Supply Contracts with Financial Hedging

Similar documents
A Cournot-Stackelberg Model of Supply Contracts with Financial Hedging

A Cournot-Stackelberg Model of Supply Contracts with Financial Hedging

An overview of some financial models using BSDE with enlarged filtrations

Tangent Lévy Models. Sergey Nadtochiy (joint work with René Carmona) Oxford-Man Institute of Quantitative Finance University of Oxford.

MARTINGALES AND LOCAL MARTINGALES

International Monetary Theory: Mundell Fleming Redux

Martingale Pricing Theory in Discrete-Time and Discrete-Space Models

Exponential utility maximization under partial information

Credit Risk Models with Filtered Market Information

Structural Models of Credit Risk and Some Applications

Spot and forward dynamic utilities. and their associated pricing systems. Thaleia Zariphopoulou. UT, Austin

Models of Directed Search - Labor Market Dynamics, Optimal UI, and Student Credit

MATH3075/3975 FINANCIAL MATHEMATICS TUTORIAL PROBLEMS

RMSC 4005 Stochastic Calculus for Finance and Risk. 1 Exercises. (c) Let X = {X n } n=0 be a {F n }-supermartingale. Show that.

Lecture 4. Finite difference and finite element methods

Appendix: Common Currencies vs. Monetary Independence

Information Aggregation in Dynamic Markets with Strategic Traders. Michael Ostrovsky

Optimal stopping problems for a Brownian motion with a disorder on a finite interval

MKTG 555: Marketing Models

Optimal Investment for Worst-Case Crash Scenarios

Local Volatility Dynamic Models

1.1 Basic Financial Derivatives: Forward Contracts and Options

Enhancing Insurer Value Via Reinsurance Optimization

Foreign Competition and Banking Industry Dynamics: An Application to Mexico

The Role of Financial Services in Procurement Contracts

Utility Indifference Pricing and Dynamic Programming Algorithm

Stochastic modelling of electricity markets Pricing Forwards and Swaps

Exponential utility maximization under partial information and sufficiency of information

Comprehensive Exam. August 19, 2013

What Can Rational Investors Do About Excessive Volatility and Sentiment Fluctuations?

Prospect Theory, Partial Liquidation and the Disposition Effect

Part 2: Monopoly and Oligopoly Investment

PDE Approach to Credit Derivatives

Optimal robust bounds for variance options and asymptotically extreme models

STATE UNIVERSITY OF NEW YORK AT ALBANY Department of Economics. Ph. D. Preliminary Examination: Macroeconomics Fall, 2009

Robust Pricing and Hedging of Options on Variance

Regression estimation in continuous time with a view towards pricing Bermudan options

A model for a large investor trading at market indifference prices

Robust Portfolio Choice and Indifference Valuation

Microeconomics II. CIDE, MsC Economics. List of Problems

Dynamic Replication of Non-Maturing Assets and Liabilities

Evaluating Strategic Forecasters. Rahul Deb with Mallesh Pai (Rice) and Maher Said (NYU Stern) Becker Friedman Theory Conference III July 22, 2017

IEOR E4703: Monte-Carlo Simulation

Equity correlations implied by index options: estimation and model uncertainty analysis

Real Option Analysis for Adjacent Gas Producers to Choose Optimal Operating Strategy, such as Gas Plant Size, Leasing rate, and Entry Point

Market Design for Emission Trading Schemes

Hedging Under Jump Diffusions with Transaction Costs. Peter Forsyth, Shannon Kennedy, Ken Vetzal University of Waterloo

Luca Taschini. 6th Bachelier World Congress Toronto, June 25, 2010

Introduction to Real Options

Auctions 1: Common auctions & Revenue equivalence & Optimal mechanisms. 1 Notable features of auctions. use. A lot of varieties.

