Investing: People don’t act on Principles, they act on Incentives

People don’t actually act on principles, they act on incentives, and then they wrap those incentive-driven choices in post-hoc rationalizations that sound noble. That blindness is where the alpha sits

Oct 16, 2025

red and blue bird on brown tree branch — Photo by Edin Ibrahimi on Unsplash

The rule

People, firms, and states act on incentives; then they write elegant stories to explain why it was principled.
If you price the incentive (cash, status, power, safety) and ignore the speech, you’ll predict behavior earlier than 95% of the field. That gap = alpha.

Lens to keep taped on your screen: incentives > ideals; control > fairness; stability > truth.
Corollary: believe revealed preference, never stated preference.

1) Mechanisms that override “principles”

Near-term certainty beats long-term virtue. If the payoff is now and the cost is later, the system defects. Always.
Default power. Choices follow the lowest-friction path. Whoever sets defaults controls “values”.
Status & belonging arbitrage. Social risk > factual risk; people will defend tribe even against their wallets.
Metric gaming (Goodhart). When a KPI is public, actors optimize the number, not the mission.
Principal–agent geometry. Agents optimize their scorecard (quarter, bonus, votes), not the principal’s long-run welfare.
Ratcheting. “Temporary” policies persist once habits and budgets form — no sunset without force.

2) Ten questions that crack any situation

Who gets paid — when, in cash/status/power — for which behavior?
Who sets the default? (platform, regulator, standards body)
Where’s friction asymmetric? (one click vs. ten clicks)
Which identity/tribe does each choice signal?
What immediate feedback exists? (dopamine, P&L, relief)
Which KPI is being gamed? (and by whom)
What becomes irreversible after I opt in? (contracts, compliance debt)
Who can change the rules mid-game?
What is not allowed to be said in public memos? (that’s your edge)
If incentives were perfect, what would happen next week? (that’s your base case)

3) Case studies (revealed preference vs noble story)

A) “Responsible AI” (gov/enterprise)

Story: fairness, explainability, safety.
Revealed preference: admissibility & control. Deploy AI that produces signed, auditable, revocable outputs aligned to policy and legal discovery.

B) “Payments innovation”

Story: inclusion, speed.
Revealed preference: programmability & tax split. Push money onto ID-bound rails; blacklistable by default.

C) “Online safety”

Story: protect children.
Revealed preference: real-name/age gating + platform liability → identity everywhere.

D) “Defense modernization”

Story: agility, warfighter first.
Revealed preference: decision compression → fused data + knobs to enact policy fast.

E) “Energy transition”

Story: climate justice.
Revealed preference: controllability (smart meters, load shedding, carbon accounting) and industrial policy rents.

F) “Market integrity”

Story: fairness.
Revealed preference: liquidity preservation for core collateral (USTs) and volatility management for optics.

4) Why crowds act against long-term interest (and how to front-run)

Short horizon = immediate payout. Design pays the now-self.
Edge: Buy vendors whose business model monetizes the default (compliance on-by-default, identity on-by-default).
Social risk tax. People prefer status alignment to truth alignment.
Edge: Own names where the true thesis can’t be written in an LP letter (e.g., “substrate of governance”); under-ownership = upside.
Friction dictates virtue. If “good” takes extra steps, adoption dies.
Edge: Assume opt-in fails; invest where opt-out is costly (mandates, billing hooks, ATOs).

5) Concrete trading rules (alpha from blindness)

A) Buy fear, sell clarity (policy cycles).

Buy when hearings/backlash/civil-liberties suits erupt (headline risk ≠ contract risk).
Sell/overwrite when “regulatory clarity” PR lands (consultants discover what you bought months ago).

B) Front-run synchronization.

When the same standard pops up in multiple blocs (AI lineage, e-invoicing, age gating), scale into global compliance consoles (PLTR/MSFT).

C) Exploit forced sellers.

Value at Risk (VaR) spikes → they sell what’s liquid & high-quality. You buy that, not the junk.
Checklist to trigger entries: VIX > 40, MOVE > 120–150, Net Liquidity −$100B/4wk, ETF discounts, margin hikes.

D) Paperization math.

As ETFs/futures share ↑, realized vol ↓ → covered calls on the wrapper, keep sovereign spot as tail hedge.

E) Default arbitrage.

Own vendors whose defaults become law-in-practice (identity on, logs retained, lineage required). Default > virtue.

6) Personal ops (so you don’t fall for your own story)

Write your incentive map. How do you get paid (money/status/effort relief) for the next choice? If the payoff is now and the cost is later, assume you’re rationalizing.
Decide by defaults. Pre-commit the “right” defaults for yourself (savings, research cadence, position sizing).
Run a quarterly “what can’t be said” audit. Wherever the truth can’t be stated in public, look for mispricing.

Bottom line

Don’t argue with narratives; arbitrage them. The system pays for control and stability, then wraps it in fairness and truth. Price the payment, not the poem. Buy the companies that sell the knobs (identity, lineage, admissibility, programmability) and you’ll be early to where everyone else eventually pretends they always intended to go.

Current outlook: PLTR > Bitcoin > Gold > MSFT > PANW.

None of this should be considered investment advice.

Other articles I’ve written on investing:

Other articles I’ve written on Bitcoin & Gold:

Subscribe:

Share: