The Kapital · Market Education · 20 August 2026
Technology investors often experience the same confusing day: a company reports no bad news, its long-term growth story appears unchanged, and the stock still falls 4% because Treasury yields moved higher. That reaction can look irrational until you remember what a stock actually is. A share is a claim on future cash flows, and the value of those cash flows changes when the return available on low-risk assets changes. The mechanism is called duration, and it is one of the most useful concepts an equity investor can learn.
The idea matters especially now. Long-term US Treasury yields have recently moved sharply higher, with the 30-year yield trading around levels last seen near 2007. Semiconductor and other growth stocks sold off even while many underlying businesses continued reporting strong demand. Investors were not suddenly deciding that artificial intelligence had disappeared. They were changing the price they were willing to pay for future earnings.
Start with the simplest valuation formula
Imagine a company that will pay you $100 one year from now. If your required return is 5%, that future $100 is worth about $95.24 today. If your required return rises to 6%, the same $100 is worth about $94.34.
The difference looks small because the cash flow is only one year away. Now move the payment twenty years into the future. At a 5% discount rate, $100 received in year twenty is worth roughly $37.69 today. At 6%, it is worth only about $31.18.
Nothing changed about the future cash flow. The company still pays $100. The only change is the return investors demand while waiting. Yet the present value falls by more than 17%.
That is duration in intuitive form: the farther into the future an asset’s value is concentrated, the more sensitive that value becomes to changes in interest rates.
Why growth stocks behave like long-duration bonds
A mature utility earns most of the cash investors expect within the next several years. A young software or semiconductor platform may be valued because investors expect earnings to become much larger ten or fifteen years from now.
The second company has more of its value in distant cash flows. In valuation terms, it has longer equity duration.
This is why expensive growth stocks can fall hard when long-term yields rise. The market is discounting a large amount of future value back to the present. A small change in the discount rate applies repeatedly across many years.
It is also why “the P/E only fell from 40 to 32” can produce a large stock-price decline without any change in earnings. The multiple itself is the mechanism through which the discount rate changes.
Bond yields influence the equity discount rate in several ways
Most equity valuation models begin with a risk-free rate, usually a government bond yield. Investors then add an equity risk premium because owning a stock is riskier than owning a Treasury bond.
If the risk-free rate rises from 4% to 5% and the risk premium stays unchanged, the required return on equity also rises by roughly one percentage point. That lowers the present value of future cash flows.
There is a second channel: opportunity cost. If a high-quality government bond offers 5%, an investor does not need to accept a very low expected return from an expensive stock. The stock must either offer a higher earnings yield, faster growth or a lower price to remain competitive.
There is a third channel: corporate financing. Higher government yields often feed into higher borrowing costs. Companies refinancing debt, building factories or leasing data-center equipment face a higher cost of capital. That can directly reduce future free cash flow.
The earnings-yield shortcut
You do not need a full discounted cash flow model every morning. One useful shortcut is to compare a stock’s earnings yield with bond yields.
A stock at 20 times earnings has an earnings yield of 5%. A stock at 40 times earnings has an earnings yield of 2.5%. This is not directly comparable to a Treasury yield because stocks can grow and earnings are risky, but the contrast helps frame expectations.
When ten- or thirty-year bond yields rise, a 2.5% current earnings yield requires stronger future growth to remain attractive. If the market becomes less confident in that growth at the same time, the multiple can compress violently.
This is why the most expensive stocks are often the first place investors reduce exposure during a yield shock. Their valuation contains less room for disappointment.
Why Nvidia can fall on a day when AI demand is still strong
Consider a company like Nvidia. Its current earnings are enormous, so it is not a pre-revenue speculative stock. But investors still pay for years of future AI growth. A large part of the market capitalization depends on Data Center revenue, margins and platform leadership staying unusually strong for a long time.
If long-term yields rise, the present value of those distant excess profits falls. Nvidia can therefore report excellent demand and still decline because the hurdle rate investors apply to the future becomes higher.
This is not a contradiction. Business momentum and valuation momentum can move in opposite directions.
I discuss Nvidia’s current operating and financing setup in more detail in my separate Nvidia analysis. The important lesson here is general: a great company can become a worse stock when the price paid for future greatness is too high relative to available interest rates.
Why banks can react differently
Higher yields do not affect every sector in the same way. Banks can sometimes benefit when longer-term rates rise because the spread between what they earn on loans and what they pay on deposits can improve.
Insurers can also benefit from higher reinvestment yields on their bond portfolios over time. Commodity companies may respond more to inflation expectations than to the discount rate itself.
This cross-sector difference matters. A broad market selloff caused by higher yields is often not equally fundamental for every company. The index may fall because its largest technology components have long duration even while financial companies become more profitable.
Not all yield increases are bearish
Investors often react to “rates up” as if the message were always negative. The reason yields rise matters.
If yields rise because economic growth is stronger than expected, corporate revenue and earnings may also improve. A higher discount rate can then be partly offset by higher cash flows.
If yields rise because inflation expectations are becoming unanchored, the effect is more dangerous. Companies face higher input costs, central banks may keep policy restrictive and the required return on both debt and equity can increase.
If yields rise because government borrowing expands and investors demand more term premium, the effect can be particularly uncomfortable for growth equities. The discount rate rises without an equivalent improvement in corporate demand.
Always ask what is driving the bond move before deciding what it means for stocks.
Real yields matter more than nominal yields for many valuations
A nominal Treasury yield can rise because expected inflation rises or because the real return demanded by investors rises. The distinction matters.
