US inflation increased at fastest pace in three years in April as Iran war hikes up prices

The second-order effect here is not just higher inflation, but a delayed policy regime shift: the Fed’s reaction function is moving from “look through” to “validate persistence,” which mechanically keeps front-end real rates elevated even as growth softens. That combination is toxic for duration-sensitive equities, levered balance sheets, and rate-conditional demand sectors because it compresses both valuation multiples and future earnings quality at the same time.

The consumer signal is more important than the headline PCE print. Real disposable income rolling over for multiple months usually precedes a visible step-down in discretionary volumes with a lag of 4-8 weeks, and the first place it shows up is not broad retail, but basket trade-down, promotion intensity, and weaker conversion in mid-tier brands. That means the market should be more worried about margin compression in retail/apparel, restaurants, and travel than about a simple unit-volume slowdown.

Energy is the obvious near-term inflation driver, but the more interesting spillover is supply-chain re-pricing in imported input categories: freight, fertilizers, packaging, aluminum, and foodservice costs can all remain sticky even if gasoline eases. That creates a nasty mix for downstream users because input costs stay firm while consumers’ ability to absorb price increases is fading; the result is margin squeeze before outright demand collapse.

Consensus may be underestimating how political pressure changes the distribution of outcomes. If inflation stays hot into the summer, the White House has incentives to attack the Fed and potentially seek any visible relief via energy policy, tariffs, or jawboning, which raises headline volatility but does not quickly restore real incomes. The more tradable contrarian angle is that the market may already be pricing a mild slowdown, but not yet a sharper earnings reset if consumer retrenchment becomes visible in June-July data.