The Core Formula and the Answer You Came For
If you need the blunt answer to how to calculate loss ratio insurance, here it is: divide incurred losses plus loss adjustment expenses (LAE) by earned premium, then multiply by 100. The formula is (Incurred Losses + LAE) ÷ Earned Premium × 100. In my first year building actuarial reports for a regional carrier, I used written premium by mistake and reported a 42% ratio that later corrected to 68%—a gap that would have skewed pricing decisions.
That early error taught me the single most ignored nuance: the denominator must be earned premium, not written. Earned premium reflects the portion of coverage that has actually expired, while written premium counts all policies sold in the period. Getting this wrong inflates or deflates your ratio depending on growth phase.
For a quick self-check, our Loss Ratio Calculator automatically flags whether you’ve input earned versus written figures, which saves hours during audit season.
The numerator also hides a split that beginners miss. Loss adjustment expenses are not part of the pure claim payment; they are the cost of investigating and defending claims. Statutory filings separate ALAE and ULAE, and if you omit them your ratio will understate true loss cost by 5–15 points depending on line.
Another misconception is that the ratio is always an annual metric. It can be computed quarterly, but seasonality in property lines (hurricane season) makes a single quarter meaningless. I advise clients to review trailing 12-month and multi-year views simultaneously.
I learned this the hard way when presenting a Q3 ratio of 48% to a board, only to be asked about hurricane exposure. The trailing 12-month view was 71%. Always default to trailing 12-month for external reporting.
Earned vs. Written Premium: The First Modeling Decision
Insurers recognize premium as earned proportionally over the policy term. A 12-month policy sold in January yields 1/12 of premium earned each month. If you write $1.2M in a year but only 60% of it is earned by year-end, your earned base is $720k, not $1.2M.
This matters because a high-growth insurer can show a deceptively low loss ratio on a written basis—losses haven’t caught up to the flood of new premiums. Conversely, a shrinking book looks worse on written basis. Practitioners standardise on earned premium for apples-to-apples comparison, as required in NAIC blank reporting.
When I consult for MGAs, I insist on an earned-premium bridge: show written, unearned reserve movement, and earned. Skipping this step is why many insurtech dashboards display volatile ratios that confuse investors.
Consider a real scenario: a startup writes $50M in year one but earns only $18M. If it pays $12M claims, written-basis ratio = 24%, earned-basis = 67%. The 43-point gap could mean the difference between appearing profitable and being flagged for inadequacy. Always ask which base a competitor’s published ratio uses.
GAAP and statutory accounting both use earned premium, but the timing of earnability can differ slightly under ERP (effective date) rules. For most P&C lines the difference is immaterial, but for long-duration contracts (some health) it matters.
One edge case: minimum premium policies or single-premium annuities earn differently under statutory rules. If you’re calculating loss ratio for a credit insurance product with upfront earning, confirm the earn pattern with the filing actuary.
Incurred vs. Paid Losses: Reserves and Adjustment Expenses
The numerator is where analysts separate amateurs from pros. Incurred losses equal paid losses plus the change in case reserves plus IBNR (incurred but not reported). Most beginner guides stop at paid losses, which understates true cost in a deteriorating book.
Loss adjustment expenses split into allocated (claim-specific legal fees) and unallocated (overhead of claims departments). I once reviewed a workers’ comp file where ALAE was 22% of incurred losses—ignoring it would have made the loss ratio look 18 points healthier than reality.
The thing nobody tells you about reserves: they are estimates, not facts. If an insurer strengthens reserves (increases them), incurred losses jump even if no cash left the door. That creates a “ghost” loss ratio spike that signals conservative booking, not necessarily worse accidents.
When to Use Paid Loss Ratio Instead
For start-ups with no credible reserve history, a paid loss ratio over 12 months can be a cleaner signal, but you must pair it with a caveat about lag. Long-tail lines like liability need 5+ years to mature, so incurred is mandatory for trend analysis.
Decision Matrix: Which Loss Base to Use
- Early-stage insurer, short-tail line (auto physical damage): Paid loss ratio acceptable with lag note.
- Established P&C with reserves: Incurred + ALAE + ULAE, annual and 5-year aggregate.
- Health insurer under ACA: Medical loss ratio per CMS definition (incurred claims + quality improvement ÷ premium revenue).
- Reinsurance recoverables: Net incurred after ceded reinsurance, not gross.
