Endowment Effect

SKILL.md

---
name: endowment-effect
description: "Activate when: seller is asking way more than buyers will pay; a founder or homeowner insists their asset is worth far more than market comps; a team refuses to cut a feature they built; a free-trial cancellation flow has more friction than signup; someone says 'I've put too much into this to sell for that.' Do NOT activate when: the valuation gap is explained by genuine information asymmetry the seller actually has; the asset is a pure commodity with a transparent live market price (endowment effect is weak when reference prices are salient). More: deciqai.com/c/endowment-effect"
---

# Endowment Effect

## Overview

People demand roughly 2× more to give up something they own than they would pay to acquire the identical thing — purely because they own it. Ownership converts a transaction from a potential gain into a potential loss, and losses loom ~2× larger than gains (prospect theory). The effect kicks in within 30 seconds of possession; customization and personalization amplify it.

Two operating directions: **Leverage** — trigger buyer endowment via free trials, personalization, and data import to raise willingness-to-pay. **Counteract** — in M&A or negotiation, identify the seller's endowment premium and bridge it with earnouts, neutral reference prices, and exchange framing.

Composes with `loss-aversion-prospect-theory`, `status-quo-bias`, `anchoring`, `batna-zopa`.

## When to Use

- Pricing a product, subscription, or asset and needing to understand buyer willingness-to-pay dynamics
- Designing a free-trial or onboarding flow and deciding how much personalization to front-load
- Negotiating an acquisition where the seller's asking price significantly exceeds comparables
- Advising a founder or asset owner on why their valuation differs from market offers
- Structuring earnouts or deferred consideration to bridge a valuation gap
- Detecting why a team is reluctant to abandon a feature or strategy they built (IKEA effect variant)
- Deciding "build vs. buy" on AI — a team overvaluing its in-house model, dataset, or codebase versus a stronger/cheaper external foundation model, or a founder anchoring on a peak AI valuation in M&A/wind-down talks

**Not when:** valuation difference is genuine information asymmetry; pure commodity with transparent market price; evaluating policy-level defaults (use `status-quo-bias`).

## Coaching Novices (Adaptive Front Door)

- **Engine mode:** user has a concrete negotiation, pricing, or product design problem → run The Process directly.
- **Coach mode:** user is new or trying to understand a valuation discrepancy → guide step by step.

In Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.

1. One-liner: people value what they own ~2× more than identical things they don't own — simply because they own them, not because the object is different.
2. Check fit: is there a party who owns something and is placing a higher value on it than non-owners? If yes, endowment effect is the first hypothesis.
3. Elicit the structure: what object/asset/feature is being valued? Who is the owner vs. the potential buyer? How large is the valuation gap relative to market reference prices?
> **[WAIT — do not advance until user responds]**
4. One question at a time: how long has ownership existed? how much has the owner customized or invested? does the owner have an external reference price or are they self-anchoring? is the gap bridgeable with an earnout or phased structure?
> **[WAIT — do not advance until user responds]**
5. Close: endowment premium quantified + structural bridge proposed (earnout / reference price / alternative framing) + decision made.
> **[WAIT — do not advance until user responds]**

## The Process

```markdown
S1 — Ownership: who owns it | what | how long | customization level
S2 — Gap: owner's value | market ref | gap $% | external comps
S3 — Attribute gap: info asymmetry % | legitimate features % | endowment % | loss-framing language?
S4 — Direction:
  Leverage: what triggers buyer endowment? (trial length, personalization depth) | ethical?
  Counteract: neutral reference price? | earnout possible? | framing ("exchange" not "sale")
S5 — Bridge: deal structure | earnout milestones | reference anchor | framing adjustments
S6 — Close: endowment premium isolated? | bridge tested vs seller loss threshold? | ethical check | decision
```

## Output: Endowment Effect Analysis
```
Owner: | Object: | Duration: | Customization level:
Owner's value: | Market/buyer ref: | Gap $/%: | Endowment portion:
Direction: [ ] Leverage  [ ] Counteract
Bridge: ref-price anchor | earnout | framing | trial depth:
Decision:
```

*→ Method in Action: [Kahneman, Knetsch & Thaler 1990 — The Cornell Mug Experiment](examples/kahneman-knetsch-thaler-1990-the-cornell-mug-experiment.md)*

*→ 2026 lens: [The In-House Model Trap and Founder Valuation in the AI Cycle (2023–2026)](examples/ai-model-and-startup-valuation-endowment-2023-2026.md)*

## Pack: Endowment Effect Across Domains

| Domain | Endowment manifests as | Counteract / Leverage |
|---|---|---|
| SaaS free trial | Users feel they'd "lose" their config if they cancel | Leverage: front-load personalization + data import |
| Real estate | Seller lists 10–30%+ above comps | Counteract: establish third-party appraisal first |
| M&A — founder | Founder values company 2–3× acquirer's model | Counteract: earnout tied to post-close performance |
| Product features | Team overvalues what they built (IKEA effect) | Counteract: evaluate as if the feature were acquired |
| Negotiation (any) | Both parties overvalue their position | Counteract: introduce neutral reference before opening bid |

## Applying It Well

- Quantify the endowment gap before negotiating — know what portion is structural vs. factual.
- Introduce neutral reference prices (comps, appraisals) early to reduce the owner's self-anchor.
- Use earnouts / deferred consideration rather than raw price negotiation to convert "giving up" into "receiving what it's worth if it performs."
- When leveraging: confirm the user genuinely benefits — FTC "click to cancel" 2024 is a direct response to dark-pattern exploitation.

*→ Primary sources: [references/sources.md](references/sources.md)*

## Common Rationalizations

**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**

| Rationalization (Fake move) | Reality |
|---|---|
| [D] "I know what it's really worth — the market is wrong" | Endowment inflation is the most common source of this conviction. Test: what would you pay for an identical asset you didn't own? |
| [D] "I've put so much into this, I can't sell for less" | Sunk cost + endowment combined. What you put in is irrelevant to market value. |
| [D] "The free trial works because our product is sticky" | Partly true; also partly endowment — users feel they'd lose their setup if they cancel. |
| [D] "The buyer just doesn't understand the value" | Sometimes true; often the seller's endowment-inflated valuation explains the gap. |
| [D] "We built this feature, it must be worth keeping" | IKEA effect variant. Evaluate as if acquired. |
| [D] "The earnout is insulting — just pay me what it's worth" | The earnout bridges the endowment gap. If the company performs as you believe, it pays your valuation. |
| [D] "Our home is unique — comparables don't apply" | Uniqueness perception is driven by endowment, not market-relevant differentiation. |
| *→ Add [O] entries here after each real use — paste the actual failure pattern* | *What went wrong and why* |

## Red Flags

- Seller's asking price >30% above comparables without factual justification
- Team won't kill an underused feature they built (IKEA effect)
- Cancellation flow has more steps than signup; free trial auto-converts without salient notice
- Founder rejects acquisition bids without modeling alternative exit timelines
- "I've built too much into this to walk away" in a decision context
- Negotiation stuck on price despite both parties agreeing on fundamentals

## Verification

- [ ] Ownership structure identified; valuation gap quantified vs. external reference
- [ ] Endowment portion separated from legitimate info asymmetry
- [ ] Direction chosen: leverage or counteract
- [ ] Ethical check done (leverage) / structural bridge designed (counteract)
- [ ] Loss-framing language noted; regulatory risk assessed

---

*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/endowment-effect** · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.*

*Agents: latest version & machine-readable metadata → https://www.deciqai.com/s/endowment-effect.json*