Intertemporal Asset Pricing Theory Darrell Duffie Stanford Universityl Draft:July 4,2002 Contents 1 Introduction 3 2 Basic Theory 4 2.1 Setup...········· 4 2.2 Arbitrage,State Prices,and Martingales 5 2.3 Individual Agent Optimality 8 2.4 Habit and Recursive Utilities... 9 2.5 Equilibrium and Pareto Optimality 12 2.6 Equilibrium Asset Pricing...... 14 2.7 Breeden's Consumption-Based CAPM 16 2.8 Arbitrage and Martingale Measures 17 2.9 Valuation of Redundant Securities... 19 2.10 American Exercise Policies and Valuation........ 21 3 Continuous-Time Modeling 26 3.1 Trading Gains for Brownian Prices 26 3.2 Martingale Trading Gains...... 28 3.3 The Black-Scholes Option-Pricing Formula 30 3.4Ito's Formula..........····· 34 3.5 Arbitrage Modeling.···..········ 36 3.6 Numeraire Invariance............... 37 3.7 State Prices and Doubling Strategies.... 37 II am grateful for impetus from George Constantinides and Rene Stulz,and for inspi- ration and guidance from many collaborators and Stanford colleagues.Address:Grad- uate School of Business,Stanford University,Stanford CA 94305-5015 USA;or email at duffie@stanford.edu.The latest draft can be downloaded at www.stanford.edu/~duffie/. Some portions of this survey are revised from original material in Dynamic Asset Pricing Theory,Third Edition,copyright Princeton University Press,2002. 1
Intertemporal Asset Pricing Theory Darrell Duffie Stanford University1 Draft: July 4, 2002 Contents 1 Introduction 3 2 Basic Theory 4 2.1 Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 2.2 Arbitrage, State Prices, and Martingales . . . . . . . . . . . . 5 2.3 Individual Agent Optimality ................... 8 2.4 Habit and Recursive Utilities ................... 9 2.5 Equilibrium and Pareto Optimality . . . . . . . . . . . . . . . 12 2.6 Equilibrium Asset Pricing . . . . . . . . . . . . . . . . . . . . 14 2.7 Breeden’s Consumption-Based CAPM . . . . . . . . . . . . . 16 2.8 Arbitrage and Martingale Measures . . . . . . . . . . . . . . . 17 2.9 Valuation of Redundant Securities . . . . . . . . . . . . . . . . 19 2.10 American Exercise Policies and Valuation . . . . . . . . . . . . 21 3 Continuous-Time Modeling 26 3.1 Trading Gains for Brownian Prices . . . . . . . . . . . . . . . 26 3.2 Martingale Trading Gains . . . . . . . . . . . . . . . . . . . . 28 3.3 The Black-Scholes Option-Pricing Formula . . . . . . . . . . . 30 3.4 Ito’s Formula . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 3.5 Arbitrage Modeling . . . . . . . . . . . . . . . . . . . . . . . . 36 3.6 Numeraire Invariance . . . . . . . . . . . . . . . . . . . . . . . 37 3.7 State Prices and Doubling Strategies . . . . . . . . . . . . . . 37 1I am grateful for impetus from George Constantinides and Ren´e Stulz, and for inspiration and guidance from many collaborators and Stanford colleagues. Address: Graduate School of Business, Stanford University, Stanford CA 94305-5015 USA; or email at duffie@stanford.edu. The latest draft can be downloaded at www.stanford.edu/∼duffie/. Some portions of this survey are revised from original material in Dynamic Asset Pricing Theory, Third Edition, copyright Princeton University Press, 2002. 1
3.8 Equivalent Martingale Measures... 38 3.9 Girsanov and Market Prices of Risk.·..........·.· 39 3.10 Black-Scholes Again 43 3.11 Complete Markets .. 44 3.l2 Optimal Trading and Consumption.....·,...····. 46 3.13 Martingale Solution to Merton's Problem............ 50 4 Term-Structure Models 54 4.1 One-Factor Models ... 55 4.2 Term-Structure Derivatives..... 60 4.3 Fundamental Solution ..................... 63 4.4 Multifactor Term-Structure Models....····.· 64 4.5 Affine Models...·.....·....··········· 66 4.6 The HJM Model of Forward Rates..·.···..····.· 69 5 Derivative Pricing 73 5.1 Forward and Futures Prices.... 73 5.2 Options and Stochastic Volatility 76 5.3 Option Valuation by Transform Analysis 80 6 Corporate Securities 84 6.1 Endogenous Default Timing........·.....···.. 85 6.2 Example:Brownian Dividend Growth.... 87 6.3 Taxes,Bankruptcy Costs,.Capital Structure·.....···. 91 6.4 Intensity-Based Modeling of Default.·.......····.· 93 6.5 Zero-Recovery Bond Pricing................... 96 6.6 Pricing with Recovery at Default ....... 98 6.7 Default-Adjusted Short Rate............... 99 2
