AUTOBODY DIRECTORY
Home › Guides › The Rate Loop: How the Prevailing Labour Rate Is Set

The Rate Loop: How the Prevailing Labour Rate Is Set

Short answer: An insurer sets a prevailing labour rate by surveying body shops. The one program agreement litigation has forced into public view ties the shops that sign it to that same figure three ways: as a billing fallback, an approval threshold, and a mandatory input.

By Anthony Braswell for Quorum Industries LLC, The Autobody Directory · Updated 2026-09-01 · How this was written, and what the machine may not do

How this page was produced: Researched, drafted and checked with AI assistance under human direction, and signed off by the named author. How this site is written

That agreement does three things to the shops that sign it: absent a separately negotiated rate it caps their billing at the lowest of their own survey submission, the survey's own result, and the rate they give any other insurer; it makes that same figure an approval threshold on their estimates whether or not a negotiated rate exists; and it obliges them to keep their pricing current in the survey. Where those terms bind the same shop, the survey's inputs are constrained by the survey's own outputs.

That describes a contractual architecture, quoted below from primary sources. Whether it moves market rates in the aggregate is inference, not a documented fact, and this page labels every inference as inference in the sentence that makes it. We do not claim that carriers agreed with each other. This is a claim about one buyer's contract with one seller, and that second theory was pleaded and rejected by the Eleventh Circuit sitting en banc in 2019, and nothing here revives it.

What one contract actually says

This research identified nine large carriers running direct repair programs. Of those, it found exactly one that publishes its standards and exactly one whose agreement litigation has forced into public view. Everything anyone says about "DRP contracts" as a class therefore rests on a sample of two, and the reason is written into the document itself: State Farm's Select Service Agreement provides at Section 5.a, headed "Non-Disclosure," that the shop "agrees not to disclose, distribute, or reproduce any part or section of the Agreement to any other person or organization" unless required by law. The copy we can read is a court exhibit in Pulera Collision v. State Farm, executed 14 April 2015 with Armando's Collision Center of Kenosha, Wisconsin, and posted in full by Repairer Driven News.

Section 4 of that agreement is headed "Competitive Price." Section 4.a reads:

"Repair Pricing. Provider agrees to estimate and bill for repairs based on a pricing agreement between Provider and State Farm. If no pricing agreement exists, Provider agrees to charge the lower of the: (1) Most recent labor rates and paint and materials pricing information submitted by Provider to State Farm through State Farm's survey process; or (2) Current labor rates and paint and materials pricing identified through State Farm's survey process; or (3) Labor rates and paint and materials pricing offered to or agreed to with any other insurer."

Two things must travel with that quotation. First, the lowest-of ladder is a fallback. It applies only "If no pricing agreement exists." Where the shop and the carrier have negotiated a pricing agreement, that agreement governs the shop's billing instead. The agreement nowhere defines what a "pricing agreement between Provider and State Farm" is or how one is reached, and how often one exists is not in the public record. Second, the same lowest-of structure recurs at Sections 4.b, 4.c and 4.d for original-equipment, recycled and non-original parts, each carrying the same "If no pricing agreement exists" condition and each ending in an "any other insurer" limb. Section 4.c runs to four limbs rather than three.

The survey obligation sits elsewhere. Section 5.j, headed "Business to Business Site Use," provides that "Provider also agrees to maintain current pricing and capacity information within the B2B Auto Repair Facility Survey Form." That form is the only pricing-submission mechanism the agreement names, and it is an inference, not something the document states, that it is the same instrument as the "survey process" of Section 4.a. On that reading, the same shop that may be bound to limb (2) is contractually obliged to keep feeding limb (1).

A third clause closes the information channel. Section 5.g, headed "Business Ethics/Anti-Trust," reads: "Provider agrees to follow ethical and professional conduct in its business practices with State Farm and vehicle owners, Provider further agrees not to disclose, discuss, or share labor rate or pricing information with other repairers, and acknowledges this activity may be construed as illegal price fixing." That clause states real law: horizontal rate-swapping among competing sellers genuinely can be price fixing. It is an inference, not a documented fact, that the clause's practical effect is to leave the survey as the shop's only signal about the market price of its own labour.