Answers to June 11, 2012 Microeconomics Prelim

Math 152: Applicable Mathematics and Computing

Parametric Inference and Dynamic State Recovery from Option Panels. Torben G. Andersen

Sample Path Large Deviations and Optimal Importance Sampling for Stochastic Volatility Models

Parameters Estimation in Stochastic Process Model

CONTINUOUS TIME PRICING AND TRADING: A REVIEW, WITH SOME EXTRA PIECES

Principal-Agent Problems in Continuous Time

Hedging under Arbitrage

Last Time. Martingale inequalities Martingale convergence theorem Uniformly integrable martingales. Today s lecture: Sections 4.4.1, 5.

Non-semimartingales in finance

Part 1: q Theory and Irreversible Investment

A discretionary stopping problem with applications to the optimal timing of investment decisions.

Introduction to Probability Theory and Stochastic Processes for Finance Lecture Notes

The Macroeconomics of Shadow Banking. January, 2016

Dynamic Portfolio Choice II

Optimal Credit Limit Management

Financial Crises, Dollarization and Lending of Last Resort in Open Economies

AMH4 - ADVANCED OPTION PRICING. Contents

Sensitivity Analysis on Long-term Cash flows

M5MF6. Advanced Methods in Derivatives Pricing

Problem Set: Contract Theory

Kyle Model Governance Atoms Bonds. Presence of an Informed Trader and Other Atoms in Kyle Models. Kyle Model Governance Atoms Bonds

All questions are weighted equally, and each roman numeral is weighted equally within each question. log(q i ) pq i + w i, max. pq j c 2 q2 j.

"Pricing Exotic Options using Strong Convergence Properties

Optimal Order Placement

Lecture 17. The model is parametrized by the time period, δt, and three fixed constant parameters, v, σ and the riskless rate r.

Effectiveness of CPPI Strategies under Discrete Time Trading

Credible Threats, Reputation and Private Monitoring.

Stock Loan Valuation Under Brownian-Motion Based and Markov Chain Stock Models

Noncooperative Oligopoly

Optimal Acquisition of a Partially Hedgeable House

Stochastic Control for Optimal Trading: State of Art and Perspectives (an attempt of)

Real Options and Free-Boundary Problem: A Variational View

Optimally Thresholded Realized Power Variations for Lévy Jump Diffusion Models

Evaluation of proportional portfolio insurance strategies

Outline. and Governance. Presence of an Informed Trader and Other Atoms in Kyle Models. Stock and Bond Volume. Kyle Model Governance Atoms Bonds

Martingale Pricing Applied to Dynamic Portfolio Optimization and Real Options

The Valuation of Bermudan Guaranteed Return Contracts

Hedging Credit Derivatives in Intensity Based Models

DUOPOLY. MICROECONOMICS Principles and Analysis Frank Cowell. July 2017 Frank Cowell: Duopoly. Almost essential Monopoly

Risk Sensitive Inventory Management with Financial Hedging 1

Optimal investments under dynamic performance critria. Lecture IV

Logarithmic derivatives of densities for jump processes

Illiquidity, Credit risk and Merton s model

Summing Up Social Dilemmas

Lecture 7: Bayesian approach to MAB - Gittins index

CUR 412: Game Theory and its Applications, Lecture 9

Problem Set: Contract Theory

Homework 1: Basic Moral Hazard

1 Consumption and saving under uncertainty

Transcription:

Supply Contracts with Financial Hedging René Caldentey Martin Haugh Stern School of Business NYU Integrated Risk Management in Operations and Global Supply Chain Management: Risk, Contracts and Insurance June 006.

Motivation Financial Market Procurement Suppliers' Management Inventory/Production Capacity/Processes Product Design Corporation Distribution Pricing Policies Demand Forecast Problem: Select an operational strategy and a financial strategy that optimize the value of the corporation. Supply Contracts with Financial Hedging

Motivation A Methodology that effectively integrates Operational and Financial activities by Maximizing the Economic Value of the corporation. Modeling the risk preferences of decision makers. Using financial markets for hedging purposes. Recognizing the incompleteness of the problem. Taking advantage of all available information. max {Economic Value} subject to and subject to problem dynamics risk management constraint(s). Operational risk in customers taste, machine breakdowns, unreliable suppliers, etc... Financial risk in exchange rates, commodity prices, interest rates, etc... Supply Contracts with Financial Hedging 3

