Why Claims Leakage in High-Volume Health Portfolios Erodes Profitable Growth

Growth Does Not Fix a Leakage Problem, It Multiplies It

A growing health book looks like a success story from the outside, more premium, more scale, more relevance to cedants. What that outside view often misses is that claims leakage scales right alongside growth, since leakage is roughly proportional to claim volume, not a fixed cost that gets diluted as the book expands.

A reinsurer celebrating premium growth on a book carrying a real leakage problem is, in practical terms, celebrating a bigger version of the same margin drag. The dollar amount of leakage grows with the book, even if the underlying leakage rate as a percentage of claims stays perfectly flat.

That is the uncomfortable core of this problem, and it is why leakage deserves specific attention from anyone responsible for growth strategy, not just from the claims department managing it day to day. Growth strategy and claims quality are far more connected than most reinsurance organizations treat them.

How Exactly Does Claims Leakage Erode Combined Ratio?

Claims leakage erodes combined ratio directly, because leaked payments count as claims cost in the loss component of the ratio even though they represent no legitimate insured loss at all. A combined ratio calculation cannot distinguish between a dollar paid for a genuine covered claim and a dollar paid due to a coding error, duplicate submission, or outright fraud.

Every leaked dollar looks identical to a legitimate claims dollar in the aggregate financial reporting, which is exactly why leakage is so easy to overlook at the portfolio level. It hides inside a number that looks completely normal unless someone specifically investigates what is actually behind it.

Can a Book Look Genuinely Profitable While Actually Carrying Significant Hidden Leakage?

Yes, and this is one of the more counterintuitive aspects of the problem, since a book priced against a loss ratio that already includes historical leakage can appear to perform exactly to plan. Pricing built on a leaked baseline simply bakes the leakage into what "normal" looks like for that specific book, and performance against that already-distorted baseline can look entirely healthy.

The book is genuinely underperforming its true achievable margin, the margin it could earn if the leakage were eliminated, even while every reported metric suggests business as usual. That gap between reported performance and true achievable performance is invisible without a specific, deliberate effort to measure it.

How Does This Compound Across Multiple Renewal Cycles?

Each renewal cycle typically uses the prior cycle's reported loss experience as the pricing baseline for the next cycle, which means undetected leakage from one cycle becomes an accepted, embedded part of the cost structure going into the next. This is a genuinely compounding problem, not a one-time distortion that eventually washes out on its own.

Over several renewal cycles, leakage that started as a modest, correctable gap can become deeply embedded in what both the cedant and the reinsurer treat as the normal, expected cost of the book. By that point, identifying and removing it requires actively challenging an assumption both parties have implicitly accepted for years, which is a much harder conversation than catching it early would have been.

This compounding pattern parallels the assumption-drift dynamic covered in why behavioral lapse models fail to catch stress-driven deviation early, where a different kind of embedded, slow-building gap similarly becomes harder to correct the longer it goes unaddressed.

What Is the Capital Efficiency Cost of Unaddressed Leakage?

Capital held against inflated loss experience is capital that cannot be redirected toward genuinely profitable new business, which quietly reduces a reinsurer's overall return on capital across its full portfolio. This effect operates independently of the direct margin erosion already described, and it compounds with it rather than simply repeating the same cost in a different form.

A reinsurer holding capital against a book whose real loss ratio, after removing leakage, is meaningfully better than its reported loss ratio is effectively over-reserving relative to true risk. That trapped capital represents a genuine opportunity cost, not just a paper inefficiency, since it could otherwise support new treaties with real, uncontaminated risk profiles.

How Much of This Cost Could Realistically Be Recovered?

Industry estimates suggest fraud, waste, and abuse alone represent roughly 3 to 10 percent of total health care expenditures, according to NHCAA figures citing a range from a conservative 3% up to loss estimates "as high as 10% of our annual health outlay, which could mean more than $300 billion" in the United States. Genuine operational leakage from coding errors, duplicate payments, and contract misapplication adds to that fraud-specific figure in books processing claims at real volume.

