⌘K
HomeStudiesUnrepaired recalls
Study · 2026-09-14

The recall repairs that never happened

Across every recall campaign whose federal reporting window has closed, an estimated 42,633,536 vehicles are still carrying a repair nobody did.

Still unrepairedestimated, closed campaigns42,633,536
Campaigns measuredreporting window closed742
Typical campaignmedian repaired87%
Vehicles coveredby those campaigns190,856,270

How this was measured

Source: 49 CFR 573.7

Manufacturers must report cumulative repair counts to NHTSA every quarter for six consecutive quarters, starting with the quarter owners were told a remedy was available. Then the obligation ends and the number stops moving.

That clock is why most published versions of this number are wrong. A recall filed last month sits at 3% because the parts shipped last week, not because anyone ignored it — so a “least repaired” list that does not control for age is topped by whatever was filed most recently, every time. Every campaign counted here has had its six quarters and closed.

Campaigns under 25,000 vehicles are excluded, because below that one dealership's paperwork moves the percentage. “Estimated unrepaired” is the potentially-affected count the manufacturer filed, multiplied by the share it never reported repaired. It is an estimate of vehicles, not of cars still on the road: some were scrapped, exported or written off, and NHTSA does not publish that.

What this is not

Not a ranking of how dangerous a vehicle is. A low completion rate usually says more about how hard owners were to reach — older cars, several owners deep, a repair that needs a dealer visit for a fault nobody can feel — than about the defect.

And not a verdict on one car

Whether any specific vehicle has an outstanding repair is settled by VIN at nhtsa.gov/recalls, free, in about ten seconds.

The 25 lowest, once the clock has run

Source: NHTSA
RepairedCampaignManufacturerComponentVehiclesEst. unrepaired
14.1%24V896000Chrysler (FCA US, LLC)Service Brakes, Hydraulic317,630272,908
20.8%24V031000Ford Motor CompanyStructure1,889,1101,495,419
24.8%22V420000Nissan North America, IncLatches/Locks/Linkages322,671242,713
28.6%24V834000Volkswagen Group of America, IncAir Bags248,330177,183
29.6%22V703000Kia America, IncTrailer Hitches70,88749,940
31.1%24V198000Chrysler (FCA US, LLC)…Air Bags284,982196,467
32%23V775000Ford Motor CompanyLatches/Locks/Linkages45,17330,700
32.4%23V652000Kia America, IncService Brakes, Hydraulic1,730,1921,170,129
32.5%24V146000Hyundai Motor AmericaLatches/Locks/Linkages186,588125,854
32.7%22V656000Porsche Cars North America, IncExterior Lighting222,858149,894
33.1%23V093000Nissan North America, IncElectrical System712,458476,563
33.8%23V502000Ford Motor CompanyLatches/Locks/Linkages112,22574,293
36.7%22V723000Chrysler (FCA US, LLC)Exterior Lighting83,45352,801
37.5%22V082000Ford Motor CompanyBack Over Prevention330,784206,905
38.3%19V446000Porsche Cars North America, IncPower Train99,66561,453
38.8%24V608000BMW of North America, LLCEngine And Engine Cooling720,796441,127
38.8%22V137000General Motors, LLCExterior Lighting740,581453,162
38.9%22V192000Ford Motor CompanyVisibility39,72024,269
40.7%23V734000Toyota Motor Engineering & Manufac…Electrical System1,853,5681,098,610
41.6%23V487000Mazda North American OperationsBack Over Prevention236,100137,788
41.9%25V091000Ford Motor CompanyVisibility35,32820,536
44.4%19V392000Jaguar Land Rover North America, L…Latches/Locks/Linkages65,38536,367
44.5%23V338000Chrysler (FCA US, LLC)Electrical System132,09973,315
44.7%23V162000Ford Motor CompanyService Brakes, Hydraulic1,280,726708,626
46.1%23V594000Kia America, IncLatches/Locks/Linkages319,436172,048

Every campaign number links to its own page, with the manufacturer's filed defect, risk and remedy text in full.

Advertisement

By manufacturer

Source: NHTSA

Mean completion across each manufacturer's closed campaigns, for the 16 manufacturers with at least 15 of them. Read it as a measure of how reachable a maker's owners are — fleet-heavy and older-fleet brands sit lower, and a manufacturer with a single enormous hard-to-reach campaign is dragged down by it.

ManufacturerMean repairedCampaignsEst. unrepaired
Kia Motors America74.9%17759,476
Nissan North America, Inc75.9%364,278,353
Kia America, Inc76.5%271,814,736
Ford Motor Company77.2%12611,442,404
Chrysler (FCA US, LLC)80.7%532,158,190
Mazda North American Operations81.1%16600,651
Chrysler (FCA US LLC)82.5%554,529,637
General Motors, LLC83.6%292,013,575
Volkswagen Group of America, Inc85.5%571,729,580
Toyota Motor Engineering & Manufacturing85.8%483,844,631
Hyundai Motor America86.1%421,005,853
General Motors LLC87%311,834,209
Honda (American Honda Motor Co.)88.1%493,390,863
Mercedes-Benz USA, LLC89.2%29492,660
Subaru of America, Inc90.4%23615,910
Tesla, Inc97.7%24123,967

By what was wrong

Source: NHTSA

The same figure grouped by the component NHTSA filed the campaign under. What sits at the bottom is the more interesting half: a fault an owner can feel gets fixed, and one they cannot does not.

ComponentMean repairedCampaigns
Latches/Locks/Linkages69.4%35
Air Bags76.9%71
Exterior Lighting78.5%32
Structure79.9%34
Power Train81.2%72
Suspension81.8%15
Service Brakes, Hydraulic82.7%44
Visibility83.8%43
Seat Belts84.1%37
Electrical System85.6%99
Fuel System, Gasoline86%45
Back Over Prevention86.3%51
Engine And Engine Cooling86.5%32
Steering86.7%33
Equipment87.1%15

The data behind this

Everything above, as files. No key, no form, no attribution beyond the line below — if it is useful, use it.

How to cite

WhichTrim, “The recall repairs that never happened”, 2026-09-14. https://whichtrim.com/studies/unrepaired-recalls/

The underlying records are U.S. federal public-domain data from NHTSA. The joins, the filtering and the arithmetic are ours, and we would rather be credited than not — but the figures are checkable from the source either way, which is the point.

Questions about this study

Source: NHTSA and 49 CFR 573.7

Does a low completion rate mean the car is dangerous?

No, and the correlation is weaker than it looks. A campaign's completion rate mostly measures how reachable the owners were — an eight-year-old car three owners deep, with a fault nobody can feel, needing a dealer appointment. The severity of the defect is a separate field, on each campaign's own page.

Why exclude recent recalls?

Because including them would make the study wrong. Repair counts climb for six quarters after owners are notified and then stop being reported. A recent campaign's low number is the parts pipeline, not neglect. Every campaign counted here has had its six quarters.

Are these cars still on the road?

Some are not. The figure is the manufacturer's potentially-affected count minus what it reported repaired, and vehicles get scrapped, exported and written off without anyone telling NHTSA. Treat it as vehicles never recorded as fixed, which is what the federal record actually says.

Is the recall repair free?

Yes. Federal law requires the manufacturer to remedy a safety defect at no charge for vehicles up to 15 years old, measured from first sale. A dealer charging for a recall repair is in the wrong.

Can I use these numbers?

Please do. The full table is on this page as CSV and JSON, the underlying records are federal public-domain data, and every campaign links to its own page so you can check us against NHTSA directly.