The ‘real’ deal — world bonds grind towards higher neutral rates
Investors are demanding higher returns from governments
Scott Bessent is correct to say the latest jolt to U.S. Treasuries is a global phenomenon. But that’s cold comfort for the Treasury secretary and his G7 counterparts as bond investors demand higher returns from governments, pricing in stronger growth, heavier business borrowing and investment, and further central bank rate rises.
As summer draws to a close, cooler September air has chilled sovereign bond markets again this week, just as traders brace for rate hikes in Japan, Europe and potentially the U.S. over the next three weeks.
Economically sensitive U.S. 10-year Treasury yields rose to their highest since President Donald Trump returned to the White House early last year. Ten-year Japanese government bond yields topped 3 per cent for the first time since 1996, and Germany’s 10-year Bund yields hit their highest in 15 years. Thirty-year British gilt yields reached levels not seen since 1998, and French 30-year bond yields surged to 18-year highs.
But, as many economists have pointed out over recent months, a breakdown of those yields suggests the latest move is being driven, at least in part, by “real,” inflation-adjusted yields — not inflation expectations themselves, nor the murkier “term premium” that captures compensation for long-term uncertainty around debt loads or inflation.
Real U.S. 10-year rates have jumped about 40 basis points in just three months, surging again over the past week. Japanese 10-year real rates have almost trebled to 0.9 per cent in that period, and French 10-year real rates are some of the highest in the euro era.
Wherever it lies precisely, the “neutral” interest rate that brings these economies to equilibrium seems to be in the process of being revised higher — not necessarily bad news, but a shift that may lift borrowing costs far and wide over the coming years as central banks recalibrate policy for the AI-related investment boom and seek the point at which policy again gains traction on the economy.
Federal Reserve Chair Kevin Warsh appeared to acknowledge that at Jackson Hole last week, emphasizing on several occasions in his keynote speech that current U.S. monetary policy showed little or no sign of bearing down on loan or credit growth, with financial conditions remaining loose. Unless inflation subsides significantly toward its 2-per-cent target — something not expected before the September 16 Fed meeting — Warsh insisted the central bank had “work to do.”
“I would be hard pressed to describe broad financial conditions as restrictive,” he said at the annual symposium in Wyoming on Friday.
Fed futures quickly took the signal and moved to show almost a 70 per cent chance of a Fed hike this month, compared with just a one-in-three shot before that speech. Perhaps most significantly, two rate rises are fully priced by March and a third is half-priced over the next 12 months. Over the following two years of the futures strip, the policy rate — currently at a midpoint of 3.625 per cent — never falls back below 4 per cent through 2028.
The resurgence in global crude oil back above $90 per barrel on renewed exchanges in the Iran war this week didn’t help. But it’s clearly not just about oil.
Coming from the chair, Warsh’s spin on how policy is affecting the economy was eye-catching because it challenges a long-standing assumption among many Fed officials that rates remain slightly “restrictive.” If the Fed rethinks the “neutral” rate that’s neither spurring nor dragging on the economy, it will almost certainly land higher than current levels, and markets may well be shifting their own “real rate” horizon accordingly.
The median of quarterly Fed policymaker forecasts for the long-term nominal policy rate comes in at 3.1 per cent, widely seen as a proxy for “neutral.” But this assessment can change materially: it has already risen from 2.4 per cent in 2022 and stood as high as 3.8 per cent as recently as 2015. Fed models suggest the real neutral rate — “R-star” — is anywhere from 1.0 per cent to 1.65 per cent. Adding a 2 per cent inflation target implies a nominal rate close to the current rate at the upper end of that range.
But with real Fed policy rates — measured against prevailing headline PCE inflation — still effectively zero, the Fed does not appear to be dragging on the economy in any material way — if anything, it may still be stimulating growth even as AI-spurred growth, inflation, investment and financial conditions all suggest otherwise.
With chip giant Nvidia indicating last week that the AI capital-expenditure boom will continue at least through next year — and its sales are expected to rise another 70 per cent in 2028 — the nature of the global economy is potentially shifting, forcing another look at neutral rate models. At the very least, corporate borrowing is rising sharply, and the U.S.-focussed data-center boom is also likely to fan out well beyond American borders in the coming years.
Even if you’re optimistic a productivity boost ensues to allow the economy to grow faster, the buildout and investment phase will put upward pressure on the cost of capital as savings and investment rebalance. Warsh himself nodded to this in comparing the boom with the stagnant years of a savings glut, when no one wanted to invest and interest rates fell to zero.
As an example of how the AI race is lifting global activity despite a year of U.S. trade wars, the Organisation for Economic Co-operation and Development last week said G20 goods trade accelerated through the second quarter. Quarterly import growth rose to 6.7 per cent from 5.2 per cent in the first quarter, with AI-related chips and computing equipment accounting for a large part of the pickup.
It may well be that all the major central banks are in the same pickle as the Reserve Bank of Australia was at the start of the year, when it moved quickly to correct course. It had lost sight of where the neutral rate lay, but knew it had undershot it given the other economic evidence. Its task since has been to raise rates and feel around in the dark towards neutral — a calculation that may itself shift over time. It has effectively admitted that you may only know it when you’re not there.
The Fed and other major central banks may also be coming to that same conclusion.