Framework12 min read
Care as a Decision Variable
A framework for bringing moral attention into high-stakes decisions
Why some decisions deserve more caution than their expected value suggests
Expected value reasoning is a useful tool for many decisions. But it systematically underweights a feature that matters enormously in practice: whether a mistake can be corrected.
Expected value reasoning is a useful tool for many decisions. But it systematically underweights a feature that matters enormously in practice: whether a mistake can be corrected.
Two decisions with identical expected values are not equivalent if one is reversible and the other is not. The reversible decision allows for learning and adjustment. The irreversible one does not. This asymmetry is not captured in expected value calculations — but it should be captured in how we approach decisions.
A decision is irreversible when its consequences cannot be undone, or when undoing them would require costs so high that reversal is not a realistic option.
Some irreversibility is obvious: decisions that cause permanent harm to people, decisions that destroy relationships or institutions, decisions that close off options that cannot be reopened. But much irreversibility is subtle. A decision that creates a new dependency. A decision that sets a precedent. A decision that changes how people understand what is acceptable.
The practical implication is not to avoid irreversible decisions. Many of the most important decisions — hiring, committing to a strategy, making a public commitment — are irreversible by nature. The implication is to treat them differently.
Specifically: to require a higher standard of reasoning before making them, to invest more in understanding what could go wrong, and to be more willing to slow down when the picture is unclear.
This is not risk aversion. It is appropriate caution in the face of asymmetry. A reversible mistake costs you the time and resources required to correct it. An irreversible mistake costs you the option of correction entirely.
Before making a significant decision, ask: "If this turns out to be wrong, what would it cost to undo it?"
If the answer is "not much" — proceed. If the answer is "a great deal" or "it cannot be undone" — slow down. Not indefinitely. But enough to be sure the reasoning is sound.
Work with me
If this kind of thinking is useful to you, I work directly with a small number of leaders on their hardest decisions.
Get in touch