Latest / The Federal Reserve Podcast with Fexingo: Interest Rates, FOMC Meetings, and Monetary Policy / Why the Fed Is Losing Its Best Inflation Signal
Transcript
- Lucas: The core PCE price index hit 130.1 in May. That's up from 129.67 in April, which was up from March. It's been rising steadily for five straight months now. Luna: And yet headline inflation feels like it's cooling. The CPI was at 334.0 in May, up from 332.41 — but the year-over-year pace is actually decelerating. So we've got two different stories. Lucas: Exactly. And the Fed's own preferred gauge — core PCE — is telling them something different from what the grocery store and gas station are telling most Americans. That's the tension I want to drill into today. Luna: It's a weird spot. The Fed watches core PCE like a hawk, but if that number is being distorted, they could make a mistake. Lucas: Honestly, if today's conversation makes that tension click for you, that's the link — buy me a coffee dot com slash fexingo. It's the smallest possible ask, and it keeps this show ad-free. Luna: Yeah, it's a quiet way to say the show matters. And it does. Lucas: Back to core PCE. The index itself — it's a price index, not a spending report. It measures the prices of goods and services that people actually consume, and it excludes food and energy because those are volatile. The idea is to see the underlying trend. Luna: But the underlying trend right now is up. And that's not what the Fed wants to see. Lucas: No, it isn't. The Fed funds rate is at 3.63 percent — effectively flat since June. So they've paused. But core PCE is still creeping higher. The ten-year breakeven inflation rate — that's the market's expectation for average inflation over the next decade — that's at 2.25 percent, up from 2.24 the day before. Luna: Tiny move, but directionally wrong for the Fed. Lucas: Right. So the question is: is core PCE actually telling us that inflation is reaccelerating? Or is it being pulled up by components that the Fed should look through? Luna: I assume shelter is the big one. Lucas: It always is. Shelter makes up about a third of the core PCE basket. But here's the thing: shelter inflation in PCE lags market rents by about twelve to eighteen months. What we're seeing in the May reading is the echo of the rent spikes from late 2024 and early 2025. Luna: So even if rents are flattening now — which they are, according to real-time data from Apartment List and Zillow — the index won't reflect it until sometime next year. Lucas: Exactly. The owners' equivalent rent component — that's what homeowners would pay if they were renting their own home — that's especially sticky. It's basically a statistical construct that moves like molasses. So a chunk of that core PCE increase is just a lag effect. Luna: But it's still showing up as inflation in the Fed's eyes. They can't just ignore their own metric. Lucas: They can't. And that's the bind. The other piece is services inflation excluding shelter — things like car insurance, medical care, haircuts. That's been running hot too. The core services ex-shelter index is up about 4.5 percent year-over-year. That's above the Fed's comfort zone. Luna: And those aren't lagging. Those are current. Lucas: Right. Car insurance premiums are still rising because repair costs and replacement parts are more expensive. Medical services are being pushed up by labor costs in healthcare. So there's a genuine demand-driven component that the Fed can't dismiss. Luna: So we've got one big lagging component that's overstating inflation, and one current component that's genuinely elevated. How does the Fed weigh that? Lucas: They look at other data. They look at the employment cost index, which measures wages and benefits. They look at unit labor costs. They look at the Philadelphia Fed's trimmed mean PCE, which strips out the most extreme movers. And they look at market-based inflation expectations, like the breakeven rate I mentioned. Luna: The breakeven at 2.25 is above the Fed's 2 percent target. That's a warning. Lucas: It is. But it's also worth noting that the five-year breakeven is at 2.12, and the five-year-forward breakeven — that's the market's expectation for inflation five years from now — is around 2.3. So the market is pricing in inflation that stays above target for a while. Luna: Which means the market doesn't believe the Fed has won yet. Lucas: Exactly. And that's why some FOMC members are getting nervous. We saw the dot plot in June — the median projection for the fed funds rate at the end of 2026 was 4.1 percent, which is about 50 basis points above where we are now. So the committee still sees one or two more rate hikes this year. Luna: But if core PCE is being artificially boosted by lagging shelter, hiking into that could be a mistake. You'd be tightening based on a ghost. Lucas: That's the risk. And some Fed officials, like Austan Goolsbee at the Chicago Fed, have made exactly that argument. He's said that the Fed should look through the shelter noise and focus on the market-rent indicators that are already showing disinflation. Luna: But the hawks — Waller, Bowman — they look at the services ex-shelter number and say, 'No, that's real. We need to keep tightening.' Lucas: Right. And the data we have today doesn't decisively favor either camp. The May core PCE reading was 0.3 percent month-over-month, which is above the 0.17 percent pace consistent with 2 percent inflation. So even if you strip out shelter, it's still a bit warm. Luna: So the Fed is in a data-dependent waiting game. They need to see a few more months of core PCE prints to decide whether the trend is real or noise. Lucas: But there's a bigger issue. The Fed's own credibility is tied to core PCE. They've spent years telling markets and the public that this is their metric. If they suddenly start saying, 'Well, actually, we're going to look through it,' that's a communication challenge. Luna: They've done it before. In 2021, they said inflation was 'transitory' and looked through the data. That turned out badly. Lucas: It did. So they're cautious. They don't want to be seen as ignoring inflation again. But they also don't want to overtighten and cause a recession. The labor market is already softening — payrolls were just 57,000 in June, and the participation rate fell to its lowest in fifty years outside of COVID. Luna: So they're stuck between a lagging inflation gauge and a weakening jobs market. Lucas: Exactly. If they hike based on core PCE, they risk crushing an already fragile labor market. If they hold, and core PCE keeps rising, they risk losing control of inflation expectations. That's the conundrum. Luna: What would break the tie? What data point could settle it? Lucas: Probably the July and August core PCE readings. If we get two months of 0.2 percent or lower, the doves win. If we get another 0.3 or higher, the hawks will push for a hike at the September meeting. Luna: And the next PCE release is July 31, for June data. That's three weeks away. Lucas: Right. So between now and then, markets will be parsing every piece of data — the CPI release on July 14, the producer price index, the employment cost index. All of it feeds into the core PCE estimate. Luna: It's a high-stakes summer for the Fed. Lucas: It really is. And the thing is, the core PCE index itself — as a construct — might need to be rethought. It was designed in the 1990s, when the economy was different. The weight of shelter, the way services are measured, the lag structures — all of that reflects an older economy. Luna: Are there alternatives? Some economists talk about the Dallas Fed trimmed mean, or the Cleveland Fed median CPI. Lucas: Those are useful. The Dallas Fed's trimmed mean PCE, for example, removes the most extreme 20 percent of price changes each month. It tends to be less volatile and a better predictor of future inflation. In May, it was running at 2.6 percent year-over-year — still above target, but lower than the official core PCE. Luna: So if the Fed used that, they might feel less pressure to hike. Lucas: Maybe. But they've never officially adopted it as a target. The FOMC's statement says they aim for 'inflation at the rate of 2 percent, as measured by the annual change in the price index for personal consumption expenditures.' That's core PCE. Changing the target metric would be a huge deal. Luna: So for now, they're stuck with it. And we're all watching the same lagging number, trying to figure out what it means. Lucas: That's the reality. The Fed's best signal is also its most distorted one. And the next few months will tell us whether they can see through the noise, or whether they react to the echo. Luna: React to the echo. I like that. That might be the title of this episode. Lucas: It's not a bad way to think about it. The Fed is listening for footsteps, but the hallway is full of mirrors.