Companies with strong pricing power can pass some inflation through to customers. Their nominal cash flows may rise with inflation. That means an increase in nominal yields driven only by inflation does not necessarily reduce real economic value by the full amount.
A rise in real yields is harder. It increases the true opportunity cost of capital. Long-duration assets typically react more negatively because there is no automatic inflation offset in the cash-flow forecast.
For that reason, I pay close attention to inflation-protected Treasury yields when trying to understand whether a technology selloff is primarily about real discount rates or inflation fears.
The terminal value is where DCF models become fragile
Most discounted cash flow models for growth companies derive a large portion of fair value from the terminal value: the estimated value of all cash flows after the explicit forecast period.
This creates hidden duration. An analyst may forecast only five or ten years, but the terminal-value formula effectively assumes cash flows continuing forever.
If the discount rate is 8% and the perpetual growth rate is 3%, the denominator in a Gordon Growth terminal value is 5%. Raise the discount rate to 9% and the denominator becomes 6%. Holding cash flow constant, the terminal value falls by roughly 17% simply because of a one-percentage-point change in the discount rate.
This is why a DCF can produce wildly different fair values from small changes in assumptions. The model is not broken. It is revealing how much of the investment thesis sits in distant cash flows.
What happens when growth slows at the same time yields rise
This is the most dangerous combination for expensive equities. Imagine a company valued at 40 times forward earnings because investors expect 30% growth. If bond yields rise and the growth outlook falls to 20%, both sides of the valuation equation deteriorate.
Future earnings are lower than expected, and the multiple investors are willing to pay for those earnings also falls. The stock can decline far more than the change in near-term profit would suggest.
This double compression explains many violent technology drawdowns. Investors often focus only on the earnings revisions and miss the simultaneous change in discount rate.
Why profitable growth is less fragile than narrative growth
A profitable company with strong current free cash flow has shorter effective duration than a company whose value depends almost entirely on profits ten years away. That does not make it immune to rates, but it provides an anchor.
This is one reason I prefer growth businesses that already fund their own expansion. Current cash generation gives management optionality and reduces dependence on external financing.
A pre-profit company facing higher yields can be hit three times: its distant cash flows are discounted more heavily, investors demand a lower valuation multiple, and raising capital becomes more expensive.
That combination is why speculative growth often performs worst during sustained increases in real yields.
A practical checklist for investors
When yields rise sharply, I ask five questions before reacting to the stock price.
1. Where are the company’s cash flows? Are most profits generated today or assumed many years from now?
2. Does the company need external capital? Higher financing costs matter far more for a business that must issue debt or equity.
3. Can revenue grow faster because the economy is stronger? A growth-driven yield rise may not be purely negative.
4. Does the valuation rely heavily on terminal value? If yes, small discount-rate changes can have large effects.
5. What is the spread between expected equity return and the risk-free rate? If that spread has become too narrow, the stock may need a lower price even if the company remains excellent.
How I change a valuation when yields move
I do not mechanically add every basis-point change in the 30-year Treasury yield to my discount rate. Markets are noisy, and a one-day move can reverse quickly.
Instead, I ask whether the change appears structural. If long-term real yields move higher for several weeks or months, I raise the discount rate in my models and often reduce the terminal multiple. I also stress-test the result with a slower growth assumption because tighter financial conditions can feed back into demand.
For mature companies, the fair-value change may be modest. For long-duration technology stocks, the range can move dramatically.
Do higher yields create buying opportunities?
Sometimes. A yield-driven selloff can be attractive when the business fundamentals remain intact and the stock falls more than the change in discount rate justifies.
But “it is down because of rates” is not automatically bullish. If rates stay high, the lower valuation may be fundamentally correct. A stock does not have to return to the multiple it traded at when the risk-free rate was lower.
I therefore compare the new price with a new fair value, not with the old share price. Anchoring to the previous high is one of the easiest ways to overpay.
What the August 2026 selloff is teaching us
Recent market action has provided a clean example. Reuters reported a sharp decline in semiconductor shares as long-term Treasury yields rose toward multi-decade highs. Nvidia, Broadcom and other AI-linked stocks fell even though the long-term AI demand narrative remained broadly intact.
The market was repricing duration and the cost of capital. It was also questioning whether the enormous infrastructure spending cycle would generate returns large enough to justify financing at higher rates.
Those two issues are connected. When money is cheap, investors tolerate longer payback periods. When money becomes expensive, every data center, chip program and software investment must clear a higher hurdle.
My conclusion
Bond yields matter to technology stocks because valuation is a competition across time. Investors give up money today for cash flows tomorrow. The higher the return available without taking equity risk, the more future corporate cash flow must be discounted.
The companies most exposed are not necessarily the worst businesses. They are the businesses whose market values depend most heavily on profits far in the future.
That distinction helps explain why excellent companies can sell off on bond-market news and why seemingly cheap mature companies can sometimes hold up better.
I use duration as a risk-management tool, not a market-timing signal. When long-term yields rise, I do not automatically sell growth stocks. I lower the price I am willing to pay for them. When yields fall, I do not automatically buy. I ask whether the underlying cash-flow expectations still deserve the higher present value.
For a live example of this framework applied to AI infrastructure, see my AI valuation analysis.
Sources and data date
Data date: 20 August 2026. Examples are simplified for educational purposes.
- Reuters: bond yields and technology selloff, 18 August 2026
- Reuters: long-term Treasury yields near 2007 levels, 19 August 2026
This article is educational and reflects my analytical framework. It is not investment advice.
