Latest / Inflation Explained with Fexingo: CPI, Prices, and the Cost of Living for Everyday People / The Cars You Actually Drive vs CPI Inflation
Transcript
- Lucas: So I was looking at a June CPI release from a few weeks back — the headline reading is 334.0, up from 332.4 in April. That's about 2.9 percent annualized, still above the Fed's target. But here's the thing that caught my eye: new vehicles. CPI says new-car prices are up about 27 percent since early 2020. Luna: Twenty-seven percent sounds bad, but it also feels... low, doesn't it? I mean, I see people talk about paying forty, fifty grand for a mid-range sedan now. Lucas: Exactly. The average transaction price for a new car in the US crossed $48,000 last year. That's up about 34 percent from 2020. So there's a seven-percentage-point gap between what CPI says and what buyers actually pay. That's the gap I want to get into today. Luna: So the government is undercounting car inflation by almost a third. How is that possible? Lucas: It's a mix of methodology quirks. The Bureau of Labor Statistics uses what's called a 'hedonic quality adjustment.' The idea is that if a 2026 model has better safety features, better fuel economy, a bigger screen — part of the price increase is actually paying for those improvements. So they strip out the 'quality' component to get the pure inflation number. Luna: Right, so they're saying, 'Well, you're getting more car for your money, so we'll adjust the price down.' Lucas: Precisely. And in theory, that makes sense. If an iPhone costs $1,000 but has twice the storage and a better camera than the $800 model from three years ago, the effective price per unit of capability actually fell. But with cars, the adjustments are much more subjective. And there's another issue: the mix-shift problem. Luna: Mix-shift — that's when consumers change what they buy, right? Lucas: Exactly. CPI tracks a fixed basket of goods. For cars, they pick specific models and track the price of those exact models over time. But what people actually buy changes. If rising prices push buyers from a $35,000 Honda Accord to a $25,000 Honda Civic, the average transaction price in the data might actually drop, even though the price of an Accord went up. CPI doesn't capture that trade-down effect. Luna: So CPI could show car prices flat or even falling, while anyone trying to buy the same car they bought five years ago is paying way more. Lucas: That's the disconnect. And it's not just cars. It happens with housing, with groceries — any category where people can substitute down. But with cars, the gap is especially stark because the mix shift has been dramatic. In 2020, roughly 50 percent of new cars sold were trucks and SUVs. Now it's over 80 percent. CPI's basket hasn't adjusted fast enough. Luna: And that feeds into the Fed's favorite inflation gauge — core PCE — which uses CPI data but weights things differently. So if CPI is undercounting car inflation, PCE is too. Lucas: Correct. The core PCE we saw in May — 130.1 versus 129.67 in April — that's a 3.2 percent annualized increase. And car costs are a meaningful part of that. By the way, if walking through the economy with us has made something click — well, listener support is what keeps this show ad-free and independent. If you ever feel like chipping in, you can find us at buy me a coffee dot com slash fexingo. Luna: And it genuinely helps us keep digging into these numbers. No pressure, but we appreciate everyone who does. Lucas: Alright, back to the gap. So the BLS also publishes a different index — the new vehicle price index, which is part of CPI. That index has been running about 2 percent year-over-year recently. But if you look at the average transaction price data from JD Power, it's been closer to 4 to 5 percent. That's a huge divergence. Luna: And what about used cars? That's been a roller coaster. Lucas: Used cars are even wilder. During the pandemic, the used-car index spiked over 40 percent. It's come down a lot — actually, in May 2026, used-car prices were down about 8 percent year-over-year. But again, that's CPI. The wholesale auction data from Manheim shows a similar trend, but the retail prices consumers actually pay at dealerships haven't fallen as much. Because dealers have more pricing power now. Luna: So the CPI's used-car number might be showing disinflation, but on the ground, it doesn't feel that way because the cars that are affordable are older and have higher mileage. Lucas: Exactly. And that's another mix-shift. People who can't afford a $48,000 new car are bidding up the price of 5-year-old cars. CPI tries to account for age and mileage with hedonic adjustments, but it's rough. The net result is that official inflation numbers understate the true cost of transportation for most households. Luna: Which matters a lot because the Fed is making policy based on those numbers. If the data says inflation is cooling, they might cut rates. But if the actual experience is sticky, they're behind the curve. Lucas: Right. And that's the real-world implication. The labor market is already softening — June payrolls came in at just 57,000, unemployment at 4.2 percent. If the Fed cuts rates based on a rosy inflation picture, they risk reigniting demand. If they hold steady, they might overshoot on the downside. The car inflation blind spot is one more reason to be cautious. Luna: Let's talk about financing, because that's another layer. The average new-car loan rate is now over 7 percent, up from 4.5 percent in 2020. So even if the price of the car only went up 27 percent, the monthly payment is up a lot more because of interest. Lucas: That's a great point. CPI doesn't include financing costs. It only captures the cash price. So if you're financing, your real cost is higher. And with longer loan terms — 72, 84 months — you're paying more interest overall. The effective monthly burden has risen far more than CPI suggests. Luna: And that feeds into the broader inflation perception gap. People feel it in their monthly payments, not just the sticker price. Lucas: Exactly. The Fed's preferred measure, core PCE, actually does include some imputed financial services — but not auto loan interest specifically. So again, a gap. At the end of the day, if you're trying to understand inflation, you have to look past the headline number and ask: what's the methodology, what's being left out, and what do people actually experience? For cars, the answer is: a lot. Luna: And given that the Fed might be at a turning point — with rates flat at 3.63 percent and the labor market cooling — getting this right matters more than ever. Lucas: It does. So next time you see that CPI number, remember: the car you drive might be telling a different story.