Capital Budgeting: Evaluating and Selecting Long-Term Investments

Capital budgeting is the process of evaluating and selecting long-term investments. It covers the identification of opportunities, the estimation of the cash flows each would produce, the appraisal of those cash flows against the cost of the money used to fund them, the decision itself, and the review of what actually happened.

The subject is often reduced to the arithmetic of net present value, which is the least difficult part of it. The arithmetic is mechanical; the estimates fed into it are judgements, and a decision is only as good as the worst of them.

Why the decision is treated separately

Capital expenditure decisions have three properties that ordinary operating decisions do not, and the whole apparatus exists because of them.

They are large relative to the business, so an error cannot be absorbed within a year's trading. They are long-lived, committing the business to an asset, a market and a technology for years. And they are difficult to reverse: a specialised asset has little resale value, and the cost of exit frequently exceeds the cost of continuing.

Those properties also explain why the process, instead of the calculation, is what organisations invest in. A poor proposal that is properly appraised will be rejected. A good proposal that nobody identified is never appraised at all, which is why the search stage matters more than its treatment in most courses suggests.

The stages

Identification. Opportunities arise from strategy, from operational necessity, from regulatory requirement, and from people close to the work. An organisation whose proposals only ever come from the centre has a search problem it cannot see.

Screening. Proposals are tested against strategic fit and against any absolute constraint before effort is spent on detailed estimation. A project with an excellent return that takes the business into a market it has decided to leave should die here.

Estimation. The cash flows are built: the initial outlay, the incremental operating flows, working capital movements, tax, and any terminal value. This is where most of the error enters.

Appraisal. The flows are evaluated using one or more of the methods below, at a discount rate reflecting the risk of the project.

Authorisation. Approval limits allocate decisions between operating management, the board and, for the largest, the shareholders.

Implementation and monitoring. Spend against budget, and the point at which an overrunning project should be abandoned.

Post-completion audit. A comparison of outcome against forecast, conducted after the asset is in use. Its value is less in correcting the individual project, which is already committed, than in calibrating the estimates on the next one. An organisation that never looks back never discovers that its sponsors are systematically optimistic.

The four appraisal methods

Payback period measures how long the initial outlay takes to recover from the project's cash flows. It is simple, it speaks directly to liquidity, and it is the most widely used method in practice despite being the weakest in theory. It ignores the time value of money, ignores everything after the payback point, and cannot rank projects sensibly. Its genuine merit is as a risk screen: in a volatile market, an estimate three years out is worth more than an estimate eight years out, and a short payback limits exposure to forecasts nobody can make reliably.

Accounting rate of return expresses average accounting profit as a percentage of the investment. It is the only method expressed in the terms the reported accounts use, which is why it survives: managers judged on return on capital employed care how a project will look in those figures. As an investment criterion it is unsound, because it uses profit, not cash and ignores timing entirely.

Net present value discounts every incremental cash flow at the cost of capital and sums them. A positive figure means the project is expected to return more than the cost of the money used to fund it, and the number is the amount by which shareholder wealth is expected to increase. It is theoretically correct, it handles unconventional cash flow patterns, and values add across projects, which makes it the right method for a capital rationing problem.

Internal rate of return is the discount rate at which net present value is zero, compared with the cost of capital. Its appeal is presentational: a percentage return is easier to discuss than an absolute figure, and it does not require the discount rate to be specified before the calculation. Its defects are real. It can produce multiple values where the sign of the cash flows changes more than once, it cannot be added across projects, and it implicitly assumes reinvestment at its own rate, which is unrealistic where the rate is high.

The two discounting methods usually agree on whether to accept a single project. They disagree on ranking when projects differ in scale or in the timing of their flows, and where they disagree the net present value answer is the one to follow, because it measures the value added and not the rate at which it is earned.

Getting the cash flows right

This is where the marks and the money both are.

Incremental only. Include a flow if and only if it differs between doing the project instead of doing it.

Ignore sunk costs. Money already spent, including feasibility work, is irrelevant to the decision now facing the business.

Include opportunity costs. A building the business already owns is not free to the project; its cost is the rent or sale proceeds foregone.

Exclude financing flows. Interest and loan repayments are excluded from the cash flows because the cost of finance is in the discount rate. Including both counts it twice.

Exclude depreciation. It is not a cash flow. It enters only through its effect on tax.

Include tax properly. Tax on operating flows, tax relief on capital allowances, and the timing of both, which frequently lags the underlying flow by a period.

Include working capital. An increase in inventory and receivables is an outflow when the project starts and typically recovered at the end. Forgetting it is the single most common omission.

Treat inflation consistently. Either discount money cash flows at a money rate or real cash flows at a real rate. Mixing the two is the second most common error, and it is always detected.

Risk and uncertainty

Appraisal produces a single figure from estimates that are not single figures, and the techniques for handling that are examinable in their own right.

Sensitivity analysis varies one input at a time to find how far it can move before the decision changes. It identifies the variables that matter, which is useful, but it says nothing about how likely a movement is.

Scenario analysis varies a coherent set of inputs together, which is more realistic because the variables are correlated in practice.

Simulation assigns distributions to inputs and produces a distribution of outcomes. It gives a probability of a negative result rather than a point estimate.

Risk-adjusted discount rates apply a higher rate to a riskier project. This is the standard approach and it has a known weakness: it treats risk as increasing uniformly with time, which suits some projects and not others.

Real options value the flexibility a project contains — to defer, expand, contract or abandon. Conventional appraisal treats the decision as now or never, and a project with a negative net present value today may still be worth holding open if waiting resolves uncertainty at low cost.

Capital rationing

Where funds are limited, the objective changes from accepting every positive project to maximising value from the funds available.

Where the constraint applies to a single period and projects are divisible, ranking by profitability index — present value per unit of the scarce resource — gives the optimal set. Where projects are indivisible, the answer is found by examining feasible combinations. Where the constraint spans several periods, the problem becomes one of mathematical programming.

The distinction between hard and soft rationing is worth stating. Hard rationing is an external constraint imposed by the capital markets. Soft rationing is a limit the business imposes on itself, usually to control quality of proposals or to avoid dilution. Soft rationing means positive value projects are being declined by choice, which is a decision that should be made explicitly rather than inherited from a budget.

How it is examined

The standard question supplies a project's data and asks for appraisal and a recommendation.

Build the cash flow table before touching a discount factor, with a column per year and a line per item, and show the tax and working capital lines separately so the marker can follow them. State the inflation basis explicitly. Calculate the required measures, then recommend, and say what the recommendation depends on.

The discussion element usually asks for a comparison of methods or the limitations of the analysis. Answer it in terms of the decision in front of you, not in the abstract: which variable the result is most sensitive to, what the estimate depends on, what non-financial factors the numbers exclude, and what would have to be true for the decision to reverse. That final sentence is what distinguishes an appraisal from an arithmetic exercise.