Chapter 1 — Why Leadership Evaluation Fails

Leadership is most often judged by outcomes. When things improve, the leader is praised. When they deteriorate, the leader is blamed. The logic feels intuitive, even responsible. But it quietly collapses two different questions into one: Did this leader produce results, and were those results achieved in a morally permissible way? When those questions are treated as equivalent, success becomes virtue and failure is mistaken for moral innocence. Harm is excused by effectiveness; restraint is dismissed as weakness—even when the systems being praised are efficient precisely because they suppress dissent, concentrate power, or externalize suffering.

This collapse is not merely cultural; it is cognitive. Research in moral psychology has demonstrated that people consistently judge decisions based on outcomes rather than decision quality—treating a lucky result as evidence of wisdom, and an unlucky result as evidence of recklessness. Once results are known, the uncertainty present at the moment of choice disappears from judgment. The path that succeeded feels obvious; the path that failed feels reckless. Leadership judgment inherits this bias wholesale.

A project manager who delivers ahead of schedule is celebrated. What is less visible are the late nights, the burnout, the quiet attrition that follows months later. A principal who raises test scores is praised, even if the gains required narrowing which students were allowed to succeed. Results are legible; harm distribution is not. Outcome-based evaluation rewards what can be counted and ignores what was prevented.

Crisis intensifies the distortion. In emergencies, decisiveness often becomes a moral proxy. Leaders who act quickly, consolidate authority, and impose compliance are described as strong. Leaders who pause to consult, preserve dissent, or share risk are described as hesitant. Once outcomes are known, hindsight completes the error: success feels inevitable and failure feels culpable.

This distortion is easiest to see in the following case.

The Problem with Outcomes

Consider two leaders faced with the same looming threat.

Both are presented with incomplete information. Both know action will cause harm. Both know inaction may cause harm of a different kind. Neither can know the full shape of the future they are about to create.

The first leader acts early. They impose restrictions that disrupt daily life. Businesses close. Routines fracture. The costs are immediate and visible. People are angry. The harm is undeniable, even if its purpose is contested. Later, fewer deaths occur than might have been expected—but there is no control group. The harm that did not occur leaves no evidence behind.

The second leader waits. They emphasize calm. They avoid disruption. Life continues as normal for a time. The immediate costs are low, and approval remains high. When the harm finally arrives, it arrives diffusely and unevenly. Responsibility is distributed across time, institutions, and individuals. By then, the narrative has shifted: no one could have known; the information was unclear; the situation was unprecedented.1

In hindsight, one leader is often judged harshly for the harm they visibly caused. The other is forgiven—or even praised—for the harm that appeared later, diluted by distance and uncertainty.

This is outcome bias at work.²

We reward leaders not for the moral posture they adopted toward harm, but for how cleanly the harm they allowed can be obscured by time, complexity, or plausible deniability. Visible harm is treated as evidence of failure. Invisible harm is treated as inevitability.³

Leadership, however, does not occur in hindsight. Moral judgment that relies on outcomes alone quietly erases the conditions under which decisions were made. It confuses luck with virtue and substitutes narrative coherence for ethical evaluation.⁴

Another failure follows close behind: judging leaders by character rather than conduct. Leaders are labeled good or bad, principled or corrupt, as though morality were a stable personality trait rather than a pattern revealed under pressure. Social psychology has long shown that we over-attribute behavior to disposition and under-attribute it to context, especially when power is involved. Hero narratives and villain narratives spare us the harder work of examining incentives, constraints, and authority.⁵

Time compounds the confusion. As consequences unfold long after a leader’s involvement, legitimacy travels forward through institutions while accountability remains fixed at the origin. Founders are sanctified to excuse institutions they would not recognize, or condemned for harms they never authorized. Authority inherits; responsibility does not.⁶

One signal that outcome-based judgment often overlooks is whether authority was visible enough to be predictable without being threatening enough to require enforcement.

Authority that must be concealed to appear kind, or asserted aggressively to remain intact, is already unstable. In both cases, power exceeds the trust supporting it. What looks like strength or humility on the surface often masks the same failure underneath: authority that provokes fear when asserted and resistance when left implicit.

Here, authority does not mean the ability to compel obedience. It refers to legitimacy that is accepted without force—and becomes unstable precisely when it must rely on threat or concealment to function.

This failure of evaluation is not confined to politics or religion. It appears in organizations, movements, and families—any where authority outlives the conditions that once constrained it. Wherever power persists without proximity to consequence, moral clarity erodes.

A more honest evaluation of leadership begins by separating what we have learned to conflate. This lens draws attention to three questions:

• How did this leader relate to harm when they had the power to choose otherwise?

• How effective was their influence in shaping behavior or systems?

• How was their authority allowed to propagate after their control ended?

These questions interact, but they are not interchangeable.

“Leadership is often judged by its outcomes because outcomes are easy to see, not because they are sufficient for moral judgment.”

Footnotes

1. Tetlock, Philip E. Expert Political Judgment: How Good Is It? How Can We Know? Princeton, NJ: Princeton University Press, 2005. ↩

2. Baron, Jonathan, and John C. Hershey. "Outcome Bias in Decision Evaluation." Journal of Personality and Social Psychology 54, no. 4 (1988): 569–579. ↩

3. Hirschman, Albert O. The Rhetoric of Reaction: Perversity, Futility, Jeopardy. Cambridge, MA: Belknap Press of Harvard University Press, 1991. ↩

4. Kahneman, Daniel. Thinking, Fast and Slow. New York: Farrar, Straus and Giroux, 2011. ↩

5. Ross, Lee. "The Intuitive Psychologist and His Shortcomings: Distortions in the Attribution Process." In Advances in Experimental Social Psychology, edited by Leonard Berkowitz, vol. 10, 173–220. New York: Academic Press, 1977. ↩

6. Arendt, Hannah. Responsibility and Judgment. Edited by Jerome Kohn. New York: Schocken Books, 2003. ↩