Introduction: What Makes a Good Capital Budgeting Technique
Capital budgeting is the process a firm uses to evaluate potential investment projects. A firm usually has several projects at hand; the technique decides which to accept and which to reject. The techniques covered are NPV, payback period, discounted payback period and internal rate of return (IRR), plus how to estimate project cash flows.
The financial goal of the corporation is to maximise shareholder wealth, so accept projects that increase it and reject those that decrease it. A technique leads to inferior decisions if it:
| Flaw | Why it is a flaw |
|---|---|
| Ignores the time value of money | Time has money value |
| Is affected by managers' tastes and preferences | These say nothing about whether a project adds value |
| Is influenced by the choice of accounting method | Accounting method largely does not change cash flows |
| Depends on the profitability of the firm's existing business | Existing profitability is unrelated to whether the project adds value |
Introduction: What Makes a Good Capital Budgeting Technique
What CFOs actually use
A survey of US CFOs gave the share of the time each technique is used:
| Technique | Used |
|---|---|
| Book rate of return | 20% |
| Payback | 57% |
| IRR | 76% |
| NPV | 75% |
The percentages sum to more than 100% because a manager can use more than one technique on the same project. Book rate of return is least used, so it is not studied; the focus is NPV, payback, discounted payback and IRR.
Key takeaways
- Capital budgeting = deciding which investment projects to accept; the benchmark is shareholder wealth.
- A good technique uses time value of money and is free of managerial taste, accounting choices and existing-business profitability.
- Practice: IRR (76%) and NPV (75%) are the most used; book rate of return (20%) the least.
Net Present Value (NPV)
NPV is present value net of the initial investment: PV of the project's cash flows minus the initial outlay.
NPV depends on exactly two things: the forecasted cash flows from the project and the opportunity cost of capital for the project. It uses all cash flows and discounts them properly (according to their riskiness).
Decision rule: accept if ; reject if . A positive-NPV project raises shareholder wealth, a negative one lowers it, so NPV is consistent with the wealth-maximisation goal. NPV measures the change in shareholder wealth from accepting the project.
Net Present Value (NPV)
Adding-up property
Pros and cons
| Advantages | Disadvantage |
|---|---|
| Recognises time value of money | Not as easy to communicate to a layperson (telling someone "NPV is 100 rupees" does not convey whether that is good enough) |
| Clear benchmark: NPV (independent of managers' tastes) | |
| Computed from cash flows, so unaffected by accounting method |
Key takeaways
- ; depends on cash flows and opportunity cost of capital.
- Accept if .
- Unique adding-up property.
- Only weakness: hard to communicate.
Payback and Discounted Payback
Payback period
The payback period is the number of years needed to recover the initial investment.
Rule: accept if payback a specified cutoff period, which is set by management.
| Advantages | Disadvantages |
|---|---|
| Easy to understand and compute | No clear benchmark: the cutoff is a matter of managers' tastes and preferences |
| Biased towards liquidity (favours projects that pay back quickly) | Biased against long-term projects: ignores all cash flows after the cutoff |
| Ignores time value of money: gives equal weight to all cash flows before the cutoff |
Worked example
Three projects each cost $2,000 today; the opportunity cost of capital is 10%.
| Project | Year 1 | Year 2 | Year 3 | Payback | NPV |
|---|---|---|---|---|---|
| A | 500 | 500 | 5,000 | 3 years | $2,624 |
| B | 500 | 1,800 | – | 2 years | −$58 |
| C | 1,800 | 500 | – | 2 years | +$50 |
Cutoff of 2 years: payback accepts B and C, rejects A. NPV accepts A and C, rejects B. So payback rejects A (which should be accepted) and accepts B (which should be rejected).
- Project A's $5,000 in year 3 is ignored because it comes after the cutoff.
- B and C look identical to payback although the timing differs (500 then 1,800 versus 1,800 then 500): equal weight to flows before the cutoff means time value is ignored.
Discounted payback period
The discounted payback period is the number of years to recover the initial investment in discounted terms (PV of cash flows). It is a modification of payback.
