A recall is the system admitting a mistake. This article is about the 847 campaigns that admit the admission needed work: recalls whose own federal text cites an earlier campaign number, plus hundreds more announcing themselves as the remedy for, the expansion of, or the correction to a fix that came before. I mined all 23,060 campaign texts in the register for the evidence, and the verdict runs in one direction: in the 1980s, one recall in two hundred referenced a prior campaign; today it is nearly one in eleven. The fix fails more often than anyone advertises, the failures have exactly five shapes, and the most re-recalled campaigns in America are not the ones you would ever guess.
How the register confesses
The method is simple and deliberately conservative. Every campaign in the register carries three federal texts, the defect, the consequence, and the remedy, so I scanned all of them, across all 23,060 distinct campaigns, for two kinds of confession: an explicit reference to another campaign number, and the phrases that announce a do-over, remedy for, expansion of, incorrectly installed, previously repaired. That yields 847 campaigns citing prior campaigns by number and 1,357 matching the broader test. Both figures are floors. Suppliers of installed equipment file their recalls under a separate numbering series that my number-matching did not count, and plenty of second recalls never mention the first at all. Whatever the true rate of fixes failing, it is higher than what follows.
The rising curve
The share of recalls that reference a prior campaign has risen seventeen-fold since the 1980s, and the honest reading is that two things are climbing at once. Vehicles got more complex, software patches created a genre of fix that can fail invisibly and ship again, and supply chains concentrated, so one bad part now touches more campaigns. But disclosure also improved: the modern register simply writes better paperwork, naming its ancestors where a 1985 filing would have stayed silent. The curve is some mixture of more failure and more honesty about failure, and the register cannot separate the two. What it can say is that for a vehicle recalled in this decade, the odds that the recall is itself connected to a previous recall are about one in eleven, which is not the number anyone imagines when the letter arrives.
The leaderboard nobody expected
Rank the register's campaigns by how many later campaigns cite them and the top of the table is not Takata, not Firestone, not any household disaster. It is motorhome chassis. The most re-recalled recall in America is a 2007 campaign for a ZF steering gear assembled with the wrong recirculating ball spacer on Freightliner motorhome chassis, cited by fourteen later campaigns as R-Vision, Fleetwood, Four Winds, Coachmen, Tiffin, Damon, and half the RV industry filed their own paperwork for the same part. The rest of the board reads the same: Workhorse chassis steering shafts, eleven citations; a Freightliner front axle, twelve; a Bendix brake valve, twelve. The reason is industrial topology, the same one the badge-twins study found in miniature: a recall is a manufacturer's filing, and the RV industry is a handful of chassis builders underneath dozens of coach brands, so one wrong spacer becomes fifteen separate federal events. The register's most-referenced recalls are not the famous ones. They are the plumbing, and Takata's absence from this table is its own finding: its thousands of do-overs spread across so many makers and years that no single campaign number concentrates the citations.
The five shapes of a failed fix
The expansion. The recall that underestimated itself. The canonical case is verified in Ford's own correspondence with the government: the speed control deactivation switch, a cruise-control part that could catch fire, recalled in 1999, again in 2005, twice more in 2006 and 2007, and finally across four million vehicles in a single 2007 wave, each campaign redrawing a circle the previous one had drawn too small, roughly ten million vehicles by the end. The 2006 wave alone sits in this register at 1.2 million trucks, cited by nine other campaigns in both directions, its own siblings included. The modern version runs quicker: Hyundai's exploding seat belt pretensioners, recalled in 2021 and again in 2022 for the same failure with a wider net, and the Toyota pedal-entrapment campaigns of 2010, one of which enters the register at 2,230,661 vehicles while explicitly citing its 1.1-million-vehicle sibling as an expansion, with General Motors filing its own separate copy for the mechanically identical Pontiac Vibe.
The wrong-part remedy. The fix installed incorrectly, 135 campaigns confessing incorrectly installed and another 133 saying previously repaired. The register's purest specimen is the one the Canada article found: Ford's 2020 block-heater campaign, which recalled the repair performed under its previous block-heater campaign, the first fix having installed the wrong cable. The dealership does the work twice; the register writes it down twice.
The supplier echo. One component maker's defect refracted through every brand that installed it. The 2010 register contains the genre's monument: a single wheelchair-lift maker's restraint belt produced remedy-for campaigns from more than twenty bus, van, and motorhome builders in one season, Blue Bird to Winnebago to Prevost, each legally obliged to file for its own badge. This is the genre the leaderboard is made of, and the genre my counting understates, since the supplier's own parent campaign lives in a numbering series the text-match did not credit.
The software do-over. The patch that needed a patch, and the genre that is still being written. Jaguar's I-PACE traction battery entered the register in 2023 for overheating and is back in 2026 with a successor campaign citing the first, the fix revised while the article you are reading was being researched. Ford's Bronco Sport carries a lighting module recalled in 2022 and revisited by two 2025 campaigns. Software remedies fail politely, invisibly, and fast, and the 2020s bar in the chart above is partly their signature.
The remedy recheck. The government auditing whether a fix fixed anything. The register keeps a whole docket genre for this, the recall queries documented in the investigations article and haunting the zombie files, where a 1999 query closed with the immortal typo OBATINED RECALL, meaning the recheck itself produced the next recall. When a fix fails badly enough, the second campaign arrives with a federal investigation as its birth certificate.
What a second recall admits
Read as a legal document, a recall-of-the-recall is a remarkable confession: the manufacturer certifying that the remedy it previously certified was wrong, on the same population, for the same hazard. Read as a practical document, it carries the finding that matters for an owner: a completed recall is not a closed file. The letter that says your vehicle has already been repaired under a prior campaign is precisely the letter to take seriously, because it means the repair is the defect. The practical translation for this site's readers: when you check a used vehicle's recall history, read the campaign texts, not just the count, because the words remedy for, expansion of, and previously repaired mark the difference between a car that had five problems and a car that had two problems and three attempts. And the largest recalls in history, the ones in the tonnage table, are disproportionately chains rather than events, which this article's base rate now explains: the bigger the fix, the better its odds of needing one of its own.