The Agency Problem: When Your Advisor Is Working for Someone Else

There's a transaction happening every time you take advice from a professional. The explicit transaction: you pay for expertise and receive a recommendation that serves your interests. The actual transaction, when incentives are misaligned: you pay for a recommendation that serves the advisor's interests and is plausibly defensible as serving yours.

Taleb calls this the agency problem: the person making decisions on your behalf has interests that diverge from yours. And he makes a specific claim about it: the problem is structural, and regulation doesn't fix it.

The Doctor Case

The doctor case is Taleb's sharpest example, and it's worth sitting with.

A physician recommending a course of treatment is optimizing across multiple variables. Some of them align with your interests: the patient's outcome, their long-term health, their quality of life. Others do not: - Malpractice risk: over-treatment is rarely prosecuted; under-treatment is. - Five-year survival metrics: the metric that institutions use to evaluate physicians, which can be gamed with aggressive short-term intervention that produces poor long-term quality of life. - Time efficiency: recommending the well-established treatment is faster than investigating whether it's optimal for this specific patient.

The result is predictable: physicians systematically err toward more treatment, more intervention, more aggressive protocols — not necessarily because more is better for the patient, but because more is better for the physician's malpractice exposure and institutional metrics.

The patient bears the consequences. They experience the side effects. They go through the recovery. They live in the body that the treatment modifies. The physician moves on to the next patient.

This is the agency problem in medicine: the person making the decision is insulated from the downside of the decision. The person bearing the downside has no professional training to evaluate whether the recommendation is optimal for their specific situation.

The Financial Advisor Case

The financial advisor case works the same way. When you hire a financial advisor, you're told you're getting guidance optimized for your financial goals. What you may actually be getting is:

Product recommendations tied to advisor compensation: Many financial products pay commissions to the advisors who recommend them. The advisor's compensation goes up when you buy the product; your returns are a secondary consideration. This isn't illegal in many jurisdictions. It's the norm.

Complexity as a business model: Simple investment strategies (low-cost index funds, diversified asset allocation, time in market) are available to most people with minimal advisory help. An advisor who recommends simple strategies isn't billing for complexity. The incentive is to recommend products that require ongoing management, ongoing advice, and ongoing fees.

Risk profile inflation: Advisors are rarely fired for recommending products that produce good returns. They are sometimes fired for recommending products that lose money. The incentive is to recommend toward the conservative end of the risk tolerance you'd actually be comfortable with — to protect against the career risk of being blamed for losses.

In each case, the advisor's interests and your interests point in different directions. The recommendation they give reflects the intersection of what serves you and what serves them — weighted toward the latter, because that's what they control.

The Salesman Case

The salesperson case is the most transparent, which makes it the most informative.

A car salesperson is not your advocate. They are on the other side of the transaction. When they tell you this model is a great fit for your needs, they're telling you that because their commission aligns with you buying it — not because they've independently evaluated your needs against all available options.

This is obvious when the transaction is explicit — we know the car salesperson isn't an independent advisor. The agency problem becomes dangerous when the role is framed as advisory but the underlying incentive structure is the same as sales. When the financial advisor presents as your partner in wealth-building while earning commissions on products they sell you. When the doctor presents as your health advocate while optimizing for institutional metrics.

Taleb's Formulation

Taleb condenses this: "Avoid taking advice from someone who gives advice for a living, unless there is a penalty for their advice."

The word "penalty" is doing specific work here. Not accountability in general — the professional stakes their career on their track record, in a general sense. The specific thing that matters is whether a given piece of advice, if it turns out to be wrong in ways that affect you, will cost the advisor something proportional.

Under current arrangements, it typically doesn't. The financial advisor who recommended the product that underperformed will not return fees proportional to your losses. The doctor who recommended the aggressive treatment you didn't need won't compensate you for the recovery time. The consultant who recommended the restructuring that didn't work has already been paid.

Why Regulation Doesn't Fix This

The instinctive response to the agency problem is regulation: require advisors to disclose conflicts, mandate fiduciary standards, impose licensing requirements. Taleb's observation is that these don't solve the structure.

Regulation creates new rent-seeking opportunities for people who are good at navigating regulatory requirements without meeting their spirit. It adds costs that create barriers to entry, which often protect incumbent advisors more than they protect clients. It addresses the symptom of disclosed conflicts without touching the underlying incentive structure.

What addresses the underlying structure is personal liability: making the advisor's compensation contingent on long-term outcomes, requiring them to hold positions alongside you, or creating genuine legal accountability for recommendations that prove harmful through foreseeable negligence.

The mechanism that works is skin in the game. The advisor who profits if you profit and loses if you lose is playing a fundamentally different role than the advisor who collects fees regardless of outcomes.

For the full framework, read Skin in the Game Explained.