Set out frequent mistakes in capital allocation and explain real options
The idea
The recurring pitfalls are behavioural rather than technical: including sunk costs, ignoring the opportunity cost of assets the project will consume, overlooking cannibalisation of existing products, and internal politics that favour the most persuasive sponsor rather than the best project. Real options sit on the other side of the ledger - the ability to expand, abandon, defer or switch inputs has value that a single-scenario NPV omits. A project with a slightly negative NPV and a valuable option to expand may still be worth undertaking.
A practice question
Why should sunk costs be excluded from a capital investment analysis?
- Because they are not tax deductible at allWhy not: Many are deductible. The reason to exclude them is that no decision now can change them.
- Because they cannot be changed by the decision - correct
- Because they are always immaterialWhy not: They are often very large, which is exactly why the temptation to count them is so strong.
A sunk cost is money already spent that cannot be recovered. A decision can only change future cash flows. Since the sunk cost is the same whether you go ahead or not, it should not influence the choice. Formula: Incremental cash flow = Cash flow with the project - Cash flow without it. A sunk cost appears in both parts and cancels out. These costs are often large, which is why people are tempted to count them, but including them distorts the comparison.
A harder one
A new product is expected to take 20% of its sales from the company's existing line. How should this be treated?
- Ignored, because total company sales still rise overallWhy not: Sales rise by less than the project's own figure suggests, and the difference is real lost income.
- Added to the project's revenue as a transfer of salesWhy not: Adding it counts the same sales twice, once in the old line and once in the new.
- Deducted as cannibalisation, since only incremental cash flow counts - correct
Cannibalisation means a new product takes sales from the company's own existing products. Only the change in the company's total cash flow counts. Formula: Incremental sales = New product sales - Sales lost from existing line. For example, if the new product sells $10 million and $2 million (20%) comes from the existing line, company sales rise by only $10 million - $2 million = $8 million. The lost after-tax cash flow on that $2 million is charged to the new project. Ignoring it overstates the project, and adding it as a transfer counts the same sales twice.
The hardest tier
Why might a project with a marginally negative NPV still be worth undertaking?
- Because a valuable real option is not captured - correct
- Because NPV is unreliable for small projectsWhy not: It applies at any scale; what a single-scenario calculation misses is flexibility.
- Because the IRR will be positiveWhy not: A positive internal rate below the cost of capital still destroys value.
A normal NPV calculation values one expected path. But a real option is the right, not the duty, to change course later, such as expanding if demand is strong or stopping if it is weak. That flexibility has value because you gain from good outcomes and can limit bad ones. Formula: Project value = Base-case NPV + Value of real options. If the option value is large enough, it can outweigh a small negative base-case NPV, so the project can be worth doing.
In the app: 9 practice questions on this outcome, a narrated walkthrough with diagrams, spaced review that brings it back before you forget it.
Take the free diagnostic →