At first glance, the Federal Reserve still looks suspended in familiar ambiguity: waiting, watching, refusing to move too quickly. But the minutes of its March meeting reveal something more telling, and more unsettling. Beneath the official caution, a harder shift was already underway. The old debate about when to cut rates is no longer the only one on the table. Another question has returned, one that markets had preferred to treat as an embarrassment from another age: what if the next move is not down, but up? That possibility, once marginal, is no longer whispered at the edges of the room. A growing number of Fed officials were clearly worried that the war in Iran could entrench inflation more deeply, through the oldest and most brutal channel of all: energy. Oil shocks do not remain politely confined to petrol stations and freight bills. They spread. They affect transport, production costs, food prices, expectations and, eventually, the public mood. And once inflation expectations start to shift again, central banks find themselves in the most thankless of positions, forced to choose between tolerating too much inflation or crushing demand more harshly than they had intended.

The minutes show that policymakers were grappling with two rival dangers at once. On the one hand, there was the risk of a weaker labour market should the war drag on and begin to damage activity more broadly. That would, in the more familiar script, argue for lower rates. On the other side stood the inflation threat: higher energy prices lasting longer, feeding into core inflation and perhaps even into long-term expectations. That, in the darker script, could require tighter policy. Not later, perhaps, but eventually. And it is that second fear which seems to have gained ground. This is where the language matters. At the March meeting, some officials wanted the post-meeting statement to make explicit that rates could rise under certain conditions if inflation remained too high. That may sound technical. It is not. In the codified language of central banking, this was a signal. Not of imminent tightening, certainly, but of a growing discomfort with the market’s old assumption that the Fed’s next instinct would always be to rescue growth first. The inflation hawks were not merely defending vigilance. They were trying to reopen the possibility of restraint.

That shift was not born in a vacuum. The meeting took place only weeks after the outbreak of the war, when the first shock to energy markets had already become impossible to ignore. Since then, the picture has only become more complicated. A ceasefire has been announced, then questioned. Direct talks have been promised, only to be clouded by fresh tension. The Strait of Hormuz, that narrow corridor through which so much of the world’s energy must still pass, remains politically uncertain and operationally fragile. In other words, the market may enjoy moments of relief, but the Fed cannot base policy on relief rallies and diplomatic theatre. It must think in terms of persistence. And persistence is precisely what worries it.

The most important point in the minutes is not that the Fed is preparing to raise rates tomorrow. It is not. The committee kept the policy rate unchanged at 3.5% to 3.75%, and most officials still appeared reluctant to move without clearer evidence. The real point is subtler and more serious. The committee increasingly sees a world in which both sides of its mandate can deteriorate at once: labour weakens, while inflation proves sticky or even re-accelerates. That is the truly toxic mix, because it strips the central bank of clean choices. It leaves the institution neither comfortably hawkish nor comfortably dovish, but trapped in the old stagflation dilemma that modern central bankers prefer to discuss in academic terms rather than admit in real time. The labour market, for now, still offers some cover. Most officials expected unemployment to remain broadly stable, though many acknowledged that employment risks were tilted to the downside. The labour market, in other words, is not collapsing. But neither does it look invulnerable. In a period of weak net hiring, even a modest negative shock can matter more than it appears. That fragility explains why most officials are not rushing to tighten policy. Yet the same fragility does not cancel the inflation risk. It merely complicates the response.

And that is precisely the Fed’s problem now. Inflation has already been above target for years. The committee knows that another energy-driven surge would not arrive in an innocent environment. It would arrive after five years in which price stability has already been bruised, credibility repeatedly tested, and households once again taught that inflation can return far more easily than central bankers like to admit. In such a world, another oil shock is not just another data point. It is a credibility test. The market still wants a simpler story. It wants to believe that if growth slows, the Fed will cut. Perhaps it will. But the minutes suggest the institution is no longer willing to promise that comfort in advance. The real message is harsher: if the Middle East crisis leaves behind a longer inflation tail, the Fed may be forced to choose discipline over rescue. That is not yet the base case. But it is no longer implausible. And once that enters the room, monetary policy changes its tone. So does the world.