Why did new car prices increase dramatically during the pandemic and why haven’t they decreased as quickly even though inventories have improved? It’s been widely suspected that carmakers prioritized building higher trim levels or more profitable models and now, using car listing data, we have some evidence that this likely did occur and that the asking price for entry-level trucks has grown by as much as 40% since the start of the pandemic. I’m calling it trimflation, and it’s likely still impacting the car market.
To be clear, nothing explicitly nefarious is happening here. It’s not a big secret and it makes logical sense that, given the ability to only make a certain number of cars, automakers did not seem to prioritize their cheapest, lowest margin cars. What’s been lacking in conversations about this topic is any quantifying of the existence of trimflation, which is what I’m attempting to do. Additionally, I want to try to put it in context of the larger discussion around inflation and its causes.
It’s become a less controversial idea in mainstream economics that many corporations used the pandemic to raise the prices of their products and services, to a degree beyond what would normally be expected due to production shortfalls. In this school of thought, this has been a contributor to overall inflation. The rapid increase in the price of new cars is a related phenomena, but with its own unique causes and outcomes.
The conventional wisdom is that, faced with a mostly self-imposed semiconductor shortage (we’ll get into that in a bit), automakers chose to prioritize higher-margin vehicles and trim levels. For example, with only enough resources to build a certain number of cars Mercedes may have chosen to make more of its GLC SUV (which starts around $48k) than smaller models the GLA crossover (which starts at $38k). And even within models, the company might have decided to build proportionally more of the AMG GLA 35 trim, which starts at $60k.
Working with a dataset of cars listed for sale before and during the pandemic provided by CarEdge, I’ve been able to isolate what appear to be quantifiable indicators of trimflation. Specifically, I looked at entry-level half-ton trucks as the likely best example of where trimflation can be seen.
This data shows that number of listings for base trim levels of trucks fell by as much as 80% during the pandemic, while prices increased by as much as 40%. It’s time to give you hardcore car enthusiasts what you love the most: deep dives into new car sales data!
A Brief Explanation Of Sellers’ Inflation/Greedflation/Whatever You Want To Call It
Basic economic theory holds that companies, unless they have some sort of monopoly or killer innovation, generally increase (often) or lower (rarely) prices based on fairly predictable factors. If a company makes wooden chairs and the cost of the lumber it uses goes up then, to protect profit margins, it will raise prices. If the price of lumber goes down and demand stays the same then the company might lower prices to stay competitive or, at least, keep prices similar and pocket the profits. If demand for wooden chairs goes down then it might force the company to lower prices and, vice versa, if demand increases it might increase prices.
Early on in the pandemic, there was a real concern that demand for most consumer goods might fall dramatically. At the same time, numerous industries faced serious production issues related to global shutdowns and the many disruptions (notably, shipping) that followed. Ultimately, demand for many consumer goods remained strong or even increased, due partially to increases in government stimulus and partially to the fact that we were all stuck at home for big chunks of time. While there were real cost increases for certain upstream products and commodities (the stuff you need to make stuff), profits for large corporations went up way more than expected.
An economist by the name of Isabella Weber from UMass Amherst wrote a paper that basically said much of the inflation in the United States we assumed was caused by COVID-19-related issues was, in fact, caused by corporate leaders sort of tacitly agreeing to raise prices and keep them up, using the pandemic more as an excuse than a real reason. As proof, Weber and her co-author looked at the very real increases in profit margins for public companies and listened to numerous earnings calls where corporate leaders essentially indicated that they felt comfortable that demand wouldn’t decrease if they kept prices up and that, frankly, everyone else was doing it.
Here’s how she put it in her paper:
The dominant view of inflation holds that it is macroeconomic in origin and must always be tackled with macroeconomic tightening. In contrast, we argue that the US COVID-19 inflation is predominantly a sellers’ inflation that derives from microeconomic origins, namely the ability of firms with market power to hike prices. Such firms are price makers, but they only engage in price hikes if they expect their competitors to do the same
Some people call this greedflation, though Weber uses the more neutral term sellers’ inflation (which I also prefer and will use). Initially, this was an unpopular opinion among economists, with New York Times columnist/economist Paul Krugman calling Weber ‘truly stupid’ in a tweet (he later apologized and deleted the tweet). Over time, however, people have come around, and the German government even brought Weber in to successfully help the country deal with the energy crisis caused by the war in Ukraine.
I highly recommend listening to the excellent Odd Lots podcast with Weber as a guest, where a lot of this is discussed (also, listening to this podcast gave me the idea for this research):
So, did automakers reap record profits because they all just decided to make cars more expensive and build fewer cars? Not quite. Automakers faced a number of real upstream issues that caused them to initially raise prices, including energy costs, labor shortages, and shipping issues.
But the biggest problem they had was a semiconductor shortage that was predominantly their doing and, one assumes, a happy accident.
