The Kia Optima Got Every Safety Feature Congress Mandated. Its Body Count Tripled.

Electronic stability control has been mandatory on every new car sold in the United States since the 2012 model year, and nobody disputes that it works: IIHS estimates the technology prevents one in three single-vehicle fatal crashes, while NHTSA's own final rule projected between 5,300 and 9,600 lives saved annually once the full passenger fleet turned over.^[1][2] We ran the FARS data expecting a clean confirmation of that story, with model-year deaths declining steadily after the 2012 mandate took effect. For most vehicles, that's what we found.

Fifteen didn't. We compared average deaths per model year in the pre-ESC window (2005–2011) to the post-mandate window (2013–2019), filtering for vehicles with at least 50 pre-mandate deaths and five or more model years in each bracket, and what emerged was a set of nameplates where the body count climbed after they received every safety feature Congress demanded. Kia's Optima averaged 14.3 deaths per model year before the mandate and 60.4 after it, a 323% increase that dwarfs any statistical noise. Chevrolet's Equinox jumped from 33.0 to 77.7, the RAV4 doubled from 28.0 to 58.3, and Hyundai's Elantra climbed from 98.9 to 161.7.^[3]

Every one of those vehicles has ESC, front and side airbags, LATCH anchors, and backup cameras, and several earned five-star NHTSA ratings alongside IIHS Top Safety Pick designations, which means they are genuinely safer on a per-mile basis than the predecessors they replaced. A 2017 Optima is a better car to crash in than a 2007 Optima by every engineering metric NHTSA measures.

More people bought them, and the sales data explains why the body count rose: Kia sold roughly 60,000 Optimas annually during the pre-mandate window, but by 2017 that figure exceeded 130,000, Equinox sales tripled from approximately 60,000 to over 330,000, and RAV4 sales nearly tripled from 150,000 to 420,000, a combined surge in registered vehicles that overwhelmed whatever per-mile safety gains ESC delivered.^[4] The mandate worked for you; it didn't work for the model.

That distinction matters, because NHTSA sells safety mandates in population-level language like “5,300 to 9,600 lives saved annually,” and that framing hides what happens inside a single nameplate whose sales are surging while its raw death count climbs even as the per-mile risk falls. For vehicles where sales collapsed, the numbers look spectacular: Chevy Tahoe's model-year deaths dropped 73% across the same window from 81.0 to 22.0, Ford Expedition fell 72%, and Chevy Impala fell 70%, declines that owe as much to the disappearing market as to the mandate itself.^[3]

There is another confound in the dataset that deserves explicit disclosure. A 2019 model year has only four to five years of crash exposure in the FARS 2014–2023 window, while a 2005 model year has eight or nine, and older cohorts also lose vehicles to scrappage, reducing their exposure denominator over time, which collectively pushes post-mandate death counts down and pre-mandate counts up. The vehicles that still showed increases were fighting that structural headwind the entire time, and Elantra's 64% increase happened despite the data being biased against producing one.

None of this means ESC is theater, because it is possibly the most effective single safety intervention since the three-point seatbelt, but the regulatory framework measures success exclusively in per-mile fatality rates and rulemaking projections rather than in the absolute body count attached to a specific model that people actually buy and drive and crash on real roads.^[1] When Kia sells twice as many Optimas, the per-mile rate drops and NHTSA's spreadsheet shows a win. Families of the 60.4 people who die per model year are not on that spreadsheet, but they were on the previous one, when the number was 14.3.

What you can do: Check whether your vehicle's model year falls in the pre- or post-ESC window. If you're driving a 2005–2011 model without factory ESC, your single-vehicle crash risk is roughly 33% higher than an equivalent post-2012 model with the system standard. Verify your VIN at nhtsa.gov/recalls for any open safety campaigns, and if you're shopping for a used car, a 2013+ model year with ESC isn't a guarantee of anything, but it is one of the few interventions supported by a genuine body of evidence.

Methodology note: Deaths per model year are raw counts from FARS 2014–2023, not rates, and they reflect total fatal crashes involving that vehicle’s model year, which correlates with both sales volume and fleet age, meaning a vehicle with rising model-year death counts may still have a declining per-mile fatality rate if sales growth outpaced the rate reduction. We state this limitation explicitly because it is the strongest counterargument to the headline finding; see methodology for additional caveats.