the benefits of collateral will be very limited. For example, with a jump in exposure, such as a currency devaluation associated with a sovereign default, it is much more difficult to receive collateral in a timely fashion.
Central counterparties (CCPs) are particularly susceptible to WWR given their dependence on collateral and default fund contributions. Recall from Topic 28 that the CCP relies on a defaulting members posted margin (i.e., collateral) and default fund contribution to absorb potential losses. If these amounts do not cover losses, the CCP will need to use their own equity capital and/or default funds from non-defaulting members to help remain solvent.
However, the CCPs loss waterfall structure may be insufficient if member initial margins and default fund contributions fail to incorporate WWR. Since WWR tends to increase with increasing levels of credit quality, it could be argued that CCPs should demand higher levels of margin and default fund contributions from those members with higher credit quality. In addition, the collateral accepted by the CCP may also carry WWR. Some members may choose to post risky and illiquid assets as collateral, which may create higher levels of WWR for the CCP. One way to mitigate this practice is for the CCP to impose higher haircuts on specific assets that are accepted as collateral.
2018 Kaplan, Inc.
Page 237
Topic 31 Cross Reference to GARP Assigned Reading – Gregory, Chapter 17
K e y C o n c e p t s
LO 31.1
Financial institutions should pay more attention to wrong-way risk and right-way risk for planning purposes. The recent global financial crisis and European sovereign debt crisis have illustrated the significance of these risks.
Numerous macroeconomic events can impact exposure risk and default probability, producing an overall increase in counterparty credit risk. Position gains may not materialize due to an increase in the counterpartys overall risk. This is an example of wrong-way risk.
On the other hand, favorable associations between exposure risk and default probability resulting from changes in macro factors may produce a decline in overall counterparty credit risk. This is an example of right-way risk.
L 0 31.2
Wrong-way risk and right-way risk can be identified in numerous investment products and transactions, such as call options, put options, credit default swaps, foreign currency transactions, interest rate products, currency swaps, and forward contracts.
The key to identify wrong-way and right-way risk is to assess the impact on overall counterparty risk. If the co-movement between risk exposure and default probability generates an overall increase (decrease) in counterparty risk, it would be an example of wrong-way risk (right-way risk).
During the recent global financial crisis, credit default swaps offered a classic example of wrong-way risk. The buyers of credit default swaps (protection against the default of bond issuers) experienced a substantial gain as the values of the bonds backed by mortgage- backed securities started tumbling. However, the collapse of the mortgage market not only increased the risk exposure but also the default probability, leading to an overall increase in counterparty risk. There were many buyers of credit default swaps whose gains remained paper gains due to the deteriorating creditworthiness of the counterparty.
LO 31.3
When exposure is increasing gradually, the impact of collateral on WWR will be beneficial. As WWR increases, more collateral can be taken. However, when there is a jump in exposure, the impact of collateral on WWR will be limited due to the inability to receive collateral in a timely fashion.
Central counterparties (CCPs) may be impacted by WWR if they fail to incorporate this risk into their members initial margins and default fund contributions. To mitigate the impact of WWR, CCPs should require higher margin (i.e., collateral) and default funds from members with better credit quality. CCPs should also impose higher haircuts on any posted collateral that may increase WWR.
Page 238
2018 Kaplan, Inc.
Topic 31 Cross Reference to GARP Assigned Reading – Gregory, Chapter 17
C o n c e p t C h e c k e r s
1.
How many of the following statements regarding wrong-way risk (WWR) and right- way risk (RWR) are correct? I. Co-movement in risk exposure and default probability producing a decline in
overall risk is an example of wrong-way risk.
II. Co-movement in risk exposure and default probability producing an increase in
overall counterparty risk is an example of right-way risk.
III. Co-movement in risk exposure and default probability producing neither a
decline nor an increase in the overall counterparty risk is an example of wrong- way risk.
IV. Co-movement in risk exposure and default probability producing a decline in risk exposure but an increase in counterparty default probability is an example of right-way risk.
A. None. B. All. C. Two. D. Three.
2.
Which of the following events would likely lead to an increase in WWR? I. The borrower and the guarantor are business partners. II. A monoline insurer sold protection concentrated in a business or industry. A. I only. B. II only. C. Both I and II. D. Neither I nor II.
3.
Which of the following statements regarding WWR and RWR is correct? A. A long put option is subject to WWR if both risk exposure and counterparty
default probability decrease.
B. A long call option experiences RWR if the interaction between risk exposure
and counterparty default probability produces an overall decline in counterparty risk.
C. Declining local currency can decrease the position gain in a foreign currency
transaction, while increasing risk exposure of the counterparty.
D. The 20072009 credit crisis provides an example of WWR from the perspective
of a long who had sold credit default swaps (CDSs) as protection against bond issuers default.
2018 Kaplan, Inc.
Page 239
Topic 31 Cross Reference to GARP Assigned Reading – Gregory, Chapter 17
4.
How many of the following statements regarding counterparty risk are correct? I.
Speculation in normal-functioning derivatives markets automatically produces RWR.
II. RWR has been the center of attention in historical context, whereas WWR has
not been paid much relative attention.
III. The counterparty default probability does not enter into the equation for
estimating the overall counterparty risk.