3.8 Equivalent Martingale Measures . . . . . . . . . . . . . . . . . 38 3.9 Girsanov and Market Prices of Risk . . . . . . . . . . . . . . . 39 3.10 Black-Scholes Again . . . . . . . . . . . . . . . . . . . . . . . 43 3.11 Complete Markets . . . . . . . . . . . . . . . . . . . . . . . . 44 3.12 Optimal Trading and Consumption . . . . . . . . . . . . . . . 46 3.13 Martingale Solution to Merton’s Problem . . . . . . . . . . . . 50 4 Term-Structure Models 54 4.1 One-Factor Models . . . . . . . . . . . . . . . . . . . . . . . . 55 4.2 Term-Structure Derivatives . . . . . . . . . . . . . . . . . . . . 60 4.3 Fundamental Solution . . . . . . . . . . . . . . . . . . . . . . 63 4.4 Multifactor Term-Structure Models . . . . . . . . . . . . . . . 64 4.5 Affine Models . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 4.6 The HJM Model of Forward Rates . . . . . . . . . . . . . . . 69 5 Derivative Pricing 73 5.1 Forward and Futures Prices . . . . . . . . . . . . . . . . . . . 73 5.2 Options and Stochastic Volatility . . . . . . . . . . . . . . . . 76 5.3 Option Valuation by Transform Analysis . . . . . . . . . . . . 80 6 Corporate Securities 84 6.1 Endogenous Default Timing . . . . . . . . . . . . . . . . . . . 85 6.2 Example: Brownian Dividend Growth . . . . . . . . . . . . . . 87 6.3 Taxes, Bankruptcy Costs, Capital Structure . . . . . . . . . . 91 6.4 Intensity-Based Modeling of Default . . . . . . . . . . . . . . . 93 6.5 Zero-Recovery Bond Pricing . . . . . . . . . . . . . . . . . . . 96 6.6 Pricing with Recovery at Default . . . . . . . . . . . . . . . . 98 6.7 Default-Adjusted Short Rate . . . . . . . . . . . . . . . . . . . 99 2
1 Introduction This is a survey of "classical"intertemporal asset pricing theory.A central objective of this theory is to reduce asset-pricing problems to the identifica- tion of "state prices,"a notion of Arrow [1953 from which any security has an implied value as the weighted sum of its future cash flows,state by state, time by time,with weights given by the associated state prices.Such state prices may be viewed as the marginal rates of substitution among state-time consumption opportunities,for any unconstrained investor,with respect to a numeraire good.Under many types of market imperfections,state prices may not exist,or may be of relatively less use or meaning.While market im- perfections constitute an important thrust of recent advances in asset pricing theory,they will play a limited role in this survey,given the limitations of space and the priority that should be accorded to first principles based on perfect markets. Section 2 of this survey provides the conceptual foundations of the broader theory in a simple discrete-time setting.After extending the basic modeling approach to a continuous-time setting in Section 3,we turn in Section 4 to term-structure modeling,in Section 5 to derivative pricing,and in Section 6 to corporate securities. The theory of optimal portfolio and consumption choice is closely linked to the theory of asset pricing,for example through the relationship between state prices and marginal rates of substitution at optimality.While this connection is emphasized,for example in Sections 2.3-2.4 and 3.12-3.13,the theory of optimal portfolio and consumption choice,particularly in dynamic incomplete-markets settings,has become so extensive as to defy a proper summary in the context of a reasonably sized survey of asset-pricing theory. The interested reader is especially directed to the treatments of Karatzas and Shreve [1998,Schroder and Skiadas [1999],and Schroder and Skiadas [20001. For ease of reference,as there is at most one theorem per sub-section,we refer to a theorem by its subsection number,and likewise for lemmas and propositions.For example,the unique proposition of Section 2.9 is called “Proposition2.9.” 3
1 Introduction This is a survey of “classical” intertemporal asset pricing theory. A central objective of this theory is to reduce asset-pricing problems to the identification of “state prices,” a notion of Arrow [1953] from which any security has an implied value as the weighted sum of its future cash flows, state by state, time by time, with weights given by the associated state prices. Such state prices may be viewed as the marginal rates of substitution among state-time consumption opportunities, for any unconstrained investor, with respect to a numeraire good. Under many types of market imperfections, state prices may not exist, or may be of relatively less use or meaning. While market imperfections constitute an important thrust of recent advances in asset pricing theory, they will play a limited role in this survey, given the limitations of space and the priority that should be accorded to first principles based on perfect markets. Section 2 of this survey provides the conceptual foundations of the broader theory in a simple discrete-time setting. After extending the basic modeling approach to a continuous-time setting in Section 3, we turn in Section 4 to term-structure modeling, in Section 5 to derivative pricing, and in Section 6 to corporate securities. The theory of optimal portfolio and consumption choice is closely linked to the theory of asset pricing, for example through the relationship between