LGD and Recovery Risk

Question 1

What are the three methods for estimating recovery rates?

Answer 1

1. Market LGD (direct observation of defaulted debt prices in the market)
2. Workout LGD (PV of ex-post realised cash-flows back to the time of default)
3. Implied LGD (inference of recovery rates from non defaulted bond prices or from CDS premia)

Question 2

How is the market LGD calculated?

Answer 2

1. Usually computed as the ratio of post default market price - from 15 to 60 days after default - to face value
2. Suitable measure of loss for investors who sell their debt immediately after default
3. Not necessarily an accurate measure of loss if the debt is worked out by the bank to eventual resolution

Question 3

What is the formula for workout LGD?

Answer 3

RR = (Discounted net value of the recovery)/(Exposure at default) = (FR/EAD) * [(FR - AC)/FR] * (1+r) ^-T

Where
EAD = exposure at default
FR = face value of the recovered amount
AC = Administrative Workout Costs
DNR = Discounted net value of the recovery

Question 4

What are the benefits of Workout LGD?

Answer 4

1. Particularly suited for bank loans risk management and supervision: secondary market of defaulted loans typically illiquid
2. Consistent with Basel II definition of LGD
3. Sensitive to changes in bank-specific workout processes
4. Commonly used in banking practices
5. A suitable discount rate must be selected whicih should be consistent with the ex ante risk of the workout process

Question 5

What is a simple formula for market implied LGD?

Answer 5

(1+i) = (1+r) * (1-p) + (1+r) * p * k =>
k = 1-(r-i)/[p * (1+r)]

Question 6

What are the three groups of drivers affecting recoveries?

Answer 6

1. Debt characteristics:
   * Instrument type
   * Seniority
   * Collateral and covenants
2. Borrower characteristics
   * Leverage (for unsecured exposures)
   * Liability structure
   * Average recover rate for the industry
3. Environmental vaiables (cycle, markets)
   * Current and expected default frequency
   * Cycle indicators
   * Industry-specific conditions at the time of default

Question 7

What are the main data driven results behind recovery factor assumptions?

Answer 7

1. Main determinants of recover rates are security and seniority
2. Bank loans, being at the top of the capital structure, typically have higher recovery than bonds
3. Recoveries are systematically lower in recessions
4. Industry seems to matter
5. Size of exposure seems to have no strong effect on losses for bonds, but seems to negatively affect recoveries in the case of bank loans

Question 8

What is a problem making calculation of LGD more complex for loans as compared to bonds?

Answer 8

The distribution of bank loans LGD tend to be bimodal (Recovery rate clustered more around 0% and 100%) => The mean is a poor indicator for the estimation of RR

Question 9

What is the alternative method to calculating recovery risk for bank loans?

Answer 9

The binomial approach:
μ = EL = PD * LGD + (1-PD) * 0 = PD * LGD
σ = UL = LGD * sqrt(PD * (1-PD))

EL = expected loss
UL = unexpected loss (?)

Question 10

What discount rate should be chosen when calculating LGD?

Answer 10

1. If the lender holds full recourse to cahs collateral: risk-free rate
2. If the collateral is represented by a long term “default free” asset a risk premium may still reamin if the instrument’s value is correlated with the market return
3. Where collateral, unrelated to the firm’s default risk, is taken as security the risk of recovery is related to the secondary market price of the collateral, and hence its market beta, and not necessarily the firm’s asset beta
4. More generally, the discount rate should vary according to the sources of repayment, and multiple discount rates may be applicable