EV/EBIT vs. EV/EBITDA: Choosing the Right Multiple for Capital-Intensive Industrials

Introduction

A company can trade at 10x on an EV/EBITDA basis yet at 18x on EV/EBIT once depreciation is recognized. For asset-light businesses where D&A is immaterial (software, professional services, financial services), the gap barely matters and either multiple produces a useful comparison. For a heavy equipment manufacturer with billions of dollars in factories, forges, and machining centers, that 8-turn gap represents a fundamentally different statement about what the business is worth relative to its economic earning power. Choosing the wrong multiple for the wrong business produces valuations that are not just imprecise; they are systematically misleading.

This article explains when EV/EBIT should be preferred over EV/EBITDA in industrials, how depreciation policies distort EBITDA comparisons across companies, and how the maintenance vs. growth capex distinction informs the decision.

The theoretical case for EV/EBITDA is that stripping out depreciation provides a capital-structure-neutral, D&A-policy-neutral view of operating performance. This logic holds for sectors where D&A is below 2% of revenue and bears little relationship to ongoing cash requirements. Software companies amortize development costs that do not require cash reinvestment. Financial services firms depreciate office furniture. In these contexts, removing D&A genuinely improves comparability.

In capital-intensive industrials, the logic breaks down. Factories, CNC machines, paint lines, forges, assembly equipment, and test facilities physically wear out and must be replaced to sustain production capacity. A company that stops investing in its physical asset base will see throughput decline, quality deteriorate, and maintenance costs spike within 3-5 years. Caterpillar recorded approximately $2.15 billion in total D&A during 2024 on roughly $65 billion in revenue: about 3.3% at the consolidated level, though the manufacturing segments run significantly higher when the financial products division is excluded.

Depreciation as Economic Cost

In capital-intensive industrials, depreciation approximates the annual cost of consuming physical productive capacity that must be eventually replaced to sustain operations. Unlike amortization of intangible assets (goodwill, customer relationships, patents) which may not require cash reinvestment, depreciation of manufacturing equipment and facilities almost always corresponds to future maintenance capital expenditure. When EBITDA strips out this depreciation, it implicitly assumes that the physical assets generating the revenue are free to maintain, which is economically indefensible for a company running $5-20 billion in PP&E. EV/EBIT includes depreciation, producing an earnings figure that reflects the real economic cost of asset consumption.

D&A as a percentage of revenue varies significantly across industrials sub-sectors. Railroads typically report D&A at 10-15% of revenue (reflecting the massive track, locomotive, and rolling stock asset base). Heavy manufacturers run 5-10%. Specialty components companies with lighter manufacturing run 3-5%. And asset-light services companies run 1-3%. The higher the D&A intensity, the more misleading EV/EBITDA becomes and the more strongly EV/EBIT should be preferred.

Even among companies with similar economic profiles, differences in depreciation accounting can make EV/EBITDA comparisons misleading in ways that EV/EBIT avoids.

Consider two competing steel fabricators, each with $500 million in revenue and $100 million in true economic operating profit. Company A invested $400 million in a brand-new facility five years ago and depreciates it over 15 years (roughly $27 million per year). Company B operates a 25-year-old facility that is nearly fully depreciated, running only $5 million in annual depreciation. Their EBITDA figures look different: Company A reports $127 million EBITDA, Company B reports $105 million EBITDA. On EV/EBITDA, Company B looks more efficient. But Company B's facility will need a massive replacement investment in the near term (the old equipment is wearing out), while Company A's modern facility has years of remaining useful life. On an EV/EBIT basis, both companies show approximately $100 million, accurately reflecting their equivalent economic earning power.

The distortion also works in reverse. Companies that have recently completed major capital investments report elevated depreciation that compresses EBITDA margins relative to peers with older assets. On EV/EBITDA, the recently invested company looks less profitable; on EV/EBIT, it looks comparable because both multiples are being applied to an earnings figure that already accounts for the asset cost.

EV/EBITDA retains a legitimate role in industrials valuation in several specific contexts.

Comparing companies with similar capital intensity. When all companies in a comp set have similar D&A ratios and asset ages, the depreciation distortion is minimized, and EV/EBITDA provides a clean comparison of operating performance before asset consumption. A comp set of five specialty chemical companies all running 4% D&A/revenue will produce comparable EV/EBITDA figures without significant depreciation distortion.

Cross-border comparisons. Depreciation policies vary by jurisdiction (accelerated depreciation in some tax regimes, straight-line in others). When comparing a US industrial with a European peer, EV/EBITDA can be more comparable because it removes the jurisdiction-specific depreciation methodology. However, this advantage is only partial: the underlying capital intensity difference between companies still matters for economic analysis.

LBO analysis. PE sponsors use EV/EBITDA as the primary leverage metric (debt/EBITDA) because EBITDA approximates cash earnings available for debt service before reinvestment decisions. In this context, EBITDA is not being used as a valuation metric but as a cash flow proxy, and the sponsor separately models capex (split between maintenance and growth) to determine actual free cash flow. The LBO model captures the depreciation economics through the capex line rather than the EBITDA line.

In practice, many industrials bankers present both multiples in their valuation analysis: EV/EBITDA for consistency with market convention and LBO analysis, and EV/EBIT as a cross-check that accounts for capital intensity. When the two multiples tell different stories (e.g., a company looks cheap on EV/EBITDA but expensive on EV/EBIT), the discrepancy usually signals a depreciation-related issue that requires further investigation.