What happened
Core inflation in the United States held at 3.3 percent in the year to July, unchanged from June, according to the personal consumption expenditures price index published by the Bureau of Economic Analysis. The PCE index is a measure of what American households actually pay for goods and services, and it is the gauge the Federal Reserve watches most closely when setting interest rates.
The headline measure, which includes food and energy, came in at 3.7 percent over the same twelve months and rose 0.2 percent on the month, landing roughly 0.1 percentage points above what economists had pencilled in. Core PCE, which strips out food and energy because those prices swing for reasons unrelated to the underlying economy, also rose 0.2 percent month on month, exactly in line with forecasts.
Underneath the top line, the composition was revealing. Prices for physical goods actually fell 0.1 percent over the month, dragged down by a 2.7 percent drop in petrol and other energy related goods and a 0.9 percent fall in furnishings and long lasting household equipment. Services told the opposite story, rising 0.3 percent, driven by a 1.2 percent jump in financial services and insurance and a 0.3 percent increase in housing costs.
The same release showed personal income up 0.4 percent and consumer spending up 0.2 percent, both stronger than expected. American households are still earning and still spending, which is precisely why price pressures are proving so stubborn.
Why it matters
Inflation has stopped falling. That is the single most important fact in this release. Between 2023 and 2025 the story was one of steady, if slow, disinflation as pandemic era distortions unwound. Through 2026 the core rate has essentially flatlined in a narrow band around 3.3 percent, which means the final leg of the journey to the 2 percent target has stalled.
For the Federal Reserve this is an uncomfortable place to sit. The central bank has a dual mandate, meaning it is required by law to pursue both stable prices and maximum employment. With inflation plateaued well above target, cutting rates risks reigniting price growth. Holding rates where they are risks slowly squeezing the labour market instead. Neither option is costless, and the July data gives the doves on the committee very little ammunition.
The services component is the real problem. Goods inflation has largely normalised, helped by cheaper energy and healed supply chains. Services inflation is driven by wages, rents and insurance premiums, and those are sticky by nature. A landlord does not cut rent because petrol got cheaper, and an insurer does not reprice a policy halfway through its term.
The knock on effects reach far beyond Washington. American interest rates anchor global borrowing costs, so a Federal Reserve that stays on hold keeps the dollar firm, keeps pressure on emerging market borrowers, and limits how far the Bank of England and the European Central Bank can move on their own.
Explained simply
Inflation is like a fever that fell from 40 degrees to 38 and then simply stopped dropping. The patient is out of immediate danger, but nobody is sending them back to work.
Here is the mechanism, step by step. When prices rise faster than a central bank wants, it raises interest rates. Higher rates make borrowing more expensive, so households take out fewer mortgages and car loans and businesses delay investment. Less spending means less competition for the same goods, and sellers lose the ability to keep pushing prices up.
That machinery worked well on the easy part of the problem. The prices that shot up because of shipping bottlenecks and energy shocks came back down largely on their own once those shocks faded. What is left is the hard part: the price of human labour and the price of shelter, neither of which responds quickly to a change in the cost of money.
So the Federal Reserve finds itself like a doctor who has run out of the medicine that works and is left holding the one with side effects. Raising rates further would cool services inflation, but it would do so by putting people out of work. Cutting rates would support jobs, but risks letting the fever climb again.
The plateau also matters for expectations. If households and businesses come to believe that three and a bit percent is simply the new normal, they will build that assumption into wage demands and price lists, and the expectation becomes self fulfilling. That is the outcome central bankers fear most.
What it means for you
If you hold savings, this is good news for now. Rates staying higher for longer means easy access accounts and one year fixed rate bonds should keep paying competitive returns rather than drifting back toward the near zero levels of the 2010s. If you have money sitting in a current account earning nothing, the gap between that and a decent savings account remains historically wide, and closing it is the single highest return admin task available to most people.
If you are borrowing, the picture is less friendly. Long term fixed mortgage rates in the United States are priced off Treasury yields, which move on expectations of future Federal Reserve policy. With cuts pushed further out, anyone hoping to refinance into a materially cheaper deal within the next few months is likely to be disappointed. Credit card rates, which float directly with the policy rate, will not fall either.
For investors, a plateau in inflation alongside steady income growth is a mixed but not disastrous backdrop. Shares can cope with 3 percent inflation provided company earnings keep growing. Bond holders fare worse, because inflation erodes the real value of a fixed coupon. Index linked bonds and shorter dated holdings offer more protection than long dated conventional debt.
The practical move is to check what your cash is actually earning. An account paying 4 percent while inflation runs at 3.7 percent leaves you barely ahead in real terms. One paying 1 percent leaves you meaningfully poorer every year you hold it.
The bigger picture
Central banks have stood at this junction before. In the mid 1990s the Federal Reserve engineered a soft landing by pausing rather than cutting, and inflation eventually drifted lower without a recession. In the 1970s a premature easing let inflation take root, and it took a punishing downturn to remove it. Those two precedents sit on either shoulder of every policymaker reading this data.
What to watch next is the labour market. If hiring cools and wage growth slows, services inflation should follow within a couple of quarters and the case for cuts becomes straightforward. If employment holds up while prices stay put, the Federal Reserve will simply wait, and markets will keep pushing the expected date of the first cut further into the future.