Two further terms carry no pricing-agreement condition at all, and they are the answer to the obvious objection that the carve-out empties the ladder. Section 4.s, headed "Specialty Repair Pricing," requires the shop to "contact State Farm management and first seek approval" for any estimate on an individual job priced above the lower of its own submitted rate or the survey rate. Nothing in 4.s turns on whether a pricing agreement exists, and it closes with the words "Regardless of the pricing utilized for such repairs." Section 4.i requires the shop to apply the Agreement's pricing whenever a customer presents a State Farm staff estimate, and provides that it "will not charge more for the repair than the price that would result from application of this Agreement." So the survey figure operates as an approval gate on the shop's estimates whether or not a negotiated rate governs its billing.

One further term extends the reach of the whole pricing section. Section 5.d applies the provisions of Section 4 "to all locations owned or controlled by Provider, even if not listed as a 'participating' location in the Addendum," and requires that a lower price given to any other insurer at a non-participating location "must be given to State Farm."

The loop, stated precisely

In sequence: Section 5.j requires the shop to supply the survey. Section 4.a(2) can bind the shop's billing to the survey's result. Section 4.s makes the survey result an approval threshold regardless. Section 5.g forbids the shop from comparing rates with the only parties who could tell it whether that result is right.

The mechanism needs no agreement among carriers. It is a property of one buyer's own contract architecture, operating bilaterally between that buyer and each shop that signs. That matters legally: the collective theories have been litigated and lost, and this research found no case in which the bilateral one was squarely tried.

A regulator wrote three provisions against it

The strongest evidence that the loop is a real hazard rather than a thesis is that California's Insurance Commissioner wrote three separate provisions against it inside one regulation. 10 CCR 2695.81, the Standardized Auto Body Repair Labor Rate Survey, contains the following.

Subdivision (d)(6), on program rates:

"No Standardized Labor Rate Survey shall use any discounted rate negotiated or contracted for with members of the insurer's Direct Repair Program, or any other Direct Repair Program, as defined in Section 2698.90."

Subdivision (d)(7), an explicit anti-self-reference rule barring the insurer's own work product from becoming the input:

"Only direct responses provided by an auto repair shop on a survey questionnaire that complies with subdivision (d)(9) of this section shall be used in a Standardized Labor Rate Survey. Labor rates obtained from the following sources shall not be used in a Standardized Labor Rate Survey: (A) Estimates written by the insurer to estimate repair claims, (B) Third-party automobile collision repair estimating software used by the insurer to prepare estimates, (C) Subrogation reimbursement, or (D) Any source other than direct responses provided by an auto repair shop on a survey questionnaire."

And subdivision (e)(2)(C), which addresses the same family from the invoice side. It lets an insurer adjust a rate down to what the shop actually charged over the preceding sixty days, on the strength of at least three invoices, then closes with: "For purposes of this subdivision, only repair invoices evidencing labor rates charged in connection with non-Direct Repair Program or other non-discounted auto body repair work may be used."

What these provisions prove is that the Commissioner considered the hazard credible enough to write three separate rules against it. They do not prove any particular carrier ran the loop, and no court or regulator has made that finding as a finding.

The limits, which are the interesting part

We are not saying that any of this is unlawful. California's own regulation contains a lesser-of provision of the same shape as the contractual ladder, the survey scheme is voluntary, the Department disclaims the product, and the antitrust caselaw on buyer-side most-favoured-nation clauses runs mostly against sellers. Each of those is set out below, because they are the parts of the record that cut against the argument on this page.

The rule does not close the loop. It narrows it. The second sentence of (d)(6) preserves the weaker form expressly: "However, nothing in this subdivision (d)(6) shall be construed to prohibit the use of the non-discounted posted labor rate of a Direct Repair Program shop in a Standardized Labor Rate Survey." A program shop's posted rate still counts. The exclusion of contract rates therefore only works if a program shop's posted rate is independent of its program participation. Whether participation depresses posted rates over time is not documented in any source this research reached, and it is the most important open question here. A third sentence requires the insurer to disclose when it has used a program shop's rate, but that disclosure is filed as non-public information under 10 CCR 2698.91(g)(5), so nobody outside the Department can count how often the box is ticked.

