Why Rules Fail to Stop Bad Actors: From Buffett to Mechanism Design

The article opens with a saying often attributed to Warren Buffett: "We've never succeeded in making a good deal with a bad person." While the exact phrasing does not appear in his public letters, the sentiment aligns with Buffett and Charlie Munger's long-standing emphasis on integrity as the first of three essential qualities (integrity, intelligence, energy). Without integrity, the latter two amplify destructive power.

I. Incomplete Contracts: The Compliance Arms Race

The root cause lies in Incomplete Contract Theory (Nobel laureates Oliver Hart and Oliver Williamson). Because of bounded rationality, uncertainty, and communication friction, no contract can enumerate every future contingency. Rules are static, finite text; bad actors exploit a continuous, evolving spectrum of loopholes.

This triggers a destructive cycle:

Compliance tax on rule-followers: Exponential growth of procedures (e.g., three-layer receipts, five stamps) creates cognitive friction and administrative burden that drives away creative talent.

Compliance arbitrage by bad actors: They ignore the spirit of rules and exploit literal wording and cross-definition gaps. Munger noted that thick compliance manuals become "action guides" for fraudsters, enabling them to use process as a shield.

Result: "Good money is locked up; bad money arbitrages."

II. The Dual Deadlock of Supervision

Deadlock 1: Internal Self-Supervision — The "Quis Custodiet" Paradox

Juvenal's question "Who watches the watchmen?" captures the logical flaw. Public Choice Theory (James Buchanan) holds that bureaucrats are self-interested utility maximizers, not angels. Asking a department to investigate itself creates a conflict of interest: deep probes expose managerial failure and threaten the unit's budget and power. The outcome is almost always performative — token penalties for minor players while systemic networks remain intact.

Deadlock 2: External Supervision — Cost Explosion and Regulatory Capture

Information asymmetry cost black hole: Outsiders lack tacit frontline knowledge. As violations grow more concealed, marginal supervision cost rises exponentially, often exceeding the direct losses from misconduct.

Regulatory capture (George Stigler): Supervisors face social, career, and rent-seeking pressures. Over time they become co-opted, turning from checks into co-conspirators.

Both "moral suasion" and "police model" hit a wall.

III. Paradigm Shift: Mechanism Design and Incentive Compatibility

Mechanism Design Theory (Nobelists Hurwicz, Maskin, Myerson) — "reverse game theory" — argues: don't try to eliminate selfishness; design payoff matrices so that pursuing self-interest automatically achieves the system's goal. This is Incentive Compatibility .

1. British Convict Ships (1793)

Early model: pay per convict boarded in England. Death rates hit 33–40%. Rules and onboard inspectors failed — inspectors were bribed, ocean voyages were unobservable.

Fix: pay only for live, healthy convicts disembarked in Australia . Instantly, each death became lost revenue. Captains voluntarily bought fresh fruit, enforced ventilation, paid surgeons bonuses. Death rates plummeted to 1–2%, with zero-death voyages. No extra inspectors, no new regulations.

2. The Porridge Division Game

Seven people, one divider. Any fixed rule or added monitor leads to collusion. Solution: one cuts, others choose first, divider takes last. The divider's self-interest forces perfectly equal portions. Power to cut is separated from priority to pick; supervision cost drops to zero.

3. Adversarial Justice and SEC Whistleblowers

James Madison (Federalist No. 51): "Ambition must be made to counteract ambition." Adversarial courts give defense and prosecution equal opposing incentives; truth emerges from their clash at lower monitoring cost.

SEC Whistleblower Program: insiders get 10–30% of monetary sanctions. The massive reward shatters internal collusion; perpetrators live in fear that any accomplice may legally cash in by betraying them.

IV. Four Action Principles for Modern Organizations

Front-load screening cost. Buffett/Munger: never enter deep contracts with flawed character. Invest in rigorous background checks and stress-test character before hiring; it's far cheaper than post-hire compliance departments.

Break the "player-referee" loop. Physically separate execution authority from control authority. Design offsetting KPIs (e.g., loan officers measured on long-cycle bad-debt penetration, not just volume). Remove the discretionary-benefit link.

Internalize supervision benefits. Don't just hire more auditors. Create adversarial stakeholders who profit from finding flaws. Example: Security Response Centers (SRC) and bug bounties turn white-hat hackers into tireless, self-motivated defenders.

Guard against Goodhart's Law / Cobra Effect. When a metric becomes a target, it ceases to be a good metric. British cobra bounty led to cobra farming. In organizations, watch for gaming: e.g., chasing short-term delivery at expense of maintainability, or refusing all loans to keep bad-debt rate near zero.

Conclusion

Rules cannot cage the unprincipled; rule-text records past errors, while human ingenuity for gain evolves without limit. Effective governance does not turn good people into form-filling machines nor expect saints. It acknowledges human frailty, weighs incentives, and uses surgical logic to restructure the game — making betrayal unbearably costly and integrity the rational, dominant strategy. Only when compliance and good faith become the equilibrium choice does durable order emerge.