L5: Capital Budgeting - Project Staging, Risk Analysis Flashcards
What is sensitivity analysis?
Breaks the NPV calculation into its component assumptions and shows how the NPV varies as the underlying assumptions change. Change one variable at a time, while holding others at the expected value. Can learn which assumptions are the most important → can invest resources and effort to refine these assumptions.
What is scenario analysis?
Considers the effect on the NPV of changing multiple project parameters simultaneously.
- Grouped by economic scenarios: Base/best/worst case
- Attaches probabilities to analysed scenarios (weigh relative importance)
- The implied values given by these probabilities should match the expectation in the standard analysis. (if scenarios & probabilities are estimated independently –> internal validation on consistency of estimates)
What does Weitzman, M. L. (1979) suggest about optimal search for the best alternative and real option value?
- Riskier projects provide more real option value.
- The real option value comes from being able to stop the search given a “good enough” outcome.
What are some limitations of real options?
- Unlike with financial options, there are no historical prices with which to estimate volatility.
- Requires projections about how future decision makers will proceed.
- The value of a real option is often sensitive to the estimates about the likelihood of scenarios (e.g. probability that the first movie is successful).
What are the advantages of sensitivity analysis?
- Shows whether the investment decision is robust. Is the result still NPV positive for all changes?
- Identifies where more information is needed. May want to learn more about particular areas.
- Identifies the areas that deserve the most managerial attention.
What are the limitations of sensitivity analysis?
- Even the best-case and worst-case estimates are wrong
- Variables are examined in isolation when they are related
- Best- and worst-case estimates are not associated with likelihood
What are some scenarios that impact how a company performs?
- Production management effectiveness (market share, price, costs)
- Trade restrictions (cost, market size, market share, price)
- Competition (cost, market share, price)
- Recession (discount rate, market size, price)
- Government incentives (market size, price, taxes)
- Etc.
What are the limitations of scenario analysis?
- Selecting likely scenarios is not an exact science. (countless options, no rule for deciding with are more relevant, business judgment)
- Important economic connections between sets of parameters may be overlooked. (since not all possible scenarios are feasible to exhaust)
- Analyzing the many combinations of different parameter values is cumbersome (–> monte carlo beneficial)
What is Monte Carlo estimation and what are the 5 steps of the general procedure?
A technique that relies on repeated random samplings to model uncertainty in complex economic interactions.
- Specify the model
- Specify a distribution for each variable
- Computer draws a sample
- Repeat step 3 N times
- Calculate relevant outcome (e.g. NPV)
What are 3 risk analysis techniques?
- Sensitivity analysis
- Scenario analysis
- Monte Carlo simulations
What are the advantages of Monte Carlo Estimation?
- Can explicitly specify the relationship between different parameters (e.g. correlations and other relationships)
- Does not require an exact model of how outcome variables depend on the input parameters (no exact mathematical formula needed. Instead simulation of if-then statements)
- Simulations can deepen a forecaster’s understanding of complicated economic scenarios (e.g. what it depends on)
What are the limitations of Monte Carlo Estimation?
- Not exactly “model-free”; it requires assumptions about how decisions in the economic interactions are made.
- Computer outputs are devoid of economic intuition (requires outside judgment from the analyst, human action needed to interpret the results)
- The distribution of each variable and the interactions between them are difficult to model (not clear which best describes reality)
- At some point, even the Monte Carlo method requires input of information from experts.
What is unlevered net income?
= EBIT * (1-t)
= (Revenues - Costs - Depreciation) * (1-t)
What are indirect effects that can affect earnings?
- Opportunity costs
- Project externalities: cannibalization/cross-selling
What are some common sunk costs?
- Fixed overhead expenses
- Past R&D expenditures
- Unavoidable competitive effects