Poor Charlie's Almanack Ch. 3: Incentives — The Model Munger Ranked Above All Others

阅读中文版

Munger said he had underestimated the power of incentives his entire life, despite ranking it highly. The model explains most financial industry behavior without requiring anyone to be dishonest.

🔊 Listen to Article (Chinese Audio)

Poor Charlie's Almanack Ch. 3: Incentives — The Model Munger Ranked Above All Others

Investment Background

Munger's line about incentives deserves more serious attention than most of his aphorisms:

"I think I've been in the top five percent of my age cohort all my life in understanding the power of incentives, and all my life I've underestimated it."

Note the structure of that sentence. He is not saying "incentives matter" — that is a platitude everyone agrees with.

He is saying: even knowing it matters, I still persistently underestimated it.

That is a claim about how we systematically misjudge others' behavior, and it has concrete operational consequences.

The Wall Street Translation

The FedEx Story

Munger tells one example repeatedly, because it displays the mechanism so cleanly.

FedEx's entire business model depends on every package being sorted at the central hub each night. If sorting runs late, the next day's on-time delivery promise fails.

The company could not solve the problem for a long time. Night shift workers moved slowly; sorting ran late repeatedly.

Management tried everything: stricter supervision, better training, motivational speeches, more management. None of it worked.

The eventual solution changed one thing: night shift pay moved from hourly to per-shift — finish the sort and go home.

The problem disappeared immediately.

Note the key point: the workers were neither lazy nor dishonest.

Under hourly pay, finishing quickly meant earning less. Their behavior was entirely rational.

Every one of management's solutions failed because they treated it as an attitude problem when it was an incentive problem.

What This Means in Finance

This model's value is that it lets you predict people's behavior without judging their character.

The core question is always the same:

Under what circumstances does this person make money? Under what circumstances do they lose it?

Applied to roles you will actually encounter:

Role How they are paid So they tend to
Commission-based financial advisor Commission on products you buy Recommend high-commission products rather than necessarily the most suitable ones
Assets-under-management advisor A percentage of your assets Not suggest paying off the mortgage or buying an annuity (both shrink the assets they manage)
Mutual fund manager A percentage of assets managed Grow assets, even when size hurts returns
Hedge fund manager The option-shaped two-and-twenty Raise volatility (see Chapter 3 of When Genius Failed)
Financial media Advertising and clicks Manufacture urgency and the urge to trade
Brokers Trading commissions or order flow Encourage frequent trading

One thing must be stated with total clarity: this table does not accuse anyone of dishonesty.

That is precisely where the model's power lies. Every person in that table may be entirely upright and genuinely wanting to help you.

And they will still lean systematically toward the recommendations that pay them — because when judging in a gray area, people unconsciously favor the interpretation that benefits them.

This is what Munger means by having "underestimated the power of incentives all my life": we tend to believe an honest person will overcome their conflict of interest. The evidence suggests that overcoming is far harder than we imagine.

A Necessary Qualification

This model can be overused, and overuse produces cynicism.

Not all behavior is explained by incentives. People also act from professional standards, reputational concerns, empathy, and long-term relationships. Someone explaining everything through incentives will mispredict a great deal of behavior.

The correct use is as the first question, not the only one.

First ask "where do the incentives point," then ask "what forces offset them" — professional licensing, fiduciary duty, reputation mechanisms, long-term client relationships.

In the United States, the legal concept of fiduciary duty exists precisely to counter this problem. An advisor bound by it must legally place your interests ahead of their own. That is not a perfect solution, but it is a genuine offsetting force.

Relationship to Other Books in This Library

Boundaries are needed, as this sits close to several existing books.

Book What it says about incentives
When Genius Failed Ch. 3 How incentives pushed one specific institution toward excess leverage — a case study
Principles (Dalio) How to design an organization's incentives — from a manager's perspective
The Essays of Warren Buffett Alignment of management incentives with shareholders — as an investment criterion
This book Incentives as a general predictive tool — for anticipating the behavior of anyone you meet

Chapter 3 of When Genius Failed is one deep application of this model; this chapter is the model itself.

Executable Trading Rules

  1. Ask everyone who gives you financial advice one question: "How are you paid?" The chapter's most important and most easily executed line. In the United States you have every right to ask directly, and someone unwilling to answer clearly has already answered you.

  2. Before any major financial decision, map every party's incentives. Concretely: list each party to the transaction and write down when they make money. Wherever a party's interest runs directly counter to yours is where you need extra scrutiny.

  3. Be wary of "free" advice. Free financial advice is nearly always subsidized by the sale of some product. That does not make the advice bad, but it means you need to know where the subsidy comes from. Hourly or flat-fee advisors have the cleanest incentive structure.

  4. Apply this model to yourself. The hardest and most valuable line. Ask: what does my current position incline me to believe? Someone who just bought a stock looks for supporting information. Someone about to retire wants to believe markets will cooperate. Your own incentives distort your judgment too.

  5. Do not use this model as a license for cynicism. See the qualification above. The goal is predicting behavior accurately, not assuming everyone is deceiving you. The latter causes you to miss genuinely valuable professional advice.

Relevance to a Retirement Portfolio

This chapter has a particularly important application for retirement investors, because retirement is exactly the phase the financial industry markets to most aggressively.

Someone newly retired holding a substantial rollover balance is among the industry's most valuable customers.

At that moment you will receive a great deal of advice. And this chapter's framework says to ask every piece of it the same question.

Some concrete scenarios:

Scenario one: someone recommends rolling your 401(k) into an account they manage. Ask: does their compensation depend on you doing this? If so, that does not make the advice wrong, but you need to independently verify the cost comparison against leaving it in the existing plan.

Scenario two: someone sells you a "principal protected" annuity. Ask: what is the commission? Some annuity products carry substantial upfront commissions. Annuities are appropriate tools in specific circumstances — particularly against longevity risk — but their sales incentives are strong, so they warrant extra scrutiny.

Scenario three: someone recommends a complex strategy because "your asset level requires more sophisticated management." Ask: who benefits from the complexity? Complex strategies typically charge more, and Chapter 5 of A Random Walk's data shows cost is the most reliable negative predictor of returns.

This chapter's relationship to our position is direct:

We recommend low-cost index funds partly because of this model — it is the arrangement with the smallest incentive conflict.

A broad index fund provider has almost no way to increase its revenue by harming your interests. Its fee is public, fixed, extremely low, and it requires no frequent trading.

That structural cleanliness is itself an advantage, and it does not depend on anyone's character.

Which is Munger-style thinking: do not ask "who can be trusted." Ask "what structure requires no trust."

Chapter 4 takes on the most systematic part of this book: Munger's catalogue of the psychology of human misjudgment, and how the tendencies combine into disasters.