Ifm Formula Sheet A

Download as pdf or txt
Download as pdf or txt
You are on page 1of 12

Exam IFM

Adapt to Your Exam

INTRODUCTION TO DERIVATIVES INTRODUCTION TO DERIVATIVES


Reasons for Using Derivatives Short-Selling Option Moneyness


• To manage risk Process of short-selling: • In-the-money: Produce a positive payoff (not
• To speculate • Borrow an asset from a lender necessarily positive profit) if the option is
• To reduce transaction cost • Immediately sell the borrowed asset and receive exercised immediately.
• To minimize taxes / avoid regulatory issues the proceeds (usually kept by lender or a • At-the-money: The spot price is approximately

designated 3rd party) equal to the strike price.
Bid-ask Spread
• Buy the asset at a later date in the open market • Out-of-the-money: Produce a negative payoff if
• Bid price: The price at which market-makers will to repay the lender (close/cover the short the option is exercised immediately.
buy and end-users will sell. position)

• Ask/Offer price: The price at which market- Option Exercise Styles


makers will sell and end-users will buy. Haircut: Additional collateral placed with lender by • European-style options can only be exercised at
• Bid-ask spread = Ask price – Bid price short-seller. It belongs to the short-seller. expiration.

• Round-trip transaction cost: Difference between Interest rate on haircut is called: • American-style options can be exercised at any
what you pay and what you receive from a sale time during the life of the option.
• short rebate in the stock market
using the same set of bid/ask prices. • Bermudan-style options can be exercised during
• repo rate in the bond market

bounded periods (i.e., specified periods during the
Payoff and Profit Reasons for short-selling assets: life of the option).
• Payoff: Amount that one party would have if • Speculation – To speculate that the price of a

completely cashed out. Zero-coupon Bond (ZCB)


particular asset will decline.
• Profit: Accumulated value of cash flows at the Buying a risk-free ZCB = Lending at risk-free rate
• Financing – To borrow money for additional
risk-free rate. Selling a risk-free ZCB = Borrowing at risk-free rate
financing of a corporation.

Payoff on a risk-free ZCB = ZCB’s maturity value
Long vs. Short • Hedging – To hedge the risk of a long position on Profit on a risk-free ZCB = 0
• A long position in an asset benefits from an the asset.
increase in the price of the asset.
• A short position in an asset benefits from a
decrease in the price of the asset.

FORWARD CONTRACTS, CALL OPTIONS, AND PUT OPTIONS


FORWARD CONTRACTS, CALL OPTIONS, AND PUT OPTIONS
Position Position in Maximum Maximum
Contract Description Payoff Profit Strategy
in Contract Underlying Loss Gain
Obligation to buy Guarantee/lock in
Long
at the forward Long S" − F%," S" − F%," F%," ∞ purchase price of
Forward

Forward
price underlying
Guarantee/lock in
Short Obligation to sell at
Short F%," − S" F%," − S" ∞ F%," sale price of
Forward the forward price
underlying
Right (but not Insurance against
max [0, S" − K]
Long Call obligation) to buy Long max [0, S" − K] AV(Prem. ) ∞ high underlying
− AV(Prem. )
at the strike price price
Call

Obligation to sell at Sells insurance


the strike price if −max [0, S" − K] against
Short Call Short −max [0, S" − K] ∞ AV(Prem. )
the call is + AV(Prem. ) high underlying
exercised price
Right (but not Insurance against
max [0, K − S" ] K
Long Put obligation) to sell Short max [0, K − S" ] AV(Prem. ) low underlying
− AV(Prem. ) −AV(Prem. )
at the strike price price
Put

Obligation to buy Sells insurance


at the strike price − max[0, K − S" ] K against
Short Put Long −max [0, K − S" ] AV(Prem. )
if the put is + AV(Prem. ) −AV(Prem. ) low underlying
exercised price

www.coachingactuaries.com Copyright © 2018 Coaching Actuaries. All Rights Reserved. 1


OPTION STRATEGIES OPTIONS COMBINATION

Naked Writing vs. Covered Writing Graphing Payoffs from First Principles
• If an option writer does not have an offsetting position in the underlying Create a payoff table separated by the strike regions, and then graph the total
asset, then the option position is said to be naked. payoff in each region accordingly.
• If an option writer has an offsetting position in the underlying asset, then
the option position is said to be covered. Shortcut Method for Graphing the Payoff of All Calls or All Puts
• For calls, go left-to-right on the payoff diagram, and evaluate the slope of
Floor, Cap, Covered Call, Covered Put the payoff diagram at each strike price.
Key Relationship: Call – Put = Stock – Bond Going left-to-right means that a positive slope is one that increases left-
Rearranging, we have: to-right, and a negative slope is one that decreases left-to-right.
• Floor = + Stock + Put (guarantee a minimum selling price for stock) • For puts, go right-to-left on the payoff diagram, and evaluate the slope of
• Write a covered put = – Floor = – Stock – Put the payoff diagram at each strike price.
• Cap = – Stock + Call (guarantee a maximum purchase price for stock) Going right-to-left means that a positive slope is one that increases right-
• Write a covered call = – Cap = + Stock – Call to-left, and a negative slope is one that decreases right-to-left.

Bull Spread Bear Spread (opposite of bull spread) Box Spread


Call Bull: Long call (K1) + Short call (K2), K1 < K2 Call Bear: Short call (K1) + Long call (K2), K1 < K2 Long call (put) bull spread + Long put (call) bear
Put Bull: Long put (K1) + Short put (K2), K1 < K2 Put Bear: Short put (K1) + Long put (K2), K1 < K2 spread



Ratio Spread Collar Collared Stock
Long and short an unequal number of calls/puts Long put (K1) + Short call (K2), K1 < K2 Long collar + Long stock
with different strike prices



Straddle Strangle Butterfly Spread (Symmetric/Asymmetric)
Long put (K) + Long call (K) Long put (K1) + Long call (K2), K1 < K2 Buy high- and low-strike options. Sell middle-

strike option. Quantity sold = Quantity bought.












Example - For 3 strike prices 30, 43, 46:
• Buy 46 − 43 = 3 options with strike 30
• Buy 43 − 30 = 13 options with strike 46
• Sell 46 − 30 = 16 options with strike 43
Any multiple of (buy 3, sell 16, buy 13) would
work.

www.coachingactuaries.com Copyright © 2018 Coaching Actuaries. All Rights Reserved. 2


FORWARDS FORWARDS Features of Futures Contract Early Exercise of American Option

Notional Value = # Contracts × Multipler American Call
4 Ways to Buy a Share of Stock
× Futures price • Nondividend-paying stock
Receive o Early exercise is never optimal.
Payment Balt = BaltN_ ⋅ e>` + GainE
Ways Stock Payment o C{|}> = C~>
Time where
at Time • Dividend-paying stock
Outright • GainE = # Contracts × Multipler × o It is rational to early exercise if:
0 0 S%
purchase Price Changet (for 𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙 position) PV(Divs) >
Fully • GainE = − # Contracts × Multipler × PV(Interest on strike) + Implicit Put
leveraged T 0 S% e>" Price Changet (for 𝑠𝑠ℎ𝑜𝑜𝑜𝑜𝑜𝑜 position) o It may be rational to early exercise if:
purchase PV(Divs) > PV(Interest on strike)
• Price Changet = Future PriceE − Future PriceEN_
Prepaid
American Put
? (S)
forward 0 T F%," Margin Call It is rational to early exercise if:
contract • Maintenance margin: Minimum margin balance PV(Interest on strike) >
Forward T T F%," (S) that the investor is required to maintain in PV(Divs) + Implicit Call
contract It may be rational to early exercise if:
margin account at all times
𝐏𝐏 (𝐒𝐒)
Relationship between 𝐅𝐅𝐭𝐭,𝐓𝐓(𝐒𝐒) and 𝐅𝐅𝐭𝐭,𝐓𝐓 • Margin call: If the margin balance falls below PV(Interest on strike) > PV(Divs)
? (S)
FE," (S) = Accumulated Value of FE," the maintenance margin, then the investor will Strike Price Effects
? (S)
= FE," ⋅ e>("NE) get a request for an additional margin deposit. For K_ < K Ü < K á :