I keep this matrix pinned above my desk. It prevents the classic error of comparing a paid-based startup ratio to an incurred-based mature competitor ratio—a mistake I saw a venture capital analyst make in 2021, overvaluing a fledgling insurtech by 30%.
Actuaries often use chain-ladder or Bornhuetter-Ferguson methods to estimate IBNR. If you’re handed a triangle, don’t blindly accept the carried reserve; a 10% shift in IBNR moves the loss ratio by roughly that proportion of the unpaid-to-paid ratio.
How to Calculate a 5-Year Loss Ratio From Public Filings
The People Also Ask query “how to calculate a 5 year loss ratio” is answered by summing incurred losses (including LAE) across five annual statements and dividing by the sum of earned premiums for the same years. You do not average the five annual ratios—that weights small premium years equally with large ones and distorts the trend.
Below is an illustrative composite drawn from statutory filings of a mid-sized property carrier (numbers rounded to protect confidentiality but reflective of actual 10-K magnitudes I’ve modeled):
- Year 1: Earned Premium $210M, Incurred+LAE $138M (Ratio 65.7%)
- Year 2: Earned Premium $234M, Incurred+LAE $151M (Ratio 64.5%)
- Year 3: Earned Premium $258M, Incurred+LAE $179M (Ratio 69.4%)
- Year 4: Earned Premium $271M, Incurred+LAE $205M (Ratio 75.6%)
- Year 5: Earned Premium $290M, Incurred+LAE $214M (Ratio 73.8%)
To compute the 5-year loss ratio: total incurred+LAE = 138+151+179+205+214 = $887M. Total earned premium = 210+234+258+271+290 = $1,263M. The aggregate ratio is 887 ÷ 1263 × 100 = 70.2%. Notice this differs from the simple average of the five annual ratios (69.8%) because premium grew; the aggregate method is the correct PAA answer.
For a health insurer subject to the 80/20 rule, the same math applies but the numerator is “medical claims incurred plus quality improvement expenses” and denominator is “premium revenue minus taxes/fees”. Using a publicly traded health plan’s 2018–2022 figures (illustrative): total premiums $48B, total medical costs $40.3B yields MLR of 83.9%, comfortably above the 80% threshold.
The ACA requires rebates if the MLR falls below threshold averaged over three years, not a single year. That’s why many health CFOs manage a 3-year rolling MLR, a close cousin to the 5-year property aggregate we computed.
I’ve bundled this exact method into a free Excel template that auto-computes multi-year ratios from raw statement data—available alongside our Loss Ratio Calculator. Drop in your own filings and it flags reserve jumps year over year.
The template also computes a “trend index”: current year ratio minus 5-year average. In the property example, Year 4’s 75.6% versus 70.2% average signals a 5.4-point deterioration that triggered management action. That’s the kind of insight a single-year ratio hides.
What Does a 100% Loss Ratio Mean? Solvency, Not Just Break-Even
Another common search is “what does 100% loss ratio mean?” It means every earned premium dollar is consumed by claims and adjustment expenses. But that is before commissions, underwriting expenses, and taxes. So at 100% loss ratio, the insurer is already operating at an underwriting loss.
If the loss ratio stays at 100% with no investment income, surplus erodes. A single year at 105% might be survivable via investment returns, but two consecutive years above 100% in property lines is a classic precursor to remediation by the NAIC or downgrade by rating agencies. The “100%” threshold is therefore a solvency tripwire, not a neutral breakpoint.
In health insurance, however, the Affordable Care Act’s medical loss ratio rules actually require ratios near or above 80%—meaning 80 cents of premium must go to medical care. That’s a different lens; a 100% MLR in health could trigger rebate provisions but isn’t insolvency because administrative load is capped. We’ll cover that next.
Most people don’t realize that a 100% loss ratio in a cat-exposed property book might be accompanied by a capital raise, not bankruptcy, because reinsurance recovers part of the losses. Net loss ratio after reinsurance is the true solvency gauge. I recall a 2018 hurricane year where a client’s gross ratio hit 118% but net was 79% thanks to a 50% quota share.
Therefore, when interpreting 100%, always ask: gross or net? With or without LAE? One-time catastrophe or systemic mispricing? Those distinctions separate a panic headline from a reasoned analysis.
Rating agencies like A.M. Best publish supplemental rating worksheets where a sustained gross loss ratio above 100% in primary lines is a key trigger for a negative outlook. I treat their BCAR model outputs as the final arbiter when counseling boards.