That range gives decision makers a useful planning benchmark, not a precise number for any specific book, since actual leakage varies significantly by claims processing maturity, provider network composition, and audit discipline. Even capturing a fraction of that potential recovery, applied across a large, growing book, represents a material and durable margin improvement rather than a one-time windfall.

Milliman's research on effective claims auditing describes a real 2023 case where "a health insurance carrier" settled for "$172 million over inaccurate diagnostic codes," a concrete illustration of how large leakage-adjacent exposure can become in a single organization before it gets addressed. That figure is a legal settlement rather than a leakage estimate specifically, but it illustrates the scale of financial consequence that can accumulate around undetected claims quality gaps at real volume.

What Is the Fastest, Most Practical Way to Test for This Problem?

The fastest practical test is running a claims audit sample against a statistically representative slice of high-frequency claims, then comparing the recovery rate found against whatever leakage rate is currently assumed in pricing. This does not require a full operational overhaul to execute, it requires a defined, representative sample and a rigorous audit methodology applied to it.

Software-driven detection, using severity scoring to prioritize likely leakage and repayment-rate tracking to estimate genuinely recoverable amounts, makes this kind of targeted sampling dramatically more efficient than manual claim-by-claim review ever was. A Claims Cost Sustainability AI Agent applied to a representative claims sample can produce a defensible leakage estimate within a single audit cycle, giving decision makers a concrete number to weigh against the cost of remediation.

Does Distribution Channel Affect How Much Leakage a Book Carries?

Yes, digitally distributed and self-service health products often carry different leakage exposure than traditionally underwritten and administered business, because the claims patterns and provider relationships behind each channel differ meaningfully. Digitally distributed books frequently see higher volumes of smaller, routine claims processed through automated adjudication, which is efficient but can also let subtle, repeated coding errors accumulate unnoticed if the automated rules themselves are not periodically re-validated against real audit findings.

Traditionally distributed business, often carrying closer broker or agent involvement in claims support, can benefit from an additional layer of informal scrutiny that a purely digital channel does not have, though it typically processes lower claim volumes where the same percentage leakage rate produces a smaller absolute dollar impact. Neither channel is inherently safer from leakage, but each concentrates it differently, which matters directly for how a reinsurer designs audit sampling across a book that blends both distribution types.

A reinsurer reviewing a cedant whose growth is concentrated in a specific distribution channel should ask specifically how that channel's claims are adjudicated and whether the automated rules behind it have been recently validated, since channel mix is exactly the kind of detail that a portfolio-level loss ratio review will not surface on its own. Building channel-specific leakage assumptions into pricing, rather than a single blended assumption across the whole book, produces a materially more accurate view of true risk as distribution mix shifts over time.

How Should Growth Projections Account for This Risk Going Forward?

Growth projections should explicitly include a leakage sensitivity assumption, showing how projected margin changes under a range of plausible leakage rates rather than assuming a single fixed rate holds constant as the book scales. A projection built on a single, optimistic leakage assumption looks attractive on paper but leaves leadership unprepared for the more likely outcome, where actual leakage tracks toward the higher end of the plausible range as volume grows and monitoring struggles to keep pace.

Presenting growth projections alongside a leakage-sensitivity range, rather than a single point estimate, gives decision makers a realistic picture of the margin outcomes actually available under different execution scenarios. This practice also creates a natural, built-in incentive to invest in detection capability proportionally as the book grows, since the projected margin benefit of doing so is made explicit in the same document used to justify the growth strategy itself.

Growth is supposed to be the reward for building a good book of business. When claims leakage goes unaddressed, growth instead becomes the mechanism that quietly scales the problem, which is exactly why margin protection and growth strategy need to be evaluated together, not as separate conversations owned by separate teams.

Sources

• NHCAA: The Challenge of Health Care Fraud
• Milliman: How effective claims auditing can save healthcare dollars lost to fraud, waste, and abuse