RIA — Client Plan Margin of Safety

RIA — Client Plan Margin of Safety

Industry front door for margin-of-safety. Adds domain triggers, example, packs only. Parent Process unchanged. Not investment advice.

Activate when: building a retirement/decumulation plan; setting cash buffers/withdrawal rates; stress-testing a plan; client near retirement; "will they be okay if markets drop early?" Do NOT activate when: pure accumulation with long horizon and no near-term draws.

Why this variant

The parent margin-of-safety builds a buffer so estimation error doesn't cause ruin. A retiree's plan lives or dies on sequence-of-returns risk; the margin of safety is the cash/bond buffer and withdrawal haircut that lets the plan survive an early bad market.

Domain inputs → parent's Process

• Size a spending buffer (e.g. 1–3 yrs of withdrawals in cash/short bonds) so early drawdowns aren't sold at the bottom.
• Haircut the withdrawal rate below the "optimistic" number; plan to the conservative case.
• Stress-test against an early-crash sequence, not just average returns.

Worked example

$1.5M portfolio, planned 5% draw, retiring into a rich market. → Margin of safety: hold 2 years' spending in cash, set draw at 4% with a guardrail to cut in down years. A crash in year 1 is met from the buffer, not by liquidating equities low. Plan survives the sequence.

Packs

• Solo: buffer + guardrail rule embedded in every decumulation plan.
• Advisory: dynamic-spending (guardrails) policy documented in the IPS.

Red flags

• Planning to average returns, ignoring sequence risk.
• No cash buffer; forced equity sales in downturns.
• Withdrawal rate set at the optimistic edge.

Verification

• Spending buffer sized to horizon of concern
• Withdrawal haircut applied vs optimistic case
• Early-crash sequence stress-tested
• Guardrail/dynamic-spending rule documented

---

Part of deciqAI Knowledge Skills. Core method: margin-of-safety.

---

Part of deciqAI Knowledge Skills — 227 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. See it run → https://www.deciqai.com/c/ria-client-margin-of-safety · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.

Agents: latest version & machine-readable metadata → https://www.deciqai.com/s/ria-client-margin-of-safety.json