There are two numbers from this week that belong side by side. The first: the yield on the 10-year US Treasury note rose to roughly 5.22% intraday on Thursday and finished Friday near 5.18%. That is the highest level since June 2007, the months before the global financial crisis. The second number comes from FactSet’s Earnings Insight report dated September 25: the S&P 500 trades at 19.2 times expected earnings over the next twelve months. Flip that price-to-earnings ratio over and you get the market’s earnings yield. One divided by 19.2 is 5.2%.
5.2% on a Treasury. 5.2% on the broad US stock market. For the first time in a very long while, an investor who takes on the risk of owning equities is not being paid any more current yield, on paper, than someone lending money to Washington. The classic equity risk premium — the extra return stocks have to offer over bonds for anyone to tolerate their swings — has shrunk to roughly zero.
And yet: the S&P 500 closed Friday up 0.51% at 7,743.41, the Nasdaq Composite gained 0.5% to 27,068.72, and the Dow Jones Industrial Average added 478.64 points, or 0.93%, to 51,828.62. For the week, the S&P 500 rose about 0.6% and the Nasdaq 100 more than 2%. How does a stock market go up while its most important valuation yardstick crosses a red line? The answer is not that the market has lost its mind. The answer is that it is making a very specific bet — and it is worth understanding exactly which one.
What happened in the bond market this week
The Treasury sell-off did not start this week, but it accelerated sharply. It began Wednesday with a batch of data that was the worst possible mix for bondholders. S&P Global’s flash purchasing managers’ surveys showed a robust US economy in September, particularly in manufacturing, which came in well above expectations. At the same time, companies reported rising input prices, driven by energy. Strong growth plus hot prices is precisely the combination in which a central bank keeps hiking rather than pausing.
Then the Federal Reserve itself chimed in. Governor Michael Barr signaled that additional rate increases may be needed. John Williams, president of the New York Fed and usually a voice of the committee’s center, pointed to the economy’s “remarkable resilience” and said another hike by year-end may be appropriate. An auction of five-year notes met with poor demand. On Thursday, the 30-year bond yield touched roughly 5.50%, the highest since 2004, and the 10-year jumped more than 10 basis points in a single session.
What did not work is just as telling. Treasury Secretary Scott Bessent’s department ran another round of buybacks, reportedly around $4 billion in 20- and 30-year securities. Yields kept climbing anyway. When a buyback explicitly designed to support the long end disappears into the noise, it is a sign that this is not a technical dislocation but a fundamental imbalance between supply and demand.
And the sell-off was global. Japan’s 10-year government bond yield hit its highest since 1996. Germany’s 10-year Bund rose above 3.6%, its highest since June 2009, for a seventh straight weekly increase. UK gilts and other European bonds also printed multi-year highs. Chris Scicluna, head of economic research at Daiwa Capital Markets, put the logic bluntly: the higher yields go, the worse everything looks — the more expensive mortgages get and the bigger the government’s interest bill.
Why yields are rising: oil, the Fed, and a flood of new bonds
It would be convenient to blame oil alone. The war with Iran has pushed energy prices sharply higher since the spring. Diesel reportedly hit a record of about $6.51 a gallon, and the national average for gasoline sits around $4.50. That is a genuine inflation impulse, and it explains why markets assign high odds to another Fed hike. The Fed raised rates on September 16 for the first time since July 2023, to a range of 3.75% to 4.00%. Its next meeting is October 27–28, and futures pricing cited by Yahoo Finance put the odds of another quarter-point move there at around 70%.
But oil explains only part of it. The second, often underestimated driver is sheer volume. The federal deficit is on track to reach roughly $2 trillion this fiscal year, and every one of those dollars has to be sold to someone in the bond market. At the same time, a new heavyweight borrower has shown up that barely existed in this form three years ago: Big Tech. According to Bloomberg data, Amazon, Alphabet, Meta and Oracle issued about $194 billion of bonds through early July — nearly 80% more than in all of 2025. Goldman Sachs expects issuance from the five hyperscalers including Microsoft to reach roughly $250 billion this year and $400 billion in 2027.
The reason is well known. Data-center budgets have become so large that even the most profitable companies on earth no longer fund them purely from operating cash flow. Microsoft, Meta, Amazon and Alphabet have guided to roughly $700 billion of capital expenditure combined in 2026, double last year. Hank Calenti, a strategist at SMBC, framed the problem simply: Treasuries are competing with the rest of the market to be purchased. When the government and the country’s largest companies are bidding for the same capital at the same time, the price of that capital goes up — and that price is the interest rate.
