Corporate Finance 2 Notes

1. Empirical Regularities
Observations:

 No single security generates lower risk than the index ⇒ benefit in holding a portfolio relative to individual securities
⇒ the power of risk diversification
 Indices lie on the line ⇒ it suggests that:
o there is a market price for risk (for lower risk sacrifice some returns)
o the market price for risk is the same
 This relationship does not hold for individual securities ⇒ idiosyncratic risk is not priced (only market risk)
𝔼𝒕 [𝑑̃𝑡+1 +𝑝̃𝑡+1 ]
 𝑟𝑡 = ⇒ so returns depend on expectations as of today ⇒ highly priced stocks may have low expected returns
𝑝𝑡
 If market is in equilibrium ⇒ asset pricing theory can explain why people bought the dominated overpriced securities
 Governmental bonds are safe assets, not risk-free (no government is free of default risk)
 A shortfall – an amount by which a financial obligation or liability exceeds the required amount of cash that is available
 Risk – an uncertainty about the ability of the borrower to fulfill their financial obligations to the lender
 Types of risk:
o Idiosyncratic risk (individual security, is not rewarded with higher returns)
o Systematic risk (market risk, cannot be avoided by diversification, is rewarded with higher returns)
o Systemic risk (risk of a crisis, exogenous, e.g. collapse of the banks, related to large jumps of equilibria)
 A short squeeze – an unusual condition that triggers rapidly rising prices in a security. For a short squeeze to occur the security must
have an unusual degree of short-sellers holding positions in it. The short squeeze begins when the price jumps higher unexpectedly.
The condition plays out as a significant measure of the short sellers coincidentally decide to cut losses and exit their positions.
 Volkswagen 2008 short squeeze:
o During the 2008 crisis Volkswagen stock was increasing while all other stocks were falling
o Hedge funds decided to short sell
o When they wanted to close their positions there was no stock left to buy, so the price skyrocketed
 Dot-com bubble:
o Does it make sense to even attempt to determine the long-term value?
o How a bubble relates to the equilibrium idea of long-term value?
o Unlike the Great Financial Crisis and other, did not destroy value, but redistributed
 US Home Prices bubble:
o Decreasing interest rates fueled the bubble, then hike in interest rates made the bubble burst
o Exit from the quantitative easing is costly, need to hike interest rate carefully to avoid bursts
o This drastic rapid changes in price make wonder whether the prices are informative
 Black Monday, October 1987
o Such large jumps put to question the existence of equilibria
o Or maybe suggest the existence of multiple equilibria (markets are very good at aggregating information)
o Views:
 Markets are unstable, bound to crash – need to regulate markets?
 There was a too rapid grows, a correction and then a more sustainable growth – need to liberalize markets?
 Flash Crash, 6 May 2010
o Automatic stock trading caused a large fast dip, amplified a downward trend.
o Such a large rapid change again raises a question about equilibrium and long-run value
 Take-Over Announcement
o Prices tend to elevate right before the announcement (due to insider trading)
o Hedge funds try to learn from prices to seize such opportunities
 Conclusion:
o Market prices reflect information to some degree
o News are difficult to predict
o Prolonged mispricing may occur
 Volatility:
 o Not only prices per se are informative, but also variation in prices (high variation indicates risk)
o Volatility is good (as long as it is not a crash) because it is a price update after new information became available
 Functions of fin. markets:
o Intertemporal transfer of resources (borrowing, lending, saving)
o Risk management and risk allocation (insurance, hedging)
o Information production (the prices of liquid securities reveal more information than of illiquid)

Multiple Equilibria and Comparative Statics:

 Equilibrium might not exist if the demand or supply is discontinuous

 If we assume that technologies and preferences are convex, then demand and supply functions are continuous and will intersect
 Multiple equilibria are very difficult to detect, especially ex-ante because it requires non-linear models and much more data
o We have data only near the “high” equilibrium
 With linear models this is impossible to detect because they approximate linearly a non-linear demand function
 Increased volatility near the turning point and larger bid-ask spreads (illiquidity) suggest uncertainty about the equilibrium price

2. Risk and Return

Portfolio Choice: Mean Variance Approach:

