Capitalised Development Costs at Nordwind AG
Nordwind AG reports under IFRS. For 2024 it reports net income of EUR 200m, cash flow from operations of EUR 350m and cash flow from investing of EUR -300m. The investing outflow comprises EUR 180m of property, plant and equipment capex and EUR 120m of development costs that were capitalised rather than expensed. Total depreciation and amortisation charged in 2024 was EUR 150m, of which EUR 40m is amortisation of development costs capitalised in earlier years. Nordwind's tax rate is 25%. The analyst covering Nordwind wants to compare it with a US peer that expenses all research and development spending as incurred.
1. Restate Nordwind's net income and cash flow from operations onto an expense-as-incurred basis, and show the adjustments.
Read the worked answer
Net income. Expensing EUR 120m reduces pre-tax profit by 120m, so the after-tax effect is 120 x (1 - 0.25) = 90m.
Adjusted NI = 200 - 90 = EUR 110m.
The capitalised development cost is not a permanent difference - it is timing. Under expensing the charge hits the income statement now; under capitalisation it arrives later as the EUR 40m amortisation already in the P&L.
Cash flow from operations. The cash was always spent. Capitalising simply routes it through investing, so:
- Adjusted CFO = 350 - 120 = EUR 230m
- Adjusted CFI = -300 + 120 = EUR -180m
Free cash flow is unchanged: 350 - 300 = 50m before adjustment is not the right FCF (that mixes definitions); on a capex-only basis FCF = CFO - capex = 230 - 180 = EUR 50m, identical to 350 - 300. This is the tell: capitalisation moves cash between categories, it does not create it.
2. Nordwind's reported EBITDA is EUR 420m and the US peer trades at 9.0x EV/EBITDA. What happens to the comparison once the development spending is expensed, and which multiple should the analyst trust?
Read the worked answer
EBITDA adjustment. The EUR 120m is an operating expense under expensing and is not depreciation, so it is not added back:
Adjusted EBITDA = 420 - 120 = EUR 300m.
Effect on the multiple. Reported EBITDA of 420 makes Nordwind look cheaper than it is: at the same EV, EV/EBITDA falls by a factor of 300/420 = 0.714, i.e. Nordwind appears roughly 29% cheaper than the peer on an unadjusted basis purely from the accounting choice. On a like-for-like basis it is 1/0.714 = 1.40x more expensive than it first appears.
Other affected ratios: reported P/E uses inflated earnings of 200 rather than 110; reported CFO looks stronger than a capitaliser deserves; ROA is flattered by a larger asset base offset by higher earnings.
Which multiple to trust. None of the reported multiples are usable without the adjustment. EBITDA is the most contaminated of the common metrics here because the capitalised amount sits outside D&A. Prefer a multiple built on adjusted EBITDA, or an EV/FCF or EV/EBIT comparison where the timing distortion is visible and stated.
Caveat: IFRS permits capitalisation when the recognition criteria are met, so the accounting may be legitimate - the issue is comparability with a US GAAP peer, not fraud.
3. Nordwind also extended the estimated useful life of its plant from 8 to 12 years, cutting 2024 depreciation by EUR 25m. How should the analyst treat this, and what does it do to EBITDA versus CFO?
Read the worked answer
EBITDA: no effect. Depreciation is added back in EBITDA by construction, so a change in the depreciation estimate does not touch the numerator. Any analyst who shows an EBITDA uplift from a useful-life change has miscalculated.
EBIT and net income: +25m pre-tax, +18.75m after tax (25 x 0.75). If the 25m is inside the 150m D&A figure, reported EBIT and EPS are boosted by an accounting estimate, not by operations.
CFO: +6.25m only (25 x 25%), because depreciation is non-cash and the only cash effect is the reduced tax bill. A common error is to add the full 25m to CFO.
What to check.
- Whether the extension is consistent with the peer's useful lives and with the asset's actual physical/economic life; a change made in a year of margin pressure deserves scrutiny.
- Whether the change is disclosed as a change in estimate with a quantified effect - required under both IFRS and US GAAP.
- Recompute trend margins using the old depreciation rate to see whether operating performance genuinely improved.
An estimate change is legal and often justified (a plant genuinely lasting longer), but it is a low-quality source of earnings and it should be normalised out of any comparison of operating profitability.