Hybrid Approach to Option Valuation Richard de Neufville Professor of Engineering Systems and of Civil and Environmental Engineering MIT Hybrid Approach to Valuation Slide 1 of 23 Outline R & D creates options R & D Evaluation in Auto Industry Parts of R & D Process Details of Ford Case Hybrid Evaluation Strategy and Results Summary Hybrid Approach to Valuation Slide 2 of 23 1
Linear Systems are Common Many Systems feature linear development They proceed naturally through phases Example: Oil Development Acquire Lease Explore Test wells Exploitation Example: New Products Basic Research Applied Research Development Production Hybrid Approach to Valuation Slide 3 of 23 Linear Systems Create Options Completing 1 st step gives right, but not obligation to do 2 nd ; and so on down line R & D in particular creates options HOWEVER, optionality of R & D has usually been ignored projects evaluated by NPV This means: value of being able to cancel project if appropriate has been ignored In short that R & D HAS BEEN UNDERVALUED Hybrid Approach to Valuation Slide 4 of 23 2
Types of R & D Traditionally, R & D organizations divide their portfolio of into categories In Auto Industry Tier 1: General Research Tier 2: Future Vehicle Work Tier 3: Current Vehicle Work Major Development Effort In US Department of Defense Categories: 6.1, 6.2, etc Evaluation Criteria for parts may differ Hybrid Approach to Valuation Slide 5 of 23 R & D Evaluation in US Auto Industry Traditional Process Net Present Value Process Assumes Development Process carried on through production possibility of stopping project not in analysis However, successful prototype production is option for full scale Is this right? How should Ford evaluate elements of its research portfolio? This was object of Neely s research Hybrid Approach to Valuation Slide 6 of 23 3
Neely s Hybrid Approach Basic idea: Divide the valuation into 2 parts Use different approach for distinct needs Financial options analysis for market element uses volatility of stock price to reflection of uncertainty in sales and thus value of product Decision analysis for technical risks, Judgment is appropriate -- not volatility Render onto Caesar what is Caesar s unto God what is God s Hybrid Approach to Valuation Slide 7 of 23 Structure of Investment Choices Decision 1 Invest? Uncertainty Resolution Decision 2 Implement? Contingent Results Invest Do Not Invest Results Promise Results Poor 0.5 0.5 Implement High Benefits -Cost(1) -Cost(2) Stop Project -Cost(1) Implement Stop Project Low Benefits -Cost(1)-Cost(2) -Cost(1) 0 Hybrid Approach to Valuation Slide 8 of 23 4
Detailed view of Probabilities, Decisions Initial Investment Choice Yes Invest? No Technical Benefit Market Implementation Time Uncertainty Uncertainty Uncertainty Decision (Endogenous) (Endogenous) (Exogenous) -R&D Costs - Implement Cost Implement High Demand + Revenues Phd Stop -R&D Costs Implement -R&D Costs - Implement Cost High Benefits Medium Demand + Revenues Stop Pmd -R&D Costs Phb Implement -R&D Costs - Implement Cost Low Demand R&D Success + Revenues Stop Pld -R&D Costs Ps Medium Benefits Similar to High Value Branch Pmb Low Benefits Similar to High Value Branch R&D Fails Plb -R&D Costs 1-Ps 0 Hybrid Approach to Valuation Slide 9 of 23 Structure of Hybrid Analysis Risk-neutral treatment of exogenous uncertainties, that is, those due to market Discount at risk-free rate Some market-priced asset needed to represent value of project Neely took the stock price for Ford, assuming it correlated with sales of cars Direct treatment of endogenous uncertainties, that is, those due to project Discount at risk-appropriate rate, for example at CAPM Hybrid Approach to Valuation Slide 10 of 23 5
Details of Ford Case Neely collaborated with Materials section of Ford R & D First table illustrates description of projects Second table, the data for the projects Note estimates of probabilities of success ~ 0.8 for Tier 3, Current Vehicle Focus ~ 0.5 for Tier 2, Future Vehicles ~ 0.3 for Tier 1, General Research Full data sets appended to presentation Hybrid Approach to Valuation Slide 11 of 23 R & D Project Descriptions Sample R & D Project Descriptions: Code A B C H J Project Area Vehicle Systems Engine Manufacturing Powertrain Engine Class Design Design Design Process Product Money Benefits Warranty & Testing Prototype Elimination Component Reduction Productivity Increase NA Other Benefits Model Weight Savings Model NA Weight Savings Hybrid Approach to Valuation Slide 12 of 23 6