IV. Unlike exposure to OTC derivatives, which is normally assumed to be a fixed
amount for a specified time period, exposure to bank loans fluctuates depending on market conditions.
A. None. B. All. C. Two. D. Three.
3.
Which of the following statements is correct? I. Depreciation of the yen after the default of Lehman Brothers gave a substantial gain to Japanese bank foreign currency swaps positions to obtain dollar funding in interest rate swaps.
II. Fixed-rate receivers experience a value gain to the extent that the swap rate
increases.
A. I only. B. II only. C. Both I and II. D. Neither I nor II.
Page 240
2018 Kaplan, Inc.
Topic 31 Cross Reference to GARP Assigned Reading – Gregory, Chapter 17
C o n c e p t C h e c k e r An s w e r s
1. A A decline in overall counterparty risk is an example of right-way risk. An increase in overall
counterparty risk is an example of wrong-way risk. An increase in overall counterparty risk is a condition for the emergence of wrong-way risk. A decline in risk exposure but increase in counterparty default probability may or may not lower overall counterparty risk.
2. C WWR will increase if the borrower and guarantor are business partners. The guarantees offered by a monoline insurer may turn out to be worthless if the risk exposure increases and the guarantor is hit by a flood of claims due to a concentrated position in an industry or business.
3. B A long call option experiences RWR if risk exposure and counterparty default probability results in decreased counterparty risk. A long put option is subject to WWR if both risk exposure and counterparty default probability increase. Declining local currency can increase the position gain in a foreign currency transaction, while increasing counterparty risk exposure. The 2007-2009 credit crisis provides an example of WWR from the perspective of a long who had bought CDSs as protection against bond issuers default.
4. A Hedging, and not speculation, in normal functioning markets automatically produces
RWR. Historically, RWR was relatively neglected by institutions for planning purposes. The counterparty default probability is one of the key elements in estimating overall counterparty risk. OTC exposures fluctuate based on market conditions.
5. D Appreciation, and not depreciation, of the yen generated a substantial gain for Japanese
banks with foreign currency swaps positions. A fixed-rate receiver experiences a value gain to the extent that the swap rate declines.
2018 Kaplan, Inc.
Page 241
The following is a review of the Credit Risk Measurement and Management principles designed to address the learning objectives set forth by GARP. This topic is also covered in:
T h e Ev o l u t io n o f St r e s s T e s t i n g C o u n t e r p a r t y E x p o s u r e s
Topic 32
E x a m F o c u s
In this topic, we take a detailed look at counterparty credit risk measurement and management. We begin by differentiating between the various measures of exposure. Next, we look at the treatment of counterparty credit risk, both as a credit risk and as a market risk. We then review the credit valuation adjustment (CVA) and stresses to the CVA. For the exam, be able to describe a stress test that can be performed on both a loan portfolio and a derivatives portfolio. In addition, ensure that you are able to calculate the stressed expected loss. Finally, be able to calculate stressed CVA and understand how the debt value adjustment (DVA) differs from the CVA.
C o u n t e r p a r t y C r e d i t R i s k E x p o s u r e M e a s u r e s

LO 31.2: Identify examples o f wrong-way risk and examples o f right-way risk.
For this LO, well create a few hypothetical examples of WWR and RWR. For example, what if Company XYZ (the borrower) and the guarantor on XYZs loan, Company ABC, share ownership in a business (or are in the same industry)? Due to some market or economic factors, both may default together (WWR), whereas if the guarantor and the borrower are not in the same industry (nor have shared ownership), XYZs loan guarantee may still be valid, even if XYZ defaults (RWR).
What if ABC has sold protection much higher than its capital in a concentrated area (business or industry) and XYZ has bought protection (insurance) from ABC? Macro factors may increase the exposure for the guarantor (ABC), and due to positive interaction between exposure and credit quality, the overall counterparty (guarantor) risk increases to
Page 232
2018 Kaplan, Inc.
Topic 31 Cross Reference to GARP Assigned Reading – Gregory, Chapter 17
the extent that XYZs protection becomes meaningless (WWR). In contrast, the reverse of the situation may generate a favorable state an increase in exposure may be sufficiently offset by an increase in creditworthiness.
The CVA, which is based on the amount of counterparty risk, is generally approximated by the product of exposure and the default probability of the counterparty (for a given recovery rate). This estimation is based on an underlying assumption that these events are independent. However, they may not be independent (as evidenced in the recent financial crisis). Unfavorable (favorable) association between default probability (credit risk) and exposure (market risk) may produce WWR (RWR), increasing (decreasing) the overall CVA.
Quantifying WWR and RWR involves estimation of the CVA based on expected exposure, conditional on counterparty default (under the more realistic scenario of the presence of interconnected markets with systemic risk), whereas under the independence assumption, we use unconditional default probability.
It is estimated that conditional expected exposure will increase if the exposure (e.g., value of a forward contract) and the default probability of the counterparty are positively correlated, exhibiting WWR. On the other hand, negative correlation in this instance will lower the conditional expected exposure, showing RWR.
As discussed earlier, the overall counterparty risk stems from a situation in which the counterparty credit quality is linked with macro (and global) factors that also impact the exposure of transactions. The transaction can be any of the following: put options, call options, foreign currency transactions, forward contracts, credit default swaps, or interest rate and currency swaps. Let us examine WWR and RWR as they relate to some of these transactions.