Nobody outside the Department knows how many surveyed shops are program shops, and that number governs how much any of this matters. A survey in which one respondent of six is a program shop is a different object from one in which four are. California collects the answer and withholds it: 2698.91(g)(5) requires the insurer to file the "name, physical address of record, and license number" of every responding shop that is a member of its own program, and (g)(6) requires the labour rates reported by each responding shop, both under a subdivision headed "non-public information." What is public under (d) is the responding shops' names and addresses, the total surveyed, the prevailing rate, the geographic area and the method. Program membership and per-shop rates are the two fields that would size the loop, and they are the two the rule keeps back. No claim on this page rests on any assumed value for that share, and because those two fields are collected and withheld, this page makes no claim about how far the loop moves rates in the aggregate.

The whole scheme is voluntary. Section 2698.91(j) states: "Nothing in this section shall be construed to require an insurer to conduct an auto body repair labor rate survey." The preamble to 2695.81 describes a survey the Commissioner "recommends insurers use ... if the insurer elects to use a survey." The only reward for following the standardized method is subdivision (c): a rebuttable presumption that the insurer "has attempted in good faith to effectuate a fair and equitable" labour rate settlement. Nothing compels a carrier to survey at all, and the regulation applies in one state.

The regulator disclaims the product. Subdivision (k) of 2698.91 provides that filing a survey "shall not be construed to indicate, and an insurer shall not represent, that the Department has approved the survey for use in settling and paying claims or that the Department has made a determination that the survey data is accurate and reliable."

And California's own regulation contains a lesser-of rule. Subdivision (e)(2)(B) of 2695.81 provides that where a shop quotes above its own posted rate, "the insurer may adjust the labor rate in the estimate to the lesser of the repair shop's posted labor rate or the prevailing auto body rate as determined by the survey." That is structurally the same shape as the contractual ladder criticised above, imposed by a regulator rather than by a counterparty. It is an inconvenient fact for the critique and it belongs in it. Three balancing provisions sit alongside it: subdivision (e)(4) permits negotiating the rate upward; (e)(5) states the rule does not imply the shop must accept the offer; and 2698.91(i) expressly permits an insurer to contract "for a labor rate that is higher or lower than the prevailing auto body rate."

How the number is actually computed

The regulation's prevailing rate is not an average. Subdivision (d)(5) defines it as "the labor rate at or below which a simple majority of surveyed shops charge in a specific Geographic Area," and the regulation supplies its own worked example: rates of $64, $65, $66, $66, $71 and $73 produce a prevailing rate of $66, "since four of the six shops (the simple majority) charge a rate of $66 or less." The arithmetic mean of that same set is $67.50, so on the regulation's own worked example the prevailing rate lands $1.50 below the mean.

The market is small by definition too. Subdivision (d)(8)(C) sets the core area at six responding qualified shops: the subject shop plus the five nearest by straight-line distance, geocoded from Census TIGER files to a thousandth of a mile, with a periphery of the core radius plus one mile. Subdivision (d)(2) requires the questionnaire to go to every shop registered with the Bureau of Automotive Repair, but only responders count, and eight equipment tests at (d)(4)(A) gate whose responses qualify.

Under 10 CCR 2695.8(f) the insurer's own estimate "shall be of an amount that will allow for repairs to be made in accordance with accepted trade standards for good and workmanlike automotive repairs," and where the claimant's shop estimates higher the insurer must pay the difference, produce a shop that will do it for the estimate, or reasonably adjust under (f)(3). Subdivision (e) of 2695.81 hangs on that hook.

The only enforcement file, read from the primary documents

This research found one U.S. regulatory enforcement file that put a carrier's own labour rate survey in issue, and its contents are commonly misreported. The California Department of Insurance and four GEICO entities settled File No. UPAC-2016-00001 by stipulation and order in 2016. The order, signed for Commissioner Dave Jones by Deputy Commissioner Anthony Cignarale, directs payment of "Two Hundred Thousand Dollars ($200,000)."