Outline of Talk The Modelling Framework - Motivation for maximizing economic value of operating profits. - Financial market & informational structure Supply Contracts with Financial Hedging - The Wholesale Price Contract with Budget Constraint Conclusions & Further Research Supply Contracts with Financial Hedging 4

Modelling Framework Financial Market & Information Supply Contracts with Financial Hedging 5

Motivating the Modelling Framework Consider two series cash-flows arising from two operating policies I 1 and I : C (1) (I 1 ) = (c (1) 1 (I 1), c (1) (I 1),..., c (1) n (I 1)) C () (I ) = (c () 1 (I ), c () (I ),..., c () n (I )) If probability distribution of C (1) and C () identical then most operations models are indifferent between the two. However, if C (1) positively correlated with financial market for example, and C () negatively correlated with financial market then economic value of C () greater than economic value of C (1). models should take this into account as they get ever more complex and include additional sources of uncertainty Do so by using an equivalent martingale measure, Q, or stochastic discount factor consistent with only a hedging motivation for financial trading Supply Contracts with Financial Hedging 6

The Modelling Framework H(I) the time T cumulative payoff from some operating policy, I The operating policy may be static or dynamic. G(θ) the time T gain from a self-financing trading strategy, θ Operational and financial hedging problem is min I,G(θ) EQ [H(I)] subject to and subject to problem dynamics risk management constraint(s) 8 < : R(H(I), G(θ)) 0 a.s. E P [R(H(I), G(θ))] 0. consistent with only a hedging motivation for financial trading G(θ) a martingale under Q so E Q [G(θ)] = 0 assuming initial capital assigned to hedging strategy is zero Supply Contracts with Financial Hedging 7

Financial Market Model Probability space (Ω, F, P). (W 1t, W t ) a standard Brownian motion. A single risky stock with price dynamics dx t = µx t dt + σx t dw 1t. Cash account available with r 0. F t is the filtration generated by (W 1t, W t ). F X t is the filtration generated by X t. Operating profits, H(I) F T, a function of (W 1t, W t ). The gain process, G(θ) t of a self-financing trading strategy θ t is G(θ) t = Z t 0 θ s dx s. An equivalent martingale measure, Q, under which X t and G(θ) t are martingales. Supply Contracts with Financial Hedging 8

Information In general, operating profits are function of (X t, W t ), that is, H(I) F T. Incomplete Information: Demand, product quality, customer tastes not always observable. Only information related to X t is available. Trading strategies must satisfy: θ t F X t, t [0, T ]. Complete Information: The evolution of X t and B t are both available. Trading strategies satisfy: θ t F t, t [0, T ]. Supply Contracts with Financial Hedging 9

Supply Contracts with Financial Hedging The Wholesale Price Contract with Budget Constaint Supply Contracts with Financial Hedging 10

The Wholesale Price Contract Manufacturer Production Cost: C(q) (1): w (): q Retailer Revenue: P(q) q Clearance Price (3): P(q) Non-Cooperative Operation: Solution Concept: Stackelberg game. 1. At time t = 0, Manufacturer offers a wholesale price w to the Retailer.. At time t = 0, Retailer orders a quantity q. 3. At time t = T, a random clearance price P (q) is realized: P (q) = A q. The random price A is correlated to the financial market X t. Retailer operates under a budget constraint: w q B. Supply Contracts with Financial Hedging 11

Types of Contracts (w,q) contract is decided Clearance Price P(q) is realized Payoffs are determined Production takes place t = 0 Simple Contract t = T time (w τ, q τ ) contract is decided Market Signal X τ is observed Clearance Price P(q) is realized Payoffs are determined Production takes place t = 0 t = τ t = T time Flexible Contract (w τ, q τ ) contract is decided Market Signal X τ is observed Trading Gain G τ is observed Clearance Price P(q) is realized Payoffs are determined Financial hedging takes place Production takes place t = 0 t = τ t = T time Flexible Contract with Financial Hedging Remark: The time τ can be a decision variable: a deterministic time or a F X τ -stopping time. Supply Contracts with Financial Hedging 1

