Law enforcement ran more searches of the federal government's vehicle-history database last year than in any year since the system was created. It's supposed to be the safety net that catches stolen cars, salvaged wrecks, and fraudulently re-titled vehicles before they reach an unsuspecting buyer. And by one measure, it's working better than ever.
But read the fine print of the system's own annual report, and a different story emerges: the states running this system are, in their own words, describing exactly where it fails.
The record-high number
According to the NMVTIS FY2024 Annual report, in fiscal year 2024, law enforcement agencies ran 2,374,562 searches through the National Motor Vehicle Title Information System (NMVTIS) — up 39% from the year before, and the highest volume in the system's history (NMVTIS FY2024 Annual Report).
NMVTIS is the federal vehicle title and history database, holding more than 2 billion records and run by the American Association of Motor Vehicle Administrators under Department of Justice oversight. It's existed in some form since the Anti Car Theft Act of 1992, and the pitch has always been simple: one national system so a car with a hidden salvage or junk title in one state can't quietly resurface with a clean record in another.
By the numbers, the system recently had its busiest year yet. Its brand records, the flags for flood, salvage, and junk titles, have grown 72% since FY2020, and its title history records now top 1.2 billion. Everything about the topline data says the system is bigger, busier, and more relied-upon than ever.
The turn: cracks in the system's own report
That's the headline. The report itself tells a messier story underneath it and it's the states running NMVTIS who are saying so.
Texas, one of the largest vehicle markets in the country, participates in NMVTIS by batch upload only. Which means it doesn't receive stolen-vehicle data back through its normal process at all. The state had to build a separate system just to catch what NMVTIS misses on its own, and 166 vehicles previously reported as junk slipped through and received fresh titles anyway.
Indiana logged 779,370 NMVTIS "hits" in a single year, but its own titling office says a significant share of those are data-entry and record-linking errors. In other words, the state that's supposed to trust this system doesn't fully trust its own numbers coming out of it.
Wisconsin uncovered something just as telling: "Salvage" brands that existed on other states' physical titles but were never entered into NMVTIS. Direct evidence that the federal database isn't a complete record of what's actually on the paperwork.
Texas had to manually review 113,721 NMVTIS-flagged discrepancies in one year just to clean up the reports.
Same system, same type of error, opposite outcomes, depending entirely on which state happened to be processing the paperwork.
Even the coordination between states has a basic weak point. According to the NMVTIS report's own findings, state NMVTIS coordinators say their single biggest challenge is simply reaching the right contact in another state. The whole system depends on one state checking another's data, and that handoff regularly breaks down on a human level.
Meanwhile, the supply side isn't keeping up either. Junk, salvage, and insurance (JSI) reporting, the data that's supposed to flag totaled cars before they're resold, fell 4% year-over-year, even in a year with one of the costliest hurricane seasons on record. And 38% of all vehicles reported to NMVTIS in FY2024 still have no final disposition on file; they're permanently stuck at "to be determined," never confirmed as crushed, sold, or salvaged.
The report even describes a $156 million scrap-metal theft investigation involving a business that was legally required to report every vehicle it destroyed to NMVTIS and had never heard of NMVTIS at all.
And there's a gap that's easy to miss entirely: NMVTIS has no distinct national brand category for flood damage. It gets folded into "Salvage," which means a "Salvage" title in the system could mean flood, could mean an accident, could mean any number of things and there's no way to tell which from the outside.
Why this matters to a buyer or seller
It's tempting to read all of this as an audit problem. Something for regulators to fix, not something that touches an actual car buyer. The data says otherwise.
Bumper surveyed thousands of people, asking one direct question: have you ever discovered after the fact that a vehicle you purchased had a salvage or junk title that wasn't disclosed at the time of sale?
Roughly one in five said yes: 502 out of 2,400 respondents, all describing the exact failure mode the federal report spends pages admitting to. A title problem that existed, that should have surfaced, and that didn't until after the money had already changed hands.