Over-the-Counter Put O ption
A put option gives the right to the long (buyer) to sell an underlying instrument at a predetermined price whereas the short (counterparty) is obligated to buy if the option is exercised. Out-of-the money put options have more WWR than in-the-money put options.
Macroeconomic events (such as interest rates, inflation, industry- and sector-specific factors, or global factors) may deteriorate the creditworthiness of the counterparty, increasing the default probability. The same factors may trigger a fall in the underlying (e.g., stock) assets price, generating positive payoffs for the long but increasing the counterparty risk exposure. Before the long gets too excited to see an increase in payoffs, he is hit by the realization of increasingly becoming a victim of WWR, due to positive correlation between the risk exposure of the counterparty and probability of default of the counterparty producing an overall increase in counterparty risk. The payoffs may not materialize, although they are increasing. On the other hand, normalcy of the transaction would be termed as RWR if the counterparty is able to fulfill its obligation despite an increase in its position obligation.
2018 Kaplan, Inc.
Page 233
Topic 31 Cross Reference to GARP Assigned Reading – Gregory, Chapter 17
Professors Note: We are assuming in the previous put option example that the counterparty and the underlying issuer are the same in order to clearly illustrate WWR. The positive association between default probability and exposure will still give rise to WWR i f the counterparty and underlying issuer are not the same.
Over-the-Counter Call O ption
A call option gives the right to the long (buyer) to buy an underlying instrument at a predetermined price whereas the short (counterparty) is obligated to sell at the agreed-upon price if the option is exercised. Like the put option, we are assuming the counterparty and the underlying issuer are the same.
Assume that due to changes in some macroeconomic and global factors, the default probability of the counterparty declines, and the price of the underlying asset (e.g., stock) increases, producing higher payoffs for the call buyer. In this instance, his excitement of making money will be appropriate because the counterparty will be in a strong position to pay off its obligation (due to the overall increase in creditworthiness). Such an outcome will be considered the normalcy of the transaction, and it is termed RWR. The short is able to fulfill its obligation despite the increase in its position obligation. On the other hand, if the counterparty is unable to fulfill its obligation due to the increase in its position obligation (higher value of underlying for the long, but higher obligation for the short an increase in counterparty risk exposure), it would be an example of WWR (from the standpoint of the long position).
Credit D efault Swaps (C D Ss)
The 20072009 credit crisis offers a classic example of WWR from the perspective of the longs (i.e., the buyers) who had bought protection on issuers default on collateralized debt obligations (CDOs) or bonds backed by mortgage-backed securities (MBSs) via credit default swaps (CDSs).
As the real estate bubble burst and the market started taking a downward freefall, the value of MBSs started exhibiting a freefall as well. The monoline insurers, such as AMBAC and MBIA, had taken highly concentrated positions in offering protection against MBSs and CDOs. As the issuers of MBSs and CDOs started defaulting, the insurers were flooded by claims from the ones who had bought the protection (i.e., holders of CDSs).
The value of CDSs was rising, but this gain was generating an increase in risk exposure to the counterparty. Both the probability of default and the risk exposure of the insurers were rising. The unfortunate buyers of protection soon found out that the macrocredit and exposure linkage had produced unfavorable results for them. Despite huge gains on their positions, nothing materialized due to the deteriorating creditworthiness of the issuers, an example of WWR.
The normalcy of the transaction would be if the counterparty could fulfill its obligation despite an increase in position exposure (perhaps due to a negative association between risk exposure and probability of default). This would be an example of RWR. If insurance
Page 234
2018 Kaplan, Inc.
Topic 31 Cross Reference to GARP Assigned Reading – Gregory, Chapter 17
company ABC, for example, had taken a nonconcentrated exposure, it might not have experienced a decline in its creditworthiness (due to fewer claims) and would have been able to satisfy its obligations despite increasing risk exposure in the CDSs.
Foreign Currency Transactions
Consider a commercial bank in a developed economy (e.g., the United States) that enters into a cross currency agreement with a commercial bank (counterparty) in an emerging market (e.g., Uzbekistan), under which the counterparty will deliver developed market currency in return for local currency.
Macro conditions in the emerging country, such as a sovereign debt crisis, generate credit stress for the local bank, as well as a decline (depreciation) of local currency. The value of the transaction increases substantially for the financial institution in the developed economy due to the declining currency of the emerging economy. At the same time, the counterparty risk exposure increases as the gain for the financial institution in the developed economy increases.
Increases in default probability (due to credit stress) and risk exposure (due to declining currency) increase counterparty risk, resulting in WWR for the financial institution in the developed economy.
If the counterparty risk exposure and the credit quality are not unfavorably associated, then the risk exposure may increase, but the probability of default may decline (due to improvement in creditworthiness), producing a reduction in overall counterparty risk. This would be an example of RWR.