Some Notation ) (Ω, F, Q) probability space. ) X t : Time t value of a Q-martingale tradable security (t [0, T ]). ) F t = σ(x s ; 0 s t) filtration generated by X t with F T F. ) E Q τ [E] := EQ [E F τ ] for all E F. ) P (q) = A q: retail price at time T with A F. ) Āτ = E Q [A F τ ] for any F t -stopping time τ T. ) c τ : per unit manufacturing cost at time τ [0, T ]. ) w τ F τ : manufacturer s wholesale price menu. ) q τ F τ : retailer s ordering quantity menu. ) G τ F τ : retailer s trading gains (or losses) with E Q [G τ ] = 0. ) Budget constraint: w τ q τ B τ := B + G τ. Assumption. For all τ [0, T ], Ā τ c τ. Supply Contracts with Financial Hedging 13

Flexible Contract Consider a fixed τ [0, T ]. Step 1: At t = 0, the manufacturer offers a wholesale menu w τ F τ. Step : A retailer, with no access to the financial market, decides the ordering level q τ F τ solving n o Π F R (w τ) = E Q 0 max E Q τ [(A q τ w τ ) q τ ] qτ 0 subject to w τ q τ B, for all ω Ω. The optimal solution (retailer s reaction) is q(w τ ) = min(āτ w τ +, ) B. w τ Step 3: The manufacturer problem is (Stackelberg leader) Π F M = EQ 0 max wτ cτ {(w τ c τ ) q τ (w τ )}. Supply Contracts with Financial Hedging 14

Flexible Contract Proposition. (Flexible Contract Solution) Under Assumption, the equilibrium solution for the flexible contract is where w F τ = Āτ + δ F τ δ F τ := max c τ, and q F τ = Āτ δ F τ 4 q(āτ 8 B)+., The equilibrium expected payoffs of the players are then given by Π F M τ = (Āτ + δ F τ c τ) (Āτ δ F τ ) 8 and Π F R τ = (Āτ δ F τ ). 16 Remarks: -) δ F τ is a modified production cost ( c τ) that is induced by a limited budget B. -) w F τ decreases in B and q F τ, Π F M τ and Π F R τ increase in B. Supply Contracts with Financial Hedging 15

Flexible Contract vs. Simple Contract Define w F := E Q 0 [w F τ ], q F := E Q 0 [q F τ ], Π F M := EQ 0 [Π F M τ ] and Π F R := EQ 0 [Π F R τ ]. Proposition. Suppose that B Ā τ c τ 8 almost surely. Then w F w S, q F q S, Π F M Π S M and Π F R Π S R. However, if B max Ā τ c τ 8, Ā c 0 8 almost surely then w F = w S + c τ c 0 and q F = q S c τ c 0 4 and Π F M Π S M and Π F R Π S R if and only if E Q 0 [(Āτ c τ ) ] (Ā c 0). Supply Contracts with Financial Hedging 16

Flexible Contract vs. Simple Contract Wholesale Price Ordering Level 0.5 Flexible vs. Simple 1.8 1.6 1.4 Flexible 0.4 0.3 0. Simple Flexible 1.5 1 Π R F /ΠR S Π P F /ΠP S Case 1 1. Simple 1 0 0.6 1. 0.1 0 0 0.6 1. 0.5 0 0.6 1. 1.8 1.6 1.4 1. Simple Flexible 0.5 0.4 0.3 0. 0.1 Simple Flexible 1.6 1.4 1. 1 0.8 0.6 0.4 Π R F /ΠR S Π P F /ΠP S Case 1 0 0.6 1. 0 0 0.6 1. Budget (B) 0. 0 0.6 1. Ā τ Uniform[1, 3], c 0 = 0.3, c τ = 0.35 (case 1) and c τ = 0.7 (case ). Supply Contracts with Financial Hedging 17