How Automakers Shot Themselves In The Foot With A Golden Bullet
Remember the early part of the pandemic when it seemed like we’d be inside forever and you spent a good 10-15% of your day wondering if you might need to spray Lysol on the Ikea catalog you just got in the mail? Fun times.
Modern cars use lots of semiconductors, aka microchips, aka chips. In particular, they use a lot of cheap, older ones, because the processing needs of a PlayStation 5 are way higher than the processing needs of the heated seats in your car (a fact we all learned when automakers started delivering cars without heated seats in the middle of winter and blamed the chip shortage). Automakers also use older chips for a reason: they’re proven and they’re safe. If your computer crashes, you turn it back on and maybe lose some work; if your car’s computers crash, people could die, which is why those chips generally don’t.
Assuming demand would crater, most of the major automakers canceled their orders for chips at the same time everyone was buying electronics to work from home. As IEEE Spectrum reported:
With panic, lockdowns, and general uncertainty rolling across the globe, automakers cancelled orders. However, those conditions meant a big fraction of the workforce recreated the office at home, purchasing computers, monitors, and other equipment. At the same time entire school systems switched to virtual learning via laptops and tablets. And more time at home also meant more spending on home entertainment, such as TVs and game consoles.
Automakers quickly realized demand for cars wasn’t decreasing, especially when banks lowered interest rates, but few chipmakers were interested in making lower profit-margin chips and essentially shuffled automakers to the back of the line. They had PS5s chips to make instead and the money was far better.
With limited chips, automakers knew they wouldn’t be able to build as many cars; from 2019 to 2020 the total number of cars sold dropped by nearly 16%. Faced with the choice of building an expensive vehicle with a high profit margin or something with a low profit margin, it’s long been assumed that automakers chose higher profit margin vehicles. This is touched on on Weber’s paper:
Companies in the automobile sector have also amplified price pressures enabled by a form of temporary monopoly granted by the computer chips shortages. This allowed car producers to focus on expensive models with higher margins and generally raise prices without having to fear a loss in market share. General Motors, for example, increased its profit margins and magnitudes in the second half of 2020 and in 2021 due to a combination of pricing and mix.
GM did, indeed, reap massive profits while building fewer cars, and part of those profits likely came from trimflation.
How To Prove That Trimflation Happened
While this activity has long been assumed, automakers don’t generally release sales information with enough detail to determine how many of which trim (LX, EX, Sport, Black Gold Edition, etc) they actually sell. I wanted to quantify this phenomena somehow and had an idea of how to do it. If any part of the market in the United States would be the ideal indicator of trimflation, it would be half-ton pickup trucks. Trucks command large margins and GM, Ford, and Stellantis all rely on their trucks as major profit centers.
Additionally, while automakers keep adding more expensive trims (Laramie Longhorn, Platinum, Denali AT4), they are fairly consistent with the lower trim levels of their half-ton trucks. In fact, Stellantis continues to produce the previous generation Ram pickup truck and markets it as the Ram 1500 Classic. If I could somehow make a smart guess of how many Ford F-150 XL, Ram 1500 Classic, and Chevy Silverado WT pickups were produced then I might be able to show some of the choices automakers made.
Thankfully, I knew someone who’d be able to help with that. The folks at CarEdge use their partner MarketCheck to scrape nearly all dealership websites nationwide every single day. They use this data to try and bring transparency to the industry and help consumers make more informed decisions. I had them pull the number of listings and the average price for every day from Jan. 1, 2019 to June 30, 2023.
The data not only confirmed what we expected, but showed it was probably worse than many realized.
“Our team was blown away by the decrease in entry-level trim options,” CarEdge CEO Zach Shefska told me after reviewing the data. “We expected to see a drop from pre-pandemic times, but the magnitude of that decline was shocking to us. It is clear that automakers put profits before affordability when the “chip shortage” and new vehicle shortage came to be.”
Again, automakers don’t generally release sales data at this level of detail. In fact, Ford doesn’t even break down sales by truck size, with all F-250s and F-150s merely called “F-Series” in Ford sales reports.
While this data is not perfect (sometimes trucks that aren’t sold are listed, fleet sales might be missed, demand is not static over time, and people often negotiate a lower price than what’s listed), it gives a good representation of the market at any given moment. The actual decrease in production numbers depends on a lot of factors, including how many days it takes to sell an individual truck and the gap between being produced and being listed. These numbers can only be used to get a sense of the declines and may not map up identically with production decreases.
Just How Bad It Got
With all those caveats, here are some fun raw numbers. At its peak on March 21, 2020, dealers showed 58,145 Ford F-150 XLs for sale across the United States. By June 23, 2021, just 8,770 were listed for sale.
