I have been reading more of the internal State Farm documents being disclosed in the Oklahoma hail litigation, and something keeps bothering me. It is not that State Farm wants to improve the quality of its claims handling. Every insurance company should want to do that. What concerns me is whether State Farm’s measurements of success are actually measuring accuracy or simply celebrating lower claim payments.

There is a fundamental difference between a company trying to reduce claim-handling errors and a company defining errors in a way that systematically favors lower payments. The first is something every policyholder should applaud. The second could become a sophisticated corporate system for underpaying legitimate claims while everybody involved proudly calls it quality improvement.

The latest documents appear in a November 9, 2026, filing in Hursh v. State Farm. The homeowners’ attorneys are opposing State Farm’s effort to prevent additional corporate representative testimony concerning rewards and performance measurements associated with the company’s Fire Model Enhancement program, commonly called FME. The filing includes internal claims-quality presentations, financial performance reports, employee accountability procedures, deposition testimony, and a sworn declaration from State Farm consultant Kevin Mosby describing the company’s reinspection and claims measurement system.

One of the more revealing documents is a June 2021 State Farm report titled “P&C Claims Fire Claim Quality,” identified as Exhibit 5. Under the heading of positive quality indicators, State Farm reported that its ratio of total-roof replacements to partial-roof replacements had declined from 4.1 to 2.3. The same document reported that wind and hail estimate-payment bands exceeding $10,000 had decreased 13 percent year over year.

I have been studying and litigating property insurance claims for more than four decades. I cannot tell you whether those numbers represent better claims adjusting without knowing whether the underlying claim decisions were correct. A reduction in roof replacements might indicate that adjusters previously replaced roofs unnecessarily. It might also indicate that adjusters now refuse to replace roofs that legitimately need replacement. Yet State Farm placed these declining payment indicators among its “positive” quality trends.

Another internal presentation, identified as Exhibit 9, states that wind and hail accuracy results were measured using paid severity compared with compound annual growth rate targets, along with Fire Estimatic Difference, or FED. Paid severity is essentially the average amount paid on claims. Comparing that figure against financial targets is a perfectly legitimate business exercise. Insurance companies must understand their costs and operate profitably. But a financial performance target is not an independent measurement of whether a policyholder received the proper payment.

What makes these documents particularly interesting is how the measurements were incorporated into management oversight. State Farm describes monthly performance reviews, coaching, changes in settlement authority, and monitoring of decisions that move from no roof replacement or partial replacement to full replacement. Supervision is certainly appropriate. Good claims managers should review files, train adjusters, and correct mistakes. But what behaviors were these managers being encouraged to change, and what results were they expected to produce?

The November 9 filing raises another interesting issue. The homeowners’ attorneys identify internal communications proposing $27,250 in rewards for personnel associated with the FME initiative. They want to determine what those rewards were actually for and whether the program’s financial results influenced compensation. The underlying reward communications are filed under seal, so the public record does not establish that employees were rewarded for reducing legitimate claim payments. State Farm’s corporate representative, Wayne Stewart, testified that rewarding adjusters based on payouts or quotas would be inappropriate and that State Farm does not engage in that practice.

Fair enough. Then let’s examine the actual criteria and find out.

Kevin Mosby’s sworn declaration makes that inquiry even more interesting. Mosby explains that State Farm spent nearly three years and thousands of employee hours developing and refining its reinspection program and FED measurements. He describes the system as proprietary and argues that disclosing its internal methodology could harm State Farm competitively. The declaration explains the substantial investment in the measurement system, although it does not independently establish whether the system identifies and corrects underpayments and overpayments with equal rigor. That was not the purpose of his declaration, which addressed confidentiality, but it is certainly something worth investigating.

The most revealing evidence would be the underlying data showing whether State Farm’s quality-control program rewards accuracy equally when the result is paying more and when it is paying less.

If I were examining State Farm’s corporate representative, I would concentrate on six questions.

First, what makes a lower payment more accurate? Identify the independent evidence establishing that reductions in roof replacements and claim severity resulted from corrected errors rather than underpayment.

Second, what happens when reinspections reveal underpayments? Produce the number, percentage, and dollar value of reinspections resulting in increased payments, and explain how those results affect employee evaluations.

Third, how is FED calculated? Does the metric preserve the direction of the estimating error? Can a reduction in payment improve the score even when the reduced estimate is ultimately incorrect?

Fourth, what financial targets were communicated to claims management? Explain the relationship among paid severity, CAGR targets, roof-replacement rates, management approval requirements, and claims-quality scores.

Fifth, what exactly earned the FME rewards? Produce the reward approvals, calculation methodologies, performance criteria, and communications explaining the program’s “lift.” Distinguish productivity savings from indemnity reductions.

Sixth, how did these programs affect the Hursh claim? Identify the people, procedures, dashboards, and approval requirements actually used in handling the homeowners’ losses, including any connection to FME personnel or performance measurements.

The sixth question is indispensable. Evidence of an enterprise-wide practice becomes substantially more consequential in an individual bad-faith case when it can be connected to the particular claim.

These questions do not presume that State Farm has done anything wrong. Indeed, portions of its internal documents discuss legitimate training, correcting estimating mistakes, and improving the identification of overlooked damage. State Farm maintains that its programs are designed to improve claims accuracy. The question is whether the underlying data and management practices support that explanation.

There is another reason this matters. Corporate performance measurements influence human behavior. If employees repeatedly hear that declining severity is good, fewer roof replacements are positive, and higher payments require additional scrutiny, they may eventually understand what management expects even if nobody explicitly tells them to underpay a claim. A company’s claims culture is shaped by what its leaders measure, praise, question, and reward.

I have seen this issue arise in bad faith litigation for decades. Insurance companies sometimes use perfectly respectable words such as quality, accuracy, efficiency, stewardship, and accountability to describe programs that deserve much closer examination. Those words sound wonderful. I use some of them when discussing how our own law firm should operate. But labels do not establish results. Calling something a quality initiative does not make it one.

This is also why I believe insurance regulators conducting market conduct examinations should be examining these underlying corporate measurement systems. Looking at individual claim files is important, but understanding what management measures and how it rewards performance may be even more revealing. A claim file tells us what happened on one claim. The corporate performance system may help explain why it happened.

State Farm has the opportunity to demonstrate that its program genuinely measures proper claims handling. The Oklahoma homeowners’ attorneys are seeking to determine whether the actual data support that explanation, and I hope the court permits meaningful inquiry into these performance measures and rewards.

Ultimately, I would like to see an internal State Farm report celebrating the discovery that its adjusters had underpaid thousands of policyholders, explaining how those mistakes were corrected, and congratulating the employees responsible for finding them. I would also like to see how those corrections affected the company’s performance measurements.

If those reports and measurements exist, State Farm should be proud of them. If they do not, perhaps the company needs to reconsider what it means by quality. Because a claims department that celebrates lower payments without establishing that they are correct may be measuring something entirely different from the promise State Farm makes to its policyholders.

Thought For The Day

“It is wrong to suppose that if you can’t measure it, you can’t manage it – a costly myth.”
— W. Edwards Deming