 The principle of insufficient reason – if don’t know the future, assume that past data generating process will continue
⇒ use historical number to predict the future ⇒ stationarity (very strong unrealistic assumption)
 Linear relation holds only for standard deviation
 Market rewards only systematic risk of the securities (based on co-movement with the market – beta)
 Returns need to be equilibrium returns
o Have to explain why people hold dominated securities – popular or for hedging
 Risk Diversification:
o Mean: 𝜇𝜆 = 𝜆 ∙ 𝜇𝑥 + (1 − 𝜆)𝜇𝑦
o Variance: 𝜎𝜆2 = 𝜆12 𝜎𝑥2 + (1 − 𝜆)𝜎𝑦2 + 2𝜆(1 − 𝜆)𝜎𝑥𝑦 (portfolio risk decreases in covariance)
𝜎𝜆2 = 𝜆12 𝜎𝑥2 + (1 − 𝜆)𝜎𝑦2 + 2𝜆(1 − 𝜆)𝜎𝑥 𝜎𝑦 𝜌𝑥𝑦
o 𝜌𝑥𝑦 = +1 ⇒ no diversification benefit
𝜎𝜆2 = 𝜆12 𝜎𝑥2 + (1 − 𝜆)𝜎𝑦2 + 2𝜆(1 − 𝜆)𝜎𝑥 𝜎𝑦
𝜎𝜆 = 𝜆𝜎𝑥 + (1 − 𝜆)𝜎𝑦
o 𝜌𝑥𝑦 = −1 ⇒ (insurance) can find a portfolio that is risk-free
𝜎𝜆2 = 𝜆12 𝜎𝑥2 + (1 − 𝜆)𝜎𝑦2 − 2𝜆(1 − 𝜆)𝜎𝑥 𝜎𝑦
𝜎𝜆 = |𝜆𝜎𝑥 − (1 − 𝜆)𝜎𝑦 |
o There is no diversification potential with a riskless asset, since 𝜎𝑥𝑦 = 0
 In equilibrium nobody holds short positions, everybody holds long position of market portfolio
 Sharpe ration = excess return over excess volatility
 Conditions for the efficient frontier (EF) to be well-behaved:
o If there are finitely many assets ⇒ the EF is closed and connected
o If there are finitely many assets & ≥ 1 asset has 𝜇 exceeding 𝜇𝑟𝑓 ⇒ a tangency PF exists
o If there are finitely many assets & short selling is feasible ⇒ a tangency PF exists
 Computing the Tangency PF:
𝑏𝑖
o Setup: 𝑈 𝑖 = (𝜇𝜆 , 𝜎𝜆2 ) = 𝜇𝜆 − 2
𝜎𝜆2 , where ∑𝐾
𝑘=0 𝜆𝑘 = 1
𝑏 𝑖
o Optimal PF shares: 𝜆̂𝑖 ∈ 𝑎𝑟𝑔𝑚𝑎𝑥(𝜇 − 𝑅0𝑡 )′𝜆 − 2 𝜆′𝐶𝑂𝑉𝜆
1 ̂
𝜆
o With short sales: 𝜆̂𝑖 = 𝐶𝑂𝑉 −1 (𝜇 − 𝑅0 𝑝) 𝜆𝑘 = ∑𝐾 𝑘
𝑏𝑖 𝑙=0 𝜆𝑙

 Market Equilibrium:
 𝑏 𝑖
o Optimal portfolio choice: 𝜃̂ 𝑖 ∈ 𝑎𝑟𝑔𝑚𝑎𝑥(𝜇 − 𝑅0𝑡 )′ 𝜃 − 2 𝜃′𝐶𝑂𝑉𝜃
1
o Demand for risky securities: 𝜃̂ 𝑖 = 𝐶𝑂𝑉 −1 𝑏𝑖 (𝜇 − 𝑅0 𝑝)
 Normal goods: 𝑖𝑓 𝑝1 ↑ ⇒ D1 ↓ (demand for normal goods is decreasing in price)
 Substitutes: 𝑖𝑓 𝑝2 ↑ ⇒ D1 ↑ (but really depends on cov matrix)
 Complements: 𝑖𝑓 𝑝2 ↑ ⇒ D1 ↓ (but really depends on inverse cov matrix)
 Missing elements (e.g.):
 Wealth: 𝑖𝑓 𝑤 ↑ ⇒ D ↑
 Interest rates: 𝑖𝑓 𝑅0 ↑ ⇒ D ↓ (in general, but for substitutes correlation effect might override)
 News: 𝑖𝑓 𝑛𝑒 ↑ ⇒ D ↑
1 1 1
o Equilibrium price: 𝑝= (𝜇 − 1 𝐶𝑂𝑉 ∙ 𝑧), where ∑𝐼𝑖=1 is the risk bearing capacity
𝑅0 ∑𝐼𝑖=1 𝑏𝑖
𝑏𝑖

(risk bearing capacity↑ ⇒ risk premium↓)

𝑑
𝜇(𝜆𝑅0 +(1−𝜆)𝑅𝑀 )|𝜆=0 𝑅0 −𝜇𝑀
o Slope of the CML: 𝑑𝜆
𝑑 =
𝜎(𝜆𝑅0 +(1−𝜆)𝑅𝑀 )|𝜆=0 −𝜎𝑀
𝑑𝜆
𝑑
𝜇(𝜆𝑅𝑗 +(1−𝜆)𝑅𝑀 )| 𝜇𝑗 −𝜇𝑀
o Slope of PF (𝑥̃𝑗 , 𝑥̃ 𝑀 ): 𝑑𝜆
𝑑
𝜆=0
= 𝐶𝑂𝑉(𝑅𝑗 ,𝑅𝑀 −𝜎2
𝜎(𝜆𝑅𝑗 +(1−𝜆)𝑅𝑀 )| 𝑀)
𝑑𝜆 𝜆=0
𝜎𝑀

