Mod Manipulation Red Flags: When a Low X-Mod Is Too Good to Be True

A low experience mod on a submission is supposed to be good news. Sometimes it is. Sometimes it is the result of a company that discovered, usually with help, that the rating system has seams, and that a mod can be reset, diluted, or shifted without anything on the ACORD 130 looking wrong.

For a PEO, this matters more than it does for a carrier writing a single policy. On a large-deductible master program the PEO carries the client's losses inside the deductible regardless of what the client's mod says, so a manipulated mod does not reduce the PEO's risk; it only reduces the price the PEO thought it could justify. The underwriter's job is to find out whether the mod describes the business that is actually going on the payroll.

The rules below are summarized from the NCCI Experience Rating Plan and the WCIRB's published guidance on ownership changes. Where state plans differ, the state's rating bureau is the authority.

The rule everyone misunderstands: experience follows the business

The most common belief among owners, and some brokers, is that a new legal entity gets a fresh mod of 1.00. In almost every case that is wrong.

Under NCCI's rules, when ownership of a business changes, the experience transfers to the acquiring, surviving, or new entity. The experience is excluded from future ratings only when a material change in ownership is accompanied by a change in operations large enough to alter the governing classification and a change in the processes and hazards of the business. All three conditions must be met. Industry commentary on the rule describes meeting all three as highly unlikely, and the usual illustration is a woodworking shop sold to a buyer who turns it into an ice cream parlor. A corporation that converts to an LLC, a founder who sells to an employee, a company that forms a new entity and moves everyone over: each of those keeps its mod.

California's WCIRB applies the same principle through its own plan. A change in ownership is non-material where the prior owner retains at least a 50 percent interest, and in a non-material change the historical payroll and loss experience stays with the entity. Material changes are evaluated by the WCIRB to decide how the historical data is handled, and the rating bureau makes that decision, not the insured.

Businesses that try to evade this do so by not reporting. That is the thread that runs through nearly every red flag below.

Red flag one: a young entity with an old business inside it

The submission shows an entity formed eighteen months ago with a mod of 1.00 and sixty employees. The owner has been in the trade for twenty years. The address, the phone number, the equipment, and the foreman all predate the entity.

This is the classic reset. A prior entity accumulated a bad mod, was allowed to go dormant, and a new one was formed to pick up the operations. Under both NCCI and WCIRB rules, the formation of a new entity to succeed a dissolved or inactive one is explicitly a change in ownership that must be reported, and the experience follows. The 1.00 mod exists only because nobody told the bureau.

How to check: compare the entity formation date against the loss runs, the prior carrier history on the ACORD 130, and the public record. CSLB license history for California contractors shows the license's original issue date and the qualifying individuals, which often predate the entity on the application by years. State business entity filings show prior entities at the same address with the same officers. If the owner's prior company had a mod, the new one should too.

Red flag two: ownership changes that were never reported

NCCI's ERM-14, the Confidential Request for Ownership Information, is the form through which ownership changes are reported to the rating bureau, and the standard policy language requires an insured to report such changes to its carrier in writing within 90 days. The ERM-14 covers sales of ownership interests, mergers, asset sales to a successor, formation of a successor entity, receiverships, and the determination of whether separate entities share common ownership. In California the WCIRB takes ownership information through its online submission tool on WCIRB Connect rather than a paper form.

The red flag is a business whose ownership history, as told in discovery or visible in public filings, includes a transaction that should have generated an ERM-14 or a WCIRB ownership submission, and whose mod shows no sign that the bureau ever processed one. A mod that stayed flat through a merger, or that dropped sharply the year after a partner bought out another partner, is a mod to ask about.

How to check: ask directly. Ask for the ERM-14 or the WCIRB ownership submission confirmation for any change in the last five years. If the answer is that none was filed, the mod is provisional at best, and a PEO pricing against it should assume the bureau will eventually catch up.

Red flag three: payroll shifted between affiliated entities

Common majority ownership means combinability. Under NCCI's plan, the experience of all entities under common majority ownership is combined into a single rating. The manipulation is to keep the entities formally separate in the submission while moving payroll between them: the high-hazard work, or the employees with the claims, sit in one entity, and the clean entity with the clerical payroll and the low mod is the one presented to the carrier or the PEO.

Signs include multiple entities at a shared address with overlapping officers, payroll implausibly small for the described operations, and employees who appear on one entity's census and another entity's loss run.

How to check: ask for every affiliated entity, with ownership percentages. Compare total payroll across the group against what the operations require. Look at the census and the loss runs side by side for names that cross entities. In ModIQ, aggregating the loss runs by policy year across all affiliated entities and projecting a combined mod shows what the rating would look like if the bureau combined them, which is what the bureau will do once it learns they share ownership.

Red flag four: claims that move around the valuation date

Experience rating uses losses valued as of a specific date, typically six months before the rating effective date. A claim that is closed, or reserved down, just before valuation and reopened or reserved back up afterward shows up in the mod at its lowest point. This is more often adjuster timing than insured manipulation, but an insured with a cooperative adjuster and a motive can make it systematic.

Signs: a claim whose incurred value drops sharply on the valuation that feeds the mod and recovers on the next one, or a cluster of closures in the month before valuation on claims that had been open for years.

How to check: ask for loss runs at multiple valuations, not just the current one. ModIQ keeps each valuation as a snapshot, so the movement of individual claims between valuations is visible rather than inferred. A claim that moved in a way that conveniently helped the mod deserves a conversation with the carrier.

Red flag five: class shifting

The mod is a ratio of actual losses to expected losses, and expected losses are driven by payroll by class. Shifting payroll from a high-rated class into a lower-rated one does not change actual losses, but it does change the mod's denominator in a way that can make a bad loss history look average, and it changes the manual premium directly.

The signs are familiar from any misclassification review: a construction firm with a large share of payroll in 8810 clerical, or an audit history that shows payroll reclassified upward every year and re-reported low on the next application.

How to check: compare the class distribution on the ACORD 130 against the loss run, where claims are described by occupation and injury. Back strains in a warehouse do not belong to an entity whose payroll is mostly clerical. Compare against the prior carrier's audit if it can be obtained, and check the governing class against the public record of what the company actually does.

Putting it together

None of these flags is proof of anything, and most low mods are legitimate. A company with a good safety program, stable supervision, and a decade of small claims earns its 0.75. The point is that a mod is a claim about the past that was computed from inputs the insured supplied, and the underwriter's job is to check whether the inputs were complete.

The practical sequence we use in our own PEO and built into HALO IQ is this. InboxIQ parses the ACORD 130 and the loss runs. ModIQ aggregates the runs by policy year, keeps every valuation, and projects the mod from the losses rather than taking the bureau's published figure on faith. The public record, including entity filings and CSLB license history for California contractors, is matched to the company so a young entity with an old license is visible on the first screen. Affiliated entities are modeled together so the combined mod is a number, not a suspicion. And the proposal that comes out of the wizard prices the account on the risk the PEO is actually going to carry.

A mod that is too good to be true is usually a mod that has not been asked the right questions yet. They have to be asked before binding, not after the first large claim.