Rule: accept if discounted payback a management-set cutoff.
| Advantages | Disadvantages |
|---|---|
| Biased towards liquidity | No clear benchmark (arbitrary cutoff set by management) |
| Takes time value of money into account | Biased against long-term projects (ignores cash flows after cutoff) |
| Never accepts a negative-NPV project |
Worked example (same projects, 10%)
Discount each flow: year flow . For A: , , .
| Project | Cumulative discounted recovery | Discounted payback |
|---|---|---|
| A | 455 (yr 1), 868 (yr 2), well over 2,000 by yr 3 | 3 years |
| B | 455 (yr 1), (yr 2) is still below 2,000 | Not defined (never recovered) |
| C | Recovers 2,000 in discounted terms by end of yr 2 | 2 years |
With a 2-year cutoff: accept C, reject A and B. NPV accepts A and C. The only mistake is rejecting A; the $3,757 (discounted) received in year 3 is ignored because it falls after the cutoff.
Why a negative-NPV project is never accepted
A negative NPV means the discounted cash flows never add up to the initial investment, so the discounted payback is undefined and the rule cannot accept it. (Project B, , is rejected.)
- 1Project NPV < 0
- 2PV of all inflows < initial outlay
- 3Investment never recovered in discounted terms
- 4Discounted payback undefined
- 5Rejected for any cutoff
Key takeaways
- Payback: years to recover the outlay; ignores time value and post-cutoff flows; cutoff is arbitrary.
- Discounted payback: same idea on PVs; never accepts negative NPV, but can still reject good projects.
- Both favour liquidity and are easy to understand.
Internal Rate of Return (IRR)
Definition and decision rule
Intuition: invest 100 today and receive 150 after a year; the profit as a proportion of investment is . As a project, . Setting NPV :
This matches the mental rate of return. For a one-year investment the rate is easy; for a multi-year investment you need IRR.
Rule: accept if opportunity cost of capital (the same logic as accepting when the rate of return exceeds the opportunity cost).
| Advantages | Disadvantages |
|---|---|
| Recognises time value of money | A project may have no IRR |
| Clear benchmark: opportunity cost of capital (not managers' tastes) | A project may have multiple IRRs |
| Easy to understand and communicate ("accept if the return is at least 20%") | Unreliable for ranking mutually exclusive projects that differ in scale or economic life |
| Does not distinguish investment from borrowing |
Computing IRR: worked example
A machine costs $4,000 and returns $2,000 at the end of year 1 and $4,000 at the end of year 2.
Set with . This is a quadratic in . Only real and positive solutions matter in practice; this has one, IRR = 28.08%. The project is accepted whenever the opportunity cost of capital is below 28.08%.
Excel: IRR(array of cash flows). Its one required argument is the cash flow range (here year 0 to year 2: ), giving 28.08%. Excel also takes an optional guess (default 10%). For projects with multiple IRRs, vary the guess; Excel returns the IRR closest to the guess.
Why "internal"?
IRR needs only the magnitude and timing of the project's own cash flows, nothing external. It does not need the opportunity cost of capital; that is needed only to compare against IRR when deciding.
IRR and the NPV profile
Plot NPV (y-axis) against the discount rate (x-axis). The point where NPV is the IRR.
| Region | IRR vs opportunity cost | NPV | IRR decision | NPV decision |
|---|---|---|---|---|
| Discount rate below IRR | IRR cost | Positive | Accept | Accept |
| Discount rate equal to IRR | IRR cost | Zero | Indifferent | Indifferent |
| Discount rate above IRR | IRR cost | Negative | Reject | Reject |
NPV and IRR give the same accept/reject decision only if NPV is a decreasing function of the discount rate. If NPV increases with the discount rate, the two rules disagree.
IRR vs opportunity cost of capital (do not confuse): IRR is a profitability measure depending solely on the amount and timing of project cash flows. The opportunity cost of capital depends on the risk of the project: it is the expected return in the financial market at the same level of risk as the project. Higher project risk means a higher opportunity cost.
Key takeaways
- IRR is the discount rate at which ; accept if opportunity cost.