That's the real cost of the gaps above. It isn't hypothetical, and it isn't just a federal audit finding. It's hundreds of actual buyers who found out the hard way. As one investigator with the Missouri State Highway Patrol put it in the report, the data is only as good as what is reported. When reporting is inconsistent, delayed, or simply skipped, buyers are the ones who inherit the blind spot.
The Bumper bridge: consistent scrutiny, no matter which state processed the title
The report's own numbers make an uncomfortable case: how well a used car's history gets checked currently depends on which of the different state systems happened to process its title.
Texas manually reviewed 113,721 flagged errors in a year; Texas caught 18,916 clerk mistakes on its own. That level of diligence is real, but it isn't guaranteed everywhere. Some states report thousands of catches, others report almost none, and there's no way for a buyer to know in advance which kind of state handled their car's paperwork. Wisconsin's discovery that brands can exist on a physical title but not in NMVTIS shows that even checking the federal database directly isn't foolproof. And Texas's admission that it structurally doesn't receive stolen-vehicle data through its normal process is as clear a "gap by design" as a state has been willing to put on paper.
None of this means the system is broken. Michigan and Texas both show what real diligence can catch. It means diligence isn't uniform and even the most careful states are working around real gaps in the shared federal database itself. The lesson here isn't "don't trust the government system," but rather it’s that the effectiveness depends entirely on which state you happen to be dealing with. Even the best state can't fully compensate for a gap in NMVTIS itself.
This is exactly why a single-source check isn't enough. Bumper vehicle history reports pull from state motor vehicle agencies, NMVTIS, the FBI's National Crime Information Center (NCIC), NHTSA, J.D. Power, and a network of more than 50,000 dealerships nationwide. Layering multiple independent sources on top of the federal system rather than relying on it alone. That's the direct answer to Wisconsin's finding that a brand can exist on a title but never make it into NMVTIS, and to Texas's admission that it doesn't receive stolen-vehicle data through its normal process at all: if one source misses it, another one is likely positioned to catch it.
Don't leave it up to the lottery
The federal system had its most-used year ever in 2024, and it's still finding out, state by state, where it falls short. That's exactly why buyers and sellers shouldn't assume a "clean" record means a clean car.
This is where an independent, regularly updated vehicle history report earns its keep. Because Bumper draws on state DMV records, NMVTIS, FBI NCIC stolen-vehicle data, NHTSA recall and safety data, J.D. Power valuation data, and reporting from over 50,000 dealerships — not just a single database. It applies the same layered scrutiny to every car, regardless of which state's process it passed through or where the federal system's reporting fell short. Bumper's report, estimated market value comparison, and insurance rate check are built to be that second layer: not a replacement for the federal system, but the check you run because the federal system's own participants say you should.
Check a VIN, compare vehicles, or get an instant valuation on Bumper before you commit, because right now, whether the system caught everything comes down to luck of geography. It shouldn't have to.
Methodology
Federal data: All NMVTIS statistics come from the NMVTIS FY2024 Annual Report (Bureau of Justice Assistance, vehiclehistory.bja.ojp.gov), covering FY2024 (Oct 2023–Sep 2024) and verified against source page references.
Survey data: The "roughly one in five" figure is based on a Bumper survey (via Hotjar) of 2,400 website visitors, asking: "Have you ever discovered after the fact that a vehicle you purchased had a salvage or junk title that wasn't disclosed at the time of sale/purchase?" 502 (20.9%) said yes; 1,898 (79.1%) said no. Responses are self-reported and collected through September 22, 2026.
Bumper report sources: Bumper vehicle history reports draw from state motor vehicle agencies, NMVTIS, FBI NCIC, NHTSA, J.D. Power, and over 50,000 dealerships nationwide.
Limitations: This is a convenience sample of people already engaging with a vehicle-history product, not a random sample of all buyers, and responses are unverified and self-reported. The data doesn't capture when or where the purchase occurred or whether it was a dealer or private-party sale, and no margin of error applies to this observational sample.
The information contained in any Bumper report and on the Bumper website is not necessarily 100% accurate, complete or up to date, nor a substitute for your own due diligence