Flexible Contract: Efficiency 0.55 Q F τ 3.5 W F τ 0.7 P F τ 0.5 0.6 0.45 3 0.5 A τ 3 c τ 0.4.5 0.4 0.35 0.3 0.3 0 0.5 1 0 0.5 1 0. 0 0.5 1 0.8 1.8 0.4 0.75 0.7 1.7 0.35 0.65 1.6 0.3 0.6 1.5 0.5 A τ 3 c τ 0.55 0.5 1.4 0. 0.45 0 0.5 1 1.3 0 0.5 1 Budget (B) 0.15 0 0.5 1 Ā τ =, c τ = 0.6 (top) and c τ = 1. (bottom). Supply Contracts with Financial Hedging 18

Flexible Contract with Financial Hedging (w τ, q τ ) contract is decided Market Signal X τ is observed Trading Gain G τ is observed Clearance Price P(q) is realized Payoffs are determined Financial hedging takes place Production takes place t = 0 t = τ t = T time Flexible Contract with Financial Hedging Step 1: At t = 0, and for a fixed τ T, the manufacturer offers a price menu w τ F X τ. Step : In response, at t = 0, the retailer selects an optimal ordering menu q τ (w τ) F X τ solving Π H R (w τ) = max qτ 0, Gτ E Q [(A q τ ) q τ w τ q τ ] subject to w τ q τ B + G τ, for all ω Ω E Q [G τ ] = 0. Step 3: The manufacturer selects the optimal wholesale price menu w τ solving Π H M (w τ) = max wτ E Q wτ q τ (w τ) c τ q τ (w τ). Supply Contracts with Financial Hedging 19

Flexible Contract with Financial Hedging Proposition. (Retailer s Optimal Strategy) Let Q τ, X and X c be defined as follows Q τ Āτ w τ +, X {ω Ω : B Q τ w τ }, and X c Ω X. Case 1: Suppose that E Q [Q τ w τ ] B. Then q τ (w τ) = Q τ and there are infinitely many choices of the optimal claim, G τ. One natural choice is to take G τ = [Q τ w τ B] δ if ω X 1 if ω X c δ R R X c [Q τ w τ B] dq X [B Q τ w τ ] dq. Remark: In this case, it is possible to completely eliminate the budget constraint by trading in the financial market. Supply Contracts with Financial Hedging 0

Flexible Contract with Financial Hedging Proposition. (Continuation) Case : Suppose that B < E Q [Q τ w τ ]. Then Āτ w τ + (1 + λ) Āτ q τ (w τ ) = where λ 0 solves E "w Q w τ (1 + λ) τ + # = B. Proposition. Let φ inf (Producer s Optimal Strategy and the Stackelberg Solution) (φ 1 such that E Q " Ā τ (φ c τ ) 8 + # B ). Then, w τ = Āτ +φ cτ and q Āτ τ = φ + cτ 4 and the players expected payoffs satisfy Π H M τ = (Āτ + φ c τ c τ ) (Āτ φ c τ ) + 8 and Π H R τ = ((Āτ φ c τ ) + ). 16 Remark: When q τ = 0, the manufacturer decides to overcharge the retailer making the supply chain non-operative. This is never the case if the retailer does not have not access to the financial market. Supply Contracts with Financial Hedging 1

Flexible Contract with Financial Hedging Proposition. The manufacturer always prefers the H-contract to the F-Contract. On the other hand, the retailer s preferences are Undetermined H-Contract H-Contract H-Contract = F-Contract H-Contract or F-Contract Small Budget Retailer's Preferences Large Budget.4. 1.8 1.6 1.4 1. 0.35 0.3 0.5 0. 0.15 0.1 0.05 Wholesale Price H Contract F Contract 0 0.5 1 1.5 Budget (B) Producer s Payoff H Contract F Contract 0 0 0.5 1 1.5 Budget (B) 0.4 0.35 0.3 0.5 0. 0.15 0.1 0.05 Ordering Level 0 0 0.5 1 1.5 Budget (B) 0.18 0.16 0.14 0.1 0.1 0.08 0.06 0.04 0.0 H Contract F Contract Retailer s Payoff H Contract F Contract 0 0 0.5 1 1.5 Budget (B) Supply Contracts with Financial Hedging

