Latest / The Macro Memo with Fexingo: Daily Conversations on Inflation, GDP, and Federal Reserve Policy / How the Fed Reads the PCE Deflator
Transcript
- Lucas: So the Fed's favorite inflation gauge just came in at 3.3 percent for core PCE in July, and the market barely blinked. But the way that number is built tells you more about where rates are headed than the CPI ever could. Luna: Right, because the Fed has always said it prefers PCE. But why exactly? What's actually different about it? Lucas: It's a good question, and the short answer is that PCE captures the whole economy's spending, not just what households report. The CPI is based on surveys of what consumers say they're buying, but PCE is built from business receipts, so it catches things like employer-provided healthcare and even spending by government programs on behalf of people. That makes it broader and, in the Fed's view, more accurate for setting policy. Luna: So it's not just a different formula, it's a different source of data entirely. Lucas: Exactly. And there's another quirk: PCE weights tend to shift as people change their buying habits. If beef gets expensive and everyone switches to chicken, PCE catches that substitution more quickly than CPI, which uses fixed weights. That's why PCE typically runs a little lower than CPI over time. Right now, headline CPI is around 3.9 percent, but core PCE is at 3.3. That gap is the Fed's whole debate. Luna: And the Fed is clearly sweating that 3.3 number. Hammack said 'now is the time to act' on rates just yesterday. Warsh is talking about inflation concerns too. So what does the PCE actually signal for the next meeting? Lucas: Well, the core PCE has been stuck in that 3.2 to 3.4 percent range for months. It's down from the four percent peak in 2024, but it's not falling fast enough to get to the Fed's two percent target. And the 10-year breakeven rate, which is the market's inflation expectation, is sitting at 2.33 percent. That's above two percent, so the market itself is pricing in that inflation will stay a bit sticky. Luna: So the market is saying, hey Fed, you're not done yet. Lucas: Precisely. And the Fed's own projections from June had one more rate hike this year. The question is whether they pull the trigger in September or wait until December. The PCE report is the key input, and this one wasn't the smoking gun for either side. Luna: It's interesting because the jobs market is cooling at the same time. Unemployment is at 4.1, job openings are down. So the Fed has to balance inflation versus a soft landing. How does PCE factor into that balancing act? Lucas: That's the crux. If the Fed hikes too much, they risk breaking the labor market. But if they stop too early, inflation could reaccelerate. The PCE is the gauge they're using to decide. And there's a subtlety: the Fed looks at the three-month annualized rate, not just the year-over-year. That's been running slightly below the three percent level, which gives the doves some cover. But the latest month was a bit warmer, so it's a coin toss. Luna: So when we hear about core PCE, what's actually in it? Is it just services? What's driving it right now? Lucas: Core PCE excludes food and energy, so it's mostly services: housing, healthcare, financial services, and also goods like cars and furniture. Right now, the big driver is services inflation, particularly in categories like auto insurance and medical care. Those have been stubborn. Goods prices have actually been deflating for a few months, which is helping the overall number, but services are keeping it elevated. Luna: So if goods are deflating, why isn't core PCE lower? Is it because services are that heavy? Lucas: Exactly. Services are about 60 percent of the PCE basket, so even if goods prices drop, services can outweigh them. And service inflation is stickier because it's tied to wages. Wages are still growing around 4 percent, and that feeds into service prices. The Fed knows that to get services inflation down, they need to slow the labor market, which is a painful process. Luna: That's the real trade-off then. And I guess that's why the Fed is so fixated on this number, even if it doesn't match what people feel at the grocery store. Lucas: Right, and that's a common criticism: the PCE doesn't reflect the consumer experience as directly as CPI. But the Fed is trying to set policy for the whole economy, not just a typical household. And for that, PCE is the more comprehensive tool. Luna: So how should we interpret the 10-year breakeven at 2.33? Is that a sign the Fed is losing credibility? Lucas: Not necessarily. The breakeven is the market's expectation for average inflation over the next decade. It's been in a range of 2.2 to 2.5 for a while. At 2.33, it's above the Fed's target, but it's not a dramatic breakout. It says the market thinks inflation will settle a bit above target, which is consistent with a Fed that's reluctant to go too far. The Fed watches this closely because if it starts to climb above 2.5, that's a red flag. Luna: So it's a watch item, not a trigger. But with Hammack saying 'act now' and Warsh wanting a quieter Fed, there's clearly internal debate. Lucas: Absolutely. And that debate is going to play out in the data over the next few weeks. The next PCE report comes out in late September, just before the FOMC meeting. If core PCE comes in at 3.2 or lower, the doves have a case. If it stays at 3.3 or higher, the hawks will push for a hike. That's the timeline. Luna: And what about the fact that the Fed's preferred gauge is the PCE, but the public and politicians talk about CPI? Does that create confusion? Lucas: It does, and it's a real communication challenge. When the Fed says inflation is 3.3 percent and people see grocery prices up 5 percent, they think the Fed is out of touch. But the Fed is trying to explain a different measure. That's why you hear Fed officials like Warsh talking about 'quieter' communication. They know the gap matters. Luna: That makes sense. So for someone who wants to track the Fed's next move, what's the one number to watch? Lucas: I'd say the three-month annualized core PCE. It smooths out the monthly noise and gives you a sense of the trend. Right now it's hovering around 2.8 to 3 percent, which is still above target but not accelerating. If that number starts to creep up, that's the strongest signal for a hike. Luna: So it's not just the headline, it's the trajectory. Lucas: Exactly. And that's the kind of nuance that gets lost in the daily headlines. The Fed is data-dependent, and the data is pointing in one direction: slow improvement, but not enough to declare victory. Luna: Before we wrap up, I want to say, if walking through the economy with us has made something click for you, these episodes stay ad-free because of listener support. If today was useful, you can keep that going at buy me a coffee dot com slash fexingo. Lucas: Yeah, that's a genuine help. It lets us keep digging into the data without worrying about the sponsor read. Appreciate anyone who does that. Luna: So, back to the PCE. If I'm a regular listener, what's the one thing I should remember from today? Lucas: That the Fed's favorite inflation gauge is built for the whole economy, not your shopping cart. It's broader, it adjusts for substitution, and it's the number that will drive the next rate decision. And right now, it's telling us the Fed is in a holding pattern, but the next report could tip the scales. Luna: And we'll be here to walk you through it when it lands. Lucas: That we will. Thanks for listening.