Dividend Structure 𝐏𝐏
𝐅𝐅𝐭𝐭,𝐓𝐓 (𝐒𝐒) The investor has to add more funds to bring the Call
margin balance back to the initial margin. C(K_ ) ≥ C(K Ü ) ≥ C(K á)
No Divs SE
C(K_ ) − C(K Ü ) ≤ K Ü − K_
Discrete Divs SE − PVE," (Divs) European: C(K_) − C(K Ü) ≤ PV(K Ü − K_ )
PUT-CALL PARITY (PCP)
PUT-CALL PARITY (PCP) C(K_ ) − C(K Ü ) C(K Ü ) − C(K á )
Continuous Divs SE eNS("NE) ≥
PCP for Stocks K Ü − K_ Ká − KÜ

Dividend Structure 𝐅𝐅𝐭𝐭,𝐓𝐓(𝐒𝐒) ? (S)
C(S, K) − P(S, K) = FE," − KeN>("NE) Put
No Divs SE e>("NE) P(K_ ) ≤ P(K Ü ) ≤ P(K á )
PCP for Futures P(K Ü ) − P(K_ ) ≤ K Ü − K_
Discrete Divs SE e>("NE) − AVE," (Divs) C(F, K) − P(F, K) = FeN>("NE) − KeN>("NE)

European: P(K Ü) − P(K_ ) ≤ PV(K Ü − K_ )
Continuous Divs SE e(>NS)("NE) P(K Ü ) − P(K_ ) P(K á) − P(K Ü)
PCP for Bonds ≤
? (B)
C(B, K) − P(B, K) = FE," − KeN>("NE) K Ü − K_ Ká − KÜ
F%,"

Forward premium = where Time Until Expiration


S% ? (B)
FE," = BE − PVE," (Coupons) For T_ < TÜ:
1 F%,"
Annualized forward premium rate = ln BE = Bond price at time t C{|}> (S, K, T_ ) ≤ C{|}> (S, K, TÜ)
T S%
P{|}> (S, K, T_) ≤ P{|}> (S, K, TÜ)

PCP for Exchange Options

Synthetic Forward For a nondividend-paying stock:


𝐂𝐂(𝐀𝐀, 𝐁𝐁) 𝐏𝐏(𝐀𝐀, 𝐁𝐁)
Synthetic long forward is created by: C~> (S, K, T_ ) ≤ C~> (S, K, TÜ)
receive A, give up B give up A, receive B
• buying a stock and borrowing money (i.e., ? (A) ? (B) This is also generally true for European call
selling a bond), or C(A, B) − P(A, B) = FE," − FE,"
options on dividend-paying stocks and European
C(A, B) = P(B, A)
• buying a call and selling a put at the same strike. puts, with some exceptions.

PCP for Currency Options
Synthetic short forward is the opposite, created by:
Use the generalized PCP for exchange option.
• selling a stock and lending money (i.e., buying a
BINOMIAL MODEL
BINOMIAL MODEL
bond), or For example, the prepaid forward price for 1 yen
denominated in dollars is: Option Pricing: Replicating Portfolio
• selling a call and buying a put at the same strike.
An option can be replicated by buying Δ shares
$
? (¥1)
= qx % r (¥1eN>¥ " ) = $x %eN>¥ " of the underlying stock and lending 𝐵𝐵 at the
Arbitrage $F%,"
¥ risk-free rate.
A transaction which generates a positive cash flow
V − Vt uVt − dV
either today or in the future by simultaneous Alternatively: Δ = eNS` ã å B = eN>` ã å
S(u − d) u−d
buying and selling of related assets, with no net S% → x % r → rt δ → rv
V = ΔS + B
investment or risk. Ct(f, K) − Pt (f, K) = x %eN>w" − KeN>x "

Arbitrage strategy: “Buy Low, Sell High.” where x % is in d/f Call Put
Δ + −
Cash-and-Carry
B − +
The actual forward is overpriced.
COMPARING OPTIONS
COMPARING OPTIONS
Short actual forward + Long synthetic forward To replicate a call, buy shares and borrow money.

Reverse Cash-and-Carry Bounds of Option Prices To replicate a put, sell shares and lend money.
The actual forward is underpriced. Call and Put

S ≥ C{|}> ≥ C~> ≥ max(0, F ? (S) − KeN>" ) Option Pricing: Risk-neutral Valuation


Long actual forward + Short synthetic forward
K ≥ P{|}> ≥ P~> ≥ maxÄ0, KeN>" − F ? (S)Å e(>NS)` − d
p∗ =

u−d
European vs. American Call V% = eN>` ⋅ E∗[Payoff] = eN>`[(p∗ )V + (1 − p∗)Vt ]
FUTURES FUTURES F ? (S) ≥ C~> ≥ max(0, F ? (S) − KeN>" )
S% e(>NS)` = (p∗)S + (1 − p∗)St
Futures Compared to Forward S ≥ C{|}> ≥ max (0, S − K)

• Traded on an exchange European vs. American Put
• Standardized (size, expiration, underlying) KeN>" ≥ P~> ≥ maxÄ0, KeN>" − F ? (S)Å
• More liquid K ≥ P{|}> ≥ max (0, K − S)
• Marked-to-market and often settled daily
• Minimal credit risk
• Price limit is applicable

www.coachingactuaries.com Copyright © 2018 Coaching Actuaries. All Rights Reserved. 3


Constructing a Binomial Tree To find the pth percentile of S" : 5. Annualize the estimate of the expected return:
General Method 1. Determine the corresponding pth percentile of SEê` 1 Ü
S St E ®ln © = ãα
¬−δ− σ ¬ å h = r̅
the standard normal random variable Z. SE 2
u = d =
S% S% 2. Substitute the resulting value of Z into the r̅ 1 Ü

expression for S" . ⇒α ¬ = +δ+ σ ¬
h 2
Standard Binomial Tree Median = 50th percentile
This is the usual method in McDonald based on § §
forward prices. = SE eıNSN%.¢ë Å("NE) = E[S" ] ⋅ eN%.¢ë ("NE)
THE BLACK-SCHOLES (BS) FORMULA
THE BLACK-SCHOLES (BS) FORMULA
u = e(>NS)`êë√` d = e (>NS)`Në√` Prediction Interval
1 BS Formula’s Assumptions:
p∗ = The 100%(1 − p) prediction interval is given by S"¥
1 + eë√` and S"µ such that Pr[S"¥ < S" < S"µ ] = 1 − p. • Continuously compounded returns on the stock
p p are normally distributed and independent over
Probability Pr[Z < z ¥ ] = ⇒ z¥ = N N_ ì î time. There are no sudden jumps in the stock
2 2
For n periods, let k be the number of "up" jumps p price.
needed to reach an ending node. Then, the risk- zµ = −z¥ = −NN_ ì î
2 • Volatility is known and constant.
neutral probability of reaching that node is given § π
S"¥ = SE eıNSN%.¢ë Å("NE)êë√"NE⋅∏ • Future dividends are known.
by: µ ıNSN%.¢ë§ Å("NE)êë√"NE⋅∏∫ • The risk-free rate is known and constant.
n S" = SE e
ì î (p∗)ï (1 − p∗)ñNï , k = 0,1, … , n • There are no taxes or transaction costs.

k Probability • Short-selling is allowed at no cost.
No-Arbitrage Condition Pr[S" < K] = NÄ−dªÜÅ Pr[S" > K] = NÄ+dªÜÅ • Investors can borrow and lend at the risk-free
Arbitrage is possible if the following inequality is S rate.
ln ì E î + (α − δ − 0.5σÜ)(T − t)
not satisfied: dªÜ = K

σ√T − t Generalized BS Formula


0 < p∗ < 1 ⟺ d < e(>NS)` < u C = F ? (S) ⋅ N(d_ ) − F ? (K) ⋅ N(dÜ )
Option on Currencies Conditional and Partial Expectation P = F ? (K) ⋅ N(−dÜ ) − F ? (S) ⋅ N(−d_ )
PE[S" |S" < K] F ? (S) 1
Substitutions: S% → x % r → rt δ → rv E[S" |S" < K] = ln ã ? (K)å + σÜ (T − t)
Pr[S" < K] F 2
u = e(>x N>w)`êë√` d = e(>x N>w)`Në√` d_ =
e(>x N>w)` − d SE e (±NS)("NE)
NÄ−dª_ Å σ√T − t
p∗ = =
u−d NÄ−dªÜ Å F ? (S) 1
ln ã ? (K)å − σÜ (T − t)
PE[S" |S" > K] F 2
Option on Futures Contracts E[S" |S" > K] = dÜ = = d_ − σ√T − t
Pr[S" > K] σ√T − t
FE,"õ = SE e(>NS)("õNE)
SE e (±NS)("NE)
NÄ+dª_ Å
T = Expiration date of the option = Var{ln[SE ]}
Tú = Expiration date of the futures contract NÄ+dªÜ Å σ=√ , 0 < t ≤ T
t
T ≤ Tú S
ln ì E î + (α − δ + 0.5σÜ )(T − t)

dª_ = K Var∆ln•FE," (S)¶«
Substitutions: SE → FE,"õ δ → r σ√T − t =√ , 0 < t ≤ T
ë√` Në√` t
uú = e dú = e Expected Option Payoffs