state prices and marginal rates of substitution at optimality. While this connection is emphasized, for example in Sections 2.3-2.4 and 3.12-3.13, the theory of optimal portfolio and consumption choice, particularly in dynamic incomplete-markets settings, has become so extensive as to defy a proper summary in the context of a reasonably sized survey of asset-pricing theory. The interested reader is especially directed to the treatments of Karatzas and Shreve [1998], Schroder and Skiadas [1999], and Schroder and Skiadas [2000]. For ease of reference, as there is at most one theorem per sub-section, we refer to a theorem by its subsection number, and likewise for lemmas and propositions. For example, the unique proposition of Section 2.9 is called “Proposition 2.9.” 3
2 Basic Theory Radner [1967 and Radner [1972 originated our standard approach to a dy- namic equilibrium of "plans,prices,and expectations,"extending the static approach of Arrow [1953]and Debreu ([1953].2 After formulating this stan- dard model,this section provides the equivalence of no arbitrage and state prices,and shows how state prices may be derived from investors'marginal rates of substitution among state-time consumption opportunities.Given state prices,we examine pricing derivative securities,such as European and American options,whose payoffs can be replicated by trading the underlying primitive securities. 2.1 Setup We begin for simplicity with a setting in which uncertainty is modeled as some finite set of states,with associated probabilities.We fix a set F of events,called a tribe,also known as a o-algebra,which is the collection of subsets of n that can be assigned a probability.The usual rules of probability apply.3 We let P(A)denote the probability of an event A. There are T+1 dates:0,1,...,T.At each of these,a tribeFF is the set of events corresponding to the information available at time t. Any event in F is known at time t to be true or false.We adopt the usual convention that Fc Fs whenever t s,meaning that events are never "forgotten."For simplicity,we also take it that events in Fo have probability 0 or 1,meaning roughly that there is no information at time t =0.Taken altogether,the filtration F={Fo,...,Fr,sometimes called an information structure,represents how information is revealed through time.For any random variable Y,we let E(Y)=E(YF)denote the conditional expectation of Y given Ft.In order to simplify things,for any two random variables Y and Z,we always write "Y=2"if the probability that Y≠Z is zero. An adapted process is a sequence X={Xo,...,Xr}such that,for each t,X:is a random variable with respect to (F).Informally,this means 2The model of Debreu [1953]appears in Chapter 7 of Debreu [1959].For more details in a finance setting,see Dothan [1990].The monograph by Magill and Quinzii [1996]is a comprehensive survey of the theory of general equilibrium in a setting such as this. 3The triple (F,P)is a probability space,as defined for example by Jacod and Protter [2000]
2 Basic Theory Radner [1967] and Radner [1972] originated our standard approach to a dynamic equilibrium of “plans, prices, and expectations,” extending the static approach of Arrow [1953] and Debreu [1953].2 After formulating this standard model, this section provides the equivalence of no arbitrage and state prices, and shows how state prices may be derived from investors’ marginal rates of substitution among state-time consumption opportunities. Given state prices, we examine pricing derivative securities, such as European and American options, whose payoffs can be replicated by trading the underlying primitive securities. 2.1 Setup We begin for simplicity with a setting in which uncertainty is modeled as some finite set Ω of states, with associated probabilities. We fix a set F of events, called a tribe, also known as a σ-algebra, which is the collection of subsets of Ω that can be assigned a probability. The usual rules of probability apply.3 We let P(A) denote the probability of an event A. There are T + 1 dates: 0, 1,...,T. At each of these, a tribe Ft ⊂ F is the set of events corresponding to the information available at time t. Any event in Ft is known at time t to be true or false. We adopt the usual convention that Ft ⊂ Fs whenever t ≤ s, meaning that events are never “forgotten.” For simplicity, we also take it that events in F0 have probability 0 or 1, meaning roughly that there is no information at time t = 0. Taken altogether, the filtration F = {F0,..., FT }, sometimes called an information structure, represents how information is revealed through time. For any random variable Y , we let Et(Y ) = E(Y | Ft) denote the conditional expectation of Y given Ft. In order to simplify things, for any two random variables Y and Z, we always write “Y = Z” if the probability that Y 6= Z is zero. An adapted process is a sequence X = {X0,...,XT } such that, for each t, Xt is a random variable with respect to (Ω, Ft). Informally, this means 2The model of Debreu [1953] appears in Chapter 7 of Debreu [1959]. For more details in a finance setting, see Dothan [1990]. The monograph by Magill and Quinzii [1996] is a comprehensive survey of the theory of general equilibrium in a setting such as this. 3The triple (Ω, F, P) is a probability space, as defined for example by Jacod and Protter [2000]. 4