2 ), 𝐶𝑂𝑉(𝑥𝑘 ,𝑥̃ 𝑀 )
o Tangency condition yields: 𝑅0 − 𝜇𝑀 = 𝛽𝑘 (𝑅𝑀 − 𝜎𝑀 where 𝛽𝑘 = 2
𝜎𝑀
o Equilibrium prices adjust such that in equilibrium investors optimally hold the tangency PF, which then is the market PF.
o If a security were not a part of the efficient PF, then no investor would want to own it (𝐷 ≠ 𝑆), therefore, its price would fall
causing its expected return to rise until it became attractive ⇒ prices adjust so that efficient PF and the market PF coincide
and 𝐷 = 𝑆.
o Empirical implications:
 Anomalies:
 Small firm effect. One of the predictions of this theory is that there is a linear relationship between the
excess return of a particular security and the excess return of the market portfolio. The proportionality
parameter is called beta and it measures the systematic risk of a security. The intercept alpha in principle
should be zero, however, empirical studies show that this is not true. Small caps have positive alpha, so
these securities reward not only for systematic risk, but also for something else outside the model (which
invalidates it). This additional reward seems to be for the liquidity risk that you take on by buying small
caps because they usually are in less liquid markets. (Size effect)
 Home bias. In the model everyone invests in the same market portfolio, but this is not true. (quantity effect,
complexity)
 Market Equilibrium with Heterogeneous Beliefs:
o Setup: heterogeneous expectations, other moments identical
1
o Optimal individual PF: 𝜃̂ 𝑖 = 𝐶𝑂𝑉 −1 𝑖 (𝜇𝑖 − 𝑅0 𝑡)
𝑏
o The SML holds for the average beliefs.
o Implications:
 Two fund separation fails (different beliefs, different assets, not market PF ⇒ different SML)
⇒ different quantities)
⇒ tangency and market PF differ)
⇒ makes sense to distinguish active traders (tangency PF) and passive traders (market PF)
 Beta unchanged (as 2nd moments are unchanged)
 Investors have different 𝛼
 The average market belief depends on wealth (wealth ↑ ⇒ influence ↑) and risk aversion (risk-aver ↓ ⇒ influence ↑)
 Need to model information. Information is omitted in the standard CAPM. Heterogeneous belief model is a rough proxy for
information, but it is not a good model because individual investors should take into account information of others.
 So the standard model is taking out the key contributor to asset pricing by assuming that posteriors are identical (post trade
information is the same) – which is the source of two-fund separation – it is a very unrealistic assumption.
 And trying to factor that in the Heterogeneous Beliefs Model we lose the explanatory power of the CAPM.
 To defend the CAPM: in more liquid markets (less information asymmetry) the model applies quite well (not in the periods of crisis).
 Richer models:
 o Alternative Betas (include more factors: background risk, liquidity risk…)
o Higher Moment Betas (include more info about the return distribution: catastrophic risk, skewness)
o Behavioral CAPM (include asymmetric behavioral biases of PT, CPT)
(only individual recommendations, no recommendations for the holder of the market portfolio)

3. Asset Pricing
3.1 The Fundamental Asset Pricing Equation:

 𝑈(𝑐𝑡 , 𝑐̃𝑡+1 ) = 𝑢(𝑐𝑡 ) + 𝛽𝔼𝑡 [𝑢(𝑐̃𝑡+1 )]

Preferences: • time-separable
• function: → increasing (desire more consumption)
→ concave (declining marginal utility of consumption)
• curvature/
convexity: → imply risk aversion to consumption fluctuations over time
→ imply preference for consumption smoothing (mixture)
• reflect fundamental functions of the fin. system: → inter-temporal trade: discounting 𝛽, cons. smoothing 𝑢(𝑐𝑡 )
→ hedging of risks: curvature of 𝑢(𝑐𝑡 ) (risk aversion)
→ inf. about the future consumptions possibilities: 𝔼𝑡 [𝑢(𝑐̃𝑡+1 )]
information is the same for everybody, not heterogeneous
 𝔼𝑡 – expectation based on the information available at time = 𝑡
 Endowments: (𝑒𝑡 , 𝑒̃𝑡+1 )
 Market structure:
o Intertemporal transfers can be performed by buying an asset at 𝑡 for price 𝑝𝑡 that delivers an uncertain payoff 𝑥̃𝑡+1 at 𝑡 + 1
o Completeness is crucial: • if complete: → sufficiently many trading opportunities ⇔
⇔ (can achieve the optimal inter-temporal consumption and hedging)
→ the budget constraint can be reduced to a single lifetime budget constraint
• if incomplete: → not all potential gains from trade can be achieved
→ consumers typically face a sequence of budget constraint
→ First Welfare Theorem doesn’t work (have a reason for gov. intervention)

 Optimal purchase 𝜽: max(𝑐𝑡 ) + 𝛽𝔼𝑡 [𝑢(𝑐̃𝑡+1 )] s.t. (𝐵𝐶1 ) 𝑐𝑡 = 𝑒𝑡 − 𝑝𝑡 𝜃 (budget constraint 1)

𝜃
(𝐵𝐶2 ) 𝑐̃𝑡+1 = 𝑒̃𝑡+1 + 𝜃𝑥̃𝑡+1 (budget constraint 2)
⇒ max(𝑒𝑡 − 𝑝𝑡 𝜃 ) + 𝛽𝔼𝑡 [𝑢(𝑒̃𝑡+1 + 𝜃𝑥̃𝑡+1 )]
𝜃
FOC: ⇒ −𝑝𝑡 𝑢′ (𝑐𝑡 ) + 𝛽𝔼𝑡 [𝑢′ (𝑐̃𝑡+1 )𝑥̃𝑡+1 ] = 0
⇒ 𝑝𝑡 𝑢′ (𝑐𝑡 ) = 𝛽𝔼𝑡 [𝑢′ (𝑐̃𝑡+1 )𝑥̃𝑡+1 ]
𝑢′ (𝑐̃𝑡+1 )
⇒ 𝑝𝑡 = 𝔼𝑡 [𝛽 𝑥̃ ]
𝑢′ (𝑐𝑡 ) 𝑡+1
The Fundamental Asset Pricing Equation (CCAPM)
𝑢′ (𝑐̃𝑡+1 )
where 𝛽 𝑢′ (𝑐𝑡 )
is the MRS between period 𝑡 and 𝑡 + 1
o In equilibrium, different agents may have different 𝜃, but the MRS will be the same for everyone and = to the price
o MRS is the same for everyone because everyone faces the same budget constraints
 Has to hold with symmetric information
 Has to hold with asymmetric information
 Stochastic
𝑢′ (𝑐̃𝑡+1 )
discount factor: 𝑚
̃ 𝑡+1 = 𝛽 𝑢′ (𝑐𝑡 )
(it is a summary statistic of preference parameters)

