Northfield Foods: an automation business case that turns on 1.3%
Northfield Foods is considering a robotic palletising line for its dry-goods plant. The installed cost is $3,000,000, with a six-year expected life and no residual value. The project would remove two shifts of manual palletising, saving $780,000 a year in labour and workers' compensation premiums before tax; maintenance and consumables are already netted in that figure. The company tax rate is 30%, and the tax office allows straight-line depreciation over the six years. Northfield's group weighted average cost of capital is 8%, but the automation committee's policy requires a risk-adjusted 10% for stand-alone plant projects. The project also requires $250,000 of spare-parts inventory, funded at the start and recovered at the end of year six. The engineering manager's base case assumes savings begin in year one. The CFO has asked the finance business partner to 'make the numbers work' and has suggested using the 8% group rate and excluding the working capital because 'it comes back anyway'.
1. Calculate the net present value, internal rate of return and payback for the project at the 10% risk-adjusted rate, ignoring working capital, and interpret the result.
Read the worked answer
Annual depreciation = $3,000,000 / 6 = $500,000.
After-tax operating cash flow = (savings - depreciation) x (1 - 30%) + depreciation = ($780,000 - $500,000) x 0.7 + $500,000 = $196,000 + $500,000 = $696,000 a year. (Check: $780,000 x 0.7 + $500,000 x 0.3 = $546,000 + $150,000 = $696,000.)
NPV at 10%. Six-year annuity factor at 10% = 4.3553. PV of cash flows = $696,000 x 4.3553 = $3,031,300. NPV = $3,031,300 - $3,000,000 = +$31,300.
IRR. Set $696,000 x annuity factor = $3,000,000, so the factor is 4.3103. Interpolating between 10% (4.3553) and 11% (4.2305) gives about 10.4% - only 0.4 percentage points above the hurdle.
Payback = $3,000,000 / $696,000 = 4.3 years of a six-year life. Discounted payback is effectively the full six years, since NPV is barely positive.
Interpretation. The project earns almost exactly its cost of capital. A positive NPV of about 1% of the outlay is not a decision - it is the absence of one. The financial case cannot carry the decision on its own; the decision must rest on the reliability of the $780,000 saving and on strategic factors. The IRR also flatters the project slightly relative to NPV because the cash flows are level and the life is short, so NPV should govern.
2. Assess the sensitivity of the case and recommend how Northfield should manage the risks the analysis reveals.
Read the worked answer
Break-even savings. Required after-tax cash flow for NPV = 0 is $3,000,000 / 4.3553 = $688,825. Working back: savings x 0.7 + $150,000 = $688,825, so savings = $769,750. The cushion is $10,250 on $780,000 - 1.3%. Each $10,000 shortfall in annual savings reduces NPV by $30,487.
Other sensitivities:
- One-year delay in benefits - a common outcome with installation and commissioning: the lost PV is $696,000 / 1.10 = $632,700, taking NPV to about -$601,000.
- 10% capital overrun of $300,000: NPV falls to about -$269,000.
- A 2% rise in the discount rate (to 12%) is enough to make the project clearly negative.
What this tells us. The decision is dominated by delivery risk, not by the discount rate. The right response is contractual and staged, not analytical:
- Vendor performance guarantees with liquidated damages for throughput, uptime and commissioning dates, plus 10-15% retention until acceptance testing.
- Phase the project - commission one line first, verify the labour saving, then commit the balance. This is a real option to abandon, and it is worth more than the $31,000 NPV.
- Model the labour saving bottom-up: actual roster reduction, redeployment, redundancy cost and award constraints. 'Two shifts removed' is an assumption until HR confirms it.
- Quantify the non-financial benefits separately - reduced manual handling injuries, insurance premiums, difficulty filling shift roles - and disclose them as a strategic premium rather than burying them in the cash flows.
- Post-implementation review at 12 and 24 months, with actual savings reported to the board against this case, so that optimism bias is visible and corrected.
3. Evaluate the CFO's two suggestions and advise the finance business partner on how to proceed.
Read the worked answer
The discount rate. The 8% group WACC reflects the average risk of Northfield's existing operations. This project is a stand-alone technology investment with uncertain, firm-specific savings and little salvage value - the definition of above-average risk. The automation committee's 10% policy exists precisely to stop average rates being applied to non-average projects. At 8%, the NPV becomes $217,500 and the decision flips from marginal to comfortable, which is exactly why the rate must be defensible rather than convenient. If the committee believes 10% is too high it can change the policy prospectively, with reasons; it cannot be overridden case by case.
The working capital. The $250,000 is real cash, committed at the start and tied up for six years. 'It comes back' ignores the time value. At 10% the recovery is worth $250,000 x 0.5645 = $141,100, so the net cost is $108,900, and NPV becomes $31,300 - $108,900 = -$77,600. Excluding it overstates the project by that amount. The maximum justifiable installed cost, given the working capital, is about $2.92 million - a useful negotiation target with the vendor.
How to proceed. Present the appraisal at the policy rate including working capital, alongside a transparent sensitivity table showing NPV at 8% and 10%, with and without working capital, and the break-even savings. Document the disagreement. Then escalate: an instruction to alter assumptions to reach a predetermined answer compromises objectivity and integrity, and the appropriate route is to inform the CFO in writing that the request cannot be met as framed, and, if it persists, to refer the matter to the audit committee or the appropriate governance body. Offer the constructive alternative - a properly staged business case with vendor guarantees - which is how a marginal project should be approved.