Foreign Currency Swaps
A real-world example will further clarify WWR in the foreign currency swaps market. Prior to the recent credit crisis in the United States, numerous financial institutions in Japan had entered into swap agreements with U.S. financial institutions to obtain dollar funding by using yen. They pledged yen to get U.S. dollars. After the default of Lehman Brothers, the financial crisis reached its peak, raising grave concerns about the economic slowdown of the U.S. and European economies. The yen significantly appreciated against the U.S. dollar, resulting in a substantial gain to Japanese bank positions (the pledged yen will buy more dollars, and U.S. banks will have to surrender more dollars for the pledged yen), increasing the counterparty risk exposure for Japanese banks. At the same time, deteriorating macro conditions had a negative impact on U.S. banks and the economy. In addition, the default probabilities of the U.S. financial institutions increased. Positive (unfavorable) association between counterparty risk exposure and default probability generated an overall increase in counterparty risk for Japanese banks, and they experienced WWR.
If the risk exposure and default probabilities are not positively associated, the normalcy of the transaction would balance out the increase in risk exposure by improving the creditworthiness of the financial institutions (macro factors may be related to both events in a different manner), lowering overall counterparty risk. The counterparty is able to meet its obligation despite an increase in risk exposure (due to an appreciating yen). This would be an example of RWR.
2018 Kaplan, Inc.
Page 235
Topic 31 Cross Reference to GARP Assigned Reading – Gregory, Chapter 17
Interest Rate Transactions
Interest rate swaps provide another good illustration of WWR. In an interest rate swap, one party (i.e., the long or fixed-rate receiver) enters into an agreement with a counterparty (i.e., the fixed-rate payer) to receive a fixed rate and pay a floating rate. The fixed-rate receiver gains if the market interest rate (the swap rate) falls.
Assume due to macroeconomic conditions (e.g., an economic downturn), policy interest rates are lowered. The fixed-rate receiver experiences a value gain to the extent that the swap rate declines against the counterparty with the fixed-rate payer and floating-rate receiver. However, this gain for the fixed-rate receiver also produces an increase in its counterparty risk exposure. Furthermore, if the economic downturn would also increase the default probability, then overall counterparty risk will increase, generating WWR for the fixed-rate receiver.
This is exactly what happened during the recent European sovereign debt crisis. Due to lower inflation and an economic recession, the policy interest rates were lowered. The euro (interest rate) swap rate declined, producing a gain for those who were holding fixed interest rate receiver positions against Italian financial institutions (fixed-rate payer). However, the decline in the euro swap rate also increased the counterparty risk exposure. Deteriorating economic conditions also increased the default probability of Italian financial institutions. An increase in both the risk exposure and default probability resulted in an overall increase in counterparty risk, generating WWR for the holder of fixed-rate receiver swaps.
In the absence of a positive association between risk exposure and default probability, the Italian financial institutions might have been able to fulfill their obligations comfortably, despite the increase in exposure, generating RWR.
Com m odities
Airlines hedge against the risk of rising oil prices. For example, assume an airline is long an oil forward contract at a fixed price. The counterparty is a dealer who has taken heavy concentrated positions. If oil prices rise, the gains for the airline will rise. The airline will buy cheap oil because the spot price will be higher than the locked-in forward price, but at the same time, the risk exposure for the dealer will increase. Because the dealer had concentrated positions, there may be a flood of claims (several forward contract claims brought by various airlines), putting intense pressure on the credit quality of the counterparty. Thus, an increase in both the risk exposure and the default probability will increase overall counterparty risk, producing WWR.
On the other hand, a dealer with a nonconcentrated position may continue to have sound creditworthiness despite rising exposure. Thus, the dealer will be able to fulfill her obligation, lowering the overall expected amount of risk exposure from the standpoint of the airline. This would be an example of RWR.
Page 236
2018 Kaplan, Inc.
Topic 31 Cross Reference to GARP Assigned Reading – Gregory, Chapter 17
T h e Im p a c t o f W W R o n C o l l a t e r a l a n d C C P s

LO 31.1: Describe wrong-way risk and contrast it with right-way risk.
Wrong-way risk (WWR) is an outcome of any association, dependence, linkage, or interrelationship between exposure and counterparty creditworthiness that generates an overall increase in counterparty risk and, therefore, an increase in the amount of the credit value adjustment (CVA). WWR also results in a reduction of the debt value adjustment (DVA). WWR can be hard to determine due to difficulties assessing the relationship among variables and the lack of relevant historical data.
Right-way risk (RWR) is just the opposite of WWR. That is, any dependence, linkage, or interrelationship between the exposure and default probability of a counterparty producing an overall decrease in counterparty risk is described as RWR. RWR decreases the CVA and increases the DVA.
It is also worth mentioning that WWR has been the center of attention in historical context, while RWR has been paid relatively little attention. However, both risks are important, and financial institutions should strive to increase RWR and decrease WWR.
.Another way to contrast WWR and RWR is to think that normality in derivatives markets is an example of RWR. That is, derivatives transactions produce intended results if the market is functioning in an expected manner. For instance, a coffee producer would sell (i.e., short) forward or futures contracts in order to protect against the downside risk of falling prices in the future, and a textile owner (that manufactures cotton cloth) would go
2018 Kaplan, Inc.
Page 231
Topic 31 Cross Reference to GARP Assigned Reading – Gregory, Chapter 17
long in cotton derivatives contracts if she anticipates a rise in cotton prices. Thus, RWR produces a favorable relation between default probability and exposure, reducing overall counterparty risk. Hedges, in normal functioning markets, should automatically generate RWR because the fundamental purpose of hedges is to curtail counterparty risk.