⇒ 𝑝𝑡 = 𝔼𝑡 [𝑚
̃ 𝑡+1 𝑥̃𝑡+1 ] The fundamental asset pricing relation
→ prices are linked to payoffs via consumption
→ different models just generate different 𝑚
̃ 𝑡+1
1
 Under certainty: 𝑅𝑓 = 1 + 𝑟 𝑓 ⇒ discount factor = 𝑅𝑓
1
𝑝𝑡 = 𝑅𝑓 𝑥𝑡+1
 𝑥̃ 𝑝̃ +𝑑 ̃
 Under uncertainty: 𝑅̃𝑡+1 = 𝑡+1 = 𝑡+1 𝑡+1
𝑝𝑡 𝑝𝑡
𝑝𝑡 𝑥̃𝑡+1
= 𝔼𝑡 [𝑚
̃ 𝑡+1 ]
𝑝𝑡 𝑝𝑡

⇒ ̃ 𝑡+1 𝑅̃𝑡+1 ]
1 = 𝔼𝑡 [𝑚 The Fundamental Asset Pricing Equation

 Risk-free rate: 𝑅 𝑓 = 1/𝔼𝑡 [𝑚

̃ 𝑡+1 ] Fundamental relation
𝑐 1−𝛾
𝛾 ≠ 1 ⇒ 𝑢′ (𝑐) = 𝑐 −𝛾
o CRRA: 𝑢(𝑐) = { 1−𝛾 1
(can express marginal utility ~ observable consumption)
ln(𝑐) 𝛾 = 1 ⇒ 𝑢′ (𝑐) = 𝑐
o Under certainty:
o With log-normal payoffs:

 Risk corrections: ̃ 𝑡+1 ; 𝑥̃𝑡+1 ) = 𝔼𝑡 [𝑚

𝑐𝑜𝑣(𝑚 ̃ 𝑡+1 𝑥̃𝑡+1 ] − 𝔼𝑡 [𝑚
̃ 𝑡+1 ]𝔼𝑡 [𝑥̃𝑡+1 ]
𝔼𝑡 [𝑥̃𝑡+1 ]
⇒ 𝑝𝑡 = 𝑅𝑓
̃ 𝑡+1 ; 𝑥̃𝑡+1 )
+ 𝑐𝑜𝑣(𝑚

𝔼𝑡 [𝑥̃𝑡+1 ] 𝑐𝑜𝑣(𝛽𝑢′ (𝑐̃𝑡+1 );𝑥̃𝑡+1 )

⇔ 𝑝𝑡 = + (price = CFs + risk corrections)
𝑅𝑓 𝑢′ (𝑐𝑡 )

̃ 𝑡+1 𝑅̃𝑡+1
1 = 𝔼𝑡 [𝑚 𝑖
]
̃ 𝑡+1 ]𝔼𝑡 [𝑅̃𝑡+1
1 = 𝔼𝑡 [𝑚 𝑖
̃ 𝑡+1 ; 𝑅̃𝑡+1
] + 𝑐𝑜𝑣(𝑚 𝑖
)
given that: 𝑅 𝑓 = 1/𝔼𝑡 [𝑚
̃ 𝑡+1 ]
𝑖 )
𝑐𝑜𝑣(𝛽𝑢′ (𝑐̃𝑡+1 );𝑅̃𝑡+1
we have: 𝔼𝑡 [𝑅̃𝑡+1
𝑖
] − 𝑅𝑓 = − ′
𝔼𝑡 [𝑢 (𝑐𝑡 )]

 Assets that co-vary positively with consumption need to compensate for their contribution to asset volatility
by paying off higher returns.
 Assets that co-vary negatively with consumption provide extra insurance services, and, hence, can offer
lower rates of return than the safe rate.

3.2 General Equilibrium:

 Needed elements for equilibrium:

o Preferences: 𝑈 𝒊 (𝑐𝑡𝒊 , 𝑐̃𝑡+1
𝒊
) = 𝑢𝒊 (𝑐𝑡𝒊 ) + 𝛽 𝒊 𝔼𝑡 [𝑢𝒊 (𝑐̃𝑡+1
𝒊
)]
o Endowments: (𝑒𝑡𝒊 , 𝑒̃𝑡+1
𝒊
)
o Information: Each consumer entertains beliefs about future endowments and securities returns
o Market structure: n risky assets traded at 𝑡 for price 𝑝𝑡 that deliver an uncertain payoff 𝑥̃𝑡+1 at 𝑡 + 1

 Optimal PF choice: max(𝑐𝑡𝑖 ) + 𝛽 𝒊 𝔼𝑡 [𝑢(𝑐̃𝑡+1

𝑖
)] s.t. (𝐵𝐶1 ) 𝑐𝑡𝑖 = 𝑒𝑡𝑖 − 𝑝𝑡 𝜃 𝑖
𝜃
𝑖 𝑖
(𝐵𝐶2 ) 𝑐̃𝑡+1 = 𝑒̃𝑡+1 + 𝜃 𝑖 𝑥̃𝑡+1
𝑖 )
𝑢′ (𝑐̃𝑡+1
⇒ 𝑝𝑡 = 𝔼𝑡 [𝛽 𝒊 𝑢′ (𝑐𝑡𝑖 )
𝑥̃𝑡+1 ]