- Multi-year IRR needs Excel/numerical solving; only real positive roots count.
- Same decision as NPV only when NPV decreases with the discount rate.
- IRR is "internal" because it uses only project cash flows.
Pitfalls of IRR
Pitfall 1: No IRR
Project F: outlay $1,000, inflow $3,000 in year 1, then a further investment of $2,500 in year 2. It has no IRR: its NPV is negative at all discount rates, so the NPV curve never cuts the discount-rate axis. Equally, a project whose NPV is positive at all discount rates has no IRR. Without an IRR there is nothing to compare with the opportunity cost, so the rule cannot be applied.
Pitfall 2: Multiple IRRs
Project G: outlay $30 billion; $10 billion inflow at the end of each year for years 1 to 9; $65 billion cleanup cost at the end of year 10.
- The NPV equation is a polynomial of degree 10, so up to 10 roots, but only positive real roots are of interest.
- Descartes' rule: the number of positive real roots is at most the number of sign changes in the cash flows. Project G has 2 sign changes ( and ), so up to 2 IRRs. Sign changes are a necessary but not sufficient condition for multiple IRRs.
- Indeed at 3.5% and at 19.54%: two IRRs. With two IRRs it is unclear which to compare with the opportunity cost.
Pitfall 3: Cannot distinguish investment from borrowing
| Project | Today | Year 1 | Resembles |
|---|---|---|---|
| D | −$1,000 | +$1,500 | Investment or lending: outflow now, inflows later |
| E | +$1,000 | −$1,500 | Borrowing: inflow now, outflows later |
Both have IRR . At a 10% opportunity cost: and . NPV accepts D and rejects E, but IRR gives the same number for both.
Borrowing-type cash flows occur in real life: an insurance company receives premiums today and pays claims in the future (inflow now, a series of outflows later).
Pitfall 4: Ranking mutually exclusive projects
| Type | Meaning |
|---|---|
| Independent | Accepting or rejecting it does not prevent accepting others (A, B, or both) |
| Dependent | Acceptance depends on accepting one or more other projects (A only if B accepted) |
| Mutually exclusive | Acceptance prevents accepting one or more alternatives (A or B but not both) |
IRR is unreliable for ranking mutually exclusive projects, in two situations.
(a) Different scale of investment
Projects H and I (mutually exclusive), opportunity cost 10%:
| Project | Today | Year 1 | IRR | NPV |
|---|---|---|---|---|
| H | −10,000 | 20,000 | 100% | $8,182 |
| I | −20,000 | 35,000 | 75% | $11,818 |
IRR favours H; NPV favours I (larger scale). NPV tells the change in shareholder wealth, so choose I. IRR errs because it ignores scale.
Fix: IRR on incremental cash flows. Form the incremental project : today, in year 1. Its IRR is 50% (and incremental NPV is $3,636, the NPV difference). Since , I is incrementally superior: accept I over H.
(b) Different economic lives
Projects J and K (mutually exclusive), same scale ($9,000), opportunity cost 10%:
| Project | Today | Cash flows | IRR | NPV |
|---|---|---|---|---|
| J | −9,000 | 6,000; 5,000; 4,000 (years 1 to 3) | 33% | $3,592 |
| K | −9,000 | 1,800 every year forever (perpetuity) | 20% | $9,000 |
NPV says K; IRR says J. The NPV profiles cross at 15.6%: above 15.6%, J has higher NPV; below it (as at 10%), K has higher NPV. So choose K.
Incremental IRR: use so the first flow is negative. Its IRR is 15.6% (found via Excel). Since , K is incrementally superior: accept K over J.
Limits of the incremental-cash-flow trick
- Valid only if the projects have the same risk, hence the same opportunity cost of capital.
- The incremental flows can themselves have multiple IRRs or no IRR.
- So it is not foolproof: prefer NPV whenever possible.
Why IRR is still popular
Managers use IRR about as often as NPV because computing IRR needs no opportunity cost of capital, only the magnitude and timing of cash flows. NPV needs the cost of capital, which managers sometimes do not know exactly.