Professors Note: We are using derivatives markets ju st for illustration of wrong-way and right-way risks. By no means are these risks confined only to derivatives.
Markets and numerous interactions (e.g., market credit interaction) do not always produce normal behavior, as evidenced by the recent global financial crisis. Those who were seeking protection against the default of debt issuers (e.g., on collateralized debt obligations) became victims of WWR when unfavorable interaction between exposures and insurers default probabilities (which were supposed to provide protection) intensified the amount of counterparty credit risk.
The amount of counterparty risk is roughly equal to the product of exposure and the counterpartys default probability at a specified loss rate given default. Counterparty risk is a kind of credit risk that is estimated as loss reserve for loans, and in over-the-counter (OTC) derivatives markets, it is similar to estimating loan reserves.
Loan exposure, however, is normally assumed to be a fixed amount for a specified time period, whereas in OTC derivatives, the exposure fluctuates depending on market conditions. An example of WWR (RWR) would be a change in exposure and counterparty credit quality, producing an unfavorable (favorable) dependence in exposure and counterparty credit quality and resulting in an increase (decrease) in the amount of overall counterparty risk. The change in exposure and credit quality could be due to numerous external factors such as interest rates, inflation, exchange rate movements, and global events. Note that credit quality increases actually increase WWR. This is because counterparties with high credit quality are less likely to default. As a result, the occurrence of a default by a counterparty with high credit quality is less expected than a default by a counterparty with low credit quality.
E x a m p l e s o f W r o n g -W a y R i s k a n d R i g h t -W a y R i s k

LO 30.9: Calculate BCVA and BCVA spread.
Given a charge for counterparty risk that favors a stronger counterparty (typically a bank), CVA historically did not take into account that both counterparties could be subject to default risk. The 20072009 financial crisis changed risk parameters and perceptions drastically.
2018 Kaplan, Inc.
Page 223
Topic 30 Cross Reference to GARP Assigned Reading – Gregory, Chapter 14
Counterparty risk is now viewed as bilateral. Bilateral counterparty risk assumes that both counterparties may default. The formula for the credit value adjustment for a bilateral contract derives from the original CVA formula and assumes no simultaneous default (e.g., wrong-way risk).
The positive expression in the following bilateral credit value adjustment (BCVA) formula represents the CVA of the counterparty, C, and the negative expression represents the CVA of the financial institution, I. The CVA of the institution is also known as the debt value adjustment (DVA). The two terms in this expression are mirror images of one another. If the financial institution defaults first, it books a gain when the marked-to-market (MtM) exposure is negative. This is the case because the institution in default will only pay the counterparty the recovery amount of what they owe, which is a fraction of what they would have otherwise owed had they not defaulted. That difference is a gain to the defaulting party.
BCVA = CVA + DVA m
CVA = +L G D C x
2EE(ti) x PDC ( t ^ , t; ) J
DVA = LGDj x J 2 NEE(t; ) x PDj (t_ !, t; )
i=l m
i=l
where: NEE = negative expected exposure (EE from the counterpartys perspective)
Implications of the BCVA model include:
1. BCVA can be negative if the second expression is larger than the first, implying that the
risk value of a derivative is greater than its risk-free value. Stand-alone CVA may only be positive.
2. Two counterparties in agreement on the parameters of the BCVA equation will settle up owing to the equations symmetry. For example, Party 1 has BCVA of +X, then Party 2 has BCVA of X. Party 2 owes Party 1 +X due to Party 2s counterparty risk.
3. Netting with BCVA may be a disadvantage when the second expression dominates,
implying that the financial institution is riskier than its counterparty. Without netting, the institution may select contracts with a positive MtM settlement, discarding those with a negative MtM value as bankruptcy liabilities.
4.
If both parties agree on the parameters of the BCVA calculation, then counterparty risk in the marketplace (the sum of all BCVAs) is zero. However, this holds more in theory than in practice.
Professors Note: This BCVA formula excludes a survival probability, which considers the possibility that a financial institution may default before its counterparty. I f this is the case, the institution will not suffer a loss from the counterparty. The survival probability will be included in the BCVA equation in Topic 33, when we discuss stress testing the debt value adjustment.
Page 224
2018 Kaplan, Inc.
Topic 30 Cross Reference to GARP Assigned Reading – Gregory, Chapter 14
BCVA Spread
BCVA may be expressed as a spread or basis point charge to the weaker counterparty as follows:
BCVA(t,T)
CD^premium (*> T )
X DS x EPE – X f s xE N E
where : XCDS _ tke institutions own CDS spread ENE = expected negative exposure (the opposite of EPE)
Here the BCVA can be represented as a running spread. The formula implies that the institution may account for its own default through reduction of the unilateral CVA charge by its own credit spread multiplied by the ENE.
The calculation of this formula is identical to that for unilateral CVA. It differs only in that there is an additional identical subtractive calculation to reflect the BCVA of the financial institution.
Example: Computing BCVA
A risk manager needs a quick calculation of the BCVA on a swap. Assume inputs are as follows: EPE = 3%, ENE =3%, counterparty credit spread = 300 bps, and financial institution credit spread = 200 bps. Compute BCVA from the perspective of the financial institution.