𝑖
⇔ 𝑝𝑡 = 𝔼𝑡 [𝑚
̃ 𝑡+1 𝑥̃𝑡+1 ]
 Market Equilibrium:
o The asset market is in equilibrium in period 𝑡 if there is a price system 𝑝̂ 𝑡 and an allocation of portfolios 𝜃̂ = (𝜃̂1 , … , 𝜃̂ 𝐼 ) s. t.
 Individual optimality: 𝜃̂ 𝑖 ∈ 𝑎𝑟𝑔max 𝑢(𝑐𝑡𝑖 ) + 𝛽 𝒊 𝔼𝑡 [𝑢(𝑐̃𝑡+1
𝑖
)] s.t. (𝐵𝐶1 ) 𝑐𝑡𝑖 = 𝑒𝑡𝑖 − 𝒑𝒕 𝜃 𝑖
𝜃
𝑖 𝑖
(𝐵𝐶2 ) 𝑐̃𝑡+1 = 𝑒̃𝑡+1 + 𝜃 𝑖 𝑥̃𝑡+1 ∀ 𝑖 = 1, … , 𝐼
(everybody hold what he likes)
𝑍1
 Market clearing: ̂ 𝒊
∑𝑖=1,…𝐼 𝜽 = 𝑍 = ( ⋮ ) (all shares are sold)
𝑍𝑛
o Type of variables:
  Exogenous to individual investor’s problem: • preferences
• endowments
• market structure
• (price)
 Endogenous to individual investor’s problem: • consumptions plan
• security demand
 Endogenous to market equilibrium: • price
• consumption allocation
o Remarks:
 The price system is the mechanism that adjusts the valuations of the securities in a way that markets clear:
everybody hold what he likes and all shares are sold.
 If no one wants to hold a security (demand decreases), then its price falls, which makes the expected returns rise,
making it more attractive.
 For a given price system the market is in equilibrium if the allocation 𝜃 is optimal and clears the market.
 Existence:
o Preferences are strictly convex: existence guaranteed, the function is continuous
o Other: not guaranteed, but may exist
 Optimality:
o First Welfare Theorem: When individual preferences are non-satiated and when the market structure is complete then
market equilibrium is Pareto-efficient.
 Multiple equilibria:
o With heterogeneous investors multiple equilibria may occur, whenever market demand is non-monotonic.
o Remarks:
 Multiple equilibria can contribute to explain market crashes and/or other discontinuous market reactions to changes
in the market environments (such as Black Monday, 1987).
 Challenges to empirical work: linear approximations to a complex world only provide local information. Therefore,
linear statistical models are inadequate to deal with crashes or crises.
 Food for thought and discussion:
o Why are the assumptions of the First Welfare Theorem needed?
 Non-satiation implies convexity of preferences. This assumption is important because otherwise the demand
function is discontinuous and the existence of equilibrium cannot be guaranteed.
 Completeness of market structure implies that all possible second-period consumption streams can be created. This
assumption is important because otherwise not all potential gains from trade can be achieved.
o How restrictive are they?
 Convexity implies risk aversion, but in real world sometimes people exhibit risk-loving behavior, so the assumption
is quite restrictive in that sense.
 In the real world developed capital markets can be considered complete because they allow for all possible second-
period consumption streams, so this assumption should not be too restrictive, but might be if we talk about
underdeveloped capital markets.
o Which are indicators for potentially chaotic market reactions based on complex (underlying) market demand
constellations?
 Increased volatility and increased bid-ask spreads might show high uncertainty about the true price.
o Which were the economic conditions surrounding Black Monday in October 1987?
 The trade deficit of the United States widened with respect to other countries.
 Crises, such as a standoff between Kuwait and Iran, which threatened to disrupt oil supplies, also made investors
jittery.
o Are there any parallels to current market conditions?
 Now the US also has a huge trade deficit.
o How can crashes be avoided?
 To collect a lot more information about the structure of the economy, of the aggregate demand, components of the
demand that might prevent huge crashes.
o What would happen if an economy becomes poorer?
 A lot of income is destroyed ⇒ MRS ↑ (= marginal utility ↑) ⇒ prices ↓
3.3 Contingent Claims:

 Setup:
o States: 𝑠 = 1, … , 𝑆
o Physical probability: 𝜋(𝑠) ∈ ∑𝑠′ =1,…,𝑆 𝜋(𝑠 ′ ) = 1
1 𝑠′ = 𝑠
o Contingent claim: 𝑎𝑠 (𝑠 ′ ) = {
0 𝑠≠𝑠
o Contingent claim price: 𝑝𝑐(𝑠)
o Contingent claims can be synthesized by other (=real) securities when (real) markets are complete
 Securities:
o Arrow: insurance against one state
o Bond: fixed income
o Insurance: payments contingent on insured event
o Derivatives: based on payment structure of the underlying basic asset (commodity, stock, bond, etc.)

