July 12, 2006 Identification of Real Options “In” Projects Prof. Richard de Neufville Massachusetts Institute of Technology Room E40-245, Cambridge, MA 02139 [email protected]Dr. Tao Wang Morgan Stanley 3 Exchange Square, Central, Hong Kong [email protected]
38
Embed
July 12, 2006 Identification of Real Options In Projects Prof. Richard de Neufville Massachusetts Institute of Technology Room E40-245, Cambridge, MA.
July 12, Real Options A real option is a right, but not an obligation, to do something for a certain cost within or at a specific period of time Real options can be categorized as: –“ON” projects, these are financial options, but on technical things, they treat technology as ‘black box’” –“IN” projects, these are options created by designing technical system. They require understanding of technology.
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
July 12, 2006
Identification of Real Options “In” Projects
Prof. Richard de NeufvilleMassachusetts Institute of TechnologyRoom E40-245, Cambridge, MA 02139
• Future is inherently unknown and flexibility has value
• Options theory offers an convenient tool to define and model flexibility
• However, real options “in” projects are
– A priori undefined, complex, and interdependent
– Have many more options than designers could consider
• How to identify the real options most likely to offer good flexibility and the most value?
July 12, 2006
3
Real Options
• A real option is a right, but not an obligation, to do something for a certain cost within or at a specific period of time
• Real options can be categorized as:
– “ON” projects, these are financial options, but on technical things, they treat technology as ‘black box’”
– “IN” projects, these are options created by designing technical system. They require understanding of technology.
July 12, 2006
4
Development of Options Theory
Real Options "in" Projects
Real Options "on" Projects
Financial Options
Real Options "on" Projects
Financial Options
Financial Options
Options Theory
July 12, 2006
5
Comparing Real Options “In” and “On” Projects
Real Options “On” Projects Real Options “In” Projects
Value Opportunities Design flexibility
Valuation important Decision important (go or no go)
Relatively easy to define Difficult to define
Interdependency/path-dependency less an issue
Interdependency/path-dependency an important Issue
July 12, 2006
6
Difficulties Facing Identification of Real Options “In” Projects
• The problem is not that we cannot discern options
• But too many variables and possible real options
• Less than a small fraction of possible options can be considered
• The designers need to identify options most likely to offer good flexibility
July 12, 2006
7
Proposed Options Identification Process
Consists of two parts:
Part 1: A screening model
• Simplified, conceptual, low-fidelity model
• Can be run many times to explore possible flexibilities
• Can be simplified in a number of ways
July 12, 2006
8
Proposed Options Identification Process (continued)
Part 2: A simulation model
• A high fidelity model
• Main purpose is to examine, under technical and economic uncertainties, the designs preliminarily identified by the screening model for their:
– Technical robustness and reliability
– Expected benefits under uncertainties
July 12, 2006
9
General Procedure for Establishing A Screening Model
1. Reduce complexity of system by eliminating some secondary dimensions of problem
2. Develop a screening model based on the reduced dimensions
July 12, 2006
10
General Procedure for Using the Screening Model to Identify Options
1. List uncertain variables
2. Find out the standard deviations or volatilities for the uncertain variables
3. Perform sensitivity analysis on the uncertain variables to pick out the several most important uncertain variables for further analysis
July 12, 2006
11
General Procedure for Using the Screening Model to Identify Options
(continued)4. List different levels of the important uncertain
variable(s) as inputs for the established screening model to identify where the most interesting real options are:
– The design elements that vary across the sets are those that may be good real options; and
– Conversely, the design elements that are included for all sets, that are insensitive to uncertainty, or design elements where settings are always constant do not present interesting real options
July 12, 2006
12
Case Studies to Illustrate the Identification Process in Details
July 12, 2006
13
Case I:Water Resources Development
• Decisions on building dams and hydropower stations regarding:
– Choice of sites
– Capacity of power plants (H) and capacity of reservoirs (V)
– Sequence of projects construction
?
??
?
?
??