Answer:
>From the perspective of the financial institution:
EPE x counterparty credit spread – ENE x institution credit spread
3% x 300 3% x 200 = 9 bps
This is what the financial institution may charge the counterparty for overall counterparty risk.
2018 Kaplan, Inc.
Page 225
Topic 30 Cross Reference to GARP Assigned Reading – Gregory, Chapter 14
K e y C o n c e p t s
LO 30.1
Motivations for pricing counterparty risk include (1) organization of responsibilities within the institution with respect to the pricing calculation and (2) determining whether a trade is sufficiently possible when factoring in counterparty risk charge.
LO 30.2
A credit value adjustment (CVA) is the price of counterparty risk. A positive value is a cost to the counterparty bearing the risk. The basic CVA formula assumes no wrong-way risk.
LO 30.3
CVA is calculated as follows:
m
CVA = L G D x y ^ x E E (ti)x P D (ti_1, t; )
i=l
CVA as a spread is CVA divided by the risky annuity for the maturity of the contract in question, producing an annual spread or charge expressed in basis points:
CVA(t,T)
CDS p r e m i u m (t>T)
X CDSx EPE
LO 30.4
Credit spread levels, the shape of the credit spread curve, the impact of the recovery rate, and the basis risk that arises from different recovery rate assumptions must all be considered when evaluating the impact of the default probability and recovery on CVA.
LO 30.3
Netting reduces the CVA price because it nets exposure when trades are settled.
Page 226
2018 Kaplan, Inc.
Topic 30 Cross Reference to GARP Assigned Reading – Gregory, Chapter 14
LO 30.6
Incremental CVA is used to calculate the cost of a new trade versus an existing one to determine the effect that the new trade has on CVA. Standalone CVA cannot do this. The formula for the incremental CVA calculation is identical to that for standalone CVA, except for the incremental expected exposure.
Marginal CVA is used for trade level attribution (i.e., to discover the determinants of the CVA). The formula for the calculation of marginal CVA is identical to that for standalone CVA, except for the substitution of marginal expected exposure for expected exposure.
LO 30.7
Collateralization reduces the CVA, changing only the counterpartys expected exposure.
LO 30.8
Bilateral CVA is a collateral value adjustment that takes into account the possibility that both counterparties could default, though not simultaneously. The CVA of the financial institution is also known as the debt value adjustment (DVA).The BCVA is the sum of CVA and DVA components.
BCVA = CVA + DVA m
CVA = +L G D C x
2EE(t;) x PDC (ti_1,ti) ^
DVA = LGDj x J2 NEE(t; )x P D ,(ti_ 1,ti)
i=l m
i=l
Implications of the BCVA model include:

BCVA can be negative. Stand-alone CVA may only be positive (representing a cost). Parties in agreement on the BCVA settle in accordance with the BCVA equations symmetry.
Netting may be disadvantageous where the financial institutions counterparty risk
exceeds that of the counterparty. Without it, the institution can pick which contracts to settle. Parties in agreement will have in theory all BCVAs net out to zero due to the symmetrical nature of the BCVA formula.

LO 30.9
The BCVA formula as a credit spread is:
BCVA(t,T)
CD^premium (L T )
xDS x EPE – xfDS x ENE
2018 Kaplan, Inc.
Page 227
Topic 30 Cross Reference to GARP Assigned Reading – Gregory, Chapter 14
To price BCVA: (credit spread of counterparty A x EPE) (credit spread of counterparty B x ENE) = either positive number that stronger counterparty charges the weaker one or negative number that the stronger counterparty may owe the weaker one if its ENE is greater than the counterpartys EPE.
Negative BCVA: The counterparty has a higher chance of defaulting and will owe money (C VA cost).
Positive BCVA: the counterparty has a lower chance of defaulting and will be owed money (recipient of CVA fee).
Page 228
2018 Kaplan, Inc.
Topic 30 Cross Reference to GARP Assigned Reading – Gregory, Chapter 14
C o n c e p t C h e c k e r s
1.
2.
3.
4.
5.
Which of the following statements is not a motivation for pricing counterparty risk? A. Accurate pricing should only account for the cost of the trade. B. Counterparty risk pricing should account for risk mitigants. C. Best practices organize pricing responsibilities in the organization. D. Pricing bilateral derivatives contracts.
With respect to the CVA calculation, which of the following statements is correct when a risk manager wishes to understand which trades have the greatest impact on a counterpartys CVA? The manager would use: A.
incremental CVA because it accounts for the change in CVA once the new trade is priced, accounting for netting.
B. marginal CVA because he could break down netted trades into trade level
C.
contributions. incremental CVA because he could break down netted trades into trade level contributions.
D. marginal CVA because it accounts for the change in CVA once the new trade is
priced, accounting for netting.
A trader wants to know the approximate CVA for a counterparty in a swap transaction. The counterpartys expected potential exposure (EPE) is 7%, and its credit spread is 473 basis points. What is the CVA as a running spread? A. 0.33%. B. 1.48%. C. 2.23%. D. 9.75%.
Regarding the impact of changes in the credit spread and recovery rate assumptions on the CVA, which of the following statements is true? A. A decrease in the credit spread will most often increase the CVA. B. For an upward-sloping curve, the CVA will be higher compared to a downward-
sloping curve.