With trucks being delivered but the country quickly shutting down, Chevy also had a max listing in March 2020 of 30,479 Silverado WT trucks. That number dropped to just 2,745 trucks in October of 2021. The Ram 1500 Classic was already declining when the pandemic began, but saw a similarly big drop, from a January 2019 high of 48,671 trucks to a low of 5,603 trucks in April of this year.
And what about prices? No surprise: they’ve gone up. The price for the F-150 XL, as indicated by the average of all of them listed for sale, hit a high of $49,037 in March of this year, compared to a low of $36,442 October 2019. The Ram 1500 Classic had a low of $33,850 in December of 2019 and a peak of $45,597 at the end of June. Most dramatic, though, has been the Silverado. For the Silverado WT the high came in June of this year at $57,431, with a low of $34,771, also in October 2019.
Looking at averages, From January 2019 to June 2023, the average listed price of Ford, Chevy, and Ram entry-level trucks increased by 24%, 32%, and 25% respectively. In January of 2022 there were, on average, just 22,868 of the Silverado WT, F-150 XL, or Ram 1500 Classic listed for sale, compared to a high of 109,934 in January of 2020.
Because I don’t yet have trim-level data for each and every trim level for every truck, I’ll have to make some informed inferences based on total sales numbers. So, for comparison, Ford sold 896,526 “F-Series” pickups in 2019 (as mentioned, this includes F-150 through F-450 trucks, but it’s mostly F-150s). In 2020, that number dropped to 787,372 trucks, for a decrease of about 12%. By comparison, XL listings dropped, on average, about 43% from January 2019 to January 2020.
The same can be seen with Silverado, which also tends to report light- and heavy-duty sales together. From 2019 to 2022, Silverado annual sales dropped from 575,569 units to 520,936 units, or about 9.5%. From January 2019 to January 2022 the average number of Silverado 1500 WTs dropped by a whopping 78%.
A spokesperson for GM told me the company did prioritize building WTs for its fleet customers (rental companies, government agencies, et cetera) as much as possible while acknowledging that “[a]ll vehicle line and trim mixes saw adjustments due to both customer preference as well as constrained microchip supply.”
Ford, as well, gave me a similar statement:
Global parts shortages continue to affect Ford’s North American plants – along with automakers and other industries around the world. Behind the scenes, we have teams working on how to maximize production, with a continued commitment to building every high-demand vehicle for our customers with the quality they expect.
The Ram Classic demonstrated declines in listings before the pandemic, but the once popular truck was one of vehicles Stellantis called out as being delayed specifically by microchip shortages, telling The Detroit Free Press in March of 2021:
“We continue working closely with our suppliers to mitigate the manufacturing impacts caused by the various supply chain issues facing our industry. Due to the unprecedented global microchip shortage, we are currently building and holding Ram 1500 Classics built at Warren Truck Assembly Plant in Michigan and Saltillo Truck Assembly Plant in Mexico,” according to a statement provided by company spokeswoman Kaileen Connelly.
According to the company’s sales reports, there were 468,344 Ram pickup sales in 2022, a decrease of 18% from 2021 and of 27% off the 633,694 sold in 2019. Between January 2019 and January 2022, Ram 1500 Classic listings dropped a whopping 85%. Stellantis, when presented with some of this data, declined to comment.
While these numbers are not perfectly interchangeable, it’s hard to imagine a plausible scenario where these trucks are produced and not listed/sold and don’t represent a large percentage of the trucks that weren’t produced. Even if a larger number entry-level trucks were produced for fleets and never showed up as listings, the number of trucks available to non-fleet buyers decreased.
What About Non-Trucks?
The other way it’s been suspected that automakers have improved margins in the pandemic is in the production of higher margin vehicle types (i.e. a big SUV over a small car). Curious about this, I also asked CarEdge to pull data for all Honda Civic and all Honda CR-V production. I didn’t pull trim level data because it’s not consistent given that Honda actually dropped its lower levels during the pandemic, which is essentially an admission of trimflation.
“We made the initial decision to eliminate certain trims to help manage the shortage of microchips and other supply issues that limited production during the past two years,” a Honda spokesperson told me in a statement. “ In order to meet very strong consumer demand, we have reintroduced the more affordable LX trim level to certain models like the CR-V and Civic Sedan for the 2023 model year.”
Honda has also created some additional trim levels that, for instance, add or remove features like Blind Spot Information in order to lower the price of certain models.
Honda is also a good test case because the company was amongst the hardest hit companies during the chip shortage. According to the data, Honda dealers listed as many as 85,000 new Civics before the pandemic. By summer of 2022, that number dropped to just 5,000 a day, or a decline of almost 95%. CR-V listings also peaked around 85,000 new daily listings in the summer of 2019, dropping to a low of around 6,000 vehicle listings in Summer of 2022. While slightly fewer Civics were listed, it’s not a huge difference.
