← The Crash Report
Existential Dread

Every Car Has a Death Invoice. We Calculated Yours.

Ghostly receipt superimposed over a parking lot of cars

Avian Law Group dropped a report this week calculating that motor vehicle crashes cost American society $1.77 trillion per year.[1] That figure includes medical bills, lost wages, property damage, legal fees, emergency services, and the DOT-assigned value of 39,254 human lives erased in 2024. A staggering number, and a fundamentally useless one.

Useless because it treats the American vehicle fleet as a single organism, when in reality some models are subsidized by death and others barely register. So we did something nobody seems to have tried before: we took FARS fatality data for every vehicle model over ten years, multiplied by the Department of Transportation's Value of Statistical Life ($12.5 million per death), and divided by estimated fleet size to produce a per-vehicle societal cost for each model in America.[2][3]

Your car has a death invoice. It is almost certainly higher than you think.

$122,286
Societal cost per Chevrolet Tracker in fleet, 2014–2023

At the top of the invoice sits the Chevrolet Tracker at $122,286 per unit, a discontinued compact SUV that most people forgot existed. It didn't forget them. 856 deaths across an estimated 87,500-vehicle fleet over the FARS decade means each surviving Tracker on the road has externalized more cost to society than it ever cost to buy.[2] Behind it: the Hyundai Veloster at $85,429 per unit (598 deaths, 87,500 fleet), the Chevrolet S-10 at $81,543 (1,427 deaths, 218,750 fleet), and the Nissan Maxima at $73,524 (1,544 deaths, 262,500 fleet).

Now look at the bottom of the list. A Ram 1500 costs society $2,125 per vehicle, a Toyota RAV4 costs $3,037, and a Subaru Forester runs $4,041. That is a 57:1 spread between the Tracker and the Ram, and around 40:1 between the Tracker and the RAV4, which means two vehicles sharing the same road and the same traffic laws impose wildly different burdens on the civilization around them.[2]

Class averages expose a paradox that should bother anyone who thinks sedans are the responsible choice. Sports cars top out at $26,880 per vehicle, which surprises no one. But sedans come in second at $20,016, beating pickups ($14,948), vans ($9,642), and SUVs ($8,545). The vehicle category Americans consider "normal" externalizes more than double the societal cost of the category Americans consider "excessive." Weight, crumple zone volume, and ride height combine into a physics package that protects occupants regardless of reputation, while the compact, efficient sedan crumples around its occupants in ways the marketing brochures never mention.[4]

Scale the individual invoices to total fleet burden and the numbers get genuinely obscene. The Honda Accord alone accounts for $88.8 billion in societal fatality cost across 7,102 deaths over the decade. That is larger than the GDP of Luxembourg. Not because the Accord is particularly dangerous per unit ($44,112, mid-table) but because there are 2 million of them, each accumulating miles and statistical probability, and the sheer volume of Accords on American roads turns a moderate per-unit risk into a national-scale externality.[2]

The Math, Exposed

Societal cost per vehicle = (FARS deaths for model, 2014-2023) × ($12.5M VSL) ÷ (estimated fleet size). Fleet estimates are derived from cumulative sales data adjusted for scrappage rates, not registration counts, which introduces roughly ±15% uncertainty on low-volume models. The DOT's VSL of $12.5 million follows their 2023 Benefit-Cost Analysis guidance and captures only the statistical cost of lost life; it does not include medical costs, property damage, or lost productivity, meaning our figures undercount the true burden by a factor of roughly 7x compared to the Avian Law Group's comprehensive methodology.[1][3]

Limitations

FARS captures only fatal crashes, which represent roughly 0.5% of the 6.7 million annual U.S. crashes.[5] A vehicle with low fatality costs might still have catastrophic injury costs. Fleet size estimates for discontinued, low-volume models like the Tracker carry proportionally higher statistical noise, and we acknowledge that the Tracker's position at the top of this list is partly an artifact of its tiny denominator. We also attribute all fatality cost to the vehicle model, which ignores the human behind the wheel. The strongest counterargument against this entire exercise: most fatal crashes are caused by driver behavior, not vehicle design. The car is a vessel, not a cause.

But here is where that counterargument cracks: if you replaced every Tracker in America with a RAV4, the data says fleet fatalities would drop by roughly 96% for that segment. Same roads, same drivers, same reckless left turns and same three-beer decisions, but a different vehicle producing a radically different outcome. The vessel shapes the consequences even when the human creates the conditions.

What You Should Actually Do

If you are shopping for a car, cross-reference any model against NHTSA's FARS data before you sign.[2] The RAV4, Forester, Traverse, and Pilot all sit below $5,000 in per-unit societal cost. The Accord, Civic, Altima, and Maxima all sit above $40,000. That gap is not noise; it is physics. Check your current vehicle's recall status at nhtsa.gov/recalls while you're at it. And if you drive anything from the top ten of this list, maybe reconsider whether the attachment is worth the externality.

Sources & References

  1. Avian Law Group, “The $1.77 Trillion Wreck: The True Cost of Car Crashes in America,” June 5, 2026. avianlawgroup.com
  2. NHTSA, Fatality Analysis Reporting System (FARS), 2014–2023. nhtsa.gov
  3. U.S. Department of Transportation, Benefit-Cost Analysis Guidance: Value of a Statistical Life, 2023. transportation.gov
  4. IIHS, Vehicle Size and Weight. iihs.org
  5. National Safety Council, Motor Vehicle Deaths Preliminary Estimates, 2025. injuryfacts.nsc.org

Source: NHTSA FARS 2014–2023, DOT VSL $12.5M (2023 guidance). Fleet estimates use sales-adjusted scrappage models with ±15% uncertainty on low-volume vehicles. See methodology for caveats.