C. Increasing the recovery rate will reduce the CVA. D. If the actual recovery rate is higher than the settled recovery rate, the CVA will most likely be higher compared to a situation where both recovery assumptions are the same for both rates.
When incorporating netting and collateralization into the CVA calculation, which of the following statements is incorrect?
I. Netting increases the CVA price because it reduces exposure when trades are
settled.
II. Collateralization does not change the CVA because it only changes the
counterpartys expected exposure.
A. I only. B. II only. C. Both I and II. D. Neither I nor II.
2018 Kaplan, Inc.
Page 229
Topic 30 Cross Reference to GARP Assigned Reading – Gregory, Chapter 14
C o n c e p t C h e c k e r An s w e r s
1. A Accurate pricing should account for not only the cost of the trade, but also the cost of
counterparty risk.
2. B Understanding which trades have the greatest impact on a counterpartys credit value
adjustment requires use of the marginal CVA. Incremental CVA, by contrast, is useful for pricing a new trade with respect to an existing one.
3. A Calculation of the CVA as a running spread entails multiplying the counterpartys EPE by its
credit spread:
7% x 4.75% = 33 bps
4. C
Increasing the recovery rate will increase the implied probability of default but reduce the resulting CVA. The CVA will most often increase given an increase in the credit spread. When considering the shape of the credit spread curve, the CVA will be lower for an upward- sloping curve compared to a downward-sloping curve. Finally, a higher actual recovery rate will most likely lead to a lower CVA compared to a situation where the recovery assumptions are the same for both actual and settled rates.
5. C Both statements are incorrect. Netting reduces the CVA price as it reduces exposure when trades are settled. Collateralization also reduces the CVA, changing only the counterpartys expected exposure (EE), but not its default probability.
Page 230
2018 Kaplan, Inc.
The following is a review of the Credit Risk Measurement and Management principles designed to address the learning objectives set forth by GARP. This topic is also covered in:
W r o n g -w a y R i s k
Topic 31
E x a m F o c u s
The most recent global financial crisis and European sovereign debt crisis illustrated the significance of wrong-way risk and right-way risk. For example, buyers of protection against bond defaults may witness an impressive gain in their position due to falling bond prices as a result of some macroeconomic events. However, at the same time, falling bond prices increase the risk exposure and default probability of a counterparty due to the adverse impact of macroeconomic events, resulting in an overall increase in counterparty risk. This is an example of wrong-way risk (WWR). Normal derivatives markets are characterized as possessing right-way risk (RWR), in which hedges produce successful expected results. Macroeconomic events affect risk exposure and default probability in a favorable manner such that the overall expected counterparty risk declines. For the exam, be able to explain both wrong-way risk and right-way risk as well as identify these risks in transactions such as put options, call options, credit default swaps, foreign currency transactions, interest rate and currency swaps, and commodities.
W r o n g -W a y R i s k v s . R i g h t -W a y R i s k

LO 30.8: Describe debt value adjustm ent (DYA) and bilateral CVA (BCYA).

LO 26.6: Explain the features o f a collateralization agreement.
Collateral agreements are typically negotiated prior to any trading, and they are often updated prior to an increase in trading. Parameters must be clearly defined, and parties must balance the work involved in calling and returning collateral with the benefits of risk mitigation.
Terms of a collateral agreement may be linked to the credit quality of counterparties in order to minimize operational workload while maintaining the ability to tighten collateral terms when a partys credit quality declines. Counterparties most commonly link a tightening of collateral terms to changes in credit rating (e.g., to a downgrade in rating to below investment grade). While this approach is easy to set up, it can lead to issues by requiring the downgraded counterparty to post collateral exactly at a time when it is experiencing credit issues. This can lead to a death spiral of the affected counterparty, as the counterparty faces multiple collateral calls. As a result, it may be preferable to link collateral terms not to the credit rating of entities, but to credit spreads, the market value of equity, or net asset values.
Margin calls should be done at least daily. Products like repos and swaps that are cleared via central counterparties most often have intraday margining. While longer margin frequencies likely reduce operational workloads, daily margining has, more or less, become the market norm.
Threshold in margining refers to the level of exposure below which collateral will not be called. As a result, threshold represents the level of uncollateralized exposure, and only the incremental amount above the threshold would be collateralized. Thresholds generally aim to reduce the operational burden of calling collateral too frequently. A threshold of zero means any exposure is collateralized, while a threshold of infinity means all exposure is uncollateralized. Thresholds are most often linked to credit ratings in a tiered manner, with lower credit ratings corresponding to lower or zero threshold amounts.
Initial margin is the collateral amount that is posted upfront and is independent of any subsequent collateralization. It is often used to mitigate the widening of credit spreads or declines in equity values. Initial margin is typically required by stronger credit quality counterparties or by the counterparty more likely to have positive exposures and represents a level of overcollateralization. Initial margins are also typically linked to the credit rating of counterparties in a tiered manner; however, as opposed to thresholds, the level of intitial margin increases with lower ratings. Intiail margins can be thought of as converting counterparty risk into gap risk, ensuring that the less risky counterparty always remains
Page 164
2018 Kaplan, Inc.
Topic 26 Cross Reference to GARP Assigned Reading – Gregory, Chapter 6
overcollateralized by this amount without incurring losses, even when the risky counterparty defaults. Initial margins should, therefore, be large enough to minimize the gap from large value movements of trades should the risky counterparty default.
