NextFin News - US stocks and bonds sold off together on Tuesday, a rare same-day loss for the two asset classes that have spent most of the AI rally moving in opposite directions, as the 30-year Treasury yield climbed to its highest level since 2002 and a fresh batch of weak economic data undercut the case that higher rates can be sustained without damage. The long bond touched 24-year highs while consumer confidence fell to a 12-year low and job openings slipped to a five-month trough, leaving investors to reconcile a bond market pricing fiscal risk with a labor market that is cooling faster than expected.
The Session That Broke the Usual Rule
The numbers that defined the day were stark. Yields on 30-year Treasuries reached their highest point in 24 years, with the benchmark closing at 5.594%, its highest closing level since 2002. The 10-year yield, which had touched 5.23% on Friday for its highest level since 2007, held near that mark by the close. That is the biggest Treasury selloff since the tariff shock of April 2025, when the 10-year yield accelerated from less than 4% to 4.5% intraday within days and the 30-year topped 5%, and it arrived on the same day the US national debt crossed the $40 trillion threshold.
Equities could not look through it. The S&P 500 slipped to 7,676, the Dow Jones Industrial Average fell 0.3%, and the Nasdaq Composite gave back early gains to close lower. The declines were modest in isolation; what mattered was the correlation. Stocks and bonds rising together has been the bull-market default for much of this cycle, because strong growth lifted earnings and yields in tandem. Falling together is the signature of a regime in which the discount rate, not the cash flow, is setting the price.
The data released during the session sharpened the tension. The Conference Board's consumer confidence index fell to 81.9 in September from a downwardly revised 88.6 in August, its lowest reading in 12 years. The present situation index dropped to 109.3, missing the 120.2 economists expected, while the expectations index slid to 63.6, well below the 80 threshold that has historically signaled a recession ahead. Hours earlier, the Labor Department's JOLTS report showed job openings falling 256,000 to 7.079 million in August, below the 7.23 million forecast.
Against that backdrop, corporate firepower looked almost theatrical. Nvidia announced on Monday a $150 billion increase to its share repurchase authorization, lifting the remaining program to $235 billion through fiscal 2028 — the largest buyback authorization in history. The stock rose as much as 3% intraday before giving back gains, a reminder that even the largest balance sheet in corporate America cannot set the price of capital for the entire market.
The Term Premium Is Back, and It Is Not Visiting
The first-order explanation for the bond move is mechanical: more supply, fewer committed buyers. The federal government is issuing debt to finance a deficit that the Congressional Budget Office projects at $1.9 trillion in fiscal 2026, while companies are borrowing heavily to fund artificial intelligence infrastructure. Vanguard estimates that Alphabet, Amazon, Meta Platforms, Microsoft and Oracle issued about $132 billion of debt through July, up sharply from the roughly $35 billion annual average between 2020 and 2024. Broader AI-related issuance could reach $300 billion to $570 billion this year. There is simply more paper chasing a buyer base that has thinned.
"There's a supply/demand mismatch in the cash bond market, with fewer price-insensitive buyers willing to take that Treasury supply without being offered higher yields to do so," a market participant said in early September.
But the mechanism that matters is the term premium — the extra yield investors demand for locking capital away for decades. For most of the post-2008 era, that premium was suppressed, at times negative, by a wall of price-insensitive demand from central banks and reserve managers. That wall has eroded on three sides. The Federal Reserve has been shrinking its balance sheet through quantitative tightening, removing the largest automatic bidder. Foreign official holders have become more selective as dollar reserves stop growing. And domestic banks, constrained by capital rules, have less appetite to warehouse duration. The market is charging accordingly.
The Treasury's own attempt to intervene — at least doubling buybacks of long-dated bonds to a minimum of $4 billion and targeting the 10-to-30-year sector — produced only a brief pause. At the first expanded buyback on September 10, the department accepted just $5.2 billion of a $6 billion target, and the selloff intensified as skepticism mounted. A buyback that cannot absorb the marginal supply is not a floor; it is a signal that the marginal buyer now requires a higher price.
This is the structural core of the move. A cyclical selloff reverses when the trigger fades. A term-premium repricing does not reverse on its own, because it reflects a changed assessment of risk, not a temporary imbalance. The evidence is in the resilience of the move: yields kept climbing after oil pulled back, after buybacks were announced, and after equity indexes absorbed the shock. The market is no longer asking how high rates can go. It is asking how much compensation is required to hold 30 years of US debt, and the answer has reset.
The fiscal arithmetic supports the reset. Debt held by the public stands at 100% of GDP and is projected to reach 120% by fiscal 2036 under current law, while net interest payments are set to surpass $1 trillion in fiscal 2026 and double to $2.1 trillion by 2036. Interest is becoming the fastest-growing category of federal spending, which means tomorrow's deficits are being financed in part by yesterday's interest bill. That feedback loop is exactly what a term premium prices. The CBO's own baseline shows deficits averaging 6.1% of GDP over the next decade with no policy change. That is not a cyclical condition; it is the new budget identity.
Why Stocks Could Not Decouple
The conventional wisdom held that equities were insulated. Earnings were strong, AI capital expenditure was real, and megacap balance sheets could fund themselves. That view treated the bond market as a backdrop rather than a pricing engine.
The transmission channel runs through the discount rate. Long-duration growth stocks are valued on cash flows far in the future; a 100-basis-point rise in the 10-year yield compresses those present values mechanically, before any earnings revision. A simple illustration: a dollar of cash flow arriving in 10 years is worth 61 cents discounted at 5%, but only 38 cents at 10%. The AI complex, where much of the value sits in the second half of the decade, is disproportionately exposed to that math. The second-order effect is broader: higher Treasury yields pull up corporate bond spreads, mortgage rates, and the hurdle rate for every capital project in the economy. The dollar, up about 1.5% against a basket of currencies this September, tightens financial conditions for anyone borrowing outside it.