Key takeaways
- IRR may not exist, or may be multiple (Descartes' rule gives an upper bound, not a guarantee).
- For borrowing-type flows, accept when opportunity cost.
- For mutually exclusive projects, rank by NPV, or use IRR on incremental flows.
- Incremental IRR requires equal risk and still can fail; NPV is safest.
Capital Budgeting Practice
Problem 1: three projects A, B, C
Initial outlays: A $1,000, B $2,000, C $3,000; opportunity cost 10%. (The year-by-year cash flow table is not reproduced in the lecture audio; only results are given.)
| Part | A | B | C |
|---|---|---|---|
| (a) NPV | ≈ −$91 | ≈ $4,045 | ≈ $39 |
| (b) Payback | 1 year | 2 years | 4 years |
| (c) Accept with 3-year payback cutoff? | Yes | Yes | No |
| (d) Discounted payback | Not defined | 3 years | 5 years |
| (e) Accept with 3-year discounted cutoff? | No | Yes | No |
- (a) NPV rule accepts B and C.
- (d) For A, the PV of the year-1 $1,000 is , and the $1,000 outlay is never recovered in discounted terms. For B: 909 after year 1, after year 2 (still below 2,000), full recovery by year 3.
- (e) A is never accepted whatever the cutoff, because it has negative NPV: discounted payback never accepts a negative-NPV project.
Problem 2: Mr. Simply Simple's mutually exclusive choice
Opportunity cost 9% (in $ thousands):
| Project | Year 0 | Year 1 | Year 2 | IRR |
|---|---|---|---|---|
| A | −400 | 250 | 300 | 23% |
| B | −200 | 140 | 179 | 36% |
Mr. Simple is tempted by B's higher IRR.
(a) Why this is wrong: A and B are mutually exclusive and differ in scale (A is larger). A lower IRR does not preclude a higher NPV, so choosing by higher IRR is not correct.
(b) Adapt the IRR rule with incremental cash flows. Use so the first flow is negative: , (), ().
Set NPV and multiply by : . Let :
The discriminant must be for real roots. Here , whose square root is 330:
Negative roots are of no interest, so , giving incremental IRR = 10%. Since , A is incrementally superior: choose A.
(c) Check with NPV (9%):
, so A is chosen on NPV as well.
Key takeaways
- Discounted payback is undefined for negative-NPV projects, which is why they are never accepted.
- With mutually exclusive projects of different scale, IRR and NPV can disagree; the incremental IRR (here 10% vs 9%) resolves it and agrees with NPV.
Estimating Project Cash Flows
NPV needs forecasted cash flows and the opportunity cost of capital. In real life cash flows must be estimated, using five rules.
| Rule | Statement |
|---|---|
| 1 | Cash flow is not accounting income; estimate free cash flow |
| 2 | Estimate cash flows on an incremental basis |
| 3 | Treat inflation consistently |
| 4 | Separate investment and financing decisions |
| 5 | Forecast cash flows after taxes |
Estimating Project Cash Flows
Rule 1: Cash flow is not accounting income
Accounting income (net profit, profit after tax) shows how well the firm is performing; cash flow is generally not equal to it.
Project cash flow (also net cash flow, NCF, or free cash flow, FCF; same thing) is the net cash generated or required by a project in a given year after paying operating costs, taxes and necessary capital expenditure.
- Direct approach: total cash inflows caused by the project minus total cash outflows caused by it.
- Indirect approach:
(inflows positive, outflows negative).
Operating cash flow (OCF) in four forms
Bottom-up (start from the bottom line, PAT, also called net income or accounting income). Depreciation is a non-cash expense deducted in reaching PAT, so add it back:
Top-down (start from sales, the top line). Since after tax, adding back depreciation cancels it:
Tax shield approach. Taxes . Substituting and factoring out :
- The first term is after-tax income assuming no depreciation.
- The second term is the depreciation tax shield: depreciation is non-cash but tax-deductible, so it reduces the firm's tax bill. It shields income from taxes.
All three approaches give the same OCF.