Flexible Contract with Financial Hedging Efficiency On path-by-path basis, the Centralized system is not necessarily more efficient than the Decentralized Supply Chain! ω Ω such that q H C τ = 0 and q H τ > 0. Remark: This is never the case under a Flexible Contract without Hedging. On average, the Centralized solution is more efficient than the Decentralized solution. E Q 0 [q H C τ] E Q 0 [q H τ ]. Supply Contracts with Financial Hedging 3

Optimal Production Postponement The manufacturer can choose the optimal time to execute the contract solving subject to Π H P = max τ E Q 0 φ = inf [ ( Ā τ + φ c τ c τ ) (Āτ φ c τ ) + ] { ψ 1 : E Q 0 8 ξ [ (Ā τ ψ c τ 8 ξ ) + ] B }. Two possibilities: - Open Loop: τ is a deterministic time selected at time t = 0. - Closed Loop: τ is an F t -stopping time. Modeling Assumptions: Financial Market X t is a diffusion process dx t = σ(x t ) dw t. Operations Additive: A = F (X T ) + ε, E Q [ε] = 0, or Multiplicative: A = ε F (X T ), ε 0 and E Q [ε] = 1 Production: c τ = c 0 + α τ κ, for all τ [0, 1]. Supply Contracts with Financial Hedging 4

Optimal Open-Loop Production Postponement 8.5 8 κ = 8 7.5 7 Profits 6.5 κ =1 κ = 4 6 5.5 5 κ = 0.5 0 0. 0.4 0.6 0.8 1 Time (τ) Optimal open-loop production postponement for four different production cost functions. X 0 = 1, σ(x) = X, F (X) = + X and c τ = 0.3 + 0.7 τ κ. Supply Contracts with Financial Hedging 5

Optimal Close-Loop Production Postponement 14 κ = 8 7 κ = 4 1 6 X(τ) 10 8 6 ρ(t) X(τ) 5 4 3 ρ(t) 4 Optimal 0 0 0. 0.4 0.6 0.8 1 Time (τ) 5 κ = 1 1 0 0 0. 0.4 0.6 0.8 1 Time (τ) 10 κ = 0.5 Optimal X(τ) 4 3 ρ(t) Optimal X(τ) 8 6 4 ρ(t) Optimal 1 0 0. 0.4 0.6 0.8 1 Time (τ) 0 0 0. 0.4 0.6 0.8 1 Time (τ) Optimal continuation region for four different manufacturing cost functions. X 0 = 1, σ(x) = X, F (X) = + X and c τ = 0.3 + 0.7 τ κ. Supply Contracts with Financial Hedging 6

Optimal Close-Loop vs. Optimal Open-Loop Production Postponement κ Open-Loop Payoff Closed-Loop Payoff % Increase 0.5 7.9 7.9 0.0% 1 7.9 7.305 0.% 4 7.71 7.99 3.7% 8 8.09 8.33 3.8% Optimal continuation region for four different manufacturing cost functions. X 0 = 1, σ(x) = X, F (X) = + X and c τ = 0.3 + 0.7 τ κ. Supply Contracts with Financial Hedging 7

Summary Simple extension to the traditional wholesale contract. The proposed procurement contracts uses the Financial Market as: A source of public information upon which contracts can be written. A means for financial hedging to mitigate the impact of the budget constraint. Consistent with the notions of production postponement and demand forecast. Managerial Insights: Manufacturer and Retailer incentives are not always aligned as a function of B. Manufacturer prefers retailers that have access to the financial market. With hedging, the supply chain might not operate in some states ω Ω. In some cases, financial hedging eliminates the budget constraint. Optimal time τ of the contract balances Var(Āτ) and c τ. Extensions: Other types of contracts: quantity discount, buy-back, etc. Include other sources of uncertainty: exchange rates, interest rates, credit risk. Supply Contracts with Financial Hedging 8