1 − dú E[Call Payoff] = SEe(±NS)("NE) NÄdª_ Å − KNÄdªÜ Å
?
Var∆ln•FE," (S)¶«
p = =√ , 0 < t ≤ T
uú − dú E[Put Payoff] = KNÄ−dªÜÅ − SE e(±NS)("NE) NÄ−dª_Å t
V − Vt
Δ=

F(uú − dú ) True Pricing For a stock that pays continuous dividends:


B = eN>`[p∗ V + (1 − p∗)Vt ] To calculate option price, discount the true C = SE eNS("NE) ⋅ N(d_ ) − KeN>("NE) ⋅ N(dÜ )
expected option payoff at the expected rate of P = KeN>("NE) ⋅ N(−dÜ ) − SE eNS("NE) ⋅ N(−d_ )
return on the option: S 1
ln ì E î + ìr − δ + σÜ î (T − t)
LOGNORMAL MODEL V% = eNº" E[Payoff] K 2
LOGNORMAL MODEL d_ =
σ√T − t
Normal vs. Lognormal Risk-Neutral Pricing
S 1
X~N(m, v Ü ) ⟺ Y = e† ~LogN(m, v Ü ) Assume α = γ = r. ln ì E î + ìr − δ − σÜ î (T − t)
dÜ = K 2
§ To calculate option price, discount the risk-neutral
• E[Y] = e|ê%.¢£ σ√T − t
§ expected option payoff at the risk-free rate:
• Var[Y] = (E[Y])Ü•e£ − 1¶ V% = eN>" E ∗[Payoff] = d_ − σ√T − t
X = m + v ⋅ Z, Z~N(0,1)

N(−a) = 1 − N(a) Estimating Return and Volatility Options on Futures Contract


Two important properties of lognormal: Given S% , S_ , . . . , Sñ , where the observations are at Use the generalized BS formula in conjunction
• It cannot be negative. intervals of length h, we can estimate the with the appropriate prepaid forward prices.
• The product of two lognormal is a lognormal. lognormal parameters as follows:

1. Calculate the continuously compounded The prepaid forward price for a futures contract is
Lognormal Model for Stock Prices returns: just the present value of the futures price:
S" Sæ

? (F)
For T > t, ln ® © ~N[m, v Ü] ræ = ln i = 1,2, … , n F%," = FeN>"
SE SæN_
• m = (α − δ − 0.5σÜ )(T − t) 2. Calculate the sample mean of the returns: Options on Currencies
• v Ü = σÜ(T − t) ∑ñæ¡_ ræ Use the generalized BS formula in conjunction
≠ r̅ = with the appropriate prepaid forward prices.
• Æ ~LogN(m, v Ü ) n
≠Ø
3. Estimate the standard deviation of returns by For example, the prepaid forward price for 1 yen
Ü ]
For T > t, ln[S" ] ~N[m, v taking the square root of the sample variance of denominated in dollars is:
• m = ln SE + (α − δ − 0.5σÜ )(T − t) the returns:

• v Ü = σÜ(T − t) $
∑ñæ¡_(ræ
− r̅)Ü ? (¥1)
$F%," = qx % r (¥1eN>¥ " ) = $x %eN>¥ "
• S" ~LogN(m, v Ü) ¬` = √
σ ¥
n−1
E[S" ] = E[S" |SE ] = SE e(±NS)("NE) 4. Annualize the estimate of the standard
§
Var[S" ] = Var[S" |SE ] = (E[S" ])Ü Äe£ − 1Å deviation:
§
S" = SE eıNSN%.¢ë Å("NE)êë√"NE⋅≤ , Z~N(0,1) SEê`
S" Var ®ln ©=σ¬Ü h = σ
¬`
SE
Cov(SE , S" ) = E ® © ⋅ Var[SE |S% ]
SE ¬`
σ
⇒σ ¬=
√h

www.coachingactuaries.com Copyright © 2018 Coaching Actuaries. All Rights Reserved. 4


OPTION GREEKS GREEKS Portfolio Greek & Elasticity ACTUARIAL-SPECIFIC RISK MANAGEMENT
ACTUARIAL-SPECIFIC RISK MANAGEMENT

Greek for a portfolio = sum of the Greeks
Greek Definition Long Call Long Put Options Embedded in Insurance Products
Elasticity for a portfolio = weighted average of the
• A guaranteed minimum death benefit (GMDB)
∂V elasticities
Δ + − ñ guarantees a minimum amount will be paid to a
∂S Δ?‘>E ⋅ S beneficiary when the policyholder dies.
Ω?‘>E = = ⁄ ωæ Ωæ
∂Δ ∂Ü V V?‘>E • A guaranteed minimum accumulation benefit
Γ + + æ¡_
= γ?‘>E − r = Ω?‘>E (α − r) (GMAB) guarantees a minimum value for the
∂S ∂SÜ
underlying account after some period of time,
∂V Delta-Gamma-Theta Approximation even if the account value is less.
θ − * −*
∂t 1 • A guaranteed minimum withdrawal benefit
VEê` ≈ VE + ΔE ϵ + ΓE ϵÜ + θE h; ϵ = SEê` − SE
∂V 2 (GMWB) guarantees that upon the policyholder
Vega + + reaching a certain age, a minimum withdrawal
∂σ
amount over a specified period will be provided.
∂V
ρ + − DELTA & GAMMA HEDGING
DELTA & GAMMA HEDGING • A guaranteed minimum income benefit (GMIB)
∂r guarantees the purchase price of a traditional
∂V Overnight Profit
ψ − + annuity at a future time.
A delta-hedged portfolio has 3 components:
∂δ
• Buy/sell options GMDB with a Return of Premium Guarantee
* θ is usually negative. • A guarantee which returns the greater of the
• Buy/sell stocks
Note: For short positions, just reverse the signs. account value and the original amount invested:
• Borrow/lend money (sell/buy bond)
Option Greeks Formulas max(S" , K) = S" + max( K − S" , 0)
The formulas for the six option Greeks for both call Overnight profit is the sum of: • The embedded option is a put option. Its value
and put options under the BS framework as well as • Profit on options bought/sold is:
·
NÕ (x) will be provided on the exam. • Profit on stocks bought/sold
E[P(Tœ )] = fl P(t)f"‡ (t)dt
• Profit on bond %
ΔŒ = eNS("NE)N(d_ ), 0 ≤ ΔŒ ≤ 1

Δ? = −eNS("NE)N(−d_ ), −1 ≤ Δ? ≤ 0 Alternatively, overnight profit is the sum of: Earnings-Enhanced Death Benefit
ΔŒ − Δ? = eNS("NE) • Gain on options, ignoring interest • Pays the beneficiary an amount based on the
expÄ−δ(T − t)ÅNÕ (d_ ) • Gain on stocks, ignoring interest increase in the account value over the original
ΓŒ = Γ? = amount invested, e.g., 40% ⋅ max( S" − K, 0).
Sσ√T − t • Interest on borrowed/lent money
• The embedded option is a call option. Its value
θŒ = δSeNS("NE) N(d_ ) − rKeN>("NE)N(dÜ)

For a market-maker who writes an option and is:


KeN>("NE) NÕ(dÜ )σ ·
− delta-hedges the position, the market-maker’s E[C(Tœ )] = fl C(t)f"‡ (t)dt
2√T − t
profit from time t to t + h is: %
θ? = θŒ + rKeN>("NE) − δSeNS("NE)

VegaŒ = Vega? = SeNS("NE) NÕ(d_ )√T − t = ΔE (SEê` − SE ) − (VEê` − VE ) GMAB with a Return of Premium Guarantee
ρŒ = (T − t)KeN>("NE) N(dÜ ) • Similar to GMDB with ROP guarantee, but the
− Äe>` − 1Å(ΔE SE − VE )
ρ? = −(T − t)KeN>("NE) N(−dÜ ) benefit is contingent on the policyholder
1 surviving to the end of the guarantee period.
ψŒ = −(T − t)SeNS("NE)N(d_ ) ≈ − ϵÜ ΓE − hθE − Äe>` − 1Å(ΔE SE − VE )
2 • The embedded option is a put option. Its value
ψ? = (T − t)SeNS("NE)N(−d_ ) where ϵ = SEê` − SE
§ is:
eNœ /Ü
NÕ (x) = P(m) ⋅ Pr[T†∗ ≥ m]
√2π If h is small, then e>` − 1 ≈ rh.