- Bottom-up: PAT + Depreciation
- Top-down: Sales - Cash expenses - Taxes
- Tax shield: (Sales - Cash exp)(1 - Tc) + Dep x Tc
Working capital and cash flows
Working capital is the capital needed for the day-to-day operations of the project (or firm). Its main components are inventory, accounts receivable (AR) and accounts payable (AP).
Operating cycle: the firm buys raw materials from suppliers (usually on credit, paid after a credit period), converts them into finished goods, and sells these (usually on credit, so customers pay later). Cash goes out when suppliers are paid and comes back when customers pay. The time between these two points is the cash cycle, and the cash tied up in operations over it is measured by net working capital.
(AR and inventory are the principal current assets; AP is the principal current liability.)
- Most projects require an initial investment in working capital: a cash outflow.
- Working capital needs can change during the project; higher sales usually need additional working capital: cash outflow.
- At the end of the project, inventory is sold and most receivables collected, so working capital is recovered: cash inflow.
NWC up means cash outflow (negative sign); NWC down means cash inflow (positive sign).
Worked example (rupees, project 2026 to 2030). For 2026: , inventory , , so . In 2025 all three are 0, so . , hence cash flow from NWC in 2026 is −50,000. Repeat for each year to 2030.
Rule 2: Estimate on an incremental basis
The value of a project depends on all the additional (incremental) cash flows that follow from acceptance.
| Item | Treatment | Reasoning |
|---|---|---|
| Negative incidental effect (e.g. cannibalising existing sales) | Cash outflow | Example: an electric-vehicle plant reducing sales of the firm's petrol two-wheelers |
| Positive incidental effect (helps existing business grow) | Cash inflow | |
| After-sales cash flows | Include | Not only product sales but also services and spare parts revenue |
| Salvage value | Cash inflow, net of taxes | Residual value from disposing equipment |
| Opportunity cost of an existing resource | Cash outflow, even if no cash changes hands | See below |
| Allocated overhead costs | Exclude (unless extra expense results) | Overheads (e.g. CEO salary) must be paid anyway; mere allocation is not incremental. Include only extra expenses caused by the project |
| Sunk costs | Ignore: no cash flow | Incurred in the past, irreversible; unaffected by the accept/reject decision |
Opportunity cost of an existing resource. A resource is not free: its cost is the cash it could generate if the project were rejected and it were sold or used elsewhere.
- Perpetual project: opportunity cost = the resource's market price (e.g. land that could be sold for Rs. 50 lakh: opportunity cost is Rs. 50 lakh).
- Finite-life project: opportunity cost = the resource's market rent over the project's life.
Rules 3 to 5
Rule 3: Treat inflation consistently. Discount nominal cash flows at the nominal rate and real cash flows at the real rate; the NPV is the same either way. Never mix (nominal flows with a real rate, or real with nominal). Market interest rates are usually quoted in nominal terms, so nominal flows are typically discounted at the nominal rate.
Rule 4: Separate investment and financing decisions. Forecast cash flows as if the project were all-equity financed, ignoring financing side effects (extra value from financing is considered separately). This isolates whether the project itself adds value. Debt proceeds, interest and principal repayments are therefore ignored.
Rule 5: Forecast after-tax cash flows. Taxes are cash expenses just like wages and raw-material costs.
Classifying cash flows
| Item | Treatment |
|---|---|
| Capital expenditure / investment | Incremental cash outflow |
| Disposal of plant and machinery (salvage value) | Incremental cash inflow |
| Initial investment in working capital | Outflow |
| Increase in working capital | Outflow (extra investment needed) |
| No change in working capital | No cash flow |
| Decrease in working capital | Inflow (working capital released) |
| Recovery of working capital at project end | Inflow |
| Opportunity cost of using an existing resource | Outflow |
| Sunk cost | No cash flow |
| Positive incidental effect | Inflow |
| Negative incidental effect | Outflow |
| Allocation of existing overhead (no extra cost) | No cash flow |
| Tax savings from project losses (losses offset profits elsewhere in the firm) | Inflow |
| Proceeds from borrowing to fund capex | Ignore (financing decision) |
| Periodic interest on that loan | Ignore (financing decision) |
New manufacturing plant: item by item
Site already owned, but existing buildings must be demolished.
