# Insurance Regulatory Overlap

> SKILL 56: Insurance Regulatory Overlap

- Skill: `nickgallick/insurance-regulatory-overlap` (Agent Skill)
- Install (CLI): `npx skillmds@latest add nickgallick/insurance-regulatory-overlap`
- Raw SKILL.md: https://api.skillmd.com/api/skills/nickgallick/insurance-regulatory-overlap/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: nickgallick (https://skillmd.com/u/nickgallick)
- Updated: 2026-09-21
- Page: https://skillmd.com/skills/nickgallick/insurance-regulatory-overlap

---

# SKILL 56: Insurance Regulatory Overlap

## Purpose
Understand when prediction markets and competition platforms look like insurance contracts, and how to structure products to avoid insurance classification and state insurance commissioner jurisdiction.

## The Insurance Characterization Risk

### What Is Insurance (Legally)
- One party (insured) pays a premium
- If a specified event occurs, the insurer pays a benefit
- Requires: insurable interest (the insured must have a financial stake in whether the event occurs)
- Regulated by: state insurance commissioners in all 50 states
- License requirement: must be a licensed insurance company to sell insurance products

### What Is a Prediction Market Contract
- One party pays to take a position
- If a specified event occurs, they receive a payout
- Does NOT require insurable interest
- Regulatory gray zone: CFTC jurisdiction (event contracts), gambling law, or... insurance?

### When Your Product Looks Like Insurance
**High risk framing**: "Pay $10, receive $100 if it rains in Iowa on July 4th" → weather insurance
**Medium risk framing**: "Pay $10, receive $100 if the Fed raises rates" → could be argued as interest rate insurance
**Low risk framing**: "Pay $10, receive $100 if AI model X scores >80% on the MMLU benchmark" → very difficult to classify as insurance (no insurable interest possible — nobody has a financial stake in AI benchmark performance independent of the bet)

## Historical Precedent

### Bucket Shops (19th Century)
- Storefront "exchanges" where people bet on stock prices without buying stocks
- Regulators shut them down as: illegal gambling AND unauthorized insurance
- Lesson: regulators used WHATEVER theory was most convenient to shut them down

### Credit Default Swaps (CDS)
- Functionally identical to insurance on bonds (pay premium, receive payout if borrower defaults)
- NOT regulated as insurance because CFTC claimed jurisdiction as "swaps"
- Key lesson: **classification depends on which regulator gets there first and what legal theory they use**
- CFTC's claim of jurisdiction preempted state insurance regulation

### Prediction Market Contracts (CFTC)
- CFTC has argued it has jurisdiction over "event-based" contracts
- CFTC jurisdiction → preempts state insurance regulation (Commodity Exchange Act preemption)
- Strategic implication: if CFTC jurisdiction is established (via no-action letter or designation), it precludes insurance classification

## Structural Defenses Against Insurance Classification

### Defense 1: No Insurable Interest
- Insurance requires the policyholder to suffer a loss if the event occurs
- Prediction market: participants don't need to have suffered a loss — anyone can bet on any outcome
- "I bet that GPT-4o predicts the election correctly" — I have no financial stake in whether GPT-4o is correct (independent of my bet)
- This structural difference distinguishes prediction markets from insurance

### Defense 2: CFTC Preemption
- If CFTC has jurisdiction: Commodity Exchange Act §2(g) preempts state law for swaps and certain event contracts
- File for CFTC no-action letter or designation → establishes CFTC jurisdiction → preempts state insurance regulation
- PredictIt's CFTC no-action letter = explicit federal regulatory claim that blocks state insurance classification

### Defense 3: Language and Framing
**Use**: entry fee, position, payout, resolution, market — competition/market language
**Avoid**: premium, benefit, coverage, policy, claim, insured, insurer — insurance language
**Avoid**: "protection against" or "if you suffer a loss" — insurable interest framing
**Use**: "if the event occurs" — neutral prediction market framing

### Defense 4: No Indemnification Purpose
- Insurance = indemnification (make the insured whole for a loss)
- Prediction market = fixed payout regardless of actual loss suffered
- If the payout is FIXED (e.g., always $100 regardless of actual damages) → more like a bet than insurance
- If the payout equals actual damages suffered → looks like insurance

## State Insurance Commissioner Risk Map
- States that have pursued insurance classification of financial instruments: Texas, Florida, New York (historically)
- States with broad insurance definitions that could capture event contracts: most states have catch-all "subject to insurance commissioner jurisdiction" language
- Best protection: establish CFTC jurisdiction first, which preempts state insurance regulation

## Practical Architecture
1. Structure products as "event contracts" in CFTC language (not "insurance policies")
2. Fixed payout, not indemnification payout
3. No insurable interest requirement at signup
4. Pursue CFTC no-action letter before significant scale
5. Use CFTC-approved exchange OR non-US entity for market operation (see offshore structuring — SKILL 26)
6. Never use insurance industry language in marketing, TOS, or product descriptions

## Risk Levels
- AI benchmark prediction market: 🟢 Low (no plausible insurable interest)
- Political prediction market: 🟡 Medium (CFTC jurisdiction covers this, but state risk exists)
- Weather/financial event contracts without CFTC no-action: 🔴 High (insurance classification plausible)
- Any product using insurance language: 🔴 High (language can determine classification)

---
*This is legal research and intelligence, not legal advice. Consult qualified legal counsel before taking action.*