 Pricing of securities: 𝑝(𝑥) = ∑𝑠 𝑝𝑐(𝑠)𝑥(𝑠) (pricing based on state price)

𝑝𝑐(𝑠)
𝑝(𝑥) = ∑𝑠 𝜋(𝑠) ( 𝜋(𝑠) ) 𝑥(𝑠) (in terms of physical probabilities)
𝑝𝑐(𝑠)
𝑝(𝑥) = ∑𝑠 𝜋(𝑠)𝑚(𝑠)𝑥(𝑠) (because in equilibrium ( ) = 𝑀𝑅𝑆)
𝜋(𝑠)

𝑝(𝑥) = 𝔼[𝑚
̃ 𝑥̃]
1
 Risk-neutral prob.: 𝜋 ∗ (𝑠) = 𝑅 𝑓 𝑝𝑐(𝑠) (𝑤ℎ𝑒𝑟𝑒 𝑅 𝑓 = 𝔼[𝑚]) (it is a normalized state price)
1
⇔ 𝑝𝑐(𝑠) = 𝜋 ∗ (𝑠)
𝑅𝑓

𝑝(𝑥) = ∑𝑠 𝑝𝑐(𝑠)𝑥(𝑠)
1
⇔ 𝑝(𝑥) = ∑𝑠 𝜋 ∗ (𝑠)𝑥(𝑠)
𝑅𝑓
1
⇔ 𝑝(𝑥) = 𝔼∗ [𝑥̃] (CFs discounted under risk-neutral)
𝑅𝑓
1
𝑝(𝑥) = 𝔼∗ [𝑥̃] = 𝔼[𝑚
̃ 𝑥̃] (dealing with expectations is easier than product of random variables,
𝑅𝑓
but the covariances are still there, they are just hidden)
𝑚(𝑠)
 Transformation: 𝜋 ∗ (𝑠) = 𝜋(𝑠) from physical to risk-neutral probabilities
𝔼[𝑚]
 Interpreting state prices: max 𝑢(𝑐𝑡 ) + ∑𝑠 𝛽𝜋(𝑠)𝑢(𝑐(𝑠)) (𝐵𝐶) 𝑐𝑡 + ∑𝑠 𝑝 ∙ 𝑐(𝑠) ∙ 𝑐𝑡+1 (𝑠) = 𝑒𝑡 + ∑𝑠 𝑝 ∙ 𝑐(𝑠) ∙ 𝑒𝑡+1 (𝑠)
(𝑐𝑡 ,𝑐𝑡+1 )

FOCs: 𝛽𝜋(𝑠)𝑢′ (𝑐𝑡+1 (𝑠)) = 𝑢′ (𝑐𝑡 )𝑝𝑐(𝑠)

𝑢′ (𝑐𝑡+1 (𝑠))
interpreting FOCs: 𝑝𝑐(𝑠) = 𝛽𝜋(𝑠)
𝑢′ (𝑐𝑡 )
𝑝𝑐(𝑠) 𝑢′ (𝑐𝑡+1 (𝑠))
𝑚(𝑠) = 𝜋(𝑠)
= 𝛽 𝑢′ (𝑐𝑡 )
𝑚(𝑠1 ) 𝑢′ (𝑐𝑡+1 (𝑠1 ))
=
𝑚(𝑠2 ) 𝑢′ (𝑐𝑡+1 (𝑠2 ))

 Food for thought:

o Under which conditions can you work out state prices for a give structure of assets?
 We need to have a complete market where the number of assets/securities traded on the market has to be higher
than or equal to the number of states. To calculate state prices, the securities traded on the market have to be
linearly independent as well.
o Why can it be helpful to work out state prices?
 With state prices, we can find out the fair price of the securities. When state prices are known, we can create arrow
certificates. Arrow certificates are hypothetical securities that: pay 1 unit in a state and 0 in others, may be or may
not be traded in the real markets, can be used to replicate any real word security. Arrow certificates can be
synthetized by other (=traded in real market) securities when markets are complete.
o Which implications can be drawn for risk-neutral probabilities?
 Risk-neutral probabilities: state prices normalized by the sum of the state. Risk neutral probabilities take the risk
preferences already into account and can therefore be used as if they were physical probabilities and the investor
were risk neutral. With risk-neutral probabilities, any security can be valued as a fair bet with respect to the state
prices.
 o Which are the advantages of risk-neutral probabilities?
 With risk-neutral probabilities, any security can be valued as a fair bet with respect to the state prices. Correlations
& covariances are not part of risk-neutral probabilities, since only expectations are used to work out the fair
values. In addition, risk-neutral probability measure does not depend on individual characteristics (wealth,
preferences, beliefs). In equilibrium, risk-neutral probabilities are the same for everybody. Risk-neutral
probabilities are often used to price derivatives.
o Which are the limitations of risk-neutral probabilities?
 Contrary to physical probabilities, risk-neutral probabilities cannot be estimated as they are an amalgamation of
the marginal rate of substitution (in equilibrium) with physical probabilities. For this reason, risk-neutral
probabilities are difficult to work out.

