ADR-0013 — Financing instruments & the cost-of-debt engine

• Status: ACCEPTED (shape; numbers owed to the sweep)
• Date: 2026-07-12

Context

Flat 25%/turn loans were a day-cadence artifact (ADR-0009) and a placeholder (DQ-8). With DEFER routing through the Ledger (ADR-0009/0012), loan terms and defer carrying-costs need one pricing engine, or the two halves of the flagship drift apart.

Decision

One pricing engine for all liabilities (loans, defers, funding-with-strings), taking these arguments at Pip's weights:

• Org type — major. Set at character/org creation; ties directly to building early-game agency into setup (DQ-4 adjacent; "the opening is a commitment device").
• Counterparty / stakeholder risk & interest — major. Who you owe changes what it costs (VC, government, philanthropist, rival's proxy — counterparty dread lives here, ADR-0007 register).
• Reputation — typed. Safety-rep and finance-rep price debt differently, and reputation affects grants, equity raises, and loans through different channels (ADR-0010 typed attention, applied to capital).
• Hype — scoped to the raise's purpose. Fundraising is purpose-tagged and the hype cycle prices the purpose; no internal earmarked budgets (rejected as fiddly — granularity lives in the raise, not the spend).
• Existing ledger load — minor direct, mostly indirect through reputation and counterparty appetite.

Instruments beyond debt (all Ledger objects):

• Equity (for-profits): early capital comes on good terms but eats equity — and equity is a more valuable bargaining chip mid/late game. Selling cheap early is a priced regret.
• Board seats (nonprofits' equity-approximate): cash for seats that curtail the player's options in voting mechanisms (hard dependency on DQ-7 governance design).
• Research-agenda narrowing: cash for scope — the funder's strings constrain which workstreams are legal.

The early-game curve, stated as design intent: pure philanthropy is starved — the first thing players feel is the difficulty of getting enough cash in the door from any purely philanthropic approach. Trading (equity, agenda, capabilities work) buys speed — reputation and publication pipelines spike faster — and bills against late-game optionality (the capabilities route compromises later government/political influence). Financing is "every mitigation is a loan" applied to itself.

Beacons served / violated

• Rams #6: an honest cost-of-capital function — org type, counterparty, reputation, purpose — instead of a flat magic number.
• MaRo Interaction: capital now interlocks with reputation (typed), hype, governance (board seats), and the rival race.
• The PoE exponential-difficulty ladder gets its mechanism: compounding early trades are the difficulty curve (ADR-0003 thesis, now with a pricing engine).

Interaction contract

Reads/writes: Ledger (all instruments are entries), reputation (typed, both directions), hype cycle (purpose pricing), char/org creation (org type; DQ-4), governance/voting (board seats curtail votes — DQ-7 now has a customer), rival pipeline (funding starvation cascade, ADR-0012).

Rejected alternatives

• Flat rates: day-cadence placeholder, dies with ADR-0009.
• Internal earmarked budgets (money "for" things): fiddly; purpose granularity lives in the raise.
• Separate pricing for loans vs defers: one engine or the flagship forks.

Consequences / open questions

• Numbers owed to the sweep (DQ-8): rate curves per org type/counterparty, equity dilution schedule, board-seat prices. Instrument root-cause death attribution first (ADR-0012) or tuning is blind.
• DQ-7 (governance) is now load-bearing twice over: the board-seat instrument needs the voting mechanics it curtails.
• Equity model granularity (cap table vs a single dilution scalar) — start scalar (Rams #10), upgrade only if bargaining-chip play demands it.
• Character/org setup design (org type as the pricing engine's biggest input) — feeds the early-game agency thread.