What’s interesting is that Civic numbers have only rebounded to about 14,400 new listings in June of 2023 while CR-V numbers are up to 25,400 new listings each day. (Of course, the general trend in the market has been away from cars and towards crossovers, so that definitely also plays a role.)
Is this another form of trimflation? The removal of a base trim definitely was, but the model-to-model data is a little less clear. The average listing price for a CR-V is hovering around $35,890 according to the data, compared to just $28,150 for the Honda Civic.
To Honda’s credit, the average listing price of the Civic and CR-V only increased by roughly 21% and 20%, respectively, from January 2019 to June 2023, which is lower than the average price increase of 27% for trucks over the same period. With more supplies, Honda’s sales across both Civic and CR-V are up significantly.
The Limits Of The Data And Future Work
As I’ve mentioned a few times before, I’m certain that trimflation–the prioritization of higher margin vehicles in supply-constrained environment–is a real thing that happened. The data I was able to get from CarEdge and analyze seems to show that, at least with trucks, this likely contributed to the increase in the average transaction price increasing during the pandemic.
This was just a limited first attempt at trying to come up with a name for what happened, which I haven’t seen, and try to demonstrate its existence. I invite anyone with access to similar (or better) data to please look into trim level availability and try to get a more detailed understanding of production changes.
The biggest shortfall in the data is the assumption that demand stayed static between 2019 and 2023, meaning that a vehicle stayed listed for the same amount of time across the period. For example, 100 trucks listed for 30 days is the equivalent of 50 trucks listed for 15 days.
While I assume demand was pretty much maxed out for trucks from the summer of 2020 until early 2023 (i.e., few vehicles seemed to stay on the lot for very long), the larger number of listings in 2019 and early 2020 could be partially attributable to softer demand as opposed to larger supply (overall sales volume was down slightly in 2019 versus 2018, for example).
How This Might Come To Hurt Automakers
With the number of vehicles increasing, the changing rate environment, and slightly less economic uncertainty it’ll be interesting to see how quickly this changes and how many of these price increases are essentially permanent.
If the listings data is correct, then both Ford and GM are trying to reverse this trend, and listings continue to increase as production begins to rebound from the chip shortage (truck sales are up year-over-year, so it’s unlikely to be only a demand issue). And, as mentioned, Honda also brought back the entry level trim options for both the Civic and CR-V, theoretically slashing the base price of the vehicle. Still, none of the listing data for Ford, Ram, Honda, or GM show the vehicles or the prices back to pre-pandemic levels.
One of the reasons why automakers were able to get away with this without, in theory, losing a ton of demand is that unusually low interest rates and buyers getting used to these increasingly long loan terms made it possible for consumers to justify spending more on a car by keeping their monthly payments comparable to their expectations. Additionally, government intervention meant that there weren’t Great Recession-style job losses and many households saw an increase in disposable income.
This doesn’t work when interest rates go up and now, according to J.D. Power and GlobalData, the average interest rate for a new vehicle loan is 7.1%, or 180 basis points higher than July of last year. In a higher interest rate environment, automakers may need more lower-trim level models to be competitive and attract buyers who were on the sidelines during the pandemic but now want or need a new car.
Do you want a counter example? Look at Tesla Motors. The company, which built its reputation on the back of the expensive Model S, did all it could to produce as many of its more affordable Model Y and Model 3 offerings during the pandemic. This led to a drop in margins from a relatively high 25% in Q2 2022 to a lower (but still enviable) 18.2% in the Q2 2023, per TechCrunch. In the face of a lot more competition, Tesla has been able to hold on to a lot of its market share and remains the biggest builder of electric cars on the globe.
With the pandemic-induced supply disruptions decreasing and used car prices remaining high, there’s a real opportunity for affordable vehicles with nicely appointed lower trim models. A good example is the 2024 Chevy Trax, a vehicle that’s both reasonably priced (it starts at $21,495) and is also quite good.
The alternative is that the United States car market begins to look more like Europe and another player, perhaps even Chinese brands, start to poach customers who just need a good car and a decent price. If the mainstream automakers don’t alter their pricing strategy and keep focusing on pumping out newer, more expensive EVs, then the market could be primed for a company like Dacia or, even, a BYD, to suddenly gain traction.
Images: Stellantis, Ford, General Motors, Honda, Topshot by Jason Torchinsky, Graphs Jason Torchinsky/Matt Hardigree
Our current economy reminds me of the “Bust out” Episode of the Sopranos, where Tony takes over the sporting goods store and keeps squeezing the owner.
It reminds me of post WW2, when cars were scarce and came heavily optioned until the early 50s; take them or leave them. Seller’s market to buyer’s.
I posted this a few weeks ago, but I was low-key interested in trading my JSW in for an id4 over the past 6-9 months. Despite at last looking over 600 vehicles in inventory within 200 miles, there are zero of the *four* cheapest trim levels. Not going to buy a car that I can’t at least see an example of in person so I continue to drive my JSW
Just a fantastic article. Well researched and thoughtful. Comments are too. You’ve created a great site that keeps me coming back every day. I guess I have to become a member. Plus your shirts are cool!