A minimum transfer amount represents the smallest amount of collateral that can be transferred. A minimum transfer amount is used to reduce the operational workload of frequent transfers for small amounts of collateral, which must be balanced against the benefits of risk mitigation. It is important to note that the threshold and minimum transfer amount are additive; that is, exposure must exceed the sum of both before a collateral call can be made. Minimum transfer amounts are also typically tied to credit ratings, with higher ratings corresponding to higher amounts.
Collateral amounts typically use rounding (e.g., to the nearest thousand) to avoid transferring very small amounts during collateral calls or returns.
A haircut is essentially a discount to the value of posted collateral. In other words, a haircut of x% means that for every unit of collateral posted, only (1 x)% of credit will be given. This credit is also referred to as valuation percentage. Cash typically has a haircut of 0% and a valuation of 100%, while riskier securities have higher haircut percentages and lower corresponding valuation percentages.
For example, if a particular sovereign bond has a haircut of 2% and a collateral call of $100,000 is made, only 98% of the collaterals value is credited for collateral purposes. That is, in order to satisfy a $100,000 collateral call, $102,041 ($100,000 / 0.98) of the sovereign bond must be posted (or $100,000 in cash).
It is easy to see that riskier securities have greater haircuts to account for their volatility, which may lead to a decline in their value. In the order of increasing riskiness and higher haircuts, cash typically has no haircuts, followed by high-quality government bonds, triple-A rated corporate bonds, structured notes or products, and, finally, equities and commodities. Key factors to consider when assessing haircuts are time to liquidate collateral, volatility of the collaterals underlying market, and the default risk, maturity, and liquidity of the security. Assessing haircuts will often depend on current market conditions using sophisticated value at risk (VaR) calculations.
Entities usually pay interest, coupons, dividends, and other cash flows to counterparties posting collateral as long as the counterparty is not in default. Interest on cash collateral is paid at an overnight market rate. During times of high volatility and illiquid markets, cash collateral is generally preferred, and entities may pay higher-than-market interest rates as an incentive to the entity posting collateral.
We will now look at substitution, reuse, and rehypothecation of collateral. Counterparties sometimes require that the original posted collateral be returned to them for various reasons, including meeting certain delivery commitments. In this case, they can make a substitution request by posting an equivalent value of some other eligible collateral. Substitution requests cannot be refused by the other party if the substituted collateral meets all eligibility criteria. Noncash collateral may also be sold, used in repo transactions, or rehypothecated.
Rehypothecation refers to transferring posted collateral to other counterparties as collateral. While widespread, rehypothecation carries two related risks. Consider a scenario where
2018 Kaplan, Inc.
Page 165
Topic 26 Cross Reference to GARP Assigned Reading – Gregory, Chapter 6
party A pledges collateral to party B; party B rehypothecates this collateral to party C. If party C defaults, then party B will not only have a loss from not receiving the collateral from party C, it will also have a liability to party A for not returning its collateral. The practice of rehypothecation was relatively widespread prior to the 2007-08 credit crisis; however, it has been significantly less popular following the crisis. Parties now increasingly prefer cash collateral.
C S A A g r e e m e n t s

LO 26.5: Explain the process for the reconciliation o f collateral disputes.
To mitigate risk, it is generally preferred to include the maximum number of trades in collateral agreements. However, if even a single trade cannot be properly valued, it can complicate collateral calls and may lead to collateral disputes. If trades include potentially problematic assets, it may be optimal to only focus on a subset of trades that make up the majority of credit exposure and leave out asset classes that are hard to value either due to complexity (e.g., exotic options) or illiquidity (e.g., credit derivatives). Global considerations are also important, especially as counterparties trade with each other over many time zones and geographical locations. It may be optimal to handle trades separately with regions that are problematic and make up only a small portion of trades. Finally, if an entity expects one of its counterparties to have difficulty valuing certain trades or assets, it may be preferred to leave those trades uncollateralized rather than face potential and frequent disputes. Given that collateral agreements typically require that undisputed amounts be transferred immediately, it is generally advantageous to collateralize the majority of products.
If disputes do arise, they can relate to the trade population, trade valuation, netting rules, market data and market closing time, and valuing collateral that was previously posted. If the disputed amount or valuation difference is small, counterparties may simply split the difference. If the disputes involve larger differences, the exposure will remain uncollateralized until the dispute is resolved. Disputes include the following steps: (1) the disputing party notifies the counterparty of its intent to dispute the exposure by the end
2018 Kaplan, Inc.
Page 163
Topic 26 Cross Reference to GARP Assigned Reading – Gregory, Chapter 6
of the day following the collateral call; (2) all undisputed amounts are transferred, and the reason for the dispute is identified; and (3) for unresolved disputes, the parties will request quotes from several market makers (usually four) for the MtM value.
Reconciling trades minimizes the chance of disputes. Parties may also find it beneficial to perform dummy (practice) reconciliations prior to trading and periodic reconciliations during trading (weekly or monthly) to preempt future disputes.
C o l l a t e r a l A g r e e m e n t F e a t u r e s