Interactive Pricing Model

Methodology & Assumptions

This model prices one cash flow — merchant fees funding AMP buybacks for stakers — using the whitepaper's own equilibrium: market cap settles where staking the reward pool earns a normal required return, with a forward-looking premium during hypergrowth and a collateral-capacity floor underneath. The Base case is a defensible central estimate (~1.5% of US card volume by Year 20); Bull is the win-big case (10%). It does not predict market sentiment or speculative trading; it estimates fundamental value only.

1. One Engine, Not Three. Prior versions blended three models (CLW, ARMA, DCF). An independent five-model review found all three capitalize the same cash flow — merchant fees funding AMP buybacks — through different lenses, so the blend was cosmetic. v15 prices that single cash flow directly: GMV(t) × fee(t) × 90% = the annual reward pool R(t). The buyback mechanism itself is patented: converting merchant fees into collateral-token rewards for stakers is claim 1 of US 12,393,932.

2. Valuation — Yield Parity (Whitepaper §5.3). The whitepaper's own equilibrium: at maturity, staking yields approach parity with the required rate. So market cap settles at R(t) ÷ r_req(t), where the required yield declines from 10% (early crypto risk premium) to 6% (hurdle + residual spread) — a terminal multiple of ~15× fee revenue, inside the 8–15× range independent reviewers deemed defensible. Early years add a forward-looking growth premium: the market prices the reward run-rate three years ahead, discounted at r_req: MCap = MAX(R_t, R_t+3/(1+r)³) ÷ r. The premium decays to 1× automatically once reward growth falls below the required yield — no hand-tuned β schedules or model weights. Optionally, a float-scarcity premium (the κ slider, default off) adds a flow-impact term: the buyback must execute against the unstaked float, so price × [1 + κ · buyback ÷ (float share × MCap)] — an inelastic-markets liquidity effect (κ ≈ 3–8 in equity studies), presented as an explicit lever rather than a hidden assumption because anticipated flows are, in an efficient market, already capitalized in the yield-parity value.

3. Capacity Floor (Whitepaper §3.2). Staked AMP must cover in-flight payments. Required staked value = Σ_flows GMV ÷ 525,600 min/yr × hold minutes × 3 (peak factor) × 1.05, where the 5% buffer implements the patented sizing rule — the lock equals the hold plus the asset's estimated price fluctuation over the window (US 12,443,962, worked examples up to ±100–300% for volatile assets on multi-day holds). Price = MAX(MCap ÷ supply, floor). Finding: the floor never binds at any modeled scale — capacity is never the constraint; adoption is.

4. Merchant Engine — GMV Scenarios (re-based to base rates). Smooth decaying-growth schedules (v17): ln GMV(t) = ln G1 + ln(G20/G1) · (1−ρ^t−1)/(1−ρ¹⁹) — each year's growth is ρ× the prior year's (default ρ = 0.88, fitted through the retired $50B Year-9 milestone, so the curve passes the same checkpoints without kinks). Base: $250M → $470B (Yr20) ≈ 1.5% of projected US card volume, including cross-border corridors — set from adoption base rates (Square ≈2% of US card volume after 14 years; all crypto processors combined <0.1% after 15). Bull is the pre-v15 "Base": 10% US share ($3.14T, ~AmEx scale) — a defensible dream at <1% base-rate probability, not a central case. Bear: 0.3% ($95B). Sliders set the Base endpoints and shape; Bear/Bull scale from them. These are modeling anchors, not forecasts.

4b. B2B Corridor Engine (v16) — The Second Lever. Cross-border B2B payouts are a different business with a different TAM: the global B2B cross-border pool is ~$90T today (Europe ~$20T) versus $10.8T of US card volume — so v16 models it as a second, additiveGMV engine with its own anchors and its own thin, compressing fee (0.30% → 0.18%, competing with Wise, Airwallex, and bank FX rather than card interchange). Base case: $200B by Year 20 ≈ 0.13% of the pool — a Payoneer/Wise-class outcome (those firms move ~$80–180B/yr after 15+ years); Bull ≈ 0.5% ($750B); Bear stays near zero (Flexa has disclosed no payout volume to date). The Engine toggle above shows each engine alone or combined. The honest headline: B2B is a volume monster but a fee dwarf — $1B of corridor volume earns ~2.5× less than $1B of merchant volume, so even a Wise-scale win adds ~17% to the Year-20 price, not a multiple.

5. Merchant Fee Economics — Compression. GMV is split across three flows whose mix shifts over 20 years (POS retail 80%→50%, cross-border payouts 15%→35%, B2B 5%→15%), each with a declining fee curve — blended 0.655% (Yr1) → 0.455% (Yr20), the way every real payment network's take-rate behaves at scale. Patent support is structural, not numeric: merchants pay different, confidential rates, with established merchants incentivized by lower fees (US 12,380,434; its illustrative examples are 1% and 2%). Buyback allocation: 90% of fees (the whitepaper describes the entirety; 90% is a conservative haircut — the fraction is a business choice, not a protocol constant).

6. Supply. 100% of the 99,720,005,508 AMP contract cap is treated as circulating from Year 1, per Flexa cofounder & CEO Trevor Filter (X Space, July 10, 2026): aggregator circulating figures aren't reliably maintained and a fully circulating supply is the safest assumption.

7. Staking (diagnostic). S-curve from the observed 23.8% (23.69B staked ÷ 99.72B) toward 50%. With κ = 0, prices do not depend on the staking rate — it moves only the capacity floor and the displayed staker yield. With κ > 0, the long-run staked share sets the float the buyback bids against, and staking becomes a price lever. The implied staker yield runs ~12–15% mid-trajectory (stakers earn more than the required rate because only stakers receive rewards), converging toward parity at maturity, exactly as §5.3 predicts.

8. Continuity Check. Year 1 (merchant launch; ~$300.0M total GMV incl. B2B at the settings above) implies a ~$167.5M market cap (today's live market cap is unavailable right now). The model does not claim today's price is wrong. Earlier versions did: they implied AMP should already trade 2–3× higher, which made their projections impossible to test.

9. Market Context. US Card TAM: $10.77T (Year 0), growing 5.5%/yr per Nilson Report. Terminal MCap/GMV at Base ≈ 0.068× — a modest premium to Visa's ~0.04× MCap-to-volume, versus the 0.27× the previous model carried.

10. What This Model Ignores. Speculation, crypto cycles, exchange liquidity, competitive response, regulation, and any Flexa decision to change the buyback. It prices the flywheel mechanism only. All patent citations verified against grant PDFs (July 2026); every claim in Flexa's portfolio names a generic collateral token — AMP is the implementation, not a legal requirement.

Note: This model operates in a vacuum. It projects price based solely on the flywheel mechanism (GMV, fee revenue, buyback, staking dynamics). Real-world price will also be affected by external market forces: speculative buying/selling, broader crypto market cycles, macroeconomic conditions, and exchange liquidity, none of which are captured here.

This model is for educational and research purposes only and does not constitute financial advice. All revenue estimates are unverified projections. The model implements the theoretical framework of the Amp Whitepaper (2020) with the assumptions enumerated above. Source: "Amp: Digital Collateral for Real-World Transfers" (2020).