A primer on calculating workers' comp experience mod

By Michael Kelly
Risk Management Director

In working with the 60-plus counties and 50-plus public entities that are members of either the NCACC Liability and Property or Workers' Compensation pools, I have had the honor of assisting a lot of staff charged with the responsibility of risk management – or at least the procurement of their employer's insurance. Topics for this monthly column are typically driven by questions from our membership. During my travels if I am asked to explain a given topic on multiple occasions, I jot it down as an idea for a possible future column. This month's Managing Your Risk article is to address such a question that comes up often each year around renewal time.

Initially my reluctance to address this question – "what is my workers' compensation experience modifier and how is it calculated?" – made me take pause, because unless you are truly motivated to understand, the explanation can be a bit dry. Having prefaced this column with that caveat, the experience modifier is the driving factor that can have the greatest significance in the final calculation of the amount of contribution/premium you are charged each year.

It is also important to note that although we follow NCCI (National Council on Compensation Insurance) formula and calculation protocols, I can only speak definitively for the Pools, as I have not practiced retail insurance since 2004. This explanation will outline, in what I hope are easy-to-understand terms, the "method behind the math." Further, this discussion will be in general terms in an effort to explain the basic concepts, but will not delve into minutia in hopes that casual readers will not nod off. Should you want to go through and prove up your actual modification calculation, contact me directly and I can do that for you.

Your workers' compensation experience modifier – also referred to as your "experience mod" – is a comparison of the average expected losses for a county or entity of your given size (based on payroll by class of employee) compared to your county's actual losses. These two numbers are set up in fraction format, with the actual losses on top (numerator) and the average expected losses on the bottom (denominator).

The full actual formula including all its components is:

Thus, A/B (rounded to two decimal places) is your experience modifier. But in actuality it is more easily understood when boiled down and illustrated thusly:

Experience mod = Actual losses/Expected losses

Normally this fraction will generate a number larger than 1.00 or smaller than 1.00. If the answer is exactly 1.00, then your workers' compensation contribution/premium is exactly average for your size and total payroll. If it is less than 1.00 – 0.97, for example – it means your final cost will be lowered 3 percent below the average. If your number is 1.04, it means your final calculation is increased 4 percent over the average – again based on an average member account with the same total amount of payroll you have, as distributed by employee class of work. The reason for this is the modification factor is multiplied by the amount of contribution/premium, which is derived by multiplying each employee class by the rate for that employee class per $100 of payroll.

For example: Assume you had just one class of employee payroll – clerical class code 8810 – which might have a rate of $0.50 per $100 of payroll. Given $10 million in clerical payroll, the unmodified annual contribution/premium would be:

($10,000,000 annual clerical payroll/100) x 0.50 = $50,000 in annual contribution/premium

This $50,000 is multiplied by your experience mod of 0.97, which adjusts your actual total contribution/premium from $50,000 to $48,500 (3 percent less than average). Similarly, it would be adjusted up if your mod was 1.04 x $50,000, increasing it to a total of $52,000.

The range of possible increase (debit mod) versus decrease (credit mod) is limited only by your account's actual loss experience. I have seen counties with mods as high as 1.72 and as low as 0.61. This equates to a variance of 111 points and generates a contribution/premium difference of $86,000 to $30,500 in the above example.

In an effort to prevent huge swings up or down, the formula has components in it that help offset and essentially "dampen" the effect of a single, very large loss. This is often called a "shock loss" as on any given year it is possible for a truly unexpected "fluke" type loss to occur that does not represent the norm for your given operation. As such, the full amount of each total loss is limited for the actual amount used in the formula. For the current policy year of 2011-12, the capped total amount of dollars used for a single claim is $248,500. This means any single claim reserved over $248,500 gets capped and plugged into the formula for generating your numerator "actual losses" at $248,500, regardless of the actual reserved amount.

This capped maximum increases each year, taking into account inflation and the increased cost of medical care as shown:

There is a total of three year's loss experience utilized in the formula to derive the numerator in our equation, discounting the most recent year. So for example, an experience modifier that is to be used for a July 1, 2010, to June 30, 2011, policy term would utilize the actual claim experience for the years 2008-09, 2007-08 and 2006-07. The losses for the year 2009-10 would not be used in an effort to allow any open claims time to mature, and would be reserved by the adjuster accurately.

In addition, if the most recent loss year were to be included, the ability to actually calculate (called promulgate in the insurance world) the mod would not be possible until after the 2010-11 term policy had actually started. This is because the payroll could not be audited until the 2009-10 policy term ended, and it takes time to actually perform the auditing process (remember that the expected loss number for the denominator is based on a county's total payroll). Thus the policy for the 2010-11 term would not have a modifier until three to four months after it actually began to allow time to complete the payroll audit. Obviously this would not be optimal for the local government budgeting process.

The actual formula also has the ability to take into account both weighting and stabilizing factors in an effort to further adjust the modifier to reward accounts with fewer claims for a given total loss value than another with a higher claim count. Stated more simply, if "County A" averages 22 claims per year for a total loss amount of $250,000, and "County B" routinely has 56 claims per year for a total loss amount of $250,000 per year, "County A" would have a lower experience mod.

The reason for this is the well-known premise that if an insured has a high claim count, the likelihood of a claim becoming severe is greater than an account routinely having a less number of claims per year. However, for the calculation of your modifier, an "occurrence called in for information only" does not impact the promulgation of your modifier in the least. Only if in fact a dollar amount were to be set up as a reserve with the possible expectation for payment would it be accumulated in the numerator "actual losses" figure. This is a secondary question that I receive multiple times each year that is most important to understand.

Finally, know that a workers' compensation experience modifier does not "go south" or trend way above 1.00 due to one bad loss year. It normally takes time for it to get bad – and unfortunately also takes time for it to improve. The utilization of a three-year loss experience timeframe, excluding the most recent 12 months, helps smooth out the effects of one really lousy year.

The bottom line is that every year you should look at your modifier and ask yourself if it makes sense based on the loss experience that your county has sustained. If not, request the detailed backup information that goes into the calculation and check the payroll and larger claims to make sure they are correct and in fact yours, as honest mistakes can happen. Having done this, take the totals from the materials provided and plug them into the formula. It is then fairly easy to determine if what you have been provided is correct. Just do the math.

NCACC Risk Management Director Michael Kelly writes a regular column on risk management for CountyLines. With more than 32 years of risk management/insurance experience, he holds the Associate in Risk Management for Public Entities, Certified Risk Manager and Certified Insurance Counselor professional designations. He can be reached at michael.kelly@ncacc.org or (919) 719-1124. Archived versions of the column can be found online at www.ncacc.org/managingyourrisk.html.

Editor's note: The editor regrets the omission of a reference in May's edition of Managing Your Risk, "How to read an insurance policy: Part II." The DICE acronym that stands for "Declarations, Insuring agreement, Conditions and Exclusions" should have been credited to "Foundations of Risk Management, 2nd Edition AICPCU – Charles M. Nyce, Phd, CPCU, API." We apologize for the omission.