The 10-Year Treasury Is Back Above 5%. What That Does to Stock Valuations
On September 14, 2026 the 10-year U.S. Treasury yield closed above 5% for the first time since October 2023. Two days later the Fed raised its policy rate to 3.75–4.00%. Bond investors got there first, as they usually do. The question for anyone who owns stocks is not whether 5% is high — it is what a 5% risk-free rate does to the arithmetic behind every share price. This piece walks through that arithmetic with a calculator and real filing data, so you can see which kinds of companies are most exposed and why.
Start with the only formula that matters
A stock is worth the cash it will hand its owners over time, discounted back to today. That sentence contains two variables, and the Treasury yield sits inside the second one. The discount rate most analysts use is the risk-free yield plus an equity risk premium — the extra return demanded for accepting business risk. If the risk-free yield rises from 4% to 5% and the premium stays put, every future dollar is worth less today. Nothing about the business changed. The alternative to owning it did.
Here is how big that effect is. Take a company producing $100 of free cash flow this year, growing that cash 5% a year for ten years and 3% a year forever after. Its value at a range of discount rates:
| Discount rate | Value of the business (5% growth) | Value of the business (15% growth) |
|---|---|---|
| 8% | $2,414 | $5,296 |
| 9% | $2,000 | $4,292 |
| 10% | $1,705 | $3,582 |
| 11% | $1,485 | $3,056 |
Moving from an 8% to a 10% discount rate — roughly what happens when the risk-free rate climbs two points and the premium is unchanged — cuts the value of the slow grower by 29% and the fast grower by 32%. That is the whole story of a rising-rate market in one table. Prices fall even when nothing goes wrong at the company, and they fall a little harder for the companies whose cash flows are furthest in the future.
The earnings-yield test: are you being paid to take equity risk?
You do not need a spreadsheet to run a first-pass version of this. Take the P/E ratio and invert it. A stock at 20× earnings has a 5% earnings yield; at 25× it is 4%; at 50× it is 2%. Then set it against the Treasury. The difference is what the market is paying you, today, to own a business instead of a government bond.
Below are the trailing P/E ratios on Investor Sam’s ratios pages for a group of widely held companies, each computed at the share price on the company’s fiscal-year-end date. Prices have moved since — several of these names rallied hard in September — so treat the table as a demonstration of the method and check the live page for the current figure.
| Company | Trailing P/E (fiscal year-end) | Earnings yield | Versus a 5% Treasury |
|---|---|---|---|
| Bank of America (BAC) | 14× | 7.1% | +2.1 points |
| JPMorgan (JPM) | 16× | 6.3% | +1.3 points |
| Exxon Mobil (XOM) | 18× | 5.7% | +0.7 points |
| Microsoft (MSFT) | 21× | 4.7% | −0.3 points |
| Chevron (CVX) | 22× | 4.5% | −0.5 points |
| Meta Platforms (META) | 28× | 3.6% | −1.4 points |
| Alphabet (GOOGL) | 29× | 3.5% | −1.5 points |
| Apple (AAPL) | 34× | 2.9% | −2.1 points |
| Nvidia (NVDA) | 38× | 2.6% | −2.4 points |
| Costco (COST) | 51× | 2.0% | −3.0 points |
| AMD (AMD) | 76× | 1.3% | −3.7 points |
| Tesla (TSLA) | 416× | 0.2% | −4.8 points |
A negative number in the last column is not a sell signal. It means the market expects earnings to grow fast enough that today’s 2% yield becomes 6% or 8% on your purchase price in a few years. Sometimes it does: Nvidia’s net income more than kept pace with its share price through 2024 and 2025, which is why its P/E fell from 48× to 38× even as the stock rose. The point of the test is to make the assumption explicit. At a 1.3% earnings yield, AMD has to grow earnings roughly fourfold just to match what a Treasury pays today. Maybe it will — the AI-agent rally this week is a bet that it will — but you should know that is the bet.
Why the “growth” end of the market is the most rate-sensitive
Put the two ideas together and a pattern emerges. The companies with the lowest earnings yields are also the companies whose value sits furthest in the future, and those are precisely the companies a higher discount rate hurts most. That is not a coincidence; both are the same fact viewed from different angles. In 2022, when the 10-year went from 1.5% to 4%, the Nasdaq fell about a third while the energy sector rose. The businesses had not changed places. The discount rate had.
