Net present value is the right starting point for many investment decisions, but a direct NPV comparison can be misleading when mutually exclusive projects perform the same continuing function for different numbers of years. A three-year machine and a five-year machine do not provide the same amount of service. Choosing the larger NPV without adjusting for life can reward the longer project simply because it operates for more periods.
Equivalent annual annuity, also called equivalent annual value, converts each project’s NPV into a constant annual amount over its life. For cost-only alternatives, the same method is usually called equivalent annual cost. This guide explains when the method is valid, provides formulas, a reusable calculation template, a worked three-year versus five-year example, decision rules and the assumptions that can invalidate the result.
The direct answer: how do you compare projects with unequal lives?
Use this sequence:
- confirm that the projects are mutually exclusive alternatives serving the same repeatable need;
- estimate incremental after-tax cash flows for each project;
- discount each project at a risk-appropriate rate to calculate NPV;
- convert each NPV into an equivalent annual annuity over that project’s life;
- choose the highest positive EAA for value-generating projects, or the lowest equivalent annual cost for cost-only alternatives;
- stress-test replacement, discount-rate, salvage-value, capacity and technology assumptions.
The conversion is not a substitute for good cash flows. It is a way to put unequal lives on a comparable annual basis when replacement is realistic.
Equivalent annual annuity formula
For a project with net present value NPV, discount rate r and life n years:
EAA = NPV × [r / (1 − (1 + r)^−n)]
The bracketed term is the capital recovery factor. It converts a present value into an equal annual amount over n periods.
An equivalent expression divides NPV by the present-value annuity factor:
EAA = NPV / [(1 − (1 + r)^−n) / r]
For a cost-only comparison, calculate the present value of costs as a positive amount and apply the same capital recovery factor:
EAC = PV of costs × [r / (1 − (1 + r)^−n)]
Choose the lower EAC, provided the alternatives deliver equivalent service quality, volume and risk.
When EAA is the right method
Equivalent annual annuity is useful when five conditions broadly hold.
First, the alternatives are mutually exclusive. You are choosing one machine, platform, lease, facility design or process rather than accepting every positive-NPV opportunity.
Second, the assets provide comparable service. A machine producing twice the capacity is not directly comparable unless cash flows incorporate the extra output or costs are normalised per unit.
Third, the need continues beyond the first asset’s life. The organisation expects to replace a three-year machine if it still needs production in year four.
Fourth, replacement is plausible on broadly similar economic terms. Exact replication is unnecessary, but the method becomes weak when technology, regulation or demand will make future replacements fundamentally different.
Fifth, cash-flow risk can be represented by an appropriate discount rate and scenarios. If one alternative carries materially different systematic or project risk, using one rate without analysis can distort the comparison.
When not to use EAA
Do not use EAA automatically whenever lives differ. It is inappropriate when the business need ends on a known date before replacement, when only one cycle is possible, when assets provide different strategic options, or when capacity and quality cannot be made comparable.
Suppose a three-year system supports a contract that ends after three years. A five-year system’s additional life may have no value. Direct NPV over the contract horizon, including terminal or resale value, is more relevant.
EAA also struggles when rapid technological change makes replacement assumptions speculative. A short-life digital system may allow migration to a much better technology, creating flexibility that a constant-annuity comparison misses. Use scenarios or real-options reasoning rather than pretending the replacement is identical.
Finally, do not annualise accounting profit. The method operates on discounted incremental cash flows and NPV, not depreciation-based earnings.
Worked example: three-year Machine A versus five-year Machine B
A manufacturer must choose one machine to perform the same production step. Both provide sufficient capacity and comparable output quality. The need is expected to continue. The after-tax discount rate is 10%.
Machine A costs $240,000 now, generates annual after-tax operating cash inflows of $105,000 for three years and has an after-tax salvage value of $20,000 at the end of year three.
Machine B costs $360,000 now, generates annual after-tax operating cash inflows of $112,000 for five years and has an after-tax salvage value of $35,000 at the end of year five.
Step 1: calculate NPV for Machine A
The present value of the three annual inflows is:
$105,000 × [(1 − 1.10^−3) / 0.10]
The three-year annuity factor at 10% is approximately 2.4869, so the operating inflows have a present value of approximately $261,125.
The present value of salvage is:
$20,000 / 1.10^3 = $15,026
Therefore:
NPV A = −$240,000 + $261,125 + $15,026 = $36,151
Step 2: calculate NPV for Machine B
The five-year annuity factor at 10% is approximately 3.7908. The present value of operating inflows is:
$112,000 × 3.7908 = $424,570
The present value of salvage is:
$35,000 / 1.10^5 = $21,732
Therefore:
NPV B = −$360,000 + $424,570 + $21,732 = $86,302
A direct comparison favours Machine B by $50,151. But B provides two more years of service. We now annualise both NPVs.
