Leadership
10
min read

Leaders Inherit Responsibility for the Risks Their Incentives Make Rational

Marvin V. Acuna
Marvin V. Acuna
Published
November 16, 2026

Leaders Inherit Responsibility for the Risks Their Incentives Make Rational

A leader does not have to order a dangerous decision to be responsible for making it reasonable.

The familiar defense is that the behavior was never authorized. No one instructed the team to conceal the delay, compromise the standard, defer the maintenance, overstate the forecast, or transfer an obligation to someone unable to challenge it. The conduct appears to belong to the person who chose it.

That account is incomplete. A decision can be unauthorized and still be institutionally commissioned.

Incentives commission behavior without specifying it. They establish what counts as success, who absorbs the cost of failure, how quickly results must appear, which problems can be disclosed safely, and what happens to the person who refuses a bad trade. The resulting action may never appear in a directive. It can still be the most defensible response to the conditions leadership created.

I think this is where leadership accountability is often drawn too narrowly. We tend to associate responsibility with instruction, approval, or direct knowledge. Those forms of involvement matter, but they are not the only ways authority shapes conduct. Leaders also determine the practical logic within which other people make decisions.

An incentive is not merely a bonus plan. It is any recurring condition that changes the cost of saying yes, no, now, or later.

Compensation matters, but so do promotions, access, attention, deadlines, exceptions, praise, tolerance, budget protection, and the treatment of bad news. The most consequential incentive may be what the organization makes expensive to refuse. A standard can be formally mandatory while violating it remains cheaper than missing the target it obstructs.

That creates a particular kind of risk: conduct that is locally rational and systemically dangerous.

A commercial team can improve its result by making concessions that another function must fulfill. An operating group can protect this quarter by postponing work whose consequences will surface later. A senior executive can preserve confidence by suppressing information that would force an immediate revision. In each case, the decision may make sense from the position of the person taking it. The wider organization inherits the liability.

The local decision is rational because the local actor receives a concentrated benefit while the downside is delayed, distributed, or assigned elsewhere. The system becomes vulnerable not because people have stopped thinking, but because they are thinking clearly about the rewards and penalties directly in front of them.

This is more unsettling than the claim that incentives sometimes produce unintended behavior. “Unintended” suggests an unfortunate deviation from an otherwise sound design. Often the behavior is not a deviation at all. The eventual damage was unwanted, but the conduct that produced it was favored at every meaningful decision point.

Unintended is a description of desire, not responsibility.

A leader can sincerely oppose an outcome while continuing to reward the choices most likely to produce it. That contradiction is easy to miss because stated values and operational incentives perform different functions. Values describe what the organization would like people to honor. Incentives decide what survives when those values compete with a deadline, a target, a career consequence, or the expectations of someone with more authority.

The conflict rarely appears as a choice between right and wrong. It appears as a choice between two legitimate obligations that the system has made incompatible. Protect the standard, but do not miss the date. Report the risk, but do not destabilize confidence. Exercise judgment, but do not depart from the forecast. Challenge the plan, but remain known as someone who can execute it.

Leadership has already intervened by deciding which side of that conflict carries the immediate penalty.

This does not remove individual responsibility. People retain agency, and incentives do not turn every act into an unavoidable response. Deliberate deception remains deliberate. Recklessness does not become prudent because a reward was available. Systemic responsibility should not be used to excuse the person closest to the act.

But individual responsibility should not become a shield for the people who established the conditions.

Both forms of responsibility can exist at once. The person who takes an improper risk may be accountable for the choice. The leader may also be accountable for making that choice advantageous, repeatable, and easy to defend in the language of performance. One responsibility does not cancel the other.

The distinction matters because organizations often respond to failure by isolating the final actor. The individual is removed, the rule is restated, and the values are reaffirmed. Yet the same target remains, the same time pressure remains, the same asymmetry of consequences remains, and the same people continue to bear the cost of refusing.

Under those conditions, replacing the actor changes the name attached to the risk, not the logic producing it.

Ethics language cannot compensate for contradictory incentives. Nor can a demand for courage. Asking someone to resist a rational pressure while preserving the penalty for resisting it is not a serious distribution of responsibility. It asks the individual to finance leadership’s contradiction with personal standing, compensation, or career momentum.

