Latest / Inflation Explained with Fexingo: CPI, Prices, and the Cost of Living for Everyday People / The Inflation Gap Between Used Cars and CPIs New Methodology
Transcript
- Lucas: So the May CPI came in at 334.0, core PCE hit 3.4 percent, and the 10-year breakeven is sitting at 2.22. And the market barely flinched — S&P 500 at 7,491, up another 1.8 percent in the last five days. But there's one component inside those numbers that I think is actually telling us less than it used to: used cars. Luna: That's interesting because everyone I know who's tried to buy a used car recently says prices are still climbing. But the CPI's used-car component has been relatively tame. Lucas: Right. And that's because the Bureau of Labor Statistics quietly changed the methodology in early 2026. They shifted from a pure survey of dealerships to a hybrid model that pulls in auction data from Manheim — the big wholesale auto auction company. The idea was to get more real-time pricing, but the execution has created a lag. Luna: A lag in what direction? Lucas: Under the old system, the BLS surveyed dealers on what they actually sold to consumers. That captured retail prices, but it was slow — dealers would report at the end of the month, and the sample was small. The new system blends auction data, which is wholesale and happens at the dealer to dealer level, with some retail surveys. The problem is that auction prices can diverge from what a buyer pays on the lot, especially when demand is hot. Luna: So you're saying the government is using wholesale price changes to estimate retail inflation for cars? Lucas: Partly, yes. And according to the Manheim Used Vehicle Index, wholesale prices were up 4.2 percent year-over-year in May. But the CPI's used-car component showed only 1.1 percent annual growth. That's a three-point gap. Luna: Three points is massive in inflation terms. The Fed's been obsessed with getting core PCE down. If used cars are actually contributing more than reported, the real inflation picture is hotter. Lucas: Exactly. And the timing matters. The Fed's preferred gauge, core PCE, already hit its highest since October 2023 at 3.4 percent in May. If the used-car component is understated by even half a point, that's significant. Luna: Why would the BLS change the methodology in a way that might understate inflation? Lucas: I don't think it's intentional understatement. It's more about trying to capture volume. The old survey had a response rate problem — dealers were too busy to fill out forms. The new auction data gives them a much bigger sample, but the trade-off is relevance. An auction price is what a dealer pays another dealer, not what you pay. There's markup, financing, and regional variation. Luna: And the lag — auction data is compiled weekly, but the CPI is monthly. So you get a blend of stale auction prices and fresh retail prices, and it's not clear how they're weighted. Lucas: Right. The BLS hasn't released the exact weightings yet — they said they'd publish a technical paper later this year. So for now, we're flying a bit blind. Luna: It feels like the opposite of what the Fed wants. They need accurate data to set rates, and if one of the volatile consumer categories is being mismeasured, it muddies the signal. Lucas: And it matters for more than just the Fed. The CPI is used to adjust Social Security benefits, tax brackets, and even some union contracts. If the used-car component is low, it drags down the whole index, and that has real dollar consequences for millions of people. Luna: So what's the real-world impact for someone trying to buy a used car right now? Lucas: If you're shopping for a three-year-old SUV, the Manheim data suggests prices are up about 5 percent from a year ago. But the CPI says 1.1 percent. So the dealer might be using CPI to justify a smaller price increase, but their costs are actually higher. It's a weird disconnect. Luna: Or the dealer uses the higher wholesale cost to justify a bigger markup. I've seen that — 'my costs have gone up, so the price is higher.' Lucas: Exactly. And because the CPI is a benchmark, it affects negotiations. A buyer who reads 'used car prices up 1.1 percent' thinks they have leverage. But the dealer knows the real market is tighter. Luna: Speaking of benchmarks that don't always match reality — this show itself relies on a kind of real-world check. If walking through the economy with us has made something click, there's a way to keep it going. listener support at buy me a coffee dot com slash fexingo helps us stay independent and focused on these details. Lucas: Yeah, it really does. Every coffee helps us keep the podcast ad-free and digging into the methodology behind the numbers. No pressure, just if you find value in these deep dives. Luna: And speaking of deep dives — Lucas, you mentioned the BLS hasn't released the weightings. Is there any historical precedent for this kind of methodology change causing a measurable distortion? Lucas: Absolutely. In the early 2000s, the BLS shifted from a simple geometric mean to a 'hedonic' adjustment for electronics, which lowered reported inflation for computers and TVs by a lot. That was controversial too. But the difference is that was a deliberate improvement — it accounted for quality changes. This used-car change seems more like a practical fix that introduced a new bias. Luna: So we're in a period where the official data might actually be underestimating inflation for a key consumer good, at a time when the Fed is trying to decide whether to cut rates. Lucas: Right. And the 10-year breakeven — which is the market's expectation of average inflation over the next decade — is at 2.22 percent, up a bit from 2.20. The market is saying inflation will stay above 2 percent. But if the CPI itself is understating, the breakeven might be low too. Luna: It's like a hall of mirrors. Lucas: Yeah. And there's a practical takeaway for listeners: if you're in the market for a used car, don't assume the CPI number reflects the market. Check the Manheim index or just look at what similar cars are listing for online. Luna: And be ready to negotiate hard, because the dealer's costs might be higher than the government says. Lucas: Exactly. And that's the theme of this episode — the gap between what we measure and what we experience. It's not just an academic issue. It affects your trade-in value, your monthly payment, and maybe even the Fed's next move.