Modified vs. Pure Comparative Negligence States: What's the Difference?
The exact same accident, with the exact same fault split, can produce a payout in one state and nothing at all in another, purely because of which comparative negligence rule applies.
The short answer
Pure comparative negligence allows a partly-at-fault driver to recover a reduced share of damages no matter how much fault they carry, even if they were mostly responsible for the accident. Modified comparative negligence sets a fault threshold — once a driver’s share of fault crosses that line, they can’t recover anything from the other party at all. Both systems reduce payouts based on fault percentage, but only one of them has a hard cutoff.
How pure comparative negligence works
Under a pure system, fault is assigned as a percentage, and a driver’s recovery is reduced by that percentage regardless of the size of the number. A driver found 90 percent at fault can still recover the remaining 10 percent of their damages from the other party, even though they caused the overwhelming majority of the accident. This system removes any cliff-edge cutoff, but it also means even a mostly-at-fault driver retains some right to recover.
How modified comparative negligence works
Under a modified system, there’s a specific fault percentage threshold, and crossing it eliminates recovery entirely rather than just reducing it further. In a hypothetical system with a 50 percent threshold, a driver found 49 percent at fault could still recover a reduced portion of their damages, while a driver found 51 percent at fault would recover nothing from the other party. That single percentage point on either side of the threshold can be the difference between a meaningful payout and none at all, which is why the detail matters so much:
- Small fault shifts have outsized consequences near the line. A dispute over just a few percentage points of fault can determine whether a driver recovers a reduced amount or nothing whatsoever.
- Disputing the fault percentage becomes higher stakes. In a modified state, both sides have strong incentive to argue for a fault split that lands on the favorable side of the threshold.
- The exact threshold varies and can change. Different modified states set the cutoff differently, and these rules can be revised over time, so the general concept matters more than any specific number.
How this affects insurance claims in practice
Because the applicable rule depends on where the accident occurred, not where the drivers live or where the insurer is based, the same crash could be handled very differently a short distance away, on the other side of a state line. This is part of why filing a claim after a multi-party or ambiguous-fault accident often involves a careful look at the fault percentages assigned, and why disputing an unfavorable fault determination can matter more in a modified state than in a pure one.
What to weigh
Since these rules directly shape whether a partly-at-fault driver recovers anything, understanding which framework applies in a given location — and, in a modified state, roughly where the threshold sits — is useful context before assuming a shared-fault accident will result in a proportional payout either way. Being found close to a fault threshold in a modified state also raises the stakes of how an accident gets documented and investigated, since that documentation can determine whether any recovery is possible at all. Because these are matters of state law that can change, this is general context rather than a substitute for understanding the specific rules that apply to a specific situation, and it pairs with the broader question of whether state-minimum coverage would even be enough to make a claim meaningful in the first place.
The bottom line
Pure and modified comparative negligence both scale down recovery based on fault, but only modified systems include a hard cutoff that can eliminate recovery entirely. Knowing which type of system applies where an accident happens is central to understanding what a partly-at-fault claim might actually be worth.