There is an irony here. The AI boom that is carrying tech stocks is also one of the reasons yields are rising — yields that ought to be pushing those very stocks down. The engine of the rally is also the engine of its biggest headwind.
The math every investor should know: earnings yield versus bond yield
The earnings yield is the simplest tool for putting stocks and bonds on the same scale. A Treasury yielding 5.2% pays that return with contractual certainty and, in the case of the US government, virtually no default risk. A stock trading at 19.2 times earnings generates, on paper, 5.2 cents of profit for every dollar invested each year, part of which is paid out and part reinvested. The gap between the two is what economists call the equity risk premium. It is the reward for sitting through drawdowns, profit warnings and, at worst, bankruptcies.
Historically, that premium has averaged somewhere between two and four percentage points. After the financial crisis, when bond yields sat near the floor for years, it was at times far higher — one of the quiet reasons behind the long bull market of the 2010s. Stocks were simply cheap relative to bonds. Today the situation is reversed. Compare an earnings yield of 5.2% with a bond yield of 5.2% and the premium is zero.
The comparison with 2007 is especially instructive because yields were at the same level then. In early summer 2007, the S&P 500 traded at roughly 15 times forward earnings by contemporary estimates. Its earnings yield was therefore above 6.5%, and even with a 5.2% Treasury there was a cushion of more than a percentage point. Today, at the same bond yield, investors are paying nearly a third more for each dollar of earnings. The extreme case at the other end is 2000: the 10-year yielded more than 6% while the market traded at well over 20 times earnings, a deeply negative risk premium. We know what followed.
It is important not to overstretch this comparison. The earnings yield is a static snapshot; it ignores the fact that corporate profits can grow while a bond’s coupon stays fixed. That is exactly where the market’s current bet comes in.
Why stocks are rising anyway: 29% earnings growth
The key sentence in the FactSet report sits right at the top. For the third quarter of 2026, analysts expect S&P 500 earnings to grow 29.1% year over year. If that holds, it would be the third straight quarter of growth above 25%. For the fourth quarter, analysts project 26.8%; for calendar 2026 as a whole, 32.0%. And — this is the truly unusual part — estimates have risen during the quarter rather than fallen. On June 30, the Q3 expectation stood at 26.7%. Analysts normally trim forecasts as a quarter progresses.
That is the mechanism carrying the rally. The price-to-earnings multiple really is falling. At the end of June it stood at 20.4; now it is 19.2 — a compression of roughly 6%, exactly what rising rates demand. But earnings are growing so fast that prices can still rise, or at least hold. The market is getting cheaper without stocks having to fall. That is the scenario optimists have in mind when they say the market can grow into its valuation.
Where those earnings come from is revealing, though. FactSet expects the energy sector to post earnings growth of 111.4% — a direct consequence of high oil prices. Information technology follows at 63.5%, and within it semiconductors stand out with expected growth of 126%. Software, by contrast, is projected at just 14%, and IT services at 4%. In other words, the market’s earnings growth rests on two pillars: oil, which is simultaneously driving inflation and therefore yields, and AI capital spending, which is simultaneously being funded with bonds and therefore also pushing yields up.
Put differently: the two forces propping up stocks are the same two forces eating away the equity premium over bonds. As long as earnings grow faster than yields rise, the math works. But it has a built-in contradiction. If oil falls, energy profits fall. If capex is cut, chip profits fall.
Who is already paying the price
Even with the indexes holding, some corners of the market are already feeling the full force of higher yields. The most obvious is housing. The average 30-year mortgage rate reportedly climbed to about 7.37%, the highest since May 2024 and roughly a percentage point above pre-war levels. Every mortgage that does not get written at these rates is lost business for homebuilders such as D.R. Horton and Lennar, for home-improvement chains like Home Depot, and for mortgage lenders.
The second pressure point is the consumer. The University of Michigan’s final September sentiment reading fell to 48.1 from 51.7 in August, a four-month low, with respondents citing high fuel prices and trade disputes as their top worries. A household paying more for gasoline and more for credit has less left for everything else. On Friday, Bank of America downgraded Nike to Underperform — one small example of how pressure is building on consumer-facing names.
The third is the government itself. A 10-year yield above 5% means every newly issued and every refinanced Treasury is more expensive. With a deficit around $2 trillion and a large share of the debt rolling over each year, interest expense consumes a growing slice of the budget. That in turn raises the borrowing need — a feedback loop that bond markets have punished with still higher yields before.
