This framework, its simulations and every figure on this page were provided by a lone scientist the 4physics network encountered on an unmapped planet: one observer, one model, no independent replication. The analysis is relayed because it is worth reading, not because it is confirmed — please treat the numbers, including the 16.5× premium gap, as a working hypothesis rather than a settled result.
Space insurance prices risk using historical failure rates, ignoring cascade feedback. One collision can trigger chain reactions, creating correlated losses across entire portfolios.
Physics-based Monte Carlo simulation (1,000 paths, 50 years) coupled with Solvency II premium decomposition. Calibrated against three documented collision events.
Cascade-adjusted premiums are 16.5× higher than current market rates ($2,058M vs $125M per year for 500 satellites). That $1,933M gap is an unfunded liability.
Space insurance is a $1.2 billion market. The way we price it hasn't changed in decades: look at what broke last year, add a margin, move on.
There's a problem with this. It assumes tomorrow will look like yesterday. But Low Earth Orbit doesn't work that way. One collision makes hundreds of fragments. Each fragment is a projectile. Those projectiles hit other things, making more fragments. The whole system has a feedback loop that nobody's pricing in.
First described by NASA scientist Donald Kessler in 1978, the syndrome predicts that once debris density exceeds a critical threshold, collisions become self-sustaining: each impact generates more debris than natural decay removes, leading to exponential growth. In the worst case, entire orbital shells become unusable — a scenario sometimes called "cascading collisional breakup."
This framework presents the first quantitative model that couples physics-based cascade simulation with actuarial pricing. The result: insurance premiums that are 16.5 times higher than current market rates — a gap that represents both a systemic risk to insurers and a market failure in orbital sustainability.
If insurers underprice risk by 16.5×, operators have no economic incentive to choose less congested orbits or invest in debris mitigation. Correct pricing aligns private costs with systemic risks — making sustainability a market force rather than a regulatory burden.
I'm looking at one shell of LEO, centered at 800 km altitude, about 200 km thick (700–900 km). That's where most of the dead satellites are. That's also where Starlink and everyone else wants to operate.
Most models cheat: they put in a free parameter for how fast cascades grow. I don't. Everything here comes from actual collision physics — no tuning, no fitting. The debris growth is what the equations say it should be.
The expected annual collision rate between debris and satellites:
| Parameter | Description | Baseline |
|---|---|---|
| Nd | Debris population (>10 cm) | 36,500 |
| Ns | Active satellites in shell | 8,000 |
| σ | Satellite collision cross-section | 10 m² |
| v | Mean relative velocity | 10 km/s |
| T | Seconds per year | 3.16 × 10ⁿ |
| V | Shell volume | 1.29 × 10⁷ km³ |
At current population levels, this yields ~7 debris–satellite and ~16 debris–debris collisions per year.
Debris–debris collision rate scales with Nd² — doubling the debris quadruples the collision rate. Each collision produces new fragments (calibrated to the NASA Standard Breakup Model), which further accelerate collisions. This is the mathematical root of the cascade: a self-reinforcing loop with no natural brake.
Watch debris grow over 50 simulated years. The left panel shows the orbital shell with debris (red) and satellites (green). The right panel plots the debris population trajectory in real time.
Before the model can project future risk, it must reproduce known events. The fragment yield parameters are calibrated against three documented collision events that collectively created over 7,000 trackable objects:
| Event | Date | Fragments (>10cm) | Mass (kg) |
|---|---|---|---|
| Fengyun-1C (ASAT) | 2007-01-11 | ~3,500 | 880 |
| Iridium 33 × Cosmos 2251 | 2009-02-10 | ~2,300 | 1,510 |
| Kosmos 1408 (ASAT) | 2021-11-15 | ~1,500 | 2,200 |
These events anchor the fragment yield distributions: debris–satellite collisions produce a median of ~245 fragments (log-normal, μ=5.5, σ=0.8), while debris–debris collisions produce a median of ~3 fragments (log-normal, μ=1.1, σ=0.5). The asymmetry reflects the mass differential: a fragment hitting a satellite releases far more energy than two small fragments colliding.
The physics layer outputs debris population trajectories — how many objects occupy the shell at each point in time. To convert this into financial risk, we need a bridge: collision physics to loss probabilities to insurance premiums.