What the document does not contain is a finding. Paragraph 6 records that "Respondents have denied the allegations of the OSC." Paragraph 9 is explicit: "This compromise settlement is not an admission of liability, wrongdoing, or violation of law and the parties agree no factual findings or legal conclusions have been made." We are not claiming that the Department found any of it true. Any account of this matter that reports the Department as having found anything is wrong, including earlier drafts of our own research.

The underlying Order to Show Cause and Statement of Charges, dated 24 June 2016, is still worth reading for what the Department asserted its legal position to be. Paragraph 12 alleges the respondents "did not pay the labor rate charged by the claimant's body shop of choice or the labor rate as established in Respondents' own labor rate survey." Paragraph 15 goes further and states the standard directly: "A labor rate survey that is not reasonably supported by data or that results in a labor rate below the labor rate that is prevailing in a specific geographic area is not considered fair or equitable. Any attempt to use the labor rate from such a survey to pay or settle a claim is deemed as a settlement offer that is unreasonably low" within the meaning of the unfair claims regulations. That is the Department's asserted position in a pleading it never had to prove.

The same document sets out the chronology, and it is one of stipulations rather than adjudications. A 2006 Second Amended Order to Show Cause (File No. UPA 05048291) alleged, among other things, that the respondents "adjusted repair estimates without having conducted a labor rate survey to determine the prevailing labor rate in the area." That resolved by a stipulation without admission, followed by a May 2007 order carrying a $60,000 penalty. A second Order to Show Cause issued in December 2008 after, in the Department's words, the respondents "did submit a labor rate survey but the survey was not compliant with the statues [sic] or regulations governing labor rate surveys." That resolved by a January 2011 order carrying $10,000. The 2016 action followed 153 consumer complaints received between 1 January 2014 and 1 June 2015. The 2016 stipulation required a new survey within twelve months and, in the interim, an upward adjustment of the 2013 survey by 3.83 percent.

Across three enforcement passes between 2006 and 2016, the methodology was never adjudicated. It was stipulated around.

The antitrust analogy, and why it is weaker than it looks

A ladder that binds a seller to give the buyer its lowest price is a most-favoured-nation clause, and the caselaw on buyer-side clauses of that kind runs mostly against sellers. In Ocean State Physicians Health Plan v. Blue Cross & Blue Shield of Rhode Island, 883 F.2d 1101 (1st Cir. 1989), the First Circuit upheld a "Prudent Buyer" policy under which "Blue Cross required each of its participating physicians to certify that he or she was not accepting any lower fees from other providers than he or she was receiving from Blue Cross for the same service," approving the district court's view that "As a naked proposition, it would seem silly to argue that a policy to pay the same amount for the same service is anticompetitive, even on the part of one who has market power. This, it would seem, is what competition should be all about." In Blue Cross & Blue Shield United of Wisconsin v. Marshfield Clinic, 65 F.3d 1406 (7th Cir. 1995), Judge Posner wrote that such clauses "are standard devices by which buyers try to bargain for low prices, by getting the seller to agree to treat them as favorably as any of their other customers."

The case usually offered on the other side does not fit. In United States and State of Michigan v. Blue Cross Blue Shield of Michigan, filed 18 October 2010, with the parties stipulating to dismissal without prejudice on 25 March 2013, the complaint pleaded clauses requiring hospitals "to charge some or all other commercial insurers more than the hospital charges Blue Cross," in some contracts "as much as 40% more," so that Blue Cross "purchased protection from competition by causing hospitals to raise the minimum prices they can charge to Blue Cross' competitors." The same paragraph says the quiet part outright: "Blue Cross has not sought or used MFNs to lower its own cost of obtaining hospital services." That is a price floor that raises rivals' costs. A survey-referenced ceiling is the opposite instrument, and citing Michigan as authority against it would be an error.