The answer to the first sentence in the article is simple: When you have high inflation, prices overall aren’t magically going to go back down. Once the new higher prices are in place, they’re generally not going to fall. This is true with bread, eggs, clothes, soap, tires… cars.
Interestingly, eggs went up to $8/dozen last year due to bird flu and possibly transportation issues but came back down to a normal ~$2. I don’t know if food is more rigid than automobiles, but demand drops dramatically if you increase the price too much. Auto companies just keep increasing as if demand is related to inflation, but it’s probably an inverse relationship (we can’t afford if they keep increasing price because our money is worth less).
Part of that is because consumers don’t need to rely solely on Big Egg corporations. My city has more and more people growing their own chickens, leading to lower demand for commercial eggs. Car companies have quite the monopoly when they all agree to price fixing through only selling more expensive cars.
>when automakers started delivering cars without heated seats in the middle of winter
Heated seats have only been an option in the majority of cars for about 10* years. The fact that people really got upset over this proves how entitled tech has made us.
*I looked at the Hyundai Elantra, the Chevy Silverado, and the Ford F-150. Elantra introduced heated seats as an option for the mid-plus trim in 2013. The GMC Sierra HD models first got heated seats in 2017 and the regular Silverado got them in 2019. The F-150 got them in 2009 starting with the Lariat trim (Trim 5 of 7).
I’m not sure it’s fair to call people entitled for being upset that prices increased alongside content decreasing.
Heated seats have been available on all those models long before the dates you mentioned. Hyundai first offered them on the Elantra starting in 2007, the pickups before then – at least 2003 for the F-150 for example, GM earlier. Even the Ford Focus offered them in 2003, and most mainstream larger/family sedans had started by that point on upper-trim models.
While I’d agree they have become more commonplace over the last 10 years in particular, it’s from trickling down to lower and lower trims. In the chip shortage, they were deleted from higher trims even. If you bought a vehicle 20 years ago that had them, it’s not unreasonable to expect the new car to still have it.
I bought a vehicle 30 years ago with heated seats–Volvo 740 wagon.
One thing that you don’t really touch on is the role that the dealers played in the price increases. At one point around here you couldn’t find a new vehicle with less than a $5k ADP sticker and $10k or more was not uncommon on desirable models. I think because of that those list prices aren’t neccesarily reflective of mfg price increases. On the other hand it probably understates the real market prices since a lot of the vehicles with the largest ADP stickers were often listed as “call for price”.
That is not to say the mfgs didn’t act to cause trimflation. Ford for example put the Maverick XL Hybrid on Customer Order only status at launch, and the Lightning Pro required a Ford Fleet account number to be processed. I believe they also put other XL models on Customer Order only status.
Many of those custom order which also means a lot of fleet vehicles never show up as listings on the internet and true pricing is hard to tell since we don’t know if the deal was done at MSRP or not.
This is a fantastic discussion, but an awful lot of variables to analyze without trustworthy data. What we can understand from this is that the pandemic was a once-in-a-generation or black swan event (an unpredictable event that seemed extraordinary at the time, but in retrospect appeared inevitable). The business school case studies and economic papers that this will generate will be going for years to come. The thing I find the most extraordinary is that for the first time ever, even in a market swimming with production overcapacity, absolutely nobody blinked on incentives, and everyone was basically able to sell everything they made at escalating transaction prices, even with the dealerships doing stuff that was beyond stupid. That to me is more remarkable than anything.
Well a lot of the “haves” saw their disposable income increase during the pandemic when they stopped going to resturants, bars, movies, gyms, salons, paying for child care, trips, ect.
In a number of cases those workers that were now out of a job were actually making more money thanks to the flat rate bonuses on unemployment benefits. My friend’s kid who worked at a resturant part time while was going to college was actually bringing in ~$1k more per month while on unemployment.
We also saw a return of the house as a cash machine as interest rates dropped. People decided to take some cash out of their home since the payment wouldn’t increase.
Bottom line lots of people had cash burning a hole in their pocket.
Dealershits. Stealerships. Subterranean lizards in pinstripe suits. Whatever you want to call them. Not once in your article do you address the very real problem of the market demand which is created solely by dealerships, and which the manufacturers only work to satisfy.
Drivers don’t drive the market because, with 72+ month loan terms, they’re not the ones paying the manufacturers, and with prices skyrocketing like this, those terms are only going to get longer. They’re doing their goddamndest to push drivers to a rent/lease model and then rags like this and the Old Site will have nothing to talk about except old stuff because every PHEV and BEV will look identical except for the shameful tailpipe.
