Article

2026-11-07

Last updated

Social inflation: what changes in case valuation

Social inflation: what changes in case valuation

Separate insurance loss trends from case evidence with a two-track review of social inflation, damages assumptions, and missing records.

Separate insurance loss trends from case evidence with a two-track review of social inflation, damages assumptions, and missing records.

Separate insurance loss trends from case evidence with a two-track review of social inflation, damages assumptions, and missing records.

Social inflation should prompt a closer look at case valuation assumptions, but it cannot supply the value of an individual claim. The useful distinction is between an aggregate insurance loss trend and evidence that changes this matter’s liability or damages analysis. Review both, separately, then explain whether the external information warrants testing assumptions or the case records support changing them.

Practical takeaways:

  • Keep insurance loss trends separate from proof about the particular claim.

  • Check a benchmark’s population, period, and relevance before using it.

  • Tie a revised damages assumption to a record, discrepancy, or identified omission.

  • Record what would trigger the next review, including evidence that could move the assessment either way.

This article is for plaintiff and defense readers evaluating litigation matters. Sources checked November 7, 2026. It provides legal information, not individualized legal advice.

Why the meaning of social inflation matters

In its October 19, 2026 public release, AM Best defines social inflation as claims-cost growth beyond general economic inflation associated with societal and legal trends. That is an industry framing, not a finding about the cause or value of an individual claim.

The term needs careful handling because a definition and an explanation are different things. Identifying growth beyond a broad economic benchmark does not, by itself, establish which causes produced that growth or how much each contributed. It also does not identify a change in the evidence of a particular case.

Our social inflation glossary provides a definition entry. For valuation purposes, keep the qualification explicit: using the term does not establish that a previous assessment was too low.

A productive review therefore starts with a narrower question than whether social inflation “applies”: What new information exists, and which assumption could it reasonably affect?

What AM Best measured, and what it did not establish

The AM Best public release reports average general economic inflation of 2.8% annually over 2014 through 2024. It reports that commercial auto liability loss payments for the first 12 months trended 6.4% annually during that period.

Those figures describe different measures. The release does not isolate a social-inflation rate from that 6.4% payment trend. Subtracting 2.8% from 6.4% would not establish an exact component caused by societal or legal trends.

The population also matters. Commercial auto liability loss payments are not a dataset of all tort claims, all verdicts, or the matter under review. Before importing the comparison into a valuation discussion, identify that boundary.

Only the public release supports this article’s treatment of AM Best’s findings. We have not reviewed the underlying paywalled report. The appropriate use here is as a reason to examine assumptions, with its limits attached, rather than as a numerical adjustment to an individual claim.

Why the meaning of social inflation matters

In its October 19, 2026 public release, AM Best defines social inflation as claims-cost growth beyond general economic inflation associated with societal and legal trends. That is an industry framing, not a finding about the cause or value of an individual claim.

The term needs careful handling because a definition and an explanation are different things. Identifying growth beyond a broad economic benchmark does not, by itself, establish which causes produced that growth or how much each contributed. It also does not identify a change in the evidence of a particular case.

Our social inflation glossary provides a definition entry. For valuation purposes, keep the qualification explicit: using the term does not establish that a previous assessment was too low.

A productive review therefore starts with a narrower question than whether social inflation “applies”: What new information exists, and which assumption could it reasonably affect?

What AM Best measured, and what it did not establish

The AM Best public release reports average general economic inflation of 2.8% annually over 2014 through 2024. It reports that commercial auto liability loss payments for the first 12 months trended 6.4% annually during that period.

Those figures describe different measures. The release does not isolate a social-inflation rate from that 6.4% payment trend. Subtracting 2.8% from 6.4% would not establish an exact component caused by societal or legal trends.

The population also matters. Commercial auto liability loss payments are not a dataset of all tort claims, all verdicts, or the matter under review. Before importing the comparison into a valuation discussion, identify that boundary.

Only the public release supports this article’s treatment of AM Best’s findings. We have not reviewed the underlying paywalled report. The appropriate use here is as a reason to examine assumptions, with its limits attached, rather than as a numerical adjustment to an individual claim.

Use two separate tracks for the review

We suggest a two-track review, not a valuation formula or an automated product feature. Keep one list for external context and another for matter evidence. The separation makes it easier to see whether a changed assessment rests on new proof, a different assumption, or a broader concern that still needs investigation.

  1. Describe the external information before interpreting it. In the external-context list, record the source date, source population, jurisdictional relevance, and selection limits. Ask what the source actually counts. Is it describing insurance payments, verdicts, or another measure? What period does it cover? What would make the information less applicable to this dispute? If the source does not answer a question, leave that limitation visible. Do not turn an unknown population boundary into an assumption that the benchmark is representative.

  2. Identify the exact matter assumption under review. In the matter-evidence list, record the damages category, supporting record, disputed assumption, missing item, and next review trigger. “Exposure has increased” is too broad to explain a revision. Identify what changed in the underlying reasoning. For example, distinguish a change supported by actual medical-cost or wage records from an assumption about jury attitudes. The former asks what the records establish. The latter requires its own stated basis and uncertainty. Neither should silently stand in for the other.