Mortgage Loan as Put
Risk Premium For an uninsured position, the loss to the mortgage
Breakeven
The risk premium of an asset is defined as the lender is max(B + C ∗ − R, 0), where:
If the price of the underlying stock changes by one
excess of the expected return of the asset over the • B is the outstanding loan balance at default
standard deviation over a short period of time,
risk-free return: • C ∗is the lender’s total settlement cost
then a delta-hedged portfolio does not produce
• Risk Premium“”Eæ‘ñ = γ − r • R is the amount recovered on the sale of
profits or losses.
• Risk Premium≠E‘’ï = α − r
property

Assuming the BS framework, given the current This is a put payoff with K = B + C ∗ and S = R.
γ − r = Ω(α − r) stock price, S, the two stock prices after a period of
σ“”Eæ‘ñ = |Ω|σ≠E‘’ï h for which the market-maker would break even Static vs. Dynamic Hedging

Sharpe Ratio are: Static/hedge-and-forget: Buy options and hold to


OptionÕ s risk premium γ−r S ± Sσ√h expiration
ϕ“”Eæ‘ñ = = Dynamic: Frequently buy/sell assets and/or
OptionÕ s volatility σ“”Eæ‘ñ Multiple Greeks Hedging derivatives with the goal of matching changes in
Õ
Stock s risk premium α − r Δ≠E‘’ï = 1; all other Greeks of the stock = 0 the value of guarantee
ϕ≠E‘’ï = =
StockÕ s volatility σ≠E‘’ï To hedge multiple Greeks, set the sum of the
ϕŒ = ϕ≠E‘’ï ; ϕ? = −ϕ≠E‘’ï Greeks you are hedging to zero.

Elasticity
% change in option price Δ ⋅ S
Ω= =
% change in stock price V
ΩŒ ≥ 1; Ω? ≤ 0

www.coachingactuaries.com Copyright © 2018 Coaching Actuaries. All Rights Reserved. 5


EXOTIC OPTIONS
EXOTIC OPTIONS Gap Option Lookback Option
K_ : Strike Price An option whose payoff at expiration depends on
Asian Option K Ü : Trigger Price the maximum or minimum of the stock price over
A(S) arithmetic average
S‚ = „ K_ determines the amount of the nonzero payoff. the life of the option.
G(S) geometric average
_ K Ü determines whether the option will have a
‰ ‰
∑‰
E¡_ SE nonzero payoff. Type Payoff
A(S) = G(S) = ÂÊ SE Á
N 0, S" ≤ K Ü Standard lookback call S" − min(S)
E¡_ PayoffÎË” ŒËÈÈ = „
G(S) ≤ A(S) S" − K_ , S" > K Ü Standard lookback put max(S) − S"
K − S" , S" < K Ü Extrema lookback call max[0, max(S) − K]
PayoffÎË” ?E = „ _
Average Price Average Strike 0, S" ≥ K Ü Extrema lookback put max[0, K − min(S)]
PayoffŒËÈÈ max[0, S‚ − K] max[0, S − S‚] Negative payoffs are possible.

Payoff?E max[0, K − S‚] max[0, S‚ − S] GapCall = S% eNS" N(d_ ) − K_ eN>" N(dÜ ) Standard lookback options are known as lookback

GapPut = K_ eN>" N(−dÜ ) − S% eNS" N(−d_ ) options with a floating strike price.
The value of an average price Asian option is less where d_ and dÜ are based on K Ü

than or equal to the value of an otherwise GapCall − GapPut = S% eNS" − K_ eN>" Extrema lookback options are known as lookback
equivalent ordinary option. options with a fixed strike price.

Exchange Option
As N increases:
C(A, B) = F ? (A) ⋅ N(d_ ) − F ? (B) ⋅ N(dÜ ) Shout Option
• Value of average price option decreases
P(A, B) = F ? (B) ⋅ N(−dÜ ) − F ? (A) ⋅ N(−d_ ) An option that gives the owner the right to lock in
• Value of average strike option increases
F ? (A) 1 a minimum payoff exactly once during the life of
ln ã ? (B) å + σÜ (T − t)
Barrier Option F 2
d_ = the option, at a time that the owner chooses. When
Three types: σ√T − t
the owner exercises the right to lock in a minimum
• Knock-in dÜ = d_ − σ√T − t
Goes into existence if barrier is reached. payoff, the owner is said to shout to the writer.
A
• Knock-out
√Var Ïln BÌ S ∗ is the value of the stock at the time when the
Goes out of existence if barrier is reached. σ=
t option owner shouts to the option writer.
• Rebate

Pays fixed amount if barrier is reached. = ÓσÜ{ + σÜÔ − 2Cov{,Ô
Payoff for a shout call
Down vs. Up: max[S" − K, S∗ − K, 0] if shout is exercised
• If S% < barrier: = ÓσÜ{ + σÜÔ − 2ρσ{σÔ =„
max[S" − K, 0] if shout is not exercised
Up-and-in, up-and-out, up rebate

• If S% > barrier: Payoff for a shout put


Maxima and Minima
Down-and-in, down-and-out, down rebate max[K − S" , K − S ∗, 0] if shout is exercised

• max(A, B) = max(0, B − A) + A =„
max(A, B) = max(A − B, 0) + B max[K − S" , 0] if shout is not exercised
Knock-in + Knock-out = Ordinary Option

Barrier option ≤ Ordinary Option • max(cA, cB) = c ⋅ max(A, B) c > 0 Rainbow Option

max(cA, cB) = c ⋅ min(A, B) c < 0 An option whose payoff depends on two or more
Special relationships: • max(A, B) + min(A, B) = A + B
• If S% ≤ barrier ≤ strike: risky assets.
⇒ min(A, B) = − max(A, B) + A + B
Up-and-in Call = Ordinary Call
Up-and-out Call = 0 Forward Start Option
• If S% ≥ barrier ≥ strike: For a call option expiring at time T whose strike is MEAN-VARIANCE PORTFOLIO THEORY
MEAN-VARIANCE PORTFOLIO THEORY

Down-and-in Put = Ordinary Put set on future date t to be XSE :
Risk and Return of a Single Asset
Down-and-out Put = 0 C(SE , XSE , T − t) ñ

Compound Option = SE eNS("NE)N(d_ ) − XSE eN>("NE)N(dÜ ) E[R] = ⁄ pæ ⋅ Ræ


A compound call allows the owner to buy another = SE •eNS("NE) N(d_ ) − XeN>("NE)N(dÜ )¶ ¡_
ñ
option at the strike price. S
ln ì E î + (r − δ + 0.5σÜ)(T − t) Var[R] = E[(Ræ − E[R])Ü ] = ⁄ pæ ⋅ (Ræ − E[R])Ü
XSE
A compound put allows the owner to sell another d_ =
option at the strike price. σ√T − t ¡_
1 = E[RÜ ] − (E[R])Ü
ln ì î + (r − δ + 0.5σÜ)(T − t)
The value of the underlying option at time t_ = X SD[R] = ÒVar[R]
= VÄSEÍ , K, T − t_ Å σ√T − t
REê_ = Capital Gain + Div Yield
The value of the compound call at time t_ dÜ = d_ − σ√T − t
PEê_ − PE DEê_
= max•0, VÄSEÍ , K, T − t_ Å − x¶ = +
The time-0 value of the forward start option is: PE PE
The value of the compound put at time t_ ? (S)
V% = F%,E × •eNS("NE)N(d_ ) − XeN>("NE)N(dÜ )¶ R = Ä1 + RÚ_ ÅÄ1 + RÚÜ ÅÄ1 + RÚá ÅÄ1 + RÚÛ Å − 1
= max•0, x − VÄSEÍ , K, T − t_ Ŷ
" "
where Same analysis applies to a put option. 1 1
Ù=
R ⁄ RE Var[R] = Ù)Ü
⁄(RE − R
• K is the strike of the underlying option
T T−1
• x is the strike of the compound option Chooser Option E¡_ E¡_
For an option that allows the owner to choose at SD(Individual Risk)
• T is the maturity of the underlying option Standard Error =
• t_ is the maturity of the compound option time t whether the option will become a European √# Observations
call or put with strike K and expiring at time T: 95% confidence interval for the expected return
Put-call parity for compound options:

VE = max[C(SE , K, T − t), P(SE , K, T − t)] = Historical average return ± 2 ⋅ Standard Error


• CallonCall − PutonCall = C~> − xeN>EÍ
• CallonPut − PutonPut = P~> − xeN>EÍ = eNS("NE) max•0, KeN(>NS)("NE) − SE ¶
+ C(SE , K, T − t)
The time-0 value is:
V% = eNS("NE) ⋅ PÄS% , KeN(>NS)("NE), tÅ + C(S% , K, T)

www.coachingactuaries.com Copyright © 2018 Coaching Actuaries. All Rights Reserved. 6


Risk and Return of a Portfolio Portfolio The Effect of Correlation
R? = x_ R_ + x ÜRÜ + ⋯ + x ñRñ Volatility
• If ρæ,¯ = 1, no diversification. The portfolio's
E[R? ] = x_ E[R_ ] + x ÜE[RÜ ] + ⋯ + x ñ E[Rñ ] volatility is simply the weighted average
Value of Investment i volatility of the two risky assets.
xæ = Elimination of
Total Portfolio Value Non-Systematic • If ρæ,¯ < 1, the portfolio's volatility is reduced
Cov•Ræ , R¯ ¶ = E•(Ræ − E[Ræ ])ÄR¯ − E•R¯ ¶Å¶ Risk
due to diversification. It is less than the
"
1 weighted average volatility of the two risky
= Ù æ ÅÄR¯,E − R
⁄ÄRæ,E − R Ù ¯ Å Systematic
T−1 assets.
æ¡_ Risk
= ρæ,¯ ⋅ σæ ⋅ σ¯ • If ρæ,¯ = −1, a zero-risk portfolio can be
Number of Stocks constructed.
For a 2-stock portfolio:
σÜ? = x_Ü σ_Ü + x ÜÜ σÜÜ + 2x_ x ÜCov[R_ , RÜ ] Observations: 25%
For a 3-stock portfolio: • The diversification effect is most significant BA

Expected Return
σÜ? = x_Ü σ_Ü + x ÜÜ σÜÜ + x Üá σÜá + 2x_x Ü Cov[R_ , RÜ ] initially. 20%
Corr. = -1
+ 2x_x á Cov[R_ , Rá ] • Even with a very large portfolio, we cannot
No Risk
+ 2x Üx á Cov[RÜ , Rá ] eliminate all risk. The remaining risk is 15%

For an n-stock portfolio: systematic risk that cannot be avoided through Corr. = +1
ñ ñ ñ
diversification. 10%
σÜ? = ⁄ x æ Cov[Ræ, R? ] = ⁄ ⁄ x æ x ¯Cov•Ræ , R¯ ¶

æ¡_ æ¡_ ¯¡_


For a portfolio with n individual stocks with WMT
5%
In the covariance matrix, we have: arbitrary weights: 0% 5% 10% 15% 20% 25% 30%
ñ
• n × n = nÜ total elements Volatility
σ? = ⁄ x æ ⋅ σ? ⋅ ρæ,?

• n variance terms Combining Risky Assets with Risk-free Assets


æ¡_
• nÜ − n true covariance terms • Each security contributes to the portfolio Suppose we invest a proportion of funds (x) in a
• (nÜ − n)/2 unique true covariance terms volatility according to its total risk scaled by its risky portfolio and the remainder (1 − x) in a risk-
correlation with the portfolio, which adjusts for free asset. Then, we have:
Diversification the fraction of the total risk that is common to E[Rœ? ] = xE[R? ] + (1 − x)rv = rv + x(E[R? ] − rv )
Systematic risk the portfolio. σœ? = x ⋅ σ?

• Also known as common, market, or • As long as the correlation is not +1, the volatility Capital Allocation Line (CAL) is a line representing
nondiversifiable risk. of the portfolio is always less than the weighted possible combinations from combining a risky
• Fluctuations in a stock's return that are due to average volatility of the individual stocks. portfolio and a risk-free asset:
market-wide news. E[R? ] − rv
Nonsystematic risk Mean-Variance Portfolio Theory E[Rœ? ] = rv + q r σœ?
σ?
• Also known as firm-specific, independent, Assumptions

idiosyncratic, unique, or diversifiable risk. • All investors are risk-averse.


• Fluctuations in a stock's return that are due to • The expected returns, variances, and
firm-specific news. covariances of all assets are known.

Total risk = Systematic risk + Unsystematic risk • To determine optimal portfolios, investors only Risky
need to know the expected returns, variances, Portfolio
Diversification reduces a portfolio's total risk by and covariances of returns. Portfolio with x
averaging out nonsystematic fluctuations: • There are no transactions costs or taxes. invested in the
• Investors can eliminate nonsystematic risk for risky portfolio
free by diversifying their portfolios. Thus, the Efficient Frontier
risk premium for nonsystematic risk is zero. • A portfolio is efficient if the portfolio offers the
highest level of expected return for a given level
• The risk premium of a security is determined by

its systematic risk and does not depend on its of volatility (or the lowest level of volatility for a • At the intercept, the portfolio only consists of
nonsystematic risk. given level of return). the risk-free asset.
• The portfolios that have the greatest expected • At point P, the portfolio only consists of risky
For an equally-weighted n-stock portfolio: return for each level of volatility (or the lowest assets.
1 1 volatility for a given level of return) make up • The line extending to the right of σ? represents
σÜ? = ⋅ ‚‚‚‚‚
Var + ã1 − å ⋅ Cov‚‚‚‚‚ the efficient frontier.
n n portfolios that invest more than 100% in the

• In a very large portfolio (n → ∞), the covariance risky portfolio P. This is done by using leverage
among the stocks accounts for the bulk of (borrow money to invest).
portfolio risk:

‚‚‚‚‚ Efficient/tangent/optimal portfolio:


σÜ? = Cov
• Is the risky portfolio on the efficient frontier of
• If the stocks are independent and have identical
risky assets that is tangent to the CAL.
risks, then:
• Is the risky portfolio on the efficient frontier of
1
σÜ? = ⋅ Var ‚‚‚‚‚ risky assets with the highest Sharpe ratio.
n
Ü
As n → ∞, σ? → 0. Thus, a very large portfolio • Is the optimal risky portfolio that will be
with independent and identical risks will have selected by a rational investor regardless of risk
zero risk. preference.

www.coachingactuaries.com Copyright © 2018 Coaching Actuaries. All Rights Reserved. 7


Capital Market Line (CML) Capital Asset Pricing Model (CAPM) Market Risk Premium
• Assume investors have homogeneous ræ = E[Ræ ] = rv + βæ [E[R˘ïE ] − rv ] Market risk premium = E[R˘ïE ] − rv

expectations. Since βæ only captures systematic risk, E[Ræ ] under
Two methods to estimate the market risk
o All investors have the same efficient frontier the CAPM is not influenced by nonsystematic risk.
premium:
of risky portfolios, and thus the same optimal
Assumptions of the CAPM • The historical risk premium: Uses the
risky portfolio and CAL.
• Investors can buy and sell all securities at historical average excess return of the market
o Every investor will use the same optimal
competitive market prices. There are no taxes over the risk-free interest rate.
risky portfolio -- the market portfolio.
or transaction costs. Investors can borrow and • A fundamental approach: Uses the constant
o When the market portfolio is used as the
lend at the risk-free interest rate. expected growth model to estimate the market
risky portfolio, the resulting CAL is CML.
• Investors hold only efficient portfolios of traded portfolio’s expected return.
• The equation for CML is:
securities. Div_ Div_
E[R˘ ] − rv P% = ⇒ E[R˘ïE ] = + g
E[Rœ˘ ] = rv + q r σœ˘ • Investors have homogeneous expectations E[R˘ïE ] − g P%
σ˘
regarding the volatilities, correlations, and
• Individual securities plot below this line. The Debt Cost of Capital
expected returns of securities.
Two methods to estimate debt cost of capital:
Adding a New Investment The consequence of these assumptions is that the
• Adjustment from debt yield:
Suppose we have a portfolio, P, with an expected market portfolio is the efficient portfolio.

rt = y − pL
return of E[R? ] and a volatility of σ? . = Yield to mat. − Pr(Default) E[Loss rate]
Security Market Line (SML)
SML is a graphical representation of CAPM: • CAPM using debt betas:
Add the new investment to the portfolio if:
rt = rv + βt [E[R˘ïE ] − rv ]
Sharpe Ratioñ}˙ > ρñ}˙,? ⋅ Sharpe Ratio?