Why the mechanism has never been squarely litigated

The shops have already lost the collective version twice. In Quality Auto Painting Center of Roselle v. State Farm Indemnity Co., decided en banc on 4 March 2019, the Eleventh Circuit affirmed dismissal of the Sherman Act price-fixing claim on pleading grounds, holding that "Following the example set by a competitor, without agreeing to do so in advance, is textbook 'price leadership'", that "The Body Shops have not pointed to any plausible reason that one should expect that prices in this market" would diverge, and that the steering tactics alleged were "not so idiosyncratic that they suggest conspiracy" but rather "methods that would logically be employed by any insurer to dissuade its insureds from using a disfavored shop." The court affirmed dismissal of every claim except tortious interference, which it vacated and remanded.

A year later the shops put the survey itself at the centre of the pleading, and lost on the same ground. In Automotive Alignment & Body Service v. State Farm, decided 6 March 2020, the panel recorded that "State Farm allegedly uses faulty methods and outright manipulation of data to generate 'market rates'" and that the survey was "allegedly more sham than survey." It disposed of the matching allegation by repeating the en banc holding that the conduct is publicly observable, since "State Farm must necessarily tell the rate to every repair shop in a given geographic area," and affirmed the dismissal of the antitrust claims. Read it for what it does not do: it never evaluates the survey's methodology, never holds a survey-derived rate legitimate, and never reaches whether the rate was accurate. The case turned on inferring an agreement.

It is an inference, not an established fact, that this is why the loop persists unexamined: the theories that have been tested all required proving something several carriers did together, while the clause at issue requires only two parties and a signature. The closest thing to a squarely pleaded version that this research found is Leif's Auto Collision Centers v. GEICO, a complaint dated 14 November 2017 alleging that "GEICO will not reimburse any auto collision repair shops for more than the maximum labor rate charged by ARX partners," a ceiling for every shop set by the carrier's own contracted shops. That is an allegation in a pleading; we did not establish the court, the docket number or the outcome.

What the other published program shows

Progressive is the only major carrier that posts its Service and Repair Standards publicly. Its most-favoured-nation clause runs to service rather than price: the facility "will provide Progressive and/or its Claimants with a level of service (in terms of scheduling priority, cycle time and quality) that is equal to or exceeds the level of service that is provided to any other insurance carrier(s)." The estimate is Progressive's: "The claims rep will complete the estimate and obtain a meaningful agreed price with the shop." That last clause cuts against the ceiling reading, as the State Farm pricing-agreement carve-out does, and it belongs here for the same reason. The shop "Will NOT substitute such non-OEM part or repair, rather than replace, the existing part, without Progressive's prior written consent," a restraint that removes shop discretion in the direction that would reduce the bill. The page sets operational deadlines, including logging into WebTracker "at least four times throughout the day" and picking up a vehicle "Within two hours of accepting the referral," but publishes no numeric labour rate and no cycle-time target in days. Nothing on it references a labour rate survey at all.

The California Autobody Association's compilation of surveys by GEICO, Liberty Mutual, Nationwide, Sentry and USAA as of 1 January 2023 found that rates "in all seven collision repair categories vary greatly," as reported by Repairer Driven News; the underlying numeric table is members-only and we did not obtain it. It is an inference, not a documented finding, that a figure reported to vary greatly between carriers surveying one state is not a property of that market.

In Pulera, the shops pleaded that "On November 5, 2015, State Farm sent a correspondence to Pulera indicating that it would be unilaterally reducing the Labor Rate for collision repair for all Select Service Providers from $56.00 per hour to $50.00 per hour," and that Kenosha County, Wisconsin had been grouped into the Chicago-Naperville-Elgin Core-Based Statistical Area. Those are allegations, untested. What is documented is the agreement one of those shops had signed: Section 5.j put its pricing into the survey, Section 4.a(2) let the survey's result set its price absent a negotiated one, and Section 4.s made that result an approval threshold either way.

What we could not establish

Corrections

This page is new. What follows are corrections made to its own draft before publication, recorded here because a page that shows only its wins is not evidence of anything.