From Q3 2021, in earnings calls, corporations have been quite open about raising prices because they can. It’s more than 60% profit taking.
good article, thanks Matt.
OEM strategies to transition to EV rely on ICE profits. if trimflation outpaces what the market can afford, those strategies could fail.
yikes.
I’m inherently skeptical of any OEM (outside of electric only startups) crying poor on their accounting of EV profitability. Not saying it’s entirely fabricated, but I take any “we lose $Xk per vehicle sold” datapoint with a large helping of salt, as well as the general notion of “we have to sell [giant SUV] to fund the transition”. I’m sure there’s some truth in it, but I fully expect they’ll claim and inch and take a yard here
It’s been extremely obvious just looking anecdotally at inventory the past couple years, but I definitely appreciate the deep dive investigative journalism, Matt. Great job. Even now, it’s still bad. I’ve been looking at WL Grand Cherokees for the past couple of years. Most dealerships only have one or two Laredo (base trim) models on the lot (~$40K). A lot of them have none at all. But they’re stuffed to the gills with $50K Overland or $65K (fucking LOL) Summit trims that appear to be sitting on the lot. Shocker.
It’s a different game now, guys. There’s no more stimulus cash or low interest rates to buoy sales amidst the absurd prices. Used car values are dropping, making it increasingly unlikely that someone can finance a new car for a reasonable amount with what they’ll get for their trade (they may even be underwater if they bought at the peak). A lot of the FOMO buyers (admittedly some people had no choice to buy) from the pandemic era are tapped out. What’s left are financially conservative buyers who will continue to sit on the sidelines patching up their beaters until you actually start to give them more selection and more reasonable prices. I count myself among them. Stellantis has already blinked and started offering incentives to move metal. It’s only a matter of time before the others follow, even if they go kicking and screaming. Boo hoo, how sad.
If you have access to this data, could you also generate 3 graphs for the trucks that show the daily listings for each available trim? A little context seeing that the XL and Lariat availability went down while King ranch and Platinum went up says a lot more than just showing the XL, when in fact all 4 dropped an equivalent amount. (not a truck guy, I think they’re all from the same model?) For extra credit maybe 3 more graphs that show the asking price for each trim during that period as well?
I’ll try to add some details about the chip shortage.
Automotive manufacturers use components that are years and maybe decades behind the cutting edge of chip manufacturing. As you said, “Automakers also use older chips for a reason: they’re proven and they’re safe”. I’ll add that it’s expensive to redesign and certify new components so they stay with those older parts as long as they can.
These older components are build in older chip fabs. They actually can’t be built at the newer fabs. These fabrication lines were often the high end cutting edge stuff long ago. As newer higher tech fabs with better capabilities come on line the market for the cutting edge chips moves to the new fabs and the existing ones fall back into being older tech.
> few chipmakers were interested in making lower profit-margin chips and essentially shuffled automakers to the back of the line. They had PS5s chips to make instead and the money was far better.
Isn’t quite right, in the vast majority of cases automakers aren’t competing with game consoles for production. But I’ll touch back to this a couple times
Fabs make chips in big batches. Everyone trust each others forecasts so when a manufacturer orders X a month the fab might make 12 * X once a year and warehouse the rest. So now there are some components where there are a bunch in a warehouse and others that are about to come up for for that yearly build.
Auto makers reduce their orders from the fabs. Fabs make less ahead and reduce their orders from suppliers for things like silicon wafers. Things go for a while before everyone figures out they actually want increased production. They go to make these orders and now the fab has to spin back up. But now they actually need to build more than normal in order to make up those stock piles and support these new orders coming in. They need to get their supply chain running again and that takes a while. Plus those silicon wafers they need could actually be in contention with game console manufacturing.
Another problem is that these automotive component plants were close to peak production before the slow down. They get back to full production but they still aren’t able to meet the short term demand.
Chip fabs take years to build and cost billions of dollars, even fabs for older components. Everyone figures that once the automakers get caught up they’ll likely go back to something resembling the previous order levels. It would be a huge cost to set up a new fab to cover this demand and since every one knows the higher demand won’t last no one is willing to do it.
The chip fabs have more orders than they can fill so they have to choose which parts to make and which orders to fill first. They prioritize the parts at the higher end of their offerings. Parts come in families so the high end version is usually a drop-in replacement, or maybe minor change, to use in place of a lower end part. It’s a trade off, the high end part uses a bit more silicon and more time in the fab so they make less total chips but anyone using the family can use what they make. It’s just a happy accident for them that the higher profit on the high-end parts is bigger than the reduced number of sales.
The automakers have already designed in the lower end parts.mNow they have 3 options:
wait for fab capacity to catch up to get the part they actually want. With this choice they will have to wait longer than people who make the other choices.suck it up to use the higher end part. Costs more but gets them parts quicker. It also extends the shortage for everyone, because the fab is making less total parts, see above.Do a big redesign to use alternatives that are more widely available, then certify the updated part. This costs a lot
Then while all this is going on an AKM factory burns. They make DACs and ADCs, the bits where electronics interface with they world, and this could be another place where automotive and PS-5 compete for the same part.
