
Getting started · · 6 min read
The ten-year is above 5%. Whose interest bill bites?
The ten-year Treasury closed at 5.18% on 24 September, its highest since July 2007. The interest coverage ratio tells you which companies feel that. Of 630 we cover outside financial services, 218 earn less than three times their interest bill. How to read the ratio, and why 2.4 is fine for a utility and a warning for an airline.
On 24 September the US government paid 5.18% to borrow for ten years. The last time it paid that much was 6 July 2007, when the iPhone was a week old. A year earlier the same loan cost 4.16%, and on 16 September the Federal Reserve raised its own rate by a quarter point, to 3.75 to 4%, and said "inflation remains elevated".
Companies do not borrow at the government's rate. They borrow above it, and a company that sells a ten-year bond today prices it off 5%, not the 4% of a year ago. So the useful question for anyone who owns shares is not what the ten-year does next. It is which companies can carry a bigger interest bill, and there is one number that answers it.
The interest coverage ratio, in one line
Take the operating income from the annual report, the profit from the business before interest and tax. Divide it by the interest expense, two lines lower. That is the interest coverage ratio: how many times the business earns what it owes its lenders.
At 3, a third of the operating profit goes to the lenders before the taxman and the shareholders get a cent. At 1.5, two thirds. Below 1, the business does not earn its interest at all, and the rest comes from cash in the bank or from more borrowing. The lines are in the income statement, and the statement is readable in an hour.
Here is how a rate reaches the ratio, on made-up round numbers. A company earns $1 billion a year from its business and owes $10 billion at 4%. Its interest bill is $400 million, so it covers it 2.5 times. Now the whole $10 billion comes due and is refinanced at 6%. The bill becomes $600 million, and the ratio drops to 1.7. Nothing changed in the business: same customers, same profit, same managers. A third of the cushion went to the lenders because of a date on a loan agreement. Real companies refinance a slice at a time, which is why the same rate rise reaches some of them this year and others in 2030.
What the market looks like on 26 September
I ran the ratio on the 630 companies we cover outside financial services whose last annual report gives both lines. (Financial companies are left out on purpose: interest is what a bank buys to make its product, so the ratio means something else there.)
| Companies | |
|---|---|
| Covered by the ratio | 630 |
| The middle company | 5.2 times |
| Under 3 times | 218 (35%) |
| Under 1.5 times | 129 |
| ...of which the business made an operating loss | 95 |
| ...of which it made a profit, but a thin one | 34 |
Our financials table, latest annual report per company, read 26 September 2026.
The 95 are a different story: a company that loses money before interest has a business problem first and a debt problem second. The 34 are the ones a rate move reaches directly. They make money, and most of it already goes to the lenders.
Same industry, three interest bills
Airlines buy the same planes and the same jet fuel, and they fly into the same airports. Their 2025 annual reports:
| Operating income | Interest expense | Coverage | |
|---|---|---|---|
| United Airlines | $4.71bn | $1.37bn | 3.4 times |
| Southwest Airlines | $0.43bn | $0.17bn | 2.6 times |
| American Airlines | $1.47bn | $1.72bn | 0.85 times |
Source: each company's 10-K for 2025, via the SEC's XBRL data, read 26 September 2026.
American earned $1.47 billion from flying in 2025 and owed $1.72 billion in interest. United earned three times as much and owed less. Same weather, same fuel price, different balance sheet, and the balance sheet is where the debt lives. Southwest's 2.6 is the other kind of thin: it earned more interest on its cash than it paid, and its problem is a small profit, not a big debt.
Why 2.4 is fine for a utility
Now the other side. Duke Energy covers its interest 2.4 times ($8.63 billion of operating income against $3.63 billion of interest, 10-K for 2025), and that is normal. Of the 35 utilities with the figure, 31 are under 3. A utility borrows to build power lines that earn a regulated return for decades, and regulators have historically let those costs flow into what customers pay. The debt is the business model.
So the ratio needs the same treatment as the P/E ratio: compare it with the industry before you judge it.
| Sector | Companies | Middle company | Under 3 times |
|---|---|---|---|
| Utilities | 35 | 2.4 | 89% |
| Communication services | 39 | 2.3 | 51% |
| Real estate | 37 | 3.3 | 46% |
| Consumer defensive | 42 | 5.8 | 36% |
| Basic materials | 31 | 6.7 | 35% |
| Consumer cyclical | 93 | 5.1 | 31% |
| Healthcare | 82 | 6.2 | 30% |
| Technology | 120 | 7.4 | 28% |
| Industrials | 118 | 6.1 | 25% |
| Energy | 33 | 5.4 | 21% |
Same table and date as above.
A 2.4 at a utility is the middle of the pack. A 2.4 at a software company is a company that borrowed like a utility without the regulator.
And the ones that do not care
Home Depot covers its interest 8.7 times ($20.89 billion against $2.41 billion, fiscal year to February 2026). Comfort Systems, the air-conditioning contractor that went up 27 times from January 2021 to 17 September 2026, had $9 million of interest expense in 2025 and covers it about 146 times. For companies like these, a 5% ten-year barely shows up in the interest bill.
Three things to check in your own holdings
- The ratio, from the last annual report. Operating income divided by interest expense. Compare it with the middle company of its sector in the table above.
- The trend. Work it out for the last three years. A ratio sliding from 6 to 3 while the debt grows says more than a single 3.
- When the debt comes due. The debt note in the annual report has a table of repayments by year. Debt that matures next year is refinanced at next year's rates. Debt fixed until 2035 does not care what the ten-year did this week.
One caution on the first check: one bad year can make a healthy company look thin, because a write-down lowers operating income without touching the debt. That is what the second check is for.
If you want to see all of this on one real company first, here is a full report, every page of it.
Sources: ten-year Treasury yields from the US Treasury's daily par yield curve rates, read 26 September 2026 (5.18% on 24 September 2026, 5.19% on 6 July 2007, 4.16% on 24 September 2025); the Federal Reserve's statement of 16 September 2026; operating income and interest expense for United Airlines, Southwest Airlines, American Airlines, Duke Energy and Home Depot from their annual reports via the SEC's XBRL data, read 26 September 2026; Comfort Systems and every count and median from our financials table on the same date.
Educational and proprietary. This explains what our research does, not the exact formula behind it, and it is not personalised investment advice. See the full disclaimer.
Read next

8 min read
How to read a balance sheet, on one real filing
A balance sheet in one hour, using a real one: Home Depot's 10-K for the year to 1 February 2026. What each line means, which four numbers matter, and why a $105 billion company owns $13 billion on paper.

8 min read
How to read financial statements in an hour, on McDonald's
The three statements every listed company files, read in twenty minutes each on McDonald's 2025 annual report: an income statement that shows it is a landlord, a balance sheet with less than nothing left for the owners, and a cash flow statement that explains why that is fine.

8 min read
What is a good P/E ratio? The answer is another number
Every guide says 15 to 20 is a good P/E ratio. On the market of September 2026 a 15 sits in the cheapest fifth. The median of the 968 companies we cover is 25, Walmart is 39, Tesla 337, and 180 companies have no P/E at all. What the number means, the three comparisons that give it meaning, and the four ways it lies.



