What Is an Experience Modification Rate?

A roofing contractor in North Carolina pays $180,000 a year in workers' compensation premiums. His experience modification rate is 1.32. He has never questioned it, because he assumes it is simply what the insurance company calculated.

What he doesn't know is that his mod carries a classification error that has sat on his experience rating worksheet for three years. That one error inflates his premium by roughly $43,000 a year. Across the three-year experience period, he has overpaid by close to $130,000. Nobody told him, nobody checked, and the money is gone.

This is not a rare event. It is a structural feature of how experience rating works, and it may be happening on your worksheet right now.

What an experience modification rate is

An experience modification rate (EMR) is a multiplier applied to your workers' compensation premium based on your company's claims history compared to the industry average for employers of your size and classification. It is calculated by NCCI (the National Council on Compensation Insurance) in most Southeast states, and by the state rating bureau in jurisdictions that run their own. North Carolina, for example, uses the North Carolina Rate Bureau.

A mod of 1.00 means your claims experience matches the average for your industry and size. Above 1.00, you pay more than that average. Below 1.00, you pay less.

The surface math is simple. If your base workers' comp premium is $100,000 and your mod is 1.25, you pay $125,000. At a 1.00 mod, you would pay $100,000. That 0.25 spread costs $25,000 every year it stays in place. For construction contractors, where base premiums commonly run $100,000 to $500,000 or more, even a small mod difference becomes tens of thousands of dollars annually.

How NCCI builds the number

Your mod is not claims divided by premium. It is a statistical comparison of your actual loss experience against the losses NCCI expects for an employer of your size and classification.

NCCI pulls three years of payroll and claims data, specifically the three completed policy years before the most recent one. From your classification codes and payroll, it calculates your expected losses, which is what your losses should have looked like. It then compares that to your actual reported losses. Higher actual losses push the mod above 1.00; lower actual losses pull it below.

The formula weights claim frequency more heavily than claim severity. Every claim is split into a primary portion (the first dollars, up to a state-approved split point) and an excess portion (everything above it), and the primary portion carries far more weight. The practical effect is that several small claims move your mod more than one large claim of the same total value. A ballast value stabilizes the result so the number doesn't swing wildly, and a weighting factor governs how much your own experience counts versus the industry average, scaled to your size. Larger employers are rated more on their own record; smaller employers get more stabilization.

The output is a single number that multiplies every dollar of your premium.

Why your mod might be wrong

Here is the part most contractors never hear. NCCI calculates your mod from data your insurance carrier reports. NCCI doesn't audit that data. It doesn't visit your job site, confirm your classification codes, or verify how your payroll was allocated. It runs the formula on whatever the carrier submits.

When the carrier reports inaccurate data, and that happens routinely, your mod rises, your premium follows, and nothing flags it unless someone deliberately reviews the worksheet. NCCI's own classification inspection data shows certain construction codes are misassigned at rates above 69 percent when a worksite is physically inspected (NCCI Classification Inspection Program). These are not obscure edge cases. They are systematic, and they are baked into how data moves from carrier to rating bureau.

The errors that inflate a mod tend to fall into a handful of recurring categories, none of which self-correct. Classification codes get assigned incorrectly at policy inception or at audit, and the rate gap between similar-sounding codes can be substantial. Claims that should code as medical-only get reported as lost-time, which strips them of a roughly 70 percent discount in the formula and lands the full claim value on the worksheet. Reserves on open claims, set as estimates of final cost, lock at a fixed valuation date and enter your mod at whatever number was on the books that day, regardless of where the claim later settles. Subrogation recoveries against third parties happen but don't always make it onto the worksheet. Carve-outs for clerical staff and outside salespersons get denied at audit, pushing payroll back into higher-rated codes.

Each of these has a documented cost in mod points. The contractor sees one number on the policy and rarely sees the worksheet behind it.

Why nobody catches these

If these errors are this common, the obvious question is why they persist. The answer is that no one in the chain is paid to look.

Your carrier reports the data and has no incentive to find errors that would lower your premium. NCCI is a statistical organization, not an advocate for any individual employer. Your broker sells and services the policy and is focused on coverage, markets, and renewals, not on auditing the actuarial calculation itself. That leaves you. Most contractors have never seen their experience rating worksheet, don't know it exists, and wouldn't know what to look for. That gap is exactly where the money leaks.

What a wrong mod costs

Consider a concrete contractor in Georgia with $1.5 million in annual payroll, a base premium of $66,000, and a mod of 1.18. He pays $78,000 a year. If a classification error and a stale reserve are corrected and his mod settles at 1.03, his premium drops to roughly $68,000, a $10,000 annual reduction.

The part most owners miss: because the experience period spans three policy years, that error sat in all three. A correction doesn't only fix next year. It can generate refunds for prior policy years as well. The higher your payroll and rate, the larger the per-point impact, which is why roofing, framing, steel erection, and drywall contractors see the steepest premium swings. We typically see the largest recoveries concentrated in those high-rate trades.

What a 1.00 mod actually means

There is a persistent myth that 1.00 is a good mod or a target to chase. A 1.00 mod simply means your claims experience is exactly average for your industry and size. It is the baseline, not a goal.

Whether your mod should sit above or below 1.00 depends entirely on your real loss history. A mod above 1.00 driven by genuine claims is accurate and reflects a safety issue. A mod above 1.00 driven by reporting errors is a different problem with nothing to do with your safety record. The useful question isn't "is my mod high?" It is "is my mod correct?"

What an audit would check

An audit checks whether the data flowing into the mod calculation matches what the carrier's records and your business operations actually show. That means reconciling claim values on the NCCI worksheet against current carrier loss runs, verifying classification by class code against the work actually performed, confirming claim type coding against return-to-work and indemnity activity, and tracing whether subrogation recoveries and carve-out credits have been applied where they belong. In our reviews of Southeast contractor worksheets over the past few years, most contain at least one of these errors, and most contractors don't know it.

A clean mod doesn't lower the carrier's filed rate, but it can lower the multiplier the rate gets applied against. Send us your NCCI worksheet and we'll review it for free.