Step 3: convert NPV to EAA
The three-year capital recovery factor at 10% is approximately 0.4021:
EAA A = $36,151 × 0.4021 = $14,537 per year
The five-year capital recovery factor at 10% is approximately 0.2638:
EAA B = $86,302 × 0.2638 = $22,768 per year
Machine B still wins, but now the decision has an interpretable basis: it is expected to create approximately $8,231 more equivalent annual value while the repeated need exists.
A copyable spreadsheet template
Build one column per project and use these rows:
| Input or calculation | Project A | Project B |
|---|---|---|
| Initial investment | input | input |
| Annual operating cash flows by year | input | input |
| Working-capital changes | input | input |
| Terminal or salvage cash flow | input | input |
| Project life in years | input | input |
| Discount rate | input | input |
| NPV | NPV() plus time-zero flow |
same |
| Annuity factor | (1-(1+r)^-n)/r |
same |
| Equivalent annual annuity | NPV / annuity factor |
same |
| Decision | higher EAA | higher EAA |
In Excel or Google Sheets, remember that the NPV function discounts cash flows beginning at the end of period one. Add the time-zero investment separately. If the initial outlay is in cell B4 and annual cash flows are B5:B7, a basic formula is =NPV(rate,B5:B7)+B4, with B4 entered as a negative amount. Include salvage and working capital in the appropriate final-year cash flow.
For EAA, use =NPV_result*PMT_factor or calculate =NPV_result/(1-(1+rate)^-life)*rate with careful parentheses. Spreadsheet PMT functions use sign conventions that may display the opposite sign; verify the economics rather than formatting the sign until it looks favourable.
Equivalent annual cost example
Suppose two data-centre cooling systems provide identical capacity. System X has a present value of lifecycle costs of $480,000 over four years. System Y has a present value of costs of $690,000 over seven years. The discount rate is 8%.
The four-year capital recovery factor at 8% is approximately 0.3019, so:
EAC X = $480,000 × 0.3019 = $144,912 per year
The seven-year factor is approximately 0.1921:
EAC Y = $690,000 × 0.1921 = $132,549 per year
System Y has the higher lifecycle present cost but the lower annual equivalent cost. If service and replacement assumptions are credible, Y is economically preferable by about $12,363 per year.
This example shows why “choose the lowest present cost” is not sufficient when cost alternatives have unequal lives and the service must continue.
Replacement-chain analysis: an alternative cross-check
The replacement-chain method repeats each project until both cover a common horizon. For three-year and five-year projects, the least common multiple is 15 years. Project A is repeated five times and Project B three times. Calculate the NPV of each chain, accounting for replacement timing, then compare them over the same horizon.
Under stable replacement assumptions, replacement-chain and EAA methods lead to the same ranking. EAA is usually simpler and avoids building a very long forecast. Replacement chains can be more transparent when replacement cash flows change in known ways, such as scheduled cost inflation or learning effects.
Do not extend a model to an absurd common horizon merely because mathematics permits it. Assets with 7- and 11-year lives produce a 77-year least common multiple, far beyond defensible forecasts. EAA is cleaner, but the real issue is whether perpetual replacement is credible.
Adjust for inflation consistently
Use nominal cash flows with a nominal discount rate or real cash flows with a real rate. Do not mix them. If operating costs and replacement prices inflate differently, model those differences explicitly rather than applying one general factor.
For nominal analysis, expected replacement costs may rise. EAA based on one current cycle assumes the annual equivalent remains representative. If inflation or technology changes the next cycle materially, add replacement scenarios and compare decisions over a finite strategic horizon.
The relationship between nominal rate, real rate and inflation can be represented as:
(1 + nominal rate) = (1 + real rate) × (1 + inflation rate)
The simplified subtraction is adequate only when rates are small and precision is not material.
Include taxes, depreciation and working capital correctly
Investment appraisal uses after-tax cash flows. Depreciation is non-cash, but it can create a tax shield. Salvage value may generate a tax effect relative to tax book value. Working capital invested at the start is a cash outflow and may be recovered at the end.
A robust model separates:
- initial asset and installation costs;
- incremental revenue or cost savings;
- incremental operating cash costs;
- tax on operating profit;
- depreciation tax shield where relevant;
- working-capital investment and recovery;
- after-tax disposal value;
- decommissioning, transition and replacement costs.