2026-09-01, a regulator finding that does not exist. The brief handed to the writer of this page asserted that a state regulator had found a carrier adjusted estimates without conducting a survey. It did not. That wording comes from a 2006 Second Amended Order to Show Cause, which is a pleading, and the matter resolved by stipulation without admission. The 2016 stipulation in the same line of files records at paragraph 6 that "Respondents have denied the allegations of the OSC" and at paragraph 9 that "the parties agree no factual findings or legal conclusions have been made." The error came from the person commissioning the work rather than from the drafter, and the fact-checker caught it before publication. The page now states the matter as an allegation resolved without findings, and says so in the section itself.

2026-09-01, the survey identity is an inference, not a term of the contract. The draft ran the loop as though Section 5.j's "B2B Auto Repair Facility Survey Form" and Section 4.a's "survey process" were plainly the same instrument. The agreement never says they are. The identification is strong, because the B2B form is the only pricing-submission mechanism the document names, but it is an inference, and the input step of the loop depends on it. It is now labelled as an inference in the sentence that makes it, and it is the weakest load-bearing joint on this page.

2026-09-01, the number that would size the loop was never named. The draft did not identify the quantity that governs how much any of this matters: the share of a survey's respondents that are the surveying carrier's own program shops. California collects it under 10 CCR 2698.91(g)(5) and (g)(6) and files it as non-public information, while making the responding shops' names, the total surveyed and the prevailing rate public. The page now says that in the body, and says that no claim on it rests on an assumed value for that share.

2026-09-01, a quotation that did not exist as a contiguous string. The draft attributed to the Eleventh Circuit the phrase "so idiosyncratic as to support an inference." The opinion contains "let alone idiosyncratic, so as to support an inference of an agreement" in one passage and "not so idiosyncratic that they suggest conspiracy" in the passage this page actually cites. The spliced version was cut and the verbatim wording of the cited passage is used instead.

2026-09-01, half a Progressive sentence, and the half we dropped cut against us. The draft quoted Progressive's page for the proposition that the estimate is the carrier's, stopping at "The claims rep will complete the estimate." The full sentence reads "The claims rep will complete the estimate and obtain a meaningful agreed price with the shop." The clause we dropped weakens this page's own reading. It is now quoted whole, and its effect on our argument is stated where it appears.

2026-09-01, a pleading quotation truncated with a full stop we added. The draft ended the Department's paragraph 15 at "deemed as a settlement offer that is unreasonably low." That full stop is not in the document; the sentence continues into the unfair claims regulations it refers to. The quotation now runs to its clause boundary, and the passage is framed as the Department's asserted position in a pleading it never had to prove.

Related

Sources

Every quotation on this page was re-pulled from its primary source and string matched before publication, and anything that would not re-verify was cut rather than softened. The Select Service Agreement exhibit was read on 2026-09-01 as raw text from the whole seven-page document, and its Section 4.a matched this page's blockquote character for character at the fourth independent attempt; the GEICO stipulation and the underlying Order to Show Cause were read the same way. All seventeen distinct URLs above were read on 2026-09-01 and all seventeen resolved. Command-line requests to every external host on this list were refused by the session's own egress proxy with a 403 on the CONNECT tunnel, which is an organisation policy denial and not a refusal by any of these sites; the documents were retrieved by other routes that reach the same URLs, and the refusal was not routed around.

Three things on this list could not be obtained, and they are named rather than glossed. The California Autobody Association's numeric rate table sits behind membership and we did not get it. The public survey search at lrssearch.insurance.ca.gov was read on 2026-09-01 and loads, but it serves its data through a JavaScript application rather than in the page, so the filings behind it are not readable by a scripted request. And the two fields that would size the loop, program membership and per-shop rates, are filed as non-public information under 10 CCR 2698.91(g)(5) and (g)(6), so they are not obtainable by anyone outside the Department. This page is not legal advice.

General consumer information: not legal, insurance, or financial advice. Requirements, coverage, and practices vary by state, policy, and manufacturer.

Run a body shop? Your shop likely already has a page here, built from public records. Check it and claim it free: verifying only ever adds.
What does claiming add? It's free ›

Where this fits

Each link says what it is for. We add one only when a reader on this page has a real reason to need that page next.