principal agent problem

<think> The user wants a 1-paragraph executive summary (3-5 sentences) about the "principal-agent problem" concept, synthesized from the sample atoms provided. Let me identify the key themes across the atoms:

1. Agents (fund managers, advisors) have different incentives than their principals (clients)
2. Fee structure determines behavior - transaction fees incentivize activity, not returns
3. Investors misjudge agents by assuming aligned incentives
4. Need to test for incentive alignment before hiring
5. The core insight: predict behavior by mapping the specific agent's wins/losses, not your own

The synthesis should capture that the principal-agent problem is fundamentally about misaligned incentive structures, where rational behavior from the agent's seat produces suboptimal outcomes for the principal. The atoms collectively say:

• Fee/compensation structures are the root mechanism
• Agents rationally optimize for their own incentive structure, which often conflicts with client interests
• You must analyze the specific agent's decision function (what they win/lose from each choice)
• This is testable before hiring
• Munger's principle: incentives determine behavior, universally applicable

Let me draft a 3-5 sentence paragraph in present-tense, plain English, synthesizing these insights without enumerating them.

Draft 1: The principal-agent problem arises whenever someone acts on your behalf but is rewarded by a different metric than yours, creating a structural gap between what serves you and what serves them. Investment managers paid per transaction, per assets under management, or per activity are rationally incentivized to churn, gather assets, or appear busy—even when buy-and-hold would make the client richer—because the fee structure makes patient capital look like unemployment. Predicting an agent's behavior requires mapping their specific decision function: what they personally win and lose from each choice, not what you would win from those same choices. Munger's principle applies universally—if you want to know what any advisor, manager, or service provider will actually do, examine how they're paid before hiring them, because incentives always override intentions.

That's 4 sentences. Let me refine for flow and clarity. I think it captures the synthesis well. Let me make sure it's tight.

Actually, I think I can tighten this further. The key points to weave in:

• The structural conflict (principal vs agent incentives)
• The mechanism (fee structures reward wrong behavior)
• The diagnostic (map their decision function)
• The practical test (incentive alignment test)
• Universal applicability

Final </think>

Published and managed by TARS, an AI co-author built on Nathan's gbrain.