[Edited to add] Less sexy and made less news, AKM made something like 50% of the worlds crystal oscillators. This might actually be the biggest stick in the wheel for electronics manufacturing in general.
Than a Renesas fab burns. Wikipedia says that at the time they made 30% of the electronic components that go in to autos.
This same pattern played out at multiple levels. I was in electronics, but not autos, while this is going on. For one part the vendor went from a typical 8-12 week lead time (order to delivery) to a 55+ week lead time for a more expensive part.
Cluster fuck is an understatement
Are you sure it wasn’t just Goldman Sachs and theor Wall Street buddies up to no good again?
https://macleans.ca/economy/business/goldman-sachs-aluminum-and-the-real-reason-why-wall-street-has-been-hoarding-metal
If so somewhere there’s warehouses FULL of chips collecting rent and dust.
I’ve never been directly on the purchasing side. What I know is because indirect complaining from the people who are.
I haven’t heard of hoarding to drive up prices. Nor of hoarding buys during a chips normal production.
The last manufacturing run (end of life) of a chip is announced well ahead of time. This gives companies an opportunity to buy what they need for the rest of time (lifetime buy)
I don’t think it’s the same thing but
there is a market for chips after end of life. For whatever reason people find they need more than they have. There are some third parties that will do a large buy at the end of life to take advantage of this market and as their supply dwindles the prices can get pretty high.
If you’re trying to purchase at this point you really have to start worrying about counterfeit chips and scam offers. The lesson to learn here is don’t mess up your lifetime buy.
One other attribute is that the smaller the process, the less tolerant the chip is to virtually EVERYTHING that happens in the real world. Most chips used for automotive target a 90-28nm process instead of the 3-7nm we’ve been using for the last 5 years as they get far more tolerant to voltage spikes, voltage sags, static, radiation, cosmic rays, heat, cold, hard shocks,… In other words they’re built with the same class of machines TSMC, Samsung, and Intel sold during their yearly yard sale 10 years ago. To add to this, if you need capacity, it makes far more sense to build a 3nm fab for 30B that already has a 300B backlog of chips it could manufacture in the next 2 years, instead of a 28nm fab for 5B, and over the next 10 years sell about 5B of chips. Another factor is if you’re switching from 5nm to 3nm, you generally can’t replace your machines as the building is the factory, so you generally build it right next to the 5nm one and keep cranking out cheaper 5nm low end chips for as long as its profitable, then sell it off / demolish it.
Another thing to add to this is the standard lead time for TSMC went from 6 months to over a year from submitting a design to getting the first wafer off the line. About 3 months of this simply couldn’t be accelerated even if they were the only customer and the entire company focused on this project.
100% agree with everything you say.
Great stuff Matt.
The simple reason they raised prices is this. They could.
One word? Greed. Fuck them.
During the pandemic, I remember hearing reporting on Autoline Daily that brought up how Ford and GM found they could increase profits while selling fewer vehicles, but hindsight, it was obvious given low interest rates.
Dealerships got put in a tough position because with fewer new cars to sell, they kept the lights on by buying/selling used cars and adding market adjustments to the cars they were getting
It’s no surprise Tesla did very well, because they got to keep every dollar from their price increases and from a consumer standpoint, ordering one is a lot easier than trying to run the dealership gauntlet
The fact is, publicly traded corporations have worshiped at the feet of the Quarterly Gods for so many years that they no longer have a choice. Margins can not drop. Growth must continue. Otherwise, stock prices plummet and companies either need to issue new stock or acquire new debt. So when faced with a shortage of material, OEMs made the (unfortunately) rational decision to allocate increased production toward higher trim (i.e., higher profit) trims.
By the way, I’m not excusing the OEMs. I work for a Tier 1, and it gets old having your business suffer because of poor decision-making at the top of the supply/food chain.
Very intriguing and very informative, Matt! I’d be curious to see a follow up to this article, perhaps as a series, once more data becomes available. This serves both to enlighten your readers while simultaneously holding automakers culpable for their decisions.
I agree this was a great article and I’d love to see more of it. But I’m not sure if there is any value in “holding automakers culpable.” Do they actually suffer because of articles like this?
If we’re really honest about it, probably not. That said, it never hurts to inform the public at large and let them vote with their collective wallets. All revolutions start with one shot.
I agree. But I am jealous of your optimism. 🙂
“Do they actually suffer because of articles like this?”
Define “suffer”.
Also whom within the heiarchy. Executives will “suffer” very differently from rank and file.
Ha. Good (but shitty) point. Executives rarely “suffer” and when they do, they get a payout anyway! This is why my life goal is to join their ranks.