Rate-sensitive sectors are the natural casualties. Real estate investment trusts such as Realty Income and utilities like NextEra Energy are valued largely for their dividends, and those dividends look far less compelling next to a risk-free Treasury yielding more than 5%. Long-duration growth stocks with profits far in the future face the same arithmetic: the higher the discount rate, the less those future earnings are worth today. Regional lenders, tracked by the SPDR S&P Regional Banking ETF (KRE), sit in a more ambiguous spot: a steeper curve helps lending margins, but unrealized losses on bond portfolios grow with every leg up in yields.
The winners of 5% money
There are beneficiaries, too. Life and property insurers such as MetLife and Travelers invest most of their float in bonds and can now reinvest maturing money at far higher rates, lifting investment income for years. Large money-center banks such as JPMorgan Chase earn more on their deposits when short- and long-term rates are elevated, provided credit quality holds. And companies sitting on huge cash piles earn a genuine return on them: Berkshire Hathaway, with its enormous Treasury bill holdings, is the textbook case of a balance sheet that gets paid for patience.
For individual investors, the most important consequence is unglamorous. For the first time in years, fixed income is a real alternative. A 10-year Treasury above 5%, or a short-term T-bill fund near the Fed’s policy range, delivers predictable income with no equity risk. Inside an IRA or 401(k), where interest and dividends compound tax-deferred, the comparison between a 5.2% earnings yield and a 5.2% bond yield holds almost one-for-one. In a taxable account, Treasury interest is exempt from state and local income tax, which tilts the math slightly further toward bonds for investors in high-tax states. Long-duration bond funds such as the iShares 20+ Year Treasury Bond ETF (TLT) offer the mirror-image trade: if yields fall back, they rally hard; if yields keep climbing, they keep losing.
The counterarguments: why a zero premium need not mean a crash
There are serious reasons not to read the vanished risk premium as a sell signal. The first is growth itself. A bond pays the same coupon for ten years. If corporate earnings keep growing even close to current expectations, the earnings yield on today’s purchase price will be well above 5.2% in two years. An investor buying at 19.2 times earnings whose profits rise 30% soon has an earnings yield of almost 7% on cost. That is exactly the calculation holding the market up.
The second concerns the index’s composition. Today’s S&P 500 is dominated by companies with exceptionally high margins, low capital intensity and — despite the bond binge — solid balance sheets. A structurally more profitable index can justify a higher multiple than the index of 2007. And at 19.2, the forward P/E is actually below its five-year average of 19.8. The valuation by itself is not extreme. The interest rate is.
The third argument: yields could come back down. Rick Rieder, BlackRock’s chief investment officer for global fixed income, called the move “not a crisis but an eye-opener,” and some investors already see these levels as a buying opportunity in bonds. If oil falls — on Friday, West Texas Intermediate crude dropped 2.33% to about $92 a barrel on hopes of a deal with Iran to reopen the Strait of Hormuz — some of the inflation pressure would bleed out, and yields could retreat. The premium would return without stocks having to change at all.
Against that stand equally serious objections. A drop in oil would lower yields, but it would also lower the earnings expectations of the energy sector, which supplies a sizable chunk of index profit growth. And the supply of government and corporate debt does not disappear just because crude gets cheaper. Even without an inflation shock, the question remains who will buy all the new debt, and at what price.
What to watch in the coming weeks
The calendar ahead reads like a series of exams for exactly this bet. The September jobs report lands on October 2; another strong print would harden expectations of an October hike and push yields higher still. In mid-October, third-quarter earnings season begins in earnest, and it has to show whether the 29% growth actually materializes. Then, on October 28, the Fed announces its next decision.
That produces a clear checklist. First, the 10-year yield: if it holds above 5.2% or keeps climbing, the risk premium turns negative and pressure on valuations builds. A move back below 5% would give stocks room to breathe. Second, earnings revisions: as long as analysts keep raising estimates, growth is carrying prices. The first reporting season in which revisions turn lower would be the real warning sign. Third, oil, which moves both sides of the equation at once.
Bottom line: the market has spent its cushion
The most important finding of this week is not that stocks rose despite a 19-year high in yields. That is well explained by exceptional earnings growth. The most important finding is that in doing so, the market has used up its safety margin. A zero risk premium does not mean stocks must fall. It means stocks no longer have room for disappointment. Every profit warning, every further rise in yields and every setback in AI spending now meets a valuation that, compared with the risk-free alternative, forgives nothing.
Two numbers, both 5.2%. One is guaranteed, the other merely expected. As long as earnings deliver, the stock market can live with that tie — as it did this week. But for the first time in nearly two decades, the question of whether you are still being paid to own stocks is no longer academic. Anyone putting money to work in the weeks ahead should be clear that they are not buying a cheap market. They are buying growth that has to arrive exactly as analysts are promising it today.
Try TradingView Free for 30 Days
Plus get a $15 discount on your first subscription through this link.