3.4 The Role of Arbitrage:

 Law of One Price: perfect substitutes demand the same market price: 𝑝(𝑎𝑥 + 𝑏𝑦) = 𝑎𝑝(𝑥) + 𝑏𝑝(𝑦)
 Theorem: whenever the law of one price holds there exists a unique payoff 𝑧 ∗ ∈ 𝑋 such that 𝑝(𝑥) = 𝔼[𝑧 ∗ 𝑥] ∀𝑥 ∈ 𝑋
 Remark: arbitrage relates to publicly observable market prices, not to bilateral price agreements such as OTC

 Absence of Arbitrage:

o Definition: the price function 𝑝(𝑥) is free of arbitrage, if for any 𝑥 ∈ 𝑋 with 𝑥 ≥ 0; 𝑥(𝑠) > 0; 𝜋(𝑠) > 0…
…for at least one 𝒔 it is strictly positive 𝑝(𝑥) > 0
⇒ 𝑝(𝑥) = 𝔼[𝑚𝑥] and 𝑚(𝑠) > 0 imply no arbitrage

o AP in Eq.: • In complete markets, no arbitrage and the law of one price imply that there is a unique 𝑚 > 0 such
that 𝑝(𝑥) = 𝔼[𝑚𝑥]
• No arbitrage and the law of one price imply the existence of a unique strictly positive discount
factor 𝑚(𝑠) > 0 such that 𝑝(𝑥) = 𝔼[𝑚𝑥] ∀𝑥 ∈ 𝑋
o Definition (1): A trading strategy is self-financing if 𝑝′ 𝜃 ≤ 0
o Definition (2): A self-financing strategy allow risk-free arbitrage if it generates non-negative payoffs 𝐴𝜃 ≥ 0 and
a positive payoff in as least one 𝑠

 Alternative formulation of the Fundamental Theorem of Asset Pricing:

−𝑝′
o Equivalent: • There exists no portfolio such that ( )𝜃 > 0
𝐴
• There exist state prices such that 𝑝𝑘 = ∑𝑠 𝐴𝑘 (𝑠)𝑝𝑐(𝑠) ∀𝑘

 Food for thought:

o What is the economic implication of a strictly positive discount factor?
 It means that prices of positive cash flows in the future cannot be negative or equal to zero.
o How plausible is the imposed linearity of price functionals?
 In the real world factors that are influencing the price don’t need to be linear, so it is not very plausible, although in
empirical work linear models are good models to start the analysis with. Linearity might be an oversimplification
and empirics therefore might indicate that the hypothesis behind the model are not correct.
o How is absence of arbitrage related to risk aversion of potential arbitrageurs?
 Risk aversion implies preferences for mixtures, therefore, MRS cannot be zero. MRS determined the discount
factor, and the discount factor cannot be negative or equal to zero for prices to be arbitrage-free.
o Are there limits to arbitrage in real markets? If yes, which assumptions do they contradict?
 Yes, resources are not limitless, cannot generate and an infinite profit.

3.6 Factor Pricing Models:

 CAPM – single factor model:

o 𝑚̃ 𝑡+1 = 𝑎 − 𝑏𝑅̃𝑡+1
𝑚
in equilibrium m is a linear function of returns of the market portfolio
 o Includes: optimal intertemporal wealth transfer, optimal hedging, information
o Not explained: liquidity, market completeness

 APT – multiple factors:

o 𝑚 ̃ 𝑡+1 = 𝑎 + 𝑏1 𝑓̃𝑡+1
1
+ 𝑏2 𝑓̃𝑡+1
2
+⋯

 Food for thought:

o How can information explain the consumption correlation puzzle?
 The effects of local information are not taken into account. Local information might affect the extent of hedging of
income risk.
o How can information explain the domestic bias in international equity investment?
 The information production in the locale of the investors might be more focused on the local companies. Also, there
might be shortage of information about foreign companies.
o How can information affect market participation and liquidity?

o How can information explain the small firm effect?
 Small firms are riskier because there is less available information about them.

4. Empirical Puzzles
4.1 Empirical Evidence

 Empirical predictions of CAPM:

o Pricing implications: 𝐸𝑅 𝑘 − 𝑅 𝑓 = 𝛽𝑘 (𝐸𝑅 𝑀 − 𝑅 𝑓 )
𝜃𝑘𝑖 𝜃𝑀
o Quantity implications: = 𝜃𝑘𝑀 for all i (people hold risk assets in the same proportion)
𝜃𝑙𝑖 𝑙

 Violations of the Pricing Implications:

o Small firm effect: liquidity risk, less information
o Calendar effects: (abnormal returns correlation with certain months or days of the week)
o Book-to-market:
o Momentum effect: behavioral risk, liquidity, slow incorporation of information
(investing in historically higher return assets give higher returns)
o Reversal effect: slow incorporation of information, corrections
(selling winners and buying losers in the long term gives higher returns)

 Violations of the Quantity Implications:

o Home bias:
 Very low percentage of holding foreign stock.
 More developed markets produce more information, therefore in developed countries domestic bias is less
pronounced.
 The answer: geographic information.
 Solution: market incompleteness, asymmetric information models

o Home bias at home:

 People tend to hold more stocks of companies that are based in their region inside the country.

 Other Anomalies:
o Equity Premium Puzzle:
 Risk-aversion, market return and growth of consumption cannot explain the observable excess return. The risk-
aversion that would justify the excess returns of equity securities over bonds is extremely high.
  Potential explanations: people are risk-loving when they face losses; covariance between market returns and
consumption is higher we primarily take into account consumers who invest.

o Consumption correlation puzzle: single factor explanation?