R & D Project Data Sample R & D Project Data: Code Costs $, millions Annual R&D Implement Benefits $, millions P (Success) A 1.0 0.25 40 0.8 B 0.8 0.25 42 0.8 H 0.4 1.00 12 0.8 J 0.8 1.00 1 0.9 Hybrid Approach to Valuation Slide 13 of 23 Results of Hybrid Analysis Next slide gives: Value of Each Project, by NPV and Hybrid Analysis Also ranking of projects by each method Note that: When valued as options, the R & D projects always had higher value. This is because, once one recognizes the possibility of canceling projects, one discards negatives that lower value The rank order is not the same for both methods. Thus the use of NPV not only gives wrong values systematically biased downward, but recommends some projects in wrong order! Hybrid Approach to Valuation Slide 14 of 23 7
Valuation of Projects by NPV and Hybrid P(s)*Benefits P(s)*Benefits NPV NPV OPT OPT Tier 3 Tier 2 A 32.0 A 97.7 B 28.0 B 85.5 E 20.0 C 54.4 C 18.0 E 51.4 F 12.0 F 39.1 G 9.8 G 29.9 H 8.0 H 24.1 I 4.0 I 11.3 D 2.7 D 8.5 J 0.9 J 1.2 N 100.0 N 282.5 K 30.0 L 35.6 L 25.0 K 35.3 M 15.0 M 26.5 P 15.0 O 9.9 O 6.0 P 1.5 A 130.6 B 114.3 E 75.6 C 72.9 F 50.3 G 40.0 H 32.4 I 15.8 D 11.0 J 2.1 N 402.7 K 95.6 L 81.1 M 49.6 P 41.6 O 18.9 Hybrid Approach to Valuation Slide 15 of 23 Higher Uncertainty Higher Value Graph clearly illustrates this basic principle % Difference Between DA and NPV 10000 1000 P 100 K L O 10 N I J 1 E C, D M G A, B, F, H 0.1 0.2 0.4 0.6 0.8 1 Probability of Success Hybrid Approach to Valuation Slide 16 of 23 8
Neely s Guide for Use of Hybrid Method High Exogenous Market Related Uncertainties Financial Options NPV Decision Analysis + Financial Options Decision Analysis Low Low Endogenous Project Specific Uncertainties High Hybrid Approach to Valuation Slide 17 of 23 Summary The Hybrid Method is a Feasible way of Valuing Real Options in Systems It has the great advantage of combining Market effects on project (provided a good proxy variable can be found) Technical uncertainties It is far from theoretically pure a rough and ready valuation that has the merit of being plausibly better than feasible alternatives More research needed! Hybrid Approach to Valuation Slide 18 of 23 9
References de Neufville, R. and Neely, J. (2001) Hybrid Real Options Valuation of Risky Project Development Projects, International Journal of Technology, Policy and Management, Vol. 1, No. 1, pp 29 46. Neely, J. (1998) Improving the Valuation of Research and Development: A Composite of Real Options, Decision Analysis and Benefit Valuation Frameworks, Ph.D. dissertation, MIT, Cambridge, MA. Hybrid Approach to Valuation Slide 19 of 23 Some Appendix Tables follow Hybrid Approach to Valuation Slide 20 of 23 10
Complete Description of Projects Code Project Area Class Traceable $ Benefit Other Benefits Tier 3: Current Vehicle Focus A Vehicle systems Design Warranty & testing savings Model B Engine Design Prototype elimination Weight savings C Powertrain Design Component reduction & sharing Model D Vehicle systems Design Warranty & testing savings Model E Engine Process Process step elimination Model F Manufacturing Process Yield & productivity increase Model G Recycling Process Cost avoidance Environment H Manufacturing Process Productivity increase NA I Sensors Product Warranty savings Monitoring J Engine Product NA Weight savings Tier 2: Future Vehicle Possibilities K Body C Process Reduced production cost Weight savings L Body C Assembly Process Reduced production cost Weight savings M Engine Process Lower cost material Weight savings N Body A Process Lower cost material Weight savings O Body A limited Process Lower cost material Weight savings P Trucks Product Reduced production cost Weight savings Tier 1: General Research Q Environment Design Tool NA Design tool R External Programs Other NA NA Hybrid Approach to Valuation Slide 21 of 23 Complete Data for Projects Code Annual R&D Cost (millions $) Implementation Cost (millions $) Annual Benefits (millions $) Upper Limit of Annual Benefits (millions $) R&D End Date Probability of Success Tier 3: Current Vehicle Focus A 1.0 0.25 40.00 48.00 1998 0.8 B 0.8 0.25 35.00 42.00 1998 0.8 C 0.4 1.00 20.00 24.00 1998 0.9 D 0.2 0.25 3.00 3.60 1997 0.9 E 0.6 2.00 25.00 30.00 1999 0.8 F 0.2 0.25 15.00 18.00 1997 0.8 G 0.2 0.25 14.00 16.80 1998 0.7 H 0.4 0.25 10.00 12.00 1998 0.8 I 0.2 1.00 8.00 9.60 1998 0.5 J 0.8 1.00 1.00 1.20 1997 0.9 Tier 2: Future Vehicle Possibilities K 1.0 50.00 100.00 120.00 2000 0.3 L 0.2 50.00 50.00 60.00 1997 0.5 M 0.1 0.25 25.00 30.00 2002 0.6 N 0.6 50.00 250.00 300.00 1997 0.4 O 0.6 10.00 10.00 12.00 1997 0.6 P 0.4 50.00 50.00 60.00 2000 0.3 Tier 1: General Research Q 0.2 NA NA NA 1999 NA R 0.2 NA NA NA 2004 NA Hybrid Approach to Valuation Slide 22 of 23 11
P(s)*Benefits P(s)*Benefits NPV NPV DA DA OPT OPT Tier 3 Tier 2 A 32.0 A 97.7 A 97.9 A 130.6 B 28.0 B 85.5 B 85.8 B 114.3 E 20.0 C 54.4 C 54.6 E 75.6 C 18.0 E 51.4 E 51.8 C 72.9 F 12.0 F 39.1 F 39.3 F 50.3 G 9.8 G 29.9 G 30.0 G 40.0 H 8.0 H 24.1 H 24.2 H 32.4 I 4.0 I 11.3 I 11.8 I 15.8 D 2.7 D 8.5 D 8.5 D 11.0 J 0.9 J 1.2 J 1.3 J 2.1 N 100.0 N 282.5 N 311.3 N 402.7 K 30.0 L 35.6 L 59.1 K 95.6 L 25.0 K 35.3 K 58.5 L 81.1 M 15.0 M 26.5 M 26.6 M 49.6 P 15.0 O 9.9 P 24.6 P 41.6 O 6.0 P 1.5 O 13.6 O 18.9 Hybrid Approach to Valuation Slide 23 of 23 12