The bond market is not worried about oil. It is worried about the Fed

THE BOND market does not trust Mr Kevin Warsh. That is the lesson of the Treasury selloff that has sent the yield on the 30-year American government bond above 5.2%, the highest level since 2007. The trouble is that Mr Warsh is making things worse.

On July 29th the Fed's rate-setting committee voted 9-3 to hold interest rates at 3.5-3.75%. Three officials - Beth Hammack, Neel Kashkari and Lorie Logan - preferred a quarter-point hike. The split was the second sign of division under Mr Warsh, who took the chairman's post in May. After the vote, Mr Warsh held a press conference. He declined to say what would move him to raise rates, suggested the Fed might one day abandon its official inflation measure, and offered little clarity on why policy was at the right place. The market reacted in a way that would have been dismaying even to a less thin-skinned chairman. Long-dated Treasury yields surged while the two-year yield fell, a pattern that says investors believe the Fed will not act now but may be forced to tighten aggressively later. The 10-year yield climbed to 4.74%, its highest since January 2025.

The surface explanation for rising yields is the war in Iran. Oil prices have jumped from around $70 a barrel in February to over $90, disrupting tanker traffic through the Strait of Hormuz and feeding through to gasoline, shipping and food costs. Mr Musalem has warned that the oil shock is the third negative supply shock in 12 months, following tariffs and tighter immigration rules, and is likely to keep underlying inflation near 3%, well above the Fed's 2% target.

To be sure, the energy shock is real. Yet bond markets are not pricing in a temporary supply blip. They are pricing in a credibility gap. The distinction matters. Supply shocks push prices up once; credibility problems push them up repeatedly, as businesses and workers build higher inflation into wages, contracts and expectations. That is the mechanism that turned oil crises into full-blown inflationary spirals in the 1970s. The danger today is not an immediate replay, but a slower erosion: inflation expectations that tick upward just enough to make disinflation harder, which forces central banks to keep policy tighter for longer than otherwise necessary.

The irony is not lost on Mr Warsh himself. Before taking office, he repeatedly criticised his predecessor, Mr Jerome Powell, for losing credibility when long-end Treasury yields rose following rate cuts. He understood then that the problem was not merely inflation data, but whether markets believed the Fed would act decisively to restore price stability. The situation after his second meeting looks uncomfortably similar. The bond market read his press conference as confused and contradictory. Jon Hilsenrath, a longtime Fed watcher, wrote to clients that Mr Warsh "didn't convey the message clearly or explicitly, and the bond market puked on him".

What the market is demanding is not more words. It is a mechanism: a threshold at which the Fed will tighten, even if doing so risks growth. Without one, the Fed looks reactive rather than credible. Mr Warsh's stated philosophy - that forward guidance made the central bank inflexible - has a theoretical case. Central banks that telegraph every move can be cornered when data shifts. But removing guidance entirely, without replacing it with a clear decision rule, leaves markets guessing. In a system where credibility is earned by consistent action rather than proclaimed in testimony, silence can be read as weakness.

The deeper problem is political. Mr Warsh was nominated by a president who, for years, publicly demanded lower interest rates and attempted to fire a Fed governor who wouldn't comply. Mr Warsh's appointment was partly a response to that pressure, but his ability to maintain the Fed's independence depends on delivering what he promised at congressional hearings: price stability. The Congressional Budget Office projects a budget deficit of $1.9trn this year, growing to $3.1trn by 2036. The government needs the bond market to keep buying. If inflation expectations de-anchor, the Treasury will face higher borrowing costs at precisely the moment deficits are expanding. The incentives of the executive branch, which prefers lower rates and higher asset prices, pull in the opposite direction. The Fed's independence is not conferred by law alone. It is earned by showing that it will do the unpopular thing of tightening when necessary.

There is also the question of what is driving inflation. Fed Governor Lisa Cook has pointed to the AI build-out - the boom in data-centre construction - as a potential source of persistent price pressures for chips, equipment, software and utilities. Mr Warsh has been more sanguine, arguing that supply will eventually catch up with demand. The debate is unresolved. But it illustrates a structural tension: the same investment that could raise productivity and ease inflation over the medium term is pushing prices up in the short run. A central bank that dismisses the near-term pressure while waiting for the long-term payoff risks looking complacent.

The strongest case for the Fed's caution is that inflation may yet cool. June's consumer-price index fell by 0.4%. Housing inflation is easing. The tariff pass-through may be fading. If the Iran conflict settles, energy prices could retreat. On this view, a premature rate hike would needlessly slow an economy that Mr Musalem already expects to grow at only 1.5-2% this year.

Yet that argument depends on three things happening at once: the war ends, oil falls, and the fiscal deficit stops expanding. All three are uncertain. Breakeven inflation rates... still sit near 2.2% for the 10-year horizon, suggesting longer-term expectations are not yet unmoored. But breakevens can be deceptive. They have historically underpriced inflation during periods of supply shock, and they reflect investors who may be betting on Fed action rather than expecting it to happen voluntarily.

The better path is not a rush to hike. It is a clear signal of what would trigger one. Mr Warsh should articulate the threshold: a sustained rise in core services inflation, a break above some level in wage growth, or a de-anchoring in survey-based expectations. He should commit to the PCE index for the time being, rather than inviting speculation that the Fed is looking for excuses to dismiss bad data. And he should make clear that the 9-3 split reflects data, not a political factional fight.

The bond market's message is unambiguous. It does not need the Fed to promise lower rates. It needs to know that the Fed will raise them if necessary. In a world of war-driven oil spikes, persistent fiscal deficits and an AI investment boom that is reshaping demand for energy and labour, credibility is not a virtue. It is the only thing standing between contained inflation and a more expensive, more disruptive episode. Mr Warsh inherited a broken system. He should not preside over breaking it further.

Credibility, like most institutional goods, is cheap until it is gone. Then it costs everything.