The housing channel makes the mechanism visible in real time. The 30-year fixed mortgage rate moves with the 10-year yield plus a spread, and at these levels it has locked millions of homeowners into sub-3% mortgages they will not give up. Housing turnover has stalled, construction has cooled, and the wealth effect from home prices has weakened — all before the labor market shows broad layoffs. That is how a bond selloff becomes a growth drag without a single company cutting headcount.
The data released Tuesday exposed the fragility of the "earnings will outgrow rates" thesis. Consumer confidence at a 12-year low and job openings at a five-month low are not yet a recession, but they are the kind of simultaneous softening in demand and labor that has preceded every growth scare in this cycle. When the bond market sells off on fiscal risk and the labor market softens on the same day, the Federal Reserve's room to cushion a slowdown shrinks. Markets were pricing a 64% probability of an October rate hike — not a cut.
"I think this year it has more to do with the bond issuance than the inflation story," said Thierry Wizman, global FX and rates strategist at Macquarie Group. "The thing that's abnormal is that we're in the midst of a very strong investment cycle."
Wizman's point cuts both ways. A strong investment cycle supports growth, which is why equities held up as long as they did. But it also means the supply of bonds competing with Treasuries will remain elevated, keeping pressure on the very yields that support equity multiples. The AI boom is financing the discount rate that is compressing AI valuations.
The Counter-Thesis: Positioning Shock, Not Regime Change
The strongest case against the structural read is that the move is crowded, fast, and therefore reversible. Yields rose with unusual speed — the 10-year climbed from below 4% in early April to above 5% within months — and rapid moves of that kind often overshoot before mean-reverting. The buyer base has not disappeared; it has gone on strike, waiting for better levels. Oil prices retreated on Tuesday, with Brent falling 2.6% to $102.59 a barrel after reports that Saudi exports through the East-West pipeline were resuming, which would ease the inflation impulse that helped drive the initial yield spike. If oil stabilizes and the Fed pauses, the term premium could compress back toward its recent range, and long yields could settle well below 5%.
This view has a serious constituency. Wizman argued through late September that yields at these levels were not themselves unusual, particularly because they were not being accompanied by extreme inflation expectations or an aggressively tightening Fed. He is right about one thing: nothing about a 5.2% 10-year yield is inherently catastrophic. In the mid-2000s, the economy grew with the 10-year in this neighborhood.
But the counter-thesis mistakes the level for the mechanism. The question is not whether 5.2% is painful in isolation. It is whether the compensation demanded for duration risk has structurally reset. The buyback program's failure to hold the line, the debt crossing $40 trillion, and the simultaneous softening in confidence and job openings point to a repricing driven by risk assessment, not just positioning. A positioning shock fades when the crowded trade unwinds. A risk-premium reset fades only when the underlying fiscal and inflation outlook improves — and neither has.
There is also a market-structure argument the counter-thesis underweights. In April 2025, the selloff eased once the Fed's communication clarified the path and Treasury auctions found buyers at the longer end. This time, the auctions themselves are the problem: the September 10 buyback shortfall showed demand failing even at a price the Treasury set. When the issuer's own stabilization tool cannot clear the market, the issue is not positioning; it is price discovery finding a new equilibrium.
What Comes Next: Three Horizons
The immediate implication is asymmetrical. For bondholders, duration is no longer a passive safe haven; it is a position that must be paid for. For equity investors, the era in which multiple expansion could be assumed as a tailwind is over. Returns will have to come from earnings, and earnings now face a consumer at a 12-year confidence low and a labor market that is cooling.
Short term, on sentiment and liquidity, volatility stays elevated. The trigger to watch is the next inflation print and any sign that the Fed is nearing the end of its tightening. A softer consumer price index could spark a relief rally in both bonds and stocks, the kind of synchronized bounce that has rescued portfolios before. The specific level to watch is the 10-year yield: a decisive break back below 5% on soft inflation would signal that the worst of the repricing is priced.
Medium term, the earnings cycle meets the rate cycle. If job openings keep falling and confidence stays below 80, the growth scare that the bond market has been pricing will arrive in the data, and equities will reprice lower even if yields stabilize. This is the path where stocks and bonds briefly become friends again — both rallying on bad news because it forces the Fed to stop. The trigger here is two consecutive monthly payroll gains below 50,000, or unemployment rising above 4.5%.
Long term, the term-premium regime has changed. Investors who assume a quick return to the low-yield world of the 2010s are betting against a decade of fiscal arithmetic. The base case is that yields grind higher in a range, with the 10-year between 5% and 5.5%, and equities chop sideways as earnings growth offsets multiple compression. The upside case is a decisive disinflation print and a Fed pause that send the 10-year back toward 4.5%, restoring the stock-bond correlation that made diversification work. The downside case is the stagflationary mix: confidence and job openings deteriorate into an outright contraction while yields stay elevated — the one combination that has no good hedge except cash.
The falsifying signal is specific. If the 10-year Treasury yield falls back below 4.5% within a quarter without a new round of large-scale official intervention, and core inflation prints at or below 0.2% month over month for two consecutive months, then the structural repricing thesis is wrong and this was a positioning overshoot. Conversely, if the 10-year holds above 5.5% while core inflation stays subdued, the market is pricing fiscal risk, not inflation — and the equity multiple will keep compressing.
Tuesday was not simply a bad day for bonds and stocks. It was the market finally charging a realistic price for lending to the US government for 30 years — and discovering that the same price is too high for stocks to ignore.
Explore more exclusive insights at nextfin.ai.