?
July 12, 2006
14
Developing Screening Model for the Water Resources Development Case
• Reduce dimensions of problem
– Drop stochasticity of water flow and electricity price
– Neglect the transitional phase and assume steady state
• Develop the optimization screening model
July 12, 2006
15
Screening Model
Max mYm cmYm i
s.t. TY t (technical constraints)
EY e (economic constraints)
July 12, 2006
16
Formulation of the Screening Model
Max PPst crf FCs VCsVs sHs s
t
s (1)
s.t. (Technical constraints Part I Continuity constraints)Ss,t1 Sst E st X st kt s,t (2)E st X s 1,t Fst s,t (3)
(Technical constraints Part II Reservoir storage /capacity constraints)Sst Vs 0 s,t (4)Vs yrsCAPDs 0 s (5)Sst s Ast 0 s,t (6)
July 12, 2006
17
Formulation of the Screening Model (continued)
(Technical constraints Part III Hydropower constraints)Pst 2.73es kt X st Ast 0 s, t (7)Pst YsthtHs 0 s, t (8)AMINs Ast 0 s, t (9)Ast AMAXs 0 s, t (10)AMAXs 2AMINs 0 s, t (11)Hs CAPPs 0 (12)
(Budget constraints)
FCs VCs Vs s Hs s B s (13)
July 12, 2006
18
List of Variables for the Screening Model
Variable Definition Units
yrsInteger variable indicating whether or not the reservoir is constructed at site s
Sst Storage at site s for season t 106m3
Xst Average flow from site s for season t m3/s
Est Average flow entering site s for season t m3/s
Pst Hydroelectric power produced at site s for season t MWh
Ast Head at site s for season t m
Hs Capacity of power plant at site s MW
Vs Capacity of reservoir at site s 106m3
AMAXs Maximum head at site s m
AMINs Minimum head at site s m
July 12, 2006
19
List of Parameters for the Screening Model
Parameter Definition Units ValueQin,t Upstream inflow for season t m3/s (374,283)
CAPDs Maximum feasible storage capacity at site s 106m3 (9600, 25,
12500)
CAPPs Upper bound for power plant capacity at site s MW (3600, 1700,
3200)
FstIncrement to flow between sites s and the next site for season t m3/s
(212, 105) for site 3, others
are 0es Power plant efficiency at site s 0.7
kt Number of seconds in a season 106 sec 15.552
ht Number of hours in a season Hours 4320
Yst Power factor at site s for season t 0.35
July 12, 2006
20
List of Parameters for the Screening Model (continued)
Parameter Definition Units Value
P Hydropower benefit coefficient 103 RMB/MWh 0.25
FCs Fixed cost for reservoir at site s B RMB (11.19, 0, 8.41)
VCs Variable cost for reservoir at site s B RMB/106m3
(4.49 x 10-4, 0, 6.68 x 10-4)
s Variable cost for power plant at site s B RMB/MW
(7.65 x 10-4, 1.85 x 10-3, 8.80 x 10-4)
r Discount rate 0.086crf Capital recovery factor for 60 years 0.087B Total budget available 103 RMB 80,000,000
July 12, 2006
21
Using the Screening Model to Identify Real Option
• Step 1. List uncertain variables. For the purpose of illustration, we pick out three important uncertain variables for further scrutiny:
– Electricity price,
– Fixed cost of reservoir, and
– Variable water flow
July 12, 2006
22
Using the Screening Model to Identify Real Option (continued)
• Step 2. Find out the standard deviations or volatilities for the uncertain variables:
– Volatility of future electricity price derived by experts’ expectation;
– Standard deviation of the construction cost was estimated from the standard deviation for cost of mega projects (Flyvbjerg, et al., 2003, pp. 16);
– Variable water flow calculated from historical record
July 12, 2006
23
Using the Screening Model to Identify Real Option (continued)
• Step 3. Perform sensitivity analysis on the uncertain variables to pick out the most important uncertain variable(s) for further analysis:
– A tornado diagram is a useful tool for such sensitivity
analysis
July 12, 2006
24
Tornado Diagram
• This kind of diagram expresses the change of net benefit due to 1 standard deviation/volatility change of each important uncertain variables, with other uncertain variables kept at the expected value
Change of Net Benefit (in M RMB) Due to 1 S.D. Change of Important Uncertain Variables
-80
-325
-418
18
332
596
(Million RMB)
Fixed cost for reservoir
Electricity price
Waterflow
2196
July 12, 2006
25
Tornado Diagram (continued)
• The most important uncertainties are fixed cost and electricity price:
– But the fixed cost of reservoir is a one time cost, offering less opportunity for ongoing flexibility
– The electricity price is a stochastic process that offers interesting flexibility over the life of the project
• As a first step to illustrate the approach, we focus on one uncertainty of electricity price in this paper
July 12, 2006
26
Back to Using the Screening Model to Identify Real Option
• Step 4. List different levels of the important uncertain variable (electricity price) as inputs for the established screening model to identify where the most interesting real options are.
