Basic premium, converted losses, tax multiplier, minimum and maximum, priced in good, expected and bad years against guaranteed cost and a large deductible.
A retrospective rating plan, or retro, sets the final workers' compensation premium for a policy year from the losses actually incurred in that year, within a minimum and a maximum agreed in advance. The insurer still issues the policy and handles the claims, but the employer pays for its own ordinary losses, plus a charge for everything else. Done well, a retro rewards an employer that controls its claims; done badly, it turns a bad year into a large bill that arrives years later. This guide prices a retro on Brannock Supply Co., a fictional distributor with a standard premium of $1,017,131, and compares it with guaranteed cost and a large deductible in good, expected and bad years.
Worked in full in The Workers' Compensation Handbook by Julian R. Sterling, with every figure reproduced in live Excel workbooks.Get the DRM-free PDF, $19.99 →
Retro plans vary by insurer and state, but in principle the premium is built the same way:
retro premium = (basic premium + converted losses) x tax multiplier, kept between the minimum and the maximum, plus the expense constant.
| Element | Brannock term | What it pays for |
|---|---|---|
| Basic premium | 0.20 of standard premium | Insurer expense and margin, the loss limit charge, the minimum and maximum |
| Converted losses | Losses limited to $250,000 per accident, x 1.11 | The year's losses plus a claims handling loading |
| Tax multiplier | 1.06 | Premium taxes and assessments |
| Minimum and maximum | 0.60 and 1.40 of standard premium | The band within which the premium moves |
In Brannock's case the basic premium factor is assumed to include the charge for the per-accident loss limit. Terrorism and catastrophe charges, $3,080 each, are added on top, as they are on the guaranteed cost policy, so both sides of the comparison carry the same fixed charges.
Every factor in a retro applies to standard premium, so the first step is to know exactly what it contains. Brannock's 2027 manual premium, payroll by class times the rate per $100, is $877,660. Increased limits for employers liability bring it to $882,926. The 1.20 experience mod lifts it to $1,059,511, and a 4 percent schedule credit of $42,380 brings it to a standard premium of $1,017,131.
Two consequences follow. The retro does not escape the mod: the basic premium, the minimum and the maximum all carry the 1.20 debit, explained step by step in the guide on how to calculate an experience mod. And the premium discount, worth $109,626 on the guaranteed cost policy, does not appear in the retro, because the insurer prices its expenses through the basic premium factor instead. Guaranteed cost, after the discount, the expense constant, the terrorism and catastrophe charges and a 2.5 percent state assessment, totals $936,603.
The expected year uses the loss pick: $642,907 of ground-up losses for 2027, built from Brannock's own loss triangle developed to ultimate and trended to 2027. In the case, 10 percent of expected losses fall above $250,000 per claim, so the losses that enter the retro after the loss limit are $578,617.
Against guaranteed cost of $936,603, the retro saves $33,761 in the expected year. The contractual worst case is the maximum plus fixed charges, $1,430,393.
An expected year is a planning device, not an outcome. Brannock's own history swung from a heavy 2023 to a light 2024, so the comparison has to show a range.
| Scenario | Ground-up losses | Retro total | Large deductible total |
|---|---|---|---|
| Good year (0.60 x pick) | $385,744 | $630,522 | $716,253 |
| Expected year (1.00 x pick) | $642,907 | $902,842 | $976,631 |
| Bad year (1.60 x pick) | $1,028,652 | $1,129,775 | $1,193,612 |
Against the fixed $936,603, the retro saves $306,081 in the good year and costs $193,173 more in the bad year. In the bad year a larger share of losses, 25 percent instead of 10 percent, is assumed to sit above the loss limit, so the limit does more work exactly when it is needed. Neither the minimum nor the maximum binds in any scenario: the band protects against a year considerably worse than the bad one, not against an ordinary bad year.
The point to put in front of a board is the trade. For $33,761 of expected saving, Brannock takes on a swing of several hundred thousand dollars between good and bad years. The scenarios are illustrations, not probabilities; an actuary should model the distribution before anyone signs.
A $250,000 large deductible divides the same losses differently: the employer pays its claims up to the deductible directly, the insurer charges a reduced premium for the layer above and for its services, and collateral secures the retained losses. In Brannock's expected year the deductible totals $976,631, $40,028 more than guaranteed cost in nominal terms, because the guaranteed cost premium keeps its $109,626 premium discount. Paying claims over five years helps: in present value at 5 percent the deductible costs $923,096 and saves $13,506. It also ties up collateral of $520,755 in a letter of credit, costing $6,509 a year.
The retro captures most of the upside with a contractual cap on the downside and, usually, lighter collateral. For an employer that wants to be paid for improving its losses, it is often the intermediate step between guaranteed cost and retention.
A retro premium is not computed once. The insurer bills a deposit at inception, computes the retro premium from incurred losses at a first adjustment some months after expiry, and recomputes it at intervals until the parties agree to close. Incurred losses develop: in Brannock's triangle a year valued at 12 months has a cumulative development factor of 1.528 to ultimate, and one at 24 months still has 1.145. A retro looks cheapest at its first adjustment. The finance team should accrue the expected final premium, not the latest bill, and expect several policy years to be open at once.
For Brannock the retro is plausible but not compelling, because its debit comes from frequency, the very thing a loss-sensitive plan charges for. The payroll side of the base deserves the same scrutiny: see what a premium audit changes.
Compare retro quotes on the same terms: the basic premium factor and what it includes; the loss conversion factor and whether defense costs sit in losses or in the factor; the loss limit and its charge; the minimum and maximum; the tax multiplier against the state's actual taxes and assessments; the timing and number of adjustments and the right to close; and collateral, its form and how it steps down. You can test how the mod inside standard premium moves with the free Experience Mod Calculator. The full program comparison, including dividend plans, self-insurance and the board presentation, is in The Workers' Compensation Handbook.
In principle, (basic premium + converted losses) x tax multiplier, kept between the minimum and maximum, plus the expense constant. For Brannock's expected year: (0.20 x $1,017,131 + 1.11 x $578,617) x 1.06 = $896,432, within the band, $902,842 after fixed charges.
No. The basic premium, the minimum and the maximum are factors of standard premium, and standard premium includes the mod. Brannock's retro is built on a standard premium that carries its 1.20 mod.
The maximum premium plus the fixed charges. For Brannock, a maximum of 1.40 times standard premium gives $1,423,983, or $1,430,393 with the expense constant and the terrorism and catastrophe charges.
It depends on the terms and the employer. On Brannock's terms the retro costs $902,842 in the expected year against $976,631 for the deductible in nominal terms, with a capped downside and usually lighter collateral.
This guide is one arc of The Workers' Compensation Handbook. The book follows one employer through class codes and payroll, the premium algorithm, the experience mod claim by claim, the premium audit, claims and return to work, the loss development triangle and the choice between guaranteed cost, a retrospective rating plan, a large deductible and self-insurance.
Paperback PDF edition, $19.99 Book page and companion workbooks
Free download: Experience Mod Calculator (Excel).
Illustrative figures computed by the book's model; the company, its claims and the case state's rating values are fictional. Rules described in principle as read in November 2026; states differ. Education only, not insurance or legal advice.