Latest / The Federal Reserve Podcast with Fexingo: Interest Rates, FOMC Meetings, and Monetary Policy / What the Feds Preferred Inflation Gauge Misses About Services
Transcript
- Lucas: The core PCE price index hit 3.4 percent in May, the highest since October 2023, and I keep seeing headlines saying the Fed's preferred gauge is flashing a warning. But I think the bigger story is what that number actually measures — and what it misses. Luna: You're talking about how core PCE consistently understates services inflation, right? Because it's weighted differently than CPI. Lucas: Exactly. And it's not just weighting. It's the components themselves. If walking through the economy with us has made something click — especially around how inflation data actually gets constructed — I'll just say, a couple of dollars a month is genuinely what keeps these episodes ad-free. Buy me a coffee dot com slash fexingo, if you've gotten something out of them. Luna: Yeah, it really does make a difference. And speaking of construction — the big one in core PCE is owner's equivalent rent, which is essentially an imputed number, not a transaction. Lucas: Right, so owner's equivalent rent — OER — makes up about 24 percent of core PCE. The Bureau of Economic Analysis imputes what homeowners would pay to rent their own homes, and that imputation lags real market rents by six to twelve months. So when rents were surging in 2022 and 2023, OER kept pulling the index up long after actual leases had stabilized. Luna: And now we're seeing the opposite effect? Rents have been cooling in many markets, but OER is still catching up to the earlier increases? Lucas: Partly. But here's the twist — medical services are another big chunk, about 16 percent of core PCE. And the BEA uses a different methodology than CPI for medical prices. They track the actual reimbursement rates paid by insurers and government programs, which tend to be sticky and slow-moving. So even when spot prices for doctor visits or procedures change, PCE lags. Luna: That means the 3.4 percent reading might actually be understating current services inflation, because the methodology is backward-looking. Lucas: Precisely. And when you look at the more timely CPI services ex-housing — that's running closer to 4.8 percent year-over-year. So the Fed is ostensibly targeting PCE, but the lag in services components means they could be reacting to inflation that's already passed, while missing the current pressure. Luna: Which brings us to the ten-year breakeven inflation rate, which dropped to 2.20 percent this week. That suggests bond traders expect inflation to settle near the Fed's target. But if PCE is systematically undercounting services... Lucas: Then the breakeven could be giving a false sense of comfort. The fed funds rate is at 3.63 percent, effectively flat since May. The FOMC has signaled one or two cuts later this year, but if services inflation is actually higher than PCE shows, they might have to hold rates higher for longer. Luna: Let's zoom in on one specific service category that highlights the gap. What about car insurance? That's been spiking in CPI but how does it show up in PCE? Lucas: Car insurance is a great example. In CPI, it's weighted at about 2.9 percent and has been rising at double-digit rates — up 15.2 percent year-over-year in the latest CPI. In PCE, the weight is lower, around 1.6 percent, and the BEA uses a different source — they look at premiums written minus claims paid, which can smooth out the volatility. So the same real-world price increase gets diluted. Luna: So the Fed's own target might be telling them things are better than they actually are for consumers. Lucas: Right. And the FOMC knows this. At the June meeting, several participants noted that core services inflation excluding housing remains 'elevated' and 'sticky.' But they still use PCE as the primary metric because it's more consistent over time and accounts for substitution effects better than CPI. The trade-off is that consistency comes at the cost of timeliness. Luna: So if you're a bond trader looking at the 3.4 percent core PCE print and thinking the Fed will cut soon, you might be misreading the situation. Lucas: I think there's a real risk of that. The market is pricing in a cut by September, but if you strip out the lagging OER component and look at market rents today, they're still rising about 3 percent annually — not falling. And the Atlanta Fed's sticky-price CPI, which tracks services that change price infrequently, is running at 4.2 percent. Luna: What about the international side? The headlines this week about the Strait of Hormuz and the fragile shipping recovery — does that feed into services inflation through shipping and logistics costs? Lucas: It could, but that's more goods-driven. The services inflation story is more domestic — it's about labor costs, rent, and healthcare. Shipping disruptions affect goods prices, which have been deflating slightly. But if the situation escalates and energy prices spike, that could bleed into transportation services and broader inflation expectations. Luna: So the Fed has to weigh a core PCE that says 3.4 percent, but with components that suggest the real number might be closer to 4. And they have to decide whether to cut rates into that. Lucas: That's the central tension. And I think the way to resolve it is to watch what the Fed calls 'supercore' inflation — core services excluding housing. That's been running around 3.6 percent in PCE terms. If that number doesn't come down in the next two prints, the FOMC will likely hold steady. Luna: And the breakeven inflation rate falling to 2.20 percent — is that the market saying it trusts the Fed, or is it mispricing risk? Lucas: Probably a bit of both. The breakeven is an average of expectations over the next ten years, so it smooths out near-term noise. But if the Fed is targeting a metric that understates current inflation, the breakeven could be anchored too low. I'd be watching the five-year, five-year forward breakeven — that's currently around 2.35 percent — for any signs of de-anchoring. Luna: And what about the impact on the bond market? The ten-year Treasury yield is at 4.37 percent, which is actually a bit higher than where it was before the PCE release. Lucas: That suggests the bond market is already skeptical that the Fed will cut aggressively. The real yield on ten-year TIPS is about 2.17 percent — that's well above the Fed's estimate of the neutral rate, which is around 1.2 percent. So monetary policy is restrictive, but if inflation is stickier than PCE shows, then real rates might not be as high as they appear. Luna: So the bottom line is: the Fed's preferred inflation gauge has a structural blind spot on services, and that blind spot could lead them to ease prematurely if they rely too heavily on the headline number. Lucas: Exactly. And the irony is that the Fed itself has acknowledged this — Chair Powell has said they look at a 'broad set of indicators,' not just PCE. But the market still fixates on the single number. The next core PCE release in July will be the one to watch, because it will include the May data that we just got, plus any revisions. If that number stays above 3.2 percent, I think the September cut gets taken off the table. Luna: And if it drops below 3 percent, the market will immediately price in a cut. But we'd have to ask whether that drop is real or just a statistical artifact of the lagging components. Lucas: That's the million-dollar question. I think the smartest thing the Fed could do is publish a supplementary dashboard that shows core PCE alongside a real-time services inflation tracker — something like the Indeed wage tracker or the Zillow rent index. They already have the data internally. Making it public would help the market see what they see. Luna: But they won't do that because it would reduce their flexibility in messaging. Lucas: Probably true. So for now, we're left to parse the gaps ourselves. The May core PCE at 3.4 percent is the headline, but the real story is in the components that move slowly and get less attention. Luna: And the fact that the ten-year breakeven is at 2.20 percent while actual services inflation is likely higher suggests either the market is incredibly confident in the Fed's ability to bring inflation down, or it's just not looking closely enough. Lucas: I'd bet on the latter. But that's what makes this show fun — we get to look closely. And if you want to keep these deep dives coming, you know where to find us.