| Item | Incremental cash flow? |
|---|---|
| Market value of the site and existing buildings | Yes, outflow (opportunity cost of an existing resource) |
| Demolition and site clearance costs | Yes, outflow |
| Cost of a new access road put in last year | No (sunk) |
| Lost earnings on other products due to executive time on the new facility | Yes, outflow (negative incidental effect) |
| Part of the cost of leasing the president's jet | No (allocation of existing overhead) |
| Future depreciation of the new plant | No (allocation of cost, not a cash expense) |
| Reduction in tax bill from tax depreciation of the plant | Yes, inflow (depreciation tax shield) |
| Initial investment in raw-material inventories | Yes, outflow |
| Money already spent on engineering design of the plant | No (sunk) |
Inflation and NPV: worked example
Outlay Rs. 100 today; real cash flows Rs. 35, 50, 30 at years 1, 2, 3. Nominal discount rate 15%, inflation 10%.
Method 1: nominal flows at the nominal rate. Convert: .
(The lecture states the denominators loosely; the nominal discount rate 1.15 is the one used.)
Method 2: real flows at the real rate.
Both methods give the same NPV (about Rs. 5.5).
Full project example: is it worth undertaking?
Given: investment Rs. 12,000 today in plant and machinery; forecast salvage (net sale proceeds) Rs. 1,949 at end of year 7; for tax, fully depreciated to zero salvage over 6 years straight-line; on sale, tax at 21% on (sale price − net book value); sales, COGS, other costs (excluding depreciation) and working capital forecasts given as nominal amounts; nominal opportunity cost of capital 20%; project tax losses can shield income from the rest of the business; all cash flows at year-end.
Step 1: depreciation (straight line).
Step 2: profit before tax, tax, PAT (year 1 shown): sales 523, COGS 837, other costs 2,200, depreciation 2,000.
A negative tax does not mean the government pays the firm: the project's loss reduces the firm's total taxable income, saving Rs. 948 of tax.
Step 3: OCF (bottom-up; other approaches give the same):
Other years follow the same way.
Step 4: capital investment and disposal.
- Year 0: .
- Year 7 disposal, net of tax: (net book value is 0).
Step 5: working capital (year 1): , so cash flow .
Step 6: net cash flow (year 1):
Step 7: NPV at 20% using the net cash flows from years 0 to 7:
Decision: , so accept; it adds to shareholder wealth.
(The year-0 net cash flow of and the year-7 flow of 3,542 are the lecture's figures; the lecture itemises only the capex and the 1,540 after-tax disposal flow, not the full forecast table.)
Key takeaways
- Five rules: cash not income; incremental; consistent inflation; separate financing; after tax.
- , where .
- Three equivalent OCF forms; the tax-shield form isolates .
- Include opportunity costs, incidental effects, after-sales flows and after-tax salvage; exclude sunk costs, allocated overhead and financing flows.
- Nominal with nominal, real with real: same NPV; never mix.
Revise before your examAll 4 exam tips from this module in one list
- NPV is treated as the most superior technique. When another technique disagrees with NPV, NPV wins.
- Discounted payback fixes the time-value flaw (never accepts negative NPV) but not the other two: arbitrary cutoff and ignoring flows after the cutoff. It can still wrongly reject positive-NPV projects (project A).
- The four IRR pitfalls to recall: (1) no IRR, (2) multiple IRRs (Descartes' rule), (3) cannot tell investment from borrowing (reverse the rule for borrowing), (4) unreliable for mutually exclusive projects differing in scale or economic life (use incremental IRR, first flow negative, same risk).
- When asked to value a project, build NCF each year as OCF + capex/disposal + (−ΔNWC), then discount at the opportunity cost. Remember: depreciation is added back (non-cash) yet still creates a tax shield; salvage is taxed on (sale price − book value); financing flows are excluded; sunk costs and allocated overheads are ignored.
Test yourself on this module
Questions from these notes, one at a time, with feedback. Your weak topics are saved.