E[Rñ}˙ ] − rv E[R? ] − rv Required Return on Project


> ρñ}˙,? ⋅ Assuming a project is financed entirely with
σñ}˙ σ?
equity, we can estimate the project’s cost of
capital and beta based on the asset or unlevered
ASSET PRICING MODELS
ASSET PRICING MODELS cost of capital and the beta of comparable firm:

All-
Beta Comparable Levered
Equity
Definitions and Key Facts
Beta β = β~ β = w~ β~ + w˝ β˝
• Measures the sensitivity of the asset's return to
Cost of
the market return. r = r~ r = w ~ r~ + w ˝ r˝
CML vs. SML Capital
• Is defined as the expected percent change in an
CML SML Net debt
asset's return given a 1% change in the market
Based on Based on = Debt − Excess cash and short-term investments
return.
total risk systematic risk If the project is financed with both equity and debt,
• The beta for a stock, on average, is around 1.
Holds for any security then use the weighted-average cost of capital
• Cyclical industries (tech, luxury goods) tend to Only holds for
or combination of (WACC):
have higher betas. efficient portfolios
securities
• Non-cyclical industries (utility, pharmaceutical) rˇ{ŒŒ = w~ r~ + w˝ r˝ (1 − τŒ )

tend to have lower betas.
Alpha WACC is based on the firm’s after-tax cost of debt
Interpreting Beta The difference between a security's expected while the unlevered cost of capital is based on the
• If β = 1, then the asset has the same systematic return and the required return (as predicted by firm’s pretax cost of debt.
risk as the market. The asset will tend to go up the CAPM) is called alpha:

and down the same percentage as the market. αæ = E[Ræ ] − ræ Multi-factor Model
• If β > 1, then the asset has more systematic risk = E[Ræ ] − (rv + βæ [E[R˘ïE ] − rv]) • Considers more than one factor when
than the market. The asset will tend to go up If the market portfolio is efficient, then all estimating the expected return.
and down more than the market, on a securities are on the SML, and: • Also known as the Arbitrage Pricing Theory
percentage basis. E[Ræ ] = ræ and αæ = 0 (APT).
• If β < 1, then the asset has less systematic risk If the market portfolio is not efficient, then the • Similar to CAPM, but assumptions are not as
than the market. The asset will tend to go up securities will not all lie on the SML, and: restrictive.

and down less than the market, on a percentage E[Ræ ] ≠ ræ and αæ ≠ 0
Key Equations
basis. Investors can improve the market portfolio by:
Using a collection of N well-diversified portfolios:
• If β = 0, then the asset’s return is uncorrelated • buying stocks whose E[Ræ ] > ræ (i.e., αæ > 0) ‰
with the market return. • selling stocks whose E[Ræ ] < ræ (i.e., αæ < 0) E[Ræ ] = rv + ⁄ βúñ
æ (E[ Rúñ ] − rv )

ñ¡_
Calculating Beta
Required Return on New Investment where βú_ úñ
æ , … , βæ are the factor betas of asset i
Cov[Ræ, R˘ïE ] σæ
βæ = = ρæ,˘ïE ⋅ Adding the new investment will increase the that measure the sensitivity of the asset to a
σܢïE σ˘ïE
ñ
Sharpe ratio of portfolio P if its expected return particular factor, holding other factors constant.

β? = ⁄ x æβæ exceeds its required return, defined as:
If all factor portfolios are self-financing, then we
r‰}˙ = rv + β‰}˙,? [E[R? ] − rv ]
æ¡_ can rewrite the equation as:

In general, the beta of an asset i with respect to a ‰
Beta can be estimated using linear regression:
portfolio P is: E[Ræ ] = rv + ⁄ βúñ
Ræ − rv = αæ + βæ (R˘ïE − rv) + ϵæ æ (E[ Rúñ ])
Cov[Ræ, R? ] σæ ñ¡_
βæ = = ρæ,? ⋅
σÜ? σ?

www.coachingactuaries.com Copyright © 2018 Coaching Actuaries. All Rights Reserved. 8


Fama-French-Carhart (FFC) Empirical Evidence Against EMH The Behavior of Individual Investors
This model considers 4 factors: Calendar/Time Anomalies Underdiversification and Portfolio Biases
• Market • January effect: Returns have been higher in Individual investors fail to diversify their
• Market capitalization January (and lower in December) than in other portfolios adequately. They invest in stocks of
• Book-to-market ratios months. companies that are in the same industry or are
• Momentum • Monday effect: Returns have been lower on geographically close:

Monday (and higher on Friday) than on other • Investors suffer from familiarity bias, favoring
The FFC estimates the expected return as:

days of the week. investments in companies they are familiar
E[Ræ ] = rv + β˘ïE
æ E[R˘ïE ?‘>E ] + β≠˘Ô
æ E[R≠˘Ô ] • Time-of-day effect: Returns are more volatile with.
+ β"˘¥
æ E[R"˘¥ ] close to the opening and closing hours for the • Investors have relative wealth concerns,
+ β?#_$# E[R?#_$# ] market. Also, the trading volumes are higher caring most about how their portfolio performs
æ
where: during these times. relative to their peers.

Excessive Trading and Overconfidence
Mkt Port = Market portfolio Underreaction/Overreaction Anomalies
Individual investors tend to trade very actively:
SMB = Small-minus-big portfolio • New-Issue/IPO puzzle: Overreaction to new
• Overconfidence bias. They often overestimate
HML = High-minus-low portfolio issues pushes up stock prices initially.
their knowledge or expertise.
PR1YR = Prior 1-year momentum portfolio • Earnings announcement puzzle: Investors
• Men tend to be more overconfident than

underreacted to the earnings announcement.
E[R˘ïE ?‘>E ] = E[R˘ïE ] − rv women.
• Momentum effect vs. reversal effect: Many
E[R≠˘Ô ] = E[R≠|ËÈÈ ] − E•RÔæ% ¶ • Trading activity increases with the number of
studies have identified momentum and reversal
E[R"˘¥ ] = E[R"Ô˘ ] − E[R¥Ô˘ ] speeding tickets an individual receives –
effects, suggesting that stock prices are not
E[R?#_$# ] = E[Rˇæññ}> ] − E[R¥‘&}> ] sensation seeking.
purely random or unrelated to past data.

Systematic Trading Biases
Other Anomalies
Holding on to Losers and the Disposition Effect
MARKET EFFICIENCY
MARKET EFFICIENCY AND BEHAVIORAL • Siamese twins: Two stocks with claims to a
Investors tend to hold on to investments that have
FINANCE & BEHAVIORAL FINANCE common cash flow should be exposed to
lost value and sell investments that have increased
identical risks but perform differently.
An efficient market is a market in which security in value.
• Political cycle effect: For a given political
prices adjust rapidly to reflect any new • People tend to prefer avoiding losses more than
administration, its first year and last year yield
information, i.e., security prices reflect all past and achieving gains. Refuse to “admit a mistake” by
higher returns than the years in between.
present information. taking the loss.
• Stock split effect: Returns are higher before
• Investors are more willing to take on risk in the
Forms of Market Efficiency and after the company announces the stock
face of possible losses.

split.
Market Prices Reflect: Investor Attention, Mood, and Experience
• Neglected firm effect: Lesser-known firms
Past market Public Private yield abnormally high returns. • Individual investors tend to be influenced by
Form
data info info attention-grabbing news or events. Buy
• Super Bowl effect: Historical data shows in the
Weak ✔ year after the Super Bowl, the stock market is stocks that have recently been in the news.
Semi-strong ✔ ✔ more likely to do better if an NFC team won and • Sunshine has a positive effect on mood and
Strong ✔ ✔ ✔ worse if an AFC team won. stock returns tend to be higher on a sunny day

• Size effect: Small-cap companies have at the stock exchange.
Empirical Evidence Supporting EMH • Major sports events have impacts on mood. A
outperformed large-cap companies on a risk-
Supporting Weak Form of EMH loss in the World Cup reduces the next day’s
adjusted basis.
• Kendall found that prices followed a random stock returns in the losing country.
• Value effect: Value stocks have consistently
walk model, i.e., past stock prices have no • Investors appear to put inordinate weight on
outperformed growth stocks.
bearing on future prices.
their experience compared to empirical
• Brealey, Meyers, and Allen created a scatter plot Bubbles also violate market efficiency. It happens
evidence. People who grew up during a time of
for price changes of four stocks. when the market value of the asset significantly
high stock returns are more likely to invest in
o No distinct pattern in the points, with the deviates from its intrinsic value.

