Enterprise Value and EV/EBITDA, From the Filings
Enterprise value (EV) is what a whole company is worth to all its investors: the market value of its shares plus its debt, minus its cash. Divided by EBITDA it gives EV/EBITDA, a valuation multiple that, unlike the P/E, is much less affected by how the company is financed.
Market cap prices a company’s shares. Enterprise value prices the whole business, which means it has to count what the shares sit behind: the debt that lenders are owed ahead of shareholders, and the cash that would come with the company if someone bought all of it.
This page gives the formula, works it for McDonald’s from its latest 10-Q, then measures seven large companies from their own SEC filings to show how far EV moves from market cap and how EV/EBITDA can reorder a ranking built on P/E.
How it works
The formula. Enterprise value = market cap + debt + noncontrolling interests - cash and short-term investments.
- Market cap here is the 25 Sep 2026 close times the shares outstanding on the latest filing’s cover page.
- Debt is borrowing on the latest balance sheet: commercial paper, short-term borrowings and the current and long-term parts of long-term debt. Lease liabilities are left out on this page, a choice discussed below.
- Noncontrolling interests are the parts of subsidiaries owned by outsiders. Of the seven companies, only Walmart reports one, at $6.268 billion.
- Cash is cash and cash equivalents plus short-term securities. Longer-dated investments are left out.
Why add debt and subtract cash. A buyer of the whole company takes on its debt and gets its cash. Two companies with the same market cap can cost very different amounts to own outright.
EV/EBITDA. EBITDA is operating income with depreciation and amortization added back. Both EV and EBITDA belong to all investors, lenders and shareholders alike, so the ratio compares like with like. The P/E ratio compares a share price with profit left after interest, so borrowing changes it.
A worked example
McDonald’s on 25 Sep 2026, from its 10-Q for the quarter to 30 Jun 2026 (filed 7 Aug 2026).
- Market cap: $236.50 close x 707,641,531 shares = $167.357 billion.
- Debt: $39.863 billion of long-term debt. The filing’s data shows no short-term borrowings or current maturities.
- Cash: $0.822 billion.
- EV: $167.357 + $39.863 - $0.822 = $206.398 billion, or 23.3% more than the market cap.
- Trailing twelve-month operating income: $12.393 billion for 2025 + $6.292 billion for the first half of 2026 - $5.880 billion for the first half of 2025 = $12.805 billion.
- Depreciation and amortization, the same way: $2.199 + $1.131 - $1.064 = $2.266 billion.
- EBITDA: $12.805 + $2.266 = $15.071 billion. EV/EBITDA: $206.398 / $15.071 = 13.70.
The debt is most of the story. A buyer paying the market price for every share would still owe lenders almost $40 billion, which is why a debt-heavy company can look cheaper on P/E than it is to own.
The original data
Seven large US companies on 25 Sep 2026 prices, each from its latest 10-Q or 10-K. Operating income and D&A are trailing twelve months, built from the filings as in the example. P/E is the trailing figure from the file behind this site’s P/E page; McDonald’s is computed the same way. Dollar figures are in billions.
| Company | Market cap | Debt | Cash and short-term securities | EV | EV vs market cap | EBITDA | EV/EBITDA | P/E |
|---|---|---|---|---|---|---|---|---|
| Nvidia | $5,424.2 | $33.4 | $56.6 | $5,401.0 | -0.4% | $201.3 | 26.8 | 28.5 |
| Apple | $4,977.6 | $84.3 | $62.4 | $4,999.6 | +0.4% | $168.0 | 29.8 | 39.1 |
| Microsoft | $3,832.8 | $40.3 | $76.8 | $3,796.3 | -1.0% | $194.2 | 19.5 | 28.8 |
| Amazon | $2,693.0 | $132.5 | $123.0 | $2,702.6 | +0.4% | $168.9 | 16.0 | 20.1 |
| Walmart | $856.7 | $50.4 | $11.5 | $901.8 | +5.3% | $47.4 | 19.0 | 39.1 |
| Costco | $409.2 | $5.7 | $20.0 | $394.9 | -3.5% | $13.8 | 28.6 | 46.4 |
| McDonald’s | $167.4 | $39.9 | $0.8 | $206.4 | +23.3% | $15.1 | 13.7 | 19.2 |
Walmart’s EV also includes its $6.268 billion of noncontrolling interests. The balance-sheet dates run from 10 May 2026 (Costco) to 31 Jul 2026 (Walmart). Microsoft’s D&A is its depreciation plus amortization of intangible assets, both from its 10-K for the year to 30 Jun 2026. Every input, including the filing each came from, is in the enterprise value table.
For the four largest, EV and market cap were within about 1% of each other, because their debt and cash were small next to their market value and largely offset each other. The gap opened only where one side was lopsided: McDonald’s (+23.3%) and Walmart (+5.3%) on the debt side, Costco (-3.5%) on the cash side.
EV/EBITDA against P/E
The two multiples put the seven in a different order. From lowest to highest, EV/EBITDA runs McDonald’s, Amazon, Walmart, Microsoft, Nvidia, Costco, Apple. P/E runs McDonald’s, Amazon, Nvidia, Microsoft, then Apple and Walmart level at 39.1, then Costco.
Walmart moves the most. Its P/E of 39.1 was second-highest of the seven, level with Apple’s to one decimal, but its EV/EBITDA of 19.0 was third-lowest. Walmart’s depreciation and amortization is large next to its operating income, $15.093 billion against $32.280 billion over the trailing twelve months, and EBITDA adds it back while earnings do not.
Nvidia moves the other way. Third-lowest on P/E at 28.5, it was fifth on EV/EBITDA at 26.8, because its depreciation and amortization, $3.687 billion, was small next to $197.579 billion of operating income.
When it fails
It fails when the definitions differ. There is no single official formula, and the choices move the number:
- Long-dated investments. Apple also held $84.118 billion of non-current marketable securities. Counting them as cash lowers its EV from $4,999.6 billion to $4,915.5 billion and its EV/EBITDA from 29.77 to 29.27.
- What counts as D&A. Amazon’s cash-flow statement shows $75.200 billion of depreciation and amortization over the trailing year, while the depreciation it reports on its own was $49.741 billion. EV/EBITDA is 16.0 on the first and 18.8 on the second.
- Leases. Many data services add lease liabilities to debt. Doing so would raise EV most for companies that rent their stores or offices, and would need EBITDA adjusted to match.
It fails for banks. A bank’s deposits and borrowings are its raw material, not financing in the usual sense, so debt minus cash says little. JPMorgan Chase is left out here for that reason.
It fails when EBITDA is small or negative. A company with little EBITDA gets a huge or meaningless multiple, and EBITDA ignores the spending needed to replace worn-out equipment, which is why free cash flow is often checked alongside it.
And it mixes dates. The share price is from one day; the debt and cash are from the last balance sheet, up to four and a half months earlier for Costco; the EBITDA covers the year before that.
Related
Market cap is the starting point of the formula, and EBITDA is the profit it is divided by. The P/E ratio is the multiple EV/EBITDA is most often set against, and the debt-to-equity ratio shows how much borrowing sits behind a share price. Valuation covers the wider set of methods these multiples belong to.
Write down what went into the debt and the cash before comparing two EV/EBITDA figures. Leases, long-dated investments and the D&A line are where two data services quietly disagree.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.