How much profit does a company actually squeeze out of everything it owns - the factories, the inventory, the cash, the receivables - is a different question than how much profit it returns to shareholders. Return on assets answers the first one. We ran it on 8 real large-cap names using live September 2026 data. Nvidia topped 60%. JPMorgan sat at 1.3% - and that's completely normal for a bank.
Return on assets, or ROA, is a profitability ratio that strips out one variable most metrics don't: how the company paid for what it owns. Where return on equity only looks at the shareholders' slice of the balance sheet, ROA looks at the whole thing - every asset the company controls, whether it was funded with equity or with debt.
The Formula
ROA = Net Income ÷ Average Total Assets
Total assets is everything on the left side of the balance sheet: cash, receivables, inventory, property and equipment, goodwill, everything. Net income is profit over the period, usually trailing twelve months.
A worked example: a company earning $8 billion in net income with $80 billion in average total assets has an ROA of 8 ÷ 80 = 10%. Every dollar of assets the company controls - regardless of whether shareholders or lenders paid for it - generated 10 cents of annual profit.
What 8 Real Stocks Score Right Now
Using live TTM fundamentals as of September 2026:
| Company | Sector | Trailing ROA |
|---|---|---|
| Nvidia (NVDA) | Technology | 60.2% |
| Apple (AAPL) | Technology | 33.6% |
| Microsoft (MSFT) | Technology | 17.6% |
| Meta Platforms (META) | Technology | 15.1% |
| McDonald's (MCD) | Consumer Discretionary | 14.7% |
| Coca-Cola (KO) | Consumer Staples | 13.3% |
| Costco (COST) | Consumer Staples | 10.2% |
| JPMorgan Chase (JPM) | Financials | 1.3% |
Source: FMP fundamentals, trailing twelve months, checked September 2026.
JPMorgan's 1.3% ROA next to Nvidia's 60.2% is not a sign that JPMorgan is a poorly run bank - it's the clearest illustration available of why ROA can only be compared within an industry, never across one. A bank's balance sheet is stuffed with loans, securities, and deposits by design; those are the raw material of the business, not idle capital, and holding a huge asset base relative to net income is simply what banking looks like. Judge JPMorgan on ROA against Nvidia and you'd conclude one is 46 times better run than the other, which says nothing true about either company.
Two Ways to the Same ROA: The DuPont Breakdown
ROA itself splits into two multiplied components, the first half of the full DuPont decomposition used for ROE:
ROA = Net Profit Margin × Asset Turnover
- Net profit margin (net income ÷ revenue) - how much of each sales dollar becomes profit.
- Asset turnover (revenue ÷ average total assets) - how many dollars of revenue each dollar of assets generates in a year.
This is where the real data tells a genuinely useful story. Costco's 10.2% ROA comes from a net profit margin of just 3.0% - Costco runs on famously thin retail margins - multiplied by an asset turnover of roughly 3.4x, meaning its assets cycle into revenue more than three times a year. Nvidia's 60.2% ROA, by contrast, comes almost entirely from the other lever: a net profit margin of 63.7%, multiplied by an asset turnover of less than 1x. Two completely different business models - a high-volume, low-margin discount retailer and a high-margin, asset-light chipmaker - can land on comparable or wildly different ROA figures depending on which lever they pull, and the single ROA number alone doesn't tell you which one you're looking at until you decompose it.
ROA vs. ROE: The Question Each One Actually Answers
The two ratios use the same numerator - net income - and differ only in the denominator, but that one difference changes what each metric measures:
| Denominator | Answers | |
|---|---|---|
| ROA | Total assets (debt + equity) | How efficiently does the company use everything it owns? |
| ROE | Shareholders' equity only | How much profit does the company return on the owners' stake? |
Because ROE's denominator excludes the debt-funded portion of the balance sheet, leverage inflates ROE without touching ROA at all. A company can borrow more, use that cash productively or not, and watch its ROE climb purely from a shrinking equity base - while ROA, unmoved by financing choices, keeps measuring the same underlying asset efficiency. That's exactly why McDonald's negative shareholders' equity breaks its ROE calculation into a meaningless negative number while its ROA above - a healthy 14.7% - shows the business is performing completely normally. ROA doesn't care that McDonald's has spent decades returning more cash to shareholders than it retained; it only cares how much profit the company's actual assets produced, and by that measure McDonald's looks exactly like the stable, capital-efficient business it is.
The practical rule: use ROE to judge how a company is rewarding its owners, and use ROA - always within the same industry - to judge how efficiently the underlying business actually operates, independent of how it chose to finance itself.
Industry Benchmarks: Why Cross-Sector ROA Comparisons Mislead
- Asset-light software and consumer-brand companies: often 15-20%+. Little physical infrastructure relative to earnings power.
- Manufacturers and industrials: typically 5-10%. Factories, equipment, and inventory are expensive to own.
- Energy and capital-intensive sectors: often single digits. Refineries, pipelines, and drilling equipment tie up enormous capital for years before it pays back.
- Banks and insurers: usually just 1-2%. The balance sheet itself - loans, deposits, policy reserves - is the product, not a tool that sits alongside the product.
None of this means banks or energy companies are worse businesses. It means ROA is only doing its job when the comparison set is the same business model. Comparing an airline's ROA to a bank's, or a bank's to a software company's, tells you almost nothing except which industry needs more physical or financial capital to operate.
How to Use ROA as an Investor
- Compare within the same industry only. Use Stock Alarm's screener to filter and sort ROA alongside sector, so the comparison set is apples-to-apples rather than a bank next to a chipmaker.
- Decompose it before trusting it. Two companies with identical ROA can be running opposite playbooks - high margin, low turnover versus low margin, high turnover. Check net margin and asset turnover separately before assuming you understand why the number is what it is.
- Pair ROA with ROE, not instead of it. A company with strong ROA and a much higher ROE is using leverage productively. A company with weak ROA propped up by a flattering ROE built mostly on debt is a different, riskier story - checking debt-to-equity alongside both ratios closes that gap.
- Watch it over multiple years, not one quarter. A single strong quarter can reflect a one-time gain rather than durable efficiency; consistency across several years is the stronger signal.
- Set an alert instead of re-running the screen. Fundamentals like ROA update quarterly - a price or fundamentals alert on a name you've already screened for strong, consistent ROA does the re-checking for you between earnings reports.
ROA won't tell you the whole story on its own - no single ratio does. What it does well is strip financing decisions out of the picture entirely, leaving a cleaner read on how efficiently a business actually turns what it owns into profit.
Related Reading
- Return on Equity (ROE) Explained
- Return on Invested Capital (ROIC): The Complete Guide
- Debt-to-Equity Ratio: The Complete Guide
- The Graham Number Explained: Benjamin Graham's Formula for Finding Undervalued Stocks
- How to Avoid Value Traps
- What Is EBITDA? How Investors Use It to Value Companies
- How to Set Stock Price Alerts: Complete Setup Guide for Traders