This becomes especially visible around bad news. Leaders may say they want problems surfaced early. But if the person who raises a concern immediately loses control of the work, absorbs reputational damage, or becomes identified with the delay, disclosure carries a certain personal cost. The organizational benefit remains uncertain and shared.

Silence has been subsidized.

The absence of dissent can then be misread as confidence. Agreement appears stronger than it is because the incentive system has filtered out the people most likely to introduce inconvenient information. What looks like alignment may be a rational response to the price of candor.

That is why leaders cannot evaluate incentive design only by examining what people are rewarded for achieving. They must also understand who pays when someone protects the organization from its own target.

A credible prohibition requires more than a rule. Someone with sufficient authority must be willing to absorb the consequences of honoring it. If stopping unsafe work leads predictably to missed commitments, reduced revenue, or public embarrassment, leadership must own those consequences. Otherwise, the organization has not prohibited the risk. It has merely transferred the cost of preventing it to the person least able to carry it.

The issue is not that every metric is corrupting or every target dangerous. Consequential work requires commitments, pressure, and judgment under constraint. No incentive system can anticipate every interpretation or prevent every misuse. Leadership accountability cannot mean omniscience.

The more useful standard is whether a class of behavior was made predictably attractive.

A leader may not foresee the exact decision, the precise person who will make it, or the moment when the risk will surface. Exact prediction is not necessary. If a system repeatedly rewards speed while allowing others to absorb defects, one need not predict the next defect to recognize the exposure. If forecasts are punished for accuracy when accuracy is unwelcome, one need not anticipate the next distortion to understand why distortion will occur.

Responsibility grows when the conduct is repeated, legible from the incentive structure, beneficial to the actor, and costly to resist. It weakens when the act is genuinely idiosyncratic, carefully concealed, contrary to enforced norms, and unsupported by any meaningful advantage. The point is not to make leaders liable for every human failure. It is to prevent them from treating patterned behavior as an inexplicable departure from the system.

There is another complication. Leaders often inherit incentives they did not create. Targets, reporting rhythms, promotion criteria, and informal penalties can predate the person now responsible for them. Authorship and accountability are not always the same.

“I did not design this” may explain the origin of the condition. It does not resolve responsibility once the condition is visible and the authority to alter it exists.

This is one of the less comfortable properties of leadership. Authority includes custody over arrangements one did not choose. The inherited system may contain commitments that cannot be changed quickly and risks that cannot be removed without creating others. Even so, the leader inherits the obligation to recognize which behaviors those arrangements make rational.

That obligation is not satisfied by announcing better intentions. It requires an honest account of the trade being offered to the people inside the system. What do they gain by taking the risk? What do they lose by refusing it? Who receives the near-term benefit? Who carries the delayed consequence? Which standard becomes negotiable when two commitments collide?

These are not questions about motivation in the sentimental sense. They reveal how authority has distributed advantage and exposure.

The answers may show that the organization has made a risk rational for reasons leadership is willing to accept. Not every risk should be eliminated. Some are inseparable from ambitious commitments, incomplete information, and competitive action. Accountability does not require a world without danger. It requires leaders to stop pretending that accepted risks were unforeseeable merely because no one gave a direct order.

The honest position is sometimes to acknowledge that a target will produce pressure, that the pressure may create specific distortions, and that leadership has chosen to carry that exposure. Once named, the risk can be bounded, monitored, or consciously accepted. What matters is that it is no longer disguised as a personal failure that begins only when the consequence becomes visible.

Incentive design is therefore not a technical matter delegated to compensation plans, performance systems, or operating reviews. It is a declaration of which judgments the organization will make sensible. Every incentive grants permission more broadly than its wording suggests because people must interpret how far they are expected to go in pursuit of it.

Leaders cannot control every interpretation. They do, however, remain responsible for the arithmetic they place around consequential choices.

The deepest exercise of authority is not approving each decision. It is determining which decisions a capable person can defend as reasonable. When the dangerous act becomes the rational price of remaining successful, the instruction has already been given.

WHO THEY BECAME.
WHAT THEY BUILT.
WHY IT MATTERED.
Marvin's new podcast, American Hustle, is in production now.
American Hustle with Marvin V. Acuna is in production now.

Browse More insights BY MARVIN V. Acuna