Analyzing Historical Performance — Reorganizing Statements and Normalizing ROIC

GAAP financial statements are designed for creditors and regulators — not for value investors and M&A analysts. They mix operating performance with financing decisions (interest expense reduces income), include non-operating items (investment gains, lawsuit settlements) alongside operating results, and apply accounting conventions (depreciation schedules, revenue recognition timing) that can distort period-to-period comparison. The first step in any valuation is to undo these distortions and reconstruct the economic performance of the operating business.

Even after separating operating from non-operating items, trailing NOPAT may be distorted by accounting choices and one-time events. McKinsey identifies five major normalization adjustments that materially affect the ROIC calculation:

• Normalization 1 — Non-recurring charges and gains: restructuring charges, litigation settlements, acquisition-related expenses, and gains on asset sales should be excluded from normalized NOPAT. The question is whether the item is truly non-recurring. A company that incurs 'restructuring charges' in 6 of the last 8 years is effectively running a recurring restructuring program — some portion should remain in normalized NOPAT (McKinsey's rule: if the item appears in more than 2 of the last 5 years, treat it as semi-recurring and adjust accordingly).
• Normalization 2 — Revenue recognition timing: under ASC 606, revenue recognition has become more conservative for long-term contracts, subscriptions, and multi-element arrangements. Changes in revenue recognition policy create year-over-year comparability problems. Identify the year of policy change, compute the pro-forma revenue under the new policy for all comparison years, and use the restated series for trend analysis.
• Normalization 3 — Pension expense accounting: GAAP pension expense includes both current service cost (genuinely operating) and interest cost, expected return on plan assets, and amortization of actuarial gains/losses (all financing and actuarial). The operating component is the current service cost; remove the rest from NOPAT and handle the pension obligation separately in the EV bridge.
• Normalization 4 — Inventory accounting (LIFO vs. FIFO): for US companies using LIFO (legal under US GAAP, not under IFRS), COGS is overstated in inflationary periods and understated in deflationary periods. The LIFO reserve on the balance sheet quantifies the cumulative difference. Add the change in LIFO reserve to COGS to convert to FIFO-equivalent COGS — then NOPAT and inventories become comparable to IFRS peers.
• Normalization 5 — Operating lease capitalization (for pre-ASC 842 data): before the 2019 effective date of ASC 842, operating leases were off-balance-sheet. Capitalize the operating lease obligations (multiply annual lease expense by an appropriate multiple, typically 6–8×), add to invested capital, add back implied depreciation to NOPAT, and separately deduct the implied interest. This adjustment is essential for comparisons involving retailers, restaurants, airlines, and real estate-intensive businesses.

With normalized NOPAT in hand, you can compute a clean trailing ROIC series. This becomes the empirical anchor for the forward ROIC assumptions in the DCF model — assumptions that deviate significantly from trailing ROIC must be explicitly justified with a business reason.

The ROIC trend reveals something important: the company has been consistently improving ROIC at approximately 0.4–0.5 percentage points per year. The analyst must now determine: (1) How long can this improvement continue? (2) Is there a structural ceiling (industry median ROIC for this sector is ~15%)? (3) Is the improvement driven by a specific initiative that has a defined endpoint? These questions inform the forward ROIC assumption directly.

• Use average invested capital (beginning + ending IC) / 2 in the ROIC denominator: beginning IC reflects the capital deployed during the year; ending IC reflects the result of the year's investment. The average is more accurate than either endpoint alone, particularly for businesses with seasonal capital patterns or significant mid-year capital events.
• The ROIC to forecast connection: McKinsey's guidance is that the forecast period ROIC should (a) begin from the trailing normalized ROIC, (b) improve or decline based on specific, articulable business drivers, and (c) converge toward a long-run level that reflects the business's competitive position in the terminal year. A company currently at 13.8% ROIC in a 10% WACC environment forecasting sustained 25% ROIC over 10 years must provide compelling competitive advantage evidence — otherwise the assumption fails the sanity check.
• Cross-sectional benchmarking: the trailing ROIC must be compared to industry peers with similar capital structures and accounting policies. A 13.8% ROIC looks mediocre in a software industry (median 30%+) but strong in specialty retail (median 10–12%). This context is essential for both the assessment of competitive position and the forecast of ROIC sustainability.