stocks.
concentration of points around the origin. No
The Efficiency of the Market Portfolio
bias towards any quadrants. Herd Behavior
Reasons why the market portfolio might not be
o Autocorrelation coefficients were close to 0. Investors actively try to follow each other's
efficient:
• Poterba and Summers found that variance of behavior:
• Proxy Error: Due to the lack of competitive
multi-period change is approximately • Investors believe others have superior
price data, the market proxy cannot include
proportional to number of periods. information, resulting in information cascade
most of the tradable assets in the economy.
Supporting Semi-Strong Form of EMH effect.
• Behavioral Biases: Investors may be subject to
• 3 months prior to a takeover announcement, the • Investors follow others to avoid the risk of
behavioral biases and therefore hold inefficient
stock price gradually increased. At the time of underperforming compared to their peers.
portfolios.
announcement, stock price instantaneously • Investment managers may risk damaging their
• Alternative Risk Preferences: Some investors
jumped. After the announcement, the abnormal reputations if their actions are far different from
focus on risk characteristics other than the
returns dropped to zero. their peers.
volatility of their portfolio, and they may choose
Supporting Strong Form of EMH
inefficient portfolios as a result.
• Top performing fund managers in one year only
• Non-Tradable Wealth: Investors are exposed
have a 50% chance to beat their reference index
to other significant risks outside their portfolio.
the following year.
They may choose to invest less in their
• The performance of actively managed mutual respective sectors to offset the inherent
funds from 1971 to 2013 only beat the Wilshire
exposures from their human capital.
5000 index 40% of the time.

www.coachingactuaries.com Copyright © 2018 Coaching Actuaries. All Rights Reserved. 9


INVESTMENT RISK & PROJECT ANALYSIS
INVESTMENT RISK AND PROJECT ANALYSIS Monte Carlo Simulation • Volatility

General steps: o When huge uncertainty exists regarding the
Measures of Investment Risk
1. Build the model of interest, which is a function future value of the investment (i.e., high
Variance
of several input variables. Assume a specific volatility), the option to wait is more
• The average of the squared deviations above
probability distribution for each input variable. valuable.
and below the mean:
2. Simulate random draws from the assumed • Dividends
Variance = E[(R − E[R])Ü ]
distribution for each input variable. o It is better for an investor to wait unless the
Semi-variance / Downside Semi-variance 3. Given the inputs from Step 2, determine the cost of waiting is greater than the value of
• Only cares about downside risk; ignores upside value of the quantity of interest. waiting.
variability. 4. Repeat Steps 2 and 3 many times. Sizing Option
• The average of the squared deviations below 5. Using the simulated values of the quantity of • Growth options give the company an option to
the mean: interest, calculate the mean, variance, and other make additional investments when it is
Semi-variance = E[min(0, R − E[R])Ü ] measures. optimistic about the future.
• Semi-variance ≤ Variance Inversion method: Set F† (x) = u. • Abandonment options give the company an
option to abandon the project when it is
Value-at-Risk (VaR) Useful Formulas from Exam P & FM pessimistic about the future.
• VaR of X at the 100α% security level is its Poisson eN*λñ
100α th percentile, denoted as VaR ± (X) or π± . Pr[N = n] =
with mean λ n!
Pr[X ≤ π±] = α x
Exponential F† (x) = 1 − exp ì− î CAPITAL STRUCTURE
CAPITAL STRUCTURE
• If X is continuous, then: with mean θ θ

F† (π± ) = α ⇒ π± = F†N_ (α) x−a Equity Financing


Uniform F† (x) =

b−a Equity Funding for Private Companies
Tail Value-at-Risk (TVaR) on [a, b]
Source of funding for private companies:
• TVaR focuses on what happens in the adverse Geometric series:
First Term − First Omitted Term • Angel Investors
tail of the probability distribution. Sum = • Venture Capital Firms
• Also known as the conditional tail expectation 1 − Common Ratio
Infinite geometric series: • Private Equity Firms
or expected shortfall.
First Term • Institutional Investors
1 '( Sum =
TVaR± = E[X|X ≤ π±] = fl x ⋅ f† (x)dx 1 − Common Ratio • Corporate Investors
α N·
· The PV of an annuity:

1 When a private company first sells equity, it


TVaR± = E[X|X > π±] = fl x ⋅ f† (x)dx 1 − vñ
1 − α '( añ| = v + v Ü + ⋯ + v ñ = typically issues preferred stock instead of common
i
Coherent Risk Measures 1 stock.
a·| = v + v Ü + ⋯ =
g(X) is coherent if it satisfies (for c > 0): i A funding round occurs when a private company

• Translation invariance: g(X + c) = g(X) + c
Real Options raises money. An initial funding round might start
• Positive homogeneity: g(cX) = c ⋅ g(X) Real options are capital budgeting options that with a "seed round," and then in later funding
• Subadditivity: g(X + Y) ≤ g(X) + g(Y) give managers the right, but not the obligation, to rounds the securities are named "Series A," "Series
• Monotonicity: If X ≤ Y, then g(X) ≤ g(Y) make a particular business decision in the future B," etc.


VaR vs. TVaR: after new information becomes available.

Pre-Money and Post-Money Valuation
• The VaR is usually not coherent since it does not Decision tree is a graphical approach that • The value of the firm before a funding round is
satisfy the subadditivity characteristic. If the illustrates alternative decisions and potential called the pre-money valuation.
loss distributions are assumed to be normal, the outcomes in an uncertain economy. • The value of the firm after a funding round is
VaR can be shown to be coherent.
called the post-money valuation.
• TVaR is coherent. 2 kinds of nodes in the decision tree:

• The square node (■) is the decision node Post-money valuation


Project Risk Analysis where you have control over the decision. = Pre-money valuation + Amount invested
The net present value (NPV) of a project equals the • The circular node (●) is the information node = # shares after the funding rounds
present value of all expected net cash flows from × Pre-money price per share
where you have no control over the outcome.
the project.


Value of real option Percentage ownership
Breakeven Analysis Amount invested
= NPV(with option) – NPV(without option) =
Calculate the value of each parameter so that the Post-money valuation

project has an NPV of zero. # shares owned × Pre-money price per share

Timing Option (Call Option) =
Sensitivity Analysis Gives a company the option to delay making an Post-money valuation
investment with the hope of having better # shares owned
Change the input variables one at a time to see =
information in the future. Total # shares
how sensitive NPV is to each variable. Using this

analysis, we can identify the most significant Venture Capital Financing Terms
Factors affecting the timing of investment:
variables by their effect on the NPV. Venture capitalists typically hold convertible

• NPV of the investment
preferred stock, which differs from common stock
Scenario Analysis o Without the timing option, invest today if
due to:
Change several input variables at a time, then NPV of investing today is positive.
• Liquidity preference
calculate the NPV for each scenario. The greater o With the timing option, invest today only if
Liquidity preference = Multiplier × Initial inv
the dispersion in NPV across the given scenarios, NPV of investing today exceeds the value of
the option of waiting, assuming the NPV is • Participation rights
the higher the risk of the project.

positive. • Seniority
• Anti-dilution protection
• Board membership

www.coachingactuaries.com Copyright © 2018 Coaching Actuaries. All Rights Reserved. 10


There are two ways to exit from a private 4 common types of corporate debt: Modigliani and Miller (MM) Propositions
company: • Notes (Unsecured) Perfect Capital Markets
• Acquisition • Debentures (Unsecured) • Investors and firms can trade the same set of
• Public offering • Mortgage bonds (Secured) securities at competitive market prices equal to

• Asset-backed bonds (Secured) the present value of their future cash flows.
Initial Public Offering
• No taxes, transaction costs, or issuance costs.
An initial public offering (IPO) is the first time a The new debt that has lower seniority than
• The financing and investment decisions are
company sells its stock to the public. existing debenture issues is called a subordinated
independent of each other.
debenture.
Advantages of IPO:

MM Proposition I (Without Taxes)


