The Recoverable Value of the Pet Lines Loss Ratio Gap

Author: Andrew Hellman (Stride Systems)

Pet insurance loss ratios run roughly 14 points above the personal lines average. Approximately half reflects veterinary cost inflation; the remainder reflects an addressable behavioral data gap. This note sizes the recoverable value in that gap, built from the loss ratio improvements the same input produced in adjacent lines and corroborated by the size of the gap itself.

Pet insurance loss ratios run roughly 14 points above the personal lines average [1] [2], and they have held there for years rather than converging the way a maturing line usually does. A gap that wide and that persistent forces a sharper question than why it exists: how much of it can be recovered, and how much that recovery is worth.

The gap may be split into two parts, different in kind. Roughly half is veterinary cost inflation, up approximately 40% since 2020 [3] and propelled by private equity consolidation across veterinary services. This half is a structural cost problem, largely outside insurer control and difficult to normalize at portfolio scale. The other half is a behavioral data gap: the risk is priced on attributes captured once at inception, while the behavior that drives a claim accumulates longitudinally across the policy term and currently goes completely unobserved. This half is an information problem, and information problems close when the information arrives. The recoverable value lives entirely in the second half.

Pet lines is overwhelmingly a dog business, and the recoverable value of the pet lines loss ratio gap concentrates there for two reasons. The first is scale: dogs account for roughly 80% of pet lines premiums [4], so the dog book holds almost all of the recoverable loss cost, and the rest of the book can improve slowly behind it without materially changing the aggregate. The second is observability: a dog's risk is driven by its relationship with the household, how it is cared for, exercised, and kept; a relationship expresses itself as behavior that recurs and can be observed over time. Recurring, observable behavior is the missing input, and it is the same kind of input adjacent insurance lines have already converted into measured loss ratio recovery. The dog book is therefore not only the largest prize in pet lines but the one where the recovery potential is evidence-based.

Solve for the dog portion of the pet book and you have solved for a substantial portion of the line.

Pet lines loss ratio recovery potential cannot be measured directly, because the longitudinal behavioral dataset a loss ratio study would require does not yet exist at scale. It can be derived, because every adjacent line that made a previously unobserved behavioral signal observable published the loss ratio improvement the signal produced. Reassuringly, the pattern repeats across adjacent lines: a line prices a dynamic risk on static inputs, a new longitudinal/behavioral input arrives, and the loss ratio improves by a measured amount.

Six analog modules may be expressed across the four stages of the dog policy underwriting workflow. Each takes a behavioral dimension of pet lines risk, finds the adjacent line that already prices the same signal, and carries the natural, unadjusted uplift that line recorded.

New Business Qualification – analog-based loss ratio uplift potential: 4 to 7 points

• Stewardship Propensity, analog to usage-based / telematics engagement in personal auto. Natural uplift 3 to 5 points based on ITL, Vitality, and NBER research.*

• Breed-Environment Fit, analog to property geospatial / peril-fit underwriting in homeowners. Natural uplift 1 to 2 points based on Guidewire and Korem findings.*

Pricing and Segmentation - analog-based loss ratio uplift potential: 1 to 2 points

• Activity-Cost Exposure, analog to ergonomic / safety human wearables in workers' compensation. Natural uplift 1 to 2 points based on Barrett Actuarial and Gen Re/NCCI studies.*

Mid-Term Monitoring - analog-based loss ratio uplift potential: 2 to 4 points

• Behavioral Drift, analog to continuous risk monitoring in commercial auto. Natural uplift 1 to 2 points based on SambaSafety and NBER studies.*

• Stress Accumulation, analog to predictive risk stratification in health insurance. Natural uplift 1 to 2 points based on AJMC health-system studies.*

Renewal Reassessment - analog-based loss ratio uplift potential: 1 to 2 points

• Baseline Divergence, analog to property condition change detection in homeowners. Natural uplift 1 to 2 points based on Cape Analytics findings.*

* Full citations, the per-module derivation behind each range, and the underlying signal construction are set out in the white paper and simulation methodology.

Summed across the four stages, the raw analog uplift totals approximately 8.0 points in the conservative case, 11.5 in the base case, and 15.0 in the aggressive case. This result is indicative, but it is also directional. Literal mapping of these observed effects onto pet lines, without adjustment for differences in frequency/severity profile, instrumentation context, and adoption dynamics, would be an oversimplification. However, the analog evidence does support the proposition that material loss cost recovery is achievable in pet lines through the introduction of a novel longitudinal data layer that renders dynamic behavior observable between inception and claim.

Applied to the canine portion of global pet lines premium volume through 2033 (cumulative) on expected annual growth of 17%, the implied potential cumulative loss cost recovery is:

• Conservative Case - $20.6B in cumulative loss cost recovered.

• Base Case - $29.7B

• Aggressive Case - $38.7B

A figure assembled entirely from other lines invites an obvious challenge, and a second, independent method answers it. Read from the top down, the roughly fourteen-point gap splits into about half cost and about half behavioral, putting the addressable portion near seven points, or about $20 billion against the canine portion of global pet lines premium volume through 2033. One method divides the observed gap; the other ignores it and assembles external evidence. They share no inputs, and while they are both somewhat arbitrary in a strict sense, their near convergence is instructive: the bottom-up build sits at or above the top-down estimate at every confidence tier. In other words, the analog evidence indicates that longitudinal behavioral signals in pet lines carry a headroom of loss ratio recovery potential greater than the problem they are designed to address.

The recovery is therefore a measurement, not a forecast. It is the recovery that adjacent lines have already realized when the same type of behavioral/longitudinal input arrived, carried conservatively to the largest and most observable line that still lacks it: the canine portion of pet. The segment's analytics investment is already substantial and has developed more rapidly than the data signal available to feed it. The recovery requires the one input pet lines has never had: a continuous view of the animal-human pair, captured as it changes rather than inferred once at inception.

The value of acting on this observation is largest while the missing input remains scarce. A softening broader market lifts appetite for asymmetric, well-understood opportunities, and this is one. The complete methodology is set out in the white paper Pet Insurance Loss Ratios: Addressing the Behavioral Data Gap. An interactive Simulation applies the same recovery logic to a participating book over a realistic 2027 to 2033 deployment window, and lets a reinsurer or carrier model the recovery against its own pet lines GWP.

- -

Stride Systems builds longitudinal behavioral data infrastructure for pet lines underwriting.
For further discussion, contact andrew@stridesystems.io.