Somewhere in Ohio this morning, a mortgage broker is quoting a rate off a screen in New York, and the number on that screen is hovering just under 5%: the yield on the ten-year United States Treasury. The same number, adjusted for a spread here and a hedge there, sits inside a Hong Kong family office's hurdle rate, a Japanese life insurer's hedging programme, and the discount rate in every valuation model built this week. It is the nearest thing global finance has to a continuous opinion poll on the future — and at these levels, it is voting on whether the world's asset prices still add up.
The benchmark under everything
The ten-year's authority is partly an accident of design and partly the reward for depth. It is long enough to span a business cycle, liquid enough to trade in size at any hour, and backed by the taxing power of the issuer of the world's reserve currency. American mortgages are priced off it. Corporate bonds are quoted as a spread over it. The equity risk premium — the entire intellectual basis for owning stocks — is defined against it. Carry trades from Tokyo to São Paulo are measured against what it yields after hedging costs.
"Risk-free" is a term of art, not a compliment; the instrument is free of default risk in nominal terms and free of nothing else. But depth begets depth. When a family office in Singapore or a reserve manager in the Gulf wants duration without credit risk, there is one market that can absorb the order without blinking, and its ten-year point is where the world's view of the future gets concentrated into a single quotation.
The term premium, in plain terms
Strip the yield into its parts and it becomes legible. One part is the expected path of short rates — what the Federal Reserve is likely to do, averaged over the decade. The other part is the term premium: the extra yield investors demand for locking money up against everything that can go wrong in ten years, from inflation surprises to supply gluts to politics.
For most of the decade after the financial crisis, that premium was not merely small but negative. Central banks were buying bonds by the trillion, paid nothing to take duration risk, and so nobody else was paid either. The New York Fed's estimates, watched by every rates desk, showed the term premium below zero for years — investors were, in effect, paying the government for the privilege of lending to it. That condition, a creature of quantitative easing, has since reversed: as issuance swelled and inflation stopped being theoretical, the premium turned positive again. It is why a ten-year beginning with 5 is now a level to be argued over, not a misprint.
What its return changes
A world with a positive, moving term premium is a different world. Long yields can rise even while the central bank cuts, because the premium answers to supply and fear, not to the policy rate. Long-duration assets — growth equities, long bonds, property priced off low cap rates — have lost the tailwind that carried them through the years of free duration. Foreign buyers find the arithmetic of owning Treasuries hostage to hedge costs, which with the dollar near ¥154 is no small print for Tokyo. So auctions matter again, and a weak tail at a ten-year sale can travel around the world before lunch in New York.
This is the part most equity investors still underprice. With the S&P 500 near 7,657, the discount rate in their models is not a fact of nature; it is an output of a bond market that is once again charging for time. Every multiple-expansion story told since 2023 has been, knowingly or not, a bet that the premium stays quiet.
What to watch from here
The poll never closes, but some questions are now permanent fixtures. Watch the Treasury's issuance mix — bills versus coupons — because it decides how much duration the market must absorb this autumn and next. Watch the term-premium estimates, imperfect as they are, because they separate what the Fed controls from what it does not. Watch the interest bill of the United States government, which now rivals the defence budget and concentrates minds in election years. And watch the auction tails, the bond market's way of clearing its throat.
The honest position for the rest of the decade is humility about which regime we are in. If the premium's return proves cyclical, the old playbook survives and 5% is a level to be bought. If it is structural — a world of chronic supply meeting price-sensitive buyers — then the ten-year will keep setting terms that asset owners did not choose. Either way, the quotation on the screen in Ohio is where the answer appears first.