Apply the same policy to every alternative. A common error is including implementation and disruption cost for a new technology while ignoring the maintenance shutdown required by the incumbent.
Capacity and quality: compare service, not labels
EAC is valid only for equivalent service. If Machine B produces 20% more units, compare cost per equivalent unit or incorporate additional contribution cash flows. If it improves quality, quantify scrap, rework, warranty, customer retention or price effects where evidence exists.
Avoid unsupported benefits. Create three columns: measured, estimated and unquantified. Include measured and defensible estimates in the base case. Put uncertain strategic benefits into scenarios or qualitative decision criteria. This preserves the integrity of the financial result while allowing management to consider broader value.
Risk and discount rates
Using one corporate hurdle rate for every alternative is convenient but may hide differences in operating, technology or demand risk. At minimum, test a range. If risk is concentrated in specific cash flows, scenario or certainty-equivalent analysis may be clearer than increasing the discount rate until the result “feels conservative.”
For the machine example, test 8%, 10% and 12%; lower operating inflows; delayed commissioning; salvage value of zero; and higher replacement cost. Record the switch point where the ranking changes. A decision that reverses after a small assumption change requires further evidence or contractual protection.
EAA-7 decision checklist
Before approving the result, answer seven questions:
- Equivalent need: Do both alternatives serve the same continuing requirement?
- Economic lives: Are useful lives based on operating evidence rather than accounting depreciation?
- After-tax flows: Are all incremental cash flows, working capital and disposal effects included consistently?
- Replacement realism: Can the shorter-life asset be replaced on broadly comparable terms?
- Risk consistency: Are discount rates and scenarios appropriate for each alternative?
- Service normalization: Are capacity, quality, reliability and downtime reflected?
- Decision robustness: Does the preferred alternative remain superior under credible sensitivities?
If questions one or four fail, use a finite common decision horizon or scenario model instead of EAA. If question six fails, redesign the cash flows before calculating anything.
Common mistakes and repairs
Mistake: comparing raw NPVs. Repair by annualising NPVs when the service is repeatable and lives differ.
Mistake: annualising initial cost only. Repair by calculating full project NPV, including operating, tax, working capital and terminal cash flows.
Mistake: choosing the higher EAA when both are negative without considering rejection. If the investment is optional and every EAA is negative, reject all. For mandatory service, compare EAC and choose the least-cost feasible alternative.
Mistake: assuming identical replacement forever. Repair with finite-horizon scenarios, technology paths and replacement-cost sensitivities.
Mistake: mixing real and nominal inputs. Repair by choosing one consistent basis and documenting inflation assumptions.
Mistake: ignoring capital rationing. EAA ranks mutually exclusive life alternatives; it does not solve allocation across many divisible or indivisible projects under a budget. Use capital-rationing methods and consider strategic constraints.
Mistake: presenting one number without decision context. Report the NPVs, EAAs, major assumptions, sensitivities and switching conditions.
How to write the investment recommendation
Use a short decision memo:
- decision and alternatives;
- continuing business need and service equivalence;
- cash-flow basis and discount rate;
- NPV and EAA/EAC results;
- three assumptions that drive the ranking;
- downside scenario and switching point;
- recommendation, conditions and owner.
For the worked example: “Approve Machine B subject to vendor confirmation of five-year maintenance performance and a maximum installed cost of $375,000. At the base case, B creates EAA of approximately $22,768 versus $14,537 for A. The recommendation should be revisited if annual cash inflow falls below the documented switching threshold or the business need becomes shorter than five years.”
Practical next step
Take one live decision involving equipment, software, leases or replacement. Verify that the alternatives provide equivalent service and that the need repeats. Build the cash flows, calculate NPV, convert to EAA or EAC, and complete EAA-7. Ask a colleague to challenge replacement and terminal assumptions before the approval meeting.
For deeper study of investment appraisal, capital allocation, valuation and decision-quality financial modelling, review MTF Institute’s Strategic Finance programme. Compare the published curriculum with your role requirements and confirm current enrolment terms. The relevant capability is not memorising the formula; it is knowing when annualisation supports a valid decision and when its assumptions do not.
Conclusion
Equivalent annual annuity makes unequal project lives comparable by converting NPV into annual economic value. Equivalent annual cost applies the same logic to mandatory cost alternatives. The method is powerful when projects are mutually exclusive, service is equivalent, the need continues and replacement is realistic. Use the formula only after building sound after-tax cash flows, then test capacity, risk, inflation, salvage and technology change. A transparent EAA model turns an apparently simple ranking into a decision that managers can inspect and defend.