Latest / The Federal Reserve Podcast with Fexingo: Interest Rates, FOMC Meetings, and Monetary Policy / Why the Fed Trusts PCE Over CPI
Transcript
- Lucas: So the latest CPI numbers came out and they looked pretty good — headline CPI down to 332.6, core basically flat at 336.1. Markets cheered, talk of rate cuts by year-end came back. But then a week later we got the PCE reading, which the Fed actually targets, and it told a different story: PCE up to 131.5, core PCE up to 130.1. That's a divergence worth talking about. Luna: Right — the two inflation measures are moving in opposite directions. That's got to make the Fed's job harder. Lucas: Exactly. And it's not just a one-month blip. The gap between CPI and PCE has been widening for a few months now. To understand why, you have to look at what each index actually measures. CPI is a fixed basket of goods — it tracks what urban consumers pay out of pocket. PCE is broader; it captures what all households spend, and it adjusts the basket as people change their buying habits. Luna: So PCE is more flexible — it accounts for substitution. If beef gets expensive, you buy chicken, and PCE captures that. CPI doesn't, at least not as quickly. Lucas: That's right. And the current divergence is partly driven by that substitution effect. But there are two other big factors right now: housing and healthcare. CPI weights housing far more heavily — over 40 percent — and those OER, owner's equivalent rent, numbers have been sticky. PCE has a smaller housing weight and uses a different rental measure that's been softening. Luna: And healthcare? CPI uses out-of-pocket costs, but PCE includes employer-paid premiums, which have been rising faster. That's actually pushing core PCE up. Lucas: Exactly. So you get a picture where the CPI decline is being driven by falling goods prices — import prices from China have been dropping, commodities are down — but the PCE is being held up by services inflation, especially healthcare and insurance. The Fed watches PCE because it's supposed to reflect the true cost of living for the whole economy, not just out-of-pocket spending. Luna: But the market still reacts more to CPI. Every month traders look at CPI and move yields around, then they're surprised when PCE comes in hotter. Lucas: That's the tension. The ten-year breakeven inflation rate, which is market-based, has actually fallen to 2.26 percent — that's down slightly. So bond traders are pricing in lower inflation ahead, but the PCE data is still above the Fed's 2 percent target. Core PCE is running around 2.6 percent by my math, which is down from last year but not low enough to declare victory. Luna: So what does the Fed do? They've been on hold at 3.63 percent for months now. Kevin Warsh gave that speech last week — three key phrases people are parsing. Lucas: Right — Warsh said they need 'greater confidence' in inflation coming down, which everyone took as hawkish. But if you look at the data mix, the Fed's in a bind. Goods deflation from China is one-time stuff; services inflation is more persistent. The PCE tells them the underlying trend isn't cooling fast enough. So they're likely to stay on hold until they see sustained improvement in PCE, not just CPI. Luna: That could mean no cuts this year. The market's still pricing in at least one cut by December, but if PCE stays up, that might be too optimistic. Lucas: And that's the thing — the Fed has been clear they're data-dependent, and they've said they watch PCE. So if you only follow CPI headlines, you're missing half the picture. That's why we do these deep dives — to help listeners understand what the Fed actually sees. Luna: Speaking of which — a quick thing. We're able to do episodes like this because of listeners who chip in through Buy Me a Coffee. It's ad-free, it's listener-funded. If you get value from these deep dives, that's literally what keeps them coming. Lucas: Yeah — it's a modest way to support the show. No ads, no sponsors. People can find us at buy me a coffee dot com slash fexingo. Even a small monthly contribution makes a difference. Luna: And we really appreciate it. Okay, back to the divergence. So Lucas, you mentioned the housing weight difference — how much does that account for right now? Lucas: A lot. In CPI, shelter is about 44 percent of the core index. In PCE, it's closer to 18 percent. And the CPI uses this lagged measure of rents that's still elevated from the 2022 surge, while PCE uses a more current rent index that's actually decelerating. So the CPI is overstating housing inflation relative to PCE. That alone could explain maybe 0.3 to 0.4 percentage points of the gap. Luna: So the true inflation picture is probably somewhere in between. But the Fed has to pick one. Lucas: And they've picked PCE, for good reason. It's more comprehensive, it adjusts for substitution, and it's revised with better data later. But in real time, PCE is harder to forecast. That's why the Fed often talks about a 'broad set of indicators' — they look at the Dallas Fed trimmed mean, the Cleveland Fed median CPI, and the sticky-price CPI. All of those show inflation still above target but gradually slowing. Luna: Let's bring in the tariff angle. Trump's new global tariff that just got announced — how does that affect the CPI versus PCE divergence? Lucas: Tariffs directly raise import prices, which show up in CPI goods quickly. But PCE goods also capture it, with some lag. The bigger effect is through supply chains and expectations. If tariffs are seen as inflationary, that could push the breakeven rate up. But right now, the breakeven is falling, so markets seem to think tariffs are deflationary because they depress demand. It's a mess. Luna: And the World Cup boost to services spending — that's temporary, right? That might explain some of the PCE services strength. Lucas: Exactly. The Fed's Beige Book noted that bars and restaurants got a boost from the World Cup. That's a one-time spike. So if you strip that out, core PCE might actually be lower. But the Fed can't just strip out data; they have to look through the noise. It's a judgment call. Luna: So the bottom line for rate cuts: if the Fed wants to see core PCE at 2 percent or below, we're not there yet. And with the divergence, they're not going to cut based on CPI alone. Lucas: That's right. The next PCE release is in late August. If it shows another uptick, any hope of a September cut will evaporate. If it ticks down, maybe we get a November or December move. But the base case is 'higher for longer.' The Fed is comfortable waiting until the data is clear. Luna: And that's probably the most important takeaway for this episode: don't get fooled by CPI headlines. Look at PCE, look at the breakeven rate, and remember the Fed is patient. Lucas: Exactly. Thanks for joining us. We'll be back next week with another look under the hood of the economy.