Your Recall Was Completed Eight Years Ago. Ford Just Admitted It Wasn't.

In October 2018, Ford filed recall 18S32 covering 1,282,596 Ford Focus sedans from model years 2012 through 2018. The canister purge valve could stick open, build vacuum in the fuel system, deform the plastic fuel tank, and stall the engine at highway speed without warning or restart capability. Dealers would reprogram the powertrain control module software. Straightforward. Drive in, flash the computer, drive out fixed.

Last Friday, Ford filed recall 26S40 covering 255,404 of those same Focus vehicles.^[1] Not for a new defect, but for the original one that was supposedly resolved eight years earlier. According to the NHTSA filing, these vehicles "were previously serviced under Recall 18S32" but Ford's records now show "discrepancies" indicating "the correct software fix may not have been successfully applied to all vehicles." They were marked as repaired, logged as complete, and counted in the completion statistics Ford reports to the federal government every quarter. They were not repaired.

That ratio deserves a moment to register. Out of 1,282,596 vehicles in the original campaign, 255,404 registered as fixed but weren't. That is 19.9% of the recalled population showing up as phantom completions in manufacturer reporting. One in five.

NHTSA tracks recall completions using a formula it has published in multiple reports to Congress: vehicles reported as remedied, divided by total vehicles in the recall minus those scrapped, exported, or stolen.^[2] The numerator relies entirely on manufacturer self-reporting. NHTSA does not independently audit whether a dealer actually performed the repair. It does not cross-reference VIN-level service records against manufacturer claims. It counts what manufacturers tell it to count. For recalls initiated between 2012 and 2022, NHTSA reports a weighted average completion rate of 65.8%.^[3] A number that already means 34.2% of recalled vehicles are driving around unrepaired, and that number itself now has an integrity problem, because some fraction of the 65.8% includes vehicles like these 255,404 Focus sedans: logged as safe, still carrying the original defect.

NHTSA processed 881 vehicle recalls covering approximately 30.8 million vehicles in fiscal year 2025 alone.^[4] If even 5% of "completed" repairs across the industry are phantom fixes, and there is no mechanism in place to confirm they are not, that is roughly one million vehicles per year entering the denominator of "repaired" while still carrying the defect that triggered the recall. Over a decade, the cumulative count reaches into the tens of millions. You cannot calculate the actual number because NHTSA does not collect the data required to calculate it.

The Ford Focus is not an obscure vehicle in FARS. In NHTSA's fatality data from 2014 through 2023, the Focus appears in 3,046 occupant deaths across an estimated fleet of 1.05 million vehicles, producing a fatality rate of 2.52 per 100 million VMT.^[5] That rate is more than double the light-vehicle average. Among the 5,017 Focus drivers in fatal crashes during that period, 19.4% tested positive for some form of impairment, which is roughly in line with the sedan class average of 20.4%. The Focus kills at an above-average rate even when its drivers are sober. This is the vehicle whose stalling defect went unrepaired for eight years while Ford reported it as fixed.

The counterargument is straightforward: Ford caught its own error. The company identified the discrepancy in its service records, disclosed it to NHTSA, and filed a new recall to correct the situation. No crashes or injuries have been reported in connection with the re-recall. The system worked, eventually, because Ford audited itself. That argument holds exactly as far as the question it raises: how many manufacturers have not audited themselves? Recall 18S32 ran for nearly eight years before Ford noticed that a fifth of the "completed" repairs were fabrications in its own database. That is not a system catching an error. That is a system where errors persist for the better part of a decade because nobody is required to look.

Limitations are significant and should be stated plainly. We have one data point: one recall at one manufacturer, with a phantom rate of 19.9%. Extrapolating from a single campaign to the industry is speculative. The mechanism of failure, whether dealer software glitches, administrative errors, or something more deliberate, is not disclosed in the filing. FARS fatality data captures only fatal crashes, and the stalling defect is more likely to produce low-speed incidents than fatalities. The connection between the Focus's high fatality rate and this specific defect is correlational, not causal.

If you own a 2012 through 2018 Ford Focus: Check your VIN at nhtsa.gov/recalls. Do not assume that a prior dealer visit resolved the issue. VINs for the re-recall will become searchable on NHTSA.gov on July 6, 2026. Contact Ford at 866-436-7332 or bring the vehicle to any Ford or Lincoln dealer for the PCM update at no charge. If your engine stalls, do not attempt to restart immediately; move to a safe position and call for assistance.