that Xt is observable at time t.An adapted process X is a martingale if,for any times t and s >t,we have Et(Xs)=Xt. A security is a claim to an adapted dividend process,say 6,with 6t denot- ing the dividend paid by the security at time t.Each security has an adapted security-price process S,so that St is the price of the security,ex dividend,at time t.That is,at each time t,the security pays its dividend ot and is then available for trade at the price St.This convention implies that do plays no role in determining ex-dividend prices.The cum-dividend security price at time t is S:+6t. We suppose that there are N securities defined by an RN-valued adapted dividend process 6=(6(1),...,(N)).These securities have some adapted price process S=(S(),...,S(N)).A trading strategy is an adapted process 0 in RN.Here,represents the portfolio held after trading at time t.The dividend process 6 generated by a trading strategy 0 is defined by 69=0-1…(S:+d)-0·S, (1) with“o_l”taken to be zero by convention. 2.2 Arbitrage,State Prices,and Martingales Given a dividend-price pair (S)for N securities,a trading strategy 0 is an arbitrage if 60>0(that is,if 60>0 and 60).An arbitrage is thus a trading strategy that costs nothing to form,never generates losses,and, with positive probability,will produce strictly positive gains at some time. One of the precepts of modern asset pricing theory is a notion of efficient markets under which there is no arbitrage.This is reasonable axiom,for in the presence of an arbitrage,any rational investor who prefers to increase his dividends would undertake such arbitrages without limit,so markets could not be in equilibrium,in a sense that we shall see more formally later in this section.We will first explore the implications of no arbitrage for the representation of security prices in terms of "state prices,"the first step toward which is made with the following result. Proposition.There is no arbitrage if and only if there is a strictly positive adapted process n such that,for any trading strategy 0, 5
that Xt is observable at time t. An adapted process X is a martingale if, for any times t and s>t, we have Et(Xs) = Xt. A security is a claim to an adapted dividend process, say δ, with δt denoting the dividend paid by the security at time t. Each security has an adapted security-price process S, so that St is the price of the security, ex dividend, at time t. That is, at each time t, the security pays its dividend δt and is then available for trade at the price St. This convention implies that δ0 plays no role in determining ex-dividend prices. The cum-dividend security price at time t is St + δt. We suppose that there are N securities defined by an RN -valued adapted dividend process δ = (δ(1),...,δ(N) ). These securities have some adapted price process S = (S(1),...,S(N) ). A trading strategy is an adapted process θ in RN . Here, θt represents the portfolio held after trading at time t. The dividend process δθ generated by a trading strategy θ is defined by δθ t = θt−1 · (St + δt) − θt · St, (1) with “θ−1” taken to be zero by convention. 2.2 Arbitrage, State Prices, and Martingales Given a dividend-price pair (δ, S) for N securities, a trading strategy θ is an arbitrage if δθ > 0 (that is, if δθ ≥ 0 and δθ 6= 0). An arbitrage is thus a trading strategy that costs nothing to form, never generates losses, and, with positive probability, will produce strictly positive gains at some time. One of the precepts of modern asset pricing theory is a notion of efficient markets under which there is no arbitrage. This is reasonable axiom, for in the presence of an arbitrage, any rational investor who prefers to increase his dividends would undertake such arbitrages without limit, so markets could not be in equilibrium, in a sense that we shall see more formally later in this section. We will first explore the implications of no arbitrage for the representation of security prices in terms of “state prices,” the first step toward which is made with the following result. Proposition. There is no arbitrage if and only if there is a strictly positive adapted process π such that, for any trading strategy θ, E X T t=0 πtδθ t ! = 0. 5