It is worth being precise about what that does and does not mean for 2026. The move from 4% to 5% on the 10-year is a smaller shock than 2022’s, and it is arriving while the AI-related companies are actually delivering the earnings growth their valuations assume — Micron is guiding to record revenue, Nvidia’s trailing net margin is 56%, Meta’s is 30%. Earnings growth can outrun a rising discount rate. It just has to run faster than it did before, and there is less margin for a disappointment.
Three things a 5% Treasury does not tell you
- It does not tell you the equity risk premium. The premium is not observable; it is inferred. If investors have become more willing to hold stocks — as a record Nasdaq close on September 21 suggests — the premium can compress and offset the higher risk-free rate. That has happened before and it can reverse without warning.
- It does not tell you which earnings are real. A 7% earnings yield on a bank is only attractive if the earnings survive the next credit cycle; a 5.7% yield on an oil major is only attractive if $100 crude holds. Earnings yield is a screen, not a verdict. Read our note on Exxon and Chevron at peak earnings for the cyclical version of this problem.
- It does not tell you the company’s own borrowing cost. That is a separate channel entirely, covered in our five-check balance-sheet audit for the rate hike.
How to use this on your own portfolio
Three practical steps, in order:
1. Rank your holdings by earnings yield. The screener sorts the whole coverage universe by P/E; your own list is quicker to do by hand from each company’s ratios page. Anything under a 3% yield is priced for substantial growth. That is a fact, not a criticism.
2. For each low-yield name, write down the growth rate you need. A rough rule: to justify a P/E of 40 against a 5% risk-free rate with a normal premium, earnings need to compound at something like 15–20% for five years or more. Then compare that requirement with the company’s actual five-year record on its income statement. If the record is 8% and the requirement is 18%, you are relying on acceleration.
3. Re-run your own discount rate. If you value companies with a DCF, and you built your models when the 10-year was 3.5–4%, your discount rates are stale. Raise them a point, re-run the numbers, and see which theses still hold. The good ones will. The marginal ones will not, and that is useful to know before the market tells you.
Where the 10-year goes from here
Nobody knows, and this article will not pretend otherwise. What can be said is what the yield is responding to. Headline inflation is 3.4% and energy-driven; core PCE is around 3.3%; the Fed’s own median projection has the policy rate at 4.1% through 2027; and the Treasury is issuing heavily. A 5% 10-year is consistent with all of that. If oil retreats as the Iran situation de-escalates, the inflation component fades and yields likely ease. If it does not, the bond market has already told you what it thinks. Either way, valuing stocks with a discount rate built for a 3.5% world is the one mistake that is entirely within your control to fix.
Frequently asked questions
Is a P/E of 30 “too high” when the 10-year is at 5%?
Not by itself. A P/E of 30 is a 3.3% earnings yield, which is below the Treasury, so the stock is only worth owning if earnings grow meaningfully. For a company compounding profits at 15% a year, that is reasonable. For one growing at 4%, it is hard to justify. The multiple alone does not answer the question; the multiple plus the growth record does.
Do dividend stocks hold up better when yields rise?
Sometimes, but not because of the dividend. They hold up when their earnings are near-term and predictable, which is also what makes them able to pay a dividend. A high-yield stock whose payout is not covered by free cash flow is more exposed, not less, because a 5% Treasury competes directly for the same income-seeking buyers.
Where does the P/E on Investor Sam come from?
It is calculated from net income in the company’s most recent annual 10-K, as filed with the SEC, and the share price on that fiscal year-end date. The date is shown on each ratios page so you can adjust for any move since.
Sources
- U.S. Department of the Treasury — daily Treasury par yield curve rates, September 2026.
- Federal Reserve Board — FOMC statement and Summary of Economic Projections, September 16, 2026.
- U.S. Bureau of Labor Statistics — Consumer Price Index, August 2026.
- Company 10-K filings via SEC EDGAR for BAC, JPM, XOM, MSFT, CVX, META, GOOGL, AAPL, NVDA, COST, AMD and TSLA; ratios as computed on Investor Sam.
- Discounted-cash-flow figures in the first table are the author’s own calculation using the stated assumptions.