Step 1: From debris count N(t), compute each satellite's annual hit probability using the same flux equation: phit = Nd × σ × v × T / V. At current debris levels, this is ~0.09% per satellite per year — small, but non-negligible across a portfolio.
Step 2: Multiply hit probability by satellite value ($50M average) to get per-policy expected loss. Then aggregate across the 500-satellite insured portfolio, applying a Gaussian Copula to model correlated losses (when debris increases, all satellites become more vulnerable simultaneously).
Step 3: From the loss distribution across 1,000 Monte Carlo paths, extract risk metrics (VaR, CVaR, EVT) and price using the Solvency II framework.
European insurance rules say premiums need to cover four things:
| Component | Description | Amount |
|---|---|---|
| Expected Loss (EL) | Mean annual portfolio loss | $921.4M |
| Risk Loading | 1.5× (CVaR₉₅ − EL) tail surcharge | $966.0M |
| Capital Charge | 4% cost of VaR₉₉ regulatory capital | $32.0M |
| Expense Loading | 15% admin costs + profit margin | $138.2M |
| Total Premium | $2,058M / yr |
The risk loading ($966M) is the biggest piece — bigger than the expected loss itself. That's because the tail is brutal: in the worst 5% of scenarios, you're looking at average losses of $1,565M, almost twice the mean.
| Metric | Definition | Value |
|---|---|---|
| VaR₉₅ | 95th percentile of worst-year loss | $1,365M |
| CVaR₉₅ (ES) | Expected shortfall beyond VaR₉₅ | $1,565M |
| VaR₉₉ | 99th percentile of worst-year loss | $1,721M |
| CVaR₉₉ (ES) | Expected shortfall beyond VaR₉₉ | $1,854M |
| EVT VaR₉₅ | Generalized Pareto tail estimate | $1,803M |
| EVT shape (ξ) | Heavy-tail indicator (ξ<0 → bounded) | −0.032 |
The EVT shape parameter ξ = −0.032 indicates a slightly bounded tail (Type III), but the EVT VaR₉₅ of $1,803M — well above the empirical VaR₉₅ of $1,365M — confirms that tail events are substantially more severe than the Monte Carlo sample alone would suggest.
That $1,933M gap between what insurers charge ($125M) and what they should charge ($2,058M) isn't just a number. It's an unfunded liability waiting to happen. If a big cascade event hits while the market is priced at historical rates, the claims could blow past insurer reserves and trigger systemic insolvency across the whole space insurance sector.
But here's the flip side: if we price cascade risk correctly, we get economic incentives for sustainability for free. Operators who put satellites in less crowded shells or invest in end-of-life deorbiting would get lower premiums — making debris mitigation profitable without needing regulators to force it.
Step 1 — Collision-driven growth. Debris evolution is governed by Poisson-distributed collision events at each time step (Δt = 0.1 yr). Fragment yields follow log-normal distributions calibrated to observed events. No free cascade parameters.
Step 2 — Monte Carlo ensemble. 1,000 independent 50-year paths produce a full probability distribution of debris trajectories and corresponding losses.
Step 3 — Risk quantification. VaR, CVaR, and EVT (Generalized Pareto Distribution) metrics are extracted from the max-annual-loss distribution.
Step 4 — Premium calculation. Solvency II decomposition with Gaussian Copula correlation-adjusted portfolio losses.
Single-shell approximation. The model represents only the 800 km shell. In reality, debris migrates across altitudes due to atmospheric drag and collision-induced velocity changes. A multi-shell model would capture cross-altitude contamination effects.
Static populations. Both the satellite count (8,000) and debris injection rate are held constant. Dynamic modeling of mega-constellation deployment schedules and retirement would improve realism.
No active debris removal. The model does not include ADR missions, which could shift the debris trajectory below the cascade threshold.
Simplified fragment physics. The NASA Standard Breakup Model provides fragment counts but not directional distributions or size–velocity correlations that would affect collision probabilities.
Multi-shell model with cross-altitude migration; NORAD TLE data integration for empirical debris tracking; dynamic constellation deployment modeling; reinsurance layer pricing (excess-of-loss, stop-loss structures); and sensitivity analysis using Sobol indices to identify the most impactful parameters.