ACCA Applied Skills · Financial Management

Adjusting for Risk and Uncertainty in Investment Appraisal

Adjusting for risk and uncertainty means testing how far an NPV decision could change if forecasts turn out wrong. You use sensitivity analysis, expected values, scenario and worst-case analysis, simulation, a higher discount rate or a shorter payback period. Then you judge whether the project is still acceptable.

What this chapter covers

Investment appraisal gives you one NPV from one set of forecasts. Those forecasts are never certain. This chapter shows you how to measure and respond to that doubt. You start by separating risk, where outcomes have known probabilities, from uncertainty, where they do not. Then you learn the tools.

The tools fall into two groups. Some measure the effect of changes: sensitivity analysis, scenario analysis, worst-case analysis and simulation. Others build the doubt into the answer: expected values, a risk-adjusted discount rate and a shorter payback cut-off. You must know how to calculate each one and when it is weak.

This chapter sits on top of NPV, IRR, payback and the cash flow rules from earlier investment appraisal work. It also links to cost of capital, because the discount rate is one way to reflect risk. The topic can be examined in both objective test questions and constructed response questions, so you need to calculate accurately and also explain and advise.

Investment appraisal questions in FM can include a risk element. Objective test questions are marked all or nothing, so accuracy in sensitivity percentages and expected values pays. In Section C, the written comment on risk is often where marks are lost, because answers are vague. If you can calculate cleanly and give a short, reasoned recommendation, you pick up marks that many students leave behind.

Adjusting for risk and uncertainty in investment appraisal: topics in the order to study them

1Risk vs Uncertainty in Investment AppraisalStart here to learn the definitions and why forecasts fail, because every later tool depends on this distinction.
2Sensitivity AnalysisIt is the most tested calculation and uses only the NPV you already know how to find.
3Expected Values and Probability AnalysisNext you add probabilities, which is the first method that works with risk rather than just testing changes.
4Adjusting the Discount Rate and Payback PeriodThese are simple, practical adjustments, and they make more sense once you have seen the calculation-based methods.
5Simulation, Scenario and Worst-Case AnalysisFinish with the broader methods, which are mostly discussion and need the earlier ideas to compare against.

How to prepare Adjusting for risk and uncertainty in investment appraisal

Aim to be quick at the calculations and clear in the written points. Work in this order.

Write your own definitions of risk and uncertainty with one example of each, and check you can say them without notes.
Recap the NPV layout from earlier chapters. Sensitivity and expected values both start from a correct base NPV, so fix any weak spots first.
Practise sensitivity on each cash flow variable: sensitivity = NPV ÷ PV of the cash flow affected × 100%. Use the PV of contribution for sales volume, the PV of sales revenue for selling price, and the PV of the relevant costs for a cost. Use after-tax figures where tax applies. Do it for each variable and rank the results. For the discount rate, do not use this formula. Find the IRR, the rate at which NPV is zero, and compare it with the cost of capital.
Practise expected value questions with a probability table. Check that probabilities add to 1, and state clearly that the expected value may never actually occur.
Learn one advantage and one weakness for every method, so you can write a balanced comment in a few lines.
Finish with timed Section C style questions: calculate, then write a short recommendation that uses your figures and names the most critical variable.

Common mistakes in Adjusting for risk and uncertainty in investment appraisal

Using the wrong denominator in sensitivity analysis.
Fix: Isolate the present value of just that variable's cash flows, after tax if tax applies, then divide the NPV by it.
Treating the expected value as the outcome that will happen.
Fix: Say it is a weighted average over many repetitions, and mention that a one-off project may produce a very different result.
Confusing risk with uncertainty in written answers.
Fix: Tie risk to known probabilities and uncertainty to unknown ones, and use that distinction when explaining why a method is suitable.
Writing generic comments such as 'sensitivity analysis is useful'.
Fix: Name the variable with the smallest margin, quote its percentage, and say what management should monitor or negotiate.
Adding a risk premium to the discount rate without explaining it.
Fix: State that the premium is subjective, that applying a constant premium each year implicitly compounds risk over time (later cash flows are discounted more heavily), which may not be realistic for every project, and that it does not show which variable causes the risk.
Letting probabilities not sum to 1 or mixing up joint and conditional probabilities.
Fix: Check the total before multiplying, and for joint outcomes multiply along each branch before weighting.

Last-day revision: Adjusting for risk and uncertainty in investment appraisal

Risk: probabilities of outcomes are known or can be estimated. Uncertainty: they cannot.
Sensitivity (%) = NPV ÷ PV of the cash flow affected × 100. For sales volume use the PV of contribution, for selling price use the PV of sales revenue, and for a cost use the PV of that cost. Use after-tax figures where tax applies. The smaller the percentage, the more sensitive the project.
For the discount rate, sensitivity is found through the IRR, the rate at which NPV is zero. Compare the IRR with the cost of capital to see how much the rate can rise.
Sensitivity shows how much a variable can change before NPV hits zero. It ignores probabilities and changes one variable at a time.
Expected value = Σ (probability × outcome). Probabilities must add to 1.
An expected value is a long-run average. It may not match any single outcome, so it is weak for a one-off project.
Expected values give no measure of spread, so they do not show how risky the project is.
A higher discount rate for riskier projects applies the same risk premium every year. This implicitly compounds risk over time, so later cash flows are discounted more heavily. That may not be realistic for every project. The risk premium itself is also subjective.
Payback favours quick returns, so it reduces exposure to later uncertainty, but it ignores cash flows after the cut-off.
Scenario analysis changes several variables together, such as best, most likely and worst cases.
Worst-case analysis shows the lowest possible NPV but says nothing about how likely it is.
Simulation uses random values from probability distributions to produce a range of NPVs. It is powerful but costly and relies on good inputs.
In written answers, recommend a decision and give a reason linked to your figures.

Adjusting for risk and uncertainty in investment appraisal practice questions

Adjusting for risk and uncertainty in investment appraisal in other exams

The same ground in other exams, if you are preparing for more than one or want another angle on it.

Adjusting for risk and uncertainty in investment appraisal: frequently asked questions

What is the difference between risk and uncertainty in investment appraisal?

Risk means you can assign probabilities to different outcomes. Uncertainty means you cannot. Expected values suit risk, while sensitivity and scenario analysis help when probabilities are unknown.

How do I calculate sensitivity in an FM question?

Find the project NPV. For a cash flow variable, divide the NPV by the present value of the cash flows affected and multiply by 100. Use the PV of contribution for sales volume, the PV of sales revenue for selling price, and the PV of the relevant costs for a cost, after tax where tax applies. This gives the percentage change in that variable that would make the NPV zero. For the discount rate, find the IRR instead and compare it with the cost of capital.

Why is expected value not always a good decision tool?

It is an average over many repeats, so it suits repeated decisions better than a one-off project. It also hides the spread of outcomes, so two projects with the same expected value can have very different risk.

Does raising the discount rate really deal with risk?

It lowers the NPV of risky projects, so only those with stronger returns pass. However, the premium is subjective. Applying a constant premium each year implicitly compounds risk over time, so later cash flows are discounted more heavily, which may not be realistic for every project. Use it with other methods, not alone.

What should I write in a Section C comment on risk?

Give a clear recommendation, support it with your calculated figures, and name the most critical variable. Add one limitation of the method you used. Short, specific points score better than long general ones.