  3. Match each proposed revision to its support. Ask whether new information changes a factual input, challenges an assumption, or merely suggests testing an alternative. A changed external benchmark may warrant a sensitivity review: reconsidering the assessment under a clearly stated different assumption. It does not establish entitlement or value. Keep that distinction in the written explanation. “We are testing whether this assumption remains appropriate” says something different from “new records establish a different loss.” A reviewer should be able to tell which statement is intended without reconstructing the discussion.

  4. Set a specific next review trigger. Close each open item with the evidence or event that would justify reconsideration. That might be receipt of a missing wage record, clarification of a conflicting date, or additional information about the external source’s population. Use the trigger to keep uncertainty actionable. If no new matter evidence exists, say so. If the proposed change rests on external context alone, label it as an assumption to review rather than a fact established about the claim.

For the broader treatment of valuation inputs, see settlement ranges from the matter file. The narrower task here is to make the step from aggregate information to an individual assessment explicit and reviewable.

Hypothetical: new wage proof changes the question

Hypothetical: A claimant’s lost-income calculation uses an assumed period away from work. The available file lacks complete wage records. Both sides also receive an industry release describing rising commercial auto liability loss payments.

Later, wage records arrive. They support a different earnings amount or absence period from the one previously assumed. That evidence gives both claimant and defense reviewers a concrete reason to revisit the lost-income calculation. Depending on what the records show, the revised assumption could increase or decrease the claimed amount under review.

The industry release cannot fill the earlier evidentiary gap. It does not establish this claimant’s earnings, time away from work, or the explanation for a discrepancy.

The useful review entry identifies the former assumption, the new record, and the remaining question. External context stays in its own list. The change in the matter assessment has a traceable explanation.

What could move an assessment in either direction?

Use the review to look for discrepancies and omissions before looking for confirmation of a preferred result. A practical set of questions is:

  • Does a new record support or contradict the amount previously assumed?

  • Do the dates in the calculation match the supporting records?

  • Is an apparent change in cost documented, or merely inferred from a broad trend?

  • Has a missing item arrived, and does it resolve the question it was expected to answer?

  • Does the external source have a relevance limit that the earlier discussion overlooked?

These are review questions, not claims that any particular answer determines the result. Their purpose is to make a proposed revision explainable to either side.

Keep unchanged assumptions visible too. A revision to one category should not silently become a reason to revise every category. Where information remains incomplete, describe the unresolved issue and the evidence needed to assess it.

Keep lawyer judgment and disclosure duties distinct

A valuation review and a damages disclosure are related tasks, but they should not be treated as interchangeable.

In federal civil cases, Rule 26(a)(1)(A)(iii) generally requires a computation of each claimed damages category and supporting material, subject to the rule’s exceptions and protections. Rule 26(e)(1) requires timely supplementation or correction when a party learns that a disclosure or response is materially incomplete or incorrect and the additional or corrective information has not otherwise been made known during discovery or in writing, or as ordered by the court.

Those provisions do not automatically require disclosure of confidential valuation recommendations merely because damages computations are required. Lawyers should assess the applicable obligations separately from internal evaluation. Court orders, local rules, and applicable state law matter; federal Rule 26 is not a universal description of state practice.

For review, keep three questions distinct: What do the records support? What remains an evaluation assumption? What must be disclosed or supplemented under the governing requirements?

Our Strategy & Settlement page describes first-pass liability and exposure evaluation refined as records arrive, source-page-linked status reports, and lawyer control. Our Review page describes a Fact Graph: matter-specific connected facts linked to source pages and lines, with chronology and gaps updated as records arrive.

FAQ

Is social inflation the same as general economic inflation or CPI?

No. AM Best’s definition concerns claims-cost growth beyond general economic inflation associated with societal and legal trends. The public release does not isolate social inflation within its commercial auto payment trend or identify its economic comparator as CPI.

Can social inflation value one claim?

Not on its own. Aggregate information does not establish the individual claim’s facts. Use it to identify assumptions worth reviewing, then distinguish that exercise from changes supported by matter records.

Does every verdict trend apply locally?

Do not assume that it does. Check the source population, jurisdictional relevance, period, and selection limits. This article’s AM Best comparison concerns commercial auto liability loss payments, not a local verdict dataset.

What records can change a valuation assumption?

Start with the disputed input. Wage proof, medical-cost records, or documents resolving conflicting dates may warrant reconsideration of the relevant assumption. Explain what the record changes, what it leaves unresolved, and why the revision follows.

Sources

FAQ

Is social inflation the same as general economic inflation or CPI?

No. AM Best’s definition concerns claims-cost growth beyond general economic inflation associated with societal and legal trends. The public release does not isolate social inflation within its commercial auto payment trend or identify its economic comparator as CPI.

Can social inflation value one claim?

Not on its own. Aggregate information does not establish the individual claim’s facts. Use it to identify assumptions worth reviewing, then distinguish that exercise from changes supported by matter records.

Does every verdict trend apply locally?

Do not assume that it does. Check the source population, jurisdictional relevance, period, and selection limits. This article’s AM Best comparison concerns commercial auto liability loss payments, not a local verdict dataset.

What records can change a valuation assumption?

Start with the disputed input. Wage proof, medical-cost records, or documents resolving conflicting dates may warrant reconsideration of the relevant assumption. Explain what the record changes, what it leaves unresolved, and why the revision follows.

Sources