DuPont Analysis: How to Decompose ROE Into 3 Real Drivers

Most investors see ROE and think "good business." DuPont analysis decomposes that one number into three drivers — and shows why NKE at roughly 18% ROE is a fundamentally different bet from F at the same level.

What Is DuPont Analysis, Actually?

It is a framework invented in the 1920s at the DuPont Corporation to figure out why a business earns the return on equity it does. Instead of treating ROE as a single number, DuPont breaks it into three independent drivers:

ROE = (Net Income / Sales) × (Sales / Assets) × (Assets / Equity)

Which simplifies to:

ROE = Net Margin × Asset Turnover × Equity Multiplier

Each term answers a different question. Net Margin asks: how much profit do you make on every dollar of sales? Asset Turnover asks: how many dollars of sales do you generate from every dollar of assets? Equity Multiplier asks: how much leverage are you using to amplify those returns?

Two businesses can hit the same 18% ROE through completely different paths. One might do it with roughly 25% margins, 0.7x asset turnover, and 1x equity multiplier — a software business. Another might do it with about 4% margins, 3x asset turnover, and 1.5x leverage — a discount retailer. Both look identical on a screening tool. DuPont is what separates them.

How Do You Calculate the 3-Step DuPont?

You need three numbers from the financial statements. Here is the practical workflow:

Step 1: Net Margin = Net Income ÷ Revenue. Both are on the income statement. For most companies this is the cleanest input — but watch for one-time items that inflate or deflate the period.

Step 2: Asset Turnover = Revenue ÷ Average Total Assets. Average means (Beginning Assets + Ending Assets) ÷ 2, using the balance sheet at the start and end of the period. Using ending assets only is a common shortcut that creates noise during rapid balance-sheet growth.

Step 3: Equity Multiplier = Average Total Assets ÷ Average Total Equity. Same average treatment. The result tells you how many dollars of assets each dollar of equity supports.

Step 4: Multiply all three. ROE = Margin × Turnover × Multiplier.

If your arithmetic does not reconcile to the reported ROE, the discrepancy is usually buybacks (which shrink equity faster than net income captures) or one-time accounting charges. Both should be adjusted for trend analysis.

Three Real Examples Across the S&P 500

Using the most recent quarterly filings (Q1 2026), here is how three very different businesses arrive at high ROE:

Look at MSFT versus COST: both land near 30% ROE, but Microsoft gets there with software-quality margins and almost no leverage, while Costco gets there with razor-thin margins compensated by extraordinary turnover. Both are excellent businesses — but they react completely differently to recessions, interest rate moves, and supply shocks.

KO is the textbook leverage case. The brand commands premium margins, but turnover is mediocre, and the equity multiplier of roughly 4.5x is what pushes ROE near 50%. Strip out the leverage and the underlying return on assets is closer to roughly 11% — still good, but not nearly as exceptional as the headline ROE implies.

F shows the dark mirror. Equity multiplier of roughly 7.5x is comparable to many banks, yet the underlying margin is only about 2%. The leverage masks the operational fragility for long stretches — until a recession or recall cycle exposes it.

When Should You Use the 5-Step Version?

You should use 5-step DuPont anytime financing decisions matter — which, in 2026's rate environment, is almost every comparison. The 5-step formulation separates the operating business from the financing structure:

ROE = (Net Income / EBT) × (EBT / EBIT) × (EBIT / Revenue) × (Revenue / Assets) × (Assets / Equity)

The first term is the tax burden, the second is the interest burden, and the third is operating margin. The last two are turnover and leverage. By splitting them this way, you can isolate whether an ROE is being earned through operations or through cheap financing.

This is especially important when comparing companies across capital structures. BRKB at roughly 12% ROE looks unremarkable next to META at about 35% — until you decompose and find Berkshire is doing it almost entirely through operating quality with low tax and interest burdens, while Meta runs higher leverage and a different tax profile. See our fundamental-analysis primer on tax adjustments for the deeper version.

What Are the Most Common DuPont Mistakes?

Three errors show up repeatedly. First, treating the equity multiplier as a free lunch. Doubling leverage doubles ROE in good times — and doubles the path to zero in a downturn. The graveyard of high-ROE financials in 2008 is a permanent reminder that high-multiplier ROE is not free.

Second, using point-in-time balance sheet numbers instead of averages. A company that did a roughly $20 billion buyback at quarter-end will show a misleadingly tiny equity number — which flatters the multiplier and inflates the ROE estimate. Always use averaged numbers.

Third, ignoring share buybacks in the trend. Buybacks reduce the denominator (equity) faster than the numerator (net income) usually responds. ROE can rise meaningfully through buybacks alone, without any operational improvement. Investors who anchor on rising ROE without checking for buyback intensity will misread the quality of the business.

For a structural read on whether share buybacks are creating real value, look at shareholder yield — it captures buybacks plus dividends in one ratio.

Pro Tips for DuPont in 2026

In a higher-for-longer rate regime, three habits separate signal from noise.

Anchor on operating return (margin × turnover) before the multiplier. Two companies can have similar ROE today, but the one earning it through operating quality survives a rate shock. The one earning it through leverage gets repriced when refinancing terms change.

Track all three components over 8 quarters minimum. A company where margin is rising while turnover is falling is doing something specific (likely premiumization or mix shift). A company where the multiplier is rising while everything else is flat is just using more debt.

Cross-check against ROIC. Return on Invested Capital adjusts for tax and capital structure, so it strips out the leverage component entirely. If ROE is rising while ROIC is flat, that is leverage doing the work — and the market eventually figures it out. Use our ROIC primer for the deeper framework.

NKE is the cleanest case study of a clean DuPont profile under stress. Inventory cycles compressed turnover in 2024, margins followed turnover down, and ROE fell roughly 1,000 basis points without any change in the equity multiplier. That is a real operational story — not a financial-engineering story.

When DuPont Analysis Misleads You

The framework breaks down for three categories. Banks and insurers have inverted balance sheets where deposits and policy reserves are not "leverage" in the operating sense — they are the product. Use bank-specific frameworks like return on tangible equity instead.

Early-stage growth companies with negative net income generate a negative net margin, which produces a negative ROE that DuPont cannot rationalize. Use revenue growth, gross margin trajectory, and FCF runway for those names.

Conglomerates with diverse segments are similarly tough — a single consolidated DuPont averages across segments with completely different economics. Segment-level decomposition would be ideal, but disclosure rarely supports it for names like GE or CAT.

A final caveat: DuPont is a snapshot, not a verdict. A high-ROE business can deteriorate quietly for years while the headline number holds up because management uses buybacks to flatter the multiplier. Always read trends, not single quarters.

Ready to analyze these stocks yourself? Search any ticker on MainRatios to see valuations from 6 legendary investors - free.