• Greater liquidity International bonds are classified into four broadly
• The value of a firm is unaffected by its capital
• Better access to capital defined categories:
structure.
• Domestic bonds – issued by local, bought by
Disadvantages of IPO: • Changing a firm's capital structure merely
foreign
• Dispersed equity holdings changes how the value of its assets is divided
• Foreign bonds – issued by foreign, bought by
• Compliance is costly and time-consuming between debt and equity, but not the firm's total
local

value.
There are two major types of offerings: • Eurobonds – issued by local or foreign
• V¥ = Vµ
• Primary offerings: New shares sold to raise • Global bonds

new capital.
Homemade leverage: Investors can borrow or lend
Corporate Debt: Private Debt
• Secondary offerings: Existing shares sold by at no cost on their own to achieve a capital
Private debt is negotiated directly with a bank or a
current shareholders. structure different from what the firm has chosen.
small group of investors. It is cheaper to issue due

When issuing an IPO, the company and to the absence of the cost of registration. MM Proposition II (Without Taxes)
underwriter must decide on the best mechanism
• The cost of capital of levered equity increases
2 main types of private debt:
to sell shares: with the firm's debt-to-equity ratio.
• Term loan ˝
• Best-efforts: Shares will be sold at the best • r~ = rµ + (rµ − r˝ )
• Private placement ~
possible price. Usually used in smaller IPOs. ~ ˝

• rµ = rˇ{ŒŒ = r + r
• Firm commitment: All shares are guaranteed Other Types of Debt ~ê˝ ~ ~ê˝ ˝

to be sold at the offer price. Most common. Government entities issue sovereign debt and Assuming the cost of debt is constant:
• Auction IPOs: Shares sold through an auction municipal bonds to finance their activities.


system and directly to the public.
Sovereign debt is issued by the national
Standard steps to launching a typical IPO: government. In the US, sovereign debt is issued as
1. Underwriters typically manage an IPO and they bonds called "Treasury securities."

are important because they:
There are four types of Treasury securities:
o Market the IPO.
• Treasury bills
o Assist in required filings.
• Treasury notes
o Ensure the stocks liquidity after the IPO.
• Treasury bonds
2. Companies must file a registration statement,
• Treasury inflation-protected securities (TIPS)
which contains two main parts:
o Preliminary prospectus/red herring. Municipal bond is issued by the state and local
o Final prospectus. governments.
3. A fair valuation of the company is performed by

There are also several types of municipal bonds Note:
the underwriter through road show and book
based on the source of funds that back them: • Since debt holders have a priority claim on
building.
4. The company will pay the IPO underwriters an • Revenue bonds assets and income above equity holders, debt is
underwriting spread. After the IPO, • General obligation bonds less risky than equity, and thus the cost of debt
underwriters can protect themselves more Asset-Backed Securities is lower than the cost of equity.
against losses by using the over-allotment An asset-backed security (ABS) is a security whose • As companies take on more debt, the risk to
allocation or greenshoe provision. cash flows are backed by the cash flows of its equity holders increases, and subsequently the
4 IPO Puzzles: underlying securities. cost of equity increases.

• The average IPO seems to be priced too low. The biggest sector of the ABS market is the • The rising cost of equity exactly offsets the
• New issues appear cyclical. mortgage-backed security (MBS) sector. An MBS benefits of using a larger proportion of debt as a
• The transaction costs of an IPO are high. has its cash flows backed by home mortgages. cheaper source of financing, resulting in a
• Long-run performance after an IPO is poor on Because mortgages can be repaid early, the constant WACC.
average. holders of an MBS face prepayment risk.

Debt Financing Banks also issue ABS using consumer loans, such
Corporate Debt: Public Debt as credit card receivables and automobile loans.
Public debt trades on public exchanges. The bond

A private ABS can be backed by another ABS. This


agreement takes the form of an indenture, which is
new ABS is known as a collateralized debt
a legal agreement between the bond issuer and a
obligation (CDO).
trust company.

www.coachingactuaries.com Copyright © 2018 Coaching Actuaries. All Rights Reserved. 11


As the amount of debt increases, the chance that Note: Managers consider how their actions will be
the firm will default increases, and subsequently • As debt increases, WACC falls, and thus the perceived by investors in selecting financing
the cost of debt increases: value of the firm increases. methods for new investments:
Financial Distress Costs • Issuing equity is typically viewed as a negative
The present value of financial distress costs has signal as managers tend to issue equity when
three components: they believe that the firm’s stock is overvalued.
1. The costs of financial distress and • Issuing more debt is typically viewed as a
bankruptcy, in the event they occur. positive signal as the company is taking on
o Direct costs – actual cash expenses associated commitment to make timely interest and
with the bankruptcy process. principal payments.
o Indirect costs – forgone investment Adverse selection has several implications for
opportunities, costs from losing customers, equity issuance:
creditors, suppliers, etc. • The stock price declines on the announcement
2. The probability of financial distress and of an equity issue.
bankruptcy occurring • The stock price tends to rise prior to the

Note: o Companies with a higher debt-to-equity ratio announcement of an equity issue.
have a higher probability of bankruptcy. • Firms tend to issue equity when information
• Although both cost of debt and cost of equity
increase as the company takes on more debt, 3. The appropriate discount rate for the asymmetries are minimized, such as
WACC remains unchanged. distress costs immediately after earnings announcement.
o The higher the firm's beta, the more likely it

WACC with Multiple Securities: will be in distress, the more negative the beta Pecking order hypothesis: Managers prefer to
rµ = rˇ{ŒŒ = ⁄ wæ ⋅ ræ of its distress costs, the lower the discount make financing choices that send positive rather
rate, the higher the PV of distress costs. than negative signals to outside investors.

Levered and Unlevered Betas:
The Agency Costs of Leverage The pecking order (from most favored to least
βµ = w~ β~ + w˝ β˝
D • Excessive risk-taking and asset substitution. favored financing option)
β~ = βµ + (βµ − β˝ ) A company replacing its low-risk assets with • Internally generated equity (i.e., retained
E
high-risk investments. Shareholders may earnings)
MM Proposition I (With Taxes) benefit from high-risk projects, even those with • Debt
• The use of debt results in tax savings for the negative NPV. • External equity (i.e., newly issued shares)
firm, which adds to the value of the firm. • Debt overhang or underinvestment.

• V¥ = Vµ + PV(Interest tax shield) Shareholders may be unwilling to finance new, Trade-Off Theory
Interest tax shield = Corp. Tax Rate × Int Pmt positive-NPV projects. Balance the value-enhancing effects of debt on a
• The value of a firm is maximized if its capital • Cashing out. When a firm faces financial firm's capital structure with the value-reducing
structure is 100% debt. distress, shareholders have an incentive to effects.

For a firm that borrows debt D and keeps the debt liquidate assets at prices below their market V¥
permanently, if the firm's marginal tax rate is τŒ , values and distribute the proceeds as dividends. = Vµ + PV(Interest tax shield)

and the debt is riskless with a risk-free interest The Agency Benefits of Leverage − PV(Financial distress costs)
rate rv, then the present value of the interest tax Managers have interests that may differ from − PV(Agency costs of debt)
shield is: shareholders' and debt holders' interests: + PV(Agency benefits of debt)

PV(Interest tax shield) = τŒ ⋅ D • Empire building. Managers tend to take on

MM Proposition II (With Taxes) investments that increase the size, rather than
• With taxes, when more debt is incorporated into the profitability, of the firm
the capital structure, the cost of equity will still • Managerial entrenchment. Because managers
rise, but it does not rise as rapidly as it would if face little threat of being replaced, managers can
there were no taxes. run the firm to suit their interests.

˝
• r~ = rµ + (rµ − r˝ )(1 − τŒ ) Leverage can provide incentives for managers to
~
• rˇ{ŒŒ =
~
r~ +
˝
r˝ (1 − τŒ ) run a firm more efficiently and effectively due to:
~ê˝ ~ê˝
• Increased ownership concentration.
Assuming the cost of debt is constant: • Reduced wasteful investment.
• Reduced managerial entrenchment and
increased commitment.


Costs of Asymmetric Information

Lemons principle: When managers have private The optimal level of debt, D∗ , occurs at the point
information about the value of a firm, investors where the firm's value is maximized. It balances
will discount the price they are willing to pay for the benefits and costs of leverage.
new equity issue due to adverse selection.

Copyright © 2018 Coaching Actuaries. All Rights Reserved. 12


www.coachingactuaries.com Copyright
Personal © 2018
copies Coaching
permitted. Actuaries.
Resale All Rights
or distribution Reserved. 12
is prohibited.

You might also like