– Although the current price of electricity was 0.25 RMB/KWh, we studied the conditions when it might be 0.10, 0.13, 0.16, 0.19, 0.22, 0.28, 0.31 RMB/KWh
– The levels cover most of the range of the experts’ pessimistic (0.18 RMB/KWh) to optimistic estimate (0.315 RMB/KMB)
July 12, 2006
27
Results from Step 4
H 1 V 1 H 2 V 2 H 3 V 3(MW) (106m3) (MW) (106m3) (MW) (106m3)
Price = 0.300 RMB/KwH Project 3 no build Project 3 no buildProject 1 no buildProject 2 build Price = 0.300 RMB/KwH
Project 3 no build Price = 0.241 RMB/KwH Project 1 no buildProject 1 no build Project 3 no buildProject 3 no build
Price = 0.193 RMB/KwHProject 1 no buildProject 3 no build
July 12, 2006
34
Results - Satellite Communication Systems
• Results
– We would first invest $0.5 Billion to test the market
– If market is not good, only lose $0.5 Billion, rather than $2.0 Billion in the best traditional architecture
– Serve up to 7.8 million subscribers readily, while the best traditional design can only serve up to 2.8 million
• Downside significantly cut, upside possible
• If Iridium and Globalstar had used this design…
July 12, 2006
35
Decision TreeYear 0 Year 2.5 Year 5 Year 7.5 Year 10
Demand = 41.85
Demand = 13.84Demand = 4.57
Demand = 4.57Deploy A5 Demand = 4.57
Demand = 1.51Demand = 0.5
Demand = 1.51Deploy A4 Demand = 4.57
Deploy A5Demand = 1.51
Demand = 0.5
Demand = 0.50Demand = 0.5
Demand = 0.17Demand = 0.05
Demand = 0.5Deploy A2 Demand = 4.57
Deploy A5Demand = 1.51
Deploy A4 Demand = 0.5
Demand = 0.50Demand = 0.5
Demand = 0.17Demand = 0.05
Demand = 0.17Demand = 0.5
Demand = 0.17Demand = 0.05
Demand = 0.05Demand = 0.05
Demand = 0.02Demand = 0.01
July 12, 2006
36
Additional References
• Wang, T. (2005) “Real Options ‘in’ Projects and Systems - Identification of Options and Solution for Path Dependency”, PhD Thesis, MIT, Cambridge, MA
• Wang, T. and de Neufville, R. (2004) “Building Real Options into Physical Systems with Stochastic Mixed-Integer Programming,” 8th Annual Real Options International Conference, Montreal, Canada
• The options identification discovers the design elements most likely to provide worthwhile flexibility
• The proposed identification process includes a
screening and a simulation model:
– Screening model is a simplified, conceptual, low-fidelity model of the system. It that represents most important issues. It can be easily run many times to identify sets of options;
– Once the screening model has identified plausible options, a detailed simulation model can validate and refine the options