 CCAPM is built on the notion that financial markets are used for intertemporal asset allocation, hedging and
information production. Participate in financial markets implies a desire to improve risk allocation, for example. So,
if you have income risk in your country and you are risk-averse, you can diversify. If you do that, your consumption
should be less risky than your income. If everybody does it, the correlation between consumption in different
countries should be higher than the correlation of their output. In reality often it is vice versa, so this theory does not
provide a very good explanation of the phenomenon.
 The answer: in order to explain the puzzle, need to take into account heterogeneity of market participants and local
information, where their wealth matters. CCAPM might underestimate the whole risk of investors.
 Market imperfections, like transactions costs
 non-traded goods (haircuts)

Quizzes S2021:
Quiz 1 Quiz 2

1. Which of the following functions belong to fundamental 1. What does the asset price of a security measure?
functions of the financial system? 🗹 The security's long run value.
🗹 consumption 🗹 The discounted sum of future cash-flows.
🗹 speculation ☐ A security's ideal price.
🗹 savings ☐ The sum of future dividends and cash-flows.
☐ trading ☐ Average of bid and ask price.
☐ credit worthiness tests
2. Which are the fundamental functions of financial systems?
2. Which of the following statements is correct? 🗹 They allow the intertemporal transfer of resources.
🗹 Asset pricing explains long run value of securities. 🗹 They generate information about investment projects.
🗹 Asset pricing builds on the notion of market equilibrium. 🗹 They allow speculation.
🗹 Traditional asset pricing theories need to be increasingly refined ☐ They guarantee loan provision when needed.
to explain asset prices in increasingly shorter periods. ☐ They provide for stable pensions.
☐ Asset pricing requires informational efficiency.
☐ Asset pricing is purely theoretically oriented. 3. What is the stock price volatility of a given stock?
🗹 The square root of the variance of the stock price.
3. Security A historically generated a return of .3€ per stock ☐ The standard deviation of the stock index.
with a volatility of .25, while security B generated a return of ☐ The VIX measure.
.4€ with volatility .2. Both companies have announced to pay
out the same dividends of .5€ per stock for the coming year(s). ☐ The absolute deviation from the mean stock price.
Which of the following statements are incorrect? ☐ The variance of the stock price.
☐ Security B is the better investment since it implies a higher
return at lower risk. 4. What is the role of market equilibrium for asset pricing.
☐ Security B trades at a higher price since it is the dominant 🗹 Asset pricing is about the determination of equilibrium returns.
security. 🗹 Stock prices can aggregate heterogeneous expectations only
☐ Security B trades at a higher price since it is less risky when equilibrium is achieved.
☐ Security A trades at a higher price since its returns are higher. ☐ Stock markets trade future expectations and those can never be
in equilibrium.
☐ Securities A and B are equivalent, since the pay the same
expected dividends. ☐ Bubbles and crashes provide sufficient proof that asset markets
can never be in equilibrium.
☐ It is sufficient to discount future cash flows; therefore,
equilibrium is not needed.

Quizzes W2020-21:
Quiz 1 Quiz 2
1. Which are the main functions of the financial system? 1. Which variables are explained in the definition of a market
Explain, how they relate to each other? equilibrium in securities markets?
Main functions of the financial system are: (2.00/2.00)
- intertemporal transfer of resources Vector of security prices, portfolio allocation, consumption
- risk management and risk allocation allocation.
- information production
Intertemporal transfer of resources allows to increase consumption 2. Which variables are exogenous for the investor in the
today, by giving up some consumption in the future, which is definition of a stock market equilibrium?
achieved by borrowing, or, alternatively, it allows to increase (2.00/3.00)
consumption in the future by giving up some consumption today,
which is achieved by lending. Variables of the individual investors:
Risk management and risk allocation is connected to the • Preferences
intertemporal transfer of resources, because it is essentially based • Endowments
on that. Risk management utilizes intertemporal transfer of • Market structure (set of tradable assets)
resources in trying to insure, say, a firm by using financial
instruments against exposure to risk (e.g. credit risk, foreign 3. Under which conditions can the existence of a market
exchange risk). Risk allocation function of financial system allows equilibrium be guaranteed in stock markets? Why are those
for sharing the risk among the participants. conditions needed?
Information production stems from the fact that prices reflect
information and once an asset is publicly traded, so when market
participants engage in trade, they exchange their beliefs, creating 4. Can you explain the worldwide reactions in global stock
in that way new information. markets on Black Monday in October 1987?

2. Which are the elementary risk categories? Explain briefly

what they mean.
Elementary risk categories:
- idiosyncratic risk
- systematic risk
- systemic risk
Firstly, we need to define risk. Risk, in this case, is an uncertainty
about the ability of the borrower to fulfill their financial
obligations to the lender.
Idiosyncratic risk is an investment risk that is characteristic of an
individual security, for example, a particular company’s stock or a
certain bond.
Systematic risk is a risk that is characteristic of the whole market,
not just particular securities and cannot be avoided by
diversification. It is caused by bigger factors, for example,
economy, interest rates, geopolitics, etc.
Systemic risk is a risk of a crisis – an exogenous shock to the
system, that can be dangerous for the stability of the whole
financial system, for example, by the collapse of the banks.

3. Which essential information is missing on the basic return-

risk diagram? Explain, why this information is important.
(1.00/3.00)
higher risk does not always mean higher returns! Higher potential
returns could also lead to higher potential losses.
(1.00/3.00)
How returns of investments covary with each other

Quizzes W2018-19:
Quiz 2

1. State carefully the basic asset pricing equation for 1 asset.

2. Why does the basic asset pricing equation require the existence of a market equilibrium in the asset market?

3. Define a market equilibrium.

4. Why is the protection of property rights important for market equilibrium?

5. Which problems may arise in asset pricing if some investors are not globally risk averse?

⇒←↑→↓⇔