Benchmarking “Good” Ratios by Line of Business
There is no universal “good” loss ratio. For property and casualty, a 60–70% loss ratio is often healthy because it leaves room for the expense ratio (typically 25–30%) and profit. For health insurers, the CMS medical loss ratio rules mandate at least 80% (individual/small group) or 85% (large group) of premiums be spent on care, so a “good” compliant ratio is 80–85%, not 65%.
Most people don’t realize that a very low loss ratio (say 50% in auto) can indicate overpricing or stale reserves—regulators may view it as extracting excess profit. I’ve seen state insurance departments demand rate decreases when personal auto loss ratios persist below 55%.
Quick Benchmark Table
- Commercial Property: Target 55–65% (volatile cat years can spike to 90%)
- Personal Auto: 60–70% stable; >75% signals pricing inadequacy
- Workers’ Comp: 65–75% due to long-tail reserves
- Health (ACA): 80–85% minimum by law; rebates if below
- Reinsurance treaties: Often 70–80% ceded ratio
Note that specialty lines like cyber or transactional liability have thin data and can swing from 30% to 120% in three years. I treat any line with less than 5 years of credible experience as “unbenchmarkable” and rely on scenario modeling instead.
Another nuance: expense ratios vary by distribution channel. Direct writers may run 18% expense ratio, allowing a higher acceptable loss ratio than independent agency companies at 28%. The combined ratio, not loss ratio alone, should dictate the target.
For marine insurance, a line where our Marine Insurance Premium Calculator is often used to derive earned premium estimates, loss ratios are heavily influenced by International Maritime Organization safety regs; a 70% year after a piracy spike is not atypical.
From Loss Ratio to Combined Ratio: The Profitability Link
The loss ratio is only half the story. The combined ratio adds the expense ratio (underwriting expenses ÷ earned premium) to the loss ratio. A combined ratio under 100% means underwriting profit; over 100% means underwriting loss offset only by investments.
For example, if our 5-year composite had an average expense ratio of 28%, its combined ratio would be 70.2% + 28% = 98.2%—profitable. But in Year 4 where loss ratio hit 75.6%, combined would be 103.6%, explaining why management cut agent commissions the following year.
Analysts I work with never report loss ratio without the expense context. A 65% loss ratio with 40% expense ratio (combined 105%) is worse than a 75% loss ratio with 20% expense ratio (combined 95%). The template I mentioned earlier includes a combined-ratio tab for this exact reason.
Trade-off: focusing on loss ratio alone can motivate claim denial practices that shrink the numerator temporarily but inflate litigation costs later. I’ve audited books where slashing LAE dropped the ratio 4 points in year one, then added 10 points of adverse reserve development in year three. The combined ratio smooths some of this, but culture matters more than math.
Investment income is the silent partner. A combined ratio of 102% might still yield net profit if the portfolio returns 3% on surplus. That’s why life insurers tolerate higher loss ratios on annuity wraps—their asset yield covers the gap.
The Mistakes I Made So You Don’t Have To: A Practical Checklist
Experience is the best teacher, and my first loss-ratio model had five flaws that delayed a client’s IPO prospectus by three weeks. Use this checklist to avoid them:
- Confirm earned premium from the statement of income, not the premium worksheet.
- Add LAE separately—don’t bury it inside “losses” if your data source splits it.
- Reconcile reserve changes: a 20% reserve strengthening will spike incurred losses without payment activity.
- Never average annual ratios for multi-year views; aggregate numerators and denominators.
- Segment by line of business; a company-wide ratio can hide a toxic auto book behind a calm homeowners book.
If you skip the segmentation step, you’ll repeat the error I made in 2017 when a 62% blended ratio masked a 91% ratio in coastal hurricane-exposed zones. That’s the kind of detail that separates a ranking article from a board-ready analysis.
Additional pitfalls: ignoring reinsurance recoverables (use net unless specifically told gross), mixing statutory and GAAP figures, and failing to adjust for catastrophic anomalies. I now add a “cat load” footnote whenever a single event exceeds 5% of earned premium.
Putting the Template to Work
You now have the formula, the 5-year method, the 100% interpretation, and the benchmark context. Open the Loss Ratio Calculator to input your own figures or download the Excel model that automates the aggregate calculation and flags years where reserves moved more than 15%.
Remember, loss ratio is a diagnostic, not a verdict. Pair it with combined ratio, reserve development, and market context before making any pricing or solvency judgment. The practitioners who win are those who treat the number as a starting line, not a finish line.