No surprises, it is interesting to see data, rather than just speculation, great work.
Eventually things will change again as stockholders start to be hit by lowered profits, they are ultimate driver of these corporations, the real customers.
This write up may have taken longer than usual to get all the data, but it puts The Autopian above others. Great work!
In the end, OEMs will have to have lower prices or a new OEM will enter the market, like you said. 1st it was the Japanese, then the Koreans, next? Most likely China.
Again I only see prices going up do to labor contracts and the Big 3 still needing to please their investors/fund managers
Nicely researched! Way to be the first to print on providing real data around something many have postulated. I can’t say anything in your article surprised me, but I like that, at least to some extent, you had data to back up your conclusions. Trimflation is a fitting term for this phenomenon.
Yes, those $1500 stimmie checks caused all the inflation. Not the MULTIPLE YEARS OF RECORD PROFITS at most corporations. Nope. Couldn’t be that. Gotta be those fat cats that are living large off $1500.
Buying groceries. Paying back due rent. Just living LARGE, my man.
Yeah right? I used my covid bucks to buy a house, three cars, and a new wardrobe. And I still have enough left to go to the movies!
Those profits come from customers willing and able to pay more for essentially the same thing. Customers ARE willing and able to pay more when they get a windfall. So yeah, it fits those $1500 checks played some part in this.
Low interest rates and longer loan terms as well but we had those long before the pandemic hit.
I keep having a chuckle about this. One rent payment on a reasonable apartment in a medium-to-large city. Yup. That definitely had something to do with pickups costing $14,000 more each in three years’ time when they were trending downward.
Corporations have managed to convince us that poor people responsible for higher prices. It’s a hell of a trick.
Every vehicle sold with bullshit fees, adjusted pricing, markups or whatever had a buyer willing and able to pay that higher price. Some of those customers actually needed a new car but many of those customers didn’t NEED a new car, they just WANTED a new car. Those customers could have refused to pay more, they could have waited till inventory recovered but bought the new car anyway despite the shenanigans, driving up prices for everyone.
Those customers do deserve their share of the blame. Painful depreciation will be their punishment.
I’m not arguing that corporate profit seeking didn’t have a significant effect on inflation, but you’re focusing on the wrong inflationary effect of the stimulus bills. Because they were paid for with deficit spending, they resulted in two massive increases in the money supply. Anyone who’s taken a monetary policy course knew inflation was right around the corner as soon as the bills were passed.
Of course, exclusively building high-trim models only works for so long, and then you run out of customers. They’re already learning this with EVs- they’ve exhausted the supply of early-adopters.
Cars – and similarly, rent – have increased far faster than wages. We are nearing a point where more people who would initially consider replacing a car with expensive maintenance issues will be more inclined to repair it. For example, when I was younger needing to replace a transmission meant time for a new car for tons of folks. But that decision isn’t so automatic anymore.
I can confirm – talking things over with the manager at my local shop they’re seeing people spend thousands to fix issues on older vehicles that would have previously been sold or junked out. People can’t afford the new car payments due to higher costs & interest rates so they’re racking up the credit card to keep the old car going.
Fear not, you can now “pay over time” with dedicated credit/financing for auto services and repairs…
And, even deeper, the latest iterations of the high-mileage car warranty scams are proliferating. I didn’t research enough to find out if the new companies are just new forms of the old ones, but the complaints sounded exactly the same as they did last I looked 5 years ago—even though they were new names.
Hey, there’s an idea for an article —if any writers here have low enough blood pressure to jump through the hoops. SWG, maybe? Warranty the SwampAvenue??
Over on YouTube Car Wizard has pretty much said this–customers of his are investing well into four figures and opting to repair 20 year old cars that previously would have been considered ready for the junk heap.
I’m an example of this. I’m currently driving a truck that has had multiple four figure repairs over the past ~year that I probably would have replaced by now, except new ones are so ridiculously expensive.
Nice reporting. It is useful to put numbers to this. The steady drumbeat of price increases on the entry level trucks also beggars belief when it comes to hard supply side issues. I bet it reflects a successful acculturation of people to higher car prices via shittier loan terms, but I definitely lack the data to demonstrate that. I also can’t imagine auto manufacturers don’t look at ADMs and see that as a sign that there is yet more pie to grab.
Very well-researched journalism, although the conclusions were not surprising to anyone – automakers screwed up their own pipelines, raised prices to what the market would bear, aren’t anxious to lower them back down, and might be doing long-term damage to their brands and customer relationships.
My unresearched anecdotal take is that prices rise on the news and fall on the competition. When a big storm is coming, gas prices spike everywhere in its path. Even if it doesn’t hit, the prices only ratchet down slowly as local stations undercut each other by a penny at a time.
It’s called “rockets and feathers”: https://en.wikipedia.org/wiki/Asymmetric_price_transmission